Sea of Faith — Dunedin

Gregory W. Dawes
University of Otago
Dunedin Sea of Faith Group
22 May 2008

1. Two Questions about Religion
There are two questions one can ask about belief in God. The first is: Are there good reasons to believe in God? This is a question with which every philosopher is familiar and which I deal with every year, when I discuss the arguments for and against the existence of God. But there is a second question one can ask, namely: Why do people believe in God? The eighteenth-century philosopher David Hume distinguished these two questions by writing two different works on religion. One of these, his Dialogues on Natural Religion, addresses the first question: Are there good reasons to believe in God? Another work, his Natural History of Religion, addresses the second question: Why do people believe in God?
You might argue that although there are two distinct questions, the answer to the first is also the answer to the second. People believe in God because they have what seem good reasons to do so. You think you could say this about some beliefs, most clearly in the case of scientific theories. Why do physicists believe that electrons exist? (No one, it is worth noting, has ever seen an electron, just as no one has ever seen God.) Most physicists believe that electrons exist because they have what seem good reasons to believe that electrons exist.
More precisely, they believe that certain phenomena are best explained by positing the existence of electrons. And being (for the most part) realists about science, they think this gives good reason to believe electrons exist. Are there phenomena that are best explained by positing the existence of God? If there were, this would give us good reason to think that God exists, at least on a realist understanding of God-talk. Religious believers often claim that there are such phenomena. Advocates of “intelligent design,” for instance, argue that the “specified complexity” of biological organisms is best explained by positing the existence of a designer. Some think that this is the only possible explanation of such organisms. Others think that even if particular phenomena such as the existence of complex organisms can be explained without reference to God, the very existence of an orderly, life-producing universe cannot.
Are arguments of this kind the reason people believe in God? Well, they may be the reason some people believe in God. But they can’t be the whole story. Firstly, religious people generally believe many other things that apparently cannot be supported in this way. As philosopher Alvin Plantinga puts it,
what is the best explanation for all that organized complexity in the natural world and the characteristic features of human life and all the rest of what we see about us? Well, let’s see, perhaps there is an omniscient, omnipotent, wholly good being, who created the world. Yes, that’s it; and perhaps this being is one of three persons, the other two being his divine son, and a third person proceeding from the first two (or maybe just the first), yet there are not three gods but one; the second person became incarnate, suffered, was crucified, and died, thus atoning for our sins and making it possible for us to have life and have it more abundantly. Right; that’s got to be it; that’s a dandy explanation of the facts.
Of course, Plantinga may be wrong. There may be (as Richard Swinburne argues) good reasons not merely to believe in God but to believe in the truth of a particular theistic faith, such as Christianity. It may be, for example, that the best explanation of the existence of the Bible or the Qur’an is that it was inspired or revealed by God. If so, we would have good reason to believe that what it said was true. Some believers certainly claim this is the case. Some Muslims, for instance, argue that the Qur’an contains scientific information (for instance, about embryology) which Muhammad, as a seventh-century Arab, could not possibly have known. It follows that the only explanation of the existence of the Qur’an is that it came from God. I suspect that such arguments are often mere rationalizations, that those who put forward these arguments hold their beliefs for quite different reasons. Their arguments certainly appear unconvincing to anyone who does not already share their faith. But it may be the case that some Muslims have accepted Islam on the basis of such arguments.
There is, however, a second reason to think that such arguments are not the whole story. They might explain why some people believe in God, they might explain why some people accept a particular theistic faith, but they don’t explain their degree of conviction. The problem here is that the degree of conviction with which religious beliefs are held enjoy often far outstrips the force of any arguments which might be offered in their support. Let me offer an illustration. One philosopher who thinks that arguments can be offered in support of the Christian faith is Richard Swinburne. But what does Swinburne claim? He thinks that it is more likely than not, by a relatively small margin, that the Christian God exists. But this is not the attitude of many religious believers. They don’t say, “Well, on balance, it is more likely that God exists than that he doesn’t.” They claim to know that God exists, and seem to harbor no doubts about the matter. So even if their belief in God could be explained as the result of arguments, their degree of conviction cannot.
So it seems that even if there are good reasons to believe in God, the reasons are not the whole story. Something else must be motivating belief in God, which means that the answer to the second question cannot be simply identical to the answer to the first. It is this conviction which has given rise to “theories of religion,” attempts to explain, often on non-intellectual grounds, why so many human beings believe in God. J. Samuel Preuss begins his recent survey of such theories with the work of the sixteenth-century thinker Jean Bodin, but some of the better- known names of this history are Karl Marx, Émile Durkheim, and Sigmund Freud.

