Reading: Chapter 1: What is Religious Experience?

WHEN I ASKED YOU TO TALK WITH ME ABOUT HOW ANY EDUCATED person in this scientific age could still believe in God, I expected you to start our discussions with arguments for God's existence. Instead, you start by talking about religious experience! Does that mean you're admitting right from the start that the proofs for God's existence don’t work?

Yes. I'm skipping the arguments for God's existence; first, because in my opinion none of them succeed. I don't believe their failure matters much, though. It’s not true that belief in God is justified only if proven and that otherwise it's blind faith. Proving is actually an inferior way of coming to know something. Proving is standard procedure in philosophy and the sciences, because it’s their business to construct theories (educated guesses) to explain their subject matter. They then construct arguments and weigh evidence to prove or disprove the hypotheses in their theories. But genuine belief in God is not a hypothesis, which is why it doesn't need proof. It’s an experience report, not an educated guess – which is what a hypothesis is. That is to say, it's a belief that is both acquired and justified by experience.

The second reason for skipping the arguments is that you asked me why I believe God is real and is the sort of being the Scriptures say God is. If I’m to answer candidly, my answer has to be a confession - an admission - of my real reasons. If, instead, I invent arguments intended to make that belief look good to you, they won't be my reasons for believing since I don't believe in God because of any arguments (and I don't know anyone else who does either). So that's why I want to start our discussions by talking about religious experience. There is also a third reason why the project of trying to prove God exists should be rejected, but that can wait till later.

Fair enough. I did ask how you can believe something that sounds like a convenient fairy tale. But to be as candid with you as you want to be with me, I have to say that I'm at least as skeptical about an appeal to religious experience as I am about the proofs of God's existence. I've never had any such experience nor do I know anyone who has.

Anyway, most of the claims I've heard about such experiences sound a lot like the reports of people who claim to have been kidnapped by aliens from outer space! So even if belief in God is "acquired" by some sort of experience, I don't believe it can be justified by it!

Perhaps no other topic is in as much need of having its name explained as is "religious experience." So before we go any further, let’s clarify the meaning of that expression. That will take some doing, but when the dust clears I hope to have shown you that it does not refer only - or even primarily - to weird events such as mystical states of consciousness, visions, or miracles. An experience isn't religious only if the furniture flies around the room.

Of the two terms in the expression, the meaning of "experience" is easier and can be cleared up pretty quickly. I'm going to use it in the broadest possible sense: an experience is being aware of anything whatever, regardless of the nature of what is experienced or the sort of capacity by which we become aware of it. The content of experience, then, is everything of which we are in any way aware, and acts of experiencing are all the ways by which we are aware of anything. Thus the expression "religious experience" refers to any experience by which a religious belief is acquired, deepened, or confirmed.¹ For the rest of these discus­sions, we can concentrate on the experience by which religious belief is acquired, since that is what you specifically asked about. So from now on, I'll only be speaking about experiences that generate belief in God rather than those that deepen or confirm it.

Hold on a moment. You seem to be switching back and forth just now between talking about religious belief in general and belief in God in particular. Which one are we going to discuss?

The account I'm going to give will first be general and then focus on belief in God in particular. To be clear about what religious experience is, we have to inquire about its role in the acquisition of any religious belief whatever. Once that is clear, we'll be in a position to examine its role concerning belief in God specifically. After that, we can go on to discuss belief in God in comparison to other religious beliefs.

OK, I think I understand what you're up to. We start by taking experience in the broadest sense, concentrating on experiences that generate religious beliefs. Then we examine how such experience constitutes the basis for belief in God.

But doesn't that mean we have to be able to tell whether a belief is or is not a religious belief to be able to tell whether or not an experience is?

Exactly! As I said, clarifying the term "experience" is the easy part. It's a lot harder to differentiate religious beliefs from nonreligious beliefs. But we must do exactly that if the meaning of the expression "religious experience" is to be clear enough to guide the rest of our discussions. So our first task has to be to arrive at a definition of religious belief. I wish this could be defined as easily as "experience." But there is a lot of confusion about religious belief, and some of the most deeply entrenched popular ideas about it are the most misleading. For that reason, we need to talk first about why some of these widely accepted ideas fail, and try to come up with a definition that successfully covers all religious beliefs. It’s only fair to warn you that this means tackling some of the most difficult issues right away.

If you can give a definition of religious belief that covers all religious belief, it will be worth the effort. But why is that hard? And what's so bad about the popular ideas?

Before we tackle the popular ideas of religious belief, let's take a moment to recall what's involved in arriving at the sort of definition we need. We have to identify the list of features that are true of all religious beliefs but are true of only religious beliefs. This is the hardest sort of definition to get, so in philosophy and science we often have to settle for other ways of delimiting what we mean by a term. Nevertheless, this type of definition would be the most helpful to our subject if we could get it - and I think we can!

Any attempt to form this sort of definition must overcome two main difficulties. On the one hand, if it fails to cover certain beliefs that are obviously religious, then it is too narrow; on the other hand, if it covers all religious beliefs but also applies to clearly nonreligious ones, then it's too broad. These difficulties can often baffle our best attempts. But even when such a definition can be formulated, there are still other difficulties that plague its acceptance. Since it can be disturbing to pare back the characteristics of a type of things till we're left with only the features that are shared by all of them and only by them, such definitions are often both surprising and disappointing. And they are frequently rejected for those reasons.

Take the case of defining what counts as a tree. Everyone easily recognizes trees, of course, and yet it is hard to state just what features are shared by all trees but are shared only by trees. The definition can be disappointing because so much that is obvious or valuable about trees is not included in it - their beautiful foliage, shade, or uses as wood, for example. Likewise, the definition may be surprising. Did you know that there are mature trees only a foot tall? So we need to recognize at the outset that this sort of defining often has such results. In fact, the more initial confusion there is about the definition of a type of things, the more certain it is that formulating a definition to clear up the confusion will be surprising or disappointing to many people.

Consider an actual example of this. Many years ago whales were classified as fish. Their bodies were shaped like fishes' bodies, they lived their lives in the oceans, and they swam. But as time went on, they were reclassified as mammals. There were good reasons for this. Whales have four-chambered hearts and are warm-blooded; lacking gills, they breathe air with lungs; and they bear their young alive and nurse them. So despite their very fishlike tails and fins, despite the fact that they can't live on land but spend their lives swimming in oceans, whales are now defined as mammals. Perhaps this redefinition was disappointing or surprising to some people when it was first put forward since it means that whales have more in common with humans than they do with fish! But it was not wrong for that reason.

I get it. You're about to drop a definition on me that's disappointing and surprising, so you're saying in advance that it doesn't matter.

Exactly right! The task of identifying the defining features of religious belief is no different in these respects from defining trees or whales. The process must inevitably leave out many of the most prominent and treasured features of each religious belief in order to state what is common to all of them - hence the disappointment. At the same time, the only definition I know that really covers every sort of religious belief also results in many beliefs turning out to be religious that are not popularly thought to be - hence the surprise. Yet neither the surprise nor the disappointment is, all by itself, a good objection to this definition. The issue is only whether the essential core of what is being defined has been identified correctly. So please remember that we're not now trying to define religion as a whole, nor are we yet addressing the question of how to tell which religious belief is true. For now, we are only trying to distinguish religious beliefs from nonreligious beliefs.

OK, I've braced myself for disappointment and surprise.

So, what exactly is wrong with popular ideas about religious belief?

