13
$\begingroup$

By a result of Foltz, and Freyd and Street, a category $C$ is essentially small (i.e. equivalent to a small category) if and only if both $C$ and $[C^{\text{op}}, \mathrm{Set}]$ are locally small. I am interested in knowing to what extent this is true for enriched categories.

Question. Is there an example of a non-thin symmetric closed monoidal category $\mathscr V$ and a $\mathscr V$-category $C$ satisfying the following two conditions?

  1. The $\mathscr V$-enriched presheaf category $[C^{\text{op}}, \mathscr V]$ exists.
  2. $C$ is not Morita-equivalent to a small $\mathscr V$-category (i.e. there does not exist a small $\mathscr V$-category $C'$ such that $[C^{\text{op}}, \mathscr V] \simeq [(C')^{\text{op}}, \mathscr V])$.

Background. Conceptually, the obstruction to such examples is the absence of arbitrary large limits in $\mathscr V$ (needed to form $\mathscr V$-enriched functor categories). Indeed, if $[C^{\text{op}}, \mathscr V]$ exists, then for every pair of $\mathscr V$-functors $F, G : C^{\text{op}} \to \mathscr V$, we have an end $$\int_{c \in C^{\text{op}}} \mathscr V(Fc, Gc)$$ in $\mathscr V$, which in general is a large limit. Since $\mathscr V$ is itself a large category, if it admits arbitrary large limits (in fact, large powers suffice), it must be thin (which is why this case is excluded). Therefore, if it is possible to construct large powers from such ends (assuming $C$ is not "Morita-small"), the question above would be answered in the negative. However, it is not clear to me whether this is possible in general, leaving open the possibility of a (nontrivially) large $\mathscr V$-category admitting a $\mathscr V$-enriched presheaf category.

(It does seem to be true that we can construct large powers of arbitrary objects in $\mathscr V$ from such ends in special cases, e.g. when $C$ is discrete, by taking $F$ to be constant on the unit of $\mathscr V$, and $G$ to be constant on a fixed object of $\mathscr V$.)

$\endgroup$
0

1 Answer 1

-1
$\begingroup$

(this should be a comment rather than an answer but I'm new here so I can't comment yet)

I take it that by "The $\mathscr V$-enriched presheaf category $[C^{op},\mathscr V]$ exists", you mean as an object representing the $2$-functor $D \mapsto [D \times C^{op}, \mathscr V]$. Then taking $\mathscr V = \mathrm{Set}$ and $C = \mathrm{Set}$, we have indeed that

  1. The ($\mathrm{Set}$-)category $[\mathrm{Set}^{op}, \mathrm{Set}]$ exists (as ($\mathrm{Set}$-)Cat is a cartesian closed $2$-category)
  2. $\mathrm{Set}$ is not Morita equivalent to a small category : ($\mathrm{Set}$-)categories are Morita equivalent iff they have equivalent Cauchy completions, and Cauchy completion preserves smallness.

So I guess you mean something different by one of those two conditions ?

$\endgroup$
2
  • $\begingroup$ When $\mathscr V = \mathrm{Set}$, admitting a $\mathscr V$-enriched presheaf category entails that the presheaf category is locally small (i.e. $\mathrm{Set}$-enriched), which is not the case for $[\mathrm{Set}^{\mathrm{op}}, \mathrm{Set}]$. $\endgroup$
    – varkor
    Commented Mar 7 at 18:15
  • $\begingroup$ @Elies Welcome to MathOverflow! I hope you enjoy using it and have a wonderful time here =) $\endgroup$
    – Emily
    Commented Mar 7 at 21:53

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .