Reason and Fallibility

Aristotle (385-323 BCE)

At least some members of the New Atheists embrace skepticism in a general way to explain why they 1) reject belief in God exists and 2) also reject belief in the nonexistence or impossibility of God. So they endorse negative atheism, but not positive atheism. This is ironic because some theologians have embraced skepticism to explain why they believe in God.

The atheists I have in mind say that our fallibility demonstrates that reason, although it usually works, is "limited" or is not "absolute." These are suspect terms though. Reason is of course limited in the sense that everything is limited: to exist is to exist as something specific, and to exist as something specific is to have a limited number of attributes. A is A means that A is limited to being A and cannot also be not-A (at the same time and in the same respect). This is hardly a controversial point. See Aristotle for details.

As for reason not being absolute, this brings us to the heart of the matter. If this is supposed to mean that reason doesn't deliver omniscience or infallibility, then don't stop the presses. This is not a shortcoming of reason. In fact, we need reason (grounded in logic) precisely because we are fallible. We would not need it otherwise. 

Omniscient God would not need reason. God would just know everything without effort. But human beings must acquire knowledge through effort. Since we can go wrong, we need a guide to keep us on the right path. Reason is the faculty -- the power! -- we have to engage in that effort. It's the process (to use Ayn Rand's term) by which we identify and integrate our sensory evidence and then form ever-broader levels of concepts, enabling us to have knowledge that goes beyond concrete sense perception. We all want to know about and act in the world, so we don't want to err. We use reason to avoid error. But for lots of reasons we make mistakes. If we find a contradiction in our reasoning, it's a sign we've made an error along the way. So we use reason to correct it. Note well: reason corrects our mistakes. 

Error is not a shortcoming of reason. It's like a tool. Look at it this way. When a carpenter hammers his thumb instead of the nail, it's no shortcoming of the hammer. We don't say that an efficacious hammer is one that never would bang a thumb. The carpenter made a mistake, that’s all. It's also not a shortcoming of the hammer that it's not good at other tasks, say, turning a screw.

Here's a closer analogy. When I make a mistake in adding a column of numbers, does this show that arithmetic is limited and not absolute? Of course it doesn't. I made a mistake, which it takes arithmetic to demonstrate. It makes no sense to say that we cannot demonstrate that arithmetic is reliable because we can't help but use arithmetic to do so. There's such a thing as self-validation and the self-evident. 

Indeed, one New Atheist says we cannot demonstrate that "reason is reasonable"! (See this video, starting at 1:40.) What? Is it a problem that we can't "demonstrate" that a circle is circular? This sheer nonsense.

The mistake here is to treat reason as though it were an oracle rather than a tool or a process in the hands of fallible human beings. In other words, this sort of skeptic has yet to abandon the theist's mindset. 

Moreover, reason is not merely one among several possible tools, just as logic is not one possible starting point among many other possibilities. What could possibly count as an alternative to reason and logic?

Let's also note that reason is not a Swiss army knife. It need not do everything to do what it does. But again, strictly speaking, reason doesn't do anything. Human beings do things using their rational faculty (or not). But the fact that reason is not appropriate for all purposes is no shortcoming. We can't use reason to discover today what we'll discover tomorrow because then we'd have discovered it today and not tomorrow. (I think Karl Popper made this point.) But this is hardly a shortcoming of reason.

Before leaving this topic, a word about philosophical skepticism. Like many words, skepticism can have several senses, and we must take context into account to decipher someone's meaning. We have a narrow everyday sense of the words. An example is, "I am skeptical about the proposition that 9/11 was an inside job." This would mean that I haven't seen any convincing evidence for the proposition. But radical skepticism (whether general or domain-specific) is very different. The radical skeptic says not merely that we can't be certain but that we cannot know. That sort of skepticism entails a contradiction. If the sentence We cannot know. is true, then it's false (because we can know at least one pretty big thing). The claim that knowledge is impossible is a knowledge claim. 

I've seen atheists invoke David Hume, who has many interesting things to say about religion, but let's be careful. Hume was not merely a skeptic about religion. He also believed that "reason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them…." Does any New Atheist really want to endorse that position? Or how about this?

When I am convinced of any principle, it is only an idea, which strikes more strongly upon me. When I give the preference to one set of arguments above another, I do nothing but decide from my feeling concerning the superiority of their influence. Objects have no discoverable connexion together; nor is it from any other principle but custom operating upon the imagination, that we can draw any inference from the appearance of one to the existence of another…. [Emphasis added.]

How many New Atheists want to take that position? On the other hand, Hume did not believe we that we cannot know -- so he wasn't actually a skeptic. He thought that through an evolutionary process of imitation and natural selection  (conceived 80 years before Darwin wrote!), our sentiments, inclinations, customs, and habits can yield reliable theoretical and practical knowledge. Elsewhere he refers to "some instinct or mechanical tendency, which may be infallible in its operations." (For more on this, see Roderick T. Long's Reason and Value: Aristotle versus Rand.)

To repeat: fallibility is not reason's shortcoming; it's ours -- and that's precisely why we can and must confidently rely on reason just as sure as A is A.

Skepticism does not advance the cause of atheism, which is worthless unless it is part of a greater cause: rationalism.