This book is dedicated to George Mavrodes and is a collection of eleven essays on the question of whether religious belief is rational only if it is based on evidence. Evidentialists claim that it is; reformed epistemologists claim that it isn’t. The book is divided into three parts. The essays in the first part examine the demand for evidence, the essays in the second part examine the relation of beliefs to evidence, and the essays in the third part examine evidence and religious belief. The book is clearly aimed at making a case for the rationality of religious (read: theistic) belief. The various contributions, with the exception of William Rowe’s, William Hasker’s, and E. J. Coffman and Jeff Cervantez’s, argue that it is rational even if there is no evidence for it, that there is in fact evidence for it, that there are pragmatic reasons for adopting it, or that certain traditional arguments against it fail. Overall, however, the contributions are of low calibre, with instances of shoddily argued—indeed, very shoddily argued—chapters, one upshot of which is that the book fails to make a case for the rationality of religious belief.
Theists go to considerable lengths to make their religious beliefs seem less bizarre than they really are, and one of their favourite strategies when pursuing this end is to compare them to other, more mundane, beliefs. James Ross does this in a notably weak piece entitled “Willing Belief and Rational Faith” (chapter 1). He argues against the view that rational beliefs need to be evidence-based, contending that “one’s life is organized by local reliances, by faith in clocks, in trains, in phone books, in Fords, in doctors, in the government to pay its debts, in the justice system, in democracy, in capitalism, in modern medicine” (p. 19). Thus, just as it’s rational to have a faith-based belief that, say, one’s clock is working, so it is rational to have a faith-based belief in, say, the god of Christianity.
I am reluctant to ascribe to Ross an argument as obviously careless as this one (a variant of the hackneyed ‘atheists-have-faith-too’ argument), but I am unable to discern a better one in his article. The fatal defect in the argument is that it equivocates on ‘faith.’ It is true that I have faith in clocks, Fords, modern medicine, etc. if ‘faith’ means confidence or trust, but I do not have faith in these things if ‘faith’ means believing without evidence. But, of course, it’s this latter sense of faith that Ross needs in order for his argument to succeed, since the argument is intended to refute evidentialism. But, on this latter sense of the word, the argument fails utterly. It’s just false that I believe without evidence that, say, my clock is working—I know that my clock is working. And it’s just false that I believe without evidence that, say, modern medicine can cure various illnesses that were once incurable. On the contrary, I have reams of evidence for it. And finally, it’s just false that by relying on clocks, Fords, modern medicine, etc. I have faith in them. What about justified (i.e., evidence-based) belief? Inexplicably, Ross doesn’t so much as mention this crucial epistemic category.
Cognate with these difficulties is the fact that the central example Ross uses to bolster his thesis that our ordinary belief systems are faith-based—viz., induction—doesn’t do the work he needs it to do. Ross claims that induction, a guide by which we live, is rooted in cognitive reliance that is not itself based on evidence, but is used because it gets us what we want. Similarly for theistic belief. And since we do not think it is irrational to rely on induction, so we should not think it is irrational to ‘rely’ on theistic belief. But there is a very clear problem here. Induction, unlike religious faith, is epistemically indispensable: we have to rely on induction—we wouldn’t learn anything if we didn’t (just ask Lucy of Charlie Brown fame)—but nobody has to rely on theistic belief. And this, of course, is because nobody has to believe in an invisible deity (desperately wanting to doesn’t count), just as nobody has to believe in fairies or ghosts.
The traditional consensus gentium argument is easily one of the worst arguments for theism ever promulgated—if 50 million people believe a foolish thing, it is still a foolish thing, to paraphrase Anatole France—but in chapter 2, in an article entitled “Epistemic Self-Trust and Consensus Gentium Argument,” Linda Zagzebski puts a new spin on the argument. She argues that, epistemically, we consistently trust ourselves, i.e., we trust that most of our beliefs are true. And since we can assume that we aren’t peculiar in having mostly true beliefs, we should trust that others have mostly true beliefs. But most people believe in gods of one kind or another, so we should trust that this belief is true. The same argument would work for atheism; only, as it happens, most people are theists. Zagzebski scarcely defends the argument and points out that it can be defeated by identifying “a property of the believer that makes a belief formed out of that property untrustworthy, and that many theists in history have that property” (p. 35). More on this shortly.
In chapter 8, in an article entitled “Consensus Gentium: Reflections on the ‘Common Consent’ Argument for the Existence of God,” Thomas Kelly examines the consensus gentium argument as well, arguing that it should be formulated as follows: (1) A strong supermajority (sixty-plus percent) of the world’s population believes that God exists; (2) therefore, probably, God exists. Kelly discusses some common strategies for minimizing the significance of the premise—e.g., that it’s false, or that the strong supermajority was not produced by independent convergence—but finds them wanting. He concludes that we should not abandon the premise, and that “an adequate treatment of these issues would inevitably require the resolution of quite subtle questions in the theories of reference and belief attribution. If that is correct, then fully getting to the bottom of the common consent argument might very well require further inquiry, not only in epistemology, but also in the philosophies of language and mind” (p. 156).
