Hostname: page-component-745bb68f8f-5r2nc Total loading time: 0 Render date: 2025-02-11T01:50:39.214Z Has data issue: false hasContentIssue false

Guarding the Shoreline: Oyster Farming, Salt Production, and Fishing Along the South China Coast (1667–1978)

Published online by Cambridge University Press:  03 September 2021

James L. Watson*
Affiliation:
Fairbank Professor of Chinese Society and Professor of Anthropology, Emeritus, Harvard University, USA
*
*Corresponding author. Email: jwatson@wjh.harvard.edu.
Rights & Permissions [Opens in a new window]

Abstract

This article explores the shoreline industries (oysters, salt, fish, lime) that emerged along the Laufaushan coast in Hong Kong's New Territories, in the period 1667 to 1978. The shoreline in question was controlled by a local security force, staffed by young men from a nearby lineage. The study draws on ethnographic research carried out by the author and local documents (of village and government origin) gathered on site.

Type
Research Article
Copyright
Copyright © The Author(s), 2021. Published by Cambridge University Press

South China's Pearl River Delta is one of the world's most fascinating and complex ecosystems. More of an inland sea than a classic delta,Footnote 1 it supports millions of people who live on or near its margins (Map 1). Located along the southern edge of this waterworld is a tidewater expanse of oyster beds and brackish-water marshes, known as Laufaushan, “Floating Mountain”Footnote 2—a term that captures its beauty and ever-changing topography. This article explores the relationship between landowning farmers who laid claim to this stretch of coastline and the itinerants who harvested the rich bounty of sea products along the shore.

MAP 1 Pearl River Delta and Hong Kong New Territories, 1898–1978 (Colonial domain shaded)

In many respects there are parallels between the Pearl River Delta and the ecosystems described by James Scott in his history of Southeast Asia's “anarchist” highlands.Footnote 3 Cynthia Chou's ethnography of migratory fishing specialists who inhabit the saltwater margins of Indonesia also presents important comparative perspectives.Footnote 4 Scott and Chou focus on regions where state authorities were essentially absent, except for punitive raids and episodic attempts to put down rebellions.

As numerous historians have documented,Footnote 5 China's southern coast was also an unruly frontier until the mid-twentieth century. Dian Murry's description is particularly apt:

[This region] must be seen not as a simple strip of land demarcating land and sea, but rather as a large and somewhat indeterminant region embracing a variety of settlement patterns. Just as there was an ‘inner Asian frontier’ … where sedentary agriculture … gives way to pastoral nomadism, … so here, in the south, there was a maritime frontier where sedentary settlement patterns gradually yielded to those of maritime nomadism.”Footnote 6

The coastal terrain explored in this article subsumed an array of micro-ecosystems—each exploited by occupational specialists who were, in turn, monitored by a security force maintained by landowning interests. Members of this organization also guarded the industries that emerged along the coast: oyster processing, lime smelting, salt production, brickmaking, fish preservation, and marketing. Banditry was a constant threat. Laufaushan's commercial system would not have been possible without this internal security system—on guard, twenty-four hours a day, every day of the year.

Background: Ethnographic Fieldwork Along the Hong Kong Coast

In 1898, a 365 square-mile section of land adjacent to the colony of Hong Kong was leased by the British government for a period of 99 years. Laufaushan and the villages discussed in this article were thus incorporated into what became known as the “New Territories” and were subject to colonial administration until 1997—at which point control reverted to China. Approximately 600 villages of various sizes and complexities were incorporated into this colonial domain. Fewer than 100 British (English/Scottish/Welsh/Irish) officials and police officers were in charge; they, in turn, were assisted by hundreds of local Chinese interpreter-assistants and Punjabi (Sikh and Muslim) police patrolmen. A brief and somewhat disorganized resistance to the British takeover emerged among landowning elites and their tenant supporters in early 1899, but this was quickly extinguished in what became known as the “Six Days War.”Footnote 7

The New Territories, like much of China's southern coast, were dominated by large, single-surname/single-lineage villages that held the best land and controlled local commerce. Two of these villages, Ha Tsuen 厦村 and San Tin 新田, are the focus of this article. In 1911 Ha Tsuen had a population of approximately 1,200 people; San Tin had just under 1,100.Footnote 8 All resident males in these villages (save for a handful of slave-retainersFootnote 9) shared the same surname: Teng (鄧) in Ha Tsuen, Man (文) in San Tin. All daughters married-out of the community and all wives married-in from other communities—reinforcing an androcentric culture that rivaled anything found in southern Italy or northern India.

