On the Plausibility of Naturalism and the Seriousness of the Argument From Evil (2007)
In my opening case, I defended an argument from evil based on the following evidence statement:
E: For a variety of biological and ecological reasons, organisms compete for survival, with some having an advantage in the struggle for survival over others; as a result, many organisms, including many sentient beings, never flourish because they die before maturity, many others barely survive but languish for most or all of their lives, and those that reach maturity and flourish for much of their lives usually languish in old age; in the case of human beings and some nonhuman animals as well, languishing often involves intense or prolonged suffering.
Notice that E does not, contrary to what Plantinga states in the first paragraph of his reply, claim that our living world has come to be by way of the processes (e.g., natural selection) endorsed by current evolutionary theory. No doubt evolutionary theory (when combined with metaphysical naturalism) does do an excellent job of explaining the facts E reports. But our knowledge of those facts does not in any way depend on knowing or assuming that the central tenets of evolutionary biology are true.
My argument from evil can be summarized as follows:
We know that E is true.
Naturalism has much more predictive power with respect to E than theism does (i.e., E’s truth is antecedently many times more probable given naturalism than it is given theism).
Naturalism is more plausible than theism (i.e., naturalism is more probable than theism independent of all evidence).
So, other evidence held equal, theism is very probably false.
Plantinga misconstrues my goals in formulating this argument. He seems to think that I’m an “atheologian” trying to show that all theistic belief is irrational or “unwarranted” (the result, perhaps, of malfunctioning cognitive faculties). In reality, I couldn’t be less interested in trying to show such things. My argument from evil is an argument for the probable falsity of theism, other evidence held equal, where “other evidence” means other things besides our knowledge of E that can affect the probability of theism being true. Not all false or improbable beliefs are irrational or unwarranted and not all true or probable beliefs are rational or warranted. Indeed, to the extent that I am interested in theistic belief, as opposed to just theism, the particular epistemic merit I’m interested in is the merit of being probably true as opposed to the merits of being rational or warranted. Accordingly, I disown the proposition that Plantinga calls “D” and attributes to me.
This misunderstanding of my project may explain why Plantinga ignores what I take to be the most important details of my opening case, though I would, of course, prefer an alternative explanation, namely, that he agrees with my central claims. In any case, it is important to note, for starters, that he does not offer any reason for thinking that my argument’s conclusion (4) does not follow from its premises (1) – (3). Plantinga appears to challenge my reasoning when he says that “what really follows from the premises he has proposed, is that if theism is false, then theism is subject to a prima facie defeater by virtue of the Challenge Draper specifies.” But Plantinga’s implicit point here is just that it does not follow from my premises that theistic belief is subject to a rationality defeater. Since I wasn’t trying to show that theistic belief is subject to a rationality defeater, I need not (which is not to say that I do not) disagree with Plantinga here. What I was trying to do is to establish (4)—that, other evidence held equal, theism is very probably false. And that conclusion does follow from my premises, as I suspect Plantinga would admit.
Even more importantly, Plantinga does not challenge the truth of my argument’s first two premises. In fact, he completely ignores the reasons I offer in support of the second premise, which I take to be the heart of my opening case. This is significant, because the first two premises, if true, are all that I need to justify the subconclusion that E reports strong evidence favoring naturalism over theism. (This is undeniable so long as “strong evidence favoring naturalism over theism” means “evidence that raises the ratio of the probability of naturalism to the probability of theism many-fold.”) Thus, by not challenging my first two premises, Plantinga in effect concedes that E is strong evidence favoring naturalism over theism.
Since the purpose of this e-book, as its title strongly suggests, is to examine or “debate” whether or not there is evidence for or against theism and naturalism, one might be tempted to conclude that a counterreply to Plantinga is unnecessary. That, however, would be a mistake; for although Plantinga does not challenge my argument by challenging its reasoning or its central premise, he does challenge it in two other ways. First, he challenges my third premise, which claims that naturalism is more plausible than theism. And second, he challenges the seriousness of my argument from evil—that is to say, he tries to show that the fact, if it is a fact, that the premises of my argument are all true is of little or no significance. To be more precise, he argues for the following conclusion (which I will call “Plantinga’s First Conclusion” or “PC1” for short):
PC1: if theism is true, then the truth of the premises of Draper’s argument (probably) isn’t a problem for most theists.
From PC1 he draws the further conclusion that
PC2: Draper’s argument is not a serious argument against theism (even if it is sound).
After addressing his criticisms of my third premise, I will show both that Plantinga fails to establish the truth of PC1 and that PC2 does not follow from PC1.
