Skip to main content
replaced http://mathoverflow.net/ with https://mathoverflow.net/
Source Link

(I completely revamped my answer, the previous version (linklink) had a consistency result with a particular example, this feels much better as it establishes a full equivalence result instead.)

The answer is no. For projectivity and injectivity. The reason is that the axiom of choice is equivalent to the following statement:

If $V$ is a vector space, and $K$ is a subspace of $V$, then there exists a a direct complement for $K$, that is a subspace $S$ such that $S\cap K=\{0\}$ and $S\cup K$ is a generating set for $V$.

If every vector space is projective, consider the case that $V$ is a vector space, $K$ is a subspace and $T\colon V\to V/K$ is the quotient map. Then there is some $h\colon V/K\to V$ such that $h\circ T$ is the identity function. In particular, this means that $K+\operatorname{im}(h)=V$, but $h$ cannot map any nonzero vector into $K$ so this is a direct sum indeed.

Therefore the statement "Every vector space is projective" implies the axiom of choice.

Similarly for injectivity, take the identity function $K\to V$, then there is a projection map whose kernel is a direct complement for $K$.


I think that a relevant paper to mention here would be the following:

Andreas Blass, Injectivity, projectivity, and the axiom of choice, Trans. Amer. Math. Soc. 255 (1979), 31--59.

(I completely revamped my answer, the previous version (link) had a consistency result with a particular example, this feels much better as it establishes a full equivalence result instead.)

The answer is no. For projectivity and injectivity. The reason is that the axiom of choice is equivalent to the following statement:

If $V$ is a vector space, and $K$ is a subspace of $V$, then there exists a a direct complement for $K$, that is a subspace $S$ such that $S\cap K=\{0\}$ and $S\cup K$ is a generating set for $V$.

If every vector space is projective, consider the case that $V$ is a vector space, $K$ is a subspace and $T\colon V\to V/K$ is the quotient map. Then there is some $h\colon V/K\to V$ such that $h\circ T$ is the identity function. In particular, this means that $K+\operatorname{im}(h)=V$, but $h$ cannot map any nonzero vector into $K$ so this is a direct sum indeed.

Therefore the statement "Every vector space is projective" implies the axiom of choice.

Similarly for injectivity, take the identity function $K\to V$, then there is a projection map whose kernel is a direct complement for $K$.


I think that a relevant paper to mention here would be the following:

Andreas Blass, Injectivity, projectivity, and the axiom of choice, Trans. Amer. Math. Soc. 255 (1979), 31--59.

(I completely revamped my answer, the previous version (link) had a consistency result with a particular example, this feels much better as it establishes a full equivalence result instead.)

The answer is no. For projectivity and injectivity. The reason is that the axiom of choice is equivalent to the following statement:

If $V$ is a vector space, and $K$ is a subspace of $V$, then there exists a a direct complement for $K$, that is a subspace $S$ such that $S\cap K=\{0\}$ and $S\cup K$ is a generating set for $V$.

If every vector space is projective, consider the case that $V$ is a vector space, $K$ is a subspace and $T\colon V\to V/K$ is the quotient map. Then there is some $h\colon V/K\to V$ such that $h\circ T$ is the identity function. In particular, this means that $K+\operatorname{im}(h)=V$, but $h$ cannot map any nonzero vector into $K$ so this is a direct sum indeed.

Therefore the statement "Every vector space is projective" implies the axiom of choice.

Similarly for injectivity, take the identity function $K\to V$, then there is a projection map whose kernel is a direct complement for $K$.


I think that a relevant paper to mention here would be the following:

Andreas Blass, Injectivity, projectivity, and the axiom of choice, Trans. Amer. Math. Soc. 255 (1979), 31--59.

deleted 383 characters in body
Source Link
Asaf Karagila
  • 39.8k
  • 8
  • 135
  • 283

(I completely revamped my answer, the previous version (link) had a consistency result with a particular example, this feels much better as it establishes a full equivalence result instead.)

The answer is no. Or at least not necessarily. At least forFor projectivity and injectivity.

