Semantic Transliteration of Personal Names

Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics, pages 120–127, Prague, Czech Republic, June 2007. c⃝2007 Association for Computational Linguistics Abstract Words of foreign origin are referred to as borrowed words or loanwords. A loanword is usually imported to Chinese by phonetic transliteration if a translation is not easily available. Semantic transliteration is seen as a good tradition in introducing foreign words to Chinese. Not only does it preserve how a word sounds in the source language, it also carries forward the word’s original semantic attributes. This paper attempts to automate the semantic transliteration process for the first time. We conduct an inquiry into the feasibility of semantic transliteration and propose a probabilistic model for transliterating personal names in Latin script into Chinese. The results show that semantic transliteration substantially and consistently improves accuracy over phonetic transliteration in all the experiments. 1 Introduction The study of Chinese transliteration dates back to the seventh century when Buddhist scriptures were translated into Chinese. The earliest bit of Chinese translation theory related to transliteration may be the principle of “Names should follow their bearers, while things should follow Chinese.” In other words, names should be transliterated, while things should be translated according to their meanings. The same theory still holds today. Transliteration has been practiced in several ways, including phonetic transliteration and phonetic-semantic transliteration. By phonetic transliteration, we mean rewriting a foreign word in native grapheme such that its original pronunciation is preserved. For example, London becomes 伦敦 /Lun-Dun/1 which does not carry any clear connotations. Phonetic transliteration represents the common practice in transliteration. Phonetic-semantic transliteration, hereafter referred to as semantic transliteration for short, is an advanced translation technique that is considered as a recommended translation practice for centuries. It translates a foreign word by preserving both its original pronunciation and meaning. For example, Xu Guangqi 2 translated geo- in geometry into Chinese as 几何 /Ji-He/, which carries the pronunciation of geo- and expresses the meaning of “a science concerned with measuring the earth”. Many of the loanwords exist in today’s Chinese through semantic transliteration, which has been well received (Hu and Xu, 2003; Hu, 2004) by the people because of many advantages. Here we just name a few. (1) It brings in not only the sound, but also the meaning that fills in the semantic blank left by phonetic transliteration. This also reminds people that it is a loanword and avoids misleading; (2) It provides etymological clues that make it easy to trace back to the root of the words. For example, a transliterated Japanese name will maintain its Japanese identity in its Chinese appearance; (3) It evokes desirable associations, for example, an English girl’s name is transliterated with Chinese characters that have clear feminine association, thus maintaining the gender identity. 1 Hereafter, Chinese characters are also denoted in Pinyin ro- manization system, for ease of reference. 2 Xu Quangqi (1562–1633) translated The Original Manu- script of Geometry to Chinese jointly with Matteo Ricci. Semantic Transliteration of Personal …

Content truncated to 500 words