Chapter 2


In any case, let us take our theoretical problem and look at it carefully.

An effect is the set of all information--and only that information--directly relevant to an apparently contradictory situation.

This is rather carefully defined for reasons we will go into shortly. For now, simply note that we have just changed the terminology; what we were calling a "theoretical problem" we are now calling an "effect."

The reason, of course, why it is legitimate to call a theoretical problem an "effect" is that effects are what have causes, and the cause "accounts for" or "explains" or is "the reason for" the effect. This ordinary way of thinking clearly implies that an effect doesn't make sense by itself and needs the cause to make sense out of it--which, of course, means that the effect, taken by itself (without the cause) would be nonsense, or a contradiction. And this is just exactly what our "theoretical problem is": a "contradiction-unless..."

In fact, you might say that this was the original notion of effect, in Plato and Aristotle, since what is translated "cause" is in Greek aitia, which means "what is demanded," and, as Aristotle mentions, is the answer to the question "Why?"

He also points out that not every question that sounds like a "why-question" is a real "why-question." When the "reason" is a definition, the question was really a "what-question" in disguise. For instance (to put it in modern terms), if you ask, "Why is blue blue and not green?" and you answer, "Because it is electromagnetic radiation of this particular wave length and not that one," you are not really giving the reason blue is blue, you are simply saying what blue is as opposed to green. In fact, there is no reason why blue is blue, because it is simply a fact; and facts as such are intelligible (and therefore don't need reasons for them, or explanations).

So the first thing we can note about an effect is this:

Conclusion 1: A single fact of itself cannot be an effect: there must be at least two facts in conflict in order for there to be an effect.

That is, there must be something to make the mind say, "This cannot be the case, because it contradicts itself," and which motivates a search for the missing information. You don't have a problem unless you have facts that fight each other, each indicating that the other is not a fact.

Thus, we must be careful in asking "why-questions" or thinking in terms of effects and causes. "Why is there something rather than nothing?" is supposedly a question that has bothered philosophers for centuries; but what is it about "There is something" that would lead one to say "There ought to be nothing," or "I have evidence to say there is nothing."

To put it another way, you can retort to the question "Why is there something rather than nothing?" by simply asking, "Why shouldn't there be?" and the original poser of the question has no reply. There is no reason, perhaps, why there shouldn't be nothing at all; but there is no reason why there should, either; there is precisely no reason for this fact either way. Hence, the question is an idle one.

Note that (as we will show) it is not idle to ask, "Why is it that there is more than one something?" This is legitimate, because it can be shown that if there are two realities, each of them is reality, and yet they are different from each other (implying either that they contain "non-reality" which makes each different, or that the "reality" in one case both is and is not the same as the other). Here there is a conflict in what is known. Whether the conflict is due simply to misreading the evidence or there is real evidence is something we will spend a good deal of time on; but the point is that, though there being something-or-other makes sense by itself and needs no cause, there being more than one something at least seems not to make sense, and hence is a problem and needs a solution.

So to demand reasons or causes for everything is to misunderstand what is going on when we demand a reason for anything; and if we are going to do philosophy and not just wander off in an apparently profound intellectual fog, we had better get this straight from now on.

To those of a religious turn of mind, who think of God as "the cause of everything," and who are afraid that what I have just said is just a way of introducing atheism by poisoning the wells of any rational investigation into God, then I simply retort, "If you say that God is the cause of everything because 'everything has to have a cause,' then what is the cause of God?" Not even those who say they think that "everything has to have a cause" really think so; because if it were true, then every cause would have to have a cause, and there could be no causeless cause. And since, as Kant pointed out in one of his antinomies, the effect is not explained unless all the causes are given, then no effect is explained, and hence nothing has an explanation; which means that if everything has to have a cause, nothing at all in fact has a cause.

There are things people do to circumvent this inconvenient conclusion; but my point is that even if it could be circumvented, you have no reason for saying, "Everything has to have a cause." The fact that it "stands to reason" because everything we directly experience is unintelligible by itself does not allow us to make the leap and say that absolutely everything is unintelligible by itself. And, of course, as I said, those who use this as a "premise" contradict it in fact once they get to what they call "God," because they don't demand a cause for that something.

Therefore, in order to be able to talk about an effect, you have to be able to show that there are facts that are in conflict.

But in order for something to be an effect, it must also be the case that the conflict is not due to misreading of one or both of the facts. That is, that kind of "problem" we saw dealing with the keys, where the person didn't in fact have them and only thought he had them, is not an effect, it is a mistake. There isn't any cause for his problem as such because there wasn't any problem to begin with; he only thought he had one. If you will, the problem in his quandary is that it seemed to him that he had conflicting evidence when in fact he didn't; and the cause of this problem was his faulty memory; but the problem as he formulated it didn't exist, and so was not an effect needing a cause.

