Boat people
« previous post | next post »
"The endangered Tanka language in Hong Kong: phonological variations and lexical convergence with Cantonese", Cong Wang, Daxingwang Peng, Yanmei Dai & Chong Qi, Humanities and Social Sciences Communications volume 12, Article number: 1133 (July 19, 2025)
The first thing we need to take care of is to discuss their name:
According to official Liu Zongyuan (773–819) of the Tang dynasty, there were Boat Dweller people settled in the boats of today's Guangdong Province and Guangxi Zhuang Autonomous Region.
The term "Tanka" (蜑家) may originate from tan (Cantonese: "egg") and ka (Cantonese: "family" or "people"), although another possible etymology is tank ("junk" or "large boat") rather than tan. "Tanka" is now considered derogatory and no longer in common usage. The Boat Dwellers are now referred to in China as "people on/above water" (Chinese: 水上人; pinyin: shuǐshàng rén; Cantonese Yale: Séuiseuhngyàn), or "people of the southern sea" (Chinese: 南海人; Cantonese Yale: Nàamhóiyàn). No standardised English translation of this term exists. "Boat People" is a commonly used translation, although it may be confused with the similar term for Vietnamese refugees in Hong Kong. "Boat Dwellers" was proposed by Dr. Lee Ho Yin of The University of Hong Kong in 1999, and it has been adopted by the Hong Kong Museum of History for its exhibition.
Both the Boat Dwellers and the Cantonese speak Cantonese. However, Boat Dwellers living in Fujian speak Min Chinese.
Abstract
The Tanka people, colloquially known as the “boat people”, are primarily located in Southeast China. Over time, their transition from a maritime to a land-based lifestyle has significantly influenced the languages they speak, particularly through contact with dominant languages such as Cantonese and Hakka. Despite this, the Tanka language in Hong Kong remains under-researched and is now endangered, with its use rapidly declining. To fill this research gap, this study examined the linguistic interactions between the Tanka language and Cantonese in Hong Kong, focusing on phonological and lexical aspects. The study yielded several key findings. First, a comparative analysis of the phonological systems of the Tanka language and Cantonese revealed striking differences in their rhyme patterns and tone evolution from Middle Chinese. These findings underscore significant phonological distinctions between the two languages. However, despite these differences, Tanka and Cantonese speakers can generally comprehend each other due to a shared linguistic structure. Second, a lexical comparison of the Tanka language, Cantonese, and Mandarin demonstrated a considerable connection between the Tanka language and Cantonese, especially among the younger speakers. This generational influence highlights the impact of Cantonese on the Tanka language. In contrast, the Tanka vocabulary diverged significantly from Mandarin, reflecting a unique linguistic development influenced by its historical and social context, while showing greater convergence with Cantonese due to prolonged contact and shared regional influences. Understanding the interplay between the Tanka language and Cantonese provides valuable insights into the origins of the Tanka people. It also contributes to a broader understanding of the typological characteristics of Sinitic languages in Southeast China.
Judging from Fig. 2, which shows the "Population of Tanka people over time (1961–2021)", there has been a steep decline in the number of Tanka from nearly 140,000 at the beginning of this period to 5,000 at the end, though levelling out during the last 20 years.
To give an idea of what the language looks and sounds like, here are a couple of colorful expressions:
打山口 [ta35 saŋ43 hɐu35] “fishing near a recognizable mountain without using a buoy to prevent theft”
捕風尾魚 [pɔʔ32 foŋ43 mei24 ji21] “fishing after a strong wind and catching fish in the wind’s wakes”
A distinctive feature of the language is that it has a rich amount of repetitive affixes that deepen the degree of a trait or add additional meanings and emotional color to adjectives. For example:
青□□ [ts’ieŋ43 koŋ24 koŋ21] “vibrantly green”
紅□□ [hoŋ21 pɔʔ4 pɔʔ4] “blushing red”
乌□□ [wu44 tsuiʔ4 tsuiʔ4] “very dark”
眼光光 [an24 kɔŋ44 kɔŋ43] “glowing-eyed, very fresh”
眼濛濛 [an24 moŋ44 moŋ44] “lifeless-eyed”
木□□ [muʔ4 k’a21 k’a21] “stale”
濛□□ [mun32 ts’a21 ts’a21] “dull-looking”
Other topolects have such affixes, though perhaps few have so many, especially for which no characters are known, i.e., they are simply sounds.