One of the most widespread ideas about religious belief takes it to be the same as belief in God or, at least, in a supreme being. Many people even suspect that all religions actually believe in the same supreme being under different names. This idea seems plausible in Europe and North America because the three most widely held religions on those continents - Juda­ism, Christianity, and Islam - all believe in one God who created the universe even though they have different beliefs about how to stand in the right relation to God. In other words, this definition would be quite right if these three were the only possible religions. But that is far from being the case.

Many religions are polytheistic, which means that they believe in many gods and goddesses. Some of these religions do not recognize any one god as supreme. If belief in a supreme being were the right definition of all religious belief, we would have to say that polytheisms such as these are not religious beliefs at all. Still, other religions are literally atheistic and do not believe in any gods. Brahmin Hinduism and Theravada Buddhism are examples.² According to Brahmin theology, the gods of popular Hindu worship and practice are but mythological ways of thinking that accommodate religious truth to the level of the average person. The divine, called Brahman-Atman, is not a person or even an individual being. It is rather being-itself, which alone is real, in contrast to our everyday world which is an illusion. Teachings such as these also show why religious belief cannot be defined as belief in a supreme being. If it were correct, Brahmin Hinduism and Theravada Buddhism, as well as polytheisms with no supreme god, would all be ruled out as religious beliefs. So I reject this definition as too narrow.

Interesting! I know the US. Supreme Court has ruled more than once that the US. Laws do assume the existence of "a Supreme Being," although there is no official religion in the United States. So you’re saying that ruling doesn't allow for complete religious freedom?

I'm afraid it doesn't. It's biased toward Jews, Christians, and Muslims. Another widely accepted idea holds that religious belief is any belief that induces or supports worship or worship-related rituals. But insisting on worship in the definition is also defeated by the counterexamples of Brahmin Hinduism and the Theravada form of Buddhism since neither practices worship. Nor are they the only counterexamples. The ancient Greek philosopher Aristotle believed in a being he called both the "prime mover" and "god." But he also held that this god neither knows nor cares about humans, so he neither advocated worship of that god nor engaged in it. Similarly, his later compatriots, the Epicureans, believed in many gods but never worshiped them for the same reason; the gods, they thought, neither know about nor care about humans. So the trouble with making the inducement of worship the defining feature of a religious belief is that there are forms of two major world religions that lack it, as well as other cases in which belief in gods also lack worship.³

But if the beliefs central to those forms of Hinduism and Buddhism are not religious beliefs, what are they? And why should we suppose that Aristotle's and Epicurus's beliefs in gods were not religious beliefs when they themselves thought they were? It seems clear to me that belief in a god must count as a religious belief if anything can, so I'm going to insist (without further argument) that any definition that makes belief in a god turn out to be nonreligious has disqualified itself.

What happens if we don't concentrate narrowly on worship but look instead far whether a belief induces rituals of many kinds?

That sounds more promising but still leads to a dead end because there are a lot of rituals that are not religious. Think, for example, of swearing-in ceremonies, graduations, inductions into clubs, national anniversaries, and even birthday celebrations. All involve ritual. Gathering around a cake with candles on it and singing "Happy Birthday" is surely a ritual, but not a religious one. If there were a specific list of rituals that accompanied only religious beliefs, this might work. But there's a long list of activities that are at times religious but at other times not: burning down a house, setting off fireworks, fasting, feasting, having sexual intercourse, singing, chanting, cutting oneself, circumcising an infant, covering the body with manure, washing, killing an animal, killing a human being, eating bread and wine, shaving the head and many more. It seems clear that the only way to characterize certain rituals as religious and others as nonreligious is to determine what the people who take part in them believe about them. Without that, even an act of prayer can't be distinguished by an observer from someone’s talking to himself. This is why we need to know whether the beliefs that motivated the ritual actions were religious in order to know whether the actions were. And it's also why we are trapped in a vicious circle if we tried to determine which beliefs are religious by looking at the rituals to which they give rise.

But suppose we expand our idea of what must accompany a belief to make it religious. Suppose we include an ethical code as well as worship and ritual. Wouldn't that work?

There can be no question that most religious traditions actually include all those things. The question is whether these accompaniments make a belief religious. I think the examples of Aristotle and Epicurus already show that the answer is no, since those thinkers never connected any ethical teachings to their belief in the prime mover or the gods. Nor are those the only examples. Ancient Roman religion had no moral code or ethical teachings connected to its belief in the gods, and neither does the Shinto tradition. So the resulting definition still looks too narrow.

By the way, depending on just what sort of relation to ethics is supposed to render a belief religious, this definition can also be too broad. Surely there are many beliefs that are clearly not religious but are importantly connected to an ethical code. Many clubs and other organizations have a code of behavior as well as rituals (think of the Boy Scouts), and even some criminal enterprises have both initiation rituals and an unwritten code of "honor among thieves." But these don't make a criminal's accep­tance of the purposes and code of his organization an article of religious belief.

I must admit that I'm surprised at how badly these ideas have turned out. But none of these objections defeats another idea I've read about. Some scholars think the most plausible definition is that a religious belief is whatever a person believes to be of supreme or highest value. Is there anything wrong with that?

Yes, there is. This definition appears more plausible than it really is due to the way we sometimes speak of peoples' obsessions as their "religion." For example, we say that golf is a golf fanatic's religion or that a worka­holic's business is his religion. But just because such expressions have value as metaphors doesn't mean that they can provide a definition. It's true that there are ways in which someone's love of golf or business career can be like the devotion and fervor of saints or prophets, but that won't make it true that whatever people value most is their religious belief. And there are good reasons to think that’s not true.

For starters, we can notice that there are polytheistic traditions in which there are gods who are little valued or even hated. So if belief in these gods is religious belief, then this definition can't be right. But it’s not just a few tribal religions that count against this definition: Christianity also counts against it. To be sure, what a person values most figures importantly in Christian teaching. In the New Testament Christians are admonished to regard God's favor as valued above all else. Jesus himself said that a person's highest value should be the kingdom of God and the righteousness God offers to those who believe in him (Mt 6:33). Therein lies the difficulty, however. Belief in God is belief in a divine personal being, not in a value. Belief in God is neither itself a value nor the belief in a value, but is the basis for the proper ordering of all values. Unless a person already believed in God's existence and in the faithfulness of his covenant promises, that person could not possibly value God's favor and kingdom above all else. As the New Testament puts the point: to properly approach God one must believe that he exists and rewards those who seek him (Heb. 11:6). The belief that God is real and his promises trustworthy is thus the precondition for valuing God's favor above all else. To put the same point another way: God's favor is to be valued above all just because God is the divine Creator on whom all humans and their destiny depend. So belief in God is not religious because of what a Christian values most; rather, what a Christian values most is a result of believing in God. For that very reason the belief in God and the valuing that results from that belief cannot be identical.

This is not to deny that what people value most can often be an indicator of what they believe to be divine. But the fact that a person's highest value can reflect his or her religious belief does not mean that it always does, let alone that the belief can be defined that way.⁴

I get what you're saying. But it still seems to me that there's a sense in which a believer can be said to be valuing God, not just the right relation to God. We do really value other humans, don't we? I know I value my wife, for example.

0K. But in that case, I'm going to say that "value" is too weak to describe the believer's relation to God. Remember, belief in God is much more than mere belief that God exists. Belief in God is a wholehearted love for God that commits the believer's entire being to God in unconditional trust. That's not simply a value. It goes beyond all valuing and is different from valuing our loved ones. In the case of valuing people, the valuing is still a combination of appreciation and preference. But God is never just a preference, not even the one that outranks all others. The commandment to have no other gods "before" the Lord doesn't mean "have nothing ahead of God on your list of other gods and values." In Scripture, the term "before" signifies a covenant relationship between God and his people: to "swear before the Lord" or "eat a meal before the Lord" means to ratify or reaffirm unconditional trust in God's covenant promises.

Look, maybe we're on a wild-goose chase here! If religious belief can't be defined as belief in a god or a supreme being, as belief that induces worship or ritual, that sanctions ethics, or concerns what is valued most highly, what can work? There are such wide differences among religious beliefs that I don't see any way to pick out something they, and only they, share in common. Perhaps religious belief can't be defined this way at all, and we should try some other way to delimit it.

That suggestion puts you in distinguished company! Around sixty years ago a number of scholars of religion tossed in the towel, and gave up on any essential definition of religion at all. ⁵ They were attempting a slightly different task, of course, since we're concerned only with religious belief and they were trying to define religion as a whole. But their conclusion - that religions have only family resemblances - is one they would surely apply to our quest too. And it's easy to see why they felt driven to say that.

Suppose, for example, we were to reply to them that every religious tradition regards something or other as divine. That seems true enough but not very enlightening; it simply shifts the problem to defining the term "divine." How, they would ask, can we find a common element among the ideas of divinity, even if we confine that search only to the major world traditions of the present? What common element can be found in the biblical idea of God in Judaism, Christianity and Islam, in the Hindu idea of Brahman-Atman, in the idea of Dharmakaya in Mahayana Buddhism, and the idea of the Tao of Taoism? To isolate a common element among just those few seems daunting enough, but if we could do it we would then also have to be able to discover that same common element in all the other ideas of divinity: those of ancient Egypt, Babylon, Palestine, and Greece; the divinities of China and Japan, of the Pacific islands, of Australia, of the Druids, and of the tribes of Africa and North and South America. So, they would ask, isn't it painfully obvious that there is no common feature to the divinities of all these traditions, and no single belief they all share in common?

Part of their point I agree with. If we look for a common element in what each tradition regards as divine, then it's true that the natures of those divinities are so diverse as to have no feature they all share in common. But I now want to suggest there is another way to go at the matter, a way that I think succeeds in finding something common to all religious beliefs. Suppose that instead of looking for common features among the natures of all the various putative divinities, we were to seek the common denominator in the status of divinity itself ?

Since this seems to be an important point for you, I want to be sure I'm understanding it. Can you be a bit more precise about the difference you're pointing to?

The difference between these two approaches is like the difference between two possible ways we could answer someone visiting the United States who asks, "Who is the president of this nation?" We could answer by naming the person who holds that office, then going on to describe that person by giving the president's gender, height, age, build, and other distinguishing features. Or we could answer by describing the responsibilities, powers and limitations of the office of the presidency. The difference is important. Even if an election were in dispute so that people did not agree on who had really been elected president, they would still be in agreement on what it means to be the president. In a parallel way, it is possible that although the ideas of who or what is divine are so diverse as to have no common element, there could still be common agreement among all religions as to what it means to be divine. If this were the case, the disagreements among religious beliefs would be disagreements about what it is that has divine status; they would all still agree about what it means for anything to have that status.⁶

Now this is exactly what I have found to be the case. In all my reading and study, I have never found a single exception to this: In every religious tradition, the divine is the self-existent reality that produces whatever is not self-existent, so the divine is that on which all nondivine reality depends. Put another way: the divine is whatever has independent reality and all that is not self-existent depends on the divine. (Logically speaking, the cleanest way to put this is to say that “self-existent” means unconditionally nondependent.)

Please do not misunderstand this point. I am not saying that there are no disagreements whatever about what defines the divine status. There are. But although there are disagreements over what else may be true of whatever is believed to have divine status, all ideas of it include non-dependence. Nor am I saying that every myth or body of teachings has used the expression "non-depen­dence" or its equivalent. Some religions trace everything nondivine back to an original something, the status of which is neither emphasized nor ex­plained. But in such accounts that original something is still spoken of as though it has independent reality: there is nothing that it is said to depend on. Thus, at the very least, that something is tacitly given non-dependent status; it is non-dependent so far as the teaching goes.

Does this definition cover every known religion, though? It sounds plausible, but so did the others till a few minutes ago.

It seems to me that it does succeed. For openers, it can locate a common element among beliefs in the biblical God, Brahman-Atman, the Dhar­makaya and the Tao, which was the brief list that appeared so daunting just now. Moreover, it also covers such ideas of the divine as Nam in Sikhism, Ahura Mazda (Ohrmazd) in early Zoroastrianism or Zurvan in its later development, the soul-matter dualism of the Jains, the high god in the myths of the Dieri aborigines, the belief in Mana among the Trobriand Islanders, Kami in the Shinto tradition, the Raluvhimba of Bantu religion, and the idea of Wakan or Orenda found among various Native American tribes. It also holds for the ancient Roman idea of Numen, and for Okeanos found in the myths of Homer.

I can't, of course, claim to know about every religion that ever existed or to know that there's no religion yet to be discovered that doesn't have this idea of divine status. But I can say that no idea of divinity I have ever come across fails to regard the divine as the nondependent reality that all the nondivine reality depends on.

Your proposal is clear enough and seems plausible. But as I said, so did several other definitions until a few minutes ago. Can this one really stand up to criticism?

I believe it can. For one thing, it simply avoids the sorts of difficulties we found with the other definitions (which is perhaps the reason it keeps getting rediscovered).⁷ More than that, it both covers and helps clarify some of the most important differences and unique features of various religious beliefs.

For example, it is well-known that in Judaism, Christianity and Islam there is but one God who is the only divine reality, so that God and divinity are identical. In these traditions, everything other than God is creation, and the creation is not divine. By contrast, however, many religions believe there to be a difference between divinity per se and the gods. They believe in a divine reality that is the source of the gods and goddesses as well as of humans and the rest of the nondivine world; for them every nondivine thing is partly divine. The ancient Greek and Roman myths are examples of this. For example, Homer called this divine reality Okeanos, whereas it was called Numen in ancient Roman religion. And there are similar beliefs in many other polytheisms, both ancient and contemporary.

This explains why the gods of these religions do not fit the definition just given for "divine." The gods of these traditions are called divine only in the sense that they have more powers than humans have; they do not have unconditional existence and produce all else. Their religious importance lies in their superhuman powers and in their being the ones through whom humans can safely approach and relate to their divine powers.

This definition also sheds light on how it can be that in certain polytheisms where there is more than one divinity or god, and where the divine and the gods are not identical, there are (as I mentioned earlier) idle or evil divinities or gods: ones that have no important relation to human affairs or are malevolent.⁸ Some scholars have puzzled over how belief in such divinities or gods could persist despite the fact that they aren't valued and aren't thought to do anything good for those who believe in their existence. The definition I’m defending makes it clear why such belief is possible. It is not beneficence or usefulness to humans that is the defining characteristic of divinity or of a god, but non-dependence that characterizes divinity and greater partici­pation in divine power that characterizes a god. In this way the definition allows for the possibility that a god can be either one on which nothing, or nothing important to humans, depends. The signifi­cance of this point is that it is a confirming feature of our definition that is able to allow for, and make sense of, the fact that idle or evil divinities or gods occur in certain religious traditions.