We will see, beginning in the next section, that one of the practical difficulties in metaphysics is in making sure that the problem confronting us is an effect and is not a pseudo-problem analogous to the keys. We have to show that there is real evidence on both sides of the contradiction

Of course, having said this, I have to add that the evidence, while it must be real, cannot be conclusive on both sides, or there would be a real contradiction, and this is impossible. There has to be a way out of the dilemma somehow, and so there has to be room for a missing piece of information which can straighten out the conflict in what is (at the moment) known.

This leads to another fact worth observing about effects:

Conclusion 2: A situation is an effect because not all the information is known.

Obviously, if the facts known were all the facts there were, there would be a contradiction; and this is precisely what cannot be accepted. That is, the man who says, "I had these keys in my hand a minute ago, and they didn't just fly away, and they're not there now" is aware that there's some fact that accounts for how the keys got out of his hand without his realizing it, which fact he does not now know. Otherwise the keys would be in his hand while they are not in his hand.

Hence, you could say that we experience an effect only due to our ignorance of the whole situation. If we were omniscient, nothing would be an effect and nothing would need explaining, because we would see the "missing fact" which made sense out of the effect, and there would be no problem. For instance, if the man noticed the thief lifting the keys from his hand, then he wouldn't consider the missing keys an effect, because he would understand how the keys got out of his hand.

There aren't, then, any effects (in our sense of the term) in reality, because of course the real situation is complete and contains the cause, and so the real situation makes sense.

Nevertheless, there are things that really depend on other things, and so there is a sense in which we can talk about "real effects." For instance, if the thief took the keys, his action explains why the keys are not in the man's hand now, even if the man knows he took them. That is, the man would have the keys in his hand if the thief had not taken them, and, given this plus the fact that he doesn't have them, this can be called an effect whose cause is the thief's action.

What is being done here? In fact, in talking about such "real dependence," a person is making an abstraction from some aspect of (what he knows to be) the real situation, and saying "Without this aspect, the remaining part of the situation would be a contradiction. This "remaining" part (the effect) is then treated as if the cause were non-existent or unknown in order to show the "dependence."

To put this another way, the only way "X depends on Y" makes any sense is to say "Without Y, X is impossible" in one respect or another; and of course the precise respect in which X would be impossible is the respect in which it is an effect of Y. But "to be impossible" means that it can't exist in the way it exists, and the only way we can know that something can't exist is that there's a contradiction of some sort involved in saying that it does exist. The upshot of this is that you can't show that X depends on Y without showing the respects in which X would be a contradiction without Y, which, of course, is saying the same thing as "Suppose Y were not there; X would then contradict itself(1)."

You have mentally removed Y from the situation, and left yourself with evidence for a contradiction.

Hence, even when we in fact know both the effect and its cause, we think of the situation as a cause-effect situation because we are pretending to view the effect without the cause and showing how the cause is necessary to the intelligibility of the effect.

All this is by way of saying that the effect is not something concrete; it occurs because of our way of thinking about things, where we either don't know or ignore certain aspects of the concrete situation.

Do not misunderstand me. I am not saying that there is not real dependence, and things just "are, side by side." It is certainly true, supposing that the keys were taken by a thief, that their non-presence depends on his taking them. But the "dependence" is not a "something" in the sense of an object; it is a fact about the two objects, and is not itself an object known--a reality in its own right. That way would lead to Plato's Forms. We will learn considerably later that facts, which are the way we know objects, are not themselves objects or realities; we know about realities in a way that is not itself a reality.

This is very mysterious, and the reason for it will be apparent later. But for now, what it says can be illustrated by a simple example or two: If it is a fact that the unicorn I am imagining is not real, then is the unreality of the unicorn the reality which I know? If it is a fact that you are not me, is the reality the "not-me-ness" in you, or the "not-you-ness" in me, or the "real relation" of not being each other "between" us--or are the only realities involved you and me, and your not being me is just a fact about these realities and is not a reality itself?

That is, if you make facts some kind of "reality," then you make it impossible to say what you mean by reality, because you've got unreal or semi-real realities, and then the Principle of the Excluded Middle goes out the window.

So basically, when you say that one object really depends on another, what you mean is not that the "dependence" is a something-or-other "out there" on a par with the objects, but that it is true (or is a fact) that the objects depend on each other. In this case, you are considering the objects in the way in which you know them, and saying that there is something about at least one of them which would be self-contradictory without the other, or which would not exist as it exists without the other.

