XFiles Friday: Hume and skepticismNovember 30, 2007 — Deacon Duncan
(Book: I Don’t Have Enough FAITH to Be an ATHEIST, by Geisler and Turek, chapter 2)
It’s time to tackle Chapter 2, which promises to take on no less than David Hume and Immanuel Kant, in the name of defending the intellectual integrity of Christianity’s exclusivist claims. To hear Geisler and Turek tell it, though, these guys were tyros compared to the philosophical genius of Warner Brothers cartoons.
We’re pretty much finished with Chapter 1, which ends with a mildly Emperor’s-New-Clothes-ish reference to “once I was blind but now I see,” i.e. “some of us DO know the truth about religion.” The authors never quite specify who these enlightened ones might be (three guesses?), so we’ll just skip over that part and move on to Chapter 2.
Chapter 2 opens with some remarks leading to the conclusion that the best way to determine whether or not something is true is to examine it for (self-)consistency, coherence, and completeness (i.e. does it address all the relevant evidence in a satisfactory manner). They’re trying to imply that Christianity does the best job of meeting these standards, which is going to prove false, however the standards themselves are excellent. They are, in fact, simply an elaboration of our fundamental doctrine that truth is consistent with itself.
Moving on, we find Geisler and Turek indulging in another anecdote about a college professor (who might or might not have been an atheist) making a bad argument against Christianity. The stated point of the story is to prove that logic is the same in India as it is in the United States (versus the professor’s claim that Eastern logic and Western logic were somehow different), but the subtext appears to be that university professors don’t know what they’re talking about.
And then, it’s on to David Hume.
Perhaps more than any other person, David Hume is responsible for the skepticism prevalent today. As an empiricist, Hume believed that all meaningful ideas were either true by definition or must be based on sense experience… Hume asserted that propositions can be meaningful only if they meet one of the following two conditions:
- the truth claim is abstract reasoning, such as a mathematical equation or a definition (e.g., “2+2=4” or “all triangles have three sides”); or
- the truth claim can be verified empirically through one or more of the five senses
While he claimed to be a skeptic, Hume certainly wasn’t skeptical about these two conditions–he was absolutely convinced he had the truth.
Geisler and Turek, in trying to make Hume sound self-defeating, have bit off more than they can chew, since they must now prove that it is impossible to derive Hume’s conclusions through logical reasoning and/or empirical verification. (Notice, too, that they are once again confusing skepticism with mere denial.) Are they up to the task?
Do you see the implications of Hume’s two conditions? If he’s correct, then any book talking about God is meaningless. You might as well use all religious writings for kindling.
Oops, that’s an appeal to the Fallacy of Consequences: If Hume is correct, then that would make our religious beliefs untrue. We don’t want that! Therefore Hume must be wrong.
We could, if we want to, stop reading right here. Geisler and Turek have just admitted that God cannot satisfy either of Hume’s two conditions for meaningful truth claims. After spending the first part of Chapter 2 agreeing that the truth about God must be (self-)consistent, coherent, and complete, and after showing that all Hume is asking for is a demonstration that one’s claims are either (valid) abstract reasoning or verifiable via empirical evidence, the authors reject Hume is based on the fact that no empirical support for God’s existence can be provided. The whole premise of their book, therefore, is a lie.
But let’s keep going anyway, and look again at the principle of empirical verification.
The principle of empirical verifiability claims that a proposition can be meaningful only if it’s true by definition or if it’s empirically verifiable.
That may be a tad oversimplified–I’d have mentioned the possibility of indirect verification, for instance–but it’s essentially sound. What it’s really saying is that truth is consistent with itself. If we’re going to talk about what’s true, we have to use words, which are symbols representing concepts. The concept, in turn, is an abstract mental reference to something. So words point to concepts, and concepts point to something that–hopefully–exists outside of the mind.
Now, let’s hear Geisler’s refutation, given, true to form, as an anecdote about a report he did in a university classroom. Once again, the university professor is (surprise surprise) a self-confident and self-defeating unbeliever.
At the beginning of the next class, the professor said, “Mr. Geisler, we’ll hear from you first. Keep it to no more than twenty minutes so we can have ample time for discussion.”
Well, since I was using the lightning-fast Road Runner tactic, I had absolutely no trouble with the time constraints. I stood up and simply said, “The principle of empirical verifiability states that there are only two kinds of meaningful propositions: 1) those that are true by definition and 2) those that are empirically verifiable. Since the principle of empirical verifiability itself is neither true by definition nor empirically verifiable, it cannot be meaningful.”
…In just the second class period, the foundation of that entire class had been destroyed!
Ta da! Super-Christian Geisler saves the day from the forces of evil atheistic university professors and their self-refuting philosophies! While we wait for the applause and cheering to die down, let’s take a moment to think about what Geisler is actually claiming here. He’s not claiming to have found a loophole in empiricism. He’s not even claiming to have fully understood what empiricism is really saying. He’s claiming that, by playing a simple word-game with an over-simplified statement of empiricism he personally had destroyed the very foundation of empiricism!
Did he really destroy empiricism, though? Or did he merely give himself a smart-ass excuse for not putting any effort into learning the material? The best way to answer that question is to see if we can derive empirical verifiability using the two principles stated above.
