A Response to Jeffery Lowder’s Case for Naturalism

Jeffery Jay Lowder, a prominent and highly respected philosophical advocate for naturalism and atheism, has recently posted the written transcript of the opening statement of his 2016 debate with Frank Turek (popular Christian apologist). I highly appreciate Lowder’s work (he has made me a better philosopher) and so I thought I’d take a post to respond to the general case for naturalism he presents in the opening statement.

A couple of notes before I begin: I haven’t watched the whole debate, so I’m not exactly sure how Turek responds to Lowder’s case. Some of the points I make here may have already been made by Turek, and Lowder may have already responded to them. If so, I apologize. I’m simply giving my own thoughts on just the opening statement.

Second: Lowder is a proponent of a bayesian approach to philosophy of religion and arguments for/against the existence of God. This approach utilizes bayesian probabilistic/evidentialist arguments to analyse which hypothesis better fits the piece of data in question. I’m personally not a big fan of the bayesian approach, and don’t really use it in any of my own work. I take more of a traditional/classical metaphysical approach. In other words, I think the existence of God can be metaphysically demonstrated without the use of probabilistic arguments. I favor this method for a number of reasons that I might go into in a future post, but won’t discuss here. For the sake of this post, I’m going to respond to Lowder on his own terms, using the bayesian method.

The overarching question which Lowder and Turek debated was which better explains reality: naturalism or theism. As I’ll try to show below, this question was probably not the best way to frame the debate. But in any case, it was the question given, and Lowder appropriatley begins with a definition of terms.

Following philosopher Paul Draper, Lowder defines naturalism as the hypothesis that “the physical exists and, if the mental exists, the physical explains why the mental exists” [1]. A work of more extended scope would probably need to give some further explanation as to what is meant by terms like physical and mental. It would also need to give some sort of explanation as to why this definition of naturalism is used, since defining naturalism is a notoriously controversial enterprise. More could be said here, but for now we’ll just accept Lowder’s definition as is and work with it.

Next, Lowder defines supernaturalism as “the view that the mental exists and, if the physical exists, the mental explains why the physical exists” [2]. Supernaturalism is thus the direct contrary to naturalism. As Lowder states, naturalism and supernaturalism “are symmetrical claims . . . [both] are equally modest and equally coherent” [3]. As such, both claims have an equivalent prior probability (the likelihood of their being true prior to/before examining any evidence).

Third, the term personal supernaturalism is described: “‘personal supernaturalism’ is a type of supernaturalism; it adds on the claims that one or more personal mental entities exist and, if a physical world exists, it or they produced the physical world for a purpose” [4]. Personal supernaturalism is a species of the genus supernaturalism.

And finally: “‘theism’ is a type of personal supernaturalism; it adds on the claim that there is just one mental entity, God, who is all-powerful, all-knowing, and morally perfect” [5]. Theism is a species of personal supernaturalism which is itself a species of supernaturalism. This is significant, because we see already that theism and naturalism are not symmetrical claims, which has an effect when we start weighing probabilities.

Lowder’s overall case in favor of naturalism over theism as the best explanation of reality goes like this:

  1. The best explanation is the explanation with the overall greatest balance of intrinsic probability and accuracy;
  2. Naturalism is an intrinsically more probable explanation than theism; and
  3. Naturalism is a more accurate explanation than theism [6].

As we mentioned earlier, intrinsic probability is the likelihood of a hypothesis being true prior to any examination of evidence. Lowder suggests, again following Paul Draper, that intrinsic probability “is determined entirely by its modesty and coherence” [7]. This is, I think, a relatively controversial claim amongst those who use the bayesian method in philosophy of religion; and Lowder doesn’t give a defense of it in the course of his presentation. But this point isn’t too relative to my own response, so I won’t say much else about it here.

Whereas prior or intrinsic probability examines the likelihood of a hypothesis independent of evidence, accuracy is a function of “the degree to which a hypothesis’s predictions correspond to reality” [8]. The accuracy of a hypothesis is determined by its relation to extrinsic evidence (Lowder gives a good overview of “evidence” in his statement; I won’t repeat it here but encourage the reader to take a loot at it for themselves).

So Lowder’s first contention is simply laying out the criteria for a “best explanation” given his bayesian framework. The next two contentions are that naturalism meets these criteria over theism; and hence that naturalism is a better explanation than theism.

Why does Lowder think naturalism is intrinsically more probable than theism? Because theism is a “less modest” hypothesis than naturalism, i.e. it claims more than naturalism does. This follows from the observation we made earlier: symmetrical claims are equally modest. Naturalism and supernaturalism are symmetrical, while theism is a subspecies of supernaturalism and hence is asymmetrical with naturalism.

Of course, prior probabilities can be overcome by a sufficient weight of evidence. But Lowder’s next contention is that here too naturalism supercedes theism. Here he gives seven lines of evidence which he argues support the accuracy of naturalism over theism:

