David Hume on the Cosmological and Teleological Arguments

Hume's Dialogues Concerning Natural Religion were not published during Hume's lifetime. We don't know why he made the decision not to publish, but one speculation is that he was concerned that they would not be well-received -- that they would be too controversial. Whether that is true or not, they have become a standard work in the philosophy of religion. There are three characters in the Dialogues: Cleanthes, the defender of the teleological argument (the argument from design), Demea, the opponent t of anthropomorphism, who defends the cosmological argument, and Philo, who argues that none of the arguments for God's existence are valid. Philo is depicted as a believer nonetheless, believing as a matter of faith. It is clear to readers familiar with Hume's overall philosophical position that Philo's arguments are Hume's, though sometimes Cleanthes gives Humean arguments as well. Many readers suspect that the veneer of piety in the depiction of Philo is just that -- a veneer.

The Dialogues are considered by many to have provided a definitive critique of the argument from design.

Our excerpt opens with a statement of the argument from design by Cleanthes. The argument is based on an analogy and begins with the fact that nature is, as Cleanthes puts it, "one great machine, subdivided into an infinite number of lesser machine s." The crucial premise in the argument is that in these various "machines," the parts of things are fitted together to serve a variety of purposes in just the way that machines made by humans are. Think, for example, of any of our own organs -- the eye became a famous illustration a few decades after Hume wrote. The parts of the eye are intricately and exquisitely fitted to work together as an organ of vision. The resemblance between such "natural machines" and devices made by humans leads Cleanthes to argue as follows:

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. We might begin by noting that as Cleanthes states the argument, it is flawed from the outset. In general, very different causes can produce similar effects. For example: a piano and a digital synthesizer can produce the same tone. But the means of production production is very different in the two cases. However, in his restatement of it, Philo provides a version that is less subject to this criticism.

Philo's initial response begins with what might seem to be an irrelevant discourse about how we know what we know. In fact, this passage lays out some fundamental principles of Hume's own philosophy, and these principles have an important role to play in t he Dialogues. Hume's general idea is that there are two sorts of knowledge. One is of the relations among ideas, as he puts it. What this really means, more or less, is knowledge of the consequences of definitions. I know that all bachelors are unmarried. But I know this simply as a consequence of knowing the meanings of the terms. Knowledge of this sort never tells me anything about the facts of the world. Knowing that -- by definition -- all bachelors are married doesn't tell me who is a bachelor , how many bachelors there are, or even whether there are any at all. (Compare: all unicorns have golden horns.)

The other sort of knowledge is of matters of fact. This is the kind of knowledge we mostly want, and it is certainly the kind that is at issue in Cleanthes argument for God's existence. And Hume insists: all of this kind of knowledge is based on exp experience. His argument boils down to this. Take any particular matter of fact and put yourself in the position of someone who had no acquaintance with it. You would perfectly well be able to imagine that it wasn't so. For example, suppose you had never seen an egg. You could perfectly well imagine that you would be poisoned if you ate it. Or suppose, to pick one of Hume's favorite examples, that you had never seen one billiard ball strike another. You could imagine all sorts of effects other than the second ball's simply moving. You could imagine that the first ball would bounce back without moving the second. Or that it would change color. Or that it would turn into a turkey. And in any case, you could imagine that it wouldn't do what we know from experience that it will do: move in a more-or-less straight line.

The point about matters of fact is that you can always conceive of them not being so. And that means, according to Hume, that the only way you can know a matter of fact is through experience. A priori reasoning, by itself, will never settle a factual question.

What does this have to do with the argument from design?

Cleanthes argues from analogy. The basis of the analogy must be in experience. For all we could know a priori, things could organize themselves into exquisite arrangements. For all we could know a priori, if you threw a pile of boards and mortar together, it would turn itself into a house. Now of course, we know that this won't happen, but we know this by experience. On the other hand, Philo says, the ideas in our minds seem able to arrange themselves into orderly plans or thoughts without outside intervention.