2. Natural and Supernatural Theories of Religion
Do believers themselves have a theory of religion? Often, they do. They have, in many cases, a theological explanation of at least one religion, namely their own. Christian thinkers, for instance, might argue that the reason at least some people believe in God is that God has implanted in all human beings a reliable cognitive mechanism, the sensus divinitatis, which spontaneously gives rise to belief. Or at least, it does so if they are further enlightened by the gift of the Holy Spirit, which God gives to his elect so that they may believe. This was, more or less, John Calvin’s view; Roman Catholics have traditionally held a similar view about what they called the “gift of faith.”
The problem facing the theologian is that such a theory will work for only one religion, or at most for a very few. For it is surely inconceivable that a morally perfect and omniscient God would inspire mutually incompatible religious beliefs. He would not lead the Christian to believe that Jesus is the incarnate Son of God and the Muslim to believe that Muhammad is the last and greatest of the prophets. For both these beliefs cannot be true. So even the theologian needs another kind of theory of religion, to explain what is—from his point of view—religious error. He may attribute such error to the work of the devil, another supernatural agent. Or he may join the rest of us and offer a natural history of religions: an account of why at least other people believe which makes no reference to supernatural agents.
Needless to say, the theories in which I am interested are natural theories of religion. They make no reference to supernatural agents. Do they entail that there are no supernatural agents? Are such theories inherently atheistic? Not necessarily. The two questions are, at least in principle, distinct. There may exist a God, and there may be good reasons to believe there exists a God, even if we have an entirely natural explanation of why people believe in God. But some of these theories suggest that belief in God is the result of cognitive dysfunction. We believe in God, they say, because something has gone wrong with our generally reliable belief-forming mechanisms. If this were true of all forms of religious belief, then presumably such theories would not be compatible with belief in God. Whether our best theories of religion do, in fact, suggest that religion is an error is a question to which I shall return.

3. Scientific Theories of Religion
The problem with the classical theories of religious belief—such as those offered by Marx, Durkheim, and Freud—is that they fall short of the standards of science. It is not clear what would falsify such theories; hence it is impossible to subject them to the kinds of experimental tests that a scientist would favour. Indeed it is hard to see that a work such as Freud’s Moses and Monotheism would have been taken at all seriously, if Freud had not already established his reputation as the founder of psychoanalysis. (Whether Freudian psychology or Marxist sociology themselves meet the standards of science is another question. The philosopher Karl Popper certainly believed they didn’t and many psychologists and sociologists are sceptical about them today.)
In the last fifteen years, however, the situation has improved. We now have theories of religious belief that do meet the standards of science. The more controversial of these are evolutionary theories, which suggest that religious belief could be the product of natural selection, acting on either individuals or groups. But there are also cognitive theories of religion, which rely on testable observations about the structure of human minds. Such observations explain why it is we are so inclined to develop or to accept religious ideas. It is these cognitive theories of religion on which I wish to focus. They are, it should be noted, compatible with evolutionary theories of religion. It may be that evolutionary theory can explain why we have the minds we do. But starting from the fact that we have minds of this kind, cognitive science can explain why human beings seem to be “naturally” (or, perhaps one might say, “incorrigibly”) religious.
Let me give you a few names. Among the cognitive theorists of religion are Stewart Guthrie, Pascal Boyer, Justin Barrett, E. Thomas Lawson, Robert McCauley, and Scott Atran, the last of whom marries cognitive and evolutionary perspectives. All build on the work of Dan Sperber, who has developed what he calls a general “epidemiology of belief.” Pascal Boyer’s book Religion Explained is a masterful synopsis of his own research, while Stewart Guthrie’s Faces in the Clouds is a very readable introduction to at least one form of cognitive theory. For a useful survey of the field, you might want to read Todd Tremlin’s Minds and Gods or Justin Barrett’s Why Would Anyone Believe in God?, from which I have borrowed my title.