Let’s start with the premise that reality is truth. That’s a “truth by definition,” since there could conceivably be other, more subjective or fanciful “truths” unrelated to reality. For this discussion, however, we’re concerned with real-world truth, so we’ll adopt “reality is truth” as our starting definition.
Next, let’s take the premise that truth is consistent with itself (i.e. does not contradict itself). That’s truth by definition (since, by definition, any contradiction of the truth is “not truth”) and by observation (since we consistently observe in real life that truth is self-consistent).
Now, since truth does not contradict itself, we find that there are two ways we can empirically verify the truth: by direct means (we directly observe reality) and by indirect means (some unobservable reality has observable consequence, by which we can verify the unobservable reality). This is a possibly subtle point which is where Geisler likely missed the boat. Indirect empirical verification allows us to empirically verify things that we cannot directly observe via the five senses, such as atoms, gravity, and constants like pi.
Geisler and Turek themselves made use of this very principle earlier in Chapter two, in arguing that logic is the same in the East as it is in the West.
In India, just like in the United States, buses hurt when they hit you, 2+2=4, and the same gravity keeps everyone on the ground. Likewise, murder is wrong there just as it is here. Truth is truth no matter what country you come from. And truth is truth no matter what you believe about it.
This is the principle of empirical verification in action. Truth is consistent with itself, and thus the patterns that appear in some parts of the truth are reflected in other parts as well: we can empirically verify the pattern “truth is truth” by observing how consistently this pattern manifests itself everywhere we look. It’s reflected in the universality of math, and of the laws of nature, and even in the biological manifestations that cause us to feel pain when we are injured.
So, then, we can empirically verify that empirical verification allows us to discern meaning, including intangible meanings like numbers, logic, and natural laws. Can we empirically verify that such meaning can only come from either definition or empirical verification? Certainly. (I might need to over-simplify a bit here, but hopefully I’ll get my point across.)
To be meaningful, a proposition must employ meaningful concepts, since any meaningless term in the proposition would render the meaning of the whole ambiguous at best. Note that this is an observation, not a mere preference or hope. When we observe the real world, we find that the chain is only as strong as its weakest link, so you can’t make a meaningful proposition out of meaningless concepts.
Concepts, in turn, can derive their meaning in one of two ways: either they refer to something that is part of objective reality, or they have their meaning arbitrarily assigned to them. The latter case is the “by definition” part of the principle of empirical verification. When we refer to a unicorn as being a magical creature that looks like a horse with a single horn, the concept of “unicorn” is not a reference to something in objective reality, it’s an arbitrarily-assigned definition. The definition may, in turn, refer to things that exist in reality (like horses and horns), or to other things whose meaning is based on definition (like “magical”), but the meaning of the concept itself is arbitrarily assigned via definition.
We must now show, empirically, that everything which is not assigned a meaning via arbitrary definition must obtain a meaning that is derived empirically (i.e. that is empirically verifiable). Thus, turning to concepts whose meaning is defined by reference to something in the real world, we see that there are two problems. Number one, in order for an individual to learn the meaning of a concept, they must first perceive the real-world referent to which the concept refers, otherwise the concept, for them, will be a reference to the unknown, and hence meaningless.
Number two, if only a single individual is able to perceive the real-world referent, then they will not be able to communicate that concept to anyone else, since for everyone else the concept will merely be a reference to the unknown, and hence a meaningless concept. A “meaning” which cannot be communicated is not really a “meaning” in the broad sense of the term. At best it can only be some sort of subjective, personal significance, and not part of a meaningful proposition such as could be discussed with someone else.
In order for a concept to be meaningful, therefore, it must not only be (potentially) perceptible, but must also be (potentially) perceptible to multiple persons. If multiple persons are capable of perceiving the same objective reality, however, then the reality they perceive is potentially verifiable empirically. Because the truth is consistent with itself, the same objective veracity that allows multiple people to perceive the same thing (and thus derive the same meaning for a particular concept) will also allow for the possibility of empirical verification.
Notice that it does not matter whether or not the perception involves the use of the five traditional senses. If there were a telepathic alien, say, who was communicating with Joe Bob and Jim Bob via direct access to their minds, we could empirically verify the objective reality of the alien and its telepathic communications by having Joe Bob and Jim Bob relay messages to one another via the alien, or by careful observation and analysis of the messages the two brothers reported receiving. Granted, it’s not a sure thing (if the alien decides to be perverse, for example), however it is at least possible in principle that such things could be verified.
Again, notice that I’m making only empirical observations here. This is the way logic works as we observe it in the real world (and it works the same in India as it does in the United States). Hume’s observations are not self-defeating, but self-supporting and self-deriving. You do have to expend the necessary mental effort to think things through instead of turning to Saturday morning cartoons for your philosophical “wisdom,” but it can be done.
What’s especially ironic is that, by attempting to undermine empiricism, Geisler and Turek are contradicting virtually everything they themselves have been saying up to now about how the truth can be known and what sort of evidence we ought to look for in order to correctly distinguish between truth and untruth. It’s as though they’ve been gritting their teeth and forcing themselves to pretend they believed that faith should be based on verifiable evidence, but when they got to Hume, the strain was too great, and they had to vent their hostility to the whole idea of measuring truth by its consistency with objective reality.
I can’t wait until they take on Kant.