A four-stage model for language evolution under the effects of human self-domestication

Abstract We propose that languages (and seemingly our language capabilities) evolved gradually as a result of being engaged in an active feedback loop with human self-domestication. Our proposal primarily builds on the management of aggression: it ties early stages of the evolution of languages with the taming of reactive aggression, whereas it ties late stages with the rise of proactive aggression. Overall, we posit a four-stage model, from simple (even single-word) utterances (Stage One), to simple two-slot grammars (Stage Two), to languages optimized for in-group communication (Stage Three), to languages better suited for communicating non-shared knowledge to strangers (Stage Four).