• 検索結果がありません。

TOWARD WAVE MODELS OF REPRESENTATIONS OF REAL SEMISIMPLE LIE GROUPS

N/A
N/A
Protected

Academic year: 2021

シェア "TOWARD WAVE MODELS OF REPRESENTATIONS OF REAL SEMISIMPLE LIE GROUPS"

Copied!
3
0
0

読み込み中.... (全文を見る)

全文

(1)

TOWARD

WAVE

MODELS

OF REPRESENTATIONS

OF REAL

SEMISIMPLE

LIE

GROUPS

by Takayuki Oda $\not\in \mathrm{R}*^{r}\backslash \mathrm{H}\hslash_{1}^{*}\mathfrak{c}\mathrm{k}_{\backslash }\mathcal{T}\backslash \cdot\#\backslash \mathrm{g}_{\mathrm{f}}\mathfrak{x}$

)

This is an introduction for a nulnber of talks given in this conference.

A typical situation to constructautomorphic $\mathrm{L}$-funcions froIn automorphic forms,

is the following.

Weare given asemisimplealgebraic group $G$overafield$k$ofKroneckerdimension

1. Sometime $G$ is assumed to be quasi-split. $\mathrm{T}\}_{1}\mathrm{i}_{\mathrm{S}}$ is

$\mathrm{P}^{\mathrm{a}\mathrm{r}\mathrm{t}\mathrm{i}\mathrm{c}\mathrm{U}111}\mathrm{a}\cdot \mathrm{y}$ thc case, when

one use Whittaker models. Also a closed algebraic subgroup $R$of $G$ is given which

is large enough so that it is spherical.

Assume that $G$ is quasi-split for simplicity. Fix a minimal parabolic subgroup $P$

of $G$ over $k$. Then the $\mathrm{s}\mathrm{u}\mathrm{b}_{\circ}\sigma \mathrm{r}\mathrm{o}\mathrm{u}\mathrm{p}R$ of $G$ is called spherical, $\mathrm{w}1_{1}\mathrm{e}\mathrm{n}$ the orbit of the

extension of scalars $\overline{R}=R\otimes_{k}\overline{h^{\wedge}}$ of $R$ on thc flag manifold $G\Theta\overline{k^{\wedge}}/P\otimes\overline{k}$ has open

orbit. As shown by $\mathrm{M}$ Brion $([\mathrm{B}1])$, this condition is equivalent to the follwing.

$(^{*})$ The restriction to $\overline{R}$ of any finite dimensional irreducible representation of $\overline{G}$

is multiplicity-free.

Also as shown in [B1], this condition implies and is equivalent to the condition that the double coset $\overline{R}\backslash \overline{G}/\overline{P}$is finite.

Remark. The notion of”spherical” seems to be discussed only for the case ofgroups

over algebraically closed fields, in the literature. Of $\mathrm{c}o$urse one can generalize this

concept for any $k$, and can use the terminology ”$k$-spherical groups”, if one can

provc any non-trival statement about that...

Now, let us replace the global field $k$ by its completion at a $\mathrm{f}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{e}_{\vee}$ place, which

we also denote by the same $k$. $\mathrm{T}\mathrm{h}\mathrm{e}_{\vee}k$-valued points of algebraic groups $G,$$P,$ $R$ are

denoted by the same symbol by an abuse of symbol. An optimist might hope the following.

Folklore. Given an a$d\mathrm{m}$issi$\mathrm{b}le$

irre

ducible representation

$\eta$ ofR. Let $C_{\eta}^{\infty}(R\backslash G)=$

$Ind_{R()}^{G}\eta bc$ the $s\mathrm{m}$ooth induction of

$\eta$ to G. Then for any irreducible admissible

representation $\pi$ of$G$, the dimension ofin$t$ertwin$ig$space

$Hom_{G(\pi},$$C\infty(\eta\backslash RG))$

is at most 1.

In view of Frobenius reciprocity with respect $G$ and $R$, this is no other than a

kind of multiplicity free statement for the restriction of $\pi$ to the subgroup $R$.