Experience, therefore, proves that there is an original principle of order in mind, not in matter An "original principle of order" means a capacity for creating orer without outside intervention. And experience does tell us, or seems to, that minds have this power. Furthermore, it does tell us that in many circumstances, matter doesn't. So Cleanthes' argument seems to be off to a fair start.

Or does it? If you think about it, there is something puzzling here. It is true: if you simply throw matter together at random, you can't expect anything orderly to come of it. But on the other hand, matter does seem to have the capacity to organize itself. The very examples that Cleanthes sees as proving design are natural objects undergoing natural processes. Put a seed in the ground and, all on its own, it will turn itself into a flower. This is certainly how it seems.

In any case, this is not how Philo responds -- at least not initially. Instead, he points out that analogies become more and more doubtful the greater the difference between the things compared. From watching a hair grow, can we infer how people grow? he asks. And his basic point is that it seems extremely rash to take a very small part of the universe -- a part understood imperfectly -- and make it the rule for the whole universe. There are all kinds of principles at work in nature, many of which we have not even discovered, no doubt. As Philo puts it,

What peculiar privilege has this little agitation of the brain which we call "thought," that we take it to be the model of the whole universe? No mere argument by analogy could hope to establish that. Hume/Philo sums the difficulty up with a rhetorical flourish:

Admirable conclusion! Stone, wood, brick, iron, brass, have not, at this time, in this minute globe of earth, an order of arrangement without human art and contrivance; therefore the universe could not originally attain its order without something similar to human art. But is a part of nature a rule for another part very wide of the former? Is it a rule for the whole? Is a very small part a rule for the universe? Cleanthes offers his own reply to this argument, but before we go on to that, we might pause a moment. If we took Philo's reasoning completely seriously, where would it leave us? We do chemistry on earth. We conclude that matter even in the most distant p arts of the universe will have the same properties as matter on earth. How could we do that if Philo/Hume were right?

Even if we take Hume to be over-playing his hand, however, surely we can wonder whether we know enough about the workings of the universe to draw so large a conclusion from so limited an analogy.

Cleanthes offers a different objection, however. He asks us to imagine that a voice comes from the sky, addressing all people in their own language all over the earth, and delivering a message fit for a diety. Surely any reasonable person would conclude the at the voice is the voice of a being with intelligence and purpose. But according to Cleanthes, we could draw no such conclusion if Philo were right; all of his objections would hold against this case.

Alternatively, he asks us to imagine that there is a common language, and that in addition to being spoken by men and women, it is found in books that literally grow on trees. These vegetable volumes are reasonable, wise, and eloquent. If Philo were right , we would have to reject any inference from these natural productions to any sort of intelligence behind them. But as Cleanthes sees it, the world as we have it is full of things that are actually even more remarkable than this, since the structure of an a animal is much more complex than the reasoning found in the subtlest book.

Demea responds to this argument, but before we consider what he has to say, we should pause a moment. If we heard a voice from the heavens saying sane and sensible things rather then merely moaning, we would of course conclude that it was the product of some sort of intelligence. But it isn't clear how much good this does Cleanthes' case. To begin with, we don't have any examples of intelligent speech produced by anything other than intelligence. But more to the point, anything that can produce meaningful and intelligent speech comes close to being intelligent by definition.

The same point applies to the vegetating library. No matter whether there is a God or not, any process that produced the sorts of books Cleanthes describes would be a form of intelligence without further argument. Arguing, reasoning, and the like are paradigm cases of intelligent behavior. If they don't count as intelligence, nothing does. But however intricate an animal may be, its complexity is the kind that normally seems to get produced by brute natural processes alone. While intelligence sometimes produces similar things, it would be very hard to make the case that a process that produces complex natural mechanisms is automatically an intelligent process.

