2
$\begingroup$

Take a vector space built from an abelian group $(V,+)$ whose elements are the vectors, a field $K$ whose elements are the scalars, and there is an operation (multiplication by scalars) that distributes on the group operation.

What is the name of the structure obtained if $(V,+)$ instead of being an abelian group is an abelian monoid?

Thanks and forgive my ignorance :)

$\endgroup$
1
$\begingroup$

Adam showed quite nicely that this isn't interesting when $K$ is a field or, as we can see looking at his argument, any ring with an identity $1$ which has an additive inverse.

We can try to find a nontrivial example of your concept by talking about semirings and semimodules. As semiring is like a regular ring, except we only require the underlying group to be a monoid (no inverses necessarily.) Usually the axiom that $0*r=0$ is added back in (since it's no longer derivable from the existence of additive inverses.)

A semimodule is an additive abelian monoid with a semiring acting on it in the same way a ring acts on a regular module. Again, we may have to add back in that $0*m=0$ to keep things nice.

For a nice example of a semiring, you can just consider the semiring of natural numbers $\Bbb N=\{0,1,2,\ldots\}$. It is a semimodule over itself.

In the same line of thought, you can consider the set of nonnegative rational numbers as a semifield, where there aren't any additive inverses (except for $0$) and yet everything has a multiplicative inverse, and it is closed under addition and multiplication, etc. They probably call semimodules over a semifield "semivector spaces".

$\endgroup$
3
$\begingroup$

There cannot be such a structure without the group information. If you use the same axioms replacing the group info with monoid info, you end up being able to show it's a vector space.

Proof: Let $v\in V$. Then as $1_F\cdot v=v$ we have that $v+(-1_F)v=(1_F-1_F)\cdot v=0\cdot v=0$. Because $V$ is closed under the field operation that $(-1_F)\cdot v$ is an additive inverse for $v$, hence $V$ is inherently an abelian group.

$\endgroup$
  • $\begingroup$ Thanks a lot, I did not stop to think in this top-to-bottom perspective. Your observation is highly appreciated. I was in the search of name, a keyword, to open the doors for more reading, so I will mark rschwieb answer as "the answer" but yours is also extremely educative! $\endgroup$ – JuanPi Dec 5 '14 at 21:08

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.