It The reason is consistent that therethe axiom of choice is aequivalent to the following statement:

If $V$ is a vector space, and $K$ is a subspace of $V$, then there exists a a direct complement for $K$, that is a subspace $S$ such that $S\cap K=\{0\}$ and $S\cup K$ is a generating set for $V$.

If every vector space is projective, consider the case that $V$ over $\Bbb F_2$ (or really any finite field) whose underlying set is amorphous, namely it cannot be partitioned into two infinite subsets.

Thisa vector space is not finitely generated, but every proper subspace has$K$ is a finite dimension.

Pick any such subspace $V'$ and consider $W=V/V'$$T\colon V\to V/K$ is the quotient map. Then itthere is not very difficult to showsome $h\colon V/K\to V$ such that the following holds:

  1. $W$ is a vector space whose underlying set is amorphous, and every proper subspace is finitely generated. But $|W|\neq|V|$.
  2. There is no non-zero homomorphism from $W$ to $V$.
  3. The natural quotient map is surjective from $V$ to $W$.

So $W$$h\circ T$ is not projective. This also gives an example of an exact sequence which doesn't split. As Jeremy Rickard observes in the commentidentity function. In particular, this means that $K+\operatorname{im}(h)=V$, but $h$ cannot map any nonzero vector into $K$ so this is also a counterexample to injectivitydirect sum indeed.

(The above can be done easily over infinite fields too, but then $V$ Therefore the statement "Every vector space is not amorphous, just weird, and we might not have $|W|\neq|V|$ in generalprojective" implies the axiom of choice.

Another generalization is that we might allow subspaces to be not just finitely generatedSimilarly for injectivity, but generated by a well-ordered set up to some cardinal $\lambda$ (which would meantake the subspace is free, of course)identity function $K\to V$, but the space itselfthen there is not generated by any such seta projection map whose kernel is a direct complement for $K$.)

 

I think that a relevant paper to mention here would be the following:

Andreas Blass, Injectivity, projectivity, and the axiom of choice, Trans. Amer. Math. Soc. 255 (1979), 31--59.

The answer is no. Or at least not necessarily. At least for projectivity.

It is consistent that there is a vector space $V$ over $\Bbb F_2$ (or really any finite field) whose underlying set is amorphous, namely it cannot be partitioned into two infinite subsets.

This vector space is not finitely generated, but every proper subspace has a finite dimension.

Pick any such subspace $V'$ and consider $W=V/V'$. Then it is not very difficult to show that the following holds:

  1. $W$ is a vector space whose underlying set is amorphous, and every proper subspace is finitely generated. But $|W|\neq|V|$.
  2. There is no non-zero homomorphism from $W$ to $V$.
  3. The natural quotient map is surjective from $V$ to $W$.

So $W$ is not projective. This also gives an example of an exact sequence which doesn't split. As Jeremy Rickard observes in the comment, this is also a counterexample to injectivity.

(The above can be done easily over infinite fields too, but then $V$ is not amorphous, just weird, and we might not have $|W|\neq|V|$ in general.

Another generalization is that we might allow subspaces to be not just finitely generated, but generated by a well-ordered set up to some cardinal $\lambda$ (which would mean the subspace is free, of course), but the space itself is not generated by any such set.)

I think that a relevant paper to mention here would be the following:

Andreas Blass, Injectivity, projectivity, and the axiom of choice, Trans. Amer. Math. Soc. 255 (1979), 31--59.

(I completely revamped my answer, the previous version (link) had a consistency result with a particular example, this feels much better as it establishes a full equivalence result instead.)

The answer is no. For projectivity and injectivity. The reason is that the axiom of choice is equivalent to the following statement:

If $V$ is a vector space, and $K$ is a subspace of $V$, then there exists a a direct complement for $K$, that is a subspace $S$ such that $S\cap K=\{0\}$ and $S\cup K$ is a generating set for $V$.

If every vector space is projective, consider the case that $V$ is a vector space, $K$ is a subspace and $T\colon V\to V/K$ is the quotient map. Then there is some $h\colon V/K\to V$ such that $h\circ T$ is the identity function. In particular, this means that $K+\operatorname{im}(h)=V$, but $h$ cannot map any nonzero vector into $K$ so this is a direct sum indeed.

