How strict is strictification? (Q1730856)

The study of this relation is motivated by the fact that the bases of $\mathbf{Gray}$-category theory and tricategory theory are distinct. They are respectively the $\mathbf{Gray}$-category of 2-categories and tricategory of bicategories. The relation between underlying categories is given by strictification adjunction: \[\mathbf{2-Cat}\to^{\bot}\mathbf{Bicat}_{st}\tag{1.1}\] [\textit{R. Gordon} et al., Coherence for tricategories. Providence, RI: American Mathematical Society (AMS) (1995; Zbl 0836.18001)]. Here \textbf{2-Cat} is the category of 2-categories and 2-functors, and \textbf{Bicat} is the category of bicategories and pseudofunctors, \(\bot\) is the inclusion of \textbf{2-Cat} into \textbf{Bicat} and \textbf{st} sends a bicategory to its strictification. The higher structure of this adjunction is studied in this paper by A. Campbell. Central role is played by the three-dimensional universal property of strictification (Corollary 3.6). The proof of this property contains, however, several subtle points that have to be mentioned. This universal property ``states that for every bicategory \(A\) and 2-category \(B\), the hom-set bijection \textbf{2-Cat}(\textbf{st}A,B)\(\cong\)\textbf{Hom}(A,B) of the strictification adjunction underlies an isomorphism of 2-categories \[\textbf{Ps}(\textbf{st}A,B)\cong\textbf{Hom}(A,B)\,\,(1.2)\] where \(\textbf{Ps}(-,-)\) and \(\textbf{Hom}(-,-)\) denote the homs of the textbf{Gray}-category of 2-categories and the tricategory of bicategories respectively (whose morphisms are 2-functors and pseudofunctors respectively, and whose 2-cells and 3-cells are in both cases pseudonatural transformations and modifications)''. One is tempted to look for some kind of a ``three-dimensional adjunction'' underlied by the strictification adjunction but ``the ``weakness'' provided by the definitions of tricategory theory is still required in order to realise the higher structure of the strictification adjunction in this setting''. Alternative approach developed by author permits to realise ``the same three-dimensional higher structure of the strictification adjunction ... by a {\em strictly} bicategory-enriched adjunction''. (In this approach bicategory-enriched categories are the categories whose hom-objects are bicategories.) In the author's approach a framework for bicategory-enriched categories based on the ``{\em symmetric monoidal closed multicategory} \textsf{Bicat} of bicategories introduced by Verity''. Main advantage is that the category of bicategories becomes {\em strictly} bicategory-enriched category. From the author's summary: ``Using standard arguments of enriched category theory generalised in \S2 to the context of enrichment of and over symmetric multicategories, in \S3 we prove our main theorem (Theorem 3.8) as a formal consequence of the three-dimensional universal property of strictification (1.2): Main Theorem. The strictification adjunction (1.1) underlies an adjunction of \textsf{Bicat}-enriched categories and, moreover, an adjunction of \textsf{Bicat}-{\em enriched symmetric multicategories}.'' As an application of the strictness of strictification revealed by this main theorem, we obtain (Proposition 3.11) a hitherto undiscovered \textbf{Gray}-{\em category of bicategories}, whose underlying category is the category \textbf{Bicat} of bicategories and pseudofunctors, and whose hom 2-categories \(\textbf{st\,Hom}(A, B)\) are the strictifications of the hom bicategories \(\textbf{Hom}(A,B)\). This \textbf{Gray}-category is triequivalent (via a bijective-on-objects, bijective-on-morphisms trihomomorphism) to the tricategory of bicategories''. Paper includes an appendix: ``The multicategory of pseudo double categories''.