Alvin Plantinga has argued for a view in religious epistemology that has become known as reformed epistemology. One of the central claims made by Plantinga is that there is no good reason to think that religious belief, and belief in God in particular, is not in fact rational even if the believer can provide no evidence or supporting arguments for her belief.
Plantinga argues that it is reasonable for the theist to take belief in God as a basic belief (that is, a belief that is not based on any other beliefs), and furthermore, that belief is warranted-if-true. Plantinga defines warrant as whatever it is that makes the difference between true belief and knowledge, so according to Plantinga, if God does in fact exist then that belief very likely has warrant and amounts to knowledge. This is because what is warranted for us depends upon the sorts of beings we are and what we are capable of, and this is likely to depend on whether or not God exists. According to the Christian account, human beings have been created by God with the intention that we can know and love him. Given that, it is natural to think that God has designed us in such a way that when our cognitive faculties are functioning properly, they will, in the appropriate circumstances, produce warranted beliefs about God. For this reason, claims Plantinga, we cannot assess the warrant or rationality of Christian belief independently of assessing the truth of those beliefs.
This account has attracted a certain sort of criticism (or group of criticisms) known as the ‘Great Pumpkin’ objection.Footnote 1 These sorts of objections do not criticize Plantinga's account directly, but instead seek to show that Plantinga's method could be used to defend some very strange beliefs, and hence that Plantinga's method must be flawed or incomplete. This objection is especially troubling because it is claimed that Plantinga's method could be used by a very wide variety of belief systems such as voodoo, astrology, beliefs about the man in the moon, and, of course, the Great Pumpkin belief itself.
The objection that I wish to articulate should be thought of as a Great Pumpkin objection because I intend to argue that something analogous to Plantinga's religious epistemology could be adopted by others who hold clearly irrational beliefs to defend those beliefs. This sort of objection has been gestured towards previously. I begin by presenting the objections that have previously been offered by Keith DeRose and Linda Zagzebski, but we will find that they are under-described. Following that I will describe the objection in more detail and show why it is a problem for Plantinga's religious epistemology. Finally, I will then offer a response to the argument, using some insights from Pritchard (Reference Pritchard2010).
Voodoo epistemology
In an unpublished essay entitled ‘Voodoo epistemology’ Keith DeRose asks of Plantinga what he calls ‘The Real Pointed Question’:
Now couldn't this be argued with equal cogency with respect to some weird/bizarre aberrations of irrationalism?, where ‘this’ refers to Plantinga's defensive strategy. (DeRose, unpublished manuscript)
DeRose claims that the problem is that there seem to be some clearly irrational beliefs that could be defended using Plantinga's strategy, and if that is the case, then there must be something wrong with the strategy, even if we cannot say exactly what it is.
A similar objection has been made by Linda Zagzebski. Zagzebski writes that Plantinga's project:
does not permit a rational observer outside the community of believers in the model to distinguish between Plantinga's model and the beliefs of any group, no matter how irrational and bizarre – sun-worshippers, cult followers, devotées of the Greek gods … , assuming, of course, that they are clever enough to build their own epistemic doctrines into their models in a parallel fashion. But we do think that there are differences in the rationality of the beliefs of a cult and Christian beliefs, even if the cult is able to produce an exactly parallel argument for a conditional proposition to the effect that the beliefs of the cult are rational if true. Hence, the rationality of such beliefs must depend upon something other than their truth. (Zagzebski (Reference Zagzebski2002), 122)
Like DeRose, Zagzebski is concerned about who might be able to copy Plantinga's strategy. She claims that any group could defend their beliefs in the way Plantinga does, so long as they are clever enough (and willing) to build in the appropriate epistemic doctrines. Zagzebski also says that if Plantinga's strategy is to be successful then the rational outside observer ought to be able to judge that Plantinga's strategy succeeds in some way that the defenders of these irrational and bizarre beliefs fail.
