Add to ORKGThe purpose of this paper is to account for certain gaps in the syntactic distribution of certain combinations of clitics in varieties of Aragonese, Catalan, Occitan and Spanish. A view in which ordering of clitics is only determined by templates is insufficient since templates enter into play after the syntactic component, either in the morphology or in some other post-syntactic component. The alternative I propose is to assume that the ordering of clitic combinations can be done in the syntax, as anticipated by Kayne (1994) and Terzi (1999). I will specifically propose that some of the anomalies involve clitic combinations that are actually non-constituents (split clitics) while in others one element is not actually a clitic but an elemen...