Language out of Music: The Four Dimensions of Vocal Learning

A growing consensus drawing on research in a wide variety of disciplines has,over the last fifteen years or so, argued the need to revisit Darwin's conjecture of1871 that language may be descended from an existing, musical medium ofcommunication that developed from animal calls. This paper seeks to examine, inan extension of Hocketfs analysis of the design features required for linguisticcommunication, the nature ofthe acoustic information produced and perceived inhuman vocalisation, and to consider the anatomical and neural mechanisms onwhich these depend. An attempt is made to sketch an evolutionary chronology forkey prerequisites of human orality. Cross-species comparisons are employed toilluminate the role of four acoustic variables (pitch, duration, amplitude andtimbre), viewing the potential for human vocal productivity from the perspectiveof animal communication. Although humans are the only species to combineenlrainment to pulse with attunement to precisely-tracked pitches, we alsodepend both for musical interaction and the production and perception of vowelsounds on precise and conscious control ofthe property of timbre. Drawing on,amongst others, Scherer's analyses of emotionally triggered sounds in a varietyof species, and Femald's presentation ofthe similarities oflnfant cries and adultproduction of infant-directed speech in a variety of cultures and languages, a caseis made for the instinctive components of human communication being moremusic-like than language-like. In conclusion, historical and comparative data areemployed to outline the adaptive and exaptive sequence by which human voealcommunication evolved. The roles of selective pressures that conform todifferent adaptive modeis are compared—natural selection, sexual selection,group selection—leading to the proposal that all of these must have played theirpart at different stages in the process in a 'mosaic' model consistent with thedevelopment of other human traits.