Ha Tsuen and San Tin were also branches of larger, multi-community surname alliances known in the anthropological literature as higher-order lineages.Footnote 10 The Teng higher-order-lineage (H-O-L) included four major village-complexes, all of which fell under British control in 1898 and continued to cooperate (and, at times, feud among themselves) throughout the twentieth century and beyond. The Man H-O-L has a different, and more complex history: in 1898 the new border was demarcated 300 yards north of San Tin. Regular social interaction with the six other Man lineage-villages immediately north of the border was severed in 1941 (following the Japanese invasion of Hong Kong) and did not resume until 1997—when the New Territories was “repatriated” (回歸) to Chinese control.Footnote 11

The landscape of the New Territories was similar to that of other Guangdong coastal regions: Cantonese-speaking communities such as Ha Tsuen and San Tin dominated the best paddy lands in the alluvial plains created by delta rivers, while Hakka-speakers settled in the hills interspersed throughout the landscape. Until the 1910–1912 completion of the Castle Peak Road system, which looped through the northern New Territories, land transport was restricted to single-file paths and cattle trails.Footnote 12 Prior to that date, Ha Tsuen and San Tin depended on ferries and cargo barges to import supplies and carry produce (crops and industrial goods) to market. Lorry traffic, introduced in the late 1920s, transformed the economy and brought the two villages into closer contact with Hong Kong's urban centers, 20 miles to the south.

Ha Tsuen and San Tin were highly organized, close-knit rural communities—in the sense that residents knew each other intimately and outsiders could not walk into these villages without notice (or challenge). Walls and gates were everywhere and fierce watchdogs guarded the narrow paths, threatening anyone whose scent they did not recognize.Footnote 13 Older women sat outside the gates of walled compounds (圍) and did not hesitate to challenge strangers: “Who are you looking for?!” (Cantonese: Wan-bingo-a?! 搵邊個呀?). Hawkers were not allowed into the village after dark and had to notify local authorities before they could sell their wares. The local security force (巡丁)Footnote 14 patrolled at night, ringing a gong every three hours to reassure residents: “Three o'clock [a.m.] and all is well.”Footnote 15 Empty houses were not rented or sold to outsidersFootnote 16 (non-lineage members) until well into the 1990s. By contrast, the nearby tenant villages and settlements of shoreline workers (see below) were essentially unguarded and subject to banditry and intrusion.

Ethnographic research for this article began in 1969, when Rubie Watson and I lived in San Tin—a specialized farming community that emerged on the edge of the delta's saline marshes in the seventeenth century (see Map 1). By the middle of the eighteenth century this otherwise hostile environment had been converted into six polders containing over 480 acres of brackish-water paddy land.Footnote 17 The red rice produced on these polders (沙田) sustained the local farmers for the next three centuries—until political circumstances and economic opportunities led to a full-scale shift to overseas emigration in the 1960s.Footnote 18

In 1977 Rubie Watson began her ethnographic study of Ha Tsuen, 7 miles south of San Tin. During that period (1977–1978) I resided in Ha Tsuen but spent most of my days in the nearby hinterland, where I conducted interviews in tenant villages and coastal settlements along the Laufaushan coast (Map 2). This article draws primarily on first-hand information (based on interviews and observations) collected during those two ethnographic experiences.

