• 検索結果がありません。

in spaces of operators, II

N/A
N/A
Protected

Academic year: 2022

シェア "in spaces of operators, II"

Copied!
3
0
0

読み込み中.... (全文を見る)

全文

(1)

Comment.Math.Univ.Carolin. 35,2 (1994)259–261 259

On complemented copies of c

0

in spaces of operators, II

Giovanni Emmanuele*

Abstract. We show that as soon as c0 embeds complementably into the space of all weakly compact operators fromX toY, then it must live either inXor inY. Keywords: spaces of weakly compact operators, complemented copies ofc0

Classification: 46B25, 46A32

LetX andY be two infinite dimensional Banach spaces. It is well known (see for instance [E1], [E3], [E4], [EJ], [F], [H], [K]) thatc0 can embed intoK(X, Y), the space of all compact operators fromXtoY equipped with the operator norm, even if it does not embed intoX andY; furthermore, such a copy ofc0 can be complemented inK(X, Y) (see [E4], [E6]).

Recently ([E2], [E5]), we obtained some results proving that ifc0 embeds into either X or Y then it embeds complementably into some spaces of operators larger than K(X, Y), for instance W(X, Y), the space of all weakly compact operators fromX toY. The technique we used in order to construct the comple- mented copy ofc0requires the presence of a copy ofc0in eitherX orY, because otherwise it does not work.

All the above facts lead us to the following natural question: Is it possible to have a complemented copy of c0 inside W(X, Y) even when it does not embed intoX andY?

In this short note (in which we continue the research started in [E5]) we want to show that the answer to this question is negative; indeed, we prove that as soon asc0 embeds complementably intoW(X, Y), then it must live inside either X orY. Actually, we shall prove a slightly more general result about the space Lw(X, Y), i.e. the space of all weak-weak continuous operators fromX toY equipped with the operator norm.

The announced result is the following

Theorem 1. Let H be a complemented copy of c0 in Lw(X, Y). If Tn is a basis forH,then there is either ax0 ∈BX or a y0 ∈BY and a subsequence (Tnk)of(Tn)such that either the sequence (Tnk(x0))spans a copy of c0 in Y or the sequence(Tnk(y0))spans a copy ofc0 in X.

Proof: It is clear that for eachx ∈BX (resp.y∈BY) the seriesP

Tnk(x) (resp.P

Tnk(y)) is weakly unconditionally converging inY (resp. inX). It will

* Work performed under the auspices of GNAFA of CNR and partially supported by MURST of Italy (40%-1990).

(2)

260 G. Emmanuele

be enough to show that there is either ax0 ∈BX or a y0 ∈BY and a subse- quence (Tnh) of (Tn) such that either the seriesP

Tnh(x0) is not unconditionally converging inY or the seriesP

Tnh(y0) is not unconditionally converging inX, because we can thus use a well known result due to Bessaga and Pelczynski ([BP]) to conclude our proof. By contradiction we assume that for eachx ∈BX and y ∈BY the series P

Tn(x) and P

Tn(y) are unconditionally converging in Y and X, respectively. So for any ξ = (ξn) ∈ l and x ∈ BX the series PξnTn(x) is unconditionally converging inY. DefineTξ(x) =P

ξnTn(x) for allx ∈BX. We now show thatTξ belongs toLw(X, Y). To this aim it will be enough to consider a w-null net (xα) inBX and ay inBY and to prove that

(1) lim

α |Tξ(xα)(y)|= 0.

SinceP

ξnTn(y) is unconditionally converging inX by our assumption, we have

(2) lim

p sup

x∈BX∗

| X n=p+1

ξnTn(y)(x)|= 0.

Thanks to (2), givenγ >0 we can find ap∈N such that

(3) sup

α | X n=p+1

ξnTn(y)(xα)|<γ 2. On the other hand,

(4) lim

α

Xp

n=1

ξnTn(xα)(y) = 0

sinceTn∈Lw(X, Y), for alln∈N. (3) and (4) together give (1).

Furthermore, using the Closed Graph Theorem we can prove easily that the linear map Ψ :l → Lw(X, Y) defined by Ψ(ξ) =Tξ is bounded. It is clear thatK= Ψ(l) containsH. IfP :Lw(X, Y)−→H is the existing projection, the operatorP|KΨ : l −→ H is a quotient map of l ontoc0. This is a well known contradiction ([D]) that concludes our proof.