Theoretical framework
This study anchors its theoretical foundation in language contact theory (Thomason and Kaufman, 1988) as the primary framework for explaining phonological and lexical changes in the endangered Tanka language under sustained Cantonese influence. Building upon this core premise that prolonged language interaction drives bidirectional change, the framework strategically layers complementary perspectives to address multidimensional research needs.
It has been my honor to know personally the two leading American authorities on the Boat People, Eugene Anderson and John McCoy, both of whom are listed in the references to this paper, respectively (1970) and (1965)
Selected readings
- "San Francisco Cantonese" (4/2/24)
- "Dragon Boat Festival and moral abduction" (6/6/22)
- "Rivers and lakes: quackery" (6/16/23) — NOT the Tanka, who are hardworking and honorable
- "Roll out of here like an egg, Xi" (4/9/20) — see the references for the negative semantics of eggs, but also good and indifferent
- Sarah Grey Thomason and Terrence Kaufman, Language Contact, Creolization, and Genetic Linguistics (Berkeley and Los Angeles: University of California Press, 1988).
[Thanks to Ted McClure]
wgj said,
August 4, 2025 @ 10:19 pm
I've met Tanka people of various ages at several bays in Hainan, including Puqian, Tanmen, Sanya, Yinggehai, and Dongfang. All of them proudly introduced themselves as Tanka to me, without me asking or otherwise mentioning the term (there were probably more Tanka that I met who chose not to volunteer that information). None of them called themselves "people on water".
So it seems that at least in this respect they're culturally different from the Tanka in Hong Kong, but this isn't too surprising. I've asked them how different they are from other groups of Tanka, and the consensus are always, pretty different even within Hainan, and very different to others in the South China Sea. Their are significant disagreement on what, if any relations they have to the Bajau (insomuch as they know about the Bajau at all).
It is true that most of those people I met worked in fishing-related jobs, i. e. fisher, seafood trader, restauranteur etc., so they had some incentive to boast their professional credentials by mentioning their ethnicity. On the other hand, it's questionable how effective such a hidden boast would be, since most people have never heard of the Tanka. My standard answer to their mentioning of their Tanka-ness is, "So are you born on a boat?", which invariably excites them (as they recognize that I know a bit about the Tanka) and starts a conversation on the rapid changes in Tanka lifestyle of their particular community.
AntC said,
August 4, 2025 @ 11:12 pm
I well remember the flotilla of boats tied up in Aberdeen Harbour. It looked like many of the boats were not fit to put to sea. What was weird was to see smartly-dressed office workers stream out of them to go work in Central.
The boats were still there in ~2017 when I visited briefly — but looking very sad. I suspect that the PLA jackbooting their way into HK will have further suppressed the culture.
wgj said,
August 5, 2025 @ 12:49 am
@AntC: It's an interesting contrast between Tanka who no longer live on boats (AFAIK, all Tanka in Hainan have moved onshore, although many a literal stone-throw away from the water's edge) but still live off the sea (working fishing-related jobs), and those vice versa. Which ones are holding up the cultural tradition more? In the long term it's clearly the former because the latter will be gone completely in another generation.
Philip Taylor said,
August 5, 2025 @ 3:43 am
In British English, a boat person is a refugee who (like my wife's aunts, uncles, etc.) fled Vietnam by boat and ship following the end of the Vietnam War in 1975. Thank God that, at that time at least (and also during the period of the Kindertransport, of course), Britain was more welcoming to refugees than it is today.
DDeden said,
August 5, 2025 @ 8:30 am
Just to mention the Moken, a boat dwelling people of the Andaman Sea, Austronesian speakers I believe.
Yq said,
August 5, 2025 @ 1:05 pm
What phonological (?) rule changes the tone in the reduplicated kɔŋ⁴⁴ kɔŋ⁴³ and koŋ²⁴ koŋ²¹ but not in moŋ⁴⁴ moŋ⁴⁴ and the rest?
