Gorgias the Nihilist, an ancient Greek philosopher, was said to have argued the following four points:
- Nothing exists;
- Even if something exists, nothing can be known about it; and
- Even if something can be known about it, knowledge about it can’t be communicated to others.
- Even if it can be communicated, it cannot be understood.
Of course, if you can understand his argument, he’s wrong. So too, many modern thinkers hold to positions that fall apart into self-refutation when critically examined.
Today, I want to look at three such popular claims. In showing their inherent contradictions, I hope to show why we can (and must) affirm that knowable, non-empirically testable, absolute truths exist.
Step 1: Answering Relativism
The claim: “Absolute truth does not exist.”
Why it’s self-refuting: The claim “Absolute truth does not exist” is either absolutely true or it’s not. But, of course, it can’t be absolutely true, since that would create a contradiction: we would have proven the existence of an absolute truth, the claim itself. Since it cannot be absolutely true, we must concede that there are some cases in which the proposition “Absolute truth does not exist” must be false…in which case, we’re back to affirming the existence of absolute truth.
What we can know: Absolute truth exists. Put another way, the claim “absolute truth exists” is absolutely true.
Step 2: Answering Skepticism
The claim: “We can’t know anything for certain.” Or “I don’t know if we can know anything for certain.”
Why it’s self-refuting: This one is a subtler self-refutation then the first, because it looks humble. After all, if I can say, “I don’t know the number of stars in the universe,” why can’t I take it a few steps further, and say, “I can’t know anything for certain”?
Simple. Because in saying that, you’re claiming to know something about your own knowledge. When we say, “I don’t know x,” we’re saying, “I know that my knowledge on x is inconclusive.”
Take the most mild-seeming statement: “I don’t know if we can know anything for certain.” What you’re really saying is that, “I know that my knowledge on whether anything can be known for certain is inconclusive.” So you’re still affirming something: that you know your knowledge to be inconclusive.
There are two ways of showing this. First, because it could be a lie. The claim “I don’t know who took the last cookie” could very well be proven false, if we later found the cookie in your purse. So these “I don’t know” claims are still affirming something, even if they’re just affirming ignorance.
Second, apply the “I don’t know” to another person. If I said, “You don’t know anything about cars,” I’m making a definitive statement about what you do and don’t know. To be able to make that statement, I have to have some knowledge about you and about cars. So if I was to say, “You don’t know if we can know anything for certain,” I’d be claiming to know that you were a skeptic—a fact that I can’t know, since I’m not sure who’s reading this right now.
So when you say “I don’t know if we can know anything for certain,” you’re saying that you know for certain that you’re ignorant on the matter. But that establishes that things necessarily can be known for certain.
This is unavoidable: to make a claim, you’re claiming to know something. So any positive formulation of skepticism (“no one can know anything for certain,” “I can’t know anything for certain,” “I don’t know anything for certain,” etc.) ends up being self-refuting. For this reason, the cleverest skeptics worded their skepticism as rhetorical questions (e.g., de Montaigne’s “What do I know?”). If they were to say what they’re hinting at, it would be self-refuting. They avoid it by merely suggesting the self-refuting proposition.
Finally, remember that in Step 1 we determined that the claim “absolute truth exists” is absolutely true. We’ve established this by showing the logical contradiction of holding the contrary position. In other words, we’ve already identified a truth that we can know for certain: “absolute truth exists.”
What we can know: Absolute truth exists, and is knowable.
Step 3: Answering Scientific Materialism
The claim: “All truth is empirically or scientifically testable.”
Why it’s self-refuting: The claim that “All truth is empirically or scientifically testable” is not empirically or scientifically testable. It’s not even conceivable to scientifically test a hypothesis about the truths of non-scientifically testable hypotheses. In fact, “All truth is empirically or scientifically testable” is a broad (self-refuting) metaphysical and epistemological claim.
What about the seemingly moderate claim, “We cannot know if anything is true outside of the natural sciences”? Remember from Step 2 that “I don’t know x” means the same as saying, “I know that my knowledge on x is inconclusive.” Here, it means, “I know that my knowledge on the truth of things outside of the natural sciences is inconclusive.” But the natural sciences can never establish your ignorance of truths outside the natural sciences. So to make this claim, you need to affirm as certain a truth that you could not have derived from the natural sciences. So even this more moderate-seeming claim is self-refuting.
Furthermore, all scientific knowledge is built upon a bed of metaphysical propositions (for example, the principle of noncontradiction) that cannot be established scientifically. Get rid of these, and you get rid of the basis for every natural science. There’s no way of rejecting these premises while still affirming the conclusions that the natural sciences produce.
Finally, remember that in Step 2, we established the truth of the claim “Absolute truth exists, and is knowable.” This is a truth we know with certainty, but it’s not an empirical or scientific question. It can be established simply by seeing that its negation is a contradiction. So that’s a concrete example of an absolute truth known apart from the empirical and scientific testing of the natural sciences.
What we can know: There exists absolute and knowable truth, outside of the realm of the natural sciences, and not subject to empirical and scientific testing.