Pullback . category theory
Collection
context | F:(a→z←b)⟶C |
definition | ⟨Fa×FzFb,π⟩:=limF |
Here we consider a functor F from the category a→z←b, consisting of three object and two non-identity arrows fa and fb, to a category C.
Universal property
For readability, let's write A≡Fa,B≡Fb,Z≡Fz,α≡fa and β≡fb.
(In the picture we have X≡Fa,Y≡Fb,Z≡Fz,f≡fa,g≡fb and the pullback object is P≡X×ZY.)
Consider two arrows γ:C[X,A] and δ:C[X,B], which fulfill the structural condition α∘γ=β∘δ. I.e. when forwarded to Z via α, resp. β, they collapse into a single arrow.
Such two arrows γ,δ can be partially glued together, in the sense that they can both be written as a unique arrow u (with codomain A×ZB) followed by the projections πa,πb.
Discussion
The pullback object A×ZB is the full solution to the equation posed by α and β. In Set, it's literally the set of pairs ⟨x,y⟩∈A×ZB⊆A×B, for which α(x)=β(y).
When the category contains a terminal object 1 (where α and β are trivial arrows and form a trivial condition), we have A×1B≅A×B.
The universal property says that all other solution embed in this object, in this is what is meant by full solution.
Special cases
- If πa is an iso, then A×ZB≅A. As A is already the pullback, it alone fully determines the “full solution”.
- If moreover πb is an iso too, the projections, we can consider the equivalent pullback with πb=πa=1A. The universal property now says that arrows γ,δ can be wholly glued together: Up to iso, α∘γ=β∘δ⟹γ=δ.
- In Set, if α=β, the pullback definition says that its elements ⟨x,y⟩ fulfill α(x)=α(y), i.e. here the pullback object is the full collection of pairs of term with give the same α value. If moreover πa is iso, any x determines an ⟨x,y⟩ and hence an y and the universal property translates to α(x)=α(y)⟹x=y. This is just the definition of an injection.
- Back to a general category. If the pullback of α along itself (α=β) is such that a projection πa is iso, we call α a monomorphism. The associated condition reads α∘γ=α∘δ⟹γ=δ.
(In the picture, f is α and g,f are our γ,δ.)
Examples
A finite pullback in Set that I just made up:
- Generally: If F(fb) is the inclusion of a subset Fb⊆Fz in Fz, the pullback is iso to (i.e. in bijection with) F(f−1a)Fb. Further, if F(fa) is an inclusion too, this is in bijection with Fa∩Fb. If the subset-interpretation doesn't apply, the function F(fb) from Fb to Fz should be viewed as defining fibre bundle over Fz and the pullback gives a fibre bundle from Fa×FzFb to Fa.
- A concrete example: Let
Fa={2,4,6,8}, Fb={10,20}, Fz={77,88,99}
F(fa)(2)=77, F(fa)(4)=77, F(fa)(6)=88, F(fa)(8)=99
F(fb)(10)=88, F(fb)(20)=77
Then
Fa×FzFb={⟨2,20⟩,⟨4,20⟩,⟨6,10⟩}
and π are projections like for the product.
Two more prominent examples:
- If M is a manifold, p:B→M is a fibre bundle over M and f:X→Y is an embedding of another manifold X into Y, the the pullback object is a fibre bundle over X, namely the differential geometric pullback.
- In Set, if the “central object” is Ω={0,1}, the right function comes from a singleton 1 and if the left function χ:X→Ω is a characteristic function, then a pullback object is a subset of X: It's defined as collection of arguments where the characteristic function agrees that the value is 1. This works also for more general “fuzzy” Ω. These Ω are the “objects of truth values” and are called subobject classifiers.
Digression: The exponential object BA is a prominent example of an object which isn't a limit, but it can be specifies via universal morphism construction. For sets or types, that's the function space A→B and for propositions it's the implication. If a category has product, exponential object and a terminal object, then it's called Cartesian closed. A Cartesian closed category with subobject classifier is a topos. We see now how a topos is a general kind of set theory, and simultaneously defines an internal logic.
Reference
Wikipedia: Pullback (category theory), Subobject classifier