Although the innovation framework offered by Ramsey et al. is aimed at nonhuman primates, their scheme, or a modification, is needed for humans, especially in connection with vocal innovation. This behavior, fully expressed, involves the use of the voice in a new way, usually to solve some environmental problem. One can witness vocal innovation, and the emergence of social and sensorimotor precursors, in development; and it has been proposed that aspects of this capability played a role in the evolution of speech and language (Locke & Bogin Reference Locke and Bogin2006; Oller Reference Oller, Oller and Griebel2004).

Whether humans have the capacity to invent new sounds and sound sequences is hardly in question. One only need consider the universal distribution of languages, numbering in the thousands, that are spoken; and their voluminous phonological and lexical systems, which appear to exceed communicative needs. But this says nothing about the development of the control systems required to effect new vocal forms and combinations, nor does it tell us about the evolution of the capacity to produce a range of phonetic elements at will.

In the past, avian researchers were responsible for much of the animal work on vocal innovation. Marler (Reference Marler, Carey and Gelman1991) reported that in songbirds, “recasting or re-editing of components of learned models into new sequences is commonly exploited as one means for generating novelty and also for producing the very large individual repertoires that some birds possess” (p. 61). As described, these cases would evidently meet the definition of innovation set forth by Ramsey et al.

A challenge for linguistic evolutionists inheres in the fact that chimpanzees, unlike the songbirds, are frequently silent, seemingly more inclined to gesture than to vocalize (Tomasello et al. Reference Tomasello, George, Kruger, Farrar and Evans1985; Reference Tomasello, Gust and Frost1989). But two recent reports indicate that captive chimpanzees issue putatively novel or unusual signals – extended grunts and labial trills (“raspberries”) – if they want food or a tool that is in the possession of a human (Hopkins et al. Reference Hopkins, Taglialatela and Leavens2007; Russell et al. Reference Russell, Braccini, Buehler, Kachin, Schapiro and Hopkins2005). This work raises a question for Ramsey et al.: How does their scheme handle behaviors that previously occurred only in an “old” context that, because of some change, came to be emitted in a “new” context?

Clues to the functionality of vocal signals are available in other studies, indicating that chimpanzees tend to use them selectively, when the experimenter is looking away (Hostetter et al. Reference Hostetter, Cantero and Hopkins2001; Leavens et al. Reference Leavens, Hostetter, Wesley and Hopkins2004). It is also of some interest that in chimpanzees, unlike the other primates, there are significantly more pre-adult, and fewer adult, cases of innovation than would be expected by chance (Reader & Laland Reference Reader and Laland2001), though it is evidently not known, at present, whether juveniles are also more likely to innovate vocally.

Because new forms emerge in development (West-Eberhard Reference West-Eberhard2003), our attention is drawn to the role of infancy in vocal innovation. An important problem that is solved by infants is the negotiation of their own care during a time when care is likely to be interrupted (Trivers Reference Trivers1974). I have claimed elsewhere that at one or more points in evolutionary history, increased competition for care stimulated novel and more flexible use of vocal behaviors (Locke 2006; Locke & Bogin Reference Locke and Bogin2006).

But signaling involves more than signals. For repertoire change, infants would also need to observe the effect of any new vocalizations on others. Ramsey et al. say little about cases where one individual uses other individuals as a tool. The exclusion of such cases would seem to ignore the social nature of many primate groups, wall off our own exceptionally social species from other primates, and complicate the process of evolutionary theorizing.

Human mothers are disposed to repeat their infants' speech-like vocalizations (Pawlby Reference Pawlby and Schaffer1977), a practice that appears to reinforce their use (Veneziano Reference Veneziano, Smith and Locke1988). In fact, some frequently babbled patterns are taken up by family members and may diffuse through the community (Ferguson Reference Ferguson1964; Locke Reference Locke2004; 2006). One is curious to know, in this connection, if Ramsey et al., who exclude cases where individuals witnessed a behavior and its effect on a conspecific, also wish to exclude cases in which an individual's own behavior elicited reactions by others.

Vocal innovation by infants may provide parents with a free evaluation of their offspring's fitness (Locke 2006). In primates and other animals, there are associations between innovativeness, brain development, and learning potential (Lefebvre et al. Reference Lefebvre, Reader and Sol2004). It is interesting, in this regard, that human infants who produce a high rate of syllables per utterance appear more pleasant, friendly, and likeable than infants who vocalize less complexly or rhythmically (Bloom & Lo Reference Bloom and Lo1990; Bloom et al. Reference Bloom, D'Odorico and Beaumont1993).

It is paradoxical, given the robustness of vocal innovation in humans, that developmental evidence has been reported only sporadically. Observers have noted infants' use of a distinctive intonation pattern when making “requests” (Dore Reference Dore1974; Ninio et al. Reference Ninio, Snow, Pan and Rollins1994), and there is evidence that human infants – like adult chimpanzees – make grunts and labial trills before they begin to speak (Kim et al. Reference Kim, O'Connor, McLean, Robson and Chance1996; McCune et al. Reference McCune, Vihman, Roug-Hellichius, Delery and Gogate1996). For some reason, these sounds may be particularly characteristic of retarded infants and children (O'Neill & Happé Reference O'Neill and Happé2000; Smith & Oller Reference Smith and Oller1981). It has also been reported that typically developing infants use a specific class of (nasally emitted) sounds as “requests” (Goldman Reference Goldman2001).

In babbling, human infants effect a variety of closures within the vocal tract while phonating, thereby producing audible activity that, in its most canonical form, resembles consonant-vowel syllables and, therefore, speech (Oller Reference Oller2000). A more advanced form, variegated babbling, involves different points of articulation within a single sequence of syllables. This activity demonstrates, and may facilitate, what Oller (Reference Oller, Oller and Griebel2004) calls “contextual freedom,” seemingly a phonetic form of Ramsey et al.'s “behavioral flexibility.” Later in development, there is evidence for invented words by young children (Becker Reference Becker1994; Leopold Reference Leopold1949), who are also responsible for the phonological, lexical, and grammatical inventions that occur in the creation of creoles from pidgins (Bickerton Reference Bickerton1984). It is clear, additionally, that adolescents play a major role in linguistic change, particularly at the phonological level of language (Eckert Reference Eckert1999; Labov Reference Labov2001).

Ramsey et al. have provided us with a framework for the evaluation of innovation across primate species. It would be interesting to see what modifications are required for vocal innovation. Working within a properly designed system – one that values synergies between evolution and development – it may be possible to identify candidate processes in the evolution of speech and language (Locke, in press).