Plantinga on Plausibility
It is worth noting that my third premise, which states that naturalism is more plausible (i.e., more probable independent of the evidence) than theism, is overkill. Since my first two premises imply that E is strong evidence favoring naturalism over theism, my conclusion that, other evidence held equal, naturalism is many times more probable than theism and hence that theism is very probably false will follow so long as naturalism is at least as plausible as theism. Indeed, even if theism is more plausible than naturalism, so long as it is not much more plausible, it will follow that, other evidence held equal, naturalism is more probable than theism and hence that theism is probably false, though it won’t follow that theism is very probably false. I point all this out, not because I intend to back off my claim that naturalism is more plausible than theism, but because it is important to see how short the logical distance is from the first two premises of my argument to its conclusion and how very short the distance is from those premises to a more modest though still significant conclusion.
I defended my premise that naturalism is more plausible than theism by arguing that naturalism is both a simpler hypothesis than theism and no greater in scope than theism. Plantinga disagrees. Concerning scope, he says that my comments are “dialectically deficient” because most theists who have thought about it believe that theism is a necessary (i.e., noncontingent) proposition and in particular a necessary truth. Necessary propositions include logical truths and logical falsehoods like “some dogs are not dogs” (which is a logical and so necessary falsehood by virtue of having the logical form “some As are not As”), and also conceptual truths and conceptual falsehoods like “all triangles have three sides,” which is a conceptual and so necessary truth because the concept of having three sides is included in the concept of a triangle. Necessary propositions are distinguished from contingent propositions like “some dogs are brown” (which is a contingent truth) and “Al Gore is President” (which is a contingent falsehood). If theism were a necessary proposition and also true, then atheism would be, not just false, but self-contradictory, and so theism would (at least in Plantinga’s opinion) have an intrinsic probability of one—i.e., independent of all evidence, it would be absolutely certain that theism is true.
I don’t believe Plantinga is right that all necessary truths have a probability of one, but put that aside. I assumed in my opening case that theism is a contingent proposition, a proposition that is possibly true and possibly false. Is this assumption dialectically deficient? Generally speaking, in the absence of some positive reason for believing that a hypothesis is necessarily true or necessarily false, objective inquiry into whether that hypothesis is true or false should (and in fact almost always does) proceed on the assumption that the hypothesis in question is a contingent proposition. So in the absence of good reasons to believe that theism is a necessary proposition, my assumption that it is a contingent proposition is dialectically appropriate. Plantinga disagrees, on the grounds that most of the (tiny percentage of) theists who have thought about this issue have held that theism is a necessary proposition. I suspect that the sample of theists upon which Plantinga bases this judgment is not representative, but it really doesn’t matter one way or the other. Regardless of how many or what percentage of theists believe that theism is a necessary proposition, the only important question is whether there is any good reason to believe that it is a necessary proposition. Plantinga offers no reasons in his reply; so I’ll just note here that the reasons that have been offered are not, in my opinion, good ones.
Suppose, however, that my assumption is false. Suppose that theism is not a contingent proposition. Then it is much more likely that it is necessarily false than that it is necessarily true. This is made clear by any objective comparison of the available reasons for thinking that theism is necessarily true to the available reasons for thinking that it is necessarily false. The former are limited to various versions of the ontological argument, which is almost universally rejected by philosophers. Indeed, even Plantinga admits that this argument fails to prove its conclusion. The latter include a whole host of serious arguments for the incoherence of theism. Keep in mind that I’m not convinced by these arguments for the necessary falsehood of theism, but they are clearly more persuasive collectively than the notoriously unpersuasive ontological argument. Further, theism asserts that the natural world was created by an omnipotent, omniscient, and morally perfect person, which assumes, not only that there is a maximum possible degree of power, knowledge, and moral goodness, but also that these three attributes are compatible with each other and with the existence of natural entities. Even ignoring specific arguments, clearly it is much more likely that some hidden incoherence lurks in the assertion that there exists a creator of nature possessing the highest possible degree of several distinct scaling properties than in the simple assertion that no such creator exists. Therefore, if I am mistaken and theism really is a necessary proposition, then it is very probably a necessary falsehood, which means that my assumption in my opening case that it is a contingent proposition is not only dialectically appropriate (for the reasons given in the previous paragraph), but dialectically generous.
