Infinity – Maverick Philosopher

Topics in Current Technical Threads

1) Potential versus actual infinity.

2) Are there mathematical sets?

3) Does mathematics need a foundation in set theory?

4) Is there irreducibly plural reference, predication, and quantification? If yes, does plural quantification allow us to avoid ontological commitment to sets?

5) Discreteness, density, and continuity at the level of number theory, geometry, and nature (physical space and physical time)?

6) Phenomenal versus physical space and their relation. Homogeneity and continuity in relation to both. Lycan's puzzle about the location of the homogeneously green after-image. Wilfrid Sellars and the Grain Argument. It was (6) that got us going on the current jag.

The above topics which we have recently discussed naturally lead to others which I would be interested in discussing:

7) How is it possible for mathematics to apply to the physical world? Does such application require a realist interpretation of mathematics? (See Hilary Putnam, Philosophical Papers, vol. I, 74.)

8) Zeno's Paradoxes. Does the 'calculus solution' dispose of them once and for all?

9) The 'At-At' theory of motion and related topics such as instantaneous velocity.

10) Mathematical existence.

11) The Zermelo-Fraenkel axioms, their epistemic status, and the puzzles to which they give rise.

12) How the actual versus potential infinity debate connects with the eternalism versus presentism debate in the philosophy of time.

Notes on Infinite Series

The resident nominalist writes,

Your post generated a lot of interest. What I have to say now is better put as a separate post, rather than a long comment. Feel free to post.

1) Plural reference provides a means of dealing with numbers-of-things without introducing extra unwanted entities such as sets. Even realists agree that we should not have more entities than necessary, the disagreement is about what is ‘necessary’.

BV: We agree that entities should not be multiplied beyond necessity, i.e., beyond what is needed for explanatory purposes. The disagreement, if any, will concern what is needed.

2) Using plural quantification we can postulate the existence of an infinite number-of-things. We simply postulate that for any number-of-things, there is at least one other thing. That gives a larger number-of-things, which itself is covered by the quantifier ‘any’, hence there must be a still larger number-of-things, etc.

BV: You give no example, so let me supply one. Consider the series of positive integers: 1, 2, 3 . . . n, n + 1, . . . . Given 1, we can generate the rest using the successor function: S(n) = n + 1. I used the word 'generate' since it comports well with your intuition that there are no actual infinities, and that therefore every infinity is merely potential.

3) In this way we neatly distinguish between actual and potential infinity. Using plural quantification, we can prove that there is no plural reference for ‘all the things’. For that would be a number-of-things, hence there must be an even larger number-of-things, which contradicts the supposition that we had all the things.

BV: Your argument is rather less than pellucid. Here is the best I can do by way of reconstructing your argument:

a) If the plural term, 'all the positive integers,' refers to something, then it refers to a completed totality of generated integers. But

b) There is no completed totality of generated integers.

Therefore, by modus tollens,

c) It is not the case that 'all the positive integers' refers to something.

Therefore

d) There is no actually infinite set of positive integers.

If that is your argument, then it begs the question at line (b). One man's modus tollens is another man's modus ponens. If the above is not your argument, tell me what your argument is. So far, then, a stand-off.

4) In this way we also avoid the pathological results of Cantorean set theory. If there is a set of natural numbers, then this is also a number, but it cannot itself be a natural number, so it is the first ‘transfinite number’. The nominalist approach avoids such weird numbers.

BV: But surely polemical verbiage is out of place in such serene precincts as we now occupy. You cannot shame Cantor's results out of existence by calling them 'pathological' or 'weird.' Most if not all working mathematicians accept them, no?

5) The problem for the nominalist arises when in trying to explain the sum of an infinite series, e.g. 1 + ½ + ¼ + ⅛ … The realist wants to argue that unless this series is ‘completed’, we don’t have all the members, so the sum will amount to less than 2.

BV: Note that the formula for the series is 1/2ⁿwhere n is a natural number with 0 being the first natural number. Recall that any number raised to the zeroth power = 1. (If you need to bone up on this, see here.)

Question for our nominalist: what does '1/2ⁿ' refer to? Can't be a set! And it can't be a property! Does it refer to nothing? Then so does '1-1/2n.' How then explain the difference between the two formulae (rules) for generating two different infinite series?

