Minimal lattice-subspaces (Q4257579)

This paper is a very interesting one. It deals with the lattice-subspaces of a Banach lattice \(E\). A Banach subspace \(F\) with a closed cone \(F_{+}\) of \(E\) is a lattice-subspace if it is a vector-lattice in the induced order. Theorem 2.5 asserts that such lattice-subspaces, which are moreover minimal for the inclusion, exist. A more precise question studied in this paper is the following: Let \(C(\Omega)\) be the Banach lattice of continuous real functions defined on \(\Omega\), a compact Hausdorff topological space. Let \(x_{1},\dots,x_{n}\) be \(n\) fixed linearly independent positive elements of \(C(\Omega)\). Does a finite-dimensional lattice-subspace (sublattice) \(X\) of \(C(\Omega)\) containing \(x_{1},\dots,x_{n}\) exist? The answer to this question in the case \(X\) is a sublattice is given in Theorems 3.5-3.7. Finally an answer in the case \(X\) a minimal lattice-subspace containing \(x_{1},\dots,x_{n}\) is given in Theorem 3.20. Theorem 3.20 gives a necessary and sufficient geometric condition for the existence of a finite dimensional minimal lattice-subspace.