February 11, 2009


Posted in Atheism, Worldview tagged , , , , at 1:30 pm by Andrew

4. Atheism is Not a “Worldview.”

This post summarizes the fourth set of answers to potential objections to my Summary Case for Atheism, in which some Christians argue that it is reasonable to believe that non-material things exist. In many cases, this argument is extended to include the claim that atheists subscribe to a “worldview” that arbitrarily excludes the supernatural.

I contend that a) atheism isn’t a worldview, and b) that my arguments survive the shared worldview both of the atheist and the theist. My ongoing, in-depth analysis of these sets of responses is permalinked to the right and can be found here.

Atheism is not a worldview, and does not entail hardcore materialism. This is such a basic issue that I’m surprised the “worldview” argument has taken such hold in contemporary apologetics.

As Wikipedia helpfully tells us, a “worldview” is defined as one’s comprehensive framework for evaluating the world, including theories of why we are here, where we’re going, how to get there, how to behave, and so on.

On face, then, atheism cannot be a worldview, because different atheists have different answers to those questions! In terms of ethics alone, we know that some atheists are Randians; some are Bethamite utilitarians; some follow more closely to John Stuart Mill; some are even Kantians. Many atheists follow Hume’s theory of epistemology, but many others do not. And so on. If atheism does not provide a single answer to these fundamental questions, it cannot be a worldview, by definition!

I think many Christians want to label atheism a worldview as part of a chain of argumentation that goes something like this: atheism means hardcore materialism, and hardcore materialism is self-contradictory; therefore, atheism is self-contradictory.

But, of course, nothing in atheism mandates hardcore materialism. At most, atheists believe in methodological naturalism, which says (roughly) that empirical claims require empirical evidence. (All of us, Christians and non-Christians alike, share this limited aspect of “worldview” — otherwise, we’d be awash in a sea of credulity.)

Contrary to the arguments of some Christians, nothing in this limited methodological naturalism rules out the supernatural a priori. Consider, for example, J.B. Rhine’s parapsychology experiments at Duke beginning in the 1930s with Zener cards and the like. (These experiments are parodied in the opening scene of Ghostbusters.)

Now as it turns out, Rhine’s studies were inconclusive at best (and in many cases, faked), but *had* they revealed statistical evidence for ESP, atheists would have had to take those results seriously using the principles of methodological naturalism. Similarly, double-blind efficacy-of-prayer studies seem to me to be a good way to test for the supernatural; it’s just that — like ESP — all of those studies to date have either been inconclusive or faked.

In other words, methodological naturalism contains no inherent bias against miracles.

Finally, Christians sometimes contend that other non-material things “exist” — like mathematics, love, morality, and the like. My short response is that these things exist as abstractions, as contingent upon the human mind. For example, we can arguably say that “five” exists — but only because we can either rationally comprehend “five” as a component of pure mathematics or because we can count out five apples, five puppies, and the like.

Note that this limited claim is, in turn, contingent on empirical verification. For example: it would obviously make no sense to claim that “five” exists, but that it was simultaneously impossible to tell if a pen full of puppies contained five, or fifty, or five million puppies. Similarly, “love” may be an abstraction, but it too manifests itself to our senses, both electrochemically in the brain and in terms of extrinsic behavior.

If someone wants to claim that God exists in the limited, contingent way that “five” or “love” exist — as an abstraction or a human mental concept — then I don’t disagree with them. But that is not Christianity.

I thus conclude that the fourth and final set of objections against my general case for atheism are not sufficient to warrant belief in God. My in-depth, updated (and lengthier) discussion of this issue can be found here.


  1. Philip said,

    This is really something you will have to take up with fellow atheists since so many people designate themselves as “atheists” when people wonder what religion they are or whatever. Thus the term “atheist,” by accident or whatever, has come to have a meaning far beyond ‘a person who believes there is no God’ (which is really all it should mean). What the majority of atheists ought to call themselves is metaphysical naturalists. However, there are a good amount of atheists who have no interest in philosophy at all, and wouldn’t even know what that means – at the popular level, then, people who imagine the world according to metaphysical naturalism are called ‘atheists.’

    Atheists who are not metaphysical naturalists are the exception to the rule, in my experience, and so we have a misnomer on our hands, but there’s not much a single person can do to change that.

