Re: (Not quite) Cantor's diagonal proof

From: |-|erc (spam_at_fodder.abc)
Date: 10/20/04 

Date: Wed, 20 Oct 2004 05:36:53 GMT

"Saint Cad" <saintcad@emailblackhole.com> wrote in > >> >>
> >>
> >> >>
> >> >> Here's another question. Suppose I list all of the repeating and
> >> >> terminating decimals (i.e. rationals). What in Cantor's proof
> >> >> prevents
> >> >> me
> >> >> from finding some sort of order for them so that I can use the
> >> >> diagonal
> >> >> method to create a repeating decimal not in the list, thereby
> >> >> "proving"
> >> >> that
> >> >> the rationals are uncountable? I know that I can't actually do this
> >> >> (the
> >> >> rationals are countable), but it would take someone else to explain
> >> >> why
> >> >> in
> >> >> order to dismiss this "counter" to his method. Again, Cantor's method
> >> >> may
> >> >> be shown to not be invalid, but it take more than just his method by
> >> >> itself.
> >> >
> >> > Since that are several well published surjective functions from the
> >> > naturals to the rationals, a proof that the rationals are uncountable
> >> > would, one hopes, be impossible, and would devastate mathematics if one
> >> > were ever found.
> >>
> >> Exactly! But my question is can Cantor's proof by itself be shown to not
> >> suffer this flaw?
> >
> > Please read the replies to you made by Tim Peters, Dave Seaman, Daniel
> > W. Johnson, and Robert Low (all of which were made after you wrote
> > this).
> > Especially Tim Peters has pointed out your flaws in not understanding
> > Cantor's Diagonal Argument with the clarity of a thousand suns.
> >
> > -Leonard Blackburn
>
> Let me start by saying that I don't believe Cantor's Method is in fact
> invalid. I know from when I got my BA in math that the proof as presented
> in class left a lot to be desired. I guess the issue I was trying to raise
> is: are there arguments against Cantor's Method that need results other that
> the method by itself to dismiss?
>
> The first "disproof" of Cantor was one presented by another student (not
> me!) at the time. I never liked that argument either. It struck me as
> similar to the argument to "the number of naturals is finite since listing 1
> to N is a finite cardnality and adding one more (i.e. 1 to N+1) is finite.
> QED." I do have to admit though that the idea a couple of posters presented
> (including Tim) that the original list of reals is arbitrary is a good
> point, one the professor didn't make when I originally learned it.
>
> The second "disproof" was in honor of James Harris. I was thinking of his
> idea of a counterexample to Wiles proof to see where it broke down. I had
> tried it myself and I kept running into the same problem - I could never
> show that the number created by the diagonal was repeating. I posted it in
> hopes that someone could show me why I kept running into the same problem
> over and over without the circular argument of "It fails because the
> rational are countable" and could instead point out where exactly Cantor's
> Method fails on rational numbers.
>

I've shown this... 3 times now. Yes it fails because the rationals are countable, so what does
'they are countable' mean?

1 Given a countable list containing all rationals
2 Given any *new number* that has a repeating decimal
3 Given any number with repeating decimal, we can calculate the integer numerator and denominator.
4 Given any numerator and denominator pair, we can calculate its position in the list.

   numerator 21111
--------------------------->
| / / / / /
| / / / /
| / / /
| / /
| /
|
| 10,000 *
v
denominator

5 Given it has a fixed position in the countable list, it cannot be new. Contradiction to 1&2.
6 Corollary : Every complete list of rationals has an irrational diagonal

HTH
Herc

Relevant Pages

  • Re: (Not quite) Cantors diagonal proof
    ... >> Please read the replies to you made by Tim Peters, Dave Seaman, Daniel ... > Method fails on rational numbers. ... Yes it fails because the rationals are countable, ... Given any number with repeating decimal, we can calculate the integer numerator and denominator. ...
    (sci.logic)
  • Re: Defining "<" for the rationals
    ... > than relation on rationals? ... and denominator share no positive integer factor larger than 1 AND that ... the denominator be positive. ... you can define your equivalence relation on the Cartesian ...
    (sci.math)
  • Re: forcing representation of denominator 1?
    ... (d (denominator odds))) ... (let* ((n (numerator odds)) ... CL rational numbers you could use rationals for pay-win ... risk-reward figures (for which you care about representation ...
    (comp.lang.lisp)
  • Re: printf "%.100f" the value 2 ^ (-127)
    ... > infinite set of rationals, would you find the member with the ... E.g. by trying a denominator of 1 first, ... makes the number fall into the range, if any (multiply range bounds by ... the denominator, and try surrounding integers). ...
    (microsoft.public.vc.language)
  • Re: (Not quite) Cantors diagonal proof
    ... >> Please read the replies to you made by Tim Peters, Dave Seaman, Daniel ... dismiss the kinds of objections raised to it. ... now that you see that the original list of reals ... you wouldn't have a proof that the set of rationals is ...
    (sci.math)