3.2.4 Pullbacks

Let $\webleft (X,x_{0}\webright )$, $\webleft (Y,y_{0}\webright )$, and $\webleft (Z,z_{0}\webright )$ be pointed sets and let $f\colon \webleft (X,x_{0}\webright )\to \webleft (Z,z_{0}\webright )$ and $g\colon \webleft (Y,y_{0}\webright )\to \webleft (Z,z_{0}\webright )$ be morphisms of pointed sets.

The pullback of $\webleft (X,x_{0}\webright )$ and $\webleft (Y,y_{0}\webright )$ over $\webleft (Z,z_{0}\webright )$ along $\webleft (f,g\webright )$ is the pair consisting of:

  • The Limit. The pointed set $\webleft (X\times _{Z}Y,\webleft (x_{0},y_{0}\webright )\webright )$.
  • The Cone. The morphisms of pointed sets

    \begin{align*} \text{pr}_{1} & \colon \webleft (X\times _{Z}Y,\webleft (x_{0},y_{0}\webright )\webright )\to \webleft (X,x_{0}\webright ),\\ \text{pr}_{2} & \colon \webleft (X\times _{Z}Y,\webleft (x_{0},y_{0}\webright )\webright )\to \webleft (Y,y_{0}\webright ) \end{align*}

    defined by

    \begin{align*} \text{pr}_{1}\webleft (x,y\webright ) & \mathrel {\smash {\overset {\mathclap {\scriptscriptstyle \text{def}}}=}}x,\\ \text{pr}_{2}\webleft (x,y\webright ) & \mathrel {\smash {\overset {\mathclap {\scriptscriptstyle \text{def}}}=}}y \end{align*}

    for each $\webleft (x,y\webright )\in X\times _{Z}Y$.

We claim that $X\times _{Z}Y$ is the categorical pullback of $\webleft (X,x_{0}\webright )$ and $\webleft (Y,y_{0}\webright )$ over $\webleft (Z,z_{0}\webright )$ with respect to $\webleft (f,g\webright )$ in $\mathsf{Sets}_{*}$. First we need to check that the relevant pullback diagram commutes, i.e. that we have

Indeed, given $\webleft (x,y\webright )\in X\times _{Z}Y$, we have

\begin{align*} \webleft [f\circ \text{pr}_{1}\webright ]\webleft (x,y\webright ) & = f\webleft (\text{pr}_{1}\webleft (x,y\webright )\webright )\\ & = f\webleft (x\webright )\\ & = g\webleft (y\webright )\\ & = g\webleft (\text{pr}_{2}\webleft (x,y\webright )\webright )\\ & = \webleft [g\circ \text{pr}_{2}\webright ]\webleft (x,y\webright ),\end{align*}

where $f\webleft (x\webright )=g\webleft (y\webright )$ since $\webleft (x,y\webright )\in X\times _{Z}Y$. Next, we prove that $X\times _{Z}Y$ satisfies the universal property of the pullback. Suppose we have a diagram of the form

in $\mathsf{Sets}_{*}$. Then there exists a unique morphism of pointed sets

\[ \phi \colon \webleft (P,*\webright )\to \webleft (X\times _{Z}Y,\webleft (x_{0},y_{0}\webright )\webright ) \]

making the diagram

commute, being uniquely determined by the conditions

\begin{align*} \text{pr}_{1}\circ \phi & = p_{1},\\ \text{pr}_{2}\circ \phi & = p_{2}\end{align*}

via

\[ \phi \webleft (x\webright )=\webleft (p_{1}\webleft (x\webright ),p_{2}\webleft (x\webright )\webright ) \]

for each $x\in P$, where we note that $\webleft (p_{1}\webleft (x\webright ),p_{2}\webleft (x\webright )\webright )\in X\times Y$ indeed lies in $X\times _{Z}Y$ by the condition

\[ f\circ p_{1}=g\circ p_{2}, \]

which gives

\[ f\webleft (p_{1}\webleft (x\webright )\webright )=g\webleft (p_{2}\webleft (x\webright )\webright ) \]

for each $x\in P$, so that $\webleft (p_{1}\webleft (x\webright ),p_{2}\webleft (x\webright )\webright )\in X\times _{Z}Y$. Lastly, we note that $\phi $ is indeed a morphism of pointed sets, as we have

\begin{align*} \phi \webleft (*\webright ) & = \webleft (p_{1}\webleft (*\webright ),p_{2}\webleft (*\webright )\webright )\\ & = \webleft (x_{0},y_{0}\webright ),\end{align*}

where we have used that $p_{1}$ and $p_{2}$ are morphisms of pointed sets.

Let $\webleft (X,x_{0}\webright )$, $\webleft (Y,y_{0}\webright )$, $\webleft (Z,z_{0}\webright )$, and $\webleft (A,a_{0}\webright )$ be pointed sets.

  1. Functoriality. The assignment $\webleft (X,Y,Z,f,g\webright )\mapsto X\times _{f,Z,g}Y$ defines a functor
    \[ -_{1}\times _{-_{3}}-_{1}\colon \mathsf{Fun}\webleft (\mathcal{P},\mathsf{Sets}_{*}\webright )\to \mathsf{Sets}_{*}, \]

    where $\mathcal{P}$ is the category that looks like this:

    In particular, the action on morphisms of $-_{1}\times _{-_{3}}-_{1}$ is given by sending a morphism

    in $\mathsf{Fun}\webleft (\mathcal{P},\mathsf{Sets}_{*}\webright )$ to the morphism of pointed sets

    \[ \xi \colon \webleft (X\times _{Z}Y,\webleft (x_{0},y_{0}\webright )\webright )\overset {\exists !}{\to }\webleft (X'\times _{Z'}Y',\webleft (x'_{0},y'_{0}\webright )\webright ) \]

    given by

    \[ \xi \webleft (x,y\webright )\mathrel {\smash {\overset {\mathclap {\scriptscriptstyle \text{def}}}=}}\webleft (\phi \webleft (x\webright ),\psi \webleft (y\webright )\webright ) \]

    for each $\webleft (x,y\webright )\in X\times _{Z}Y$, which is the unique morphism of pointed sets making the diagram

    commute.

  2. Associativity. Given a diagram

    in $\mathsf{Sets}_{*}$, we have isomorphisms of pointed sets

    \[ \webleft (X\times _{W}Y\webright )\times _{V}Z\cong \webleft (X\times _{W}Y\webright )\times _{Y}\webleft (Y\times _{V}Z\webright ) \cong X\times _{W}\webleft (Y\times _{V}Z\webright ), \]

    where these pullbacks are built as in the diagrams

  3. Unitality. We have isomorphisms of pointed sets
  4. Commutativity. We have an isomorphism of pointed sets
  5. Interaction With Products. We have an isomorphism of pointed sets
  6. Symmetric Monoidality. The triple $\webleft (\mathsf{Sets}_{*},\times _{X},X\webright )$ is a symmetric monoidal category.

Item 1: Functoriality
This is a special case of functoriality of co/limits, of , with the explicit expression for $\xi $ following from the commutativity of the cube pullback diagram.
Item 2: Associativity
This follows from Chapter 2: Constructions With Sets, Item 2 of Proposition 2.1.4.1.3.
Item 3: Unitality
This follows from Chapter 2: Constructions With Sets, Item 3 of Proposition 2.1.4.1.3.
Item 4: Commutativity
This follows from Chapter 2: Constructions With Sets, Item 4 of Proposition 2.1.4.1.3.
Item 5: Interaction With Products
This follows from Chapter 2: Constructions With Sets, Item 6 of Proposition 2.1.4.1.3.
Item 6: Symmetric Monoidality
This follows from Chapter 2: Constructions With Sets, Item 7 of Proposition 2.1.4.1.3.


Noticed something off, or have any comments? Feel free to reach out!


You can also use the contact form below: