Reference for lifting an adjunction to a monoidal one
Hi all, I need a textbook or otherwise standard reference for the fact that if one has a pair of monoidal categories C, D, and an adjunction L -| R: UC <--> UD between their underlying categories, then if one of L or R lift to a (strong) monoidal functor, then the adjunction lifts to an adjunction in the 2-category of monoidal categories, strong monoidal functors and monoidal natural transformations. (Mac Lane of course only treats the case of strict monoidal functors, at least in my, older, edition of his book) Thanks, David [For admin and other information see: http://www.mta.ca/~cat-dist/ ]
{Note from moderator: Apologies to Richard and Steve whose posts were inadvertently placed in the wrong folder...} Dear David, I am sure you will get a few responses telling you that the result, as you state it, is not quite correct. What is correct is that, given an adjunction L -| R: UC <---> UD: a) endowments of L with oplax monoidal structure are in bijection, under the mates correspondence, with endowments of R with lax monoidal structure b) given endowments of L and R with lax monoidal structure, the unit and counit of the adjunction satisfy the conditions to be monoidal transformations if and only if the given lax constraint cells on L are inverse to the oplax constraint cells induced from R via a) whence: c) liftings of the adjunction L -| R to an adjunction in the 2-category of monoidal categories, lax monoidal functors and monoidal transformations are in bijective correspondence with endowments of L with strong monoidal structure There is a dual b') of b) giving the dual c') liftings of the adjunction L -| R to an adjunction in the 2-category of monoidal categories, oplax monoidal functors and monoidal transformations are in bijective correspondence with endowments of R with strong monoidal structure of c). All of this follows from the general considerations in Kelly "Doctrinal adjunction" SLNM 420, though it would be more perspicuous to prove it directly following Kelly's schema. Richard On Sat, Jan 28, 2017, at 12:08 PM, David Roberts wrote:
Hi all,
I need a textbook or otherwise standard reference for the fact that if one has a pair of monoidal categories C, D, and an adjunction L -| R: UC <--> UD between their underlying categories, then if one of L or R lift to a (strong) monoidal functor, then the adjunction lifts to an adjunction in the 2-category of monoidal categories, strong monoidal functors and monoidal natural transformations.
(Mac Lane of course only treats the case of strict monoidal functors, at least in my, older, edition of his book)
Thanks, David
[For admin and other information see: http://www.mta.ca/~cat-dist/ ]
Dear David Roberts, Sorry. I did not read the details of your statement. As Richard Garner observed, it is incorrect. Another way of stating the relevant result of Kelly is the following: "Consider the forgetful 2-functor U: Lax-Alg\to X. Let U(f) be a right adjoint 1-cell. f is left adjoint if and only if f is a pseudomorphism (and not just a lax morphism)." This means that there is a right adjoint g to f (if (U(f) is left adjoint and f is a pseudomorphism) in Lax-Alg. To get an adjunction in Ps-Alg, you should, now, ask whether this lifted g is also a pseudomorphism (which means to verify if the mate of the structure of f is an isomorphism). In other words, in the context of strong monoidal functors, considering the forgetful functor F: StrongMonoidal\to Cat, assume that f\dashv F(g) is an adjunction in Cat. g is right adjoint if and only if its mate is an isomorphism (that is to say, the induced oplax structure in f is a strong structure: Beck Chevalley Condition) Anyways, "Doctrinal Adjunction" (Kelly) is what you are looking for. You will probably find what you want about lifting of adjoints there. I would also recommend "Two-Dimensional Monadicity" of John Bourke (Advances in Mathematics) 2014. Best Regards 2017-01-30 2:37 GMT+00:00 Richard Garner <richard.garner@mq.edu.au>:
Dear David,
I am sure you will get a few responses telling you that the result, as you state it, is not quite correct. What is correct is that, given an adjunction L -| R: UC <---> UD:
a) endowments of L with oplax monoidal structure are in bijection, under the mates correspondence, with endowments of R with lax monoidal structure
b) given endowments of L and R with lax monoidal structure, the unit and counit of the adjunction satisfy the conditions to be monoidal transformations if and only if the given lax constraint cells on L are inverse to the oplax constraint cells induced from R via a)
whence:
c) liftings of the adjunction L -| R to an adjunction in the 2-category of monoidal categories, lax monoidal functors and monoidal transformations are in bijective correspondence with endowments of L with strong monoidal structure
There is a dual b') of b) giving the dual
c') liftings of the adjunction L -| R to an adjunction in the 2-category of monoidal categories, oplax monoidal functors and monoidal transformations are in bijective correspondence with endowments of R with strong monoidal structure
of c). All of this follows from the general considerations in Kelly "Doctrinal adjunction" SLNM 420, though it would be more perspicuous to prove it directly following Kelly's schema.
