Monday, 9 November 2015

Chinese Sources on Indonesia - Romanisation!

       Perhaps the most important language for working out a narrative history of ancient Indonesia and Malaysia, besides Malay and Javanese, is classical Chinese. It might actually be the most important overall, and the ability to read classical Chinese, or even to understand something of the intricacies of Chinese pronunciation, is certainly useful. Unfortunately, as far as I can tell, it's a skill most modern scholars of Indonesia are lacking. That's fairly obvious from the books that get written about ancient Java: it's rare to find one in which pinyin romanisation is used consistently. Actually, it's quite rare to find one in which any proper system of Chinese romanisation is used consistently. That's a problem.
       Chinese sources tell us a lot of things, often in quite a cryptic way, and it's helpful to have some grasp of the language when making sense of even basic facts about the archipelago - such as which states were where, what religions they practiced, and so on. Records of 'tribute' and emissaries sent to China tell us about the rise and fall of assorted Indonesian polities, and travellers accounts tell us about religious practice, funerary rites, food, topography, sports, you name it. They give a different angle to the native sources on most things - including, for example, the Mongol invasion of Java in 1292 - and they cover the entire period between the first and sixteenth centuries CE. They tell us about states that we wouldn't otherwise have known anything about, that are otherwise completely unknown.

      And that's why knowledge of Chinese is so useful and also so troubling and difficult. The Chinese sources give us the names of plenty of early states in Southeast Asia, but they naturally did so in Chinese, which isn't written with an alphabet or abugida but rather with the huge logosyllabic Chinese writing system. The Chinese names are often (but not always) attempts to replicate in logosyllabic characters the (presumably) native pronunciation of the names, and it is a far from simple historiographical problem to uncover the original pronunciation.

       It's difficult even today to work out some of the modern placenames of non-Chinese-speaking peoples in Chinese. Check out these placenames and imagine that the Chinese names are the only evidence we have for the originals: Oxford comes out as 牛津 Niujin, a direct calque (ox-ford); New York is 紐約 Niuyue, an attempt to replicate something of the pronunciation of the English; Las Vegas becomes 拉斯維加斯 Lasi Weijiasi, which is a whole two syllables longer than the original and renders English /v/ as [w] or [ʋ]. Imagine trying to work out the original name of New York City from that Chinese name and you have some idea of the problems facing students of the protohistory of Indonesia.

       Now imagine that we're a) often dealing with Sanskrit names b) with Southeast Asian pronunciations c) rendered into some variety of Chinese d) all from different points in time e) occasionally with different variants used in different texts and f) often with scribal errors in the existing manuscripts.

  And then imagine that modern English-speaking scholars who don't know Chinese use a variety of different romanisations (including modern Hanyu Pinyin and the antiquated Wade-Giles) to replicate the modern Chinese pronunciation of some variations of some of these attempted Chinese transcriptions.

      Some of the states and placenames make some sense. Chitu (赤土), probably on the Malay Peninsula, was a polity whose Chinese name makes no attempt to replicate the original's pronunciation. It means 'red earth', and could either refer to a local name for the state or to the soil Chinese sailors saw there. Tanah Merah, the literal Malay equivalent of the Chinese, is a common placename on the Peninsula - leaching and erosion leading to infertile laterite is quite a normal process there. Chitu makes sense as a name for either of those reasons and its identification with somewhere on the Malay Peninsula is sound. It helps, too, that the description accompanying the name is long and detailed, and that there's a (probable but unproven) Sanskrit equivalent referring to a place on the Peninsula.  But that's not normally the case.

         The fact that plenty of people writing about the ancient history of Southeast Asia continue to use antiquated and inconsistent romanisation in their scholarly works is, I think, a serious issue. It shows a lack of familiarity not only with some of the most important primary sources out there but also with the historiographic problems inherent in using them. This tendency also influences other books written by non-Indonesia-specialists. I've just been reading The Indian Ocean in World History (2013) by Edward Alpers, which, as you can imagine, has a lot to say about Java, Melaka, and so on, and the Chinese placenames used in the text are overwhelmingly in Wade-Giles. Sometimes, though, they're in pinyin. Doubling confusing. (Still a very good book, as it happens.)

        Imagine that you wanted to know about 21st century Nevada but the only text on the place that had survived the Third World War was an early-twentieth century Chinese piece about what seemed to be Las Vegas. And now imagine that everyone writing about ancient Nevada called it Laszuweichiaszu in their English-language works. That's really what we're dealing with when historians ignorant of Chinese talk about ancient Indonesia.

1 comment:

  1. Most Chinese phonetic rendering of South East Asian place name would be done using SouthEastern Chinese languages such as Hokkien or Cantonese dialect NOT Northern Mandarin pronouciation.

    For example 马六甲 in Hokkien is render as almost perfectly as Melaka. Hakka pronouciation is pretty close Ma lok ga. But Mandarin Ma liu Jia strays far from original


You can post anonymously if you really want to, but I would appreciate it if you could provide some means of identifying who you are, if only for the purpose of knowing who has written what.