  2. Andrew said,


    You might be aware that there is considerable debate in atheist/secular humanist/freethought and related communities over the use of language and descriptors, so I think this is accurate.

    One of the things I am (implicitly) responding to in this objection is the common Christian argument that it is okay to just postulate God’s existence as part of a “worldview” because atheists are doing the same thing in reverse. You sometimes see this claim as: “Well, I put my faith in God, and you put your faith in science, but we both have faith.”

    I think that argument carries no water for the reasons outlined above; both the Christian and the atheist accept methodological naturalism as a means of evaluating empirical claims like whether certain things ‘exist’ or not.

    As I go back and read the original post, I see that I didn’t draw that out particularly clearly; so I’ve added a new section on the permalinked page on this issue. Thanks for helping me clarify my thoughts.


  3. Philip said,

    By making sure that ‘atheism’ isn’t a worldview, it doesn’t seem like you are responding to the argument you reference. You will still just be taking the word ‘atheism’ off the table for designating whatever the person’s worldview actually is. The Christian only means to say, ‘Whatever your worldview is is not complete either, since there are missing pieces of evidence in both yours and mine, and therefore the level of our faith/belief in our respective worldviews is equivalent.’

    It’s a lazy response, but can sometimes be an important one for someone who thinks the presumption of atheism is worth anything.

    Your response is that both the atheist and the Christian craft their worldviews using methodological naturalism; that’s pretty clearly false. The reason Christians prefer God as explaining design in the universe over a mutliverse is because we’re not methodological naturalists. We do prefer natural explanations over supernatural as long as they’re not the most ad hoc explanations conceivable, but methodological naturalism is committed to finding some kind of a natural explanation for things (this is the conclusion of the dispute over ID and whether it can be taught in schools – the evolutionists in the trial argued that science operates according to methodological naturalism, and therefore ID could not count as science, and therefore could not be taught).

    The way I think people actually operate is something closer to explanatory rationalism, or the idea that every piece of reality has an explanation. Thus, a Christian is someone who believes that God is responsible for creating the universe, designing it, ensuring that humans will come about as a part of it, and so on. So there are explanations for those things, but then there are things which are not so easily explained, such as why does God seem so absent sometimes, why is there rampant suffering all over the planet, and so on. So the Christian believes that these explanatory gaps can be filled somehow, and by trusting God has faith that there are in fact explanations for them.

    Similarly a person who is a scientific naturalists thinks the universe can be explained completely in terms of science. The multiverse explains the apparent design in it, evolution explains how we evolved, evolutionary psychology is why we think the way we do, and so on. But there are a few gaps here as well. Why is there a universe at all? How did the evolutionary process get started? How did purely physical substances operating in a deterministic universe give rise to consciousness? And so we have the same problem here all over again. The Christian is pointing out, ‘Well look, according to your worldview you don’t have explanations for everything either.’ They then use the word ‘faith’ for describing what the atheist has in his worldview, which is probably a mistake given that atheists feel they are operating solely according to reason and it is clearly false they have faith, but is basically true.

    So I don’t see where methodological naturalism is the way people go about shaping their worldviews, and I rather think there’s another quite simple way of explaining it. And it doesn’t seem that’s an argument *for* a worldview, more so just defense against seeming irrational, on the Christian’s part.

  4. Andrew said,


    As the SEP points out, “methodological naturalism” is a phrase with no fixed definition. The way I’m using it in this post is (as I explain above) the notion that empirical claims require empirical evidence. So when you say–

    …methodological naturalism is committed to finding some kind of a natural explanation for things (this is the conclusion of the dispute over ID and whether it can be taught in schools – the evolutionists in the trial argued that science operates according to methodological naturalism, and therefore ID could not count as science, and therefore could not be taught).

    –that’s simply not how I’m using the term. If that differs from how Judge Jones used it in Kitzmiller, I certainly regret any confusion.

    Viewed in the narrow sense, I stand by my contention that Christians inherently employ methodological naturalism with respect to all non-Christian claims. For example: if I tell you that leprechauns exist, you’re going to demand proof, and you’re not going to be satisfied with a book of Irish folklore.

    Thus, my use of ‘methodological naturalism’ (MN) is distinct from the question of constructing an explanation for the phenomena that we do observe; instead, I’m concerned with the level of proof necessary to substantiate a belief.

