It is well known that $\mathbb{C}$ and $\mathbb{C}_p$ are "algebraically" isomorphic (that is, ignoring the topology), but an isomorphism depends on the axiom of choice and there is no canonical way to relate and archimedean and a p-adic valuation. Still, doing so for all $p$ you have a field where you have an archimedean valuation, and can be equipped with a $p$-adic valuation for each $p$ which makes it complete, extending the archimedean and $p$-adic valuations of $\mathbb{Q}$ (or a finite extension thereof).

I was wondering if this can be done in order to build a field where a "product formula" for valuation holds $\prod_v |x|_v=1$, and which is complete for all valuations $|\cdot|_v$. Note that we cannot expect that $|x|_v=1$ for all $v$ except a finite set, or our field would be forced to be either an algebraic number field, either an algebraic function fields, by [Artin-Whaples, Axiomatic characterization of fields by the product formula for valuations]. But still, such a product could converge for all $x$.

I don't know if this may look like a "random question", but I was interested in understanding what are the limits in the study of statements about complex numbers via the reduction to $p$-adic equivalents, which sometimes happens in algebraic dynamics (and often can be made to work by just reducing to the dense set of algebraic numbers and using density).


Your Answer

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

Browse other questions tagged or ask your own question.