Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Let $k'$ be an field extension of an infinite field $k$. If $\{v_i\}$ is a set of vectors over $k$, and they satisfy $\sum_{i=1}^{n} c_iv_i =0$ for some nontrivial $\{c_i\} \subset k'$, then is the same set of vectors $\{v_i\}$ linearly dependent over $k$? In other words, can the $c_i$'s be chosen in $k$ only?

I see that if $k'$ is a finite extension of $k$, then I can quickly prove why this fact is true. But for infinite extensions I'm less sure. Maybe I'm missing something very obvious? Thanks in advance.

share|improve this question
add comment

1 Answer 1

up vote 5 down vote accepted

Any non trivial linear combination with coefficients from $k'$ will involve only finitely many elements of $k'$. Consider the subfield $k''$ of $k'$ generated by $k$ and by these coefficients. Then $k''$ is a finite extension of $k$, the $v_i$'s satisfy a linear relation with coefficients in $k''$, and you claim to be able to do what you want in that case.

N.B.: In any case, your question, as stated, does not make a lot of sense really: your $v_i$'s are presumably elements of a $k$-vector space, so what sense does it make to compute a linear combination with them using coefficients from $k'$? What I wrote above answers a version of what you meant:

suppose that $V$ is a $k$-vector space, that $(v_i)_{i\in I}$ is a family of elements of $V$, and that the family $(1\otimes v_i)_i$ of elements of the $k'$-vector space $k'\otimes_k V$ is linearly dependent over $k'$: then $(v_i)_{i\in I}$ is linearly dependent in $V$ over $k$.

There is no need, though, to proceed by reducing first to a finite subextension to do this. An alternative is the following: suppose that we have a non-trivial zero linear combination $\sum_i a_i\otimes v_i=0$ of the family $(1\otimes v_i)_{i\in I}$ with coefficients $(a_i)_{i\in I}$ in $k'$. Let $(z_j)_{j\in J}$ be a basis of $k'$ as a $k$-vector space, and suppose for all $i\in I$ we have $a_i=\sum_{j\in J}a_{i,j}z_j$, with all the $a_{i,j}\in k$. Then $$0 = \sum_{i\in I} a_i\otimes v_i = \sum_{j\in J}z_j\otimes\Bigl(\sum_{i\in I}a_{i,j}v_i\Bigr).$$ As $(z_j)_{j\in J}$ is linearly independent, this means that for all $j\in J$ we have $$\sum_{i\in I}a_{i,j}v_i=0,$$ and since we have assumed one of the $a_i$s is not zero, one of these linear combinations, which have coefficients in $k$, is itself non-trivial.

share|improve this answer
    
thanks for the answer! –  Heidi Mar 10 '11 at 6:31
add comment

Your Answer

 
discard

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.