Running the usual pdflatex-bibtex loop on
\documentclass{article}
\usepackage[american,ngerman]{babel}%%% The book is in German.
\usepackage{filecontents}
\begin{filecontents}{\jobname.bib}
@misc{TheTerminator,
author={James Cameron},
title={The {Terminator}},
year=1984,
language={american}
}
@misc{TerminatorTwo,
author={James Cameron},
title={Terminator 2: Judgement Day},
year=1991,
language={american}
}
@misc{TerminatorThree,
author={Jonathan Mostow},
title={Terminator 3: Rise of the Machines},
year=2003,
language={american}
}
\end{filecontents}
\usepackage{babelbib} % Multilingual bibliographies
\begin{document}
Wir schauten \cite{TheTerminator,TerminatorTwo,TerminatorThree}.
\bibliographystyle{babalpha-fl-gs-sort}%%% Avoid problematic abbreviations such as SS and SA, see https://tex.stackexchange.com/a/441877/165772. Also disabmiguate "Ber89"; see https://tex.stackexchange.com/a/472956/165772.
\bibliography{\jobname}
\end{document}
yields the output
Wir schauten [Cam84, Cam91, Mos03].
followed by the literature list. How can one get
Wir schauten [Cam84, Cam91 und Mos03].
automatically (i.e., not by writing something like \cite{TheTerminator, TerminatorTwo}, and \cite{TerminatorThree}) without changing the workflow, i.e., still using pdflatex, babel, babelbib, bibtex? We are allowed to patch the style file (thx to @moewe!!!) and allowed to write macros.
![Wir schauten [CE84, Cam91 und MES03].//Wir schauten [CE84 und Cam91].//Wir schauten [CE84].](../../images/1ba95757f797047abee607f220ba88e6.webp)
babelmight change it, because the first language set (in the new way ngerman) determines the main language of the document. – Max R Mar 14 '19 at 20:50:(– Max R Mar 14 '19 at 20:56