MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I start to read the paper "On p-adic analogues of the conjectures of Birch and Swinnerton-Dyer" by Mazur, Tate and Teitelbaum (referred as [MTT]) to learn how we can associate p-adic L-function to certain eigenforms.

For an eigenform $f$ (with certain conditions) and a choice of the root $\alpha$, the authors define the $V_f$-valued measure $\mu_{f,\alpha}$ in page 13. Here $V_f=C_p\otimes_\bar{Q}L_f\bar{Q}$ is a finite dimensional $C_p$-vector space defined in line 5 of page 13. And $L_f\bar{Q}$ (I think) is the $\bar{Q}$-vector subspace of $C$ generated by elements of $L_f$, where $L_f$ is defined right before the proposition in page 6.

Finally, they define the p-adic L-function $$L_p(f,\alpha,\chi,s):=L_p(f,\alpha,\chi\chi_s)$$ as in page 19.

Apparently, this function has values in the $C_p$-vector space $V_f$. But in the rest of the paper, it seems to me that they treat this as a $C_p$-valued function. On the other hand, other references, such as Greenberg-Stevens' paper in Inventiones 1993, and the earlier paper ``Arithmetic of Weil Curvess'' by Mazur--Swinnerton-Dyer use the two periods $\Omega_f^{\pm}$ to get out two measures $\mu_f^{\pm}$. Then use them to define a $C_p$-valued function.

Question: do we need to change the functions in [MTT] into $C_p$-valued functions, say, by using the above periods as in Greenberg-Stevens?

share|cite|improve this question
up vote 6 down vote accepted

In order to get $\mathbb{C}_p$-valued functions, you need to choose a basis for the vector space $V_f$. If $f$ corresponds to an elliptic curve, there is a reasonably canonical way of doing this (using the periods of a Neron differential), as in the paper by Mazur and Swinnerton-Dyer. If $f$ is a more general modular form it is much less clear what the "right" normalisation is for a basis of $V_f$. This is an important issue, though, because one needs to fix such a normalisation to make sense of whether or not two L-values are congruent modulo a prime. The question has been studied in detail by Vinayak Vatsal in his paper "Canonical periods and congruence formulae" (Duke Math Journal 98 no. 2, 1999), which determines a canonical normalisation up to p-adic units.

share|cite|improve this answer

Your Answer


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.