Take the 2-minute tour ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

I am going to make this question deliberately vague.

A category is concrete if its objects can be realized as sets with extra structure (in particular, it admits a faithful functor to Set).

The examples I know of non-concretizable categories (the homotopy category, Cat with naturally isomorphic functors identified) are obviously derived from forgetting structure of higher categories. Is there some sense in which, given any locally small category, the objects of that category can be realized themselves as (small) higher categories with extra structure, analogous to considering faithful functors to Set?

EDIT: I think my question is as Jeremy suggested below: Given a locally small category, is it always the homotopy category of a concrete $(\infty,1)$-category, for some reasonable definition of concrete?

share|improve this question
If your categories aren't required to be locally small, any category which is not locally small fails to be concretizable for reasons that seem to me to have nothing to do with higher categories. –  Qiaochu Yuan Oct 18 '12 at 22:24
Sorry I meant locally small. –  John Berman Oct 19 '12 at 1:25
You may be interested in the discussion here: golem.ph.utexas.edu/category/2011/02/concrete_categories.html Several definitions of concrete infinity categories are proposed. Perhaps you are asking which 1-categories occur as homotopy categories of concrete infinity categories? –  Jeremy Hahn Oct 19 '12 at 2:38
This seems very unlikely to me. It's kind of a coincidence that the standard example of non-concrete categories are homotopy categories--they just happen to be the most (only?) naturally occurring examples. Essentially, most categories you will ever encounter are either small or accessible; the only way "real life" large categories fail to be accessible is if they are secretly actually accessible higher categories. –  Eric Wofsey Oct 19 '12 at 16:38
Eric is there a very tight relation between accessibility and concreteness? I would be interested to know it. I think John's question is unsolvable without a better concept of concrete infinity category. The cafe discussion suggests it might be reasonable, for instance, to think of every infinity category as concrete. It seems that concreteness is one of the few classical notions that hasn't yet been suitably generalized to the infinity context. –  Jeremy Hahn Oct 20 '12 at 21:42
show 1 more comment

1 Answer

If you require you category to be actually small (and not just locally small) then the answer is yes.

Suppose $C$ is a small category. We can then define the category $C'$ as follows:

  • Objects($C'$) = $\{ C/t: t \in$ Object(C)$\}$
  • Morphisms($C'$) are functors between the corresponding categories

There is then a functor $F:C \rightarrow C'$ where

  • $F(t) = C/t$ for any object $t$ of $C$
  • For any map $\alpha:s \rightarrow t$ in $C$, $F(\alpha):C/s \rightarrow C/t$ is the functor where:

    • When $\beta:p \rightarrow s$ is an object of $C/s$ then $F(\alpha)(\beta) = \alpha \circ \beta:p \rightarrow t$ is the corresponding object of $C/t$.

    • Suppose $P:p\rightarrow s$ and $Q:q\rightarrow s$ are objects of $C/s$ and $g: P \rightarrow Q$ is a morphism in $C/s$ (i.e. a map $g:p \rightarrow q$ such that $q \circ g = p$).

      Then $F(\alpha)(\gamma) = g$ (as a map from $F(\alpha)(P) \rightarrow F(\alpha)(Q)$)

When dealing with locally small (but not necessarily small) categories however you have to be careful about set theoretic size issues.

share|improve this answer
I believe that your functor is faithful making any small category concrete. –  Spice the Bird Nov 2 '12 at 16:59
add comment

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.