Yet another feature of religious belief that this definition accounts for is the large variety of ways the nondivine can be thought to depend on the divine.⁹ For example, there are religions that believe in two divine principles and understand every nondivine thing as partially dependent on both divinities. By con­trast, others hold that one range of nondivine things depends entirely on the first divinity while another range of nondivine things depends entirely on the second divinity. Still other religions believe in a whole realm of divine beings, thus increasing the number of ways these can be thought to relate to one another and to the nondivine world - including that some of them are idle or evil with respect to humans.

So it appears that this definition not only isolates a common element in all ideas of divinity but also allows for the great diversity of ways the nondivine is thought to depend on the divine.

No doubt belief in divinity is crucial to religion, but you now seem to have made it everything! Even if worship and ethics and values can't define religious belief, surely they shouldn't be left out altogether. Aren't such beliefs also religious in some sense - at least when attached to belief in a divinity?

Absolutely! We've already noticed there's more to religious belief than just holding something or other to be divine. I alluded earlier to that fact that beliefs about how humans can stand in proper relation to the divine are also religious. So the full statement of my definition is as follows:

              A belief is religious provided that it is:

Perhaps this helps make clear the main problem with the popular definitions of religious belief that I found to be faulty: they focus almost entirely on the third part of the definition. In a way, that's under­standable since those are the most obvious, public features of religious traditions: their creeds, rites, rituals, holy days, meditations, pilgrimages, codes of behavior and range of values. Important as these are, however, they are not the most fundamental components of religious belief or practice. They are secondary for the purpose of definition. What is fundamental is identifying what is divine, since every belief about how to stand in proper relation to the divine is determined by the nature of what is taken to be divine, and of how the nondivine is thought to depend on it.

OK, so you allow that beliefs about the right way to relate to the divine are also religious. This still seems to demote them somehow in a way that doesn't sound plausible. After all, doesn't the word "religion" literally mean "reconnec­tion"? Isn't it essential, then, that religious belief be about the human connection to the divine? And doesn't the whole business have to be more personal than you're making it?

Oh, I see what's bothering you. You are thinking more of actual, personal religious belief and practice than of the abstract task of defining it. In that case, you're right on target. I didn't mean to say that the proper relation to the divine is secondary in actual lived belief and practice, that’s true only for the purpose of definition. Besides, our relation to the divine is never without personal importance. Let me try to clear that up.

Identifying the unconditional nondependent reality is never just a matter of intellectual curiosity. Human beings cannot help but want to know about that on which everything depends, since that includes us. For the same reason, the divine is also what determines our destiny - however that is conceived. What's more, our idea of the divine strongly influences the view we take of human nature and that, in turn, influences what we believe about values and many other issues ranging over the entirety of our experience. For these reasons, a belief in something as divine is intensely personal. We can't help but think of our relation to it in ways that far exceed mere dependence.

This side of religious belief is no less important than identifying the divine. In order to get a definition, we needed to distinguish belief in something as divine from beliefs about how to relate properly to it, so as to see why the idea of the divine is logically basic to ideas about how to relate to it. But distinguishing these two points in thought is not the same as separating them in practice. From the standpoint of religious experience, belief, and practice, the idea of the divine is always encountered as embedded in a context of ideas about how to relate to the divine. Both are grasped simultaneously; they are intertwined and inseparable. So al­though it is true that (1) the divine is whatever the nondivine reality depends on, and (2) belief in something as divine is logically basic to beliefs about how we should relate to the divine over and above mere dependence, it is also true that (3) a belief in divinity is almost always acquired in conjunction with, and remains embedded in, beliefs about how to relate to it. As far as actual experience and belief are concerned, these are equally important.

Plausible as all that sounds, there is an objection that seems more plausible. It’s the very one you mentioned earlier: the definition makes beliefs that are not religious into religious beliefs, so it’s too broad.

For example, I know of some theories in philosophy and the sciences that assert or assume something as non-dependent. Matter, for example. But surely belief in the non-dependence of matter is not a religious belief, so your definition is too broad. If you try to say that it is a religious belief, then what difference is left between religion, philosophy, and science?

I am indeed saying that such beliefs are religious regardless of what is regarded as divine and regardless of the context in which they occur. To see why this is so, let's consider your example of materialism and use the teachings of a famous ancient materialist, Epicurus, as our example. For Epicurus, there is a distinction between the divine, which has independent reality, and what he called gods. In his theory, only physical atoms in space are self-existent; all things other than atoms are combinations of them. This includes the human soul and the gods; they too are combinations of atoms.

You seem to be willing to concede that Epicurus’ theory gives matter the same status that I called "divine" in religions, but you want to say that it's not a religious teaching despite that fact. Have I got you right?

Exactly! I want to say that beliefs in something as non-dependent are only religious when they're accompanied by other beliefs about how to relate properly to what is nondependent.

In that case, we need to notice why even Epicurus's theory could not avoid beliefs about how humans should relate to the divine. The most important element in the relation was, of course, simply to know that matter is divine. That knowledge then served as a guiding assumption to Epicurus's view of human nature, destiny, values, and ethics, since he clearly believed that human happiness depended on that knowledge. In all these respects, his teaching exactly mirrors religious traditions.

The same is true of modern materialisms, not just Epicurus' version. Most people think that being a materialist is the reverse of being religious. But why think that? Every materialism asserts the belief that what has independent existence is the “purely physical.” ¹⁰ That's a different idea of divinity from the ideas held by a Christian, a Hindu, or a Wiccan, but it’s still an idea of divine reality. Moreover, Materialism also implies a distinctive range of acceptable conceptions of human nature, of human destiny and of what can or can't be done to improve the human condition. And it includes an idea of human happiness. At the very least it prescribes that all other conceptions of the divine are to be rejected as false and that people will be better off by knowing materialism to be the truth. In other words, the secondary beliefs that religions emphasize (and which you say make a divinity belief religious) can't be utterly absent from any belief in divinity, even one that occurs in a theory. So even on the view you're defending, there is no denying the religious character of materialism's belief that matter is non-dependent. It's not religious only if its adherents sing hymns to force fields or offer prayers to quarks. It's religious both because it regards matter as the ultimate reality and because that belief implies other beliefs about human nature, happiness, values and destiny.¹¹ Needless to say, these implications are highly personal and so are just like the secondary beliefs found in religious traditions.

Can't we strip away the personal stuff if we just try harder? Why can't we just drop every idea of human destiny, for example?

Maybe materialists of the past were too influenced by the religions of their culture and allowed their theories to be larded over with religious trappings. Even if some of that has persisted into the present day, that doesn't show it’s unavoidable.

How will you strip away all the implications for human destiny? Doesn't materialism require that when we die we rot and that is the end of us? Isn't that just as much a conception of human destiny as the Hindu belief in Nirvana or the Christian belief in the resurrection of the body? It's just a different idea of human destiny. The same goes for ethics or the idea of happiness. Doesn't materialism require either that there are no distinctly ethical properties and laws or that they all must have a physical basis? Doesn't that make a difference to one's ethical beliefs? And doesn't materialism require a view of happiness that is very different from any view that presupposes nonmaterialist ideas of what is divine?

This is really weird! I've always thought of materialism as the reverse of religion. Now you're saying it is one!

Not quite. Materialism is a theory not a religion, but at the heart of that theory lies a divinity belief. It should be clear by now why thinking of atheism and materialism as the reverse of religion is narrowly culture-bound. It reflects the fact that the religions you are most familiar with denounce materialism and atheism. But the truth is that there have long been religions in which matter was believed to be divine. Some of the Greek mystery religions, for example, referred to the divine as the "ever-flowing stream of life and matter." There is still a branch of Hinduism that regards the divine (their term for it is Brahman-Atman) as matter.

But then is there no difference at all between a theory and a religion? This almost sounds as though everything is religion!

I'm not saying that everything is religion, or that there's no difference between a religion and a philosophical theory. I’m saying that beliefs in divinity occur in both. But such beliefs are used differently in theories than they are in religious traditions. In theories, they are used as the basis for constructing explanations about how the cosmos works. They do this by controlling how we conceive of the nature of the things philosophical and scientific theories propose. Religious traditions, by contrast, emphasize how believers can acquire the proper relation to the divine in order to obtain present happiness and/or an ultimate destiny not otherwise obtainable. That is an enormous difference, but it does not cancel the fact that belief in a divinity doesn't lose its religious character just because it occurs in a theory.

Perhaps this all looks more convincing than it should. Maybe the theories you picked as illustrations fit your definition better than others would have.

Do you really think that? I started with materialism just because it seemed to you, as it does to most people, to be the great counterexample to my definition.

All I can add here is something parallel to what I said about religions. I said I couldn't claim to have read about every single religion that ever existed but, that in all my years of study I've not come across one that lacks a belief in something divine at its core. (And it’s not just my own reading that supports this point. It’s also true of all those other experts in religion that I listed earlier). Likewise, although I can't claim to have read every theory of reality ever written, I've yet to come across one in which a belief in something as divine doesn't play a pivotal role in guiding the hypotheses that get constructed, adopted, or rejected. What's more, I can't imagine how any theory of reality could possibly avoid such a belief insofar as they try to identify the basic nature of reality, that is, the nature of what is metaphysically ultimate.

But surely most theories don't try to do anything as esoteric as "identify the basic nature of reality." Maybe I'm more upset about this point than I really need to be; maybe all you've shown is that one small set of (strange) philosophical theories may have some obscure involvement with religious belief!

It's true that most theories don't explicitly concern themselves with the basic nature of reality. Some even go out of their way to avoid anything that even sounds like that. But even if a theory can avoid all talk of whatever is non-dependent and generates all else, it cannot avoid presupposing some­thing as having that status, even if it is only assumed subconsciously.

Here’s another way to put the point I’m after. Suppose a theory uses a particular kind of facts or laws to explain everything whatever, while saying nothing about whether the explainers of that kind are self-existent. Then that theory would be doing what the myths do that fail to be explicit about what they say all else depends on. That is, they never explicitly say it is nondependent or self-existent. They just treat it as the explanation of the existence of everything else. In that case, the same point I made about those myths would apply to theories: whatever is presented as the explanation of all else is tacitly being given the status of divinity which is independent reality. Just as in the case of those myths, the ultimate explainers of such a theory don't depend on anything whatever so far as the theory tells us.

The most influential philosophical theories, however, have been very explicit about their candidates for divine status. A list of just a few of their more famous candidates would include numbers, space, space/time and energy, space and matter, matter alone, forms and matter, souls (or minds) and material bodies, sensations, logical laws plus matter, monads, and logical laws plus sensations. What is more, a great many scientific theories also presuppose one or another of these candidates for divinity, even if they don‘t specifically mention them. So the point I'm making has very far-reaching consequences. To the extent that any theory presup­poses a view of the nature of reality, it would thereby also be presupposing some religious belief. This would hold not only for the physical sciences but also for theories arising in other fields, for example, mathematics, politics, law, psychology and sociology.¹² Once again, these theories would not thereby be religions any more than theories of reality are. But they are unavoidably and importantly influenced by some religious (divinity) belief or other.

Your admission that your definition could have surprising and disappoint­ing results was an under- statement. I find this to be downright offensive. Instead of the theories of philosophy and the sciences being able to help us judge religious beliefs, you expect me to accept that theories are themselves controlled by one or another religious belief!

No matter how surprising or disconcerting this result may be, it doesn't show the definition to be mistaken. Remember that it has been derived from an enormous empirical base including every major world religion of the present, a great many influential religions of the past, and tribal myths from all over the globe, both past and present. It includes doctrines that have been given an extensive theological elaboration, those that are embodied in simple myths or stories, and those that are both. It can't claim to be based on my having canvassed every religion, of course, as I already admitted. But then it is rare that any definition can claim to have canvassed every instance of the type of things being defined, and no one supposes that we don't have a good definition of trees just because we haven't seen every tree. Besides, far from being based only on my own reading, it has been held over the last one hundred years by a number of distinguished thinkers who arrived at this definition despite having widely different approaches to the study of religion (they’re listed in note 7). The cumulative reading on which it is based – my own plus theirs - is therefore enormous.

What is more, the definition appears to have identified the only element that is even plausible as the common denominator in all religious beliefs; if that is correct, it is then the only feature that could define them. It should at least be obvious that the common element I've identified is not an oddity at the fringe of religious belief and practice, but is central to each in the view of each.

But simply being true of all religious beliefs is not enough to show it isn't also true of nonreligious ones, and that's what is at stake here. Surely this definition is too broad when it forces nonreligious beliefs to be religious.

Remember, the definition can justly be accused of being too broad only if it covers beliefs that are clearly not religious.¹³ So I ask: is it really clear that the beliefs in something as non-dependent that occur in theories are not religious? You suggested yourself that regarding something as divine would be truly religious only if it were accompanied by beliefs about how to stand in proper relation to the divine, beliefs that were personal and implied a way of life. But surely that's true of beliefs in divinities as they are expressed in theories, whether or not those who hold the theory draw out those implications. (Theories can have such implications whether or not those advocating the theory actually drawn out those implications.) So what reason can there now be for insisting that these implications aren't religious just because they don’t actually occur in many theories?

Isn't this objection a holdover from one of the popular views of religion? Or isn't it based on the assumption that there must be a specific set of ideas or practices about how to relate to the divine that are the only truly religious ones? We've already seen why that can't be right: there's no limit to which ideas or practices can be religious. Keep in mind that we’re not now talking of which of these beliefs is true, but only of whether they are religious.

This sounds to me like a great slander on the neutrality and reliability if science. It's bad enough to say some philosophical theories can't avoid religious assumptions, but to extend that to science is intolerable!

What you're saying now sounds like the objection that whales just couldn't have more in common with humans than with fish because it's disappointing or offensive to think so.

No, my reaction is not just emotional. There's still something about this definition that bothers me a lot -even aside from its consequences for theories. But I'm having a hard time putting my finger on what it is.

Take your time - this subject is not easy! While you try to get clarity on your point, I’ll offer one more reason in favor of my insistence that the definition isn't too broad.

Any belief in divinity, whether consciously asserted or unconsciously assumed by a theory, is the logical denial of every different divinity belief, no matter in what context the others occur. How, then, could any such belief fail to have genuinely religious meaning? If any particular belief about what is self-existent is true, then every other belief that disagrees with it is false. It wouldn't matter whether the others occur in cultic religious traditions or not. The others would all be alternative, competing divinity beliefs. Denying this is like claiming you alone speak with no accent: just as having an accent is any difference in diction, having a distinctive view about what is self-existent is having a different religious belief.

