Interpolating hereditarily indecomposable Banach spaces

From MaRDI portal



Abstract: It is shown that every Banach space either contains ell1 or it has an infinite dimensional closed subspace which is a quotient of a H.I. Banach space.Further on, Lp(lambda), 1<p<infty, is a quotient of a H.I Banach space.


A Banach space \(X\) is said to be hereditarily indecomposable if for any pair of closed infinite-dimensional subspaces \(Y\subset X\), \(Z\subset X\), with \(Y\cap Z= \{\emptyset\}\) the subspace \(Y+ Z\subset X\) is not closed. Such spaces are structurally irrelevant to classical Banach spaces with unconditional basis, in particular Hilbert spaces. The very detailed investigation of the structure of such spaces given in the paper leads to the conclusion that this class of spaces has also great practical importance and a logical connection with the spaces with unconditional basis through the construction of the quotient space. The answer to the general question of whether every separable Banach space is a quotient of a hereditarily indecomposible Banach space is negative, due to the lifting property of \(\ell^1(\mathbb{N})\). However, the following theorem is proved in the paper: Every Banach space \(X\) either contains a subspace isomorphic to \(\ell^1(\mathbb{N})\) or has a subspace which is isomorphic to a hereditarily indecomposable Banach space. The main result concerning factorization of operators is the following: Let \(T:X\to Y\) be a bounded linear operator between Banach spaces such that \(T[B_X]\) is a \(a\)-thin set. Then there exists a hereditarily indecomposable space \(Z\) and bounded linear operators \(F_1: X\to Z\), \(F_2:Z\to Y\) such that \(T= F_2\cdot F_1\) (i.e. \(T\) is factorized through a hierarchically indecomposable space).



Cites work


Cited in
(33)








This page was built for publication: Interpolating hereditarily indecomposable Banach spaces

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4942939)