How do we come to know things? Is it primarily through our use of reason that we acquire knowledge? Or does experience gained through our senses play the most significant role in the way we get to know the world? Much of the history of Western philosophy has been colored by this basic opposition between reason and experience as the foundational principle of knowledge. Specifically, it is the main bone of contention between two extremely influential philosophical strands—rationalism and empiricism.
Three key distinctions To understand what is at issue between rationalist and empiricist theories of knowledge, it is useful to look at three key distinctions that are used by philosophers to elucidate the differences between them.
a priori vs. a posteriori Something is knowable a priori if it can be known without reference to experience—that is, without any empirical investigation of how things actually stand in the world; “2 + 2 = 4” is known a priori—you do not have to go out into the world to establish its truth. By contrast, if such investigation is needed, something is only knowable a posteriori; so “coal is black,” if true, is an a posteriori truth—to verify it, you need to look at a lump of coal.
analytic vs. synthetic A proposition is analytic if it does not give any more information than is already contained in the meanings of the terms involved. The truth of the statement “All spinsters are unmarried” is apparent merely by virtue of understanding the meaning and relation of the words used. By contrast, the statement “All spinsters are miserable” is synthetic—it brings together (synthesizes) different concepts and so provides significant information (or misinformation in this case). To establish whether it is true or not, you would have to inquire into the mental state of every unmarried woman.
necessary vs. contingent A necessary truth is one that could not be otherwise—it must be true in any circumstances, or in all possible worlds. A contingent truth is true but might not have been if things in the world had been different. For instance, the statement “Most boys are naughty” is contingent—it may or may not be true, depending on how the majority of boys actually behave. By contrast, if it is true that all boys are naughty and that Ludwig is a boy, it is necessarily true (in this case, as a matter of logic) that Ludwig is naughty.
The analytic/synthetic distinction has its origins in the work of the German philosopher Immanuel Kant. One of his main objectives in the Critique of Pure Reason is to demonstrate that there are certain concepts or categories of thought, such as substance and causation, which we cannot learn from the world but which we are required to use in order to make sense of it. Kant’s main theme is the nature and justification of these concepts and of the synthetic a priori knowledge that stems from them.
“Mathematics has not a foot to stand upon which is not purely metaphysical.”
Thomas De Quincey, 1830
There seems to be an obvious alignment between these distinctions: so, on the face of it, an analytic statement, if true, is necessarily so and is known a priori; and a synthetic proposition, if true, is contingently so and is known a posteriori. In fact, however, things are not nearly so neat, and the chief difference between empiricists and rationalists can be captured by the different way they choose to line up these terms. Thus the task of rationalists is to show that there are synthetic a priori statements—that significant or meaningful facts about the world can be discovered by rational, nonempirical means. Conversely, the aim of the empiricist is often to show that apparently a priori facts, such as those of mathematics, are in fact analytic.
In the conflict between empiricism and rationalism, the field of mathematics has been the battleground where the most intense fighting has taken place. For the rationalist, mathematics has always appeared to offer a paradigm of knowledge, presenting a realm of abstract objects about which discoveries could be made by the exercise of reason alone. An empiricist cannot let this go unchallenged, so is obliged either to deny that mathematical facts can be known in this way, or to show that such facts are essentially analytic or trivial. The latter course usually takes the form of arguing that the supposedly abstract facts of mathematics are actually human constructs and that mathematical thinking is at root a matter of convention: in the end there is consensus, not discovery; proof, not truth.
Alternatives to foundationalism Rationalists and empiricists may differ on many things, but they do at least agree that there is some basis (reason or experience) on which our knowledge is founded. So, for instance, the 18th-century Scottish philosopher David Hume may criticize Descartes for his chimerical search for a rock of rational certainty from which he could corroborate all our knowledge, including the truthfulness of our senses. But Hume does not deny that there is any basis, merely that this foundation can exclude our common experience and natural systems of belief.
So both rationalism and empiricism are essentially foundationalist, but there are other approaches that dispense with this basic assumption. An influential alternative is coherentism, in which knowledge is seen as an interlocking mesh of beliefs, all the strands of which support each other to form a coherent body or structure. But it is, though, a structure without a single foundation, hence the coherentist slogan: “every argument needs premises, but there is nothing that is the premise of every argument.”
Historically, the British empiricists of the 17th and 18th centuries—Locke, Berkeley and Hume—are often ranged against their Continental “rivals,” the rationalists Descartes, Leibniz and Spinoza. But as usual, such easy categorizations obscure much of the detail. The archetype on one side, Descartes, often shows himself to be sympathetic to empirical inquiry, while Locke, the archetype on the other, sometimes appears willing to grant the space that rationalists would give to some form of intellectual insight or intuition.
the condensed idea
How do we know?
| Timeline | |
|---|---|
| c.350BC | Forms of argument |
| AD1670 | Faith and reason |
| 1739 | Science and pseudoscience |
| 1781 | Reason and experience The categorical imperative |
| 1963 | The tripartite theory of knowledge |