I'm studying modules by reading Dummit and Foote, and I'm having a problem understanding the definition of a free module. I read this stackexchange question, but I couldn't figure it out.

The textbook defines a free module as following:

definition of a free module

The following is the example that I'm confused about.

Let $F = \{ 0 \}$. Then $F$ is a $\mathbb{Z}$-module since $F$ is an abelian group under addition.

In the stackexchange question, the empty set is given as the basis for $F$. That makes sense because $F$ has no nonzero elements, so it satisfies the definition vacuously.

But, using the same logic, wouldn't $F$ be a free $\mathbb{Z}$-module on the set $\{ 0 \}$?

I think that would be a problem because if that was the case, the rank of $\{ 0 \}$ would be both 0 and 1.

  • $\begingroup$ No, $\{0\}$ is not a free module on one generator. $\endgroup$ – Lord Shark the Unknown May 23 '19 at 17:13
  • $\begingroup$ Can you explain how I can get to that conclusion using the definition I posted? $\endgroup$ – hidenoris May 29 '19 at 15:34

The set $\{0\}$ is not linearly independent, because $r0=0$ for any nonzero $r\neq 0$.

The empty set does span the zero module, because an empty sum is zero.

  • $\begingroup$ It looks like $\{0\}$ is a basis over the null ring, though. $\endgroup$ – hmakholm left over Monica May 23 '19 at 17:29
  • $\begingroup$ I don't think the definition I posted (the one from Dummit and Foote) mentions anything about linearly independent sets. $\endgroup$ – hidenoris May 23 '19 at 18:16
  • $\begingroup$ That definition is equivalent to: (1) $A$ spans $F$, and (2) $A$ is linearly independent. Think about it: clearly $1\cdot 0 = 0 = 0\cdot 0$, so $0$ does not have a unique representation as a combination of elements of $\{0\}$. $\endgroup$ – Ehsaan May 23 '19 at 18:54
  • $\begingroup$ I'm not sure if that answers my question. The definition I posted states "nonzero element x of F, there exist..." $\endgroup$ – hidenoris May 23 '19 at 20:43
  • $\begingroup$ Okay, it's just a weird definition. I think most algebraists would agree that a basis is a linearly independent spanning set. A free module with a finite basis should be isomorphic to $R^n$, but this fails if you're using D&F's definition. Rank should be the cardinality of a basis, and should uniquely determine a free module (over a commutative ring). $\endgroup$ – Ehsaan May 24 '19 at 1:56

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.