Corollary 2. Let c0 embed complementably into W(X, Y). Then c0 embeds into eitherX orY.

Proof: It is enough to observe thatW(X, Y) is isomorphic withLw(X∗∗, Y).

With a proof similar to that of Theorem 1 we can prove the same result for the spaceL(X, Y) of all bounded operators fromX toY. One could also consider the spaceU C(X, Y) of all unconditionally converging operators fromXtoY; in such

(3)

On complemented copies ofc0in spaces of operators, II 261 a case we have been able to get just a slightly less precise result than Theorem 1;

indeed, the same technique used for proving Theorem 1 shows that as soon as c0 embeds complementably in U C(X, Y), then either Y contains a copy of c0 or there are ay0 ∈ BY and a subsequence (Tnk) of (Tn) so that the sequence (Tnk(y0)) spans a copy of c0 in X, but in such a case we do not know how the copy ofc0 contained inY is spanned.

At the end, we observe that in the paper [E5] we also considered other spaces of operators, such as spaces of Dunford-Pettis operators; we do not know if The- orem 1 can be extended to cover this case.

References

[BP] Bessaga C., Pelczynski A,On bases and unconditional convergence of series in Banach spaces, Studia Math.17(1958), 151–164.

[D] Diestel J.,Sequences and Series in Banach Spaces, Graduate Text in Mathematics 97, Springer Verlag 1984.

[E1] Emmanuele G.,Dominated operators onC[0,1]and the(CRP), Collect. Math.41(1990), 21–25.

[E2] ,Remarks on the uncomplemented subspaceW(E, F), J. Funct. Analysis99(1991), 125–130.

[E3] ,On the containment ofc0 by spaces of compact operators, Bull. Sci. Math.115 (1991), 177–184.

[E4] ,A remark on the containment ofc0 in spaces of compact operators, Math. Proc.

Cambridge Phil. Soc.111(1992), 331–335.

[E5] ,On complemented copies ofc0 in spaces of operators, Comm. Math.32(1992), 29-32.

[E6] ,About the position of Kw(X, Y)inLw(X, Y), Atti Sem. Mat. Fisico Univ.

Modena, to appear.

[EJ] Emmanuele G., John K.,Uncomplementability of spaces of compact operators in larger spaces of operators, to appear.

[F] Feder M.,On subspaces of spaces with an unconditional basis and spaces of operators, Illinois J. Math.34(1980), 196–205.

[H] Holub J.R.,Tensor product bases and tensor diagonals, Trans. Amer. Math. Soc. 151 (1970), 563–579.

[K] Kalton N.J.,Spaces of compact operators, Math. Annalen208(1974), 267–278.

Department of Mathematics, University of Catania Viale A. Doria 6, 95125 Catania, Italy

E-mail: [email protected]

(Received April 27, 1993)

参照

関連したドキュメント

[5] to integrate directly the regularization into the equation by convolving the image with the Gaussian filter on the gradient of the noisy image to smooth the image first in order

These objects include not only axioms, proof rules, proofs, definitions, theorems, and theories but must also accommodate inspection of algorithmic content which includes

2. Garay B.M., Hofbauer J., Robust permanence for ecological differential equations, mini- max, and discretizations, SIAM J. Hirsch M.W., Smith H.L., Zhao Xiao-Qiang,

The repeated homogeneous balance method is used to construct new exact traveling wave solutions of the (2+1) dimensional Zakharov- Kuznetsov (ZK) equation, in which the

In Section I we shall prove the theorems of integral operators that are bounded and completely continuous in H¨older spaces with weight.. Section II will contain the proof of

Turmetov; On solvability of a boundary value problem for a nonhomogeneous biharmonic equation with a boundary operator of a fractional order, Acta Mathematica Scientia.. Bjorstad;

We study these spaces and the C ∗ -algebra (in the usual sense) of bounded operators on them which have an adjoint.. Hilbert space and we study the associate C

The commutative case is treated in chapter I, where we recall the notions of a privileged exponent of a polynomial or a power series with respect to a convenient ordering,