Language Evolution and Computation Bibliography

Abstract: In order to understand the evolutionary pathway to the capability for language, we must first clearly understand the functional capabilities that the child brings to the task of language acquisition. Behavioral studies provide insight into infants' ability to extract ...

In this paper an experiment is presented in which two mobile robots develop a shared lexicon of which the meanings are grounded in the real world. The robots start without a lexicon nor shared meanings and play language games in which they generate new meanings and negotiate words for these meanings. The experiment tries to find the minimal conditions under which verbal communication may begin to evolve. The robots are autonomous in terms of computing and cognition, but they are otherwise far simpler than most, if not all animals. It is demonstrated that a lexicon nevertheless can be made to emerge even though there are strong limits on the size and stability of this lexicon.

The basic internal structure of a word consists of an alternation between consonants and vowels. Words tend to begin with a consonant and end with a vowel. The fundamental evolutionary status of the consonant-vowel alternation is indicated by its presence in rhythmically organized pre-linguistic vocalizations of 7 month-old babbling infants. We have argued that the basic alternation results from a mandibular cyclicity (``The Frame'') originally evolving for ingestive purposes. Here, we consider beginnings and endings of words. We conclude that preferences for consonantal beginnings and vocalic endings may be basic biomechanical consequences of the act of producing vocal episodes between resting states of the production system. Both the characteristic beginning-end asymmetry and some details of the choice of individual sounds in the non-preferred modes (vocalic beginnings and consonantal endings) are mirrored in babbling and early words. The presence of many of these properties in modern words, even though they are delivered in running speech, as well as in a proto-language corpus, indicates retention, for message purposes, of properties originally associated with the single word stage of language evolution.

The origin of phonological systems is examined from the paradigm of self-organization. We claim that phonological systems could have emerged as the product of self-organizing processes. Self-organization may have facilitated the evolution of structures within the sounds that humans were able to produce. One of the main points of the paper concerns the identification of the processes which could account for the self-organized behavior of sound systems used in languages spoken by humans. In this paradigm, phonological systems or sound patterns of human languages are emergent properties of these systems rather than properties imposed by some external influence. Regulations are defined as the constraints that adjust the rate of production of the elements of a system to the state of the system and of relevant environmental variables. The main operators of these adjustments are feedback loops. Two types of processes can be distinguished in regulatory networks, homeostatic and epigenetic. Since the origin of sound patterns, of human languages, is in the vocal tract constraints, we make the hypothesis that sound change does not reflect any adaptive character but rather is the phonetic modality of differentiation understood as epigenetic regulation.

This paper describes computer simulations that investigate the role of self-organisation in explaining the universals of human vowel systems. It has been observed that human vowel systems show remarkable regularities, and that these regularities optimise acoustic distinctiveness and are therefore adaptive for good communication. Traditionally, universals have been explained as the result of innate properties of the human language faculty, and therefore need an evolutionary explanation. In this paper it is argued that the regularities emerge as the result of self-organisation in a population and therefore need not be the result of biological evolution. \\ The hypothesis is investigated with two different computer simulations that are based on a population of agents that try to imitate each other as well as possible. Each agent can produce and perceive vowels in a human-like way and stores vowels as articulatory and acoustic prototypes. The aim of the agents is to imitate each other as well as possible. \\ It will be shown that successful repertoires of vowels emerge that show the same regularities as human vowel systems.