Separation by Foundation

We come now to the most popular separation principle of all.

Many religious writers have said that religion requires a 'leap of faith', that religious belief is religious precisely because it is held in the absence of proof or supporting observations; in fact, it may be held so strongly as to persist despite a great deal of contrary evidence. We might even say, echoing some of these writers, that a faith untested by reason is a faith not worth having.

In our examination of this most popular proposal, we must check on three things: first, we have to be clear on what it is to have a belief on faith; then, second, we need to ask why faith is thought to be essential to religion; and third, we need to make sure that the principle really marks a difference between religion and science. We have already found reason to think that the proposal is overstated. In our discussion of Scientific Creationism, we were reminded that inconclusive justification is the best that is even hoped for in science - so it does not seem to require proof. Also, given how indirect the connection between theory and observation is, given how remote observational evidence may be from many theoretical claims it helps to justify, there may be more room for observation in religion, and more similarity between it and science, than a distorted picture of science would allow. Let's drop the second part of Separation by Foundation and focus on faith.

The Evidentialist Objection to Religious Belief

It's useful to consider an old, familiar objection to belief in the existence of God. For the sake of simplicity, I will now narrow the focus to the God of Western theology. Let's begin with an analogy.

Consider the peculiar case of Mr.Maxwell Jones, multi-millionaire. Mr.Jones has become fantastically wealthy by playing the stock market. Beginning with an investment of $25, he has amassed a multi-million dollar fortune. How did he do it? What is the secret of his success? He offers to share his secret with you. "It's all in this book" he says, the Book of Fluctuations. The Book, he tells you, was a gift from beings from another planet, a race of supereconomists who amuse themselves by observing the economies of other worlds and finding the fundamental principles that govern them. They themselves are too rich to care about making more, and it pleases them to give whatever they find out to a Select Few on each planet they study. They came to Max one day in a vision on Cablevision, and soon after the Book arrived in the mail, post paid. Ever since, he's been on easy street. You are pretty skeptical - after all, all you've got is Max's word on this - but he is very rich, so you decide to take a look at the Book. What you find is deeply disappointing: a bunch of vague financial platitudes, open to a variety of interpretations. Instead of something really useful, like "Buy IBM at 104 on October 15, 1999; sell at 108 on October 23, 2001.", you find instead such statements as "Buy low, sell high", "Avoid excessively large risks" and "High tech stocks are risky during the end of the Second Millennium." All good advice, but not the detailed stuff of real get-rich-quick schemes. Max, you say, money's not the only thing you're full of.

As you first listened to Max, you had clearly in mind what appears to be an eminently sensible principle for the conduct of your cognitive affairs:

But you soon brought in another principle which seems equally worthy of assent:

Both of these principles seem eminently sensible. And they direct you not to believe Max's story. It's more reasonable to suppose that he's a rich nut.

Mustn't the same principles apply to the claims of religious believers? Surely, one might say, it is the height of irrationality to accept claims for which one has no evidence. And if the claims are both vague and extraordinary, as some of the Bible's claims are, isn't it your epistemic duty to demand a great deal of evidence before you believe what it offers? There is, I think, a case to be made for the extraordinariness of the belief that God exists. We are used to this belief, the overwhelming majority of us accept it, and even the atheists and agnostics among us are quite familiar with the belief. Suppose, however, that someone grew up in isolation from religious belief, and then enters our society. This would be very unusual but not, I think, impossible. Mr. Theistically Ignorant (luckily, his name is aptly descriptive for our purposes) is walking down the street one day, minding his own business, when you, acting on a deeply felt obligation to proselytize for your religion, introduce yourself to him and say, "I believe that there's one and only one omniscient, omnipotent, omnibenevolent, necessarily existent, incorporeal creator of the universe, and I want you to believe this, too." Mr. T would be within his cognitive rights to ask for some evidence. We'd not fault him if he responded, "Very interesting, but what justification can you offer me for that remarkable assertion." Indeed, we should fault him if he did not express some skepticism. This helps to motivate

This is an agnostic's favorite argument; atheists find it too weak. Coupled with the principle that one ought not believe what is irrational to believe, this argument would lead to the conclusion that one ought not believe that God exists.

