Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. 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

Suppose that V and W are vector spaces (say over the reals) and that T is a linear surjection from V onto W with kernel K. By picking a basis for W (using choice) and pulling it back through T to a linearly independent set in V (again using choice) it is not difficult to get a linear embedding i of W into V satisfying T(i(w)) = w. It is then easily checked that the map $(k,w) \mapsto k + i(w)$ gives an isomorphism from $K \oplus W$ onto V. This all seems a little hamfisted though for such an algebraic result. I thought I would check whether there is some way to get such an isomorphism which avoids choice. I would guess there is no chance since there this isomorphism is so devastatingly nonunique... In fact I'm not sure I even want to ask this question anymore now that I think about it...

share|cite|improve this question
The choice-free result is that $T$ induces an isomorphism from the quotient space $V/K$ to $W$. – Chris Eagle Mar 17 '11 at 22:49
up vote 4 down vote accepted

You do need choice in general:

without it, there may exist vector spaces none of whose nontrivial subspaces has a complement (Herrlich, Axiom of Choice, LNM 1876, Disaster 4.43).

-JS Milne, MathOverflow, Dec 1, 2009

share|cite|improve this answer
Saints preserve us! Thanks for the reply – Mike F Mar 17 '11 at 22:55

Your Answer


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

Not the answer you're looking for? Browse other questions tagged or ask your own question.