Philosophy and science are both ways of learning about ourselves and the world.
Here we’ll review the two main perspectives on the question of whether and to what degree science and philosophy overlap in their methods and their sources of knowledge.
We’ll start with what has historically been the most dominant view of the nature of philosophy: let’s call this view ‘rationalism.’ After looking at this traditional perspective, we’ll review a more recent view of what philosophy is or should be: ‘naturalism.’
1. The Traditional View: Rationalism
According to rationalism about philosophy itself, most or all of our characteristically philosophical beliefs are formed and justified independently of sensory observation, i.e., a priori, or just by thinking about the issues.
If you know some fact a priori, then you know it independently of, or prior to, any kind of empirical, or sensory-based, or perceptual observation of that fact: e.g., we know a priori that all even numbers are exactly divisible by two; we don’t need to perform scientific experiments to learn that fact. In contrast, to know whether the number of planets in the solar system is even, a priori investigation wouldn’t be enough: we need empirical or sensory- or experience-based information.
If we think that characteristically philosophical beliefs may be justified or known a priori (if and when they are justified or known), then this will tell us a lot about the differences between science and philosophy. Here’s a summary of a way we could draw the contrasts:
- Science is about empirical knowledge; philosophy is often about that but is also about a priori knowledge (if it exists).
- Science is about contingent facts or truths; philosophy is often about that but is also about necessary truths (if they exist).
- Science is about descriptive facts; philosophy is often about that but is also about normative and evaluative truths (if such truths exist).
- Science is about physical objects; philosophy is often about that but is also about abstract objects (if they exist).
To explain these contrasts, contingent facts are facts that are actually the case, but they didn’t have to be: they could have been otherwise: necessary truths are true claims that must be true. Descriptive facts are about how things actually are; normative and evaluative truths are about how things should or shouldn’t be and their goodness and badness. Physical objects are—well, that’s not easy to describe, but perhaps they’re objects that are made of matter and have locations in space and time; abstract objects, if they exist, are somehow not in the physical universe.
Notice the recurrence of the word “also” above. Philosophy often is about empirical knowledge of contingent, descriptive, and physical facts; it’s just that those are chiefly the focus of science. And notice the recurrence of the parenthetical “if they exist”; some philosophers have denied the existence of a priori knowledge; of necessary truths; of normative truths; and of abstract objects.
This approach to contrasting philosophy with science is supported by examining the traditional definition of the scope of philosophy. Philosophy is often thought to have three main sub-areas: metaphysics (the nature, structure, and contents of reality), epistemology (the nature and scope of knowledge and evidence), and ethics (good, bad, right, and wrong), generally speaking.
If that’s accurate, then that fits very well with the description of rationalism above. Metaphysics is largely about necessary truths and abstract objects. Epistemology is about the arguably normative and evaluative questions of what we should believe: what are we justified in believing, what do we know, and is knowledge better than true beliefs? And ethics is clearly normative: it’s about evaluative questions of good and bad and right and wrong.
And if there is any knowledge or justified belief about these issues, this is a priori; there’s no need to run scientific experiments in laboratories, and it’s not clear what such experiments could be.
2. A Newer Perspective: Naturalism
The word ‘naturalism’ is used in many ways in philosophy: here we are addressing what can be called ‘metaphilosophical’ naturalism. This means that this theory is a naturalism about philosophy itself, not merely about some philosophical issue or topic: it’s a philosophical view of philosophy itself.
The word “naturalism” also gives us a clue about what the theory says: it says that there’s some strong connection between philosophy and the natural and social sciences. This could mean, e.g., that scientific observations, such as from neuroscience or cosmology, are relevant to traditional philosophical concerns.
Most philosophers agree that science can have important things to say about philosophical issues, and some even argue that philosophy should largely be replaced by empirical science: e.g., perhaps we should simply believe all and only what our best scientific theories tell us about reality, and if they don’t say anything about some traditional philosophical concern, then we won’t say anything about that, either. This kind of view can also be called ‘naturalism.’