In response to this complaint Plantinga writes that his method is not available to a wide variety of other belief systems:
My argument depends essentially upon premises about God's knowledge, intentions, and power; parallel premises about the sun or the moon or Greek gods will be manifestly false. Naturalists, furthermore, will also be unable to construct such an argument, as, indeed, will non-theists of any stripe. (Plantinga (Reference Plantinga2002), 130–131)Footnote 2
An important part of Plantinga's argument is that what we are discussing is belief in a personal God who created us with – at least in part – the intention that we can know and love him. It is because of this that Plantinga is able to argue that the truth of this belief would have important implications for the sorts of beings that we are and our cognitive capabilities. This means that something very much like Plantinga's arguments could be used by most monotheists, but not by followers of voodoo, sun-worshippers, and followers of the Great Pumpkin.
DeRose and Zagzebski claim that Plantinga's strategy could be used by a wide variety of strange belief systems, but it is not clear that they are right. This means that we are left with something of a stalemate – on one side we have Plantinga insisting that his arguments could only be adopted by major world religions and on the other DeRose and Zagzebski claim that many others could use similar arguments as well.
In order to make progress in this debate we need to outline what Plantinga's strategy is. To achieve this we must first answer two questions:
1. What is Plantinga's method?
2. What is Plantinga's method supposed to show?
By answering these two questions we will then be able to see how to advance this objection properly and assess whether Plantinga's response is adequate. That will be the subject of the following section.
Return of the Great Pumpkin
What is Plantinga's method?
Plantinga's method involves arguing that it is not reasonable to believe that Christian belief is unwarranted or irrational because there is a model under which Christian belief is both rational and warranted, and this model is true for all we know. Plantinga means by a model a possible state of affairs such that if it obtains then some target proposition will be warranted.
He constructs a model which he calls the Aquinas/Calvin model (A/C model for short).Footnote 3 The target proposition for the A/C model is that God exists and it involves claims such as that God has created us with something like what Calvin calls the Sensus Divinitatis which is a cognitive faculty that, in the correct circumstances, produces the belief that God exists. Plantinga argues for three important claims about the A/C model, which he summarizes as follows:
(a) [the extended A/C] model is possible, both logically and epistemically; (b) given the truth of Christian belief, there are no philosophical objections to this model's also being not merely possible but true; and (c) if Christian belief is indeed true, then very probably it does have warrant, and has it in some way similar to the extended A/C model. (Plantinga (Reference Plantinga2000), 350)
Although these claims involve Plantinga's extended A/C model,Footnote 4 which makes reference to God and his activities, there is no reason why some other model could not be inserted in its place, so long as it gave an account of how some other belief could be warranted. This means we can now restate this method:
Plantinga's method for belief p:
(i) There is a model under which p is warranted, which is both logically and epistemically possible.
(ii) Given the truth of p, there are no philosophical objections to the model also being not merely possible but true.
(iii) If p is true, then very probably it does have warrant in some way similar to the model.
One can use Plantinga's method to defend a belief p if one can give a model in which the target belief is warranted, and this model must be logically and epistemically possible. Also there needs to be an important connection between the truth of p and the truth of the model, such that there are no successful arguments that show that the model is false that do not assume the falsity of p. And finally, it must be that the truth of p makes the truth of the model likely, or, at least something like the model.
Plantinga often summarizes this by saying that the belief is likely warranted-if-true. Many other beliefs are warranted-if-true, for example the belief that there is a table before me in clear view is warranted-if-true because the truth of that belief makes it likely that I am in a position to know that it is true given that I have good eyesight. Most beliefs are not, however, warranted-if-true; such as beliefs about what the weather will be like tomorrow, that Mount Everest is the tallest mountain in the world, and that 56×82 = 4592. Though these beliefs may in fact be both warranted and true, they are not warranted-if-true because their truth does not give us reason to think that the person who has those beliefs has warrant for them. In these cases we can clearly separate questions about the truth of these beliefs from those about how warranted they are.