Or more simply, consider n. It is a variable. It has values and substituends. The values are the natural numbers. Only the ones we counted up to, or generated thus far? No, all of them. The ones we have actually counted up to in a finite number of countings, and the rest which are the possible objects of counting. The variable is a one-over-the-many of its values, and a one-over-the-many of its substituends, which are numerals, not numbers. Numerals bring in the type-token distinction. And so I will ask the nominalist what linguistic types are. Are they sets? No. Are they properties? No. What then?

6) It’s a difficult question for the nominalist, but here is my attempt to resolve it. Start with the notion of non-overlapping parts. Two non-overlapping parts have no part that is part of the other. Then there can be a number of non-overlapping parts such that there is no other such part, i.e. these are ‘all’ such parts.

BV: OK.

7) Then suppose we have a method of defining the parts. Start with a line of length 2. Note that the nominalist is OK here with the existence of lines, because lines are real things and not artificially constructed entities like ‘sets’. And suppose we can divide the line into two non-overlapping parts of equal length, i.e., a part of length 1, and another part of the same length.

BV: You shouldn't say that sets are artificially constructed. After all, you think numbers are artificially constructed, no? They are artifacts of counting. Your beef is with abstract objects, not artificial objects. Sets are abstract particulars. You oppose them for that reason. As a nominalist you hold that everything is a concrete particular. (Or am I putting words in your mouth?)

Second, you are ignoring the difference between a geometrical line and a line drawn with pencil on paper, say. The latter is a physical line, which is actually a 3-D object with length, width and depth. In addition to its pure geometrical properties, it has physical and chemical properties. It is a physical line in physical space. The former is not a physical line, but an ideal line: it has length, but no width or depth. Ideal lines are not in physical space. Suppose physical space, the space of nature, is non-Euclidean. Then Euclidean lines are obviously not in physical space. But even if physical space is Euclidean, Euclidean lines would still not be in physical space.

8. So the proposition “2 = 1+1” says that a line of length two can be divided into two equal non-overlapping parts. Then suppose that we divide the second part into two equal parts. Thus “2 = 1 + ½ + ½” says that the line can be divided into three non-overlapping parts, one of length 1, and the other two equal. Do the same again, thus 2 = 1 + ½ + ¼ + ¼. And again and again!

BV: An obvious point is that the arithmetical proposition '2 = 1 + 1' is not about lines only. It could be about a two-degree linear cool-down of a poker. (I am thinking about Wittgenstein's famous poker-brandishing incident.) It could be about anything. Two pins. An angel on a pin joined by another.

Besides, "2 = 1 + 1" cannot be about the non-overlapping parts of a particular line, the one you drew in the sand. It is about a geometrical line, which is an ideal or abstract object. The theorem of Pythagoras is not about the right triangle you drew on the blackboard with chalk; it is about the ideal right triangle that the triangle you drew merely approximates to.

9. It is clear that for every such division, the parts ‘add up’ to the same number, i.e. 2.

10. Then consider what the proposition “2 = 1 + ½ + ¼ + ⅛ … “ expresses. Surely that every such series, however extended, has a sum of 2. Do we need the notion of a ‘set’? No.

BV: I don't see how this answers the question that you yourself raised in #5 above. What makes it the case that the series you mention actually has a sum of 2? The most you can say is that series potentially has a sum of 2. The Cantorean does not face this problem because he can say that there is an actual infinity of compact fractions that sums to 2. No endless task needs to be performed to get to the sum.

On Potential and Actual Infinity, and a Puzzle

Consider the natural numbers (0, 1, 2, 3, . . . n, n +1, . . . ). If these numbers form a set, call it N, then N will of course be actually infinite. This because a set in the sense of set theory is a single, definite object, a one-over-many, distinct from each of its members and from all of them. N must be actually infinite because there is no greatest natural number, and because N contains all the natural numbers.

It is worth noting that 'actually infinite set' is a pleonastic expression. It suffices to say 'infinite set.' This is because the phrase 'potentially infinite set' is nonsense. It is nonsense (conceptually incoherent) because a set is a definite object whose definiteness derives from its having exactly the members it has. A set cannot gain or lose members, and a set cannot have a membership other than the membership it actually has. Add a member to a set and the result is a numerically different set. In the case of the natural numbers, if they form a set, then that set will be an actually infinite set with a definite transfinite cardinality. Georg Cantor refers to that cardinality as aleph-zero or aleph-nought.

