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

I've read in a few different places that the standard fact $\text{Nat}\,(F,G) \cong \int_x \text{Hom}\,(Fx,Gx)$ can be upgraded to $\textbf{LaxNat}\,(F,G) \cong \oint_x\textbf{Hom}\,(Fx,Gx)$. where the left hand side is the category of lax natural transformations and modifications, and the right hand side is a lax end.

I am looking for a reference that gives the definition of lax end and proves this equivalence. I do know of the reference, S. Bozapalides, Théorie formelle des bicatégories, but I can't read French and I also can't find a copy. If someone can link me to the Bozapalides reference would be great. Or even better would be if there is a reference in English. Thanks!

share|cite|improve this question
I've added the ct.category-theory tag. – Finn Lawler Jun 7 '11 at 20:39
I guess there's an obvious candidate for what a "lax wedge" $c\Rightarrow F$ should be. Maybe the rest is straightforward too... – Alan Wilder Jun 8 '11 at 0:00
Yes as long as one sticks to strict 2-categories and strict functors, the details in proving the lax transformation identity are not too terrible using the obvious definition of lax wedge/lax end. I'll take that as evidence that obvious is right in this case. Still, a reference would be nice. – Alan Wilder Jun 8 '11 at 1:49
BOZAPALIDES, S., Les fins cartésiennes, C. R. A. S. Paris 281 ( 1975) and… – Buschi Sergio May 16 '12 at 18:57
Bozapalides 'Some remarks on lax-presheafs' – Ma Ming Apr 4 '14 at 23:06

This page says that you may be able to get a copy by emailing Andrée Ehresmann.

I don't know the exact answer to your question, but if you can't find a reference then it may be worth recalling that:

  • For Cat-valued F and G, $\mathrm{Nat}(F,G) \simeq \{F,G\}$, the limit of G weighted by F,
  • ends are $\hom$-weighted limits, and
  • there are lax morphism classifiers for 2-functors, meaning that $\mathrm{Lax}(F,G) \simeq \mathrm{Nat}(QF,G)$ for another 2-functor $QF$.

So if you define the lax end $\oint_x T(x,x)$ to be the representative of $\mathrm{Lax}(\hom_K, L(1,T))$, then you get $\oint_x [F x, G x] \simeq \mathrm{Lax}(\hom_K, [F-, G-])$, which is not quite what you want, but it's close.

Hope that helps.

share|cite|improve this answer
Thanks, Finn. I'm still holding out hope for a reference in English, but all of the above is helpful. – Alan Wilder Jun 6 '11 at 22:55

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.