6

Gowers and Maurey proved in their remarkable paper(s), that there is a Banach space $X$ such that $X$ is isomorphic to its cube $X\oplus X\oplus X$ but not to isomorphic to its square $X\oplus X$. This space seems to be not a dual space, am I correct? Is there a solution to this problem which is a dual space?

And the second quick question: what is the status of the following conjecture:

There is a Banach space $X$ such that $X$ is not isomorphic to $X^2$ but $X^2$ is isomorphic to $X^3$?

Thank you very much. S.

Tomasz Kania
  • 11,291
Sellapan Nathan
  • 113
  • Related: http://mathoverflow.net/questions/10128 – Martin Brandenburg Dec 12 '11 at 17:46
  • 1
    I think both questions are open, Sellapan. For the first one, the closest result I know is

    Ferenczi, V.(F-PARIS6-E) A uniformly convex hereditarily indecomposable Banach space. (English summary) Israel J. Math. 102 (1997), 199–225,

    which contains a proof of the theorem in the title.

     – Bill Johnson Dec 12 '11 at 18:28
  • Thank you, I'll read it soon. May I ask one more question? Is there a pair of non-isomorphic Banach spaces, say, $X, Y$ such that $X^2$ is isomorphic to $Y^2$? – Sellapan Nathan Dec 12 '11 at 21:58
  • 1
    You "know" the answer to this last question, Sellapan. For $X$ take the Gowers-Maurey space you mentioned and for $Y$ take $X^2$. – Bill Johnson Dec 13 '11 at 00:37
  • But, on the other hand, Gowers' example seems to be reflexive. Is it? – Sellapan Nathan Dec 13 '11 at 20:27
  • 1
    Sorry; I don't know whether the answer to your first question is known. The basic Gowers-Maurey space is reflexive and I would guess so are the ones constructed in their second paper. Kevin Beanland, who participates on MO, might know.

    It is strange that Gowers and Maurey do not mention reflexivity in their papers. I guess it just did not occur to them.

     – Bill Johnson Dec 17 '11 at 01:01
  • 1
    I must admit, I've assumed, without verifying that the spaces defined by Gowers and Maurey are reflexive. They don't seem to mention it in the papers, but if it is true it shouldn't be hard to prove. Other than these papers, I know Zsak wrote at least two papers studying these construction and Galego, Elói Medina has a series of papers studying these types of spaces. You can also check out Gowers' ICM paper at http://www.dpmms.cam.ac.uk/~wtg10/papers.html called "Recent results in the theory of infinite dimensional Banach spaces" where he gives an overview of what they prove. – Kevin Beanland Dec 20 '11 at 13:55
  • The idea is, that he takes the space $X$ (the original Gowers-Maurey space $X$, which is claimed to be reflexive) and defines the space $E$ (the solution to the cube-not-square problem) as $E = X \oplus \mathbf{W}$, where $\mathbf{W}$ admits a Schauder decomposition into summands of the form $X\oplus_\infty X$ (hence reflexive), so $\mathbf{W}$ seems to be reflexive as a direct sum of reflexive spaces proving that $E$ is reflexive. Please correct me if I am wrong. – Tomasz Kania Dec 20 '11 at 14:06
  • Tomek: I don't think I understand. If you are talking about the definition of the cube-not-square space, it does not use the original GM space. The trick was to carry out the same construction as the GM space, all while ensuring that a certain algebra of shift operators where bounded. The boundedness of these operators yields the desired properties. The fact that the constructions are very similar is why I think the these spaces should be reflexive. – Kevin Beanland Dec 21 '11 at 10:28
  • 1
    Kevin: I think he claims that he takes [http://blms.oxfordjournals.org/content/28/3/297.full.pdf; page 298, par. 1] the space $X$ from the [W. T. Gowers and B. Maurey "The Unconditional Basic Sequence Problem"] which is reflexive [p. 869]. – Tomasz Kania Dec 21 '11 at 13:21
  • Tomek: I was confused. thanks for clarifying – Kevin Beanland Dec 21 '11 at 14:12
  • Do you agree that the Schauder decomposition of $W$ into summands of the form $X\oplus_\infty X$ gives the reflexivity of the cube-not-square solution? – Tomasz Kania Dec 21 '11 at 14:20

0 Answers0