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 category with just one object is a monoid. A category with two objects (which are distinguished) can be described by the following data (imagine the picture $\stackrel{M}{\curvearrowright} \bullet {\stackrel{X}{\longrightarrow} \atop \stackrel{Y}{\longleftarrow}} \bullet \stackrel{N}{\curvearrowleft}$):

  • a monoid $M$ and a monoid $N$
  • a $(N,M)$-set $X$ and a $(M,N)$-set $Y$
  • a homomorphism $\alpha : X \otimes_M Y \to N$ of $(N,N)$-sets and a homomorphism $\beta : Y \otimes_N X \to M$ of $(M,M)$-sets
  • such that under the obvious identifications $\alpha \otimes X = X \otimes \beta$ and $\beta \otimes Y = Y \otimes \alpha$

The same description works for enriched categories, where you would call $X$ and $Y$ bimodules. And by some coincidence, the above description reminds heavily of the definition of dualizable bimodules; but there $\beta$ (or $\alpha$) goes in the other direction. One can easily write down the $1$-morphisms as well as the $2$-morphisms in the category above.

Question. Does this category of "mixed bimodules" have been studied anywhere?

More specifically: How can we use this algebraic category to understand the category of all categories with two objects? What are some specific examples defined by generators and relations? For example, one could write down the free category on two objects $x,y$ such that $x$ is a retract of $y$. Background is Tilman's $\mathbb{Z}[C]$-conjecture, where $C$ may be assumed to be a category with just two objects. (EDIT: Oh, meanwhile Winfried Dreckmann has solved it! So let's call it the Tilman-Dreckmann-Theorem :-))

share|cite|improve this question
Winfried's answer there has indeed gone largely unnoticed :) – Gjergji Zaimi May 15 '12 at 10:29
Isn't this just a monoid with an idempotent? – Sasha May 15 '12 at 13:50
@Sasha: Hm, can you explain? – Martin Brandenburg May 15 '12 at 14:54
See the last paper of Berger and Moerdijk. They even work in an enriched context. – Fernando Muro May 15 '12 at 15:39
@Fernando: Thanks! So you mean Although they don't study those specific questions I've mentioned, they also consider categories with two objects as these "mixed bimodules", so it answers my question and you may add this as an answer. – Martin Brandenburg May 15 '12 at 18:34

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.