nLab AT category

Redirected from "AT-categories".
Contents

Context

Category theory

Additive and abelian categories

Topos Theory

topos theory

Background

Toposes

Internal Logic

Topos morphisms

Extra stuff, structure, properties

Cohomology and homotopy

In higher category theory

Theorems

Contents

Idea

Freyd (1999) gave a sharp and clever description of the commonalities and differences between abelian categories and toposes (or, in the first place, between abelian categories and pretoposes), by introducing a finitely axiomatized theory of “AT categories”. Thus abelian categories and pretoposes are examples of AT categories; underscoring how much these structures have in common, Freyd’s theory hews so closely to both of these two basic examples that in fact every AT category splits cleanly as a product of a pretopos part and an abelian part.

Here, the properties held in common between abelian categories and pretoposes are all exactness conditions; in the AT set-up, the decisive difference between abelian categories and pretoposes is concentrated solely in the behavior of the initial object with respect to cartesian product (being in one case the cartesian monoidal unit, i.e. an element that is universally absorbed into any other, and in the other an absorbing element, i.e. an element that absorbs any other element into itself, aka a zero element or annihilating element).

There is no research we know of on AT categories. One might say that they were invented by Freyd really for the sole theoretical purpose of demonstrating the commonalities and differences between abelian categories and pretopos categories, not because AT categories are found in nature. Indeed, a “typical” AT category is something of a mythological creature with the head of an abelian lion and the body of a pretopos dragon, and serves no purpose as such, even though pure lions and pure dragons themselves serve a purpose. Some of the axioms, particular axiom “AE” below, seem expressly designed to fuse them together in ad hoc fashion.

Definition

Freyd introduces some baseline assumptions on the categories in question:

Each of these assumptions obviously holds in any abelian category and in any pretopos.

Remarks

If we introduce predicates to express these four baseline assumptions, such as a predicate formula P(p,q;f,g)P(p, q; f, g) whose meaning is “(p,q)(p, q) is a pushout pair for pairs of arrows (f,g)(f, g)”, then the remaining AT axioms given below all have a simple logical structure: they can be expressed as Horn clauses in the language generated by the baseline predicates. The first 8 axioms are in fact universal Horn clauses in these predicates.

This might look like a side comment for logicians, but it might also signal that Freyd intends to invoke, if pressed for details of proofs, one of a battery of representation theorems, such as the representation theorem for regular categories whose precise formulation is in terms of Horn theories – see Categories, Allegories for examples.

Furthermore, by choosing a specific initial object, a specific pullback for each pair of arrows with common target, etc., an AT category can be cast as a model for a certain essentially algebraic theory. In other words, such “algebraic AT categories” can be internalized in any finitely complete category.

The crucial difference between “abelianness” and “toposness” is concentrated in the following definition:

Definition

Let 00 denote the initial object, and for any object XX let

  • π 1:0×X0\pi_1 \,\colon\, 0 \times X \longrightarrow 0,

  • π 2:0×XX\pi_2 \,\colon\, 0 \times X \longrightarrow X

denote the two product projections.

We say XX is

  1. of type T if π 1\pi_1 is an isomorphism,

  2. of type A if π 2\pi_2 is an isomorphism.

With this:

  • A pretopos will turn out to be precisely an AT category in which every object is of type T.

  • An abelian category will turn out to be an AT category where every object is of type A.

Here then are the AT exactness axioms. Again, each of them is satisfied in every abelian category and in every pretopos, and according to Freyd’s thesis, any exactness condition satisfied in both classes of categories is a logical consequence of this set of axioms. Some of Freyd’s remarks in his original posting are included in parentheses.

Definition

A category meeting the baseline assumptions above is an AT category if the following 8 axioms plus axiom “AE” are satisfied.

  1. The category is an effective regular category.

    (“Yes this can be stated as universal Horn conditions on pullbacks and the special pushouts mentioned above.”)

  2. The arrow 010 \to 1 is monic. (“Note that it follows that all maps from 00 are monic.”)

  3. If i:ACi\colon A \to C is mono, then any pushout square

    A i C B j D\array{ A & \stackrel{i}{\to} & C \\ \downarrow & & \downarrow \\ B & \underset{j}{\to} & D }

    is also a pullback, and jj is mono.

  4. The functor 0×0 \times - preserves pushouts of kernel pairs, and pushouts of pairs of arrows one of which is monic.

  5. If f:B0×Cf\colon B \to 0 \times C is epi and

    A B f 0 0×C\array{ A & \to & B \\ \downarrow & & \downarrow f \\ 0 & \to & 0 \times C }

    is a pullback, then it’s also a pushout.

