Schur's property

When we are working in a normed space X and we have a sequence ${\displaystyle (x_{n})}$ that converges weakly to ${\displaystyle x}$, then a natural question arises. Does the sequence converge in perhaps a more desirable manner? That is, does the sequence converge to ${\displaystyle x}$ in norm? A canonical example of this property, and commonly used to illustrate the Schur property, is the ${\displaystyle \ell _{1}}$ sequence space.

Suppose that we have a normed space (X, ||·||), ${\displaystyle x}$ an arbitrary member of X, and ${\displaystyle (x_{n})}$ an arbitrary sequence in the space. We say that X has Schur's property if ${\displaystyle (x_{n})}$ converging weakly to ${\displaystyle x}$ implies that ${\displaystyle \lim _{n\to \infty }\Vert x_{n}-x\Vert =0}$. In other words, the weak and strong topologies share the same convergent sequences. Note however that weak and strong topologies are always distinct in infinite-dimensional space.

The space ℓ¹ of sequences whose series is absolutely convergent has the Schur property.

This property was named after the early 20th century mathematician Issai Schur who showed that ℓ¹ had the above property in his 1921 paper.[1]

• Radon-Riesz property for a similar property of normed spaces
• Schur's theorem

• Megginson, Robert E. (1998), An Introduction to Banach Space Theory, New York Berlin Heidelberg: Springer-Verlag, ISBN 0-387-98431-3