If we accept something like naturalism as defined in either of these ways, then it will be more difficult to define philosophy simply by contrasting it with science. But, in the bulleted list above, if we simply replace each occurrence of philosophy “is also about” with ‘traditionally, “philosophy” has been more about,’ we’ll end up with something that nearly all philosophers would agree with, including naturalists.
Naturalists believe that we should use science to investigate traditionally philosophical questions, but most would admit that their position is relatively new on the scene. They acknowledge that philosophers have traditionally understood their subject matter as pursued primarily through the use of intellect, conceptual investigation, logic, and intuition.
The sorts of topics studied in philosophy classes are definable, in general, by contrasting them with the topics and methods of the natural and social sciences.
Are philosophical issues ultimately suited to a priori inquiry, or are they instead more appropriately considered the objects of scientific observation? Understanding how rationalists and naturalists answer this question, and comparing their insights and methods in justifying their answers, can help us understand and develop our own philosophy of philosophy.
 The word “rationalism” is used in many different ways, not only in philosophy. But this label strikes me as best; “intuitionism” would commit to a stronger claim; “nonnaturalism” also suggests stronger claims and isn’t inherently very informative; and “apriorism” would be confusing. Beyond this, some rationalists have argued influentially that non-rationalistic philosophy is self-defeating (BonJour 1998: ch. 1; Bealer 1996; Huemer 2017). If there is even a minimally convincing case to be made for this conclusion, then this is further reason to suggest that philosophy itself is fundamentally a priori in an important way. We may affirm this conclusion without going so far as to say that one must assume the truth of rationalism in order to acquire philosophical knowledge.
 That doesn’t mean that we can know something a priori without making any kind of observation at all; we might have to use our eyes to discover what various colors look like. But some philosophers say that once we have the concepts of red and green, for example, we can know just by thinking that nothing can be all red and all green at the same time (cf. BonJour 1998: 2). You know it just by thinking about it; or just because when you consider the idea of something all red and all green, it just seems impossible; or just because you can’t even begin to imagine such an object. Note that strictly speaking, a priori learning may not require “thinking.” For example, if I see “1=1” written somewhere, I instantly realize that the equation is true, and I don’t really have to think much about it. But the “just by thinking about it” explanation is typically accurate of a priori knowledge.
 For an explanation of the distinction between contingent and necessary truths, see Possibility and Necessity: An Introduction to Modality by Andre Leo Rusavuk.
 Cf. Plato 1997, bks. V-VII.
 Quine 1961: 43.
 Quine 1969: §§ 1-4.
 Cf. Ridge 2013.
 Cf. Loux 2006: ch. 2; Armstrong 2002.
 Loux 2006.
 BonJour 2002: ch. 10; Huemer 2002; but contrast Kornblith 2002 and Goldman 1986.
 Cf. Huemer 2005: ch. 4; Ridge 2013.
 Quine 1969; Goldman 1986; Kornblith 2002; cf. Papineau 2013.
 See e.g. Papineau 2013.
 Papineau 2013.
Ridge, Michael. 2013. “Moral Non-Naturalism.” In E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy, Fall 2013 edition. URL = http://plato.stanford.edu/archives/fall2013/entries/moral-non-naturalism/.
What is Philosophy? by Thomas Metcalf
Epistemology, or Theory of Knowledge by Thomas Metcalf
Possibility and Necessity: An Introduction to Modality by Andre Leo Rusavuk
Ethical Realism, or Moral Realism by Thomas Metcalf
This essay was originally posted on 2/23/2018. A significantly updated version was posted on 10/8/2021.
About the Author
Tom Metcalf is an associate professor at Spring Hill College in Mobile, AL. He received his PhD in philosophy from the University of Colorado, Boulder. He specializes in ethics, metaethics, epistemology, and the philosophy of religion. Tom has two cats whose names are Hesperus and Phosphorus. shc.academia.edu/ThomasMetcalf