  6. In the full subcategory of type T objects (Def. ), pushouts of pairs of morphisms, one of which is monic, are universal (i.e., stable under pullback).

  7. Define a functor TT by the pushout diagram

    0×X π 1 0 π 2 X TX\array{ 0 \times X & \stackrel{\pi_1}{\to} & 0 \\ \pi_2 \downarrow & & \downarrow \\ X & \to & T X }

    Then TT preserves pullbacks.

  8. Given a morphism f:XYf\colon X \to Y, if TfT f and 0×f0 \times f are isomorphisms, then so is ff.

This is the basic list of exactness assumptions which permit a sharp comparison between pretoposes and abelian categories. From this list alone one can prove that every AT category embeds faithfully in a product of a pretopos and an abelian category. To show that every AT category is equivalent to such a product, Freyd appends the following axiom involving an existential Horn condition:

Definition

axiom (AE)

For every object XX, there is a map ξ:TXX\xi\colon T X \to X such that

0×Xπ 1XξTX0 \times X \stackrel{\pi_1}{\to} X \stackrel{\xi}{\leftarrow} T X

is a coproduct diagram, and the canonical map XTXX \to T X is a retraction of ξ\xi.

To bring actual toposes into the picture, Freyd adds some more axioms, but let’s first take stock and see what this gives us so far.

Basic consequences of the AT axioms

Proposition

For any XX the unique map 0X0 \to X is monic.

Proof

By axiom 2, the map 010 \to 1 is monic. The pullback-preserving functor ×X- \times X preserves monos, so

0×X!×1 X1×XX0 \times X \stackrel{! \times 1_X}{\to} 1 \times X \cong X

(which is just the second projection π 2:0×XX\pi_2: 0 \times X \to X) is monic. The unique map 00×X0 \to 0 \times X is monic since it has a retraction π 1:0×X0\pi_1: 0 \times X \to0. The result follows.

Proposition

Binary coproducts exist.

Proof

Because we can take the pushout of a pair of monos 0X0 \to X, 0Y0 \to Y.

Proposition

Coproducts are disjoint.

Proof

By axiom 3, the two coprojections i X:XX+Yi_X \colon X \to X + Y, i Y:YX+Yi_Y \colon Y \to X + Y are monic, and their pullback is initial.

Category of type A objects is abelian

In this section we show that the full subcategory of type A objects forms an abelian category.

Lemma

An object is of type A if and only if there exists a map to 00.

Proof

If XX is of type A, then we clearly have X0×Xπ 10X \cong 0 \times X \stackrel{\pi_1}{\to} 0. Conversely, suppose there exists p:X0p: X \to 0. Note that since 010 \to 1 is monic, there is at most one map Y0Y \to 0 for any object YY. It follows quickly that maps YXY \to X are in natural bijective correspondence with maps YX×0Y \to X \times 0 by composition with π 1:X×0X\pi_1: X \times 0 \to X, so that π 1\pi_1 is an isomorphism by the Yoneda lemma.

Proposition

The full subcategory of type A objects is coreflective.

Proof

By Lemma , the subcategory of type A objects in an AT category CC is equivalent to the subcategory C/0C/1CC/0 \hookrightarrow C/1 \simeq C, which is the category of coalgebras for the functor A(X)=0×XA(X) = 0 \times X. Thus the category of type A objects is comonadic.

Lemma

Objects of type A are closed under binary products, finite coproducts, subobjects, and quotient objects (= cokernels of kernel pairs).

Proof

This follows from the fact that the category of type A objects is the slice category C/0C/0.

Before we prove the next lemma, recall that in a category with zero objects, a kernel of an arrow f:ABf: A \to B is an equalizer of the pair f,0:ABf, 0: A \stackrel{\to}{\to} B. This is the same as a pullback

K A f 0 B\array{ K & \to & A \\ \downarrow & & \downarrow f \\ 0 & \to & B }

Cokernels are defined dually, and can be formulated dually as certain pushouts. Since pushouts along monos exist, we can take the cokernel of any mono, and in particular the cokernel of any kernel.

Lemma

In the category of type A objects, every mono is the kernel of its cokernel, and every epi is the cokernel of its kernel.