There are two premises here, and both have been attacked. I am going to wait until Separation by Explanation to attack the second premise, though attack it I will. Here, the flaws in the first premise are relevant.

Is it irrational to believe where one lacks evidence? It is often irrational to do so. The story of Maxwell Jones reminds us of this. But there are homier examples, too. If I get it into my head to believe that wearing white socks leads to success, without having any evidence at all, then, surely, I am being irrational. But is this always the way it is? Let's consider a third example. I meet with one of the students about his test grade, and I come to believe that he is angry. "Why do you think he's angry?" you ask, in a skeptical and challenging mood What's your evidence? I respond by reporting the usual sorts of evidence: he shouted obscenities at me and made unkind remarks about my parentage, he waved a baseball bat in my direction, he hopped up and down, etc. This, of course, is usually enough to settle the matter; usually, no further evidence is demanded. But, pressing me further, you ask about my claim that he shouted at me. I respond that I heard him shouting, and he mentioned my name, and there was no one else around, and I remember all this clearly and vividly. "What makes you think that you heard him shouting?" you ask. I respond that it seemed then to me that I heard a voice emanating from the neighborhood of his mouth, and have a clear, present memory impression of that audio-visual perception. "What makes you think that that is how it now seems to you?" ask you. At this point, I get angry, hop up and down etc.. You have outrun my patience. "I don't have any further evidence," I say. "What's more, I don't need any: this is how things now seem to me to be; and for how things now seem to me to be, I need no justification. Beliefs like these are self-evident, are not the sort of thing about which I could be mistaken, so I don't need evidence for them. So your last request is not reasonable. Get lost before I hit you with my baseball bat." (I don't really have a baseball bat. I do have a crow bar.)

This story brings out two important points:

First, justification must come to an end somewhere. This is not just because we get tired, or our patience runs out. It is because there must be beliefs that need no independent justification, if any justification is to be possible. Such beliefs we can call foundational, because they lie at the foundation of all our knowledge and rational belief. It is on such a basis that all of our other beliefs rest: all the beliefs, that is, that are justified or known for us at any one time, all beliefs that we are rational in holding on to. It is because the 'higher' beliefs are linked in the appropriate ways to the lower ones that we can be justified in believing anything at all. This view is, naturally enough, called "Foundationalism," and has a long history, dating back about 2500 years. It is difficult to resist.

The second point of our story is this: whatever else the foundation includes, it will surely include beliefs about how things then seem to the believer. In order to get clearer on what this means, I want to say a bit about the most famous argument in philosophy. It was given a few hundred years ago by our friend Descartes. He himself did not buy the argument's conclusion, but he thought that it was useful for getting at some truths about Foundationalism.

First, however, I need to run a sanity check on you. I'd like you to bring to mind someone you really care about, whose well-being really matters to you. Now,what do you think of the following scenario: you arrive home one day to find your loved one (I'll write "she") deep in anguished thought. She says that she's now convinced that it's all just a dream. "Sure, it looks like I'm here on the couch, talking with you, but I'm really back in my room, fast asleep, in bed, dreaming it all up. And it's been that way for quite a while. You're just a figment of my imagination." Wouldn't you take that as an emergency, and try to get help for your loved one? After all, such a belief is obviously crazy. Anyone who takes it seriously does not deserve points for philosophical insight; they need, perhaps, a long rest in a quiet place. Consider a slightly different scenario. This time, your loved says she thinks it all might well be a dream - "maybe, maybe not - I'm no longer sure." Even this much of a loss of one's grip on the difference between dreams and reality is a sign that your loved one needs help. This, too, is a sign of serious mental malfunction. I hope all that seems right to you.

Philosophical Inventions - the First

Of all the sources of information about the world - testimony of others, memory, present use of the senses - the last is clearly the most trustworthy. Among our least trustworthy sources of experience are dreams. If we think that an experience is just dreamed up, we know not to trust it as a guide to how the world is then. But, Descartes asks, how can you now be sure that your present experience isn't just dreamed up? And if you can't be sure, then there is no reason to trust it as a guide to how the world is now. Since present sensory experience is rendered suspect in this way, and it is the most trustworthy source of information, it seems to follow that we can never know anything. Obviously, it would be cheating to rely on less reliable sources - memory and testimony - to shore up the most reliable source, so relying on those sources is 'against the rules'.