David Hume noticed this fact, I think, that the "dependence" of the effect on the cause was not a "something" which you actually observed, but had to do with your way of thinking about the two objects; but he drew the wrong conclusion from it. That it is obvious that he drew the wrong conclusion was that he felt compelled to explain why we still persist in thinking about "dependence" and "cause and effect" even though we observe no dependence as such. He said we just get into the habit of noticing the first event (the one on which the other "depends") and then the second; and so the first leads us to expect the second, and so we think there is some "real relation" between the two when in fact there isn't.

The problem with this explanation is that it is an attempt to show how our notion of real dependence really depends on habit, not some "real dependence." That is, Hume is trying to show the relation between noticing sequences and expecting the second, and our thinking in terms of cause and effect and supposing that there is real dependence. But of course if there is not a real dependence of our notion of cause-and-effect thinking on this habit we get into of noticing sequences, his "explanation" is just so much palaver, and accounts for nothing. So Hume is begging the question. If his theory is valid, then it gives the real cause for cause-and-effect thinking; but what it purports to establish is that there are no real causes. So if it is valid, it is false.

Of course, it is also true that if his theory were valid, then every invariable sequence in our experience would set up an "X is the cause of Y" idea. And since day always follows night, then we would necessarily think that night causes day(2) , or roads cause automobiles, or that robins cause the spring to come.

But what Hume did notice is that in the real world "by itself" there aren't causes and effects; in the real world as we know it, however, there are; but the effects are effects because we abstract from some aspect of the real world to consider them as such. Effects are abstractions, not objects. Not that they are false; they are just incomplete ways of knowing or looking at the real, concrete situation. An incomplete view of something is only false if you take it to be the complete view; and anyone who is considering an effect knows precisely that it is not all there is to the concrete situation.



1. Of course, the usual notion of "effect" is this "dependence" notion, in the sense that the effect is what is somehow "produced by" its cause. What I was trying to show here is that, looking at the effect itself, we know that it is "produced by" something when we know that it couldn't be what it is without the cause. Ordinary knowledge, then, looks at the situation from cause (the "producer") to effect (the "product), which is legitimate in most cases.

But first of all, my way of looking at effects is more general, because a given situation might be impossible-in-itself without there being a real agent that does something to it. For instance, the conclusion of a syllogism is "caused by" the arrangement of the premises; but they don't actually generate the conclusion (they merely imply it); but it is the case that the conclusion would not be known to be true if the premises were false or not arranged in the proper way. So my view of "effect" applies to all cases where something "needs" something else in order to be what it is and not just those cases in which it is known to have been acted on by something.

Secondly, note that I am "deriving" the notion of effect (and cause, as we will see shortly) from the Principle of Contradiction, only in this sense: It is not a logical conclusion of the Principle, but a recognition that our incomplete view of reality can sometimes make it appear that the Principle does not hold in this case, while we know with absolute certainty that it does.

Ayn Rand tries to derive causality from the Principle of Identity, making the leap from saying "A is A" to saying that "A being acts according to its nature," (since its "nature" is defined as "the source of its activity"). But this has several difficulties. First of all, if it is in fact a logical implication of the way "nature" is defined here, then it is just trivially tautologous, and all it says is "When this being acts in this way, it acts in this way," and there is no possibility that anything could be "unnatural," whether in the world of matter or human conduct. When a person chooses to violate his humanity, for instance, he does what he does, and this is not (by her definition) an "unnatural" act; it is just as "natural" as any other, because A is A. Similarly, the baby born with four legs recently was just as "natural" as any other birth by this definition, since what it is is what it is. In other words, the words "natural" and "unnatural" now have no meaning. Also, you can't get to "a being acts according to its nature" from A is A, because all "A is A" says is that a being is what it is, and says nothing about what it does. As Kant would say, the proposition "X acts according to its nature" is synthetic, not analytic (tautological), because information is added when you move from "X is what it is" to "X does what it does," and you presuppose still more information when you add "What it does is according to its nature" (you at least add "the source consistently results in its actions," which of course is one way of looking at the Principle of Causality, and is something additional to and different from merely saying "A is A."

My view is, as I said, not an analysis of the Principle of Contradiction, but supposes the additional fact beyond the Principle that our lack of knowledge of the whole situation can seem to violate the Principle--but since the Principle is absolutely and universally true, we know that it has not in fact been violated, and the "cause" of this is that our knowledge is incomplete. That is, the fact that we have evidence that there is a contradiction here, (while we know from the self-evidence of the Principle that there is not a contradiction here) is an effect, whose cause in this case is that we don't know all the facts. Since this is so, as I will say shortly, we know that there has to be an additional fact which resolves the contradiction.

2. I was at a meeting of a philosophical society once where I pointed this out, and a Humean in the audience said, "Well it does; the passing of the night causes the day." I countered, "Nonsense; the sun (which appears after the dawn) causes the day." He held to his position, however. I suppose he was "comfortable" with it; but I pity his students.