Reviving Siraya: A Case for Language Engineering

Abstract

Siraya is a language once spoken in Southwest Taiwan, which is being revived. Some Siraya data is inconsistent, requiring strategies as to how it will be implemented. I discuss some of these strategies in support of the revival attempt. The following issues deserve attention: 1. Siraya phonology includes a schwa (ə), although it is ignored in the original orthography. The choice here is between keeping this orthography and ignoring schwa, or re-establishing schwa and changing the orthography. 2. Siraya had maintained part of the original Proto Austronesian voice system. However, this system was losing some voice oppositions and was being re-aligned when Siraya was still spoken. Two approaches are possible: keeping the original Siraya voice system, or adapting to the tendencies to change, which were strong but had not yet taken their full course. 3. Siraya had at least three dialects, two of which are particularly useful for revitalization. In order to build a lexicon for a revitalized Siraya, should the vocabulary of these dialects be combined without further ado? Or should the words from one dialect phonologically be adjusted to the other? Is there a cause for revitalizing various dialects? 4. Siraya had “anticipating sequences”, whereby a formal part (an initial consonant, a syllable, or two syllables) of the lexical verb is prefixed to the adverbial head. Anticipating sequences abound in one dialect but are absent in the other. As it is a rather complicated and irregular feature, should it be taught in modern Siraya classes? And if so, how should it be taught: in all its complexity, or in a somewhat simplified version? Or can it be ignored without causing too much structural imbalance to the grammar?