Coastal Ecosystems: Laufaushan

Founders of the Teng lineage settled in the rich plain near Laufaushan in the fourteenth century and dominated the economy of their district (鄕) until the late-twentieth century.Footnote 19 A stone tablet (dated 1751) in Ha Tsuen's ancestral hall notes that founding ancestors chose this area because of its ecological and commercial advantages: “[After a] close study of the landscape, and viewing the rich advantages of Ha Tsuen's broad expanse of land, fish, and salt, [our ancestors] moved from Kam Tin [their original home] to live in Ha Tsuen.”Footnote 20 The connection between land and sea thus distinguished the Ha Tsuen Teng from competing lineages that controlled the inland portions of the New Territories.Footnote 21

Teng pioneers constructed a double-crop rice paddy system that produced a prize variety of white rice that was always in demand in regional markets. Ha Tsuen landlords and ancestral estates claimed ownership rights to all land in their district. Besides rice fields and vegetable plots, these claims included the shoreline, estuaries, tidal flats, and expanses of reeds extending 200 yards or more into Deep Bay (see Map 2)—an odd name for what was essentially a flat estuary with two constantly changing channels for the passage of shallow-draft boats and barges.Footnote 22 Anyone wishing to exploit this coastal territory had to pay rent to the Teng ancestral hall, which held ownership rights to most of the Laufaushan coastline and hundreds of acres of prime rice land.Footnote 23

MAP 2 Deep Bay and Yuen Long District, 1969–1978

Locally recognized categories of shoreline property included tidal flats (oyster beds), fishponds, duck ports,Footnote 24 commercial lots (fish shops, tea houses, restaurants), lime kilns, brick kilns, ferry piers, rocky shorelines (fish traps and stationary dip-nets), brackish-water reed fields (raw material for baskets and sleeping mats), and grassy hills above the shoreline (a primary source of domestic cooking fuel).Footnote 25

The sandy beaches of Laufaushan constituted a carefully monitored source of income for Ha Tsuen's ancestor hall. Until the introduction of nylon netting in the 1960s, fishnets were made of ramie and linen—natural fibers that required regular drying after use.Footnote 26 The beaches also supported acres of drying platforms for salt fish, a common staple of the southern Chinese diet.Footnote 27 Larger varieties of oysters were also preserved in this manner. In the 1950s, the resources associated with Laufaushan constituted approximately 25 percent of the Teng ancestral hall's funds (until income declined in the late 1970s).Footnote 28

Oysters: the Delta's Premier Resource

In 1667 the imperial Chinese government granted a deedFootnote 29 to Teng pioneer settlers for the right to collect rent for use of the oyster beds established along the Laufaushan coast. This deed was valid until 1909, when the recently established British colonial administration extended tenancy rights in the form of renewable, twenty-one-year crown leases.Footnote 30 The five largest beds had auspicious names (裕安塘,公和塘,裕和塘,合和塘,豐裕塘), all ending with the character tang (塘), commonly used for ponds or enclosures. During the first decade of the twentieth century, the Laufaushan oyster beds were clustered in tidal waters covering 2,100 hectares (approximately 5,200 English acres).Footnote 31 Map 3 shows the general outline of this oyster terrain; most of these beds date from the mid-seventeenth century and were registered in Qing dynastic archives. Qing authorities leased the beds to designated claimants (including the estate of Ha Tsuen's ancestral hall) and demanded an annual payment for their use.Footnote 32 It appears that British colonial officials devised this map, in part, as a political maneuver in 1904 to claim future territorial rights in Deep Bay and adjoining waters. It is almost certain that oyster production was not carried out in all of the territory covered by these numbered beds. Only a few hundred acres of Yuwutong (裕和塘), which fronted on the Laufaushan coast, was actually devoted to oysters.Footnote 33

Map 3 Oyster Territory in Deep Bay, British Colonial Depictions, 1904

Laufaushan oysters were prized for their tenderness and delicate taste. Some of the larger varieties took up to five years to reach full maturity and were used primarily to make oyster sauce—an essential ingredient of Cantonese cuisine. Favorable water conditions, mild tides, and the skills of local oyster workers were responsible for Laufaushan's primary product. Although oysters were considered to be a luxury, the demand was strong, even during periods of economic recession. This is due, in large part, to a peculiarity of the Cantonese ethnozoological system: villagers classify the oyster as a plant, not an animal. The critical distinction is based on the premise that oysters, unlike other sea creatures, do not move (at least perceptibly) of their own volition. Oysters are “planted” (種) in “fields” (田) like any other crop;Footnote 34 they are neither trapped in nets, nor caught with hooks. Furthermore, oysters are analogous to rice in that the immature spat (like rice sprouts) are transplanted from nurseries to finishing beds, which are the size of large rice paddies, where they grow to full size.Footnote 35 Ha Tsuen residents also claimed that oysters—like plants—do not have souls or spirits (shen 神) and, thus, cannot be killed (sha 殺) when they are cooked or eaten.Footnote 36

