Take the 2-minute tour ×
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.

Could you please confirm if this proof is correct?

Theorem: If $q \neq 0$ is rational and $y$ is irrational, then $qy$ is irrational.

Proof: Proof by contradiction, we assume that $qy$ is rational. Therefore $qy=\frac{a}{b}$ for integers $a$, $b \neq 0$. Since $q$ is rational, we have $\frac{x}{z}y=\frac{a}{b}$ for integers $x \neq 0$, $z \neq 0$. Therefore, $xy = a$, and $y=\frac{a}{x}$. Since both $a$ and $x$ are integers, $y$ is rational, leading to a contradiction.

share|improve this question
1  
Looks good but what happened to the $z$ and $b$ in the line $xy = a$? –  Eric O. Korman Jun 13 '11 at 12:59
    
It's surely not quite correct. For example, you missed factors of $1/z$ and $1/b$ in evaluating or "simplifying" $xy/z=a/b$. Otherwise the logic is OK. –  Luboš Motl Jun 13 '11 at 13:01
    
You just get the relation $y=\frac{za}{xb}$ which is a ratio of integers and therefore rational. –  Beni Bogosel Jun 13 '11 at 13:17
    
@Eric: Hm, I assumed that if $\frac{a}{b}=\frac{c}{d}$ then $a=c$. Which is not true... –  persepolis Jun 13 '11 at 13:24
    
Minor point: this is not a proof by contradiction, you prove that qy is irrational by proving that it is not rational, this is just the definition of being irrational. –  Guillaume Brunerie Oct 28 '11 at 16:11

6 Answers 6

up vote 2 down vote accepted

It's wrong. You wrote $\frac{x}{z}y = \frac{a}{b}$. That is correct. Then you said "Therefore $xy = a$. That is wrong.

You need to solve $\frac{x}{z}y = \frac{a}{b}$ for $y$. You get $y = \frac{a}{b} \cdot \frac{z}{x}$.

share|improve this answer

You can directly divide by $q$ assuming the fact that $q \neq 0$.

Suppose $qy$ is rational then, you have $qy = \frac{m}{n}$ for some $n \neq 0$. This says that $y = \frac{m}{nq}$ which says that $\text{y is rational}$ contradiction.

share|improve this answer

A group theoretic proof: You know that if $G$ is a group and $H\neq G$ is one of its subgroups then $h \in H$ and $y \in G\setminus H$ implies that $hy \in G\setminus H$. Proof: suppose $hy \in H$. You know that $h^{-1} \in H$, and therefore $y=h^{-1}(hy) \in H$. Contradiction.

In our case, we have the group $(\Bbb{R}^*,\cdot)$ and its proper subgroup $(\Bbb{Q}^*,\cdot)$. By the arguments above $q \in \Bbb{Q}^*$ and $y \in \Bbb{R}\setminus \Bbb{Q}$ implies $qy \in \Bbb{R}\setminus \Bbb{Q}$.

share|improve this answer
    
Shouldn't it be $\mathbb{R}^{\ast} \setminus\mathbb{Q}^{*}$ –  user9413 Jun 13 '11 at 13:29
    
yes, but you can see that $\Bbb{R}\setminus \Bbb{Q}=\Bbb{R}^*\setminus \Bbb{Q}^*$. –  Beni Bogosel Jun 13 '11 at 13:59
    
You are right, great answer :) –  user9413 Jun 13 '11 at 14:02

As I mention here frequently, this ubiquitous property is simply an instance of complementary view of the subgroup property, i.e.

THEOREM $\ $ A nonempty subset $\rm\:S\:$ of abelian group $\rm\:G\:$ comprises a subgroup $\rm\iff\ S\ + \ \bar S\ =\ \bar S\ $ where $\rm\: \bar S\:$ is the complement of $\rm\:S\:$ in $\rm\:G$

Instances of this are ubiquitous in concrete number systems, e.g.

enter image description here

share|improve this answer

Let's see how we can modify your argument to make it perfect.

First of all, a minor picky point. You wrote $$qy=\frac{a}{b} \qquad\text{where $a$ and $b$ are integers, with $b \ne 0$}$$

So far, fine. Then come your $x$ and $z$. For completeness, you should have said "Let $x$, $z$ be integers such that $q=\frac{x}{z}$. Note that neither $x$ nor $z$ is $0$." Basically, you did not say what connection $x/z$ had with $q$, though admittedly any reasonable person would know what you meant. By the way, I probably would have chosen the letters $c$ and $d$ instead of $x$ and $z$.

Now for the non-picky point. You reached $$\frac{x}{z}y=\frac{a}{b}$$ From that you should have concluded directly that $$y=\frac{za}{xb}$$ which ends things, since $za$ and $xb$ are integers.

share|improve this answer

I don't think it correct. It seems like a good idea to indicate both x as an integer, and z as a non-zero integer. Then you also want to "solve for" y, which as Eric points out, you didn't quite do.

share|improve this answer

Your Answer

 
discard

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.