Revisiting bridging contexts in language change: core properties and the role of frequency vs salience

This paper discusses bridging contexts in language change. Bridging contexts have been treated as central to meaning change and grammaticalization, but their definition – and, by implication, their identification – is not uniformly agreed upon, and there are open questions regarding their empirical status, the mechanism through which they trigger change, and whether they apply to all forms of change involving meaning. I critically review existing definitions and propose a revised one, arguing also that bridging contexts are principally a second-order phenomenon. I argue, further, that bridging contexts are not relevant to all forms of change, but principally – perhaps exclusively – to hearer-driven and metonymy-based forms. Finally, I discuss what specific mechanism allows bridging contexts to trigger change. Some recent studies have proposed that frequency is key, and that an expression will only undergo change if at least 50% of its uses occur in bridging contexts. Adducing empirical support in the form of two case studies, one from the lexicon and one from grammar, I show that this is unnecessary and propose that a more relevant factor is the salience of the innovative interpretation made available in bridging contexts, a hypothesis for which recent experimental work offers initial support.