[Math] Problems with Cantor’s diagonal argument and uncountable infinity

cantor setinfinity

Cantor's diagonal argument:
As a starter I got 2 problems with it (which hopefully can be solved "for dummies")

First: I don't get this: Why doesn't Cantor's diagonal argument also apply to natural numbers?
If natural numbers cant be infinite in length, then there wouldn't be infinite in numbers. By using a randomly ordered list, you wouldn't end with an endless sequence of 0's you have to change.
Also it initially goes for "set of numbers". It is applied to the "right" side (fractional part) to prove "uncountability" but can't be used for the "left" side (integer part) because of "reasons" (I simply do not get it).

Second: the way is is used so many times, would just work in the case that the length and width of the list equals. Just listing all natural numbers where $0<=x<100$ will have a width of 3 but a length of 100. At base x each increase of digits will increase the length by x times. At base 10, all 4 digit numbers will create a list with 1000 entries. The length increases exponentially while the width does linearly. This is wrong in so many ways but: Doing this infinitely makes it a square?!? (Natural numbers are a "part" of integers but as you can map both with each other they are considered being the same size aka 2 different countable infinities always have the same "size" while one can be just a part of the other)

As the last part: lets assume we divide all real numbers in 2 parts. The integer part which defines the "set" we use. (there will be "countable" infinite of them)
Now, all we need to do is mapping the fractional part. Just use the list of natural numbers and flip it over for their position (numeration). Ex 0.629445 will be at position 544926. You could argue that this isn't possible for numbers like $\sqrt{2}$.
Lets pick $\pi$:

3.1415926535897932384626433832795… will be
3rd set at position: …5972383346264832397985356295141

There is no reason you cannot pick the next digit and put it in front for the position. There is no limit in length for natural numbers -> you can write a natural number which is the index for just that fractional part. Simply put: you can map EVERY number with $0<=x<100$ with a natural number. And a countable infinite amount of sets containing countable infinite entries still is countable.

So there are 3 Questions (I probably need to split this question):

  • Why doesn't Cantor's diagonal argument also apply to natural numbers? (for dummies: why you can't simply use it to the left)
  • If the count of digits equals the the length of the list, doesn't it just proves that this construction cannot contain all possibilities?
  • Wouldn't the construction of a set like in "the last part" be a prove that all real numbers are countable infinite? (Which part can't be done / is invalid?)

The most important part would be the third question. (If it only qualifies for one answer) Thanks in advance.

An additional big "thanks" in advance for correcting all the spelling, orthography and typos…

Best Answer

"If natural numbers can't be infinite in length, then there wouldn't be infinite in numbers." You are extrapolating properties of the natural numbers to what is called "potential infinity." But that is not what Cantor means by Aleph0.$\newcommand\N{\mathbb{N}}$

"Potential Infinity" is a hypothetical number that represents the hypothetical end to the sequence 1,2,3,... . Nobody treats it as a real object, but only in hypothetical evaluations of expressions where its impact on the result diminishes to nothing. For example, f(x)=3/(2+x) approaches 3/2 "as x goes to potential infinity." The point is that it is treated like a number with the same properties as "regular" numbers, but you can never reach it. "Actual infinity" is a well-defined mathematical concept, but it is nothing like a "regular number."

The "cardinality" of the set {1} is the cardinal number 1. That is, the set has one member. The cardinality of the set {1,2} is the cardinal number 2. In general, the cardinality of the finite set {1,2,3,...,n} is the largest number in it, n.

The Axiom of Infinity says that there is a set $\N$ with a similar form, but that has no largest member. Its cardinality is Aleph0, which can be called "Actual Infinity." It is not a member of the set it describes. It is not a "regular number" in any way. Even though we can extend the definition of numbers to include it as an "irregular number," it does not have the all the same properties. One that is different, is that you can't represent it with a string of digits. Not even an infinite string whose value you might think you could claim is "Potential Infinity."

Diagonalization does not work on natural numbers because it requires a digit for every member of $\N$, and that does not represent a natural number.

About your second problem: you can't consider an $\N$ by $\N$ array of characters to be square. "Square" means that you can only pair rows with columns one a 1:1 basis. While you can make such a pairing with an $\N$ by $\N$ array, you can also pair them on a 1:2 basis. Or 3:5. Or 14:9. The point of using Aleph0 for a cardinality, is that it represents any countably infinite set the same way.