According to Plantinga, belief in God is different because the truth or falsity of this belief has important implications about the proper functioning of our cognitive faculties, and therefore, which beliefs are warranted for us.Footnote 5 This means that questions about the warrant of belief in God cannot be separated from the truth of those beliefs. The Return of the Great Pumpkin objection seeks to respond to this by claiming that it has unpalatable consequences because there are some very strange beliefs that are warranted-if-true.
What is Plantinga trying to show?
This question is important to answer because part of the Return of the Great Pumpkin objection involves finding a belief that can not only be defended using Plantinga's method but also clearly does not have the status that Plantinga argues Christian belief has. Plantinga considers a number of different statuses such as justification and internal or external rationality, but he finds all these to be inadequate for various reasons. I will not repeat Plantinga's comments here.Footnote 6
Plantinga's most important contribution to religious epistemology is called Warranted Christian Belief, so one might expect him to argue that Christian belief is warranted, and therefore that what is at issue here is whether or not Christian belief is warranted. But this is not what we are concerned with here since Plantinga does not attempt to argue that theistic belief is warranted, only that it is warranted-if-true.
We need then, to turn to what Plantinga himself has said was his aim. We find this is the first line of Warranted Christian Belief: ‘This book is about the intellectual or rational acceptability of Christian belief’ (Plantinga (Reference Plantinga2000), vii). Plantinga says that the purpose of his book is to assess all the ways in which the acceptability of Christian belief has been challenged. Acceptability is Plantinga's term that is intended to cover all the various positive statuses that it is claimed that Christian belief lacks. For example, some claim that Christian belief is unacceptable because it is not based upon evidence, others that it is unacceptable because it is false or others that it is unacceptable because it is unwarranted.
Plantinga argues that Christian belief should be thought of as acceptable because for any understanding of acceptability one of the following applies:
1. There is good reason to think that a belief need not meet the relevant requirement in order to be acceptable (e.g. that beliefs must be based upon evidence to be acceptable).
2. It can be shown that the belief is acceptable in that sense (e.g. deontological justification).
3. There is no good reason to think that the belief does not meet the requirements for that understanding of acceptability (e.g. warrant).
Plantinga does not give a definition of acceptability and it is likely that it would be very difficult to do so because acceptability is supposed to cover a wide variety of statuses that a belief could have. To be fully acceptable, a belief would have to be known, and therefore true, but this would not be a very helpful definition of acceptability, since there are many cases in which there is no agreed method for determining what is true, as this debate demonstrates. What Plantinga is aiming to show is that Christian beliefs should be considered acceptable by others even if those beliefs are not based upon evidence. The acceptability of Christian belief in this sense is compatible with the acceptability of other contradictory belief systems such as other religions and non-theistic belief systems.
In the context of advancing the Return of the Great Pumpkin objection we do not need a definition because the objection does not seek to challenge the suggestion that Christian belief is acceptable, only that it has not been shown to be so using Plantinga's method. It achieves this by suggesting that there are beliefs that can be defended using Plantinga's method which are uncontroversially not acceptable, and for this we do not need a precise definition of acceptability. It is enough for present purposes that we have agreed examples of beliefs (such as the Great Pumpkin belief) that should not be considered acceptable, whatever the best understanding of that term happens to be.
The Enhanced Great Pumpkin belief
Now that we have answered these questions we are in a position to complete the Return of the Great Pumpkin objection. All that remains is to find an example of a belief that is clearly unacceptable, yet which can be defended using Plantinga's method. Here is an example of just such a belief:
The Enhanced Great Pumpkin belief: For as long as Linus can remember he has believed that the Great Pumpkin exists, and he does not base this belief on any of his other beliefs. Many people have objected to his belief, saying things such as that he is unreasonable because he has no evidence, but he weighs up their arguments and still finds his belief in the Great Pumpkin compelling. He is also able to explain why others do not believe because the Great Pumpkin has implanted in him an imperceptible faculty whereby he is able to communicate with Linus but he has not done this for anyone else. The Great Pumpkin has also explained all this to Linus using this imperceptible faculty. Each Halloween when Linus is again disappointed that the Great Pumpkin fails to turn up he communicates with Linus to apologize and explains that he will try harder next year.
