# Katz Modular Functions and Emerton's Completed Cohomology

Early this year, I started to learn about p-adic modular forms. Very recently, a mathematician tells me Emerton constructed an object called completed cohomology group with very rich structure, and the author could use it to prove fantastic results about Galois representations. (see Emerton's paper "Local-Global Compatibility In The p-adic Langlands Programme For $GL_{2/\Bbb{Q}}$". In page 45, he defines the huge $\hat{H}^{1}_{A}$)

With little understanding of it, can I ask if we can have a "natural" embedding of the space of Katz generalized p-adic modular functions (as defined in Section I.3 of Gouvea's Arithmetic of p-adic Modular Forms) into Emerton's completed cohomology group, assuming the base ring is $\Bbb {Z}_p$? If so, can we describe the image? This would be some type of Eichler-Shimura theorem for Katz generalized p-adic modular functions. Could this be one of the motivations when he defined such completed cohomology group?

-
You should check out Iovita's talk, "An overconvergent Eichler-Shimura isomorphism", at the IAS Galois representations and modular forms conference this past March, for some ideas of what results of this flavor might be true (and are provably true!). But I am fairly sure the answer to your actual question is "no", or at the very least "the actual relation between overconvergent modular forms and Emerton's completed cohomology groups is rather mysterious, aside from the fact that they seem to be producing the same Hecke eigenpackets in certain situations". –  David Hansen Jun 22 '11 at 3:57
Dear David. Somehow, I believe the answer is yes. –  ndk Jun 22 '11 at 4:18
Dear ndk: Fine, but why do you believe this? Take a look at the final section of Emerton's 2006 Inventiones paper, and the rather circuitous route by which he compares his construction for $GL2/\mathbf{Q}$ to the Coleman-Mazur eigencurve. Anyway, here is the video for Iovita's talk - video.ias.edu/graf/2011/03/21/iovita - which I for one found quite clear and enjoyable. –  David Hansen Jun 22 '11 at 5:10
Dear ndk, While Langlands's Antwerp paper was pivotal, it's not something one should read as a beginner. Have you read Scholl's paper in Inventiones? This follows Deligne's 1969 paper quite closely, but is more detailed and (I think) easier to understand for a newcomer. Another possibility is to learn the theory of modular symbols: this makes the cohomology of modular curves a quite concrete object, and so dispells some of the mystery. Of course, you won't see the Galois action concretely, but you will see Eichler--Shimura theory. I think there is something very satisfying about ... –  Emerton Jun 24 '11 at 11:58
... computing modular symbols for $X_0(11)$ (say) by hand, directly computing a matrix of $T_2, T_3$, etc., finding the eigenvalues, and then discovering the $q$-expansion of $q\prod_n (1-q^n)^2(1-q^{11n})2$ coming out, one coefficient at a time! It makes the objects in question appear much more concrete (or at least, it had that effect with me). Best wishes, and good luck, Matthew –  Emerton Jun 24 '11 at 12:01

You already have two helpful answers related to general aspects of Eichler--Shimura isomorphisms in a $p$-adic context. Here is an answer that more directly addresses your original question.

I will begin by recalling/stating some facts on the $p$-adic modular form side:

Fix a tame (i.e. prime-to-$p$) level $N$, and let $\mathbb T(N)$ be the completed Hecke algebra generated by the $S_{\ell}$ and $T_{\ell}$ for $\ell \nmid N p$ acting on Katz's space $V(N)$ of generalized $p$-adic modular fuctions of level $N$. Fix a maximal ideal $\mathfrak m$ in $\mathbb T(N)$, and let $V(N)_{\mathfrak m}$ be the localization of Katz's space at the maximal ideal $\mathfrak m$.

In fact, in order to deal sensibly with oldforms at $N$, I find it helpful to do the following: fix the finite set of primes $\ell_1,\ldots,\ell_n$ dividing $N$, and take a direct limit of $V(N)_{\mathfrak m}$ as $N$ ranges over all levels divisible by just $\ell_1,\ldots,\ell_n$. Note that $\mathbb T_{\mathfrak m}$ may grow as $N$ increases (if $N$ divides $N'$ then $\mathbb T(N)\_{\mathfrak m}$ is a quotient of $\mathbb T(N')\_{\mathfrak m}$), but eventually stabilizes (even thought the $V(N)_{\mathfrak m}$ don't stabilize), because if $\rho$ is any lift of the Galois representation $\overline{\rho}$ attached to $\mathfrak m$ then the difference between the prime-to-$p$ conductor of $\rho$ and $\overline{\rho}$ is bounded.

So now let $V_{\mathfrak m}$ be this direct limit of $V(N)\_{\mathfrak m}$, and let $\mathbb T_{\mathfrak m}$ be the Katz Hecke algebra acting on it. Note that $V_{\mathfrak m}$ is a smooth representation of the product $\prod_{i = 1}^n GL_2(\mathbb Q_{\ell_i}).$

Note also that there is an action of $U_p$ on $V_{\mathfrak m}$. (Let me reiterate that I did not include $U_p$ in my Hecke algebra $\mathbb T_{\mathfrak m}$!)

Now completed cohomology:

If we take completed cohomology at tame level $N$, we get a $p$-adically complete $\mathbb Z_p$-module $\widetilde{H}^1(N)$, with an action of $\mathbb T_{\mathfrak m}$, as well as of $G_{\mathbb Q}$ and $GL_2(\mathbb Q_p)$. We can complete it at $\mathfrak m$ to get $\widetilde{H}^1(N)\_{\mathfrak m}$.

If we then take the direct limit over all $N$ which are divisible exactly by $\ell_1,\ldots,\ell_n$, we get a module I'll denote $\widetilde{H}^1_{\mathfrak m}$, which has an action of $\mathbb T_{\mathfrak m}$, of $G_{\mathbb Q}$ and $GL_2(\mathbb Q_p)$, and also of the product $\prod_{i=1}^n GL_2(\mathbb Q_{\ell_i})$.

Now suppose that $\overline{\rho}$ satisfies some technical conditions, irreducibility being the most significant one. (The precise conditions are in the local-global compatibility paper that you mention. Note also that the irreducibility assumption eliminates the distinction between cohomology and cohomology with compact support, and --- more or less equivalently --- the distinction between working on closed vs. open modular curves.)

Then you can show that there is an isomorphism of $\mathbb T_{\mathfrak m}[G_{\mathbb Q}\times GL_2(\mathbb Q_p) \times \prod_{i = 1}^n GL_2(\mathbb Q_{\ell_i})]$-modules

$$\widetilde{H}^1\_{\mathfrak m} = \rho^u \otimes_{\mathbb T\_{\mathfrak m}} \pi^u \hat{\otimes}_{\mathbb T\_{\mathfrak m}} V\_{\mathfrak m}^{U_p =0}.$$ Here $\rho^u$ is the univeral modular deformation of $\overline{\rho}$ over $\mathbb T_{\mathfrak m}$, $\pi^u$ is a representation of $GL_2(\mathbb Q_p)$ on an orthonormalizable $\mathbb T_{\mathfrak m}$-Banach module constructed from $\rho^u_{| G_{\mathbb Q_p}}$ via the $p$-adic local Langlands, and $V_{\mathfrak m}^{U_p = 0}$ is, as indicated, the kernel of $U_p$ on $V\_{\mathfrak m}$. Also, the completed tensor product $\hat{\otimes}$ has to be suitably interpreted. (One should cut back to a fixed tame level $N$, then form the completed tensor product, and then take a direct limit over all $N$.)

So this is a kind of $p$-adic Eichler--Shimura isomorphism relating $p$-adically completed cohomology and $p$-adic modular forms.

It brings out the difference between the two sides quite clearly: on the completed cohomology side, we have a Galois action, which is encoded in the appearance of $\rho^u$. (This reflects the classical fact that every cuspform appears "twice" in cohomology.)

Also, completed cohomology has an action of $GL_2(\mathbb Q_p)$, while $p$-adic modular forms just have an action of $U_p$.

So to compare the two, we have to first get rid of the $U_p$-action on $p$-adic modular forms (which we do by passing to $U_p = 0$), and then add in a $GL_2(\mathbb Q_p)$-action, which we do by tensoring with $\pi^u$.

