Tao Gong

龔 濤

Alexander von Humboldt researcher

PhD @ The Chinese University of Hong Kong
M.A.& B.A. @ Tianjin University

Max Planck Institute for Evolutionary Anthropology
Deutscher Platz 6
04103 Leipzig

Tel.: +49 (341) 3550 - 319
Fax: +49 (341) 3550 - 333

  Email: tao_gongeva.mpg.de

Office: U 1.31

Research Projects

Evolutionary linguistics - Language Games - Complex Networks

My research covers both evolutionary linguistics and computational simulation. I am interested in some central questions concerning language evolution, including whether syntactic abilities gradually evolved from domain-general abilities, how different linguistic components such as semantics, lexicon, and syntax interacted with each other during language evolution, and what were the roles played by cultural transmission or social structure during language evolution. This research, with a multidisciplinary nature, can provide insights on both human language and its evolution, and assist the research on artificial intelligence and evolutionary computation. This research direction has gradually become a beacon for modern scientific research.

One aspect of my current work, cooperated with Language Engineering Laboratory in The Chinese University of Hong Kong, is to take a "bottom-up" approach to explore the collective effect of individual cognitive and communicative behaviors on the evolution of linguistic components such as semantics and syntax. My focus lies on the relation between the emergence of simple syntax and that of lexical items. Following Emergentism that language is triggered by some domain-general abilities shared among humans and other species. I have developed a multi-agent model to trace a coevolution of compositionality (in the form of lexical items) and regularity (in the form of constitute word orders). This model suggests that the primitive syntax can be adapted from domain-general abilities and coevolve with the emergence of lexical items. It provides an alternative way of viewing language evolution, different from Innatism that emphasizes the innateness of human language. This model is under modification to study other essential topics in linguistics, such as construction grammar and online semantic and syntactic processing. In addition, I have explored the relation between language use and social structure. By manipulating individuals' popularities to participate in communications, my model can trace language emergence under different social structures, and discuss the relation between linguistic communications and the development of social structure. Based on this model, more complex networks and related theories will be adopted to further explore the social structure effects on language evolution and vice versa.

Another aspect of my current work, cooperated with PIL Group in Physics Department, "La Sapienza" University of Rome, is to use the language game approach to study the conventionalization of various kinds of linguistic knowledge. The simulated language games include the naming game that studies the emergence of coherence, and the category game that explores the coevolution of semantic categories and their word labels. Using the category game, I have discussed the relation between physiological and cultural factors on categorization, and their collective roles on the emergence of common categorical knowledge, such as color terms. Based on statistical physics, I have further explored the effect of cultural transmissions on the emergence and maintenance of linguistic/cognitive categories. The language game approach serves as an efficient method to tackle problems concerning language evolution.

Curriculum Vitae
Publications
Research activities