As Descartes observed, the first premise is overstated and so false. There are things about which you can be certain even if you are crazy enough to be convinced that your whole life's a dream. Even if you think that your just dreaming up this whole book - what a nightmare!- and so don't know that there really is a book that you are now seeing, you can still be absolutely certain that it now seems to you that you are seeing a book. In general, certain statements about how things seem to you to be now will be self-evident to you. Let's say

The Definition of Classically Self Evident Statement (always indexed to particular person and time)

A statement S is classically self-evident for me now =def. necessarily: if I think that S is true, then S is true.

Statements that are in this way self-evident you can make true simply by thinking that they are true, and so you could not be wrong about them. About the current, conscious contents at the surface of your own mind, at least, you are infallible - as long as you confine myself to mere appearance, and don't stick your neck out on matters of mental causation.

This definition is so strict that even your own body is beyond self-evidence. Consider, for example, your belief (indeed, your knowledge) that you have a nose.

The medical literature is full of case histories of people with nervous system malfunction so severe that they make mistakes about the presence or absence of parts of their own bodies. In phantom limb pain, an amputee may continue to feel pain as if it is in her missing limb, even though she can see quite clearly that the limb is gone. Some stroke victims lose all sense of ownership and all voluntary control over a limb, sometimes asking for its removal from the hospital bed because, after all, it must belong to someone else. It's as if the brain maintained a map of the body that can get out of synch with the actual state of the body.

So imagine that you've been in an accident, and you're just awakening from anesthesia after surgery. You take inventory: "Two legs, two arms, and, thank goodness, I still have my nose." But, despite heroic efforts, your nose was too badly damaged to save, and the surgeons were forced to do an emergency nosectomy. You are now noseless, though you don't yet know it.

The bare conceivability of such mistakeness about your own nose is enough to deprive your belief "I now have a nose" of self-evident status. From the perspective of self-evidence, even your own body is in the external world, outside your mind.

Among the foundational beliefs are surely the classically self-evident ones. But do any other beliefs deserve to be foundational as well?

Classical Foundationalism takes the highly - maximally - restrictive view that the classically self-evident statements are the only ones that belong in the foundation.

Classical Foundationalism thus sets the highest possible standards for the foundational and other beliefs: they must be wholly immune to doubt or error.

Classical Foundationalism has been the object of many attacks. Alvin Plantinga presents one of them in "Is Belief in God Properly Basic?" He says that Classical Foundationalism is self-referentially incoherent. His objection is an ingenious one. Consider the statement (CF), which is Classical Foundationalism itself. Presumably, its proponents hold that it is possible to know that their view is correct. But then, according to Classical Foundationalism, (CF) must be either self-evident or derivable with certainty from self-evident statements. It is clearly not self-evident, nor does there appear to be any hope of deriving it from a bunch of statements about how things now appear to me to be. So, according to Classical Foundationalism, it itself is unknowable or unjustifiable.

Of course if Classical Foundationalism is false, and we consider that fact, it would be irrational to hold it. Either way, Classical Foundationalism seems to be a view that it would be irrational to hold. It presents us with the very odd spectacle of a theory of rationality that says to you, "Don't believe me. Why, you'd have to be nuts to believe me!"

This argument, as well as common sense, suggest that Classical Foundationalism is too restrictive in setting irrationally high standards for rational belief: it shuts the door on any number of perfectly reasonable foundational beliefs that, although not classically self-evident, are fit for inclusion in the foundation. So, we might also object to Classical Foundationalism in this second way: There are many possibilities that might interfere with our justification. The dream hypothesis is one we've seen already. You might like to entertain yourself by imagining others. For example consider the possibility that you are a disembodied brain floating in a vat of nutrients, and that the Mad Scientist has stuck electrodes in your brain to make you think that you're embodied and reading a book. But unless you have some compelling and specific reason to take that possibility seriously in a given situation, it should not be allowed to interfere with your justification: some possible alternatives are normally irrelevant and (thought-) ought to be ignored. It is possible to get into abnormal situations - you suddenly are told by a reliable source that the water supply has been contaminated by hallucinogens, and you begin to wonder how much of the day really happened - but in such situations we typically know what to do to check on the abnormalities and to discover when they're no longer an issue.

We began with the question: do we always need evidence for rational belief? We saw that in some cases - the classically self-evident ones - independent evidence is not needed. We have also seen that knowledge and justified belief have a structure described by Foundationalism: some beliefs are justified because they rest on yet other beliefs, but there must be an end to requests for justification; they terminate in beliefs that need no independent justification. And we have discovered that the foundation must contain more than the classically self-evident beliefs. Let's call this, by way of summary, Common Sense Foundationalism. Common sense beliefs are, roughly, those that it would be truly insane to have doubts about in a given situation. Although they are less than classically self-evident, they nevertheless qualify for foundational status.

What has all this to do with faith?

To have faith in something is to make it foundational. It is not because we can provide further evidence for them that we have faith in them. It is because they need no independent justification. In this way, faith can gain intellectual respectability by associating with all those other commonsensical beliefs, like "I have two hands" and "there's a page I'm reading now". For, as we have already discovered, the foundation must be made thicker if it is to sustain the weight it needs to bear, and we're going to have to accept some such commonsensical statements without justification anyway. Plantinga suggests that among the beliefs in the thickened foundation of a theist may be some religious beliefs; e.g., "God forgives me", "God is to be worshiped", "God has created the universe and all around me", etc. As he puts it, belief in God is "properly basic". Let's call his view "Theistically Thick Foundationalism".

Of course, on this view, one may not be able to give an airtight proof that God exists - but the same can be said, for example, of your belief that there are people in the room with you, or that you are now reading a book. There is little to be done, and normally nothing that needs to be done, against the Dream Hypothesis except to assume that the properly basic beliefs somehow get beyond it, and this supposition seems eminently rational. So the Evidentialist Objection to Belief in God turns out to have a false first premise. In some cases, believing without proof or evidence is quite rational, and belief in God may be one of them. Plantinga offers the theist a rather clear way to characterize faith and to make it intellectually respectable. He can thus explain why the numerous attempts to prove the existence of God are not necessary: they are attempts to prove what needs no proof. He even gives believers a way of making good use of religious experience: statements of such experience can be properly basic as well, fitting as they do with similar beliefs in a theistically thick foundation.

Plantinga's view is not without its problems, and, characteristically, he is among the first to point them out. The main problem is, how does one rationally decide in general what to include in the foundations? There must be some constraints, unless 'anything goes' - and that would be irrational indeed. On what principled grounds can one charge an adherent of Vegetabilism, or Mr. Maxwell Jones, of cognitive irresponsibility, while keeping foundationalism safe for Western theology? What reason can be given for preferring the Bible to the Book of Fluctuations, or to Vegetabilist scripture? Plantinga's response is about three books long, wholly appropriate given the depth and difficulty of the question. But since this difficulty does not threaten Plantinga's proposed characterization of faith, I leave it to you to read what Plantinga has to say.

Now that we know what faith is, we can evaluate Separation by Foundation, and we can see quite clearly that it is false. If accepting something on faith is including it in the foundations, then it is clear that everyone accepts a great deal on faith, scientists doing science included. They do, of course, accept the testimony of their fellow scientists on many matters - not every claim made in the journals gets checked before being accepted. Even in those cases where checks are made, where, for example, experiments are repeated, scientists do not allow themselves to be taken in by Classical Foundationalism, not even when they are doing perceptual psychology. We would rightly think a scientist crazy if he began to raise serious questions about the existence of his laboratory or co-workers. Science may move beyond common sense, but it has no alternative to starting with it. Scientists, like the rest of us rational believers, have no choice but to accept a great deal without proof or justification. So accepting things, even important things, without proof, marks no difference between science and religion. It may be that what is accepted in this way, on faith, is different in science and religion, but that's tantamount to admitting that Separation by Foundation is false (-since it says that it is how statements are accepted, not what they say, that matters), and the difference lies elsewhere. Perhaps it does. Many theists have thought that the best response to the Evidentialist Objection is to refute its second premise by producing the evidence. We will consider what they have to say next.

Additional Sources

2000 David F. Austin

Go to next section Separation by Explanation