Note also that this isomorphism is not canonical. In this sense, it is analogous to looking at classical cohomology of modular forms with say $\mathbb Q$-coefficients, and modular forms with $\mathbb Q$-coefficients. These will be isomorphic as Hecke modules --- up to the issue of cuspforms appearing twice in cohomology --- but not canonically so. In order to make the Eichler--Shimura isomorphism canonical, one has to extend scalars to an appropriate period ring. Whether this is possible with completed cohomology I'm not sure about at the moment.

One more remark: trading in a $U_p$-action for a $GL_2(\mathbb Q_p)$-action is a fairly significant upgrading of structure, and this is why completed cohomology provides a useful tool for proving modularity theorems for Galois representations, over and above the already-existing theories of $p$-adic modular forms and $p$-adic modular symbols.

Added: You asked about motivation. The original motivation for defining completed cohomology was to construct eigenvarieties. Later it became clear that it was an important object in its own right, providing a global counterpart to the representations of $p$-adic groups that were beginning to appear as part of $p$-adic local Langlands. E.g. the theorem that locally algebraic vectors in cohomology are classical was first proved as an ingredient in the proof of an analogue of Coleman's "small slope implies classical" result for the eigenvariety constructed from completed cohomology. Only later was it realized that this could be combined with a local-global compatiblity result to prove modularity theorems for Galois representations.

Note that the rough relation with $p$-adic modular forms, namely that one gets the same Hecke algebra via either approach, was clear from the beginning, even though the more precise Eichler--Shimura-like statement above was not. Since eigenvarieties (as their name indicates) only care about Hecke eigenvalues, this meant that completed cohomology was good enough for constructing them.

-

I don't think I even know a natural non-zero map from the space $S_2(\Gamma_1(N);\mathbf{Q}_p)$ ("classical" modular forms with $p$-adic coefficients, defined for example as global sections of an appropriate sheaf on the modular curve $X_1(N)/\mathbf{Q}_p$) to the space $H^1(X_1(N),\mathbf{Q}_p)$ (etale or singular cohomology).

The problem is that the usual Eichler-Shimura isomorphism works when the coefficient fields are both the complex numbers, but the definition of the map itself involves an integral and is "transcendental" in nature. The two spaces above are, loosely speaking, $\mathbf{Q}_p$-subspaces of some $\mathbf{C}$-vector space, but they're not the same $\mathbf{Q}_p$-subspace -- the difference between them is some period, which is related to some $L$-value, which is probably not an algebraic number in general.

On the other hand, the systems of Hecke eigenvalues showing up in both spaces are the same, because they are isomorphic once you base extend to the complexes.

But if one can't write down a map between the finite-dimensional spaces, is there any hope of doing it with these fancy infinite-dimensional generalisations? This isn't a proof that one can or can't do it, but perhaps it's some food for thought.

So then you have to start thinking about this work of Iovita, as David already mentioned, and things get much more complicated quite quickly.

Coleman also wrote a paper called something like "a $p$-adic Eichler Shimura isomorphism" -- I forget the precise title -- but again you might find that it's not really what you're looking for.

-
Doesn't part of this works at least in the ordinary case though? Say the ordinary Hecke algebra is Gorenstein, then I think you do have such a canonical Eichler-Shimura map by the work of Mazur-Wiles and Ohta. All in all, this seems to me to indicate that indeed any natural such map will live in some $p$-adi period space, which happens to be your ring of coefficients in the ordinary case, so I guess I share the general pessimism. –  Olivier Jun 22 '11 at 8:23
Olivier -- I know of no such Eichler-Shimura map, even in the ordinary case. Can you give a precise reference? –  Kevin Buzzard Jun 22 '11 at 15:46
Sure. M.Ohta On the p-adic Eichler-Shimura isomorphism for $\Lambda$-adic cusp forms (Crelle 463)+the fact that the Hecke algebra is complete intersection (under the assumptions that you know) thanks to Wiles/Taylor-Wiles to have a canonical splitting of the short exact sequence Ohta considers. –  Olivier Jun 22 '11 at 17:20
Olivier -- thanks for bringing this paper to my attention -- I was unaware of it. Although I've only taken a superficial glance at it, it seems to me that Ohta does not do what the OP wants: the OP specifically wants something over $\mathbf{Z}_p$ (this is my reading of the question). Ohta base changes to the integers of a p-adically complete field containing all roots of unity, and such an integer ring must be very far from $\mathbf{Z}_p$. Somehow it's less surprising that one can do something now, because one is closer to being able to use $p$-adic Hodge theory to replace Shimura's... –  Kevin Buzzard Jun 22 '11 at 19:53
classical Hodge Theory. On the other hand I still find it a little surprising that one "only" needs to adjoin all roots of unity and then complete -- I should look at the paper more closely. I guess Tate got away with adjoining all $p$-power roots of unity and completing in his $p$-divisible groups paper though, so perhaps I shouldn't be so surprised. –  Kevin Buzzard Jun 22 '11 at 19:55