The demand for fresh oysters was particularly high during religious observances that proscribe the consumption of animal flesh. Until the 1980s, many local families observed vegetarian restrictions on the first and fifteenth of each lunar month. During the five-day sequence of community purification rites (jiao 醮),Footnote 37 a festival that occurs every ten years in Ha Tsuen District, nothing that fits the category “meat” or “seafood delicacy”Footnote 38 can be brought into the community—let alone eaten. In Ha Tsuen this period was marked by the consumption of (literally) tons of fresh oysters delivered from Laufaushan. Rather than treating the jiao as a period of culinary restraint or abstinence, therefore, older people in Ha Tsuen sometimes spoke of the festival as “oyster eating time.”

Oyster Sauce, Lime Smelting, and Brick Kilns

Laufaushan was also the center for small-scale industries that were dependent on the oyster beds for raw materials. One of the wealthiest families in Ha Tsuen started an oyster sauce (蠔油) factory at Laufaushan in the early 1900s; the enterpriseFootnote 39 employed up to twenty local men and women by the 1960s, many of whom had moved to Laufaushan from other parts of the Pearl River Delta. The company continued production until 1983, when it was closed due to a pollution crisis in the local oyster beds.Footnote 40 The bottled product carried a label identifying it, in English, as “Tang's Sauce.”Footnote 41 It was marketed in Hong Kong, Southeast Asia, and overseas Chinese settlements—including San Francisco's Chinatown and London's Soho District.

The sauce was made by boiling oysters and reducing the fluid to a thick, brown residue. Oyster sauce has a long history in the Pearl River Delta. It was an essential flavoring for the “common pot” banquets that marked weddings, housewarmings, and births of male offspring.Footnote 42 Every village had its own banquet chef who stewed and thickened dried oysters with local herbs to produce a distinctive blend. Recipes were passed from father to son, becoming the equivalent of intellectual property—although village chefs were never paid for their services.Footnote 43

Residents of Laufaushan also worked in kilns (灰窰) that produced lime from crushed oyster shells. The kilns were owned by Teng entrepreneurs who purchased mounds of spent shells, which—like the oysters—were deemed to be the property of Ha Tsuen's ancestral hall. There were four lime kilns in Laufaushan, one of which was still operating in the 1980s.Footnote 44 The lime had many uses, including egg preservation, cloth-dying, leather-tanning, wall-plastering, waterproofing for nets and rope, and as a caulking agent for wooden boats.Footnote 45 Spent shells were also shipped (on narrow barges) to these kilns from oyster beds in the creeks leading to Yuen Long Market.Footnote 46

The Man lineage could not have survived on the margins of Deep Bay without oyster shell lime, a flocculating agent that made it possible for water buffalo teams to plow the otherwise sticky, viscous soil.Footnote 47 Lime was also the active agent in zhuangli 樁籬—a concrete-like building material of uncommon strength and longevity.Footnote 48 Villagers claimed that zhuangli is so strong that bandits knew it was not worth their time to attack walls made of it. Many zhuangli walls in Ha Tsuen remain intact today, 300 years after construction.

Another set of kilns near the village of Mong Jeng (see Map 2) produced two types of bricks from mud and clay harvested in nearby marshlands: low-fired red bricks and high-fired green bricks (青磚 reflecting their color). Green bricks were finished in “step-kilns” (梯窰) situated on hillsides, thereby allowing an up-draft required for high temperatures.Footnote 49 Wealthy households paid high prices for green bricks which were far more resistant to dynamite and iron crowbars—the two essential tools of housebreaking in south China. Bandits who operated in the delta knew that walls built with green brick also had interior layers of lime-based zhuangli concrete, making it more profitable to focus on red brick structures.

