The “Contact Matrix” as a Model for Understanding Logographic Script Borrowing and a Tool for Decipherment
Because most graphs in a logographic script represent spoken morphemes, each graph is tightly associated with a pair of attributes: pronunciation and meaning. In a contact situation, speakers of another language who learn the script will associate each graph with a filtered pair of attributes: a localized pronunciation and a translated meaning. These associations form what I call a “contact matrix”, which is the basis for semantically-adapted and phonetically-adapted applications of the graphs to write linguistic units of the second language. The contact matrix is therefore a useful model for understanding how logographic scripts that are used to write one language are repurposed to write other languages, as in the adaptation of Sumerian cuneiform to write Akkadian or the adaptation of Chinese characters to write Korean, Vietnamese, and Japanese. In this talk I will illustrate how the contact matrix functions, and explore how it might be used to attempt decipherment of Sinoform (Chinese-like) scripts like Khitan, Parhae, and Tuyuhun. (Decipherment results are not guaranteed!)
