similarities could have come about completely fortuitously and arisen
purely by parallel evolution.This parallelism model rejects any notion
of shared ancestral features.Second,the similarities could have arisen
from continuing interaction between discrete music and language
modules,such that effective binding mechanisms evolved to confer
musical properties onto language and linguistic properties onto music
(binding model).Third,music could have evolved as an outgrowth of lan-
guage (music outgrowth model).Fourth,language could have evolved
as an outgrowth of music (language outgrowth model).Fifth,these
similarities could have arisen due to the occurrence of an ancestral stage
276 Steven Brown
Figure 16.2
Five models for the evolution of the shared properties of music and language.In the par-
allelism model,language’s evolution from a protolinguistic precursor and music’s evolu-
tion from a protomusical precursor are thought to occur by completely independent
processes.The binding model is quite similar except that it posits evolution of binding
mechanisms that confer linguistic properties onto music and musical properties onto lan-
guage (shown by the reciprocal horizontal arrows at the top of the figure).Neither of these
two models invokes any notion of shared ancestral features.The next three models do.In
the music outgrowth model,music is thought to evolve out of a linguistic precursor,
whereas in the language outgrowth model language is thought to evolve out of a musical
precursor.The musilanguage model is another outgrowth model in which shared properties
of music and language are attributed to a common precursor,the musilanguage stage.
Fig.16.2