I know that the two norms: $p$ - adic norm and the usual norm ($\left| \cdot \right|$) defined on $\mathbb{Q}$ are not quivalent. This is clearly because the $p$ adic norms staisfies the strong triangle inequality while the usual norm does not. However, it contradicts a theorem that I have studied.

"On a finite dimensional vector space, any two norms are equivalent."

We know that a field over itself is a one - dimensional vector space and therefore, $\mathbb{Q}$ is a vector space. From the above statement $p$ adic norm and usual norm should be equivalent.

I know that there is something hiding behing the "$p$ - adic" norm, rather numbers. But I am not able to figure it out. I would like some help regarding the same.


The theorem you quote carries a bunch of implicit assumptions, more specifically, it only considers special kinds of norm. Specifically, the norms it concerns are norms $\|\cdot\|$ on a finite-dimensional vector space $V$ over real numbers, which further satisfies the homogeneity property $\|av\|=|a|\|v\|$ for all $v\in V,a\in\mathbb R$.

As you see, the example of the standard and of the $p$-adic norm show that this is not true much more generally. In fact, both requirements are necessary: the norm must both be homogeneous (otherwise there are wild counterexamples even on $\mathbb R$) and the vector space must be over $\mathbb R$ (we can find homogeneous examples over $\mathbb Q$. Actually, some other fields will work in place of $\mathbb R$, for instance $p$-adic numbers $\mathbb Q_p$, but it's an exception more than a rule).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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