Primates (2009) 50:283284 DOI 10.

1007/s10329-009-0136-3

BOOK REVIEW

Nobuo Masataka (Ed.): The origins of language, unraveling evolutionary forces

Received: 4 February 2009 / Accepted: 4 February 2009 / Published online: 4 March 2009 Japan Monkey Centre and Springer 2009

Language development is a process of acquiring motor and cognitive skills which include both gestural and vocal elements. Nobuo Masataka argues that research into gestures and acoustic communication may therefore lead us to a better understanding of the origins of language. The book consists of an introductory chapter followed by eight others, each of which addresses one of the two aspects of language development. The book is relatively short; nevertheless it provides a helpful, succinct overview of various aspects of language evolution and offers interesting new hypotheses. The rst main chapter, by Corballis, reviews the gestural theory of language origin. Four main chapters provide empirical data or reviews on aspects of nonhuman primate vocal communication. Three other chapters deal with aspects of human acoustic communication, i.e. one on the relevance of sound symbolic words, one on motherese, and one on singing behavior. The gestural theory of language origin states that spoken language evolved from manual gestures. While nonhuman primates have little cortical control over vocalization, they have excellent control over arm and hands. The assumption is that natural selection for increased control over vocalization has evolved to excellent control over both vocalization and manual motor patterns in humans. Corballis discusses how the vocal and visual/manual domain are linked. He points out that in modern humans both channels of communication accompany each other and both provide critical, non-redundant information. He links the theory to various topics from behavioral studies,

T. Riede (&) University of Utah, Salt Lake City, UT, USA e-mail: t.riede@utah.edu

research on the mirror neuron system and to the FoxP2 gene mutation of chromosome 7. Research on the origin of human language received important inputs from attempts to test the gestural theory. In Chapter 3, Kita presents an interesting hypothesis that sound symbolic words (or mimetics): (1) provide insight in to the cognitive world of human ancestors, and (2) qualify as components of a protolanguage. Sound symbolic words can be those that mimic actual sounds or those that represent non-auditory senses or psychological states. If those words do not change much over time, or change in a predictable manner, they could provide us with insight into ancient human subjective, emotional states. Mimetics exist in all languages, but they are rare in Indo-European languages and frequent in Japanese. Japanese mimetics refer to a wide range of events and states, being tightly linked to speakers subjective experience of that event or state. According to Kita, this linkage makes them candidates to be windows into the psychological capacity of the speaker. Mazuka et al. report on a data analysis investigating the vocabulary used by adult humans when communicating with children under the age of 18 months by means of infant-directed speech. Infant-directed speech in Japanese shares the same prosody-related features (higher pitch; exaggerated pitch contour; more and longer pauses) of other languages, but the infant-directed vocabulary in Japanese seems to be much larger than, for example, in Indo-European languages. The authors performed a survey of infant-directed speech in mothers and other females and found no differences in the vocabulary. In a second investigation, the authors measured various acoustic features using the same three samples of words. Results indicate that infant-directed vocabulary is a default form for new word formation and imitation of early production forms by children.

123

284

Primates (2009) 50:283284

Two chapters address vocal learning, a capability very important in the acquisition of language during the ontogeny of humans. The complexity of this ability is critical to human language evolution among mammals. Songbirds are another taxon with such a complex vocal learning capability, including a tutor, a sensitive period in the sender, the exercise of motor patterns, and auditory and sometimes social feedback. Koda provides a review on long-term and short-term changes of acoustic properties of nonhuman primate vocalizations. In one of his own studies on Japanese macaques, he studied coo-calls. Such calls are given in various contexts, often in response to other individuals starting to produce coo-calls. He found that acoustic properties (for example fundamental frequency or call duration) change, and changes are particularly prominent in calls uttered at rates higher than 0.5 calls per second. The acoustic changes accelerate with the rate of coo-calls if no vocal response by a conspecic occurs, similar to when crying babies do not get immediate attention, or a human mother changing her prosody to gain an infants attention (infant-directed speech). Koda points out that we are lacking research on acoustic preferences in nonhuman primates. Such studies could shed light on one of the forces driving vocal signal evolution. Yamaguchi and Izumi provide another review on vocal learning in nonhuman primates focusing on (1) vocal responses to conspecics, and (2) context-specicity. Although changes to acoustic properties have been described in nonhuman primate calls, such modication of existing call types or the acquisition of new call types remains controversial. For example, the development of acoustic differences between populations of the same species (dialect-like changes) are interesting examples of long-term vocal changes. However, alternative causal explanations of such acoustic differences, for example morphological differences or aging effects need to be studied further. Facial expressions are important and elaborate components in primate communication, and often occur simultaneously with vocal signals. Ghazanfar and Lewkowicz provide an extensive review on bimodal signal perception in primates. The combined presentation of a visual facial signal and a vocal signal can increase salience and provide more effective detection, learning, and discrimination. The ability to discriminate facial and/or auditory signals is suggested to be an example of developmental narrowing, i.e. an organized pattern (the ability to integrate visual and auditory signals simultaneously) emerges out of an initially diffuse, less organized pattern. For example, before the age of 9 months, infants universally can discriminate various phonetically relevant features (e.g. voice onset time or velar versus uvular ejectives) an ability infants selectively keep if relevant for their native language. They can also discriminate faces of

their own and other races at about equal error rate, however after 9 months they are better at discriminating faces of their own race unless trained with various faces from different races. Empirical data suggest that a similar time line for integration of voice and face information occurs at an early stage. A comparative approach might indicate at what stage bimodal signal integration evolved. Nishimuras review covers the anatomical aspects of speech evolution. The human oral, pharyngeal, and laryngeal anatomy has long been thought to be uniquely distinct, thus explaining, for example, much of the apparent inability of nonhuman primates to speak. However, recent studies suggest otherwise. Evidence is accumulating that nonhuman primates are capable of producing variable formant patterns, even those resembling the acoustic structure of human vowels. They just do not exploit this ability as much as humans do. The vocal tracts of nonhuman mammals are exible, and anatomical differences between humans and nonhuman primates are small. It is important to point out that much of the functional relevance of anatomical differences is not yet fully understood. In order to pinpoint the onset of early speech-like behavior, fossil records have been studied. The interpretation of fossil records with regard to speech remains challenging. Nishimura discusses three structures that received much attention in recent decades. First, the hyoid bone is an anchor point for the tongue, the most important articulator in human speech. The bones shape and size presumably enables conclusions about the exibility of tongue movement. Second, the size of the hypoglossal canal is correlated with the size of the respective nerve. It is assumed that the larger the hypoglossal nerve, the more bers exist, supplying more muscle subunits, thereby allowing greater exibility of the tongue. Third, the size of the motor speech area of the frontal cortex is estimated from endocranial casts and presumably correlates with cognitive abilities. In the nal chapter, Masataka discusses possible parallels between nonhuman vocalization and human music/ singing. The comparison is intriguing because many would intuitively agree that animal vocalization sometimes sounds somewhat like human singing. However, are those parallels drawn on acoustic features alone (rhythm, repetition, pitch contour, etc.) or also on functional aspects? From the studies to date it is difcult to decide whether music and nonhuman vocalization have a similar origin or, rather, represent behavior developed in parallel with similar underlying selective factors. In summary, I believe the authors in this book could have formulated more specic hypotheses to be tested in future research. However, Masataka was successful in bringing together researchers using very different approaches for an empirically focused and comparative approach to language and its origins.

123