I see now that Cartier duality for groups more general than finite group schemes is discussed here:

Generalization beyond finite group schemes is discussed in

- Amelia Álvarez Sánchez, Carlos Sancho de Salas, Pedro Sancho de Salas, _Functorial Cartier duality (arXiv:0709.3735)

and in

- Dima Arinkin, appendix in Ron Donagi, Tony Pantev,
*Torus fibrations, gerbes, and duality*, (arXiv:math/0306213)

The page Dual abelian variety has the statement

The n-torsion of an abelian variety and the n-torsion of its dual are dual to each other when n is coprime to the characteristic of the base. In general - for all n - the n-torsion group schemes of dual abelian varieties are Cartier duals of each other.

but again that’s not what you want.

]]>Thanks for offering help, David!

Yeah, this doesn’t cover the case I was looking for, this is still finite group schemes (over $Spec A$).

Hm, maybe what I was hoping for does not exist…

]]>Theorem 4.63 of that book discusses Cartier duality for locally free commutative group schemes (see also here in the author’s other book, which I suspect is a copy/paste on the part of the author), but I suspect this doesn’t cover the case you need.

]]>Sorry, I am sure my reading did not cover that case…you need some expert…

]]>Thanks. I’d need it specifically for elliptic curves over rings. I see that there is a book titled “Elliptic Curves and Arithmetic Invariants” which has a section 6.1.6 that sounds like it should talk about Cartier duality of elliptic curves over rings. But I cannot open that book right now (am on a train with shaky and slow connection…)

]]>I have seen some reference days ago, but having read so much last one month it is difficult to remember. I think it was about the case of pro-finite groups.

]]>Is there anything in the literature on generalizing Cartier duality to non-finite group schemes? Pointers would be welcome.

]]>