This is a bogus conclusion—or, at any rate, it doesn’t follow from anything Kelly says in his article. Kelly, like Zagzebski, fails to recognize that the kind of belief in question is relevant to whether the consensus gentium argument is worth taking seriously. Indeed, like Zagzebski, he portrays belief in God as being like any other belief (e.g., p. 151), which it isn’t. People have any number of powerful motivations for believing in God (e.g., fear of annihilation, hope for an eschatological payoff, self-adulation the maintenance of which is achieved by believing that one is the object of a divine being’s affections, etc.) that they don’t have for believing that, say, there are fish in the sea, such that even if this latter belief’s being accepted by a supermajority is evidence that it is probably true—since patently there are reasons to accept it and presumably a supermajority accepts it for those reasons—it does not follow that the former belief’s being accepted by a supermajority is evidence that it is probably true. Beliefs for which there are motives and no reasons, like the belief in God (or that 9/11 was an ‘inside job,’ or that Shakespeare didn’t write all those plays and sonnets), may be true, but in the absence of reasons for a belief (where reasons are logical, empirical items), there is, well, no reason to think that the belief is true, and hence no reason to think that the belief’s being accepted by a supermajority is evidence that it is probably true. For my part, I put no stock whatever in the premise. A supermajority, I would guess, of the world’s population believes that the theory of evolution is false—the motivations for this are, as it happens, almost always religious. This is hardly evidence that the theory of evolution is false.
In chapter 3, in an article entitled “Religious Experience and the Question of Whether Belief in God Requires Evidence,” C. Stephen Evans, drawing on the work of Mavrodes (in particular, Mavrodes’s Belief in God), argues for a partial rapprochement between evidentialists and reformed epistemologists. According to Mavrodes, experiences can be mediated, and Evans argues that:
…if we allow for the possibility of mediated experiences of some entity, it seems quite possible for there to be experiences of God, if God is real. For…if God exists and has the properties traditionally ascribed to God, such as being creator and sustainer of the universe, then God is the ultimate source of whatever energy streams come to us from any object in the created universe….Thus, to use the same examples that [Alvin] Plantinga cites as ‘grounds’ for belief in God, a person who contemplates a flower or a person who is looking at the ‘starry heavens’ might indeed be experiencing God through the experience of the flower or the heavens. (p. 43)
This kind of experience, on Evans’s view, can be grounds for belief in God, and in “cases where the grounds of belief in God consist in or at least include a mediated experience of God, there does seem to be a clear sense in which the ground could be described as evidence of God’s reality” (p. 45). Thus, we have a partial rapprochement between evidentialists and reformed epistemologists.
I commend Evans for attempting to curtail the excesses of reformed epistemology (to put a charitable spin on what he is up to here), but his argument is underwhelming, since bare possibilities are a dime a dozen, and even if (contrary, as I see it, to fact) it is possible that a person who contemplates a flower is experiencing God through her experience of the flower, this possibility need not be taken seriously. Presumably Evans would dismiss as unserious the suggestion that a person who stands aghast at film footage of a pile of bodies outside the crematorium at Dachau is experiencing a perfectly evil deity through her visual experience of the footage, so why should we take seriously his claim that it is possible that a person who contemplates a flower is experiencing a perfectly benevolent deity through her experience of the flower? In short, the evidence, if one can call it that, that Evans uses in his attempt to reconcile evidentialism and reformed epistemology is crummy evidence. But if the evidence Evans uses for this purpose is crummy, then it cannot be grounds for belief in God (even if the experiencer thinks it is).