Richard
On Sat, Jan 28, 2017, at 12:08 PM, David Roberts wrote:
Hi all,
I need a textbook or otherwise standard reference for the fact that if one has a pair of monoidal categories C, D, and an adjunction L -| R: UC <--> UD between their underlying categories, then if one of L or R lift to a (strong) monoidal functor, then the adjunction lifts to an adjunction in the 2-category of monoidal categories, strong monoidal functors and monoidal natural transformations.
(Mac Lane of course only treats the case of strict monoidal functors, at least in my, older, edition of his book)
Thanks, David
[For admin and other information see: http://www.mta.ca/~cat-dist/ ]
Of course, there was a mistake in "Consider the forgetful 2-functor U: Lax-Alg\to X. Let U(f) be a *left adjoint* 1-cell. f is left adjoint if and only if f is a pseudomorphism (and not just a lax morphism)." Sorry. Best wishes 2017-02-04 0:36 GMT+00:00 F. Lucatelli Nunes <flucatellinunes@gmail.com>:
Dear David Roberts,
Sorry. I did not read the details of your statement. As Richard Garner observed, it is incorrect.
Another way of stating the relevant result of Kelly is the following:
"Consider the forgetful 2-functor U: Lax-Alg\to X. Let U(f) be a right adjoint 1-cell. f is left adjoint if and only if f is a pseudomorphism (and not just a lax morphism)." This means that there is a right adjoint g to f (if (U(f) is left adjoint and f is a pseudomorphism) in Lax-Alg.
To get an adjunction in Ps-Alg, you should, now, ask whether this lifted g is also a pseudomorphism (which means to verify if the mate of the structure of f is an isomorphism).
In other words, in the context of strong monoidal functors, considering the forgetful functor F: StrongMonoidal\to Cat, assume that f\dashv F(g) is an adjunction in Cat. g is right adjoint if and only if its mate is an isomorphism (that is to say, the induced oplax structure in f is a strong structure: Beck Chevalley Condition)
Anyways, "Doctrinal Adjunction" (Kelly) is what you are looking for. You will probably find what you want about lifting of adjoints there. I would also recommend "Two-Dimensional Monadicity" of John Bourke (Advances in Mathematics) 2014.
Best Regards
2017-01-30 2:37 GMT+00:00 Richard Garner <richard.garner@mq.edu.au>:
Dear David,
I am sure you will get a few responses telling you that the result, as you state it, is not quite correct. What is correct is that, given an adjunction L -| R: UC <---> UD:
a) endowments of L with oplax monoidal structure are in bijection, under the mates correspondence, with endowments of R with lax monoidal structure
b) given endowments of L and R with lax monoidal structure, the unit and counit of the adjunction satisfy the conditions to be monoidal transformations if and only if the given lax constraint cells on L are inverse to the oplax constraint cells induced from R via a)
whence:
c) liftings of the adjunction L -| R to an adjunction in the 2-category of monoidal categories, lax monoidal functors and monoidal transformations are in bijective correspondence with endowments of L with strong monoidal structure
There is a dual b') of b) giving the dual
c') liftings of the adjunction L -| R to an adjunction in the 2-category of monoidal categories, oplax monoidal functors and monoidal transformations are in bijective correspondence with endowments of R with strong monoidal structure
of c). All of this follows from the general considerations in Kelly "Doctrinal adjunction" SLNM 420, though it would be more perspicuous to prove it directly following Kelly's schema.
Richard
[For admin and other information see: http://www.mta.ca/~cat-dist/ ]
participants (3)
-
David Roberts -
F. Lucatelli Nunes -
Richard Garner