Wang Dayuan’s Account of Banda
Prior to the British occupation of the Dutch East Indies during the Napoleonic Wars, when the plants were transplanted to other British colonies in Asia and the Caribbean, true nutmeg trees (Myristica fragrans) only grew on the islands of a tiny archipelago in eastern Indonesia, just south of Seram and a little to the west of New Guinea, known as Banda. The Banda Islands are truly small — the city of The Hague, just down the road from where I’m writing this, is about twice as big in area as all the islands put together — but in the Middle Ages Banda produced all the nutmeg and mace in the world. This made it a noteworthy place indeed, and Banda is mentioned by name in medieval texts in Javanese, Malay, Sundanese, Latin, Venetian, and Chinese, and alluded to in texts written in many other languages.
In this post I will look at one of the earliest descriptions of Banda in any language: the account in the 島夷誌略 (Dǎoyí Zhìlüè ‘Summary Record of the Island Foreigners’) written by Wāng Dàyuān (汪大淵), a Chinese man from Quanzhou who travelled around much of Asia in the early fourteenth century, when China was ruled by the Yuan (Mongol) dynasty. The text was finished in around 1339 CE. As with my earlier post on Zhao Rukuo’s description of Sunda (West Java), I have relied on the Wikisource text of Wang’s account, by far the most accessible version (but perhaps not the most dependable).
I’ve put the text below. It is followed by my tentative translation and copious notes commenting on some of the more interesting aspects of Wang’s description.
‘A tall ring of swollen mountains — the creek water seems fresh but the fields are barren. The people mostly eat sago (3) and coconut. The climate is incessantly hot. Their customs are depraved. Men and women wear their hair in topknots and their bodies exposed with [only] natural bark cloth tied around them. During the day they fear the heat and don’t do any planting or sowing. On moonlit evenings they plough and hoe, go fishing and hunting, collect firewood, and fetch water. There’s no danger of snakes and tigers in the mountains and there’s no risk of robbers at home. They boil seawater to make salt and brew coconut milk into wine (4). Women work at weaving cotton (5). They have headmen (6). The land produces nutmeg (7), black servants (8), mace (9), and clove bark (10). [You should] trade goods like drinking water, damask twill cloth, floral print cloth (11), black jars (12), drums and zithers, and qing [or: green] porcelain (13).’
(1) This translation is a first draft and comes with the caveat that I’m not a bona fide Sinologist. The account here is extremely terse — subjects of sentences are often lacking and there are few conjunctions. I have tried to make it a tiny bit more readable in translation.
(2) WDY’s 文誕 is reportedly the same as the name for the islands in the 大德南海志 (Dàdé nánhǎi zhì — ‘South Sea Records of the Dade Era’), which was written before 1307 CE. This is thus the name under which Banda enters history. The islands were completely undocumented until the fourteenth century, after which they appear sporadically in texts from across Afro-Eurasia. The characters 文誕 are pronounced wéndàn in modern Mandarin and /mɨundɑnX/ in Middle Chinese.
The variety of Chinese spoken at the time WDY was travelling, however, is probably best represented by so-called ’Phags-pa Chinese — that is to say, Chinese as written in the ’Phags-pa script, an alphabet designed on the orders of Khubilai Khan in the thirteenth century. It was based in part on the Tibetan alphabet, although it was written vertically and with unique square-shaped letters; its initial purpose was as a dedicated writing system for recording the Mongol language. Over the course of the Yuan dynasty it came to be used to write Persian, Tibetan, and Chinese (among others). A manual for writing Chinese in ’Phags-pa survives and has been transcribed and published by William South Coblin (2006). A number of inscriptions are also extant. This alphabetic data gives us some insights into the pronunciation of one semi-official dialect of Chinese at the beginning of the fourteenth century, when WDY was travelling. And in this ’Phags-pa Chinese, 文誕 was written <wuntan> and would (according to Coblin’s notes) have been pronounced [ʋundan] (Figure 1).
The name for Banda in all the earliest surviving accounts ends in [n]: Chinese <文誕> [ʋundan]; Old Javanese <wwaṇḍan> (Deśawarṇana 14.5, 1365 CE); Old Sundanese <ba(n)dan> (Bujangga Manik, fifteenth century CE); Classical Malay <بندن> [bandan] (Hikayat Raja Pasai, date difficult but attributed to late fourteenth or early fifteenth century CE); Latin <Bandan> (De Uarietate Fortunæ, 1448 CE); and Venetian <bandã> (the Fra Mauro map, c.1450 CE) (Figure 2). This is significant because it is often said that the name for the islands comes from Persian bandar (بندر) ‘harbour, port’; it is claimed that the name was bestowed on the archipelago because Banda Neira in particular was an important entrepôt at which goods from throughout the region gathered. The entrepôt thing may well be true, but the [-n] in all the earliest mentions of the name makes the Persian derivation a little implausible.
(3) 沙糊 (pinyin: shāhú; Middle Chinese: srae hu) clearly refers to sago (Malay: sagu), the edible starchy pith of several trees in Southeast Asia. The most common tree exploited for sago production is (unsurprisingly) the sago palm, Metroxylon sagu, but others can be used as well, including Arenga pinnata, an incredibly useful and fairly common species throughout the region. Sago can be eaten in many ways: rolled into tiny balls (‘sago pearls’) for boiling; pressed and partially dried out into a larger ‘bread’, as described by Marco Polo and Odoric of Pordenone in Sumatra around the same time as WDY’s account; or prepared with water as a gluey gruel (papeda). WDY is probably referring to the latter, both because that’s a common form of the stuff in Maluku and New Guinea and because the characters used suggest a gruel or porridge: the second character (and its variant form, 餬) has the meaning ‘gruel, made with barley or millet; congee’ (Kroll 2017:165 sub 餬). Taken literally the word 沙糊 means ‘sand gruel’, which sounds a little unpleasant. The characters appear to be an attempt at transliterating the Malay word sagu, so this could well be a culinary pun.
(4) Coconut milk is not, in fact, used to make alcoholic beverages (as far as I know). Palm wine is made from other sugary substances produced by palm trees, usually tapped from inflorescences or, in the case of coconut, the water inside the fruit.
(5) In Indonesia, and more widely in the Austronesian-speaking world, there appears to have been a consistent divide between men’s and women’s work. Men were expected to fight and take heads in war; women were expected to weave cotton cloth.
(6) Chiefs or village headmen as opposed to kings. WDY is tersely telling us what sort of societies the Bandanese live in.
(7) Literally ‘flesh cardamom’.
(8) I don’t know of another way to interpret this, and it doesn’t appear to have any possible botanical meaning — 黑 means ‘black’ and 小廝 means ‘(lowly) servant’. It must be a reference to the trade in enslaved human beings, who seem to have been taken captive in New Guinea (going by the evidence of later times — also cf. Old Javanese boṇḍan [OJED 251:13], apparently a reference to Papuan slaves).
(9) Literally ‘nutmeg flowers’ (or ‘cardamom flowers’ — see note 4). This was a common formula meaning ‘mace’ (i.e. the aril surrounding the nutmeg, which is the seed of the tree) in languages across medieval Afro-Eurasia, as in (among others) Early New High German.
(10) 丁皮 means ‘clove bark’, or more literally ‘nail bark’. (Cloves are referred to in Chinese as ‘nail fragrance’, as etymologically in English.) In truth the bark in question doesn’t come from clove trees (Syzygium aromaticum) but rather from Cinnamomum culitlawan, a tree from central Maluku and Banda whose Malay name is kulit lawang ‘clove bark’, whence the Latin binomial and presumably the Chinese name of the commodity. (Thanks to Wen-Yi Huang and Lars Christensen for their suggestions here.)
(11) Garrett Kam has pointed out that the cloths here are probably Indian chintzes, not (as I originally wrote) patolas, which are double ikats.
(12) Is this a reference to Cizhou ware? It would be interesting to know if any pottery of that sort has been found at archaeological sites in Banda.
(13) This is certainly a kind of porcelain, 靑 ‘natural/blue/green’ in colour. Presumably it was the same as the kind we now know as qingbai ware, although Wen-Yi Huang suggests it may have been celadon or greenware instead, which would fit better with the colour term WDY uses.
Wang Dayuan’s account is one of the first pieces of direct evidence in a long line of outside engagements with Banda, engagements which reached a head in 1621 with the genocide of the islands’ population by the Dutch under Jan Pieterszoon Coen. The islands have a rather tragic history about which plenty of books have now been written, including the well-known account of struggles between the English and Dutch in early colonial Banda, Giles Milton’s Nathaniel’s Nutmeg (1999). More of the Banda Islands’ (considerably more interesting) pre-colonial/medieval history is coming to light these days, thanks in particular to recent archaeological work in the islands by Peter Lape and others (see e.g. this (unfortunately not-open-access article on Neolithic Banda and the earliest known culinary use of nutmeg).
Don’t miss the follow-up to this post: How Big are the Banda Islands?
Coblin, William S. 2006. A handbook of ’Phags-pa Chinese. Honolulu: University of Hawai’i Press.
Kroll, Paul W. 2017. A student’s dictionary of classical and medieval Chinese. Leiden: Brill.
Lape et al. 2018. New data from an open Neolithic site in eastern Indonesia. Asian perspectives. 57(2):222–243.
Milton, Giles. 1999. Nathaniel’s nutmeg. London: Hodder and Stoughton.
YOU MAY ALSO LIKE
I have a Ko-Fi account, in case you’re interested in showing your appreciation for this and other Medium stories: https://ko-fi.com/P5P6HTBI