Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

On $\mathbb{R}^2$, define $(a_1, a_2)\sim(b_1, b_2)$ if $a_1^2 + a_2^2 = b_1^2 + b_2^2$. Check that this defines an equivalence relation. What are the equivalence classes?

My work:

Let $a_1=b_1 = 1$ and $a_2=b_2 = 2$. We have $(1,2)\sim(1,2)\implies 5 = 5$ hence it is reflexive.

For symmetry, let $(a_1,a_2)\sim(b_1,b_2)$ then we should have $(b_1,b_2)\sim(a_1,a_2)$ to be satisfied the condition of symmetry. Like I did for reflexive, it is also symmetry.

Also this holds transitivity: let $c_1=1,c_2=2$.

I feel like I'm not on the right track and I don't know how to find the equivalence classes....

share|cite|improve this question
The equivalence classes correspond to circles centered at the origin of varying radii. This much is obvious. – oldrinb Sep 5 '13 at 1:24
up vote 2 down vote accepted

Most of what's presented in the question amounts to proofs by example; it's a logically fallacy, since one can prove untrue statements using proofs by example.

To prove reflexivity, your task is to show that $(a,b) \sim (a,b)$ for all $(a,b) \in \mathbb{R}^2$. You need to show that $$a^2 + b^2 = c^2 + d^2$$ holds whenever $(a,b)=(c,d)$.

To prove symmetry, you need to show that if $(a,b) \sim (c,d)$ then $(c,d) \sim (a,b)$ for all $(a,b),(c,d) \in \mathbb{R}^2$.

Try completing this: If $(a,b) \sim (c,d)$ then [some equation] holds, this implies [some other equation] holds, which implies that $(c,d) \sim (a,b)$.

To prove transitivity, you need to show that if $(a,b) \sim (c,d)$ and $(c,d) \sim (e,f)$ then $(a,b) \sim (e,f)$ for all $(a,b),(c,d),(e,f) \in \mathbb{R}^2$.

Try completing this: If $(a,b) \sim (c,d)$ then [some equation] holds. Further, if $(c,d) \sim (e,f)$, then [some other equation] holds. Together, these imply [some other equation] holds, which implies that $(a,b) \sim (e,f)$.

Finding a succinct way of describing the equivalence classes is often not straightforward. In this case, oldrinb's comment is quite helpful here:

The equivalence classes correspond to circles centered at the origin of varying radii.

Formally, the equivalence class containing $(a,b) \in \mathbb{R}^2$ is the set $$\{(c,d) \in \mathbb{R}^2:(a,b) \sim (c,d)\}$$ but this doesn't give much insight into what's in this class, and to why it's interesting. We can rewrite it $$\{(c,d) \in \mathbb{R}^2:c^2+d^2=k\}$$ where $k=a^2+b^2$, which is where oldrinb's hint comes from. (Note also that $(0,0)$ needs to be treated separately.)

share|cite|improve this answer
So I don't need to use norm? – therexists Sep 5 '13 at 3:40
You need to recognize them as circles. You can do that in your preferred way. I'd do it by recognizing that $x^2+y^2=k$ is the equation of a circle. – Rebecca J. Stones Sep 5 '13 at 3:53
Your helps totally helped me out. Thanks Rebecca. : ) – therexists Sep 6 '13 at 1:50
My pleasure! Thank you. – Rebecca J. Stones Sep 6 '13 at 1:50

Careful! The relation will be reflexive if it is true that $(a,b)\sim (a,b)$ for any pair $(a,b)$ you choose. Proving this is true for just one doesn't do it. But you know equality is already reflexive, right?

Think about complex numbers. If $z=a+bi$, then $|z|^2=\cdots$? What geometrical figure is $C_r=\{z\in \Bbb C:|z|=r\}$?

If you don't know about complex numbers, pick any $(a,b)$; and set $r^2=a^2+b^2$. Define $\lVert (a',b')\rVert=\sqrt{a'^2+b'^2}$. We're then saying that $(a,b)\sim (a',b')$ iff $\lVert (a,b)\rVert^2 =\lVert (a',b')\rVert^2$. So, the equivalence classes are of the form $$\widehat{(a,b)}=\left\{(x,y):\lVert (x,y)\rVert^2 =x^2+y^2=r^2\right\}$$

What does $\sqrt{x^2+y^2}$ measure for a point in $\Bbb R^2$?

share|cite|improve this answer
it might be more 'enlightening' for those in lower-level math to instead consider $\{(x,y)\in\mathbb{R}^2:x^2+y^2=r^2\}$ ;-) complex numbers may confuse him... +1 though – oldrinb Sep 5 '13 at 1:25
I found a website – therexists Sep 5 '13 at 1:28
This relates to my question, right? – therexists Sep 5 '13 at 1:28
@therexists I think you can answer that! =D – Pedro Tamaroff Sep 5 '13 at 1:28
@PeterTamaroff well of course I do not argue -- you are correct in both your points. My point was only that it may confuse the OP who may not immediately see the connection between the problems – oldrinb Sep 5 '13 at 1:29

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.