In chapter 4, in an article entitled “Phenomenal Conservatism and Evidentialism in Religious Epistemology,” Chris Tucker defends two theses: (1) that phenomenal conservatism (which says that if it seems to me that P, then I have evidence for P) makes evidence so easy to obtain that the truth of evidentialism is not an obstacle to justified religious belief, and (2) that phenomenal conservatism is better suited than proper functionalism (which says that a mental state, M, evidentially supports P just in case it is proper function for the subject of M to believe P on the basis of M) “to explain how an appreciation of the glories of nature leads to justified religious beliefs” (pp. 72–3). (We are not told whether the fact that nature is replete with disease, starvation, predation, fratricide, and just a lot of murder and suffering in general is one of its glories, or whether we are supposed to appreciate the fact that life did not arise on our planet until it was billions of years old, only for 99% of it to be obliterated in a catastrophe billions of years later.) Tucker’s argument fails, however, because it assumes, falsely, that appreciating the glories of nature can justify theistic belief. The idea is that, on the basis of how things seem to us (e.g., perceptually), we can make reasonable existence claims or claims about how the world is. So, for example, on the basis of my friend’s aberrant behaviour, it may seem to me that she is depressed, and this seeming is evidence for (and hence justifies) my belief that she is depressed. Similarly, on the basis of one’s appreciation of nature, it may seem to one that God exists, and this seeming is evidence for one’s belief that God exists. But this is confused, for the simple reason that it cannot seem to one, in the relevant sense, that God exists (any more than it can seem to one, in the relevant sense, that DNA exists). It may seem to me, in the relevant sense, that my friend is depressed, but I am justified in holding that things seem this way to me only if I am acquainted with some of the outward signs of depression—indeed, my claim that it seems to me that my friend is depressed would lack sense if I was not acquainted with some of these signs (assuming, of course, that I lack testimonial evidence that she is depressed). But there are no outward signs of a bodiless person who necessarily is omniscient, omnipotent, morally perfect, etc. More generally, it can legitimately seem to one, in the relevant sense, that X exists only if one is acquainted, in some way, with X or its outward signs. But, of course, nobody is acquainted with God or his ‘outward signs.’ The person who claims, on the basis of her appreciation of nature, that it seems to her that God exists is simply placing an unwarranted religious interpretation on her appreciation of nature.Footnote 2
In chapter 5, in a piece entitled “Theistic Proofs, Person Relativity, and the Rationality of Religious Belief,” William J. Wainwright disputes the view that since Christian theism, unlike agnosticism, exceeds what the evidence warrants, the former is less reasonable than the latter. It is unclear, however, how Wainwright arrives at the conclusion that this view is false. His claims about the person relativity of arguments—e.g., that “assessments of interesting…philosophical arguments…invariably reflect our personal histories [and our] passional nature” (p. 87); that “an argument’s cogency and convincingness can depend on the state of one’s heart, and the states of people’s hearts vary” (87–8); and so on—don’t support it. Yes, an argument for the existence of Xenu, the god of Scientology, may be cogent and convincing to Tom Cruise and uncogent and unconvincing to people who haven’t been duped by an inane cult, but nothing of interest follows from this. Certainly it doesn’t follow that arguments for the existence of Xenu should be taken seriously, or that believing that Xenu exists is epistemically on par with believing that Xenu does not exist. Xenu agnosticism—indeed, Xenu non-belief—is vastly more reasonable than belief in Xenu even if there are people who accept arguments for the existence of Xenu because they feel like it. The pandering to sentiment found in this chapter simply underscores how altogether intellectually impoverished the ‘case’ for theism really is.
But, too, a pivotal step in Wainwright’s argument is a non sequitur. He writes:
Suppose that the Christian God exists but that the evidence for his existence is no stronger than that against it. (Many Christians believe this. Kierkegaard, as sometimes interpreted, is an example.) If this is correct, then, while agnostics may be more rational than Christians, a Christian faith which believes in the absence of adequate evidence is better than agnosticism. (p. 88)
But this follows only if we assume that God, if he exists, does not object to believing (in him, say) without adequate evidence, which is something that neither Wainwright nor anyone else knows or can assume (at least not without begging the question). Perhaps God, if he exists, abhors believing without evidence, such that agnosticism is both better (in terms of divine rewards) and more rational than believing in the absence of adequate evidence.
In chapter 6, in a piece entitled “Hiddenness, Evidence, and Idolatry,” E. J. Coffman and Jeff Cervantez raise some worries about Paul Moser’s reply to the Hiddenness Argument against the existence of God (an argument to the effect that if God existed, every cognitively normal adult human being would have evidence of her existence, but since it is not the case that every cognitively normal adult human being has such evidence, God doesn’t exist). Philosophers interested in the Hiddenness Argument will no doubt find something of value here.
In chapter 7, in an article entitled “An Evolutionary Objection to the Argument from Evil,” Thomas M. Crisp launches an unsuccessful objection to the evidential argument from evil. Let RR be the proposition that our cognitive faculties are reliable with respect to recondite philosophical matters. Crisp’s argument is that the evidential argument from evil is a recondite philosophical argument, and if our cognitive faculties are the result of unguided evolution—Crisp discusses the possibility that our cognitive faculties are the result of guided evolution, but I will disregard that discussion here—then the (epistemic) probability that RR is true is low or inscrutable. But if the probability that RR is true is low or inscrutable, then there is a defeater for the following key premises in various versions of the evidential argument: (1) probably, no reason would justify God’s permitting the amount suffering found in our world, and (2) the probability, given that God exists, that there should be that amount of suffering is much lower than that probability given naturalism.
