MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

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

Let $\Delta$ be the standard cosimplicial space sending $[n]\mapsto \Delta^n\subset \mathbb R^{n+1}$.

Then the correspondence $\delta\colon [n]\mapsto {\rm Sh}(\Delta^n)$ defines a cosimplicial object in the category of toposes.

It follows from general nonsense that the pair of functors $${\rm Lan}_y\delta : \widehat{\Delta}\to {\bf Topoi} : X_\ast\mapsto \int^{n\in\Delta}X_n\cdot {\rm Sh}(\Delta^n)$$ where $y\colon\Delta\to\widehat{\Delta}$ is the Yoneda embedding, and $$N_\delta : {\bf Topoi} \to \widehat{\Delta} : {\cal E}\mapsto\Big([n]\mapsto {\bf Topoi}({\rm Sh}(\Delta^n), \cal E)\Big)$$ are mutually adjoint.

Has this construction a particular name? Has it a particular interest? Can it be characterized in a more hands-on way? What if I replace the easy cosimplicial topos $[n]\mapsto {\rm Sh}(\Delta^n)$ with a generic $[n]\mapsto {\rm Sh}(Y^n)$, for a generic cosimplicial (topological) space $Y^\ast$?

share|cite|improve this question
Minor comment: I'd say that $\delta$ is a cosimplicial topos, because it is a composite of functors $\mathbf{\Delta} \to \textbf{Top} \to \textbf{Topoi}$, both of which are covariant. – Todd Trimble Sep 14 '13 at 10:57
You should have a look at Moerdijk's Classifying spaces and classifying topoi. – Zhen Lin Sep 14 '13 at 11:09
Does Moerdijk's book address my specific question? I saw that he defines precisely the same toposic-realization I had in mind, but I would like to have a couple of simple examples (the standard cosimplicial space seems perfect) – tetrapharmakon Sep 14 '13 at 11:17
Well, your question contains some incorrect assertions. Firstly, $N_\delta$ is generally not a small simplicial set, if by $\mathbf{Topoi}(-, -)$ you mean the collection of geometric morphisms (even if you quotient out by natural isomorphism) – for instance, consider the classifying topos of any algebraic theory. Secondly, the category of Grothendieck toposes and geometric morphisms is not cocomplete as a 1-category (again, even if you quotient out by natural isomorphism), so you can't just say that the left adjoint exists by abstract nonsense. – Zhen Lin Sep 14 '13 at 11:33
You could restrict your attention to locales instead of toposes. Then the usual abstract nonsense goes through, and you don't have to worry about the difference between 1-categories and 2-categories. (Incidentally, the embedding of locales into toposes does not preserve bicategorical colimits.) – Zhen Lin Sep 14 '13 at 13:22

Your Answer


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

Browse other questions tagged or ask your own question.