Turning now to simplicity, Plantinga challenges my account of it and offers an alternative account. My position is that the simplicity of a hypothesis depends on how much objective uniformity it attributes to the world. Plantinga worries that we can’t distinguish objective uniformity from language-relative uniformity. He mentions Goodman’s paradox, which some philosophers believe establishes that all uniformity is language-relative. Goodman (as usually interpreted) defined a new color word, “grue,” roughly as follows: an object x is grue at a time t if and only if either (i) x is green at t and t is earlier than Y2K (the year 2000) or (ii) x is blue at t and t is simultaneous with or later than Y2K. Similarly, an object is bleen when it is blue prior to Y2K or green after Y2K. Given these definitions, did emeralds change color in Y2K? Whether they did, seems to depend on what language one speaks. English speakers would claim that emeralds remained green in Y2K. But those who speak the grue-bleen language would say that they changed colors from grue to bleen. And if one objects that the definition of grue and bleen make reference to a specific time, the reply is that for those who speak the grue-bleen language, the definitions of green and blue will make reference to a specific time. The lesson, according to some philosophers, is that the claim that nature is uniform, or that uniformity has a higher prior probability than variety or change, is empty of content, since whether one finds uniformity in a situation depends on the language one uses to describe that situation. There is no such thing as “objective” uniformity.
As George Schlesinger has pointed out, this argument, though influential, is completely without merit. Surely a world in which emeralds remain green is, other things being equal, an objectively more uniform world than one in which emeralds remain grue. That is why no reasonable person, regardless of what language he or she speaks, would have predicted in 1999 that in the coming year all emeralds would remain grue. If one doubts that this preference for projecting greenness instead of grueness is grounded in a judgment of objective uniformity and not in a judgment of what seems, because of the language one speaks, to be natural, then I would suggest a simple thought experiment. Imagine someone who speaks the grue-bleen language trying, in 1999, to teach their young children their color words in the usual way, by pointing to various colored objects and saying the appropriate color-word. Then imagine their dismay on New Year’s Day in the year 2000 when their children suddenly “forget” what they previously learned and can no longer correctly identify the grue or bleen objects in the house. Gee, I wonder what went wrong.
As far as Plantinga’s own account of simplicity is concerned, there are at least two problems with equating simplicity with naturalness and using that to compare the plausibility of theism and naturalism. First, what seems natural to a person depends on cultural factors that are completely irrelevant to how likely it is that a hypothesis is true. Of course, one could try to overcome this first problem by seeing what beliefs are held across all or most cultures, but then theism won’t pass such a plausibility test any more than naturalism will, for the idea that theism (as opposed to belief in invisible agents) is a natural belief in some culturally neutral sense is not supported by any serious study of the history of religion (as some of the authors Plantinga cites in support of his position are careful to point out). A second problem is that whether a hypothesis seems natural to a person may very well depend on the conceptual or mathematical or philosophical or other intellectual skills or background possessed by that person. Hypotheses that require any significant degree of such skills or background in order to be expressed or fully understood won’t of course seem natural to most people. But surely it doesn’t follow that such hypotheses are all implausible. Naturalism is just such a hypothesis (as is the sort of theism defended by Plantinga). Belief in entities that philosophers might classify as supernatural is common in many cultures, but those cultures did not distinguish the natural from the supernatural in the way philosophers do; so it is misleading to say even that belief in the supernatural is natural, let alone belief in a theistic God. Most people who have the conceptual framework and intellectual skills needed to really understand metaphysical naturalism hardly find it “unnatural.” Many accept it and even some who reject it find it difficult to resist the simplicity of denying the existence of all supernatural entities without exception as opposed to denying the existence of all supernatural entities with the sole exception of God and any supernatural entities created by God. Thus, Plantinga’s suggestion that the scarcity of naturalists proves the unnaturalness and hence implausibility of naturalism, just isn’t credible.
Plantinga’s Main Point
Let us now turn to the heart of Plantinga’s reply to my opening case. Plantinga says that “a belief P is subject to a Draper Challenge when there is a proposition Q incompatible with P that is simpler and has less scope than P, and a true proposition R that is much more probable with respect to Q than with respect to P.” He proposes to show that a Draper Challenge (“Challenge” for short) is not a “problem” for theists “because . . . very many, indeed most of the propositions P we believe are subject to Draper Challenges.” His point, of course, is that we still believe them and reasonably so in spite of their being subject to Challenges. Plantinga concludes that my argument is not a “serious” argument from evil.
