MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

A morphism of set-valued functors $\eta: F \to G$ on $\mathcal{C}$ is called smooth if for all epimorphisms $B \to A$, the natural morphism $F(B) \to F(A) \times_{G(A)} G(B)$ is surjective.

Obviusly "smooth => formally" smooth for $\mathcal{C} = \mathrm{Sch}$.

Now my question: Does the converse hold?

My thoughts: 1. Assume the morphism is of finite presentation.

  1. Assume it is in the local standard form: an étale morphism followed by an affine projection

  2. It is clear that an affine projection is smooth in the above sense, so we have reduced the problem to étale morphisms.

share|cite|improve this question
Possible duplicate:… – Francesco Polizzi Sep 5 '11 at 10:42
I think my definition of "smooth" is a priori different from this. – user12832 Sep 5 '11 at 11:08
I don't have the book in front of me, but there is an appendix to Loday's "Cyclic Homology" where various definitions of 'smooth' are compared quite carefully. – Peter Samuelson Sep 5 '11 at 13:08
Your definition of "smooth" is for me the definition of "formally smooth", and "smooth" would be "formally smooth"+"locally of finite presentation". What in fact is your definition of formally smooth? – Keerthi Madapusi Pera Sep 5 '11 at 17:40
up vote 4 down vote accepted

In the case where $F\to G$ is representable by an epimorphism $C\to D$ of rings, then your smoothness condition implies that $C\to D$ has a section. (Take $A\to B$ to be the given map $C\to D$.) But it is not hard to find a formally smooth epimorphism that does not admit a section. Any localization will do. For instance, you can take $D$ to be the zero ring and $C$ any ring but the zero ring.

If you visualize the geometry here, it's pretty clear that your smoothness condition is much stronger than anything normally called smoothness.

share|cite|improve this answer
Ah, I missed the part where the OP allowed any epimorphism. – Keerthi Madapusi Pera Sep 5 '11 at 23:03

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.