Showing ##\sqrt{2}\in\Bbb{R}## using Dedekind cuts

Click For Summary

Homework Help Overview

The discussion revolves around proving that ##\sqrt{2}\in\Bbb{R}## using Dedekind cuts. The original poster presents a specific cut in ##\Bbb{Q## and attempts to demonstrate that the square of this cut relates to the number 2.

Discussion Character

  • Exploratory, Assumption checking, Mathematical reasoning

Approaches and Questions Raised

  • Participants discuss the necessity of showing that every member of ##A^2 < 2## and explore the implications of the least upper bound property. There are suggestions to simplify the problem by focusing on positive rationals and considering the squares of elements in the original cut.

Discussion Status

Some participants have offered guidance on how to approach the problem, while others express uncertainty about specific elements being included in ##A^2##. The conversation reflects a mix of interpretations and attempts to clarify the relationship between the sets involved.

Contextual Notes

There are mentions of difficulties in selecting appropriate numbers for the proof and the complexity of the problem compared to initial expectations. Participants also note the importance of the least upper bound property in the context of Dedekind cuts.

Terrell
Messages
316
Reaction score
26
1. The problem statement, all variables and given
Prove that ##\sqrt{2}\in\Bbb{R}## by showing ##x\cdot x=2## where ##x=A\vert B## is the cut in ##\Bbb{Q}## such that ##A=\{r\in\Bbb{Q}\quad \vert \quad r\leq 0 \quad\lor\quad r^2\lt 2\}##.

I believe that I have to show ##A^2=L## however, it seems that ##L\nsubseteq A^2##. What am I doing wrong?

Homework Equations


Let ##r^*## denote the cut of ##r\in \Bbb{R}##.

The Attempt at a Solution


By definition of multiplying cuts, if ##x=A\vert B##, then ##x\cdot x## is ##A^2\vert F## such that ##A^2=\{r\in\Bbb{Q}\quad \vert \quad r\leq 0\quad \lor \quad \exists a,a'\in A(r=aa' \land a,a'\gt 0)\}##. Note that ##2^*=L \vert U=\{q\in\Bbb{Q}\vert q\lt 2\}\vert\{q\in\Bbb{Q}\vert 2\leq q\}##. If ##x\in A^2##, then either ##x\leq 0## or ##x\geq 0##. Suppose ##x\geq 0##, then ##\exists ]alpha,\alpha'\in A## s.t. ##x=\alpha\alpha'## and ##\alpha,\alpha'\gt 0##. Keep in mind that ##\alpha,\alpha'\in A## means that ##\alpha^2,(\alpha')^2<2## so ##\alpha^2(\alpha')^2<4 \Longrightarrow \alpha\alpha'\lt 2##. Thus, ##x^2 \lt 2## which implies ##x\in L##. On the other hand, if ##x\leq 0##, then clearly, ##x\in L##. In all cases, ##A^2\subseteq L##.
 
Physics news on Phys.org
I'm thinking you have to show that every member of ##A^2 < 2##. Are you able to do that? I would give a hint but I think this already is.
 
  • Like
Likes   Reactions: Terrell
if you want what i think is a nicer approach, the cauchy sequences approach due to cantor, it is in fundamentals of real analysis by sterling berberian. if you want to stick with the dedekind approach, i suggest you make things easier on yourself and consider only cuts of the positive rationals, where you can just assume the lower part of the square of a (positive) cut consists of the squares of the elements in the lower part of the original cut.
 
  • Like
Likes   Reactions: Terrell and fresh_42
verty said:
I'm thinking you have to show that every member of ##A^2 < 2##. Are you able to do that? I would give a hint but I think this already is.
I believe I did, then concluded with ##A^2\subseteq L##. I'm having trouble with ##L\subseteq A^2##. Since defining ##L## as the cut of ##2##, ##L=\{q\in\Bbb{Q}:q<2\}## meant ##3/2\in L##, but ##3/2\notin A^2##. Thanks!
 
mathwonk said:
i suggest you make things easier on yourself and consider only cuts of the positive rationals, where you can just assume the lower part of the square of a (positive) cut consists of the squares of the elements in the lower part of the original cut.
I believe this is the approach I took. Thank you.
 
Is it correct that the l.u.b. of ##A^2## is ##4##? Which would let me argue that ##\{a\in\Bbb{Q}:q\lt 2\}\subseteq A^2##?
 
Terrell said:
I believe I did, then concluded with ##A^2\subseteq L##. I'm having trouble with ##L\subseteq A^2##. Since defining ##L## as the cut of ##2##, ##L=\{q\in\Bbb{Q}:q<2\}## meant ##3/2\in L##, but ##3/2\notin A^2##. Thanks!

Sorry, I think this is more complicated than I expected. But regarding ##3 \over 2##, ##{3 \over 2} = {9 \over 8} \times {4 \over 3}##.
 
  • Like
Likes   Reactions: Terrell
verty said:
Sorry, I think this is more complicated than I expected. But regarding ##3 \over 2##, ##{3 \over 2} = {9 \over 8} \times {4 \over 3}##.
But ##3/2## cannot be in ##A^2##, right?
 
Terrell said:
But ##3/2## cannot be in ##A^2##, right?

##{9 \over 8}## is in A and ##{4 \over 3}## is in A, therefore ##{3 \over 2} = {9 \over 8} \times {4 \over 3}## is in ##A^2##. I don't yet know how to turn this idea into a proof. It seems difficult to select the right numbers.

PS. It seems you should be able to assume that cuts have the least upper bound property because that is usually proven before multiplication is defined.
 
Last edited:
  • Like
Likes   Reactions: Terrell
  • #10
verty said:
9898{9 \over 8} is in A and 4343{4 \over 3} is in A, therefore 32=98×4332=98×43{3 \over 2} = {9 \over 8} \times {4 \over 3} is in A2
Right. I should have tried with a calculator.
verty said:
It seems difficult to select the right numbers.
It's where I am having difficulty too.
Also, the book hinted that I can use the following to solve the problem. Though, I haven't figured how to use it.
hint,pugh.png
 

Attachments

  • hint,pugh.png
    hint,pugh.png
    7.9 KB · Views: 453

Similar threads

Replies
16
Views
2K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 14 ·
Replies
14
Views
3K
  • · Replies 28 ·
Replies
28
Views
3K
Replies
20
Views
4K
Replies
3
Views
4K
Replies
14
Views
3K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K