This is not an answer to the original question, since I do not yet know much about Matt's completed cohomology (but learning it seriously is high on my to-do list), but to the question as it has evolved after David's comment and Kevin's answer.

It is true that Eichler-Shimura provides no natural bijective map from $S_{k+2}(\Gamma,Q_p)^2$ to $H^1_p(\Gamma,V_k(Q_p))$, as it does over $C$: we need to choose a normaliwation of the period of each eigenform appearing in the source to get one. Above, $V_k(Q_p)$ is the symmetric $k$-th power of $Q_p^2$, and $H^1_p$ is the parabolic cohomology, image of $H^1_c$ into $H^1$.

However:

1) the fact that every system of Hecke eigenvalues appearing in $S_{k+2}(\Gamma,Q_p)$ also appears in $H^1_p(\Gamma,V_k(Q_p))$ (with twice the multiplicty), and conversely, extends to the infinite-dimensional generalizations, as follows: every system of Hecke eigenvalues with finite slope appearing in $S_{k+2}^{dag}(\Gamma,Q_p)$ also appears in $H^1_p(\Gamma,D_k(Q_p))$ (with twice the multiplicty), and conversely. Here $D_k$ is the space of distributions over $Z_p$ with a suitable action of $\Gamma$, that we assume contained in $\Gamma_0(p)$, depending on $k$.

This also makes sense for negative $k$, or even $p$-adic $k$, and is also true. A proof can be found in my prepint on critical $p$-adic $L$-function. It uses idea of Stevens and Chenevier. The result is actually more precise (see below), and actually there is a small exception for $k=0$.

2) Somehow (I don't know exactly how), Iovita, Andreatta, and Stevens have constructed a Hecke-equivariant map $S_{k+2}^{dag}(\Gamma,Q_p)^2 \rightarrow H^1_p(\Gamma,D_k(Q_p))$ (or just the finite-slope part of them, I am not sure) for all $k$ (integer or not) which agrees with some normalisation of the classical Eichler-Shimura isomorphism when $k$ is a positive integer and we restrict to the sub-space of classical forms.

3) The precise form of my result 1) implies that the map 2) (on finite slope parts) is an isomorphism if it is injective, or if it is surjective. It is not known for which values of $k$ this holds.

-
Joel -- for a general reader of this page, the task of finding the work you refer to as "my preprint" is made much harder by the fact that you don't tell anyone your last name! It's clear to me who you must be, but I will not say just in case you really do want to keep it a secret... –  Kevin Buzzard Jun 22 '11 at 17:20
You are right: my webpage is www.brandeis.edu/~jbellaic –  Joël Jun 22 '11 at 17:22
Salut Joël, ça va? –  Laurent Berger Jun 22 '11 at 18:17
Joel: the result (1) was announced by Glenn Stevens in his famous unfinished paper on overconvergent modular symbols that he wrote in the 90s :-) It's basically the last statement of the paper. Glenn never wrote the proof down, as far as I know, but I constructed a proof in the early 2000s ("true for classical forms and small slope forms are classical on both sides so one can argue by continuity") and emailed him to ask if this was his proof too. It was. I am very pleased someone has finally written it down! The control theorem for the overconvergent symbols was first proved in Glenn's preprin –  Kevin Buzzard Jun 22 '11 at 20:44
t, as far as I know (this is the same preprint that actually introduced the notion). I think it's a great shame that the document was never finished -- it hugely influenced my mathematical career. However my impression from your comments above is that now everything in it has been written down by others. –  Kevin Buzzard Jun 22 '11 at 20:45