3
$\begingroup$

The following is a corollary of the Briançon-Skoda theorem:

If $R$ is a regular Noetherian ring of Krull dimension $d$ and $f_1,f_2,...,f_{d+1}\in R$. Then, $f_1^df_2^d...f_{d+1}^d \in (f_1^{d+1},f_2^{d+1},...,f_{d+1}^{d+1})R$

Now, given $R=k[[x_1,...,x_d]]$ where $k$ is a field, then $R$ is a regular local ring, so the corollary applies. I was wondering if there is direct proof of the above corollary for the case of the power series ring. I remember reading that there is a direct proof for the case when $f_1,...,f_{d+1}$ are polynomials. Can anyone point to a reference for this.

$\endgroup$
1
  • $\begingroup$ I took the liberty of adding the missing cedillas. (I also erased the corresponding apology.) $\endgroup$ Nov 16, 2010 at 4:35

1 Answer 1

3
$\begingroup$

It depends on what do you mean by "direct". There is no elementary proof, as far as I known, even for polynomials when $d=2$, see page 8 of this note. When $d=1$, the statement is an easy exercise: one can write $f_1= da, f_2=db$ with $(a,b) = R$ since $R$ is an UFD.

There are a few proofs of this very interesting theorem, analytic (the original one, over complex numbers), using duality theory (Lipman-Sathaye, for all regular rings) and reduction to characteristic $p$ + tight closure (Hochster-Huneke, when $R$ contains a field, see the note in the first paragraph for some exposition of this method) but I am not sure any of them can be called direct. Any new proof will be very exciting!

$\endgroup$
1
  • $\begingroup$ Thanks a lot Hailong for all the information. I shall look into these. $\endgroup$ Nov 16, 2010 at 5:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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