Proof

Of course, 00 is a zero object in the category of type A objects. If i:ABi\colon A \to B is a mono in the category of type A objects, then the pushout

A i B 0 coker(i)\array{ A & \stackrel{i}{\to} & B \\ \downarrow & & \downarrow \\ 0 & \to & coker(i) }

exists and is also a pullback, by axiom 3, and hence ii is the kernel of its cokernel. Now suppose f:ACf: A \to C is an epi in the category of type A objects. Since 0×CC0 \times C \cong C, we have an epi f:A0×Cf\colon A \to 0 \times C. Then, by axiom 5, the pullback

ker(f) A f 0 0×C\array{ ker(f) & \to & A \\ \downarrow & & \downarrow f \\ 0 & \to & 0 \times C }

is also a pushout, so ff is the cokernel of its kernel.

Theorem

The category of type A objects is an abelian category.

Proof

Any category with zero objects, binary products and coproducts, and in which every mono is the kernel of its cokernel and every epi is the cokernel of its kernel, is in fact an abelian category. See Freyd-Scedrov, Categories, Allegories, 1.598 (p. 95).

Category of type T objects is a pretopos

Now we show that the full subcategory of type T objects is a pretopos. It is clear that 0×000 \times 0 \cong 0 (since 010 \to 1 is monic), so 00 is a type T object.

Lemma

If AA is type A and TT is type T, then there exists exactly one map ATA \to T. Type T objects are characterized by this property.

Proof

There is exactly one morphism A0A \to 0. Hence morphisms ATA \to T are in bijection with maps A0×T0A \to 0 \times T \cong 0, of which there is exactly one. For the second statement, suppose that XX has the property that there is exactly one map AXA \to X for each type A object. Such objects XX are closed under products, and 00 is such an object; therefore 0×X0 \times X is such an object. On the other hand, 0×X0 \times X is of type A since it projects to 00. Hence there is at most one morphism 0×X0×X0 \times X \to 0 \times X, and it follows that 0×X00×X0 \times X \to 0 \to 0 \times X is the identity, so that 0×X00 \times X \cong 0. Thus XX is of type T.

Corollary

The initial object 00 is strict in the category of type T objects.

Proof

Given TT of type T and T0T \to 0, we know TT is type A, and therefore by Lemma there is exactly one map TTT \to T. Hence T0TT \to 0 \to T is the identity, and of course so is 0T00 \to T \to 0. So TT is initial.

Corollary

The full subcategory of objects of type T is closed under products, coproducts, subobjects, and quotient objects.

Proof

Closure under products and subobjects is immediate from Lemma . Closure under quotients and coproducts follows from axiom 4.

Theorem

The full subcategory of objects of type T is a pretopos.

Proof

Corollary gives finite completeness, coproducts, and quotients of kernel pairs. Axiom 6 gives that in the full subcategory of TT objects, pushouts along monos are stable under pullback, and the initial object is stable under pullback in the category of T objects, because it is strict by Corollary . It follows that coproducts are universal in the category of T objects. They are also disjoint by an earlier result, so the category of T objects is extensive. It is also effective regular by axiom 1, hence a pretopos.

Splitting into type A and type T objects

In this section we show that every AT category CC is the product of the abelian category C AC_A of type A objects and the pretopos C TC_T of type T objects.

Let us first observe that the coreflector (see Proposition )

0×:CC A0 \times {-}\colon C \to C_A

is left exact (as all coreflectors are), and

  • Preserves the initial object (by axiom 2),

  • Preserves pushouts of kernel pairs, and pushouts of pairs of arrows one of which is monic (axiom 4),

  • And therefore also preserves coproducts.

A functor between AT categories which is left exact and which preserves such classes of finite colimits may be called a morphism of AT categories. Hence the coreflector is a morphism of AT categories.

Next let us observe that the subcategory C TCC_T \hookrightarrow C is reflective; the reflector is the functor TT defined above by means of a suitable pushout. The reflector clearly preserves any colimits that exist, and preserves pullbacks (by axiom 7), so

T:CC TT\colon C \to C_T

is also a morphism of AT categories. Therefore we have a morphism of AT categories

F=0×,T:CC A×C TF = \langle 0 \times {-}, T \rangle\colon C \to C_A \times C_T

and now we wish to prove, under a suitable additional axiom (also satisfied by every abelian category and every pretopos), that this is an equivalence.

Lemma

The functor FF is faithful.

Proof

The functor FF is left exact and therefore preserves kernels. By axiom 8, FF reflects isomorphisms. It follows immediately from these two facts that FF is faithful.

