Junks and sampans
« previous post | next post »
These are two premodern words for Chinese watercraft that have worked their way into the English lexicon. Their etymology, however, is not as straightforward as it might seem.
"Language Matters | Where did English get the words ‘sampan’ and ‘junk’ from? Probably Cantonese and Javanese: Scholars are split on the roots of ‘sampan’ and ‘junk’, with some pointing to Chinese and others to Old Malay and Javanese respectively", by Lisa Lim, SCMP (9/30/24)
Sampans – typically small, light, wooden boats with a relatively flat bottom, propelled by a pole, oars, or a single long stern sculling oar – have a long history in East and Southeast Asian coastal and river waters.
Usually open, with a shelter aft, they were – and still are – used as a means of transporting passengers and goods over short distances; fishing; or to get to larger vessels out at sea. They also constituted homes for sea-dwelling communities, including the Tanka or Séuiseuhngyàn “people born on or of the water”, of coastal southern China and Hong Kong and Macau.
The name sampan began being used in European accounts of the China seas for such small boats. The earliest documentation in English is from a 1620 diary entry by Richard Cocks, merchant and East India Company servant, describing how “Yt was thought fytt and brought in question by the Hollanders to trym up a China sampan to goe with the fleete”.
Many accounts explain the name as deriving from Sinitic, most likely Cantonese 三板 sāam báan – or Hokkien 舢板 sam-pán – meaning “three boards”, referring to the construction of the craft’s keel-less hull which usually comprised three planks or pine boards.
However, several scholars have demonstrated an Austronesian origin. In the earliest written texts in Malay, inscriptions from the 680s, which relate the dispatching of war fleets to establish the newly founded polity of Srivijaya, a specific boat type is named, namely the sampan, in the Old Malay form sāmvau. This is believed to have served as the origin for cognate words in South, East, and Southeast Asian languages.
Although the "three boards" theory seems simple and straightforward, it comports neither with the history of their physical construction nor with the linguistics chronologically and phonologically as well as the Old Malay derivation. Consequently, the resemblance between sāmvau and samban is the result of the latter being chosen as a sound transcription with convenient semantic content, a common phenomenon in Sinitic borrowing of words from other languages.
Here are some published etymologies for "sampan":
From Sinitic, likely Cantonese 三板 (saam1 baan2) or Hokkien 舢板 (sam-pán).
a word applied by Europeans to any small, light boat on the Chinese pattern, used on the coasts of East Asia, 1610s, from Chinese san pan, literally "three boards," from san "three" + pan "plank." In 16c. Spanish made it cempan; Portuguese had it as champana.
Cantonese saam1 baan2 (akin to Mandarin sābǎn [sic]), from Middle Chinese sam pa⋮n´ : sam, three (ultimately from Proto-Sino-Tibetan *sum; akin to Tibetan gsum and Burmese sûm) + pa⋮n´, board (since the hulls of sampans were originally constructed from three planks of wood, two for the sides and one for the bottom);
(AHD 5th [retrieved 9/30/24])
It is sometimes claimed that the word "sampan" is derived from the Cantonese term sāam báan (三板), literally "three planks", but this is likely to be a false etymology. A possible Austronesian origin of the word has been suggested, as it is attested in an Old Malay inscription from 684 CE.
Now, returning to "junk" in Lisa Lim's SCMP article.
Another vessel associated with local waters – a Hong Kong icon, found in the Tourism Board’s logo – is the traditional Chinese junk, the larger, efficient, sturdy, multi-masted vessel with fully battened sails, compartmentalised hull, stabilising lee- and centreboards, and stern-mounted rudder.
Again, some accounts have suggested the English word junk comes from Sinitic 船 “ship”, specifically Southern Min chûn – or Mandarin chuán – not least because of the vessel’s prominence in the Chinese world for naval warfare and trade.
…
However, the origin of the English name again lies in pre-colonial Southeast Asia’s vibrant maritime trade, which was dominated by the Javanese from the Srivijaya empire of the 7th to 12th centuries through to the Majapahit empire of the late 13th to 16th centuries.
Their word for “ship” was jong, tracing back to a 9th century Old Javanese inscription, with the word entering Malay by the 15th century.
