From a linguist’s point of view, the ability to vocalize new sounds may not seem to be a critical component of language. Yet when this ability is impaired, the social and emotional consequences for the affected individual can be severe, as evidenced by those suffering from developmental or injury-related speech disorders. How are we to understand this vocal learning trait and where should it be placed within a framework for language evolution? Here, I argue that studying the supporting brain pathways that are affected in vocal learning disorders is a good place to start. Since such study is largely limited to non-invasive methods in humans, investigating other animals that possess this rare trait paves the way for a comparative analysis of the molecular, cellular and synaptic basis of vocal production learning, including human speech. Such inquiry can highlight shared evolutionary pathways as well as key detours.