The difficulty is that Crisp scarcely defends the argument’s central premise, viz., that the probability that RR is true is low or inscrutable if our cognitive faculties are the product of unguided evolution. What we get, in lieu of either evidence or argument for this hypothesis—and without so much as a nod, even in a footnote, to the extensive and ever-growing scientific literature on the topicFootnote 1—is a Kirk Cameron impression, with Crisp making bold scientific pronouncements from the armchair:
Is there some reason, given that atheism holds and that we and our cognitive faculties arose via unguided evolutionary processes of the sort described by contemporary evolutionary theory, to expect those processes to have endowed us with much by way of reliable insight into recondite philosophical matters—matters quite beyond the exigencies of normal human life? Surely the answer here is ‘no.’ Given unguided [evolution], the main explanation why our evolutionary ancestors evolved cognitive faculties of the sort they did was that those faculties were adaptive: useful for feeding, flying, fighting, reproducing and so forth. But why would cognitive faculties selected for their success at those tasks have required reliability with respect to abstruse philosophical matters, matters quite unconnected to the concerns of everyday life? (pp. 116–17)
Here Crisp simply trots out the hoary bromide that it is improbable that unguided evolution could have endowed us with faculties conducive to success in anything other than avoiding predators and finding food or potential mates. But how does he know that? What are his grounds for such an extraordinarily narrow view of the evolution of human cognition? Human beings are highly social animals with big brains. Where is Crisp’s evidence that it is improbable that societies composed of such creatures would come to value, through unguided evolution, abstract thought? There is no such evidence, of course, at least none that’s adducible from the armchair. Accurately calculating probabilities involving deep time and a planet-sized petri dish is extremely difficult, and the estimation of such probabilities is not within the purview of a priori philosophical thought. Crisp’s argument, therefore, since it depends on an eminently contestable (and, to my amateur eye, false) scientific assumption for which he provides no support, is a non-starter.
In chapter 9, in a piece entitled “Morality and Happiness,” Kelly James Clark and Andrew Samuel argue that believing in an afterlife is a necessary condition for being properly morally motivated. “There must be a next life,” they argue, “with other-worldly goods, if justice is to prevail. It should motivate us because we will believe that our best but invariably puny efforts to flourish will not be in vain. Deprive us of that hope, and we will believe that since the moral struggle cannot be won it is not worth fighting” (p. 172).
This argument is uncompelling for at least three reasons. First, it relies on the dubious assumption that “a rational being…is a self-interest maximizer” (p. 158). This assumption has no demand on our acceptance—in fact, it’s false: it is not the case that, necessarily, the person who performs an action that she knows will not maximize her self-interest does something irrational—so anyone not antecedently friendly to the assumption can reject Clark and Samuel’s argument outright. Second, ‘ought’ implies ‘can,’ so those not antecedently friendly to the notion of an afterlife will be unmoved, since they will balk at the notion that they should be motivated by something—viz., an afterlife—that they cannot believe exists. And third, it’s false that there must be an afterlife for justice to prevail. If Jones commits a crime and is punished for it (or rehabilitated, or whatever), justice has prevailed. Now, of course, justice doesn’t always prevail, but those who value truth over wishful thinking, intellectual honesty over comfort, will not let their yearning for ‘cosmic’ justice trump their reasoned belief that there is no afterlife.
In chapter 10, in an article entitled “Divine Perfection and Freedom,” William Rowe presents an atheological argument to the effect that a being who is essentially morally perfect and essentially free in many of his actions, as it is alleged God is, cannot exist, since moral perfection and divine freedom are incompatible. The argument is clever and, to my mind, sound.
In chapter 11, in a piece entitled “The Many Gods of Hick and Mavrodes,” William Hasker discusses the question of whether Mavrodes’s characterization of John Hick as a polytheist is accurate. This chapter will probably interest specialists; it bored me, as I expect it will other non-specialists, to the point of distraction.
There is nothing in this book that will dissuade the discerning nontheist that the ‘evidence’ for theistic belief, when theists bother producing it, is typically comically bad evidence that would be roundly dismissed in any non-religious context. (Imagine a scientist claiming there are multiple universes solely on the basis of its seeming to her that there are.) There is also nothing here that will dissuade the discerning nontheist that just as there is no surrogate for evidence-based argument when it comes to the question of whether there are fairies, ghosts, goblins, and the like, so there is no surrogate for evidence-based argument when it comes to the question of whether there are gods. Finally, and relatedly, there is nothing here that will dissuade the discerning nontheist that reformed epistemology is a disgrace. As Brian Leiter has recently pointed out, it “is nothing more than an effort to insulate religious faith from ordinary standards of reasons and evidence in common sense and the sciences” (2013, p. 81).