4. The Cognitive Science of Religion
What do such authors say? Their central claim is that our thinking about gods is shaped by a number of apparently universal cognitive mechanisms or mental devices. These mental devices not only shape the formation of our “non-reflective, “ that is to say spontaneous and taken-for-granted beliefs. They also determine what kinds of reflective beliefs we are likely to find plausible. One of these cognitive mechanisms is what Barrett calls a “hypersensitive agency detection device” or HADD. Human beings seem hard-wired, as it were, to detect agency. We tend to interpret what we see or hear as the work of beings like ourselves, with beliefs and desires, acting for a reason. We see “faces in the clouds” (as Hume remarked), we perceive a rock as a bear, and believe that the bump in the night is an intruder or a ghost. Indeed when experimental subjects observe geometrical shapes moving around a computer screen, they naturally describe such movements in terms of purpose: one shape is “chasing” another or is “trying” to get over a barrier, and so on.
Incidentally, it is easy to see how this observation could be coupled with an evolutionary theory of religion. The fact that our tendency to detect agency is hypersensitive—it gives rise to false positives, seeing agency where no agency is present—presumably has survival value. If one is going to make perceptual errors, it is better to see a rock as a bear than to see a bear as a rock. It is easy to see why natural selection might favour such a tendency. But whatever one thinks of such evolutionary explanations, the existence of such a tendency seems to be well established.
One feature of HADD—the tendency to interpret the world in terms of purpose—seems to be particularly characteristic of children’s thinking. Given a choice between a mechanistic explanation and one that assumes that things have a purpose, children will tend to choose that which assumes purpose. “Why are some rocks pointy?” “So that animals can’t sit on them and smash them” It follows, researcher Deborah Kelemen has argued, that children are “intuitive theists,” insofar as they are predisposed to interpret the world in terms of design. No wonder creationists want their views to be taught to children in schools. Children would be particularly receptive.
Of course, the mere fact of interpreting what we see in terms of purpose doesn’t mean we will believe in God. We still need the idea of a god to apply to our experience and such ideas are culturally transmitted. So what can we say about the spread of religious ideas, the “epidemiology of belief” of which Dan Sperber writes? Cognitive scientists have argued that our concepts of supernatural beings combine two features: they are memorable and they are easy to work with. In particular, Pascal Boyer has argued that our concepts of gods and other supernatural beings are “minimally counterintuitive.” They depart from our usual conceptions of agents in ways that make them interesting, but they remain sufficiently similar to such agents that we can reason about them, relate to them, and predict how they might act.
Some work done by Justin Barrett is interesting in this respect. Barrett designed some experiments intended to distinguish between what people claimed to believe about God and how they actually thought about God when reasoning about religious matters. It turns out that while people may claim to believe, for example, that God was omnipotent and omnipresent, in practice they assume that he can assist only one person at a time. This leads to what Boyer calls “the tragedy of the theologian”: whatever sophisticated notions theologians may develop about God, in practice people keep thinking about God in the same old, anthropomorphic way.
An interesting feature of such theories of religion is that they can incorporate the reasons which people may offer in support of their faith. When faced with puzzling and otherwise apparently inexplicable facts—whether some disturbing tragedy such as an earthquake or the wondrous complexity of the natural world—people can make use of their God-concepts to try to make sense out of what they see. Whether or not the existence of God is the best explanation of any of these phenomena, many believers think that their beliefs have explanatory power. If the apparent explanation at which they arrive is at least emotionally satisfying, their belief will be reinforced. In their more sophisticated and reflective forms, these “efforts to explain” become the familiar arguments for the existence of God.
There are other features of human cognition which these theories highlight, which predispose us to believe in God. The fact, for instance, that the gods are thought to know and be concerned about how we behave means that we have a particular reason to pay attention to what they say. And the fact that we find it so difficult, perhaps impossible, to imagine not existing predisposes us to believe in a life after death. But the factors I have mentioned should be enough to give you some sense of what a cognitive theory of religion would be like.

5. Implications for Religious Belief
Let me end with the question I mentioned earlier: What implications do such theories have for religious belief? The first point to make is that nothing that has been said above entails the falsity of religious belief. One could, consistently, accept these theories of why we believe while holding that at least some religious beliefs are true and that God exists. It is true that these theories posit a hypersensitive agency detection device, a tendency to detect agency even when no agency is present. But of course sometimes we detect agency when agency is present, as when I interpret your actions as purposeful. (I am assuming that you do have beliefs and desires and that you sometimes act purposefully. But you, at least, are unlikely to contest this assumption.) So to say that the believer’s intuitions regarding God are false is to assume that this detection of apparent agency is a false positive: that there is, in fact, no divine agent. This may be true, but it needs to be shown rather than merely assumed. Perhaps there are good reasons to believe that there is no God, but such reasons have nothing to do with cognitive theories of religion.
This is not to say, however, that such theories have no implications for belief. Firstly, if you do decide that belief in God is false, or at least unwarranted, then you need to explain why it is that so many people embrace a false belief. These cognitive theories of religion would explain why human beings are so prone to fall into this particular error, assuming it is an error. But even if you think that belief in God is warranted, that belief in God is not an error, a cognitive theory of religion is hard to reconcile with at least some theological claims. In particular, they are hard to reconcile with theological explanations of why we believe, such as that offered by John Calvin. As we’ve seen, Calvin thought that belief in at least the Christian God arose from a divinely implanted sensus divinitatis, coupled with the internal testimony of the Holy Spirit. But cognitive theorists of religion regard belief in God as arising from entirely natural mechanisms, which have evolved for other purposes.
There are, it seems to me, two options for the theologian here, if he concedes that these cognitive theories of religion are convincing. The first is to regard these cognitive mechanisms as Calvin’s sensus divinitatis, as God’s way of ensuring that most people come to faith. But this idea seems implausible. Why would the Christian God design us in such a way that we believe, not merely in him but in thousands of false deities? If this is a divine design, it seems to be a poor one. The second option is to insist that the theologian’s favored faith is an exception to such explanatory schemes. While other religions may be the result of natural mechanisms, Christian faith (for instance) is the result of God’s action in the heart of the believer. But unless some reasons can be given for regarding Christianity as different, such an exemption seems arbitrary, nothing more than a form of special pleading. Sadly, much twentieth-century theology has been precisely this, a form of special pleading. Whether the theology of the future will be any different is a question that remains to be answered.