I grant, however, that it is not obvious that the natural numbers form a set. Suppose they don't. Then the natural number series, though infinite, will be merely potentially infinite. What 'potentially infinite' means here is that one can go on adding endlessly without ever reaching an upper bound of the series. No matter how large the number counted up to, one can add 1 to reach a still higher number. The numbers are thus created by the counting, not labeled by the counting. The numbers are not 'out there' in Plato's topos ouranios waiting to be counted; they are created by the counting. In that sense, their infinity is merely potential. But if the naturals are an actual infinity, then they are not created but labeled.

Moving now from arithmetic to geometry, consider a line segment in a plane. One can bisect it, i.e., divide or cut it into two smaller segments of equal length. Thus the segment AB whose end points are A and B splits into the congruent sub-segments AC and CB, where C is the point of bisection. The operation of bisection is indefinitely ('infinitely') iterable in principle. The term 'in principle' needs a bit of commentary.

Suppose I am slicing a salami using a state-of-the-art meat slicer. I cannot go on slicing thinner and thinner indefinitely. The operation of bisecting a salami is not indefinitely iterable in principle. The operation is iterable only up to a point, and this for the reason that a slice must have a certain minimal thickness T such that if the slice were thinner than T it would no longer be a slice. But if we consider the space the salami occupies — assuming that space is something like a container that can be occupied — then a longitudinal (non-transversal) line segment running from one end of the salami to the other is bisectable indefinitely in principle.

For each bisecting of a line segment, there is a point of bisection. The question can now be put as follows: Are these points of bisection only potentially infinite, or are they actually infinite?

A Puzzle

I want to say that from the mere fact that the operation of bisecting a line segment is indefinitely ('infinitely') iterable in principle, it does not follow that the line segment is composed of an actual infinity of points. That is, it is logically consistent to maintain all three of the following: (i) one can always make another cut; (ii) the number of actual cuts will always be finite; and that therefore (iii) the number of points in a line will always be finite, and therefore 'infinite' only in the sense that there is no finite cardinal n such that n is the upper bound of the number of cuts.

At this 'point,' however, I fall into perplexity which, according to Plato, is the characteristic state of the philosopher. If one can always make another cut, then the number of possible cuts cannot be finite. For if the number of possible cuts is finite, then it can longer be said that the line segment has a potentially infinite number of points of bisection. It seems that a potential infinity of actual cuts logically requires an actual infinity of possible cuts.

But then actual infinity, kicked out the front door, returns through the back door.

I have just posed a problem for those who are friends of the potentially infinite but foes of the actual infinite. How might they respond?

World + God = God: A Mathematical Analogy

The Big Henry offers the following comment on my post, World + God = God?

"World + God = God" is (mathematically) analogous to "number + infinity = infinity", where "number" is finite. If God embodies all existence, then God is "existential infinity", and, therefore, no amount of existence can be added to or subtracted from God's totality.

The numerical concept of infinity does not comply with the rules of arithmetic addition or subtraction. Similarly, if God is presumed to be the embodiment of all existence, He does not comply with the rules of arithmetic addition or subtraction.

To supply an example that supports Big Henry's point, 8 + $\aleph_0$ = $\aleph_0$ . $\aleph_0$ (aleph-nought, aleph-zero, aleph-null) is the first transfinite cardinal. A cardinal number answers the How many? question. Thus the cardinal number of the set {Manny, Moe, Jack} is 3, and the cardinal number of {1, 3, 5, 7} is 4. Cardinality is a measure of a set's size. What about the infinite set of natural numbers {0, 1, 2, 3, 4 . . . n, n + 1, . . .}? How many? $\aleph_0$ . And as was known long before Georg Cantor, it is possible to have two infinite sets, call them E and N such that E is a proper subset of N, but both E and N have the same size or cardinality. Thus the evens are a proper subset of the naturals, but there are just as many of the former as there are of the latter, namely, $\aleph_0$ . How can this be? Well, EACH element of the evens can be put into 1-1 correspondence with an element of the naturals.

So far the analogy holds. But I think Big Henry has overlooked the transfinite ordinals. The first transfinite ordinal, denoted

, is the order type of the set of nonnegative integers. (See here.) You could think of

as the successor of the natural numbers. It is the first number following the entire infinite sequence of natural numbers. (Dauben, 97) The successor of

+ 1. These two numbers are therefore different. Here the analogy breaks down. God + Socrates = God.

+ 1 is not equal to

Moreover, it is not true to say that "The numerical concept of infinity does not comply with the rules of arithmetic addition or subtraction." This ignores the rules of transfinite cardinal arithmetic and those of transfinite ordinal arithmetic. Big Henry seems to be operating with a pre-Cantorian notion of infinity. Since Cantor we have an exact mathematics of infinity.

In any case, I rather doubt that mathematical infinity provides a good analogy for the divine infinity. God is not a set!

Does Potential Infinity Presuppose Actual Infinity?

Returning to a discussion we were having back in August of 2010, I want to see if I can get Peter Lupu to agree with me on one point: It is not obvious or compellingly arguable (arguable in a 'knock-down' way) that there are infinite sets. Given my aporetic concerns, which Peter thoroughly understands, I will be satisfied if I can convince him that the italicized sentence is true, and therefore that the thesis that the infinite in mathematics is potential only is respectable and defensible and has never been shown definitively to be false. Let us start with a datanic claim that no one can reasonably deny:

1. There are infinitely many natural numbers.

If anyone were to deny (1) I would show him the door. For anyone who denied (1) would show by his denial that he did not grasp the sense of 'natural number.' The question, however, is whether from (1) we can validly infer

2. There is a set of natural numbers.

If there is such a set, then of course it is an infinite set, an actually infinite set. (Talk of potentially infinite sets is nonsense as I have argued in previous posts.) So, if the inference from (1) to (2) is valid, we have a knock-down proof of actual infinity. For if there are infinite sets then there are actual infinities, completed infinities.

Now I claim that it is obvious that (2) does not follow from (1). For it might be that the naturals do not form a set. A set is a one-over-many, a definite single object distinct from each of its members and from all of them. It should be obvious, then, that from the fact that there ARE many Fs it does not straightaway follow that there IS a single thing comprising these many Fs. This is especially clear in the case of infinitely many Fs.

But from Logic 101 we know that an invalid argument can have a true conclusion. So, despite the fact that (2) does not follow from (1), it might still be the case that (2) is true. I might be challenged to say what (1) could mean if it does not entail (2). Well, I can say that however many numbers we have counted, we can count more. If we have counted up to n, we can add 1 and arrive at n + 1. The procedure is obviously indefinitely iterable. That means: there is no definite n such one can perform the procedure only n times. One can perform it indefinitely many times. Accordingly, 'infinitely many' behaves differently than 'finitely many.' If something can be done only finitely many times, then there is some finite n such that n is the number of times the thing can be done. But 'infinitely many' does not require us to say that that there is some definite transfinite cardinal which is the number of times a thing that can be done infinitely many times can be done. For 'infinitely many' can be construed to mean: indefinitely many.

On this approach, the naturals do not form a single complete object, the set N, but are such that their infinity is an endless task. The German language allows a cute way of putting this: Die Zahlen sind nicht gegeben, sondern aufgegeben. In Aristotelian terms, the infinity of the naturals is potential not actual. But if you find these words confusing, as Peter does, they can be avoided. A wise man never gets hung up on words.

Now if I understood him aright, one of Peter's objections is that the approach I am sketching implies that there is a last number, one than which there is no greater. But it has no such implication. For the very sense of 'natural number' rules out there being a last number, and this sense is understood by all parties to the dispute. There cannot be a last number precisely because of the very meaning of 'number.' Every natural number is such that it has an immediate successor. But from this it does not follow that there is a set of natural numbers. For 'has an immediate successor' needn't be taken to mean that each number has now a successor; it can be taken to mean that each number at which we have arrived by computation is such that an immediate successor can be computed by adding 1.

But Peter has a stronger objection, one that I admit has force. His objection in nuce is that potential infinity presupposes actual infinity. Peter points out that my explanation of what it means to say that the naturals are potentially infinite makes use of words like 'can.' Thus above I said, "however many numbers we have counted, we can count more." This 'can' refers either to the abilities of men or machines or else it refers to abstract possibilities of counting not tied to the powers of men or machines.

Consider the second idea, the more challenging of the two. Suppose the universe ceases to exist at a time t right after some huge but finite n has been computed. Now n cannot be the last number for the simple reason that there cannot be a last number. This 'cannot' is grounded in the very sense of 'natural number.' So it must be possible that 1 be added to n to generate its successor. And it must be possible that 1 be added to n + 1 to generate its successor, and so on. So Peter could say to me, "Look, you have gotten rid of an actual infinity of numbers but at the expense of introducing an actual infinity of unrealized possibilities of adding 1: the possibility P1 of adding 1 to n; the possibility P2 of adding 1 to n + 1, etc."

The objection is not compelling. For I can maintain that the unrealized possibilities P1, P2, . . . Pn, . . . all 'telescope,' i.e., collapse into one generic possibility of adding 1. P1 is the possibility of adding 1 to n and P2 is the possibility of adding 1 to the last number computed just before the universe ceases to exist.

What I'm proposing is that 'Every natural number has an immediate successor' is true solely in virtue of the sense or meaning of 'natural number.' Its being true does not require that there be, stored up in Plato's Heaven, a completed actual infinity of naturals, a set of same. Since I have decidedly Platonic sympathies, I would welcome a refutation of this proposal.

Infinity and Mathematics Education

A reader writes,

Regarding your post about Cantor, Morris Kline, and potentially vs. actually infinite sets: I was a math major in college, so I do know a little about math (unlike philosophy where I'm a rank newbie);
on the other hand, I didn't pursue math beyond my bachelor's degree so I don't claim to be an expert. However, I do know that we never used the terms "potentially infinite" vs. "actually infinite".

I am not surprised, but this indicates a problem with the way mathematics is taught: it is often taught in a manner that is both ahistorical and unphilosophical. If one does not have at least a rough idea of the development of thought about infinity from Aristotle on, one cannot properly appreciate the seminal contribution of Georg Cantor (1845-1918), the creator of transfinite set theory. Cantor sought to achieve an exact mathematics of the actually infinite. But one cannot possibly understand the import of this project if one is unfamiliar with the distinction between potential and actual infinity and the controversies surrounding it. As it seems to me, a proper mathematical education at the college level must include:

1. Some serious attention to the history of the subject.

2. Some study of primary texts such as Euclid's Elements, David Hilbert's Foundations of Geometry, Richard Dedekind's Continuity and Irrational Numbers, Cantor's Contributions to the Founding of the Theory of Transfinite Numbers, etc. Ideally, these would be studied in their original languages!

3. Some serious attention to the philosophical issues and controversies swirling around fundamental concepts such as set, limit, function, continuity, mathematical induction, etc. Textbooks give the wrong impression: that there is more agreement than there is; that mathematical ideas spring forth ahistorically; that there is only one way of doing things (e.g., only one way of construction the naturals from sets); that all mathematicians agree.

Not that the foregoing ought to supplant a textbook-driven approach, but that the latter ought to be supplemented by the foregoing. I am not advocating a 'Great Books' approach to mathematical study.

Given what I know of Cantor's work, is it possible that by "potentially infinite" Kline means "countably infinite", i.e., 1 to 1 with the natural numbers?

No!

Such sets include the whole numbers and the rational numbers, all of which are "extensible" in the sense that you can put them into a 1 to 1 correspondence with the natural numbers; and given the Nth member, you can generate the N+1st member. The size of all such sets is the transfinite number "aleph null". The set of all real numbers, which includes the rationals and the irrationals, constitute a larger infinity denoted by the transfinite number C; it cannot be put into a 1 to 1 correspondence with the natural numbers, and hence is not generable in the same way as the rational numbers. This would seem to correspond to what Kline calls "actually infinite".

It is clear that you understand some of the basic ideas of transfinite set theory, but what you don't understand is that the distinction between the countably (denumerably) infinite and the uncountably (nondenumerably) infinite falls on the side of the actual infinite. The countably infinite has nothing to do with the potentially infinite. I suspect that you don't know this because your teachers taught you math in an ahistorical manner out of boring textbooks with no presentation of the philosophical issues surrounding the concept of infinity. In so doing they took a lot of the excitement and wonder out of it. So what did you learn? You learned how to solve problems and pass tests. But how much actual understanding did you come away with?

Kline on Cantor on the Square Root of 2

Morris Kline, Mathematics: The Loss of Certainty, Oxford 1980, p. 200:

. . . when Cantor introduced actually infinite sets, he had to advance his creation against conceptions held by the greatest mathematicians of the past. He argued that the potentially infinite in fact depends on a logically prior actually infinite. He also gave the argument that the irrational numbers, such as the square root of 2, when expressed as decimals involved actually infinite sets because any finite decimal could only be an approximation.

Here may be one answer to the question that got me going on this series of posts. The question was whether one could prove the existence of actually infinite sets. Note, however, that Kline's talk of actually infinite sets is pleonastic since an infinite set cannot be anything other than actually infinite as I have already explained more than once. Pleonasm, however, is but a peccadillo. But let me explain it once more. A potentially infinite set would be a set whose membership is finite but subject to increase. But by the Axiom of Extensionality, a set is determined by its membership: two sets are the same iff their members are the same. It follows that a set cannot gain or lose members. Since no set can increase its membership, while a potentially infinite totality can, it follows that that there are no potentially infinite sets. Kline therefore blunders when he writes,

However, most mathematicians — Galileo, Leibniz, Cauchy, Gauss, and others — were clear about the the distinction between a potentially infinite set and an actually infinite one and rejected consideration of the latter. (p. 220)

Kline is being sloppy in his use of 'set.' Now to the main point. Suppose you have a right triangle. If two of the sides are one unit in length, then, by the theorem of Pythagoras, the length of the hypotenuse is the sqr rt of 2 = 1.4142136. . . . Despite the nonterminating decimal expansion, the length of the hypotenuse is perfectly definite, perfectly determinate. If the points in the line segment that constitutes the hypotenuse did not form an infinite set, then how could the length of the hypotenuse be perfectly definite? This is not an argument, of course, but a gesture in the direction of a possible argument.

If someone can put the argument rigorously, have at it.

Does Potential Infinity Rule Out Mathematical Induction?

In an earlier thread David Brightly states that "The position on potential infinity that he [BV] is defending is equivalent to the denial of the principle of mathematical induction." Well, let's see.

1. To avoid lupine controversy over 'potential' and 'actual,' let us see if we can avoid these words. And to keep it simple, let's confine ourselves to the natural numbers (0 plus the positive integers). The issue is whether or not the naturals form a set. I hope it is clear that if the naturals form a set, that set will not have a finite cardinality! Were someone to claim that there are 463 natural numbers, he would not be mistaken so much as completely clueless as to the very sense of 'natural number.' But from the fact that there is no finite number which is the number of natural numbers, it does not follow that there is a set of natural numbers.

2. So the dispute is between the Platonists — to give them a name — who claim that the naturals form a set and the Aristotelians — to give them a name — who claim that the naturals do not form a set. Both hold of course that the naturals are in some sense infinite since both deny that the number of naturals is finite. But whereas the Platonists claim that the infinity of naturals is completed, the Aristotelians claim that it is incomplete. To put it another way, the Platonists — good Cantorians that they are — claim that the naturals, though infinite, are a definite totality whereas the Aristoteleans claim that the naturals are infinite in the sense of indefinite. The Platonists are claiming that there are definite infinities, finite infinities – which has an oxymoronic ring to it. The Aristotelians stick closer to ordinary language. To illustrate, consider the odds and evens. For the Platonists, they are infinite disjoint subsets of the naturals. Their being disjoint from each other and non-identical to their superset shows that for the Platonists there are definite infinities.

3. Suppose 0 has a property P. Suppose further that if some arbitrary natural number n has P, then n + 1 has P. From these two premises one concludes by mathematical induction that all n have P. For example, we know that 0 has a successor, and we know that if arbitrary n has a successor, then n +1 has a successor. From these premises we conclude by mathematical induction that all n have a successor.

4. Brightly claims in effect that to champion the Aristotelean position is to deny mathematical induction. But I don't see it. Note that 'all' can be taken either distributively or collectively. It is entirely natural to read 'all n have a successor' as 'each n has a successor' or 'any n has a successor.' These distributivist readings do not commit us to the existence of a set of naturals. Thus we needn't take 'all n have a successor' to mean that the set of naturals is such that each member of it has a successor.

5. Brightly writes, "My understanding of 'there is' and 'for all' requires a pre-existing domain of objects, which is why, perhaps, I have to think of the natural numbers as forming a set." Suppose that the human race will never come to an end. Then we can say, truly, 'For every generation, there will be a successor generation.' But it doesn't follow that there is a set of all these generations, most of which have not yet come into existence. Now if, in this example, the universal quantification does not require an actually infinite set as its domain, why is there a need for an actually infinite set as the domain for the universal quantification, 'Each n has a successor'?

6. When we say that each human generation has a successor, we do not mean that each generation now has a successor; so why must we mean by 'every n has a successor' that each n now has a successor? We could mean that each n is such that a successor for it can be constructed or computed. And wouldn't that be enough to justify mathematical induction?

Addendum 8/15/2010 11:45 AM. I see that I forgot to activate Comments before posting last night. They are on now.

It occurred to me this morning that I might be able to turn the tables on Brightly by arguing that actual infinity poses a problem for mathematical induction. If the naturals are actually infinite, then each of them enjoys a splendid Platonic preexistence vis-a-vis our computational activities. They are all 'out there' in Plato's heaven/Cantor's paradise. Now consider some stretch of the natural number series so far out that it will never be reached by any computational process before the Big Crunch or the Gnab Gib, or whatever brings the whole shootin' match crashing down. How do we know that the naturals don't get crazy way out there? How can we be sure that the inductive conclusion For all n, P(n) holds? Ex hypothesi, no constructive procedure can reach out that far. So if the numbers exist out there, but we cannot reach them by computation, how do we know they behave themselves, i.e. behave as they behave closer to home? This won't be a problem for the constructivist, but it appears to be a problem for the Platonist.

Doron Zeilberger’s Ultrafinitism

This is wild stuff; I cannot say whether it is mathematically respectable but the man does teach at Rutgers. It is certainly not mainstream. Excerpt:

It is utter nonsense to say that sqrt 2

is irrational, because this presupposes that it exists, as a number or distance. The truth is that there is no such number or distance. What does exist is the symbol, which is just shorthand for an ideal object x that satisfies x² = 2.

Now what the hell does that mean? A rational number is one that can be expressed as a fraction a/b where both a and b are integers and b is not 0. An irrational number is one that cannot be expressed in this way. By the celebrated theorem of Pythagoras, a right triangle with sides of 1 unit in length will have an hypotenuse with length = the square root of 2. This is an irrational number. But this irrational number measures a quite definite length both in the physical world and in the ideal world. How can this number not exist? It is inept to speak of a symbol as shorthand for an ideal object since, if x is shorthand for y, then both are linguistic items. For example 'POTUS' is shorthand for 'president of the United States.' But 'POTUS' is not shorthand for Obama. 'POTUS' refers to Obama. Zeilberger appears to be falling into use/mention confusion. If the symbol for the sqrt of 2 refers to an ideal object, then said object is a number that does exist. And in that case Zeilberger is contradicting himself.

What's more, it seems that from Zeilberger's own example one can squeeze out an argument for actual infinity. We note first that the decimal expansion of the the sqrt of 2 is nonterminating: 1.4142136 . . . . We note second that the length of the hypotenuse is quite definite and determinate. This seems to suggest that the decimal expansion must be actually infinite. Otherwise, how could the length of the hypotenuse be definite?

As an ultrafinitist, however, Zeilberger denies both actual and potential infinity:

. . . the philosophy that I am advocating here is called

ultrafinitism. If I understand it correctly, the ultrafinitists deny the existence of any infinite, not [sic] even the potential infinity, but their motivation is `naturalistic', i.e. they believe in a `fade-out' phenomenon when you keep counting. [. . .]

So I deny even the existence of the Peano axiom that every integer has a successor.

As I said, this is wild stuff. He may be competent as a mathematician; I am not competent to pronounce upon that question. But he appears to be an inept philosopher of mathematics. But this is not surprising. It is not unusual for competent scientists and mathematicians to be incapable of talking coherently about what they are doing when they pursue their subjects. Poking around his website, I find more ranting and raving than serious argument.

The ComBox is open if someone can clue us into the mysteries of ultrafinitism. There is also some finitist Russian cat, a Soviet dissident to boot, name of Esenin-Volpin, who Michael Dummett refers to in his essay on Wang's Paradox, but Dummett provides no reference. Is ultrafinitism the same as strict finitism?