This account fulfils Plantinga's tripartite method above. There is a model under which Linus's beliefs are warranted, namely the one where Linus has this faculty for communicating with the Great Pumpkin, and it is logically and epistemically possible. Given the truth of the beliefs there seem to be no sound objections to the model; that is, there does not seem to be a way to show that the model is false without begging the question against Linus. Finally, if the beliefs are true then it is very likely that they have warrant in a way similar to that described above. It should also now be obvious that there are many ways to construct warranted-if-true beliefs, as long as we are prepared to build in to them details about the origin of the belief.
But is Linus' belief acceptable? Intuitions may vary here, and are likely to be affected by how one imagines the scenario. Linus, in the comic strip, is a child with an overactive imagination. But what if we redescribe the scenario to make Linus a well-educated adult, who aside from his Great Pumpkin beliefs seems to have quite common beliefs? Some might think that if Linus is sincere and otherwise rational then we ought to give him the benefit of the doubt, and regard his belief as acceptable. This would be mistaken, although we might regard Linus as being rational, on some ways of understanding what it means to be rational, we should not regard his belief as acceptable. This is because we have good reason to think that something has gone wrong in the way this belief has come about (though it may not be Linus's fault); reasons such as that we know lots about pumpkins, but have never encountered one like this, or that there has never previously been a case of someone having an imperceptible faculty whereby she can communicate with another person. These reasons should be enough to support the belief that something is wrong with Linus's belief (even though they do not show that the beliefs are false), and therefore it is unacceptable.
This objection is that Plantinga's method for defending religious belief is not distinguishable, in any epistemically relevant way, from the defensive strategies that might be used by someone who holds clearly unacceptable beliefs like the Enhanced Great Pumpkin belief. This also demonstrates that Plantinga's response to DeRose and Zagzebski is not adequate. Although Plantinga is right that the Great Pumpkin belief cannot be defended in exactly the same way as Christian belief, it can still be defended in a very similar way and that is enough to call into question the success of his argumentative strategy.
It does not follow from this that Christian belief is irrational or defective in any way even if Plantinga's description of it is correct. It may, after all, be the case that the epistemic practices of the Christian are indistinguishable from those who hold clearly unacceptable beliefs, but yet there is some important difference that we are unable to observe. So, if this objection is successful, it means, not that the Christian (on Plantinga's account) is irrational, only that Plantinga's defensive strategy does not achieve what it sets out to do.
The conclusion can now be seen to be quite modest, though still worrying; it is not that Christian belief as described by Plantinga is unacceptable, it is rather that Zagzebski's outside rational observer has no good reason to treat it as acceptable. This outside observer cannot see any relevant epistemic difference between the behaviour of Plantinga's Christian and the Great Pumpkin believer, so the outside observer is correct to judge the epistemic practices of the two equally, that is, as clearly unacceptable.
The solution
This solution takes its inspiration from an article by Duncan Pritchard (Reference Pritchard2010). In this paper Pritchard discusses Fred Dretske's famous cleverly disguised mule case. We are asked to imagine a person – Pritchard calls her ‘Zula’ – who has ordinary cognitive abilities and background knowledge. She is at the zoo and can clearly see a zebra in the animal enclosure marked ‘zebra’. Upon seeing this animal Zula forms the belief that it is a zebra, and we are asked to consider whether she knows that it is a zebra. Intuitively it seems that this is a clear case of knowledge. However, a problem arises when we suppose that Zula knows that if she is looking at a zebra, then she is not looking at a cleverly disguised mule.
