Is an atheist someone who doesn’t believe in gods, or someone who believes there are no gods? It’s a hairline distinction, irrelevant for most practical purposes, but there clearly is a theoretical difference between the two definitions. Anyone who lacks belief satisfies the first, but only those who consciously reject belief satisfy the second. In the first post in this series, I argued that both definitions have some currency, but in this long-delayed sequel I want to go a little further and make a pragmatic case for preferring the second, narrower meaning of atheism.
I don’t, of course, think either definition is superior in any cosmic or objective sense. As Brian Green Adams astutely pointed out in a comment on my earlier post, words don’t have “true” meanings, but rather evolve with usage. However, we can still argue fruitfully about which meanings for words are most useful, in the sense of facilitating clear, substantive communication and keeping misunderstandings to a minimum. In this case, atheism is one term in a classification of possible beliefs about religious matters, and classifications work best when their terms correspond to categories that are prominent in our discourse and separated by easily grasped distinctions that happen to matter to us. It’s also best to avoid terms that are redundant, or that overlap too much in their definitions. Individuals might come to different conclusions about which categories are prominent enough to be worth labelling, and which distinctions are easily grasped and important, but that’s precisely why semantic issues are worth discussing.
The key issue here, in my opinion, is what it actually means to lack belief in gods. I can see three distinct positions that might be grouped together under this heading.
First, a person who actively denies the existence of gods certainly lacks belief in gods. Such a person is clearly an atheist, by all reasonable definitions.
Second, a person who is unsure whether gods exist might be said to lack belief, without embracing active disbelief. This applies regardless of whether he or she thinks the existence of gods is an unresolvable question in principle or merely an iffy question in practice: both possibilities boil down to ticking the “don’t know” box, and can be comfortably placed under the heading of agnosticism. Whether agnosticism and atheism should be regarded as mutually exclusive positions or as compatible positions that a person might hold simultaneously is really the crux of the distinction between the two alternative definitions of atheism given above, and I’ll return to this point below.
The third way to lack belief, at least hypothetically, is to be ignorant and unreflective. Imagine a person who has never considered the possibility that gods might exist, and therefore necessarily goes about his or her life as if they don’t. Is such a person an atheist, an agnostic, or something else entirely? It’s a perfectly legitimate question in theory, but in practice it’s a bit of a non-issue because virtually everyone becomes aware of at least one religion at a very early age. Even staunchly atheist parents in the most secular countries on Earth would have to go to considerable lengths to prevent their children from discovering that some people believe in such things as gods, souls and afterlives, and of course most atheist parents would probably prefer to teach their children a thing or two about religion anyway. In the real world, the issue of whether people who are unaware of religion should be considered atheists concerns only infants, and is therefore basically irrelevant to discussions of how particular viewpoints on the existence of gods should be classified. Infants don’t have meaningful viewpoints at all.
Setting aside this third form of non-belief leaves the first and second, which might respectively be summarized as rejection of belief in gods and uncertainty about whether gods exist. The former clearly can be called atheism, and the latter clearly can be called agnosticism. However, defining atheism as a mere lack of belief in gods would imply that agnostics – or at a minimum, agnostics who tend towards disbelief – are also atheists. This creates unnecessary overlap between the atheist and agnostic categories, and introduces an inconvenient and cumbersome need to distinguish between agnostic and non-agnostic atheists. The same objections can be raised against proposals that atheists who are “unwilling… to be too dogmatic” about their disbelief in gods should be called agnostic atheists, to acknowledge their lack of total certainty. Furthermore, any position can be held with varying degrees of dogmatism, and people generally understand this without any need for special terminology. Those of us who believe the Loch Ness monster doesn’t exist, but are non-dogmatic enough to accept a remote possibility that some kind of large, unknown aquatic vertebrate might lurk in the depths of the loch, don’t feel obligated to call ourselves “agnostic Nessie doubters”. Why should atheism be any different?
The tidier and more reasonable alternative, it seems to me, is to use “atheism” only to refer to overt rejection of belief in gods, and “agnosticism” only for uncertainty about whether gods exist (stopping short of overt rejection). In this scheme one doesn’t have to be absolutely certain of the non-existence of gods to call oneself an atheist, but merely certain enough to regard their existence as too unlikely to be worth highlighting. Embracing the agnostic label, on the other hand, implies a greater willingness to entertain the notion that someone is home on top of Mount Olympus (or wherever) after all. One can be on the fence between agnosticism and atheism, or wavering between the two, or even agnostic on some days and atheist on others. But the two positions are mutually exclusive in the sense that they can’t be held simultaneously, and in that they define adjacent rather than overlapping parts of a theoretical spectrum that runs from 100% theistic certainty to 100% atheistic certainty. Theists are close to one end of that spectrum, atheists are close to the other, and agnostics occupy the vast middle ground. If a term is needed to refer to both atheists and agnostics, “non-theists” should fit the bill just fine – and similarly, agnostics and theists can be grouped together as “non-atheists”. This taxonomy has the immense virtues of clarity, straightforwardness, and consistency with how the words “atheism” and “agnosticism” are used by many and probably most speakers of the English language. It’s no more “correct” than the alternatives, but in my opinion it’s a damn sight better.