Indexed text attachment
galois-theory-in-higher-categories.tex
Source for Galois Theory in Higher Categories
View paper · Download original file
Lines 3401–3500 of 4071. Source text is displayed literally.
3401invertibility.3402 3403\begin{proposition}[Effectiveness of internal groupoids]3404\label{prop:geo-effective-groupoid}3405The map $G_0\to Q$ is an effective epimorphism, and there3406are compatible equivalences3407\[3408 G_n \simeq3409 \underbrace{G_0\times_Q\cdots\times_QG_0}_{n+1}.3410\]3411\end{proposition}3412 3413\begin{proof}3414Define a simplicial object $E_\bullet$ by $E_n=G_{n+1}$,3415where the simplicial operators retain the first vertex and3416act on the remaining vertices. In degree $n$, its map to3417$G_n$ forgets that first vertex. There is also an augmentation3418$E_\bullet\to G_0$ given by the retained vertex. Duplicating3419the retained vertex is an extra degeneracy contracting this3420augmentation, so $|E_\bullet|\simeq G_0$.3421 3422The map $E_\bullet\to G_\bullet$ is Cartesian. To check this,3423first pull it back along a vertex of a simplex $G_n$.3424Over the first vertex of the remaining chain, the assertion3425is a Segal identity since the extra data is the one incoming3426arrow. Over any other vertex, the incoming arrow can be3427transported forward along the chosen segment of that chain.3428Composition by an invertible arrow is an equivalence on the3429corresponding space of arrows, with inverse given by3430composition by its inverse. This argument is functorial on3431generalized points and proves the pullback identity for any3432chosen vertex. To check an arbitrary simplicial operator3433$G_n\to G_m$, evaluate the two candidate pullbacks over a3434vertex of $G_m$. Both identify with the same pullback of3435$E_0\to G_0$ along the composite vertex of $G_n$.3436 3437Descent now identifies $E_0$ with the pullback of3438$|E_\bullet|$ along $G_0\to Q$. Its two projections are the3439endpoints of the arrow in $G_1$. Hence3440$G_1\simeq G_0\times_Q G_0$. Compatibility with longer3441chains follows by the Segal maps and their composition3442coherences. The augmented Čech nerve therefore realizes to3443$Q$, proving effectiveness. This is the shift argument3444of \cite[\S6.10]{rezk} in the notation of internal groupoids.3445\end{proof}3446 3447A $G_\bullet$-object is a Cartesian transformation3448$A_\bullet\to G_\bullet$. It is given first by an object3449$A_0\to G_0$, then by transport of this object along3450arrows of $G_1$, with composition homotopies on $G_2$3451and their higher compatibility. Equivariant morphisms are3452maps of such transformations. The corresponding3453$\infty$-category will be denoted $\operatorname{Act}_{\mathcal X}(G_\bullet)$.3454Applying \eqref{eq:higher-slice-vankampen} gives3455\begin{equation}3456 \operatorname{Act}_{\mathcal X}(G_\bullet)3457 \ \simeq\ \lim_{[n]\in\Delta}\mathcal X_{/G_n}3458 \ \simeq\ \mathcal X_{/Q}.3459 \label{eq:higher-quotient-actions}3460\end{equation}3461The inverse sends $A\to Q$ to the family $A\times_QG_n$.3462The forward map sends a Cartesian family to its realization.3463Under Theorem~\ref{thm:geometric-colimit-slices}, the3464right-hand side is also the colimit of the simplicial3465diagram of étale geometric morphisms $\mathcal X_{/G_n}$.3466The two variance conventions agree because the geometric3467morphisms point in the direction opposite to their inverse3468image functors.3469 3470For the special case of $U\to1_{\mathcal X}$ effective epi,3471let $G_n=U^{n+1}$. Then $Q\simeq1_{\mathcal X}$ and3472\eqref{eq:higher-quotient-actions} recovers $\mathcal X$ itself3473from the Čech diagram of $\mathcal X_{/U}\to\mathcal X$.3474Indeed Theorem~\ref{thm:geo-sections} identifies3475\[3476 \underbrace{\mathcal X_{/U}\times_{\mathcal X}\cdots3477 \times_{\mathcal X}\mathcal X_{/U}}_{n+1}3478 \simeq \mathcal X_{/U^{n+1}},3479\]3480where the fiber products on the left are in3481$\mathbf{Top}_{\infty}$. The inverse image of this projection3482is conservative since evaluation of a Cartesian family at3483degree zero detects equivalences. Conversely, if3484$\pi_U^*$ is conservative, it reflects the equivalence3485obtained by pulling back $\operatorname{im}(U\to1)\to1$3486along $U$. This image must be all of $1$, so $U\to1$ is3487an effective epimorphism. For étale geometric morphisms,3488surjectivity in the sense of conservative inverse image3489thus agrees with being induced by an effective epimorphism3490of objects. All the formulas hold in a slice over any3491specified base object.3492 3493Presentations here are invariant under refinement in a3494concrete sense. If $W\to U\to1_{\mathcal X}$ are effective3495epimorphisms, their Čech nerves have a map3496$W^{\bullet+1}\to U^{\bullet+1}$. Pullback on actions is3497an equivalence because under \eqref{eq:higher-quotient-actions}3498it identifies with the identity functor of $\mathcal X$.3499Two covers $U\to1_{\mathcal X}$ and $V\to1_{\mathcal X}$3500admit a common refinement by $U\times V$. Both projections