Back to contents

In any discussion of the fundamental structure of the Universe the word "infinity" invariably pops up. It's usually meant to evoke a sense of wonder, but we may well ask what precisely do we mean by that word? Actually, we can mean any one of several things by it, but here I want to consider only the mathematical meaning. I can do that because in the late 1800's Georg Cantor worked out the technical descriptions of an endless number of infinities. The one I want to use is the smallest of them, the one that Cantor called Aleph-Null.

Imagine that all of the counting numbers, absolutely all of them, have been written down on a straight line. You can see that the sequence has a beginning, at the number zero, but does it have an end? To answer that question suppose that I show you a very big number and claim that it's the last number in the sequence. You know that's nonsense because you know that you need only add one to my number to generate the next number in the sequence and you know that you can apply that procedure over and over again and never reach a point at which you are compelled to stop. Thus, the sequence has no end, which is just what the Latin word infinitas means.

Even though we can never come to the end of the sequence of counting numbers, we nonetheless want to pretend that we can do so and talk about the hypothetical last number in the sequence, the number that can never be reached. We call that unreachable number Aleph-Null. We know that we cannot draw that number in the way that we draw the counting numbers, even the very largest that mathematicians use, so Cantor designated it with the Hebrew letter Aleph to which a subscript of zero has been appended (that subscript distinguishes it from Aleph-One, Aleph-Two, and the other alephs that represent progressively bigger infinities). You might think that there's nothing more that we can say about Aleph-Null, but what made Cantor's work significant was that he showed how to extend the rules of arithmetic to include operations carried out on Aleph-Null.

For our purpose I need only work out one of those rules. Recall the sequence of numbers from above and imagine that Cantor has come along, counted along that sequence to any one of its members, and then, starting from that point, drawn the same sequence next to ours. Thus, for example, if Cantor chooses to draw his zero next to our fifty-three, then he will draw his one next to our fifty-four, his two next to our fifty-five, and so on. He can go on drawing his numbers next to ours all the way along our sequence, all the way toward the end that has no end, and as he does so he can make an extraordinary claim: he can claim that a) we and he have drawn exactly the same sequences, containing exactly the same elements, and b) that, even though his sequence started somewhere past where ours did, his sequence will not extend beyond ours.

But, we object, surely his sequence must extend fifty-three elements beyond ours; after all, he began his sequence where we had already drawn fifty-three elements and we and he have drawn all the same elements, matched in pairs. Why would he not have fifty-three elements left over when he comes to our last number? And, Cantor asks, what would that last number be? The very asking of that question opens our mind's eye to the understanding that his sequence will not extend beyond ours because there is no last number relative to which there can be a beyond.

Having thus made the necessary preparation, we now subtract Cantor's sequence from ours. We accomplish that subtraction through the simple expedient of eliminating all matched pairs of numbers from the double sequence. Thus, we eliminate his zero and our fifty-three, his one and our fifty-four, his two and our fifty-five, and so on, eliminating all of the paired elements toward the end that has no end. When we have finished the elimination we have left only the unpaired elements, the first fifty-three elements of our sequence (0, 1, 2, 3, 4, ..., 50, 51, 52). We have thus subtracted, one from the other, two things that are the same and have come up with a nonzero remainder. From that fact we infer a simple rule for infinities:

Aleph-Null subtracted from Aleph-Null yields

any finite number as the difference.

That rule means that subtracting one Aleph-Null from another, whether they come from counting different sequences or from counting the same sequence at different times, is not guaranteed to yield zero as the difference. Indeed, because we have no reason to believe that any one finite number is more probable than any other as the difference, we can state that the likelihood of the difference being zero is itself effectively equal to zero. That proposition necessitates an important constraint upon the laws of physics, one that will enable us to use the conservation laws to describe space and time as something more than merely there and then.

The generic conservation law tells us that some property possessed by matter can be neither created nor destroyed, but can only be transferred from one body to another. That law can also be stated as a subtraction: if we are given an isolated system of bodies possessing some quantity of a conserved property, then subtracting the amount of that property that the system possesses at one time from the amount that it possesses at some other time must always necessarily yield zero. Viewing that rule in the light of the subtraction rule for Aleph-Null, I can now state


No conserved property can ever exist in an infinite quantity.

If that theorem were not true to Reality, then the subtraction in the conservation law would not yield a necessary zero and the property in question would not be necessarily conserved.

To get a feel for what that rule means, let's consider something that is not conserved. Space is a good example. We describe it as being made up of points and we describe the points as having zero extent. How many things of zero extent must we take to span a nonzero extent? An infinite number at the very least, of course. Every volume of space, however small, contains an infinite number of points, which means that the fundamental measure of space, extent, is not subject to a conservation law. Space can expand or contract freely.

In contrast, consider angular momentum. That is a conserved property that we describe as the product of a mass, a velocity, and a distance. The finite-value theorem tells us that the Universe must be so structured that there is no possibility of that product being infinite. But we are free to set up our revolving body in any way that we want within the Universe, so we must infer that none of the factors in that product can possibly be infinite: that is, we infer that there cannot be an infinite amount of mass in the Universe; there can be no motion at infinite speed; and there can be no infinite distance.

Those three inferences may not seem impressive. I found them disappointing at first myself; they seemed to say so little about the physics of space and time. But, as I discovered, they are part of the combination that lets us unlock the deeper secrets of the Universe.


Back to Contents