The first half of this compound argument (and hence the entire second section of Plantinga’s reply) turns out to be something of a red herring for two reasons. First, although Plantinga says that he will argue that most of our beliefs are subject to Challenges, he ends up giving only three imaginary examples followed by the appropriately hesitant comment that “My guess is that most of the propositions we believe are challenged” (italics are mine). Second, and more importantly, it’s no mystery why a rational person might have a belief, learn that it is subject to a Challenge, and yet still reasonably retain that belief. Strong evidence favoring one hypothesis H1 over another H2 can be outweighed by other even stronger evidence favoring H2 over H1. Indeed, even if H1 is also more plausible than H2, the other evidence favoring H2 might be so strong that H2 would be, all things considered, more probable than H1. This is why my conclusion includes the ceteris paribus clause “other evidence held equal.” Notice too that this other evidence might be inferential or noninferential, propositional or experiential. Still, while all this is fairly obvious, it is equally obvious that in many cases, including cases in which H2 is not a scientific hypothesis or anything like a scientific hypothesis, there is no outweighing evidence on the other side and so H2 is probably false all things considered. So the important question is not how many or what percentage of our (reasonably held) beliefs are subject to Challenges, nor is it how often there is outweighing evidence in such cases. What matters is whether there is outweighing evidence in this particular case. Plantinga does not try to show that there is, and I cannot be expected to examine all of the alleged relevant evidence in a single paper; hence the need for my ceteris paribus clause and the need to divide this e-book into different sections on different areas of evidence.
I conclude that, contrary to what Plantinga suggests in the second paragraph of his reply, the fact (if it is a fact) that most of our beliefs are subject to Challenges fails to support his claim that PC1 is true. PC1, recall, is the claim that, if theism is true, then the fact that the premises of my argument are true is probably not a problem for most theists. Of course, if my opening case had consisted solely of my pointing out that theistic belief is subject to some Challenge or other, then the (alleged) fact that most of our beliefs are subject to Challenges would support PC1. But that’s not what I did. I gave a specific Challenge, the significance of which depends partly on its specific features. Moreover, if one were to ask why, in each of Plantinga’s three examples, the Challenge in question creates no intellectual problem for the “believer,” the answer is clear. In each of those cases, it is obvious that there is outweighing evidence favoring the belief in question over the relevant alternative hypothesis. In the case of my Challenge to theism, the existence of such outweighing (inferential or noninferential) evidence is far from obvious. Hence the problem.
Since, however, it isn’t always obvious what’s obvious, let us suppose, for the sake of argument, that PC1 is true. Plantinga infers from PC1 that PC2 is true, that my argument is not a serious argument from evil. Plantinga makes it clear that he wants to draw this further conclusion when he says that, “To produce ‘a serious argument from evil against theism . . .,’ Draper would first have to show that theism is false.” I will close by showing that Plantinga’s inference here is incorrect: PC2 does not follow from PC1. The reason it does not follow is that there are very many people who, like me, don’t believe they already know that God exists (or that God doesn’t exist), and for that reason believe that it is appropriate and important to engage, not in apologetics, but in genuine inquiry designed to determine, to the best of their ability, whether or not God exists. Included here are agnostics as well as theists and atheists who have doubts about God’s existence or nonexistence. These skeptical souls have no choice but to do their best to objectively assess the available evidence. Thus, for them, the fact that E is strong evidence favoring naturalism over theism, which my argument demonstrates, is of great significance. For them, my argument from evil is very serious indeed. Notice also that, for them, indeed for anyone who doubts that God exists, it doesn’t help to be told by Plantinga that, if God does exist, then some theists probably know he does!
 No doubt there are some interesting connections between these three ways in which a belief can have merit, but that doesn’t undermine my point here.
 When I said that God, if he existed, would be a necessary cause of every natural event, I meant that God, if he existed, would at least be a necessary if not a sufficient cause of every natural event. I did not mean that God would be a cause that exists necessarily!
 For a fairly comprehensive collection of such arguments, see Michael Martin and Ricki Monnier, eds., The Impossibility of God (Amherst, NY: Prometheus Books, 2003).
 I assume here that the optimal degree of power, knowledge, and moral goodness would, if there is one, be the maximum degree.
 A “scaling” property is a property that comes in degrees—an object can have more or less of it. For example, beauty is a scaling property while the property of being pregnant is not.
 In The Sweep of Probability (Notre Dame, IN: University of Notre Dame Press, 1991), chap. 4, section 6.
 In April of 2006, Nicholas Wolterstorff delivered a paper at an APA symposium in which he carefully distinguished the epistemic merit of probability from the merits of warrant and rationality. Reading that paper helped me to write this one, as did e-mail discussion with Michael Bergmann and Michael Rea.
Copyright ©2007 Paul Draper. The electronic version is copyright ©2007 by Internet Infidels, Inc. with the written permission of Paul Draper. All rights reserved.