Let $V$ be a vector space over the field $F$. Suppose there are a finite number of vectors $\alpha_1, . . . , \alpha_r$ in $V$ which span $V$. Prove that $V$ is finite-dimensional.

Does it suffice to just say that no linearly independent set of $V$ can have more than $r$ elements. So a basis for $V$ must have $r$ or less elements in it. And hence $V$ has a finite basis?

Or am I missing/ignoring something??

  • $\begingroup$ If you have a proof that no linearly independent set in $V$ can contain more than $r$ elements, that's enough. Otherwise, you could argue with the fact that every spanning set contains a basis (might also require a proof, but you can restrict yourself to proving it for finite spanning sets here). $\endgroup$ – Daniel Fischer May 30 '14 at 16:19
  • $\begingroup$ @Daniel Fischer: Yes. I have a proof for the former. Thanks. Problem seemed too easy that's all. $\endgroup$ – Ishfaaq May 30 '14 at 16:21
  • $\begingroup$ @Daniel Fischer: One more thing if it's alright. To prove your second suggestion - Consider a finite spanning set $A$ for a vector space. Consider its largest linearly independent subset, $B = \{b_1, b_2, ..., b_n\}$. Suppose there is $a \in A \setminus B$ which is not a linear combination of vectors in $B$. Then suppose $k_1b_1 + k_2b_2 + ...+ k_nb_n + ka = 0$. It can be argued that $k$ must be $0$ and hence all the scalars $k_i = 0$ contradicting the maximality of $B$. Hence $B$ can be proved to be a basis. Sound good enough?? $\endgroup$ – Ishfaaq May 30 '14 at 16:35
  • $\begingroup$ "It can be argued" is not a phrase that fits particularly well in a proof. For a proof, you need to spell that argument out (it's after all the crucial point). Then it works. But it's IMO easier/nicer to use a minimal spanning set contained in $A$, and show that that is linearly independent. $\endgroup$ – Daniel Fischer May 30 '14 at 16:43
  • $\begingroup$ @Daniel Fischer: Too long for a comment though. And you're right. That is more concise. Thanks loads for the input. Much appreciated. $\endgroup$ – Ishfaaq May 30 '14 at 16:46

Suppose the vectors $ \alpha_1, ... , \alpha_r $ form a basis for $V$. ie. They are linearly independent and span $V$. Then by definition $V$ is finite dimensional. So instead assume that for some $\alpha_i$ there exists scalars $c_1, ... , c_{r-1}$ in $F$ such that (without loss of generality let $\alpha_i $ be moved to become the rth element)

$c_1\alpha_1 + ... + c_{r-1}\alpha_{r-1} = \alpha_i $ Then we can remove $\alpha_i$ from $V$ and form a new subset of vectors $ \alpha_1, ... , \alpha_{r-1} $ that clearly still span $V$.

If the vectors $ \alpha_1, ... , \alpha_{r-1} $ are linearly independent then they form a basis for $V$ and again by definition $V$ is finite dimensional. If the vectors $ \alpha_1, ... , \alpha_{r-1} $ are linearly dependent then we repeat this process until we form a basis for $V$ and hence prove our theory that $V$ is finitely dimensional.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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