1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: Proof involving dedekind's cuts

  1. Sep 12, 2012 #1
    1. The problem statement, all variables and given/known data
    Show that for any non-empty sets A and B as in (P13') and for any [itex]n \in N[/itex],
    there is a pair (a,b) with [itex]a\in A[/itex] and [itex]b\in B[/itex] such that [itex]|b-a|=b-a<\frac{1}{n}[/itex].

    2. Relevant equations
    Property (P13'). For any two non-empty sets A and B of rational numbers such that
    Q = [itex]A\cup B[/itex], if all elements of A are less than all elements of B, then there is a unique
    [itex]x\in R[/itex] s.t. for any [itex]a\in A[/itex] and any [itex]b\in B[/itex] we have [itex]a\leq x\leq b[/itex].

    This is called P13' because I am currently learning from Spivak and have finished chapter 1-5 where we get P1-P12. P13 is in a later chapter so I haven't seen it yet. P13' was given to me to help introduce me to Dedekind's cuts.

    3. The attempt at a solution
    I have considered the pairs like (a,b) where [itex]a=a[/itex], [itex]b=a+\frac{1}{n+1}[/itex] or [itex]a=b-\frac{1}{n+1}[/itex], [itex]b=b[/itex] but this isn't correct. My professor says there is a very easy proof but I have thought about this so much that my mind is clogged and I can no longer see clearly. Could someone please point me in the right direction?

    Edit: Taking another look at the problem, I feel like maybe I could use epsilon delta to prove that the limit goes to 0 as a and b get arbitrarily close? Since |b-a|=b-a<1/n is talking about distance after all... hmm. Any comments on this?
    Last edited: Sep 12, 2012
  2. jcsd
  3. Sep 12, 2012 #2
    I have yet again taken another look and now I am thinking, since we have for any [itex]a\in A[/itex] and [itex]b\in B[/itex] that we have a unique [itex]x\in R[/itex] such that [itex]a\leq x\leq b[/itex] then would it be possible for my to take say a certain x from one pair and subtract in from an x of another pair to show that it'll be less than 1/n? I'm not sure how I would make a proof from that.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook