Separation by Explanation

In considering the Evidentialist Objection to the belief that God exists, I presented the evidentialist premise that there isn't adequate evidence for God's existence, either because there is no such evidence, or because there is some evidence, but that it is not sufficiently strong or compelling evidence. I want now to look at a direct response to this premise, one of the most famous and widely honored arguments for the existence of God: the Argument from Design, or Teleological Argument. ("Teleological" comes from Greek roots meaning, roughly, purpose-directed.) According to the various versions of this argument, there is compelling evidence of God's existence because we can see God's purposes and design in many parts of the universe, and in the universe itself, considered as a whole. We've already met this kind of argument in Plutarch's Purpose Argument, concerning extraterrestrial life, but it's so important, it's worth going over in more detail. Probably the best discussion of the argument in a religious context was given during the late 18th century by the British philosopher and historian David Hume, in his beautifully written Dialogues Concerning Natural Religion.

Look round the world: contemplate the whole and every part of it: You will find it to be nothing but one great machine, subdivided into an infinite number of lesser machines, which again admit of subdivisions to a degree beyond what human senses and faculties can trace and explain. All these various machines, and even their most minute parts, are adjusted to each other with an accuracy which ravishes into admiration all men who have ever contemplated them. The curious adapting of means to ends, throughout all nature, resembles exactly, though it much exceeds, the productions of human contrivance; of human designs, thought, wisdom, and intelligence. Since, therefore, the effects resemble each other, we are led to infer, by all the rules of analogy, that the causes also resemble; and that the Author of Nature is somewhat similar to the mind of man, though possessed of much larger faculties, proportioned to the grandeur of the work which he has executed. By this argument a posteriori, and by this argument alone, do we prove at once the existence of a Deity, and his similarity to human mind and intelligence.

[Only the paragraph above is required reading from Hume's Dialogues.]

We can gain a better appreciation of this argument if we begin with bits and pieces of the universe, and work our way up to the universe itself.

Recently, I was on a walk in the woods with my beagle, Snickers, and my friend, Skeptical Sam. (Perhaps you already suspected that I have an odd social life.) At one point in our perambulations, Sam tripped. I retrieved the object over which he tripped and examined it for a few minutes. I noted that it had many parts, among them a viewfinder, a lens, a focusing-ring, various buttons and levers, etc. After a few minutes of non-expert examination - I'm no optical engineer - I was able to see the soundness of:

This seems a perfectly reasonable way to argue. Precisely parallel arguments could be, and are, made about any number of other familiar objects.

Living up to his name, and feeling pretty annoyed about having been tripped, Skeptical Sam objected in three ways to the argument:

"See this lever on the right. It's twisted and no longer works. The camera has a broken part, so it is not likely to have been designed." But this is not a good objection; perhaps the object was broken in the factory, or in shipping, or by being abused by its previous, obviously quite careless, owner.

"There's a little black button here on the side of the lens. Push it, and nothing happens. The camera has a useless part, so it is not likely to have been designed." This is not a good objection, either; perhaps there was an error in the manufacturing process, or it's the designer's idea of a joke.

"It's all just a cosmic accident. Camera molecules drifting through the atmosphere happened to glom together here on the forest floor. The camera came to be as it is where it was purely by chance, so it is not likely to have been designed." Ha! This is the worst objection of all; the possibility it mentions (if it's even a possibility) is so unlikely that we need not take it seriously.

The Camera Argument is safe from these objections, and appears to be a perfectly good way to reason.

The 18th century theologian, William Paley, wrote that "the surest cure for atheism is contemplation of the human eye". (Natural theology, or, Evidences of the Existence and Attributes of the Deity) The eye is certainly a marvelous object, in many ways superior to cameras. To cite a very few respects in which it is superior: color distortion is a major problem in lens design, costing huge sums of money and lots of computer time, but eyes are virtually free of chromatic aberration. The very young daughter of a friend once decided to take some pictures in her wading pool with his very expensive 35mm camera, so she immersed it in the water and began clicking away. When she was finished, she gently toweled it off, and put it back on the shelf - where its guts rusted. Unlike cameras, eyes are safely submersible in water. If it's the right kind of water, it's even good for eyes to be under it. Unlike cameras, eyes are to a remarkable extent self-repairing, as I discovered after my daughter poked me in the right eye when I carelessly got too close to her waving hands while changing her diapers one day twelve and half years ago. The emergency room doctor told me that my corneal abrasion would heal within hours, as it did, and that they saw several fathers every week for similar injuries. So not only are eyes similar in purpose to cameras, they do a better job in many ways, and are more complex and more highly, even hierarchically, organized. And the Camera Argument appears to run every bit as smoothly if we replace "camera" in it by "eye".

