Vocal Development as a Guide to Modeling the Evolution of Language

Abstract

Modeling of evolution and development of language has principally utilized mature units of spoken language, phonemes and words, as both targets and inputs. This approach cannot address the earliest phases of development because young infants are unable to produce such language features. We argue that units of early vocal development-protophones and their primitive illocutionary/perlocutionary forces-should be targeted in evolutionary modeling because they suggest likely units of hominin vocalization/communication shortly after the split from the chimpanzee/bonobo lineage, and because early development of spontaneous vocal capability is a logically necessary step toward vocal language, a root capability without which other crucial steps toward vocal language capability are impossible. Modeling of language evolution/development must account for dynamic change in early communicative units of form/function across time. We argue for interactive contributions of sender/infants and receiver/caregivers in a feedback loop involving both development and evolution and propose to begin computational modeling at the hominin break from the primate communicative background.

Keywords: Computational modeling; Illocution; Language evolution; Parent-infant interaction; Perlocution; Spontaneous vocalization; Vocal development.

Fig. 1. Schematic fragment of communicative natural logic: Tree of vocal development/evolution

The infrastructural tree shows vocal capabilities in order of emergence in human development from bottom up, infraphonological/signal capabilities on the left, infrasemiotic/function capabilities on the right. At birth, Spontaneous Vocalization in modern humans includes the endogenous tendency to produce phonation, which we propose occurred in the human line at the beginning of differentiation from non-human primate vocal communication. On the basis of Spontaneous Vocalization, additional capabilities emerge by 3 months: (a) Vocal Type Expansion (resulting in protophones such as vowel-like sounds, squeals, and growls) appears to result from self-organization through phonatory exploration, (b) Functional Flexibility, where each protophone is used with a full range of illocutionary/perlocutionary functions, expressing infant positivity, neutrality or negativity, and eliciting corresponding caregiver responses (Oller et al. 2013), and (c) Face-to-Face Vocal Interaction, where infants appear to bond with parents in conversation. The figure illustrates a hypothesis about how language foundations do and presumably must develop and evolve. Such capabilities have been called “preadaptations” for language (Christiansen & Kirby, 2003). See SM2 for further explanation and for description of subsequent stages leading to spoken language.

Fig. 2. Computational modeling of interactive feedback in communicative development/evolution

A: In our proposed computational modeling of communication, a Signal produced (and sensed) by a Sender/Infant, transmits an Illocutionary Function recognized by a Receiver/Caregiver, whose response is a Perlocutionary Effect, which upon reception by the Sender serves as a selection force on the Sender/infant’s capability/inclination to produce the Signal/Function (terms adapted from Austin [1962], see SM1). Learning by both Sender and Receiver can occur at various points in the computationally modeled feedback loop, which can be tested for outcomes and initial settings on 1) infant volubility, 2) infant self-monitoring in signal production, 3) infant learning sensitivity to parental responses, 4) parent responsivity to infant signals, 5) parent adjustment of responsivity based on infant capabilities previously exhibited, and 6) parent self-monitoring of the effects of their own responsivity on infant signaling. Correlations between these outcomes/settings and reproductive success can also be tested and assessed in reference to the extensive literature in vocal and interaction development (see section SM4). B: Feedback will operate in the computational models in both intra- and inter-generational selection, in the former case selecting for vocal development in individual Senders (S₁-S_n) under the influence of Receivers (R₁-R_n) within generations, and in the latter case selecting Senders of the species in vocal evolution across generations. Senders of the first generation (G1) will be differentially selected to reproduce in G2, such that progeny of more vocally capable Senders from G1 will be overrepresented in G2. Additionally, more vocally capable Senders from G1 will also be overrepresented as influential Receivers in G2, since they will be more likely to have children in G2 than less vocally capable Senders from G1.