3
$\begingroup$

Recently I came across a proof of the infinite product for $\sin z$ (https://www.sciencedirect.com/science/article/pii/0022247X77902347). It applies the fundamental theorem of algebra to $$p_{n}(z)=\dfrac{1}{2}\left(\left(1+\frac{z}{n}\right)^{n}-\left(1-\frac{z}{n}\right)^{n}\right),$$ which it factors to a product. But it also states that $n=2m+1$ and lets $n$ and $m$ go to infinity.

Then how is the factorization to a product, by applying the fundamental theorem of algebra, possible?

I'm confused because the fundamental theorem of algebra holds only for finite polynomials, but $p_{n}(z)$ is not a finite polynomial if we let $n$ go to $\infty$.

$\endgroup$
4
  • 1
    $\begingroup$ It is a standard fact that $a^n-b^n=(a-b)(a^{n-1}+\cdots+b^{n-1}) = \prod_{k=0}^{n-1} (a - z^k b)$ where $z:=e^{2\pi i/n}$ is a primitive $n$-th root of unity. $\endgroup$ – Somos May 29 '19 at 1:13
  • $\begingroup$ I'm familiar with roots of unity but can't find a proof of $(a-b)(a^{n-1}+\cdots +b^{n-1})=\prod_{k=0}^{n-1}\left(a-be^{\frac{2k\pi i}{n}}\right).$ $\endgroup$ – Abo Rakan May 29 '19 at 12:06
  • $\begingroup$ Even if it is proved, how can we let $n$ go to $\infty$? Using the roots of unity, we assumed finite $n$, didn't we? $\endgroup$ – Abo Rakan May 29 '19 at 12:19
  • 1
    $\begingroup$ They are both equal to $a^n-b^n$. In limit proofs we always let $n\to\infty$. For example, $2^{-n}\to0$ as $n\to\infty$ where $n$ is always finite. $\endgroup$ – Somos May 29 '19 at 12:22
0
$\begingroup$

You are of course right that

the fundamental theorem of algebra holds only for finite polynomials,

I suspect that if you read that paper carefully you would find that Eberlein's purpose is (in part) to prove rigorously that Euler's seemingly casual "letting $n$ go to infinity" produces mathematics that's correct by 20th century standards.

$\endgroup$
2
  • $\begingroup$ But what permits the author to do the factorization? $\endgroup$ – Abo Rakan May 28 '19 at 23:01
  • $\begingroup$ Sorry I can't help you with details. I took just a quick look at the paper. $\endgroup$ – Ethan Bolker May 29 '19 at 0:14

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.