Theism, Atheism, Agnosticism and Dogmatism: An Anatomy Of Some Terms

In a comment on a previous post, a reader wrote:

I've never liked the word "agnostic" anyways. To me it has always been just a weasely way to get out of being asked if you believe in God.

Broken down to its latin roots, "agnostic" means one who does not know. In my experience, no one ever asks you if you know if there's a god. They ask if you believe there's a god. Answering "agnostic" doesn't answer the question in the least. You either believe (theist) or don't believe (atheist). None of us know one way or the other, so we are all agnostics. The term therefore conveys zero information and isn't even worth uttering.

I see the appeal of this approach, but I think it’s based on a mistake that I’ll spell out below. In any case, I though it provided an excuse for looking at what we mean by the terms theist, atheist, and agnostic, which seems appropriate given the previous discussions about Dawkins’s recent programme.

Theism is straightforward: a theist believes in the existence of a God or gods. Atheism looks similarly straightforward: an atheist believes that God(s) does not exist – but this can be taken a number of ways, as we’ll look at. Similarly, an agnostic is someone who does not claim to know, or is undecided about, whether God(s) exists.

The suggestion that theists and atheists are similar in holding beliefs about the existence of God(s), as well as in not being justified to claim knowledge about these matter, seems to me to be based on a mistaken view of the relationship between belief and knowledge.

Although it’s not spelled out, it seems that the suggestion “You either believe (theist) or don't believe (atheist). None of us know one way or the other, so we are all agnostics” is based on the idea that a belief is merely something that you think likely to be right or wrong, whereas knowledge is something more certain, more indubitable. But I don’t think these interpretations accord with normal usage of the words, and lack a philosophical justification.

Some things we might claim to know for certain. According to the logical positivists (a primarily European school of philosophy centred in Vienna – it was sometimes known as the Vienna Circle – popular in the first half of the 20^th century, but less influential after some heavy attacks), statements such as ‘2 + 2 = 4’ we can know for certain so long as we understand what the terms mean. If you know what ‘2’ and ‘4’ mean, and what the ‘+’ denotes, you can see the logical truth of the claim. You don’t have to go out and do an experiment to confirm it. Similarly, you don’t need to go into the wild to confirm that a vixen is a female fox; it’s true by definition. In these cases, your knowledge is certain.

But once we move outside of these logical realms, and into the real world, we lose this certainty. Anything we believe about the real world could, in principle, need to be revised in light of new discoveries. It is always possible that another explanation, laying unthought-of, better accounts for the world, or that new facts will emerge that are inconsistent with our best current explanations. Even the things most of us would bet our houses on, such as the sun rising tomorrow, are not grounded in logical certainty (the famous problem of induction). It is logically possible that the laws of physics might change overnight, throwing the solar system into chaos. So we’re not certain, but I think most of us would say we know that the sun is going to rise tomorrow.

If this lack of certainty precluded us from claiming to have knowledge of the world around us, then we’d have robbed the word ‘knowledge’ of all real meaning – for we could only claim to know logical truths, tautologies completely obvious to an omniscient mind. So what is knowledge? Well, belief that you hold to very probably true (I realise the subjective move here) in light of good reasons (again, a subjective element). We might want to add that the belief needs also to be true, but this has always struck me as an odd requirement.

Imagine I have a belief, say that life evolved on earth. Why do I believe this? Because it is the best explanation of the evidence presented by the world, and the theory that the belief is based on is the best explanation for the problems any such theory would have to solve. Do I say I know that life on earth evolved? Yes, in as much as I can know anything. Then what if I want to determine whether I additionally ‘know’ this fact, as well as merely believing it? I’d have to determine whether my belief was true independent of my reasons for believing that it is true. What process would or could I undertake to determine this? I could search for further supporting reasons, but I’d be back where I started. The problem is that I can’t get outside of myself and the world and then look in to see whether on top of the reasons I have for holding a belief to be true (and therefore constitute knowledge) the belief is, in fact, true – if I could, what need would I have of the other reasons? And what would unbridled access to the truth be? All I’ve got to assert that I think something is true, and therefore claim to know it, are reasons. What more could I have? There isn’t any reason-, argument- or evidence-independent way of identifying truth. The best we can do is corroborate our beliefs with others, and look at how they hang together with the rest of our beliefs, and at some point make a call about what you think you know.

So, some of my beliefs are less well supported, and I don’t claim to have knowledge on these matters, and others are supported to a degree that in the only real sense it could mean I know these things. Where to draw the line between believing and knowing is tricky and takes us into epistemological waters that I am ill-equipped to chart, but the distinction along a continuum seems to be valid, and to enable us to talk about knowing anything at all apart from logical truths.

So what does all this have to do with theism, atheism, and agnosticism? Well the first point is that it is not invalid to say that you believe in God(s) and also that you know (in the non-certain sense) that it/they exist, or to say that you believe, and also know (in the non-certain sense), that they do not exist. Similarly, although it might seem weak willed, it is also valid to adopt a third position, of not claiming to believe or know either way – agnosticism.

It is true that neither theists nor atheists should claim to know whether God (assume the plural as well from here!) exists in the strong sense of absolute certainty. Such certainty is surely going to far (although listening to the faithful and the devout makes it clear that many religious folk have a certainty that they’re not really entitled to). But this doesn’t mean that they shouldn’t claim knowledge in the weaker sense of ‘best explanation and highly likely to be true’. So if agnosticism is taken to mean not knowing, in any sense, then there is no need for theists or atheists to call themselves agnostic. And agnosticism stands as a useful and conceptually distinct term to designate a position between theism and atheism.

So the theist claims to know that God exists – but on what basis? Well, some religious people do employ arguments, such as the argument from design. Others hold that they know through the mysterious power of faith (I don’t understand at least). The atheist is more likely to respond with arguments, reasons and evidence, and to claim that to extent that there are religious explanations for the world, they are far outgunned by scientific explanations. Further, the atheist is likely to claim that the evidence not only speaks for the scientific explanation, but also speaks against the theistic explanation. To the extent that theists’ beliefs are predicated on pure faith, then it is hard to engage in any meaningful debate about whether it is likely that God exists, because by introducing faith they cut themselves free from the chains of reason and argument in float off into their own orbit where they can only hear each other.

There is another, and quite different, way in which you might be an agnostic or an atheist that is not based on weighing up arguments and evidence. You might find the question of God’s existence to be meaningless in metaphysical sense. If it was deemed in advance, as faith does, that nothing you found out about the world should be taken to refute your belief in God, then you can ask what the claim that God exists really entails, what factual significance it is supposed to convey. In another context, Dan Dennett has used the example of gremlins in a carburettor to illustrate the complaint. Imagine someone proposed that carburettors got their power from seven invisible, massless, undetectable gremlins. How should you treat this claim? Well, you wouldn’t treat it like a normal empirical claim about the world, because you can’t undertake any physical tests for their presence. In an important sense, the claim has no real factual content – and because it is supposed to relate to the factual world it is literally meaningless, even though it is grammatically comprehensible. Compare it with the claim that there are ten gremlins in carburettors. How would you adjudicate between these claims? What difference would it would it make to hold one or the other? A belief that makes no difference to anything has no factual content, and can’t be considered to make a meaningful assertion. It’s not even proposing something that is in principle up for debate, so you just reject the whole gremlin question. Ditto for some conceptions of God.

Because in one sense the rejection of the God as an inherently confused concept entails not taking the notion of God seriously enough to consider evidence in favour or against its existence, is seems like a form of agnosticism, and maybe some people would like to see it this way. But it differs from the agnosticism characterised earlier, in that it is not about sitting on the fence, but in denying that there is fence there at all. So the question is considered meaningless and therefore pointless. But isn’t this also a form of atheism? Possibly, and probably yes. If you accept this line of reasoning, then you certainly do not believe that there is a God – so this non-theism, or a-theism. But it is clearly not the same as saying that while you accept the logical possibility of God, you think that the arguments and empirical evidence support the claim that there is, in fact, no God.

So, to wrap up. You can be a theist in at least two ways: by believing and knowing with complete certainty that God exists, presumably on the basis of faith, because empirical evidence would never justify such an extension (this dogmatic theism should be universally regarded as an unjustified position); or by believing that in all probability God exists, but that you could be wrong, however unlikely you think that to be the case. And you can be an agnostic in at least two ways: by accepting the possibility of there being a God, but remaining unswayed either way by the arguments and evidence (a somewhat spineless and unthinking position, because serious reflection seems to take people either down a naturalistic, scientific route or something more faith-based or mysterious); or by rejecting the question as meaningless and effectively pleading the fifth on it – you’re not going to engage with a pointless concept. Finally, you can be an atheist in perhaps four senses. First, you reject the concept of the God(s) of traditional religions as meaningless and are therefore a priori incapable of believing in God, and are therefore a non-theist or a-theist. Second, you accept the logical possibility that God exists, but claim to know with absolute certainty that God does not in fact exist; this is as unjustified as the dogmatically certain theist’s position, and dogmatisms, to paraphrase Ferris Bueller, are not good in my opinion. Third, you accept the logical possibility of God but claim that the best arguments and evidence strongly suggest that God does not exist – while admitting to a lack of certainty about this (even if it’s the thing you think is most likely to be true of all your beliefs); such an admission doesn’t mean you can’t talk about knowing that God doesn’t exist, but it needs to be clear what you intend by ‘know’. Fourth, you might reject the existence of God simply because you don’t think there are any good reasons for asserting that God exists – it’s not that you’ve balanced the reasons for and against and come out in favour against, but that there is nothing arguing in favour of God, and therefore there’s nothing to way up. Nearly anyone who thinks this probably also thinks that what evidence there is speaks against the existence of God, but I think that it’s possible to distinguish these motivations for describing yourself as an atheist.

How do I come out according to this classification? I am not a dogmatic atheist, the second version above, for sure. Most of the time I’m definitely an atheist in the third sense – I think the best evidence and theories argue against God – as well as in the second sense – there are no good reasons for asserting the existence of God, or at least what reasons there were have been superseded by scientific explanations. But I’m not always an atheist in just the third and fourth senses, in that sometimes I’m confronted with a conception of God that is more like a gremlin. In this case I'm an atheist in something more like the first sense. I don’t know how to even begin to engage in a meaningful dialogue about whether this proposed entity exists, which entails a denial of the theism that this entity’s existence is the basis of. So in this sense I’m a non-theist, or a-theist when confronted with a certain conception of God.

The overall message I guess is that when applying labels to ourselves, as it seems we have to, it might be worth checking that that we're using the terms in the same way as people listening to us, as a lot of the criticisms of atheism in particular are based on a conflation of different notions of atheism. For instance, the usual claim of critics of atheism that science doesn’t disprove good is only relevant to the dogmatic atheist, and I think we can agree that dogmatisms are bad. It has no force against atheism that makes the weaker claim that the best arguments and evidence point to God not existing, because this doesn’t claim to have established beyond all doubt that God does not exist. It admits that science hasn’t disproved God’s existence – but it doesn’t have to! We all claim to know things for which we haven’t disproved any number of alternative accounts (perhaps because the account cannot be disproved in principle), but that’s not what knowing things is about. We should all learn to live with uncertainty, and acknowledge it in cases even when the uncertainty is very small, as it is for many, if not, most, atheists.