Natural way of looking at projective transformations.

82 Views Asked by At

Let $k$ be a field and let $V$ and $W$ be finite-dimensional $k$-vector spaces, where $\dim(V)\ge1$ and $\dim(W)\ge1$. Let $q:V\to\mathbb{P}(V)$, $u\to[u]$ be the quotient map. By my teacher, a map $f:\mathbb{P}(V)\to\mathbb{P}(W)$ is defined to be a projective map if there exists an injective linear map $L:V\to W$ such that for all $[p]\in\mathbb{P}(V)$ we have that $f([p])=A(P)$, where $P$ is the subspace spanned by any $p\in V$ for which $q(p)=[p]$ holds. Consequently, he defines a projective transformation to be a projective map $f:\mathbb{P}(V)\to\mathbb{P}(V)$. These projective transformations, as I understand it, form Aut$(\mathbb{P}(V))$.

That's all nice and well, but I was wondering if this really is the most "elementary" way of looking at Aut$(\mathbb{P}(V))$, and if it really is necessary to define what an element of Aut$(\mathbb{P}(V))$ looks like. It now seems so "derived" from Aut$(V)$. I hope that my point is coming across.

For example, when we define a group, ring, $R$-module, topological space homomorphism it is just a map (of sets) which has the structure-preserving properties you want, but these are not "derived" from maps on some other category, such as it seems to be here, as this definition implies the surjection Aut$(V)\to$Aut$(\mathbb{P}(V))$. We don't say (or at least I have not seen it) that any group homomorphism for example has to come from I don't know, a map of sets.

If someone can clears this up even a little bit that would be great, and I apologise for any basic errors which someone like me who hasn't formally and rigorously done category theory (but is fascinated by the concepts introduced by it) yet.

1

There are 1 best solutions below

0
On BEST ANSWER

You're right to worry about this; in more general settings ($k$ not a field but a more general commutative ring) the natural map $\text{Aut}(V) \to \text{Aut}(\mathbb{P}(V))$ might fail to be surjective, so it won't be possible to use this definition. Unfortunately, the correct definition at this level of generality is relatively sophisticated (even correctly defining $\mathbb{P}(V)$ itself is relatively sophisticated); your professor's definition is certainly the most elementary in that it's the one that requires developing the least additional material to use.