The singing Neanderthals: The origins of music, language, mind and body, and: Music, language, and the brain (review)

Aniruddh Patel in Music, language, and the brain and Steven Mithen in The singing Neanderthals both suggest that the neural bases of human language and music are linked, but that is the only common element that binds these books. Although P could have made a stronger case for his primary thesis—which is that a common neural substrate underlies human musical and linguistic capabilities—if he had a better grasp of speech physiology, acoustics, and analysis, he presents a useful survey of present research on some of the common and distinctive elements of music and language and an agenda for future research. M, however, ignores the principles of evolutionary biology that have yielded well-attested data that refute his thesis, which is that Neanderthals communicated using an inventory of immutable 'holistic' song-like utterances that lacked discrete words or syntax. M instead directs a barrage of factoids on the reader—short accounts of hundreds of studies, often misinterpreted and often irrelevant, and citations of studies that he may or may not have actually read.

The neurophysiologic studies that support P's thesis, that language and music devolve from shared neural resources, are discussed late in his book, starting on page 267. P's organization is justified since he must first present the necessary background material. Many readers might not be familiar with either musical terminology or the 'syntax' of music. Other readers might not be familiar with the physiology or perception of speech, speech analysis, syntax, and acoustic analysis. P conveys this information with varying degrees of success.

P first must establish the similarities and differences that he believes mark music and speech. According to P, musical form is determined by pitch sequences, which in the domain of language equate with the fundamental frequency of phonation (F0)—the rate at which the vocal cords of the larynx open and close. This supposition is reasonable since the 'first' musical instrument undoubtedly was the human voice. As the vocal cords rapidly open and close (the rate varies from a low of 60 Hz to 1.5 kHz for young children), they release a series of 'puffs' of air containing energy at F0 and its 'harmonics'—integral multiples of F0. This source of acoustic energy enters the airway above the larynx, the supralaryngeal vocal tract (SVT). The SVT acts as an acoustic filter, allowing maximum energy to pass through it at particular frequencies. The shape of the SVT determines these frequencies—termed 'formants'—which are the major determinants of the speech sounds, the 'segmental' phonetic elements that convey words. The absolute values of the formant frequencies of different speakers for the same sound differ, scaled to the length of the speaker's SVT; the relative values of the formants specify phonetic elements.

P views the formant distinctions that mark vowels and consonants to be analogous to musical 'timbre', which he claims plays a minor role in musical form. He notes that timbre is determined by the balance of acoustic energy in the harmonics of F0. Since formant frequencies determine the balance of spectral energy in the speech signal by reducing the energy of individual harmonics (its timbre), P, in discussing the commonality that may exist between music and speech, focuses on the pattern of F0 variation in the speech signal—prosodic linguistic elements such as stress and intonation. Detailed discussions of linguistic analyses of stress and intonation thus occupy a major part of the book. This brings P into a minefield—many of the linguistic studies of stress and intonation that are discussed have serious errors, but he lacks the background in speech physiology and acoustic analysis that would have revealed these problems. For example, P takes at face value the incorrect claim that stressed syllables are produced by the expiratory...