# Compatibility of monoid and comonoid structures when monoidal product is a product

Let $\mathcal{C}$ be a category with finite products. Then $\mathcal{C}$ is a braided monoidal category with the product as the monoidal product and terminal object as the monoidal unit, and braiding $\tau_{A,B}$ the unique isomorphism $A\times B \cong B\times A$ induced since both of sides are the product of $A$ and $B$. For every object $A$ in $\mathcal{C}$, $A$ has the unique structure of a comonoid with comultiplication given by the diagonal map, $\Delta$, and counit the unique map to the terminal object. Now suppose $A$ is a monoidal object with multiplication $\mu$. I'd like to show that the monoidal and comonoidal structures are compatible in the sense that $\Delta$ is a monoid morphism. This is part of the justification for calling Hopf algebras "group objects" in braided monoidal categories. I'm stuck trying to show that $\Delta\circ\mu = \mu\times\mu\circ(\text{id}\times\tau\times\text{id})\circ \Delta \times \Delta$.

-

This has very little to do with monoidal categories and much more to do with the properties of the cartesian product (and, very specifically, $\Delta$). Observe:
$$\begin{split} \Delta(\mu(x, y)) & = (\mu(x, y), \mu(x, y)) \\ & = (\mu \times \mu) (x, y, x, y) \\ & = ((\mu \times \mu) \circ (\textrm{id} \times \tau \times \textrm{id})) (x, x, y, y) \\ & = ((\mu \times \mu) \circ (\textrm{id} \times \tau \times \textrm{id}) \circ (\Delta \times \Delta)) (x, y) \end{split}$$
You can interpret $x$ and $y$ as generalised elements $U \to A$.
 Thank you for your answer. I've never heard of generalized elements before. Are you saying that we can use generalized elements to reduce the problem to a proof that the property holds in the category of sets with Cartesian product? I tried looking online on Wikipedia and nLab, but so far haven't been able to understand how generalized elements work yet. Is there a good reference for this? Is it in Maclane? – Seth Jun 3 '12 at 18:39 -ish. Roughly speaking, what we are doing is taking the Yoneda representation of a category. Then a morphism $U \to A$ genuinely is an "element" – specifically an element of the representable presheaf $\textrm{Hom}(-, A)$. Mac Lane does a special kind of generalised element in Chapter VIII. Lawvere and Rosebrugh [Sets for mathematics] also discuss generalised elements a bit. – Zhen Lin Jun 3 '12 at 18:51