I’m now going to try to come at this from a different angle. I’m going to say that your position is not consistent with the biblical teaching about God. After all, it’s belief in God that we’re ultimately concerned with here, so this is an important issue: Is it really part of biblical teaching to say that God’s being the Creator is basic to everything else it says of God?

Just about every major Theistic theologian has thought so, and I think they were right about that because the scriptures also convey that idea. For example, when Moses asks God for his name - his self-identification - God tells him that it is “Yahweh.” The eminent archae­ologist and biblical scholar William Albright said that the best translation of this is "the one who causes to be."¹⁴ The prophet Isaiah makes the same point another way. He quotes God as saying, "I will not yield my glory to another" (Is 48:11 NIV). What is the glory that belongs to God alone? Isaiah answers the question in a passage that is well-known because it has become part of the Christian liturgy. Because it is so familiar, however, we need to notice that it has entered the liturgy in a translation that is not quite accurate. The familiar rendering is "Holy, holy, holy is the LORD God of hosts; the whole earth is full of His glory'' (Is 6:3 NKJV). Some Old Testament scholars have said that translated more precisely the last clause should read, "the fullness of the whole earth is your glory."¹⁵ In other words, the glory God will not share with any other is that He alone is the One who fills heaven and earth with creatures, the One on whom all else depends. Attributing that status to anything else is idolatry, says Isaiah, because it replaces God with a false divinity. Of course, by the same token, from other points of view God would be the false replacement for Brahman-Atman, the Tao, Mana or Wakan Beings, just as any of them could be viewed as false replacements for numbers, space, matter, sense data or logical forms.

This same point is explicit in the New Testament when it says that all people either believe God to be the sole and true divinity on which all else depends or replace him with something from creation that they regard as divine in his stead (Rom 1). And the point is repeated with specific reference to the famous "four elements" of ancient Greek meta­physical theories (earth, air, fire, water) when St Paul says that it is not these elements but God in Christ on which the entire cosmos depends (compare Gal 4:3, 8-9; Col 2:8).¹⁶ The point is that any belief in something as nondependent has genuine religious significance, if only because it is an alternative to God. And finally, the New Testament explicitly says that God is worthy to receive glory and honor because He is the One by whose will all things exist and have been created (Rev 4:11).

This is really hard to take. It makes out something I find repugnant as pervading all human experience - even down to corrupting theories of science.

I realize that many people have such a strong antipathy to the whole subject that there is nothing that would ever convince them to accept a definition that entails this sort of importance for it. A fellow graduate student once said to me, "Show me that any belief I have is religious in any sense and I'll give it up on the spot!" And a faculty colleague once commented to me that the only way religion should be taught is "with great hostility." It may surprise you to hear me say that there is a sense in which I sympathize with this attitude. The history of religious institutions has been such an abysmal panorama of bigotry, persecution, and cruelty that I can see why it could lead someone to wish to be rid of the whole business.

But as much as I can sympathize with those reasons, they do not count against either the definition or the importance of divinity beliefs. Failing to acknowledge the power and pervasive­ness of religious belief will hinder and not help the restraint of its worst examples. Please bear in mind that the definition defended here does not seek to praise religious belief as something good in itself, but only to recognize it for what it is. Some of the worst evils plaguing the human race have come from distorted or fanatical religious belief, but these cannot be corrected by pretending to avoid religious belief altogether.

Wait a minute! If this is right, wouldn't you then have to say that everyone who holds a theory that assumes a divinity belief is, therefore, a religious person? You can't want to say that! So surely your definition is too broad in at least this way.

What's more, if you try to escape this by biting the bullet and accepting that result, then you're going to end up with nothing more than a cheap trick that tries to make everybody religious overnight merely by definition! But who can really believe that everyone has become religious just by rearranging words on a page?

The definition does not perform the magic trick of making everyone religious in the sense you mean. There are indeed many people who completely reject the dominant religion(s) of their time and place. They adhere to no community, practice no worship or other rites and may have no conscious – let alone fervent - commitment about what is divine or about how to relate to the divinity. So they are not actively "religious" in the sense you're placing in opposition to my definition. You're using it to describe a person who has a conscious conviction about what is divine and joins others in adhering to a tradition that cultivates a life of devotion to the divine. There certainly are many people not like that. But that doesn't show that the people who are nonreligious in your sense are nonreligious in the minimal sense I'm pointing to. To be utterly nonreligious in my sense a person would have to lack even an unconscious assumption about what everything depends on. I think all people are religious only in the minimal sense of having some such belief, even if it remains an unconscious assumption.

Maybe what I said doesn't show that there are people who don't have any such belief, but nothing you've said shows they all do. Aren't there many people who just never think about this issue at all? Isn't it true that there are even scientists and philosophers who, if you ask them what they think everything depends on, would honestly answer that they haven't a clue? I have friends who, if I asked them how they’d answer the question “What is the self-existent origin of everything else?” would say, “Let’s have a drink and forget it!”

That's quite true. I was trying to acknowledge that point when I said that such beliefs could be unconscious assumptions as well as fervent convictions, and vague ideas as well as precise concepts.¹⁷ What I meant was that many people's beliefs as to what has divine status are very imprecise. They locate the divine in a certain "region'' of their experience without ever sharply formulating that belief. For example, they may have a vague sense that what everything depends on is roughly spatial, or physical, or is something like a mind, or is the universe as a whole. And they can do this without even articulating these ideas to themselves, let alone trying to refine them into sharply drawn concepts.

In theories that make explicit what is taken to be divine, on the other hand, great care is usually taken to elaborate the idea of divinity as sharply as possible. Often that takes the form of a hypothesis – an educated guess - about how best to conceive it. (This fact that has misled many thinkers to suppose that beliefs in divinity are always theories and therefore to be evaluated as we would any other theory). So it is important to notice that even where a precise concept of divinity really is a hypothesis, the original sense of divinity it tries to refine into a careful definition is not. For example, a materialist may be ever so tentative about how to conceive of exactly which purely physical entities have divine status, but will not budge from the conviction that it is purely physical entities that have that status regardless of how they are to be more precisely described.  

If we keep these distinctions in mind, then it seems to me quite plausible that everyone has some divinity belief. This is especially so if my point is right about theories presupposing something as divine. Virtually everybody believes theories on a wide variety of topics: the proper role of government, how to raise children, whether a particular behavior should be illegal, and what sort of education is best, for example. It's hard to see how any of these could avoid assuming some view of reality, and it's hard to see how any view of reality could avoid assuming something as the ultimate, nondependent reality. If that's right, then it's another reason to think that everyone has some religious belief, whether consciously held or not, and that the only plausible candidates for people totally without religious belief would be those who believe no theory whatever.

But let's not get bogged down on this point. It's not essential to what I want to say about religious belief that everybody has one. I think that’s true, but know of no way to prove it. By contrast, I do think I can show that no theory can fail to presuppose a divinity belief. We can come back to all this later if you like.

While you were answering my last question, I finally put my finger on what it is that's been bothering me about your claim that all beliefs in independent reality are religious, even when they occur in theories. And it's a decisive objection!

Even if it's true that belief in an independent reality is central to religion and also occurs in theories, and even if such beliefs have implications for the whole of life in both religious traditions and in theories, it still won't follow that the belief is religious in both cases. I can concede your definition of divinity and still maintain that when such a belief occurs in a theory, it isn't religious because in a theory it's justified by reasons and arguments, whereas in religion it's taken on faith. So even the same idea of divinity - say, that it's matter - would be religious if taken on faith but not if supported by rational arguments and evidence. In that case, all you were saying about the theory of Epicurus, about materialism, and about theories of reality and the scientific theories that presuppose them is defeated!