This is a problem because there are two very plausible principles that together yield a contradiction in this scenario. The first is the closure principle, and the second the discrimination principle:Footnote 7
The Closure Principle
If S knows that p, and S competently deduces q from p (thereby coming to believe q while retaining her knowledge that p), then S knows that q.
The Discrimination Principle
If S has perceptual knowledge that p, and S knows that another (known to be inconsistent) alternative q does not obtain, then S must be able to discriminate between the object at issue in p and the object at issue in q.
Given the closure principle, Zula is able to deduce that she is looking at a zebra rather than a cleverly disguised mule, but according to the story, Zula is not able to discriminate between a zebra and a cleverly disguised mule, so it follows from the discrimination principle that she does not know that she is looking at a zebra rather than a cleverly disguised mule.
Dretske has used this example to argue that we should give up the highly intuitive closure principle, but Pritchard claims that it is possible to make sense of the example and preserve the closure principle by giving up the discrimination principle instead. Pritchard contends that this is not as implausible as it might first appear. There is reason to think that Zula does in fact have knowledge both that she is looking at a zebra and that it is not a cleverly disguised mule because her beliefs could not very easily have been false. More needs to be said to understand this scenario, but this suggests that we ought to look again at our intuition that Zula does not know she is not looking at a cleverly disguised mule.
Abandoning the discrimination principle creates another puzzle – what is Zula's supporting evidence for her belief that she is not looking at a cleverly disguised mule? It cannot be the way it appears to her, since we are assuming that Zula does not have the ability to tell the difference between a zebra and a cleverly disguised mule just by looking. This is a problem because, although we may not wish to claim that all knowledge must be supported by evidence, it seems that this sort of belief ought to be evidentially grounded.
Pritchard believes that this puzzle can be dissolved if we draw a distinction between favouring and discriminating evidence. He claims that if Zula knows that she is looking at a zebra and not a cleverly disguised mule, then she must have better evidence in support of it being a zebra than that it is a cleverly disguised mule, but this does not need to be discriminatory evidence (i.e. evidence that would be present if she were looking at a zebra, but not a cleverly disguised mule). This initially seems implausible because, if she was looking at a cleverly disguised mule, Zula would not be able to distinguish her perceptual experience from the experience of seeing a zebra. However, Pritchard challenges the assumption that Zula comes to know that there is a zebra ‘just by looking’; instead there will be other evidence – favouring evidence – that plays an important role when error-possibilities are raised. Pritchard imagines how Zula might react when the cleverly disguised mule possibility is raised:
One might reason, for instance, that there would be no point in such a deception, that it would be costly and time-consuming without bringing any comparable benefit, that it would be easily found out, and then the zoo-owner would be subject to penalties, and so on. (Pritchard (Reference Pritchard2010), 256)
Here we can see that Zula possesses, as part of her background knowledge, evidence that supports her belief that what she is looking at is a zebra. For this reason it is not puzzling that Zula is able to know that she is looking at a zebra, and deduce from this that she is not looking at a cleverly disguised mule because her evidence for the former is better than her evidence for the latter.
Pritchard's observations offer us a way to solve the Return of the Great Pumpkin objection. His concern was about perceptual discrimination between objects, whereas our concern is about how some rational observer might discriminate between the epistemic practices of others. Just as Zula can acceptably believe that there is a zebra and not a cleverly disguised mule before her even though she cannot discriminate between them, so too, the rational observer may be able to believe reasonably that one person has acceptable beliefs, and another not, even though she cannot discriminate between their epistemic practices.
It might seem at first that the Zula case is not analogous to the case being considered in the Return of the Great Pumpkin objection, since the objection does not involve the claim that the Christian is in an equally good evidential position concerning both the existence of God and the existence of the Great Pumpkin. The claim made by the reformed epistemologist is that there may be some experience that grounds the Christian's belief in God, and although some might deny that such an experience could ground belief in God, it would not be argued that such an experience grounds equally well the belief that the Great Pumpkin exists. The analogy can be made if we make Zula parallel, not with the Christian, but with a third person observing the religious believer and the Great Pumpkin believer.
