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

Consider a category $C$ enriched in categories. Here we have a natural class $W$ of equivalences, namely those morphisms having inverses up to 2-isomorphisms. One can take the hammock localization $L^H(C,W)$ and get a simplicial category.

On the other hand one can take the maximal subgroupoid of each $hom$-category (or simply suppose that the hom-categories were groupoids from the start, this wouldn't change the above notion of equivalence), apply the nerve functor $hom$-wise and get another simplicial category.

Are these two simplicial categories Dwyer-Kan equivalent?

[I am also interested in a particular case of this question, which would take too long to expose here, which has the feature that the $hom$-categories are preorders, and hence the maximal subgroupoids are simply equivalence relations on the sets of morphisms (i.e. the sets of objects of the hom-categories). Yet I couldn't see by hand that the hom-simplicial sets of the hammock localization are homotopy discrete...]

share|cite|improve this question
If $C$ has one object and one 1-morphism, the first construction forgets about all the 2-morphisms but the second does not. – Eric Wofsey Mar 30 '14 at 23:06
More precisely, if we have an abelian group $A$, then we can form a 2-category $\mathbb{B}^2 A$ with a unique object $*$, unique 1-cell $\mathrm{id}_*$, and elements of $A$ as 2-cells. The hammock localisation cannot see the automorphisms of the 1-cell in this case, so the two simplicial categories are Dwyer–Kan equivalent if and only if $A$ is trivial. – Zhen Lin Mar 30 '14 at 23:50
Thank you for this example! You are right that the hammock localization cannot distinguish different 2-isomorphisms. Now if the hom-categories are preorders, the answer could still be affirmative. Any thoughts on this? – Peter Arndt Mar 31 '14 at 0:33
@PeterArndt There are some further necessary conditions. For instance, we need $\mathcal{C} [\mathcal{W}^{-1}]$ to be isomorphic to the category obtained by identifying isomorphic 1-cells. This fails for e.g. the 2-category with a unique object, a non-trivial involution, and 2-cells making the two 1-cells isomorphic. You could probably get around this by introducing some kind of path or cylinder object. – Zhen Lin Mar 31 '14 at 9:22
@Zhen Lin : This helped me. Thanks again! – Peter Arndt Mar 31 '14 at 9:46

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.