数理解析研究所講究録

(2)

When $k$is a

$p$-adic fielf, this problemis consideredbyMurase and Suganoin their

work (see their articles in this proceedings) on the $\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{S}\mathrm{t}\mathrm{r}\mathrm{u}\mathrm{C}\mathrm{t}\mathrm{i}_{\mathrm{t}}$) $1\mathrm{l}\langle$

$)\mathrm{f}$automorphic

L-functions, under the name ”Shintani function”. Among others $\dagger 1_{1\mathrm{t}^{\backslash }},\mathrm{y}$ (and S. Kato)

now have explicit formula of $K$-invariant vector in the$\mathrm{i}\mathrm{m}\mathrm{a}\mathrm{g}\mathrm{t}^{\backslash }()\{\uparrow$he unique element

$\Phi\in Hom_{G(}\Pi,$$c\infty\eta(R\backslash G))$, when $\pi$ is of class 1 (i.e. as $\mathrm{s}_{1^{\mathrm{J}1\iota 11}}’ \mathrm{i}$(.ill principal series

representation), in rather general situation.

Naive analogue at archemean places does not seem to hold in general. The typi-cal counter-example is the case when $R$is a maxilnal unipotent subgroup and $\eta$ is a

non-degenerate character. Then the above space is the space of Whittaker vectors

(or functionals if you prefer this terminology). As shown by Kostant $([\mathrm{K}])$, the

di-mension of the above intertwing spacefor irreducibleprincipal series representation

of $G$ is the order of the (little) Weyl group. In this case, to have multiplicity one

statement, we have impose an increasing condition for the functions in target space

$C_{\eta}^{\infty}(R\backslash c)$, to replace it by much smaller space. As shown by Wallach $([\mathrm{t}\mathrm{V}])$, we

have multiplicity-free statement after this modification.

At the real place, Yamashita [Y] proved various sufficient conditions for the finiteness of the above intertwining space. Because we need further terminology to describe his results, we do not give the details of his results here. They cover

various important cases in application. I recommend the readers to consult with

the original paper. Here we quote a theorem of [B-O] wllic.h is easy to state. Theorem. (Bien-Oshima [B-O]) Set $k=$ R. Assume that $R_{\mathrm{C}}$ has open

or-bits on the Hag manifold $G_{\mathrm{C}}/P_{\mathrm{C}}$. Then for any finite-dimensional

representa-tion $\eta$ of$R$ and any irreducible admissible $(\mathrm{g}, K)- m$odule $\pi$, the interwining$sp$ace

$Hom\mathfrak{g},K(\pi, c_{\eta}\infty(R\backslash G))$ is of finite dimension.

Now let me explain the titles of some talks given here. But before that recall the following. When $G=S_{p}($2 : $\mathbb{R}),$ Miyazaki-Oda$[]$ computed explicitely the

holonomic system for the radial part of the principal series Whuittaker functions with smallest $K$-types. $\mathrm{O}\mathrm{d}\mathrm{a}[]$ and Miyazaki-Oda$[]$ give integral expressions of

Whittaker function with the oomer $K$-type for the discrete series representations

and generalized principal series for the cuspidal parabolic subgroup corresponding to long root.

The talk of$\mathrm{I}\mathrm{i}\mathrm{d}\mathrm{a}_{\mathit{1}}$

(Masatoshi) will discuss the$\mathrm{h}\mathrm{o}1_{01\mathrm{n}\mathrm{o}\mathrm{r}}\mathrm{P}\mathrm{h}\mathrm{i}\mathrm{c}$ system of the radialpart

of the matrix coefficients with smallest $k$-types of the principal series representation

of $Sp(2;\mathbb{R})$, and this shift operates. Oda’s talk is on the holonomic system of the

matrix coefficients with minimal $I1’$-type of the large discrete series representations

of$Sp(2;\mathbb{R})$.

Takuya Miyazaki will talk about the holonomic system for the radial part of

generalized Whittaker functions on $Sp(2;\mathbb{R})$ with respect to the Siegel parabolic

subgroup.

Hayata (Takahiro) will treat principal seriesWhittakerfunction on $SU(2,2)$ with smallest K-type.

