2
$\begingroup$

In Gowers' explanation of tensor products, there is a passage where he writes, with vector spaces $V$ and $W$:

For every pair $(v,w)$ in $V\times W$, regard $[[v,w]]$ as a meaningless symbol. We can define a rather large vector space $Z$ by taking formal linear combinations of these symbols. By that I mean that $Z$ consists of all expressions of the form $$a_1[[v_1,w_1]]+ a_2[[v_2,w_2]]+...+ a_n[[v_n,w_n]]$$ with obvious definitions for addition and scalar multiplication.

I am a little confused by what he means at the end. When he talks about addition and scalar multiplication, I am wondering if he means the following: $$a_1[[v_1,w_1]]+a_2[[v_2,w_2]]=[[a_1v_1+a_2v_2,a_1w_1+a_2w_2]]$$ But this also seems incorrect because then there would seemingly be no difference between $Z$ and $V\times W$ other than the fact that instead of $(\cdot,\cdot)$ we write $[[\cdot,\cdot]]$.

$\endgroup$
1
$\begingroup$

No, $Z$ really is just the vector space whose basis is the set of all symbols $[[v,w]]$. $a[[v,w]]$ can be simplified no further, and there's no combination of elements when you take sums, in particular $[[v+v',w]]$ doesn't split into a sum of two elements. That comes later when we take the quotient to create the tensor product. Even $[[0,0]]\neq 0$, $0$ is the element $0[[v,w]]$ for any $v,w$, so for example $[[0,0]]$ and $2[[0,0]]$ are different elements.

$\endgroup$
2
  • $\begingroup$ Just to make sure I'm understanding you correctly, in this case $[[v+v',w]]$ and $[[v,w]]+[[v',w]]$ are different elements, and they are only made to identify with each other when the quotient is taken, which explicitly sets all symbols of the form $[[v+v',w]]-[[v,w]]-[[v',w]]$ to zero. $\endgroup$
    – Vasting
    Mar 27 '20 at 19:33
  • $\begingroup$ @Vasting That is correct. $\endgroup$ Mar 27 '20 at 19:41

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.