The Water Patrol: A Commercial Security Force

Prior to the 1950s, commercial enterprises along the Pearl River Delta could not count on the Chinese government (or, in the New Territories, the British Colonial administration) to provide routine protection against bandits and predatory neighbors.Footnote 50 Specialized security forces thus emerged to guard local industries of any consequence. In the nearby market town of Yuen Long, for instance, shop and mill owners supported their own street patrols that operated day and night.Footnote 51 C. K. Yang describes a “navigation protection corps” that operated in delta waters near Guangzhou to provide armed guards for passenger boats and freighters.Footnote 52

Along the Laufaushan coast, security was assured by an organization known locally as the “water patrol” (水巡).Footnote 53 Rents and fees for the use of oyster beds, lime kilns, fish shops, restaurants, and associated industries were collected by members of this specialized force. These fees, combined with paddy field rents, constituted the primary sources of income for the Teng ancestral hall (i.e., the estate of the founding ancestor and, by extension, the property of all living males who could demonstrate descent from that ancestor). The water patrol's leader, known locally as the “water master” (水長), was selected—by auction—during a formal meeting held in Ha Tsuen's ancestral hall, once every eight years. Auction winners paid an annual sum to the hall and sponsored a banquet for Teng elders who were asked to legitimize the transaction by “chopping” a cloth document with their signature stones.Footnote 54 In 1946 the office fetched HK$500 (per annum), an impressive sum for that period; by 1970 the winning bid had increased to HK$4,000.Footnote 55 In return, the leader and his patrolmen received 3 percent of the harvested oysters (either in kind or in cash equivalent following sale) as their fee for protecting the crop. They also collected an additional 15 percent for the Teng ancestral hall—the legally recognized holder of the subsoil rights (地骨 lit. “earth bones”) of the oyster beds. The remaining 82 percent ended up in the hands of the leaseholders who held the cultivation rights (地皮, “earth skin”).Footnote 56 Leaseholders were responsible for paying the oyster workers (see discussion below), plus additional expenses associated with harvesting, transportation, and marketing.

The water master recruited his own team—all of whom (like the master himself) were members of the Teng lineage. Patrolmen received a share of the harvested oysters and any additional funds collected for guarding businesses in Laufaushan Market. These were, by local standards, lucrative sources of income for men in their twenties and thirties. Membership varied from six full-time operatives in the 1930s to ten during the 1960s.

The Teng, like all major lineages in the Pearl River Delta, also maintained a land-based security force (巡丁)Footnote 57 that patrolled villages and fields in Ha Tsuen District. The activities of the two security forces did not overlap, although there was close coordination between leaders who understood that their primary duty was to maintain the territorial hegemony of the Teng lineage.

The duties of water patrolmen were first, to guard the oyster beds and, second, to protect commercial enterprises that emerged along the coast (seafood processing, lime smelting, shoreline fishing, boat provisioning, and fish marketing). Until the 1970s when restaurantsFootnote 58 and a fresh fish market expanded at Laufaushan, patrolmen devoted most of their time to watching over the oyster beds and making sure that oyster workers did not encroach on territory outside their assigned allotments. Thievery was a constant threat. At low tide the oysters were exposed to the boat and barge traffic along Deep Bay, leading to and from the market town of Shenzhen. Two or more patrolmen were on guard, day and night, to ensure that unauthorized persons did not venture into the beds. Nighttime raids by small groups of oyster thieves who arrived in boats was an ever-present danger. If trouble arose, a loud gong summoned other patrolmen (and Teng men who happened to be in the vicinity) who rushed to the coast.

Oyster workers were well known to Teng patrolmen who were quick to spot strangers. Starting in the 1950s, according to Ha Tsuen elders, harvesting permits were required before anyone could remove oysters from the beds or offer them for sale in the nearby market. These documents—printed ticketsFootnote 59 inscribed with the date and name of harvester—had to be renewed each morning until the harvest was finished, a procedure that allowed patrolmen to check for intrusions into neighboring beds (see Figure 1).