(3)

Taniguchi (Kenji) works out completely the Whittaker functions with minimal

$I’\backslash$-types of the discrete series on real unitary group $SU(n, 1)$ of split-rank 1 (and

recently he settled also the case of Spin$(2n, 1))$.

Tuzuki considers the case when $G=SU(2,1)$ and$H=S(U(1,1)\cross U(1)\cong U(1,1)$

when $?l$ the representations $\eta$ of $H$ are of

infinitc-dimension.

REFERENCES On spherical subqroups

The following references are noticed by Prof. T. Matsuki

[Brl] Brion, M.: Quelques proprie’te’s des espaces homog\‘enes$\mathrm{s}\mathrm{p}\mathrm{h}\mathrm{e}^{\text{ノ}}\mathrm{r}\mathrm{i}\mathrm{q}\mathrm{u}\mathrm{e}\mathrm{S}$, Manuscrip

Math. 55 (1986) 191-198.

[Br2] Brion, M.: Classification des espaces homog\’enes sph\’eriques, Comp. Math. 63 (1987)

189-208.

[K] Kr\"amer, M.: $\mathrm{S}\mathrm{p}\mathrm{h}\ddot{\mathrm{a}}\mathrm{r}\mathrm{i}_{\mathrm{S}}\mathrm{c}\mathrm{h}\mathrm{e}$ Untergruppen in Kompakten zusanlmenh\"angenden

Liegruppen Comp. Math. 38 (1979) 129-153.

The section 4 of

[M] $\mathrm{I}\mathrm{v}\mathrm{I}\mathrm{a}\mathrm{t}\mathrm{S}\mathrm{u}\mathrm{k}\mathrm{i}$, T.: Orbits on Flag Manifolds, Proc. of Intern. Congress ofMath.,

Kyoto (1990), 807-813.

is illuiminating. Here is amuch simpler proofofthemaintheorem of [Brl]. The,

conjecture 1,2 on real spherical subgroups given there are solved

by.

B. Kimelfeld and F. Bien.

On $multiplicit\cdot U$

finiteness

[W] Wallach, $\mathrm{n}.$: Asymptotic expansions of generalized matrix entries of $\mathrm{r}\mathrm{e}_{1^{)\mathrm{r}\mathrm{e}-}}$

sentations of real $\mathrm{r}\mathrm{e}\mathrm{d}\mathrm{u}\mathrm{c}\uparrow\partial \mathrm{i}\mathrm{l}\Gamma \mathrm{e}$ groups. Lie Group Representations I. Lecture

Notes in Math. 1024, Springer (1984), 284-369.

[B-O] Bien, F. and Oshima, T.: Multiplicities of induced representations of semisim

ple his groups, preprint.

[Y] Yamashita, H.: Criteria for the finiteness of restinction on - modules to

subalgebras and application to Harish-Chandra modules. J. Funct. Annual. 121 (1994),

296-329.

参照

関連したドキュメント

In the study of properties of solutions of singularly perturbed problems the most important are the following questions: nding of conditions B 0 for the degenerate

7, Fan subequation method 8, projective Riccati equation method 9, differential transform method 10, direct algebraic method 11, first integral method 12, Hirota’s bilinear method

In [12], as a generalization of highest weight vectors, the notion of extremal weight vectors is introduced, and it is shown that the uni- versal module generated by an extremal

to use a version of Poisson summation with fewer hypotheses (for example, see Theorem D.4.1 in [1])... It seems surprisingly difficult to prove directly from the definition of a

Also, if G has real rank at least 3, we provide a C ∞ classification for volume-preserving, multiplicity free, trellised, Anosov actions on compact

The following result about dim X r−1 when p | r is stated without proof, as it follows from the more general Lemma 4.3 in Section 4..

A conjecture of Fontaine and Mazur states that a geo- metric odd irreducible p-adic representation ρ of the Galois group of Q comes from a modular form ([10]).. Dieulefait proved

Another technique we use to find identities is the repre- sentation theory of the symmetric group. The process of studying identities through group representations is indi- rect