These large ocean-going Javanese trading jongs, which dominated the spice routes between Maluku, Java and Melaka, were what the Portuguese encountered when they arrived on the scene in the early 16th century. They described them as towering over their warships and withstanding their cannon.
…
The local word was adopted as Portuguese junco. And with Portuguese, and Portuguese creole, a lingua franca of Indian Ocean trade, many Portuguese vocabulary items entered the languages of other Europeans in the region, including Dutch jonk and English junk.
After the decline of Southeast Asian jongs in the 1700s – losing out to smaller and more agile Western ships in battle – the term junk has come to be used exclusively for the Chinese (and Japanese) junks.
And here are some published etymologies for "junk":
"large, seagoing Chinese sailing ship," 1610s, from Portuguese junco, from Malay (Austronesian) jong "ship, large boat" (13c.), probably from Javanese djong. In English 16c. as giunche, iunco.
The English word "junk" comes from Portuguese junco from Malay jong. The word originally referred to the Javanese djong, very large trading ships that the Portuguese first encountered in Southeast Asia. It later also included the smaller flat-bottomed Chinese chuán, even though the two were markedly different vessels. After the disappearance of the jong in the 17th century, the meaning of "junk" (and other similar words in European languages) came to refer exclusively to the Chinese ship.[4][5][1][6][7]
The Chinese chuán and the Southeast Asian djong are frequently confused with each other and share some characteristics, including large cargo capacities, multiple (two to three) superimposed layers of hull planks, and multiple masts and sails. However the two are readily distinguishable from each other by two major differences. The first is that Southeast Asian (Austronesian) ships are built exclusively with lugs, dowels, and fiber lashings (lashed lug), in contrast to Chinese ships which are always built with iron nails and clamps. The second is that Chinese ships since the first century AD are all built with a central rudder. In contrast, Southeast Asian ships use double lateral rudders.
The development of the sea-going Chinese chuán (the "junk" in modern usage) in the Song Dynasty (c. 960 to 1279) is believed to have been influenced by regular contacts with sea-going Southeast Asian ships (the k'un-lun po of Chinese records) in trading ports in southern China from the 1st millennium CE onward, particularly in terms of the rigging, multiple sails, and the multiple hull sheaths. However, the chuán also incorporates distinctly Chinese innovations from their indigenous river and coastal vessels (namely watertight compartments and the central rudders).[1] "Hybrid" ships (referred to as the "South China Sea tradition") integrating technologies from both the chuán and the djong also started to appear by the 15th century.
From Portuguese junco or Dutch jonk (or reinforced), from Arabic جُنْك (junk), from Malay or Javanese djong, variant of djung, from Old Javanese jong (“seagoing ship”), ultimately from either Hokkien 船 (chûn) or Teochew 船 (zung5), from Proto-Min *-džionᴬ (“ship, boat”).
It's interesting that the main, completely unrelated meaning of the English word "junk" also has an unexpected nautical origin:
From earlier meaning "old refuse from boats and ships", from Middle English junk, jounke, jonk, joynk (“an old cable or rope”, nautical term), sometimes cut into bits and used as caulking; of uncertain origin; perhaps related to join, joint, juncture. Often compared to Middle English junk, jonk, jonke, junck (“a rush; basket made of rushes”), from Old French jonc, from Latin iuncus (“rush, reed”); however, the Oxford English Dictionary finds "no evidence of connexion".
(Wiktionary — for both of the preceding two etymologies)
So the two most characteristic types of Chinese watercraft during the last millennia and more appear to be known in English and in Sinitic by words of Southeast Asian derivation.
Selected reading
- "Rivers and lakes: quackery" (6/16/23)
- "Topolect: a Four-Body Problem" (6/18/24) — discussion of different ways to propel watercraft
- "Phono-semantic rebranding" (10/25/16)
[Thanks to Mark Metcalf]
Philip Taylor said,
October 1, 2024 @ 7:07 am
Onions (1966) gives only Chinese san pan, f. san three, pan board.
Pamela said,
October 1, 2024 @ 10:06 am
As a non-linguist I'm confounded by the -vao to -ban/pan transformation, unless I imagine some kind of filtering between Portuguese and Spanish, something that would make an -ao ending in Portuguese into Spanish -an. Is there anything anywhere that shows something like that?
