Algorithms in the historical emergence of word senses.
Human language relies on a finite lexicon to express a potentiallyinfinite set of ideas. A key result of this tension is that wordsacquire novel senses over time. However, the cognitive processesthat underlie the historical emergence of new word senses arepoorly understood. Here, we present a computational frameworkthat formalizes competing views of how new senses of a wordmight emerge by attaching to existing senses of the word. We testthe ability of the models to predict the temporal order in whichthe senses of individual words have emerged, using an historicallexicon of English spanning the past millennium. Our findingssuggest that word senses emerge in predictable ways, followingan historical path that reflects cognitive efficiency, predominantlythrough a process of nearest-neighbor chaining. Our work contributesa formal account of the generative processes that underlielexical evolution.