  1. Existence of physical reality: naturalism entails the existence of physical reality in its definition; theism does not. While the existence of physical reality given theism is possible, it isn’t necessary; and hence theism doesn’t predict it in the way that naturalism does. Thus, the existence of physical reality favors naturalism over theism
  2. Success of the natural sciences: naturalism entails that supernatural explanations should be false while natural explanations should be true. And the history of the natural sciences shows exactly this: natural explanations are much more plentiful than supernatural explanations, and wherever there have been supernatural explanations, they are often replaced with natural explanations over time. Theism doesn’t entail this like naturalism does (since, on theism, supernatural explanations are entirely possible); and hence this data provides evidence for naturalism over theism.
  3. Biological evolution: on the supposition that life exists, on naturalism biological evolution is pretty much the only way life could have developed. But on theism, God could have used other ways of bringing about life, such as special creation. Hence, on the supposition that life exists, the fact that it developed via biological evolution favors naturalism over theism.
  4. Biological role of pain and pleasure: On naturalism, given the supposition that life exists, and since on that supposition life must have developed via evolution, the biological roles of pain and pleasure are to be expected. But given theism, we would expect pain and pleasure to play primarily moral rather than biological roles. Hence, the fact that pain and pleasure play primarily biological roles is evidence for naturalism over theism.
  5. Mind-brain dependence: On naturalism, given that life and specifically intelligent life exists, such intelligence would have to develop from and be dependent upon the physical. On theism, however, this is not the case; since God could create intelligent creatures not-dependent upon the physical. Hence, the fact that our intelligence is dependent in some sense on the physical brain favors naturalism over theism.
  6. Empathy and apathy: moral decision making is significantly influenced by emotional capacities such as empathy. However, certain neurological defects can result in a lessened capacity for empathy and hence for impediments to proper moral decision making. This is to be expected on naturalism, given that life and moral agents must have developed via biological evolution; but is surprising on theism.
  7. Nonresistant nonbelief: There are some people open to having a relationship with God who nonetheless find themselves simply unable to believe in God. This again makes sense given naturalism, since natural selection doesn’t care whether one beleives in God or not as long as they survive and reproduce; but is surprising given theism, since God could make it such that everyone open to believe in Him is able to actually believe in Him. [9].

These are simple summaries of the seven lines of evidence which Lowder defends in more depth. The overall conclusion is that the evidence favors naturalism over theism. And this, in conjunction with the earlier contention that naturalism is intrinsically more probable than theism, leads to the conclusion that naturalism is a better explanation for reality than theism.

So, how ought a theist to respond? The first issue is with the very structure of the debate. The question of the debate begins by presenting us with two options to choose from: naturalism or theism. But why start with these two? On the bayesian approach, a case needs to be cumulative and progressive. Further, the outcome of the debate may tell us which of the two proposed hypotheses is better, but it surely won’t tell us which, if either, is the actual best of all possible explanations. There are a plethora of other competing hypotheses which could also be examined. Again, these are problems with the structure of the debate itself and not Lowder’s case specifically.

But these issues do have some bearing on Lowder’s case. Given the definitions used, Lowder is correct that theism has a lower intrinsic probability than naturalism. But that’s simply because naturalism and theism are asymmetrical claims, as we’ve seen. And it doesn’t really make sense to start a bayesian examination with two asymmetrical hypotheses. Instead, we should begin our examination by comparing naturalism and supernaturalism, since these are symmetrical and hence have equivalent prior probabilities. With equivalent prior probabilities, the sole determining factor will be extrinsic evidence.

So which, between these two, does the evidence support? Well, here Lowder still might point to his argument from physical existence. Naturalism entails physical existence while supernaturalism does not; hence physical existence favors naturalism over supernaturalism. But here I would respond with my own argument: the contingency of the physical world much more highly favors supernaturalism over naturalism. Since supernaturalism as defined entails that anything physical depends upon the mental, it requires that if there is a physical world, it will be contingent in nature. Naturalism on the other hand does not entail the contingency of the physical world. In fact, one could argue that the contingency of the physical world is not only highly improbable on naturalism, but actually inconsistent with naturalism. Hence, I think we have very good reason to accept supernaturalism over naturalism.

Once we’ve selected supernaturalism over naturalism, then we can turn to examining specific kinds of supernaturalism. Here I’m just going to give a brief glance at some general lines of reasoning that one could possibly develop here. Personal supernaturalism would seem to be favored over non-personal supernaturalisms given that we know personal mental agents (human beings) do in fact exist. Next we would need to examine the subclasses of personal supernaturalism. Again, one could argue that the contingent nature of the physical world supports the existence of some omnipotent supernatural entity which creates and sustains. The order in the universe might support a specifically intelligent supernatural agent. Monotheism is certainly more initially plausible than polytheism, insofar as it posits fewer beings. It might be harder to select between theism and deism, but perhaps one could develop an argument from the widespread prevalence of religion as evidence of a God interested in relationship with human creatures. And so on.

Once again, this is just a brief, general run through of some kinds of arguments that could be developed. I’m not actually developing or defending them here, because my main point is just to show 1) that naturalism and theism shouldn’t really be being compared in the first place, since they are asymmetrical claims; 2) when supernaturalism and naturalism are compared, there are good reasons to choose the former over the latter; 3) once we’ve selected supernaturalism over naturalism, then we can further narrow down our inquiry to the specific kinds of supernaturalism. Again, the bayesian approach needs to be cumulative and progressive. We should start out by weighing more similar claims against each other, and then progress to more specific claims. If we are going to compare naturalism and theism, then we’ll need to somehow factor in the relevant probabilistic information given to us by the prior conclusions, i.e. that supernaturalism is more probable than naturalism, that personal supernaturalism is the most probable kind of supernaturalism, and that theism is the most probable kind of personal supernaturalism. On its own theism might be less intrinsically probable than naturalism; but when the relevant information is factored in, things shift. If we can show that supernaturalism itself is much more probable than naturalism, then surely the most probable kind of supernaturalism is thereby more probable than naturalism.

Much more could be said, and I might do so in further posts. But for now this much suffices as an initial response to Lowder’s well argued case for naturalism.


[1]. From the written transcript published on Lowder’s blog: https://www.patheos.com/blogs/secularoutpost/2018/12/29/opening-statement-from-my-debate-with-frank-turek/

[2]. Ibid.

[3]. Ibid.

[4]. Ibid.

[5]. Ibid

[6]. Ibid

[7]. Ibid

[8]. Ibid

[9]. Ibid

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 )

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