Notice also that the three bad objections that Skeptical Sam made to the Camera Argument are every bit as bad as objections to the Eye Argument. Some eyes are 'broken' (malfunction), but that does not show that they aren't designed. Even if there were useless parts on some eyes - a bit of tissue dangling by the side, doing nothing for your appearance - that would not be evidence against design. And, of course, it is preposterous to suppose that eye molecules, drifting through space, coincidentally came together in the two holes so conveniently located in your head. We can be confident that the eye did not come to be as it is by chance.

"Nature has made the hindmost parts of our body which we sit upon most fleshy, as providing for our Ease, and making us a natural Cushion. . . . [S]o that it is manifest that a Divine Providence strikes through all things." -Henry More (An Antidote against Atheism, Chapter 12, sections 8-9).

Once we see this kind of argumentation, it is not hard to extend it to larger, more complex objects. Wherever there is the right sort of resemblance or similarity between objects that we know to have been designed, and various 'natural' objects, we can infer the existence of an intelligent designer.

In this case, conjectures about the relevant purpose, or purposes, or harder to come by, but, as I pointed out when we discussed Plutarch's Purpose Argument, knowing which purpose(s) is not essential for the argument to have force.

One common sort of objection to this argument and the Eye Argument is that they ignore cosmology and evolutionary biology. We have alternative accounts of both the evolution of the universe and the evolution of the eye, the objector says, and so we need not suppose that the eye or the universe are designed; in fact, we know that they were not, since we have these correct, alternative scientific explanations. This sort of reply is rather odd. For it might seem that far from discrediting these purpose-directed arguments, they provide more support for them. Although these arguments are stated as arguments about particular objects - eye and universe - precisely analogous arguments could be made about processes. We often do find it convenient to describe complex processes by breaking them down into parts, usually called "stages". We can therefore observe about the evolution of the eye and of the universe that they are production processes that resemble the production processes of human manufacturing. Far from detracting from these arguments, this objection can be reinterpreted to yield additional evidence for the existence of purposefully directed patterns in nature. It is somewhat puzzling that believers do not react in this way to evolutionary theory, and instead sometimes react with fear. What keeps them from saying that science describes the production process in exquisite detail?

[The claim that 'there must be something that started it all' is the basis of a less compelling argument than these purpose-directed ones. Such cosmological arguments throw away information about natural patterns and processes, and try to scrape by with the mere fact of the existence of the universe, ignoring the many subtle complexities that they could instead put to good use.]

But there is a yet stronger version of the argument from design, presented by Plantinga. In order to state it, it will be useful to introduce a standard way of stating such arguments about likelihood. Such arguments are called "inductive arguments", and the arguments from design are among them. Let's consider a simple example. Suppose that I stand outside a gold mine, examining chunks of ore as they come off the conveyor belt. I have examined the first 1000 chunks roll off the belt, and have found them to be gold; I conclude that the next chunk to roll off the assembly line will be a chunk of gold. Let C be the class of all conveyor belt items, and G be the property of being gold. Now every item to come off the belt about which I have the relevant knowledge is in fact a chunk of gold. So

There are three fundamental components in this sort of inductive argument: a reference class, a sample property, and a sample class. Here, the reference class is C, the class of all conveyor belt items; G, the property of being gold, is the sample property; and those members of C which we have examined and know whether or not they have G, constitute the sample class. So what we do is to examine members of the reference class to find out whether or not they have the sample property; the items thus examined form the sample class. If every one of the items about which you have made a determination has the sample property, then you have strong support for your claim that the next member will, as well.

In the case of the Argument from Design, the reference class is the class of all things that exhibit 'curious adaptation of means to ends' in Hume's phrase (A), and the sample property is being the product of intelligent design (D). We do not know of any member of class A that it does not have D: even if we knew that a thing was not designed by human beings, we do not know that God did not design it, unless we already know that God did not design the universe (in which case it would be pointless to consider the argument from design). Hence we have:

or, using some obvious abbreviations,

This argument gives a partial answer to the second premise of the Evidentialist Objection to Belief in God. Its claim that there is no evidence for the existence of God is false. But this is only a partial answer. The second premise also claimed that even if there was some evidence, it wasn't adequate to the job at hand: establishing the existence of a rather special sort of being: the one and only one omnibenevolent, omnipotent, omniscient, necessarily existent, incorporeal creator of the universe, God. And here, of course, is where even the stronger version of the Argument from Design falls down. Perhaps that argument boosts the likelihood that the universe has property D - being intelligently designed. But given the nature of the reference class (A) and the sample class (things definitively examined for being D), we do not have differential evidence for

there is just one designer of the universe

over

there is more than one designer of the universe

or for

any universe designer is omnibenevolent

over

any universe designer is sometimes good, sometimes malevolent

-and similarly for most of the defining properties of God. In short, the Argument from Design does not provide adequate evidence for the existence of God, though it may give us some reason to believe in the intervention of some sort of purposive agent or agents in the evolution of the universe and human beings.

