MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Hi everyone,

Ive been looking at the following problem, but its not entirely in my area and some potential solutions seem to rely on algebraic geometry. Maybe thats just a complicated way to solve the problem.

Im wondering if anyone can give me a good reference or answer to this question

For any set of $n×n$ matrices $A_1,A_2,...,A_k$, such that

  1. they share only a SINGLE eigenvector in common
  2. the joint commutant of $A_1,A_2,...,A_k$ and $A^*_1,A^*_2,...,A^*_k$ is trivial,

how can I guarantee there exists only this single one dimensional invariant subspace (corresponding to the span of that eigenvector)?

Googling around, it seems some progress has been made on saying whether invariant subspaces of dimension greater than 2 exist, but they rely on the fact that the matrices $A_1,A_2,...,A_k$ are chosen to have pairwise distinct eigenvalues.

Thanks for the help!

edit: changed some wording so I dont mislead anyone.

share|cite|improve this question
What is the "joint commutant"? – Anton Jun 19 '13 at 11:14
Matrices with pairwise distinct eigenvalues are generic. – Peter Michor Jun 19 '13 at 12:43
Oh, I mean that the only matrices which commutes with every matrix $A_1,...,A_k$ and their Hermitian conjugates $A^*_1,...,A^*_k$ are those which are scalar multiples of the identity. Sometimes this is denoted $\{A_1,...,A_k,A^*_1,...,A^*_k\}'=\mbox{span}(\mathbb{I})$. – jeremy Jun 19 '13 at 12:45
Hm okay yea, I mean generic in the sense that I put no restrictions on the conditions of the eigenvalues. i.e. $A_1,..A_k\in\mathcal{M}^n(\mathbb{C})$ and can have degenerate eigenvalues. – jeremy Jun 19 '13 at 12:49

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.