Transliteration Problems of Place Names

The earliest inhabitants of central Taiwan were Austronesian plain groups

(Pingpu), and many place names were naturally given by them in the beginning.

However, none of the Pingpu languages had a writing system, so it can only be recorded by other ethnic groups who had contact with them in their languages. Due to the differences of language structures and limited knowledge of the Pingpu languages, information loss cannot be avoided in the translation process. In addition, place names transliterations normally would happen more than once. Whenever a new ethnic group came they would then change these place names again according to their own naming practice. Just like peeling off the skin of an onion, the meaning of an original place name was peeled off during the process so in the end we can hardly recognize its original meaning through their modern presentations.

The transformation process of Pingpu place names has several stages:

　 Pingpu place names Æ Han Chinese place names (Hoklo or Hakka) Æ Ching

Government renaming Æ Japanese renaming Æ Mandarin Chinese renaming

Take Da-ya County (大雅鄉) of Taichung County as an example. Pazah, the first

settlers, called the place “A-ho-ba” (阿河巴), meaning “a piece of uncultivated wild land”; the following Hakka people then called it “Ba-a” (壩仔), for it was close to the

original Pingpu sounds, and “Ba-a” also meant “uncultivated land”, corresponding to

“Bo-a” (埔仔) of Hoklo. The Ching Government changed the final particle “a” (仔) into “ya” (雅) for the sake of elegancy. In 1920 the Japanese government renamed it as “Da-ya” (大雅) which has been used until now. The transformation process of

modern “Da-ya” can be illustrated as follows:

阿河巴 (A-ho-ba) Æ壩仔 (Ba-a) Æ壩雅 (Ba-ya, see Map 8) Æ大雅 (Da-ya)

Map 8 The Bao Map of Central Taiwan in Ching Dynasty (清代台灣堡圖, Da-ya) (Cited from 地名檢索系統 http://tgnis.ascc.net/placename/)

“Miao-li” (苗栗) was a village name “Miyori” of the Pingpu Taokas group, which

meant “plain”. Han immigrants transliterated it to “Mao-li” (貓裡), and the Ching government changed it into “Miao-li” (苗栗) in 1889.

Miyori Æ 貓裡 (Mao-li) Æ 苗栗 (Miao-li)

Pu-li Wine Factory is famous for its “Ai-lan White Wine” (愛蘭白酒), but very

few people know that this name “Ai-lan” is related to the Pingpu people. There was a village of Pazah called “Auran”, which meant “a place where tomatoes grow”. This village originally was located in Feng-yuan ( 豐 原 ), and it might have been transliterated by Hakka people into “Wu-niu-lan” (烏牛欄). During the time of Ching Emperor Dao-guang (道光) this village moved to the Pu-li Basin, but they still named

their new inhabitation “Wu-niu-lan”. After Taiwan was returned to China this place was renamed by the government as “Ai-lan” (愛蘭).

Auran Æ 烏牛欄 (Wu-niu-lan) Æ 愛蘭 (Ai-lan)

Another example is “Zhuo-lan” Township (卓蘭鎮) of Miao-li County. The first

settlers of this place were the Paiten Tribe of the Pingpu Pazah groups. They called it

“Tarian”, which meant “beautiful wild field”. After the Hakka people then immigrated

to this place, they transliterated “Tarian” into “Da-lan” (打蘭). Later, Hoklo people changed “Da-lan” into the similar sounding “Ta-lan”(罩蘭). Because people thought that the Han character “罩” carried a bad connotation for the local development, they took away the upper part of “网” which means “net” or “enmeshed” from “罩” and renamed it as “Zhuo-lan” (卓蘭) in 1920 during the reign of the Japanese government.

Tarian　 Æ　 打蘭(Da-lan) Æ 罩蘭(Ta-lan)　 Æ　 卓蘭(Zhuo-lan)

Fortunately, not all place names had to go through changes in every stage. Like

Miao-li, even thought it went through three transformations before its present form, it has not been changed since the Ching government in 1889. Note also that besides Han Chinese, many Pingpu village names and tribe names were also recorded in Dutch and Japanese, such as the population census table from the Dutch people in the 17^th century and the field work notes of the Japanese scholars Abe and Ino. However, except for those who can read Dutch or Japanese, ordinary people hardly have access to this data and therefore these studies have had very little impact on the lives of ordinary people. But this data has been able to preserve the original place names as they only went through one transformation, especially so for the Dutch ones for Dutch was relatively similar to Austronesain in terms of phonology and morphology. They

can thus serve as a precious tool when studying Pingpu place names.

Except for these transcribed by the Dutch and Japanese, most place names were

transliterated by Hoklo, a Han Chinese dialect in the first layer, with only a few cases where Hakka was used. This reflects the fact that within Han Chinese immigrants, the Hoklo was much larger than the Hakka even though Hakka people indeed arrived at quite a few places in central Taiwan earlier than Hoklo people.

We have mentioned earlier that “-an” is a locative suffix of quite a few

Austronesian languages. But since Han people did not understand Austronesian morphology, transliteration was much more common than paraphrasing. Thus those frequently seen place name particles “lan” (蘭), “an” (岸), “man” (蠻), “dan” (丹), and “quan” (泉) were actually transliterated from Austronesian words. If we judge

place names only from their literal meaning without this consciousness, these words ending with “-an” in Hoklo would easily be misunderstood as the general part of place names, inferring that they refer to the local product or landforms. The famous scenic spot “Alisan” (阿里山) in Jia-yi County is a typical example. “Alisan” was a place name of the Tsou (鄒族). Although it was surrounded by mountains, the area

encompassing “Alisan” was not limited to mountain areas only. Thus the mountain which is famous for its sunrise should be written as Mt. Alisan, not Mt. Ali.

Another place name in question is “Tong-siao-wan” ( 通 宵 灣 ) in Tong-siao

Township (通宵鎮) of Miao-li County (苗栗縣). According to Hong Min-lin (洪敏麟,

1983), “Tong-siao-wan” is a long strip of a village which can be sectioned into upper, central and lower parts. It is located between two mountains while the crest line down curved towards the east like an arc. Thus it was called “Tong-siao-wan” for “wan” (灣)

is often used to refer to plains which are open in one side and surrounded by mountains on the other three sides with an arc crest line (Hong, 1983). However, the

“Tun-shou” Tribe (吞宵社) of Taokas called themselves “Pariwan” or “Parrewan”.

Could “wan” (灣) in “Tong-siao-wan” (通宵灣) be the transliteration of the final

particle “-wan” instead of the description of a geographical feature? We think it is possible, just like “Taiwan” (台灣) and “Na-lu-wan”(娜魯灣) are names given to

places that have nothing to do with a real bay.

We have known that “-an” is a locative suffix in Austronesian language, but could

there be other affixes related to place names? We found there are some forms occurring time and again in the village names of Austronesian groups in the old maps of the Ching Dynasty, such as:

“A-” (阿): A-me-li (阿密哩), A-li-sai (阿里史), A-ta-bu (阿罩霧), A-sock (阿束), A-ba-quan (阿拔泉).

“Ba-” (貓): Bad-si-kan (貓兒干), Ba-bu-za (貓霧揀)

“-li” (裡): Mao-li (貓裡), Ba-li (岸裡), Wan-li (灣裡), A-me-li (阿密哩)

“-bu” (霧): Ta-li-bu (他里霧), A-ta-bu (阿罩霧), Ba-bu-za (貓霧揀)

Like the prefix “-ki” in Pingpu Katagalan place names, the relation of these affixes with place names are left for further studies.