Chris Button said,
October 1, 2024 @ 11:59 am
@ Pamela
Yes, -an to -ão would be very Spanish to Portuguese in style and so eminently plausible. The vocalic nasalization could then be lost to give -ao.
But the other way round? Hmmm…. I think we need some more context.
Chris Button said,
October 1, 2024 @ 12:03 pm
Incidentally, as an example of the loss of nasalization of the vowel, Sino-Japanese borrowings of Middle Chinese words with -ŋ would be a good one.
Victor Mair said,
October 1, 2024 @ 5:51 pm
From Geoff Wade:
Pierre-Yves Manguin sees 'sampan' as deriving from Old Malay sāmvau
https://journals.openedition.org/archipel/2905?lang=en
'Both the Kedukan Bukit and the Kota Kapur inscriptions tell us about war fleets sent to establish the authority of the newly founded polity, with the first appearance of the name of a specific boat type: the term sampan, under the Old Malay form sāmvau, that will be at the origin of so many cognates all along Asia-wide trade routes (Cœdès 1930, pp. 34-35, 47-49).
He discusses such terms in his:
—, 2002, « The amorphous nature of coastal polities in Insular Southeast Asia: Restricted centres, extended peripheries », Moussons, 5, pp. 73-99.
—, 2012, “From lancaran to ghurab and ghali: Mediterranean impact on war vessels in Early Modern Southeast Asia”, in Geoff Wade and Li Tana (eds.), Anthony Reid and the Study of the Southeast Asian Past. Singapore: Nalanda Sriwijaya Centre, Institute of Southeast Asian Studies, pp. 146-182.
This is another useful text from Manguin:
Pierre-Yves Manguin "The Southeast Asian Ship: An Historical Approach". Journal of Southeast Asian Studies. 11 (2): 266–276
Much cited much in: Junk (ship) – Wikipedia.
Chris Button said,
October 1, 2024 @ 7:21 pm
And thanks to Geoff Wade, I think we have the answer to Pamela's conundrum.
In the article by Manguin, there is a link to Cœdès, who includes in a footnote on p.36 that Juynboll "donne la forme samboṅ, mais il cite sambo comme la graphie donnée par l'édition du Rāmāyaṇa, et rapproche le mot du malgache sambu.
Given that 舟 (which warrants its uvular -ʁ coda to account for the -k coda in its source) and 船 (whose variants 舩 and 舡 seem to account for the -ŋ coda in its source) are loans into Chinese, it follows that sampan is a likely originally a loan into Chinese too.
AG said,
October 2, 2024 @ 5:42 am
By chance I just went down a rabbit hole last week when a student asked me what a "Tankadere" was. He was reading "Around the World in 80 Days" and it's the name of a boat they take from Hong Kong. I told the student I thought it might be a Japanese word, or maybe a Japanese-sounding word Verne had made up – but when I found the Japanese version of the book, it was "Tankadia" in katakana. I looked up the French original, where it's "Tankadère", which French wiktionary says is a "Femme qui manœuvre un tanka (bateau chinois)". The Tanka boat people, mentioned above, must have given their name to a type of boat in French but not in English.
Victor Mair said,
October 2, 2024 @ 7:48 am
@AG
Eugene Newton Anderson, food historian extraordinaire and professor of anthropology emeritus at UC Riverside, is an authority on the tanka. He is a regular contributor and commenter on Language Log.
https://en.wikipedia.org/wiki/E._N._Anderson
Website: http://www.krazykioti.com/
Guy Emerson said,
October 7, 2024 @ 10:27 am
Both "junk" and "sampan" are discussed at length by Hoogervorst (2012; link below), including terms in more languages than mentioned above, like Swahili "sambu" and Malagasy "sambo". The various words are displayed on a map, in Figure 42 for "junk" and Figure 43 for "sampan"/"sāmvau"/"sambook". However, anyone looking for a simple answer about phonological correspondences will be disappointed to read that "the exact etymology… remains obscure".
https://ora.ox.ac.uk/objects/uuid:b8b47816-7184-42ab-958e-026bc3431ea3
Chris Button said,
October 7, 2024 @ 8:56 pm
@ Guy Emerson
Thanks for sharing!
It's interesting that Burmese has သမ္ဗန် sampan and သင်္ဘော saṅphaw