Jonathan Smith said,
August 5, 2025 @ 3:24 pm
Re: Yq's question I am also curious but maybe there's not enough info here (or anywhere?) to say… some Min languages also have tons of such forms, e.g. under âng 'red' we find in Embree (1973) âng-chu-chu, âng-kì-kì, âng-kòng-kòng, âng-phà-phà, with the penultimate tones indeed phonetically different from ultimates due to "sandhi" — but no idea if something similar might be at work in Tanka Yue.
Incidentally the authors say Tanka is clearly not Min which is true enough — but there are a few striking lexical connections which could represent Min substrate or shared indigenous substrate. Also obvious but somehow still striking is lexical lists where Standard Cantonese and Standard Mandarin again and again have parallel items but Tanka is different… reminding us the degree to which regional standards everywhere are homogenized/(inter)nationalized relative esp. to rural patois.
Jongseong Park said,
August 5, 2025 @ 7:35 pm
Speaking of the Boat Dwellers/Tanka, the name Hong Kong as used in English may come from the Tanka pronunciation. After all, the name originally refered to Aberdeen on the south side of Hong Kong Island, today called 香港仔 Hoeng1gong2zai2 or literally 'little Hong Kong' in Cantonese, which was known for its Tanka floating village.
Standard Cantonese 香港 Hoeng1gong2 would normally be rendered as Heung Kong in Hong Kong-style romanization. But it seems that Standard Cantonese /œ/ is pronounced as /ɔ/ in Tanka, which accounts for the form Hong Kong (assuming that the characters 香港 indeed reflect the true etymology of the name).
john@tkacik.com said,
August 6, 2025 @ 3:39 pm
Robert van Gulik's "Judge Dee" introduced the outside world to Kwang-tung's "tanka" in 1967
https://www.nytimes.com/1967/08/20/archives/criminals-at-large.html
" … Robert van Gulik turns to Judge Dee's late years, as President of the Metropolitan Court of Imperial China in 680 A.D., for MURDER IN CANTON (Scribner's, $3.95). The main theme (and a fascinating one) is the early contacts between China and the Moslems of Arabia; the plot also involves crickets and the gypsy-like waterfolk, the Tanka, a missing Imperial Censor and a charming and spirited pair of Persian twins. Not one of Dee's strongest puzzles, but the trimmings are pure delight…"
John Tkacik said,
August 6, 2025 @ 3:45 pm
Robert van Gulik's "Judge Dee" introduced the outside world to Kwang-tung's "tanka" in 1967
https://www.nytimes.com/1967/08/20/archives/criminals-at-large.html
" … Robert van Gulik turns to Judge Dee's late years, as President of the Metropolitan Court of Imperial China in 680 A.D., for MURDER IN CANTON (Scribner's, $3.95). The main theme (and a fascinating one) is the early contacts between China and the Moslems of Arabia; the plot also involves crickets and the gypsy-like waterfolk, the Tanka, a missing Imperial Censor and a charming and spirited pair of Persian twins. Not one of Dee's strongest puzzles, but the trimmings are pure delight…"
Chris Button said,
August 6, 2025 @ 4:35 pm
I'd have thought some long-ago corruption of 船家 would be more likely.
wgj said,
August 7, 2025 @ 11:18 am
@Chris Button: Except in both Cantonese, Hakka, and Min, the initial consonant for 船 is a variant of s or ts, which you can see in these exhaustive lists that include all subdialects of the three:
http://www.zdic.net/zd/yy/yy/%E8%88%B9
http://www.zdic.net/zd/yy/ky/%E8%88%B9
http://www.zdic.net/zd/yy/my/%E8%88%B9
Neither Middle Old Chinese nor Late Old Chinese is believed to have pronounced it with a t-sound either.
Chris Button said,
August 7, 2025 @ 4:33 pm
But 蜑 and 船 both ultimately go back to lateral onsets of sorts. And loaning was rife among boat-related words: 舟, 船, 三板 all show connections beyond Sino-Tibetan.
I haven't looked into the history of 蜑家. A corruption of 船家 might not be the origin, but it seems far more linguistically reasonable to me that the other suggestions.
Chris Button said,
August 7, 2025 @ 4:40 pm
Perhaps an obstruent stopping the vocal chord vibration versus a sonorant allowing it to continue?