The claim being made by Plantinga is that we ought to regard belief in God as acceptable because it is warranted-if-true. The Return of the Great Pumpkin objection counters this by pointing out that showing that a belief is warranted-if-true cannot be sufficient to show that some belief is acceptable because the status of being warranted-if-true does not allow us to discriminate between belief in God and other beliefs that are also warranted-if-true but clearly not acceptable, such as belief in the Great Pumpkin.
This now shows us how it is possible to respond to the Return of the Great Pumpkin objection. The reformed epistemologist can concede that there may be no way for the rational outside observer to discriminate between the epistemic practices of Plantinga's Christian and the Great Pumpkin believer. She will, however, be able to show that there is favouring evidence in support of the Christian's epistemic practices, whereas there is no such evidence for the Great Pumpkin belief.
There may be many ways of supplying favouring evidence, here is one example:
Christian beliefs are held by many different people; people from a wide variety of times, countries and cultures, people with varying levels of intelligence and education, a variety of ages, social classes and life experiences. The Christian tradition also contains people who have devoted much of their lives to considering arguments for and against Christian belief, and have continued in their beliefs. Some Christians have always held religious beliefs, some have come to hold them later in life, and still others have given them up and then come to believe them again.
These facts favour the claim that the theist's beliefs are acceptable. It is clear that Linus's beliefs, on the other hand, do not have this favouring evidence. But what if we amend the story so that he does?
Perhaps Linus lives on an, until now, undiscovered island in a community of Great Pumpkin followers. For thousands of years this community has lived there without any contact with the rest of the world. Suppose that we now find this island and meet Linus. He tells us about his belief in the Great Pumpkin. We are perplexed and begin to ask him questions and raise objections to his belief. Sometimes he can answer our questions, at other times he says that he thinks some of the Great Pumpkin writers of the past have written about these subjects, and that if we look them up he is sure we will find answers. Other questions have never occurred to him, so he agrees to bring them up at the next pumpkin meeting so that the island's best minds can go to work on them.
In this scenario it is likely that none of us would be persuaded of Linus's beliefs, but we would not judge his beliefs to be unacceptable since it is clear that the community he is in has reflected upon the problems that have arisen and sought to find answers.
One of the things that Plantinga is attempting to do in Warranted Christian Belief is to show that dismissing Christian belief simply because the Christian cannot produce compelling evidence in support of her beliefs is in some way illegitimate. He has sought to do this by showing that even if the Christian does not possess this evidence her beliefs may still be justified, rational, and warranted. From our current vantage point it is clear that this is not enough. We do reject beliefs for which it is possible that they are justified, rational, and warranted – or we would if we met people with those beliefs – even if we cannot produce good reasons for thinking those beliefs are false. The Return of the Great Pumpkin objection seeks to highlight this by showing that it is possible that there are other beliefs that similar claims can be made about, yet which we reasonably dismiss – e.g. Linus's Great Pumpkin belief.
This problem shows that there is a shortcoming in Plantinga's approach. Plantinga has sought to give an account of how Christian belief can be acceptable without the need for Christian believers to have evidence or arguments that they can produce in support of their beliefs. The Return of the Great Pumpkin objection shows that doing that does not also demonstrate that others outside the community ought to regard Christian belief as acceptable. To show to others that one's beliefs should be considered acceptable one must turn to the evidence, though, as can be seen from this example, the burden is not too great.
Summary
The Return of the Great Pumpkin objection concerns whether or not Plantinga has successfully shown that Christian belief is intellectually acceptable. This is achieved by showing that Plantinga's method for defending Christian belief can be mimicked by those with clearly unacceptable beliefs. I have also shown that we can overcome this objection by drawing a distinction between discriminating and favouring evidence. We possess favouring evidence for the claim that Christian belief is intellectually acceptable by considering the historical and social environment in which the belief occurs. This evidence is not available for the Great Pumpkin belief.