Remark

With this result, Freyd’s “first task” is complete: any AT category may be faithfully represented in a product of a pretopos and an abelian category.

At this point we bring in axiom AE, def. , for its debut appearance. It may be helpful to consider that TXT X is by definition the cokernel appearing in the exact sequence

0×Xπ 2Xcoker(π 2)0 \times X \stackrel{\pi_2}{\to} X \to coker(\pi_2)

and that axiom AE asserts that this exact sequence splits (with splitting ξ:coker(π 2)X\xi: coker(\pi_2) \to X), in such a way that XX is the coproduct of the end terms.

Proposition

The functor FF is full.

Suppose given maps f:0×X0×Yf: 0 \times X \to 0 \times Y, g:TXTYg: T X \to T Y, and contemplate the diagram

0×X π 2 X TX f g 0×Y π 2 Y ξ TY\array{ 0 \times X & \stackrel{\pi_2}{\to} & X & \to & T X \\ f \Big\downarrow & & & & \Big\downarrow g \\ 0 \times Y & \underset{\pi_2}{\to} & Y & \stackrel{\overset{\xi}{\leftarrow}}{\to} & T Y }

Since X=(0×X)+TXX = (0 \times X) + T X, the two obvious arrows 0×XY0 \times X \to Y, TXYT X \to Y combine to give an arrow (f,g):XY(f, g): X \to Y. It is straightforward to check that 0×(f,g)=f0 \times (f, g) = f (because the functor 0×0 \times - kills the type T summands) and that T(f,g)=gT(f, g) = g.

Theorem

The functor FF is an equivalence.

Proof

All that is left to check is that FF is essentially surjective, but this is clear because given a type A object XX and a type T object YY, we have 0×(X+Y)X0 \times (X + Y) \cong X (use axiom 4 plus the fact that 0×Y00 \times Y \cong 0), and

T(X+Y)TX+TY0+TYYT(X + Y) \cong T X + T Y \cong 0 + T Y \cong Y

which completes the proof.

This completes Freyd’s “second task”.

Topos-theoretic considerations

In order to beef up the type T objects to a topos, Freyd imposes some extra structure on top of the AT axioms. Since it is now the topos axioms that are in the ascendant, I will christen these categories TA categories.

Thus, a TA category is an AT category CC together with

  • Functions P,E:Ob(C)Ob(C)P, E\colon Ob(C) \to Ob(C),

  • Functions l,r:Ob(C)Mor(C)l, r\colon Ob(C) \to Mor(C), of the form lX:EXPXl X\colon E X \to P X, rX:EXXr X\colon E X \to X,

  • An operation Λ\Lambda which assigns to each pair of morphisms f:RYf\colon R \to Y, g:RXg\colon R \to X (where RR and YY are assumed to be type T), a morphism Λ(f,g):YPX\Lambda(f, g)\colon Y \to P X.

This data is to satisfy the following axioms:

  1. lX,rX:EXPX×X\langle l X, r X \rangle: E X \to P X \times X is monic and PXP X, EXE X are of type T,

  2. The composite of relations

    YΛ(f,g)PX(lX) opEXrXXY \stackrel{\Lambda(f, g)}{\to} P X \stackrel{(l X)^{op}}{\to} E X \stackrel{r X}{\to} X

    is equal to

    Yf opRgX,Y \stackrel{f^{op}}{\to} R \stackrel{g}{\to} X,
  3. If

    R g EX f lX Y h TX\array{ R & \stackrel{g}{\to} & E X \\ f \downarrow & & \downarrow l X \\ Y & \underset{h}{\to} & T X }

    is a pullback and RR is of type T, then Λ(f,(rX)g)=h\Lambda(f, (r X) \circ g) = h.

Freyd: “Note that in the full category of type-T objects, PP yields power-objects with EE, ll, rr naming the universal relations. (The third axiom provides the uniqueness condition.)”

And: “In any abelian category the only type T object is the zero object, which forces P=E=0P = E = 0 for abelian categories.”

And finally: “It’s routine that both 0×0 \times {-} and TT preserve the new structure. We can now remove the existential from AE. Define ξ:TXX\xi\colon T X \to X as the image of rXr X. The third task is finished with:

  1. For every XX, 0×XXξTX0 \times X \to X \stackrel{\xi}{\leftarrow} T X is a coproduct diagram.“

This section is likely to be rewritten and cleaned up.

References

Last revised on July 11, 2023 at 09:14:15. See the history of this page for a list of all contributions to it.