    I think — and please correct me if I’m wrong! — that you accept this narrower definition of MN and recognize that the empirical question of whether God exists requires some empirical proof. That’s why you say:

    Thus, a Christian is someone who believes that God is responsible for creating the universe, designing it, ensuring that humans will come about as a part of it, and so on.

    Now, obviously we differ as to whether those empirical observations are sufficient evidence of God; I present my counterargument to those claims on my Universe (#3) page.

    But I think we have common ground from the ‘worldview’ perspective that an empirical question (God’s existence) requires empirical evidence. That is an important prerequisite to us having a further meaningful conversation.


  5. Philip said,

    I’m not sure I’m following.

    When you say methodological naturalism is the view that, “empirical claims require empirical evidence,” how are you using the word ’empirical’? It seems awkward that you should then comment that your definition, “is distinct from the question of constructing an explanation for the phenomena that we do observe,” since repeated observation rendering a certain explanation probable is what is the word ’empirical’ normally refers to.

    While the terms may get mixed up, it seems fairly clear that people go about accepting a worldview because of its explanatory power. In that case, I still think that the Christian is merely playing a bit of defense when he points out that the atheist’s worldview (whatever it may exactly be) does not have absolute explanatory power either, and therefore is dependent on the atheist’s supplemental credence in order to be believed, and in that sense the atheist and Christian are on equal footing. I have never seen such a claim then used further as an argument *for* belief in God.

    When you say that we can agree that ‘an empirical question (God’s existence) requires empirical evidence’ I think you mean to say that we should only believe what he have empirical evidence for. I do not agree with that at all. However, I *do* think there is actually good evidence that God exists which puts anyone who thinks about the question in an evidence-based way in a position to seriously consider the possibility that God does in fact exist. But I think Plantinga has it about right that if Christianity is true, evidence is not the only way someone can be conferred a warrant for their belief in God. It is likely that if God exists, people can know and experience God directly, and that will be their warrant for belief, rather than having to amass and analyze some amount of evidence which can then afford them their warrant. Evidence can be a warrant too, but in a world in which God exists, it is not the only warrant.

  6. Andrew said,


    It seems awkward that you should then comment that your definition, “is distinct from the question of constructing an explanation for the phenomena that we do observe,” since repeated observation rendering a certain explanation probable is what is the word ‘empirical’ normally refers to.

    In the section you quote, I’m differentiating between evidence on the one hand and explanations on the other. You are correct that evidence generally leads one to formulate explanations, but MN does not require that the explanations themselves be naturalistic; that’s the point of my Zener cards example.

    As a result, my argument is not that the word “God” is empirical, but rather, that the word “exists” refers to something that both the atheist and the Christian consider to be an empirical claim (that then requires empirical evidence). Put another way: is there something (other than God) that you think exists but for which you have no evidence?

  7. Philip said,

    Oh, okay. Well I usually use the word ‘evidence’ (my it seems like we’ve been living on different planets with all these discrepancies!) to mean that which has some indicative value with respect to some claim, like God exists or whatever. Because of that, I don’t think data without an assessment of the adequacy of a theistic explanation for it compared to the adequacy of a naturalistic explanation in itself is something which constitues ‘evidence.’ It’s just data. Once it is analyzed in terms of explanation, it becomes evidence for something.

    There are some other things I believe that I can’t offer empirical evidence for, like that other minds exist at all, or that it is wrong to hurt my sister, that Laura Marling has a beautiful voice, or that the external physical world is real and not just a computer simulation. So I don’t think that my beliefs are all simply generated by empirical experimentation, if that’s what you mean to ask.

  8. Andrew said,

    Two things:

    1. The theory that “all beliefs are generated by empirical experimentation” is Lockean materialism; I don’t hold to it, either (and I don’t know of anyone who does). Obviously, one can be an atheist and still believe in innate (a priori) beliefs. You could even be a hardcore materialist and still accept those types of beliefs as being hardwired into the brain.

