Let’s say we have an item with the title being in English, but the container-title is in German. If the language of the item is set to en, this will lead to incorrect capitalization of the German title of the container. What do we do here?
One of the things I had in the back of my mind when promoting titles as object was ML support. But I don’t understand this area well enough to have in mind exactly how that would work, other than that you could tag titles with a language.
title:
lang: de
full: Der ...
The CSLm approach is that each field has a language, defaulting to the item language if empty.
Right now, the only consequence of the language field is casing of English titles. Disabling that for specific fields on an otherwise English item can currently be accomplished by wrapping the field in preserve-case markup tags.
Making fields more broadly into arrays with language specified and linguistic alternatives could be possible (e.g., https://github.com/citation-style-language/schema/issues/327#issuecomment-665990627 or a simpler structure), but it would add a lot to the complexity of applications, processors, and data. I might suggest that we reserve that for a multilingual CSL module.