“…This idealised picture is often challenged by cases in which clitics can be stressed as in (6), have the same form of strong pronouns as the French enclitic in (7), do not climb in compound tenses as in (8), may be separated from the verb by certain adverbs as in (9), or occur as the complement of a (lexical) preposition as in (10) (6)-(10) challenge or support the proposed distinction into pronominal classes? Laenzlinger 1993Laenzlinger , 1994Repetti 2006, 2014;Repetti 2016;Cardinaletti 2015aCardinaletti , 2015b have argued that the clitics in (6)-(10) are in fact weak elements in the sense of Cardinaletti 1991Cardinaletti , 1998Cardinaletti & Starke 1999. Conversely, I argue that the phenomena in (6)-(10) end up challenging class-based accounts like (1)-(2) as none of the properties distinguishing weak from clitic pronouns, listed in (11), hold systematically across languages.…”