As interesting or distressing as this result may be, it does not matter all that much for our discussion of separation principles. Finding a correct such principle does not require determining the truth of religious and scientific claims. At most, it requires that we find some patterns in their meanings. To this our previous discussion is relevant. It prompts the following:

The kinds of explanation contrasted here are quite familiar to us from a variety of contexts. (We met them when we talked of UFOlogy, and I'll say more about them in The Meaning of "Life.") Purpose-directed explanation is a resource we use every day; indeed, we couldn't survive long without it. Causal explanation is also familiar to us from laboratory science. These two kinds of explanation are very different.

In contrast to previous Separation Principles, Separation by Explanation gives proper emphasis to the fact that many religions try to give explanations of why things are the way they are. A religion that did not at least attempt to give us some such understanding would seem, I think, rather empty and unappealing. The present proposal also suggests that conflict between religion and science is at least a possibility, especially if they both attempt to explain the same phenomenon from opposed standpoints. The essential use of some notion of purpose- or goal- directed action is undoubtedly a striking feature of many of the explanations that almost all, perhaps all, religions offer.

The problem here is one that we've seen before, in our discussion of Gilkey's criticisms of Scientific Creationism. Religion does not seem to have a corner on the market in purpose-directed explanation. Two branches of science - biology and psychology - make heavy use of purpose-directed explanation. Texts in both subjects and at all levels are full of such explanations. And it certainly seems that they could not explain what they do without being allowed to use purpose-directed explanation. Purpose-directed explanation seems essential to both biology and psychology. Or is it?

Let me address this issue in a slightly different way, by describing an unlikely circumstance. Suppose that, two hundred years from now, an article is written for the Physical Review in which the last word on cosmological issues is stated: it had been confirmed, after much investigation and testing, that the Big Bang was caused by several very powerful beings who, working together, designed and caused it. They surveyed the many ways that the universe could be, and decided, in the interest of propagating more of their own kind - sentient beings - to select this one, with its finely tuned physical constants and four finely balanced fundamental forces. That's how it was. The popular press, in its typically irreverent fashion, calls these beings the "Big Bangers." Their old label was: minor deities; and after them, physicists call this theory "Deistic Cosmology."

I am not saying that such a theory is likely to be correct. As monotheists and atheists alike will agree, this particular theory is pretty unlikely. The point I want to make is not about its likelihood, but about its status as science. As unlikely a candidate for truth as it may seem, let us ask if there is any reason not to allow it scientific status. There seems little doubt that equally peculiar theories have been proposed. For example, Francis Crick, one of the world's leading theoretical biologists, Nobel Prize winner and co-discoverer of DNA's structure, has suggested the hypothesis that life on Earth developed from the crumbs left behind from an extraterrestrials' picnic. Being weird or unlikely is, we know by now, not enough to render a hypothesis unscientific.

If there is resistance to Deistic Cosmology, it is most apt to come from the same source as Separation by Explanation. Here is how the resistance might go: there is nothing unscientific simply in talking about living or thinking things, what they do and why they do it. It is alright to talk about such matters in science. But in order for an explanation to be scientific, it must rely on causal explanation alone: any apparent reference to purposes or goals must be shown to be mere shorthand for a deeper, more revealing causal explanation. So if we are to talk about the Big Bangers, then we must explain what they have done in a non-purpose-directed, causal manner if our explanation is to be scientific. And so it is in biology and psychology, as well: all their talk of purposes and goals is just misleading talk: a cover for current ignorance about how to translate the purpose talk into talk of natural laws of cause and effect. Can we reduce biology and psychology to physics, as chemistry has (apparently) been reduced to physics? This leads us to consideration of our final questions: What conceptual resources are essential in biology? Is a science of the mind even possible, and if so, what conceptual resources does it need? Until we have a better handle on their answer, we won't know whether Separation by Explanation keeps its promise.

Additional Sources

2000 David F. Austin

Go to next section The Mystery of Our Being: Limits of Causal Explanation?