    2. With respect to your list, I submit that for all of the empirical claims on that list, you do indeed have empirical evidence. Let’s run through them–

    a) “Other minds”: the philosophical debate is whether we have sufficient evidence to conclude that other minds exist, but obviously we have some evidence through observation (I see other people) and induction (in my experience, people have minds). This also covers the argument for solipsism.

    b) Wrong to hurt your sister: the moral obligations part may be non-empiricial (and that’s off-topic, but obviously I’ll cover it at some point), but you have empirical evidence that your sister herself exists. It would be insane to argue that it is wrong to hurt your imaginary friend, for example.

    c) Same thing with Laura Marling. (I had to look her up on Wikipedia, because I’d never heard of her.) You can subjectively decide that her voice is beautiful, but you must first objectively have evidence that she exists and has a voice.

    You might think Laura Marling has a beautiful voice; I might hate it. But again, we can agree that she exists. That’s subjective, and it’s okay. But if you told me that “Quodroplexy Zorbleen” has a beautiful voice, you would once again be insane (because that’s an imaginary name that I just made up).

    So that’s where we are: for subjective claims of person preference, one does not need evidence — but for empirical claims, including claims of existence, we do. That’s all.

  9. Philip said,

    I’m not sure that belief in the existence of the external world or other minds are scientifically supported or not; we do infer those beliefs, and we are rational for doing so, but whether or not ’empirical’ is the word that describes the evidence for them is something I am not quite sure of.

    As far as what you’re driving at, the idea that we are all connected by a need for evidence, I’m really not so sure. One of Plantinga’s examples is a poker hand where you’re dealt the Ace of Spades as one of your cards; now that’s extremely improbable before the cards are dealt, having only a chance of 1 in 13 that you will be dealt an ace. But does that mean you shouldn’t believe you have the ace? Of course not. So I am shying away from saying that our beliefs are simply determined by whatever has the greatest quantity of evidence – for the Christian, they know God, so even if other people can’t *see* that they know God, that doesn’t really do anything to enervate the Christian’s warrant.

    The Christian’s knowing God isn’t common ground, that is true. So perhaps it won’t be the most useful thing for just philosophical conversation, although it is useful for introducing people to God, which is very hard to do if you don’t even know God yourself.

    However I don’t think it’s improbable on the evidence that God exists and that Jesus was who he claimed to be. While I don’t agree with the principle that people need loads of evidence for belief, I do think we have that evidence.

    Listening to William Lane Craig’s recent podcasat might help. I just listened to it tonight and understood some things I hadn’t before. It’s the first one under “Reasonable Faith Podcast.”


  10. Andrew said,

    1. On other minds and solipsism, my argument is not that all people would share a belief that those claims are “scientifically supported” — but simply that there is some empirical evidence for each. Thus, it is reasonable for you to claim that “Laura Marling” has a beautiful voice, but not Quodroplexy Zorbleen. Do you disagree?

    2. Plantinga’s argument isn’t responsive, either, because I’m not claiming that we reject beliefs that are below a particular probability threshold. For example: suppose you claim to have randomly dealt out a royal flush from a fair deck of cards. Unless I watched you do it, I would probably be suspicious — given the low probability — but I would not inherently reject the claim on face.

    On the other hand, if you claim to have drawn the 97 of Pinwheels from a fair deck of cards, I reject your claim outright — not because it’s of low probability or because it has less evidence supporting it — but because it’s of zero probability and is unsupported by any evidence. Would you do otherwise?

    Again, this supports my basic claim, that we share a worldview commitment to the premise that empirical claims require empirical evidence. We may disagree on how much — but we know it’s greater than zero.

    3. With that in mind, since you think we have sufficient evidence to believe in God, why don’t you share what that is?

  11. Philip said,

    I think we agree.

    That would take a long time. It’s not a removed process however – I don’t think the evidence for God is akin to something you can analyze in a lab, because if God is then he is so in our experience, meaning our experience of everything, how we think, how we see and analyze the world and our emotions and thoughts from moment to moment. So there is nothing that just in and of itself *is* the evidence for God’s existence – at the very least the features of the universe need quite a bit of discussing and the intricacies of our lives need thorough analysis for the evidence to become fully apparent. I couldn’t do it in any succinct manner – however if I will send you a link if I ever end up finishing a piece on it.

  12. Andrew said,

    Fair enough — and thank you for helping to clarify my thoughts and writings on this issue. Given your obvious intelligence and cogent writing, I’d be happy to host whatever you write on the topic as a post here at EC.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: