Find the flaw in this mapping between the naturals and reals

91 Views Asked by At

I was studying Cantor's diagonal argument etc. I was testing the ideas and I thought of the following mapping between the naturals and the reals and I need some help to find the flaw in it. For convenience, we use binary.

We start with the root node (the dot). Recursively, every node has two children, 0 and 1. Every n-th layer of the tree are possible values for the n-th decimal. In a BFS fashion, we set the first two children as 0 and 1, the next layer 2,3,4,5 etc.

Can't this be considered a mapping from the naturals to the reals? What assumption am I missing? Thank you.

1

There are 1 best solutions below

0
On BEST ANSWER

We establish the existence of infinite sets by the Axiom of Infinity:

  • There is a set N that contains the natural number 1 and, if it contains the natural number n, it also contains the natural number n+1.

Some obvious but subtle points here:

  1. Every number n in this set has a finite value.
  2. If the process stopped at some value n, the cardinality of the set would be have the finite value n. That is, the cardinality of every such finite set has the same value as the "last" member of the set.
  3. But when the process does not end, as in the Axiom, the cardinality is Aleph0. This is not a finite value, and it does not correspond to to a value in the set.

What you are trying to do, essentially, is use a nested version of the process in the Axiom of Infinity to define real numbers. The problem is that every number you define this way has a finite length, just like every n in N has a finite value. Every number you define this way is the binary representation of a rational number whose denominator is 2^n.

+++++

BTW, Cantor did not apply his Diagonal Argument to real numbers. He even said "there is a proof of this proposition ... which does not depend on considering the irrational numbers." He did apply it to infinite-length strings like you are trying to define. But they cannot be not defined by a tree like you attempt. They are defined by a mapping from N to {0,1}.