In any case, Demea has a very different objection. He argues that there is no sense to be made of attributing what we call thought to God. A mind as we understand it is a collection of diverse but separate faculties and abilities. Thought is a process that is changeable, varied, altering form moment to moment. But God is supposed to be unchangeable and simple. He is supposed to be fully present at all times in all places, and he somehow sees all there is in a single act. The idea that such a being thinks in anything like the way we do is incomprehensible.

This objection is had to evaluate. It is surely true that if there is a God, then God's thoughts are indeed not our thoughts. We can't even imagine what God's mind would be like, and it is very hard to know how terms like "thought," let alone terms such as "angry" or "loving" could apply to a being so very different from ourselves. The point -- and it is surely correct -- is that if we stretch terms too far, they simply lose their meaning. But even if talk of God's thoughts is ultimately metaphor, this doe mean that any metaphor would be as good as any other for what is sometimes called the "inner life" of God, nor that there is no rational structure within God.

In any case, Phil offers a new and much more challenging objection to Cleanthes' argument. If the complexity of organisms needs an explanation, so does the complexity of the thought in the mind of a thinker. As he puts it, "a mental world or universe of ideas requires a cause as much as does a material world or universe of objects." But if we set off on this path, we will never reach the end. We will end up positing an infinite succession of causes that never bring us any closer to satisfaction.

This is a serious objection. It is especially serious if we think of God in anthropomorphic terms -- as being essentially a glorified version of ourselves or our minds. Because in that case, everything that leads us to wonder how we got here should lead us to wonder in spades how God got here. The moral of the story: if you want God to be the cosmic creator, don't leave God in need of creation.

Demea offers the cosmological argument as an alternative. Since we have talked about this argument at length, we won't review the details here, except to say that Demea focuses on the need for a necessary being, as does Aquinas in the Third Way.

Now a necessary being would certainly solve the problem raised by Philo. We would have a natural stopping point in our search for explanations. We would know why there is something rather than nothing. But Cleanthes, this time, is the critic, and he will have none of it. And he takes a page from Hume's own book in making his argument: any matter of fact can be conceived to be otherwise.

Nothing that is distinctly conceivable implies a contradiction. Whatever we conceive as as existent, we can also conceive as non-existent. There is no being, therefore, whose non-existence implies a contradiction. Consequently there is no being whose existence is demonstrable. I propose this argument as entirely decisive, and am willing to rest the whole controversy on it. Now one reply to this argument is that necessity is not just a matter of what we can or can't conceive. Many philosophers would hold out for a concept of metaphysical necessity, which is not to be confused with logical necessity. Logical Necessity is a matter of definitions or relations of ideas, to borrow the term we introduced earlier. But metaphysical necessity is in the things themselves. And that, it would be replied, is what God has.

In fact, however, Hume takes this into account. He reminds us of the familiar argument that matter doesn't exist of necessity: just as we can conceive any particle of matter as not existing, so we can conceive of God as not existing. And if the rep reply is that there is some unknown feature of God that makes his existence necessary, we might as well say the same thing about matter. Or put another way, if there is something that could make God's existence necessary, for all we know it might make the existence of matter itself necessary.

There is a cheap reply to this argument. We really do know, in the simplest way possible, that individual bits of bits of matter don't exist necessarily. We know it because we can destroy them. But do we know that the totality of mass/energy is not a necessary being? Here it is harder to see how Hume's point can simply be dismissed.

Cleanthes adds one more argument. Some people complain that if there is no necessary being, then the existence of the entire causal order is left unexplained. In order words, if only the parts are explained, the whole itself is left unexplained. Cleanthes ' reply is that the "whole" is not a thing in its own right needing a separate explanation. The whole of the universe is like a country made up of different states. It is "an arbitrary act of the mind" that unites the states into a country -- that unites the parts into a whole. Once we have explained the parts, there is nothing left to explain.

Nothing except why there are any parts at all, you might reply. But Hume's response to that comeback is implicit in what has already been said, and so we will leave it at that.

Back to the Main Page