Therefore the statement "Every vector space is projective" implies the axiom of choice.

Similarly for injectivity, take the identity function $K\to V$, then there is a projection map whose kernel is a direct complement for $K$.

 

I think that a relevant paper to mention here would be the following:

Andreas Blass, Injectivity, projectivity, and the axiom of choice, Trans. Amer. Math. Soc. 255 (1979), 31--59.

added 762 characters in body
Source Link
Asaf Karagila
  • 39.8k
  • 8
  • 135
  • 283

The answer is no. Or at least not necessarily. At least for projectivity.

It is consistent that there is a vector space $V$ over $\Bbb F_2$ (or really any finite field) whose underlying set is amorphous, namely it cannot be partitioned into two infinite subsets.

This vector space is not finitely generated, but every proper subspace has a finite dimension.

Pick any such subspace $V'$ and consider $W=V/V'$. Then it is not very difficult to show that the following holds:

  1. $W$ is a vector space whose underlying set is amorphous, and every proper subspace is finitely generated. But $|W|\neq|V|$.
  2. There is no non-zero homomorphism from $W$ to $V$.
  3. The natural quotient map is surjective from $V$ to $W$.

So $W$ is not projective. This also gives an example of an exact sequence which doesn't split. As Jeremy Rickard observes in the comment, this is also a counterexample to injectivity.

(The above can be done easily over infinite fields too, but then $V$ is not amorphous, just weird, and we might not have $|W|\neq|V|$ in general.

Another generalization is that we might allow subspaces to be not just finitely generated, but generated by a well-ordered set up to some cardinal $\lambda$ (which would mean the subspace is free, of course), but the space itself is not generated by any such set.)

I think that a relevant paper to mention here would be the following:

Andreas Blass, Injectivity, projectivity, and the axiom of choice, Trans. Amer. Math. Soc. 255 (1979), 31--59.

The answer is no. Or at least not necessarily. At least for projectivity.

It is consistent that there is a vector space $V$ over $\Bbb F_2$ (or really any finite field) whose underlying set is amorphous, namely it cannot be partitioned into two infinite subsets.

This vector space is not finitely generated, but every proper subspace has a finite dimension.

Pick any such subspace $V'$ and consider $W=V/V'$. Then it is not very difficult to show that the following holds:

  1. $W$ is a vector space whose underlying set is amorphous, and every proper subspace is finitely generated. But $|W|\neq|V|$.
  2. There is no non-zero homomorphism from $W$ to $V$.
  3. The natural quotient map is surjective from $V$ to $W$.

So $W$ is not projective. This also gives an example of an exact sequence which doesn't split.

The answer is no. Or at least not necessarily. At least for projectivity.

It is consistent that there is a vector space $V$ over $\Bbb F_2$ (or really any finite field) whose underlying set is amorphous, namely it cannot be partitioned into two infinite subsets.

This vector space is not finitely generated, but every proper subspace has a finite dimension.

Pick any such subspace $V'$ and consider $W=V/V'$. Then it is not very difficult to show that the following holds:

  1. $W$ is a vector space whose underlying set is amorphous, and every proper subspace is finitely generated. But $|W|\neq|V|$.
  2. There is no non-zero homomorphism from $W$ to $V$.
  3. The natural quotient map is surjective from $V$ to $W$.

So $W$ is not projective. This also gives an example of an exact sequence which doesn't split. As Jeremy Rickard observes in the comment, this is also a counterexample to injectivity.

(The above can be done easily over infinite fields too, but then $V$ is not amorphous, just weird, and we might not have $|W|\neq|V|$ in general.

Another generalization is that we might allow subspaces to be not just finitely generated, but generated by a well-ordered set up to some cardinal $\lambda$ (which would mean the subspace is free, of course), but the space itself is not generated by any such set.)

I think that a relevant paper to mention here would be the following:

Andreas Blass, Injectivity, projectivity, and the axiom of choice, Trans. Amer. Math. Soc. 255 (1979), 31--59.

Source Link
Asaf Karagila
  • 39.8k
  • 8
  • 135
  • 283
Loading