This is perhaps the best objection to my position, and I want to do it justice. As a result, my reply will have to have several parts, so we can only start it for now. In fact, the full answer is going to extend throughout the rest of our discussions.

To begin with, let me remind you that philosophers and scientists aren't the only ones who have given arguments for their beliefs about what everything depends on; religious writers and theologians have done that too. Do you really want to say that if someone were to defend his or her belief in God with an argument, then that belief would thereby become nonreligious? If so, that would mean that all the Jews, Christians, and Muslims who ever accepted an argument for belief in God have thereby rendered their belief in God nonreligious! Does that make any sense? Isn’t “non-religious belief in God” a contradiction in terms? And if you say it’s not, then notice your position also requires that those people are therefore not really religious Jews, Christians or Muslims. Is that even plausible? I think not.

That aside, there are powerful reasons to doubt that any divinity belief can ever really be justified by evidence and argument. These reasons (which I’m about to explain) show that in fact people don't have good arguments for their divinity beliefs, although they may think they do. If this is right, then either (1) every divinity belief – including those that occur in theories -  is either a sheer mistake or blind faith, or (2) there's some other ground for them. Since I started today by saying I think divinity beliefs are grounded in religious experience, it won't surprise you to hear that I don't think they're all sheer mistakes or blind faith. Their real ground is experience, whether or not arguments are appended to that experience. So the reasons I'm about to sketch briefly are intended to show you why divinity beliefs can't be justified by reasons and arguments.

The first of these goes back to what I had been saying about the way theories presuppose some such belie£ This was easy to show for theories of reality, but I went on to argue that other theories - including those of the sciences - do this too. The position I took was that the influence of religious belief extends to these other theories indirectly: theories of reality either explicitly assert or tacitly assume that something is divine, whereas other theories – including those in the sciences - presuppose some view of reality. In this way, religious beliefs directly impact theories of reality and indirectly impact scientific theories via whatever view of reality they assume. The influence of divinity beliefs on scientific theories is therefore a two-step affair: a scientific theory presupposes (or includes) some view of the basic nature of reality, and every view of the basic nature of reality presupposes (or includes) some divinity belief or other.

I now want to apply that same point to theories of knowledge. There are many views of the basic nature of knowledge, and the differences between them lead to competing views of what it means to justify a belief. They disagree about the kinds of premises they regard as basic, as well as about what argument procedures they accept as convincing. The crucial point about these differences is that they appear to derive from assumptions about the nature of reality.¹⁸ In other words, what you see as divine guides your view of reality, and the view of reality you take makes a crucial difference to the view of knowledge you subscribe to. At the same time, your view of knowledge includes or assumes standards for what counts as rationally justifying a belief! The upshot is that attempts to justify a belief about what it is that has nondependent reality will be circular; they will subtly assume what they are trying to prove.

My second argument has to do with the process by which theorists have arrived at the various candidates for divinity that have appeared in theories. The slate of these candidates is usually acquired by abstracting them from our everyday experience. That is, we find that the objects of common-sense experience exhibit basic, large-scale kinds of properties such as quantitative, spatial, kinematic, physical, biotic, sensory, logical, and others. And we find that each of these kinds exhibits an orderliness among its properties, which the sciences assume to be the result of its laws. Almost all theories of reality have singled out some one (or two) of these major kinds of properties-and-laws, and propose it (them) as the nature of the independent reality that produces everything else.

Now there is a powerful reason for rejecting the possibility that any such selection can be justified. The reason takes the form of a thought experiment. This means that in order to grasp the argument you actually have to perform the experiment, and to perform the experiment you have to try to form a particular idea, namely: the idea of the theory's divinity belief. We do this to see if we can think of it being independent of everything else, which is the way it’s advocates claim it really is. For example, if someone proposes that the divine reality is the purely physical, then we’re going to try to form the idea of a purely physical thing.

Let me see if I've got this right. The experiment is to try to think of whatever a theory says has independent reality as though it really does. Is that it?

Exactly. We're going to try to conceive of whichever kind of properties-and-laws a theory takes to be the nature of the divine reality (the reality that exists independently from everything else) as though it really is independent of everything else. In short: If a theory claims that only the purely X kind of things can exist apart from everything else, we're going to see if we can so much as frame the idea of the purely X kind of things apart from all the other kinds. Fair enough?

I guess so. But where will that get us? I mean, suppose we can't? So what?

Well, if it turns out that you can't - that you can't even conceive of any of the major kinds of properties-and-laws apart from all the other kinds - then that's a powerful reason to think you can't give an argument in favor of thinking that any one kind is really independent of the others. (And if a particular kind of properties and laws isn’t independent of all others, it can’t be the cause of the others.) So the question is: how are you going to produce an argument or evidence to show something is a fact, when you can't even form an idea of the alleged fact? How can you produce evidence for the truth of a belief that is completely empty of content?

I think I get it, but I'm not sure. How about an easier example?

Ok, let's take Plato's quest to find what he calls "justice in itself," He says he's not trying to find out which actions or people possess the quality of justice, but that he wants to know what that quality is apart from anything that has it and apart from all other kinds of qualities. Try the experiment on that. Can you think of any meaning for "justice" when that idea is divorced from all other kinds of properties-and-laws? Try it. Think of justice; and then strip away everything having to do with number, space, matter/energy, biotic life, sensation/feeling, logic and social relations. What do you have left?

I've got nothing left. But what if I said I did? What would you say then?

I'd say that this argument would have no force for you, and we'd go on without it. The force of the argument lies in your own self-reflection. If you find you have nothing left to the idea of justice once it's stripped of all connection to other kinds of properties-and-laws, it shows you really have no idea whatever of that quality all by itself. Your idea is really always of justice-in-relation to other properties and laws. Therefore, so far as you can ever know, that's the way justice is. You have no idea of it apart from all other kinds of properties and laws, so even if it could exist that way you could never know that to be a fact.

OK, suppose that's right. How does it affect, say, materialism? Surely matter is more substantial than something as ethereal as justice.

Let's take materialism, then. The most widely accepted form of this theory claims that there are purely physical somethings that are the ultimate realities because they have independent (divine) existence. They produce all the rest of reality, no matter how the rest is thought of. Some versions of the theory think the rest of reality is also purely physical, other versions think the independent and purely physical somethings produce non-physical properties or even non-physical things. An example of the last type of these theories says there are such things as non-physical minds, but they are wholly produced and controlled by purely physical brains. Put another way: if there are any nonphysical qualities, laws, or objects, they are generated by the purely physical self-existent realities.

But can we actually form an idea of what it would mean for anything to be exclusively physical? Try it. Do the same thing you did for "justice in itself." Try to form an idea of, say, the chair you’re sitting in as a purely physical object. That means stripping from your idea of it every property that has to do with quantity and spatiality, every property and law of sense perception and logic, everything of language and of social relationships. Now tell me: what do you have left?¹⁹

Whenever I try this, I get nothing - literally.

I got nothing too. But then how could I have anything left? Without at least logical and linguistic properties, nothing can be so much as identified or named.

Exactly right. And that's why I get the same result no matter which of the traditional theories' candidates for divinity I test this way.

