A geometric morphism $f\colon E\to F$ between topoi is localic if every object of $E$ is a subquotient of an object in the inverse image of $f$: of the form $f^*(x)$.
Any geometric morphism between localic topoi is localic.
Any geometric embedding is localic.
If $g:C\to D$ is a faithful functor between small categories, then the induced geometric morphism $Set^C \to Set^D$ is localic.
A Grothendieck topos is a localic topos if and only if its unique global section geometric morphism to Set is a localic geometric morphism.
Thus, in general we regard a localic geometric morphism $E\to S$ as exhibiting $E$ as a “localic $S$-topos”.
This is supported by the following fact.
For any base $S$, the 2-category of localic $S$-toposes (i.e. the full sub-2-category
of the over-category Topos over $S$ spanned by the localic morphisms into $S$) is equivalent to the 2-category of internal locales in $S$
Concretely, the internal locale in $\mathcal{E}$ defined by a localic geometric morphism $(f^* \dashv f_*) : \mathcal{F} \to \mathcal{E}$ is the formal dual to the direct image $f_*(\Omega_{\mathcal{F}})$ of the subobject classifier of $\mathcal{F}$, regarded as an internal poset (as described there) and $\mathcal{F}$ is equivalent to the internal category of sheaves over $f_*(\Omega_{\mathcal{F}})$.
The last bit is lemma 1.2 in (Johnstone).
Localic geometric morphisms are the right class of a 2-categorical orthogonal factorization system on the 2-category Topos of topoi. The corresponding left class is the class of hyperconnected geometric morphisms.
This is the main statement in (Johnstone).
Localic geometric morphisms are defined in def. 4.6.1 of
The discussion there is based on
Last revised on September 22, 2015 at 09:36:05. See the history of this page for a list of all contributions to it.