Search This Blog

Friday, January 14, 2011

Epistemology: How Do We Know Anything?

The question "how do we know anything?" is one of the most basic questions in philosophy, the starting point for the study of epistemology, the study of knowledge.  Being very abstract, and hard to spell, this subject has been passed over by the layman perhaps just as much as the exceedingly abstract subjects of metaphysics and aesthetics.  More often, when thinking of philosophy, the non-philosopher thinks primarily of ethics and occasionally of logic.  But epistemology is vital to the understanding of both of these others.  How do I know what the rules of ethics are? If I am a utilitarian, how do I identify the greatest good for the greatest number? etc. These are important questions to consider, and a great deal of effort on the part of philosophers has been dedicated to answering these questions.  And the philosophers have come to some interesting conclusions.

What may reasonably called the mainstream idea on epistemology is called logical positivism.  This idea proposes that there are precisely two sources of knowledge about the world--rationality and experience.  However, there are limitations on rationality.  Specifically, reason can only tell us things that are true by definition.  For example, starting with the tautology "all bachelors are male", we can analyze the statement "Morgan is a bachelor" to determine that Morgan is male without actually checking, and this conclusion is perfectly correct (assuming, of course, that the statement "Morgan is a bachelor" is indeed correct).  Experience also gives us  correct information (insofar as our tools for collecting it are accurate), but it only tells us about events, not causes.  We can use reason to imagine possible causes for various events, but any such explanations must be considered tentative and potentially subject to being disproved by experience.  For example, experience tells us that there is this phenomenon that we call "lightning," which consists of a bright flash of light and a bolt of energy.  We might try to explain this phenomenon by proposing that there is a being we call Thor who rides around in a chariot amidst the clouds hurling an enormous hammer.  Lightning occurs when this hammer strikes a cloud.  To test this hypothesis, we might set up various experiments, such as watching the sky to see Thor hurling his hammer (when we do not, it is because the clouds hid him from our eyes), locate Thor and see if he can cause lightning by hurling his hammer (this test is ongoing), or attempt to find lightning when Thor is elsewhere (we are awaiting confirmation of  Thor's location before proceeding).  If we perform one of these tests and find a result indicating that the hypothesis is incorrect, we might identify various extenuating circumstances; if we perform one of these tests and find a result indicating that the hypothesis is correct, we might be overjoyed at our cleverness.  In either case, we must perform the test again...and again and again and again.  We can never prove that there is a causal relationship between Thor's hammer and lightning.  Nor can we ever prove that there is no such relationship.  All of our theoretical knowledge is tentative, awaiting confirmation that can never come.

Logical positivist epistemology has several glaring problems.  For one, its assertion "all knowledge about the world is either empirical or tautological" is neither empirically justified nor tautological.  Even if it were one of these, there are difficulties.  If the statement were a tautology, it would not be telling us anything.  If the statement were empirically justified, then it would have to be rephrased in one of two ways: (1) "all thus-far examined knowledge about the world is either empirical or tautological," which says nothing about any future knowledge; or (2) "all knowledge about the world that will ever be found is either empirical or tautological."  In case (2), this statement must be considered tentative by its own standard.  That is, its validity is not known, but must be tested.  Furthermore, this process of testing will end only when all knowledge about the world has been assessed.  Considering the infinite variety of knowledge about the world, it is easy to understand why logical positivists are not eager to begin this task.  Thus, by its own standards, logical positivist epistemology cannot be considered the ultimate standard for assessing the validity of knowledge about the world.

Logical positivist epistemology can actually be disproved, and in a very elementary way--by counterexample.  Let us strip away everything logical positivism tells us we must have to know anything about the world.  We must suppose that all experience is illusory and that we will simply ignore any tautologies.  Can we know anything?  Yes.  As Decartes put it "Cogito ergo sum"; "I think, therefore I am."  Even supposing solipsism, my mind cannot be a creation of itself.  The nature of my existence is still unknown, but the fact that I exist is known.  I have thus acquired knowledge about the world without resort to any kind of experience and that is not true merely by definition since we have assumed both of those sources of knowledge away.  Therefore, logical positivist epistemology is false.  Q.E.D.

In my next post, I will elaborate on an alternative to logical positivism that is particularly applicable to praxeology.


  1. Never seen someone use Descarte's quote to disprove positivism before. But couldn't we say we don't know we exist until we experience thought? (just a question)

    Also I think you need to elaborate why the statement "'all knowledge about the world is either empirical or tautological' is neither empirically justified nor tautological." I know that the statement "all knowledge about the world is either empirical or tautological" is aprioristic, but seeing how this is an article explaining epistomology to anyone that is interested, it would be helpful to state why it is neither empirical or tautological. You did a great job explaining the contradiction of the statement flawlessly otherwise.

    Good post, I can't wait to see your next one :D

  2. Thanks Michael.

    Firstly, the statement "all knowledge about the world is either empirical or tautological" cannot be a tautology because it tells us something about the world that we didn't know before making the statement. If we define a bachelor as an unmarried man, then we know nothing more about men or the institution of marriage--we have just assigned a term to something. Secondly, the statement is not empirical because it provides no empirical justification. It is simply a proclamation ex cathedra. At best, it might be justified by induction (though I don't see how it could be), but even that is pure rationality, not empiricism.

    Hope that clears the issue up.

  3. You said that "in [your] next post [you would] elaborate on an alternative to logical positivism.." but I looked at the next couple posts and none were about epistemology. If you did make that post, can you provide a link?