What, for example, is left of the idea of a purely sensory “datum” when it is deprived of all connection to number, space, matter, and logic? What, indeed, is left of our idea of logic without anything non-logical to which it can apply? Even its famous axiom of non-contradiction says that nothing can both be true and false at the same time in the same sense. Thus it includes an essential reference both to time and to other "senses" (kinds of properties, laws, things, etc.) to which it applies.

But you admit this doesn't show that reality couldn't actually be just physical or whatever. The experiment just shows we can't think of it that way.

Right. The experiment doesn't show that if a theory's candidate for divinity is drawn from the world we experience, then it couldn’t possibly be right. We weren't trying to see whether any of these ideas of divinity could be true, but whether any of them could be justified by argument. I think the experiment succeeds in undermining any way to argue for any of the candidates, even indirectly. Suppose, for instance, that instead of offering specific arguments that conclude with "X is what everything else depends on," a theory's backers argue that awarding divine status to X yields great explanatory advantages across the board. The trouble is that all the explanatory power they can point to will never recommend belief in X's independence, since any explanatory power X can have is derived from conceiving it as essentially related to all the other kinds of properties and laws rather than as isolated and independent of them! The same fate befalls any attempt to say that the primacy of X can be justified pragmatically. How can anyone claim that there are practical advantages to regarding the purely X-things as the ultimate realities, when no one can so much as frame the idea of purely X things?  

This seems to destroy the neutral rationality of theories, so I'm not about to accept it on the spot! It's deeply disturbing, and I'll need to think about it.

Meanwhile, I'm afraid I've forgotten where we were going with all this. Remind me how we got into this experiment and why it's important.

Let's recap. We started by asking what religious experience is, and that led to finding a definition for religious belief. You objected that my definition of divinity is too broad because it would mean that theories inevitably have a religious component because they argue for, or assume something to be, the divine reality. So you suggested that belief in divinity is not religious when justified by arguments and evidence rather than taken on faith.

I objected that this would make clearly religious beliefs nonre­ligious, and then briefly sketched two arguments to show that no one can ever really justify a divinity belief with arguments and evidence. People may, of course, think they have this sort of justification for their divinity belief, but they don't. And this opens the way to taking more seriously the possibility that divinity beliefs are grounded in experience. Thus it's all the more important to examine that experience, as I intend to begin doing at our next meeting. It's all the more important because even though there may be people who hold a divinity belief on blind trust or wishful thinking, it's not plausible to suppose that everyone has done so without exception. (What of the scientists, philosophers, and theologians who have spent years of effort and thought in investigating divinity beliefs? Are they all doing nothing but exercising blind trust?) It seems more likely that there's another source for these beliefs - especially as they occur in theories. In that case it's no longer reasonable to insist that divinity beliefs be proven. Taken together, these points strongly suggest that divinity beliefs have an experiential basis. But if that's so, there's also no reason to suppose that only philosophers and scientists ground their belief about what is self-existent on experience.²⁰ There may be many people whose divinity belief is held in a traditional religious setting but is equally grounded in experience.

OK, I'm back on track.

It's more than a little upsetting that you can put religious belief in general on par with beliefs in divinity that occur in theories. I'm going to think about that between now and our next session. But I can tell you that I'm going to keep looking for some way to distinguish the two. I don't think religious belief at large is reliable, but I don't want to demote theories to being equally unreliable.

Before we quit for the day, let me add one more thing in favor of the view that religious belief is based on experience. The experiential view is better able to account for why religion has always been one of the most widespread and persistent features of human life. So far as we know, humans everywhere and at all times have been drawn to the question of their origins, not just in the sense of how they originated (the processes that produced them) but in the sense of what it is they ultimately depend on. In fact, I would say that, on its subjective side, religion is an innate impulse in humans to direct themselves to what they take to be that which everything else depends on, and to understand their own nature in light of whatever they take that to be.²¹ So even without thinking up any arguments to justify these beliefs, people have in all times and places instinctively formed beliefs about what they, along with everything else, ultimately depend on. It is only within the last two hundred years that the idea has been entertained on a large scale that it is possible to be free of all religious belief and that people would be better off if they were. The massive incoherency in that claim is that the arguments for it assume views of reality and human nature that presuppose something as self-existent and thus divine.²²

But why insist on calling all beliefs in anything as nondependent "religious"? Why not say that when they occur in theories they are ''metaphysical" beliefs or some other less objectionable term? After all, you must admit that beliefs about what is nondependent that occur in theories have a different "feel" to them than those that occur in religious traditions.

Sure they do. That's because explicitly cultic traditions concentrate on showing believers how to stand in proper relation to the divine for their personal benefit, whereas theories concentrate on explaining how the cosmos works by relating all of the other kinds of things we experience to the divine kind. The difference, however, seems to me to be one of emphasis and not of exclusion, because theological doctrines also seek to explain the cosmos, and philosophical theories also have implications for how we should live.

Of course, we could refer to divinity beliefs by some other name. If someone was determined not to admit to having a belief that is religious in any sense whatever, there is nothing to stop such a person from calling a belief in ultimate reality "metaphysical" or from coining a new term for it. I'm not simply arguing for one word rather than another here. There's a sense in which it doesn't matter what you call such beliefs so long as you see that they are all contrary ideas of what it is that has unconditionally nondependent reality and produces all else, and that identifying that reality is central to all religions. In that case even if you insist that whatever you believe to be divine isn't religious for you, you'll have to admit that for all who hold other such beliefs and admit their religious character, your belief is going to appear to them to be clearly religious for reasons that are far from arbitrary.

Finally, there are reasons why I don't think "metaphysical" is a good term for these beliefs. First, because even when they occur in theories, they carry the same freight: the implications for human nature, destiny, values, happiness and so on. Second, because it would give the impression that these beliefs are a result of making theories or that they arise as hypotheses in the course of theory making. That's not true. Beliefs about ultimate reality can be refined and developed in the context of a theory, but they arise in ordinary experience whether or not we engage in theories; they are brought to theory making, not derived from it, and they clearly arose long before theories did. Beliefs about what is divine are not invented out of nothing in order to explain certain things we experience, but are formed in response to whatever is experienced to be divine. Missing this point is what's wrong with the popular idea that religious belief arose because primitive people were puzzled about this or that and so invented religion out of the blue to explain their puzzlements. That's not even plausible.

No doubt ancient people did invent specific gods, but in order to think of particular beings as bearers of divine power, those people would need to have had the idea of divine power already. In other words, inventing gods depends on prior religious belief, even if only of the vague type I mentioned earlier. Belief in divinity is not the product of theorizing but is one of many beliefs we bring to theorizing. It guides theorizing as one of its unavoidable presuppositions, and the history of theories bears witness to that fact. (Surely there's no question which came first historically: humans have held beliefs about what is divine for perhaps a hundred thousand years, whereas theories have only been around for the last twenty-seven centu­ries.)

In short, the fact that religious beliefs play a role in theories doesn't make them products of those theories.²³ The belief that there is a real world, the belief that other people have minds, and the belief in logical consistency also occur in, and are presupposed by, theories. But they were not invented by them. Like divinity beliefs, they are also brought to theory making from our pre-theoretical experience. They are all beliefs everyone already had - and needed to have - in order to live in the world as well as to construct theories about it.

I'm going to have to think about all this between now and our next discussion. Although I find your definition of religious experience unobjectionable, I'm still not convinced I should accept your definition of divinity. But since I can't think of a rebuttal right now, I think we should just go ahead; I want to hear the rest of what you're going say about religious experience.