Figure 1 Oyster Harvesting Permit, 1977, Water Patrol, Laufaushan (Photo by J. L. Watson©)

The water master and his part-time assistant/accountantFootnote 60 lived with their families in Ha Tsuen. Ordinary patrolmen resided in a barracks-like headquarters at Laufaushan and were on call twenty-four hours a day, 365 days a year. They slept in shifts on hammocks or camp cots arranged haphazardly around the walls of their barracks and cooked for themselves on coal-burning grills. The majority of water patrolmen were preadapted to this spartan life: they had spent their youth (ages 13–20) in “bachelor houses” (男仔屋) located in the back rooms of Ha Tsuen's ancestral halls and study halls (書房/書院). Overcrowding was a serious problem in tightly nucleated villages such as Ha Tsuen, and many young men were—quite literally—pushed out (推出) of their parents’ homes at puberty.Footnote 61 Two bachelor halls were still operating in Ha Tsuen during the 1970s.

Most water patrolmen were second or third sons of Ha Tsuen tenant farmers and therefore could not count on their families for help with bride-wealth payments.Footnote 62 If they wished to marry and start their own families, they were completely dependent on their own resources. The fees generated by guarding oysters and Laufaushan's auxiliary industries thus made the water patrol an especially attractive occupation for these young bachelors.Footnote 63

Oyster Theft and Boundary Control

Oyster thievery reached its peak, according to retired patrolmen, during the late 1940s and early 1950s, when immigrants from Guangdong Province flooded into the New Territories.Footnote 64 In 1946 the Commissioner of Hong Kong Police authorized the water patrol to carry rifles,Footnote 65 legalizing a practice that had begun surreptitiously in the 1920s when surplus weapons (Enfield rifles and pistols from First World War battlefields) were readily available in the Pearl River Delta.Footnote 66 According to retired patrolmen, the level of violence increased alarmingly during the late 1940s and early 1950s, a period marked by incursions of Nationalist and Communist forces into Hong Kong territory. No member of the Teng lineage was killed while protecting the local oyster beds, but several patrolmen were involved in shootouts with “bandits” (a category that included any armed outsider) in the 1940s.

In 1947 the Hong Kong Government built a police station at Laufaushan and “encouraged” Teng patrolmen to relocate their headquarters from a hill overlooking the oyster beds to a new building in the nearby fish market.Footnote 67 The move underscored a fundamental change in the security system for Ha Tsuen District. Henceforth, the Royal Hong Kong Police assumed the burden of suppressing bandits and monitoring the entry of immigrants from China. Nonetheless, Teng water patrolmen continued to carry firearms, especially during night patrols, until the early 1960s.

Besides watching for thieves, patrolmen were also responsible for verifying the boundaries of oyster beds and for monitoring harvest procedures. These duties presented special problems, given that the Laufaushan mud flats are completely devoid of natural features and—without flag poles that had to be repositioned after every storm—it was impossible to distinguish one oyster bed from another (maps in the conventional sense did not exist).Footnote 68 The Teng relied on mental maps that were passed from father to son in particular patrilines associated with the oyster business.Footnote 69 Leaders of the water patrol were always careful to employ two or three of these specialists as full-time patrolmen. Their ancestors had devised an indigenous system of surveying.Footnote 70 Teng surveyors used distant hills and islands along the western shores of Deep Bay, plus rocks and trees on the Laufaushan coast, as sighting points to align boundary flags in the mud flats. These skills were kept strictly within specialist patrilines. Surveyors always worked alone and the boundary specifications were never recorded in writing.Footnote 71 Their main duties were to realign the flags after storms and to make periodic checks to ensure that oyster workers had not moved into adjoining beds.

Exploiting the Shoreline: Long-nets and Fish Traps

Teng patrolmen were also responsible for regulating all types of fishing that emerged along the shores of Laufaushan. They could not, of course, monitor the activities of specialists who fished in deep waters beyond the coast, but they did control the small-scale entrepreneurs who operated within the low tide zone.

Residents of Ha Tsuen District had a complex lexicon that distinguished between (a) fishing with nets towed behind boats and (b) fishing from the shore. The latter category included a wide variety of methods that employed nets and traps.Footnote 72 Large dip-nets—permanently attached to wooden huts—were located on shore or in shallow mudflats.Footnote 73 Rattan basket-traps (used primarily to catch shrimp, crabs, and small fish) were placed near sluice gates, river mouths, or narrow channels. Fish weirs were created by piling rows of stones along tidal flats that filled with water during hightide and left fish trapped at low tide.Footnote 74

Stationary tube-nets (網魚) constituted by far the most lucrative form of shore fishing, and as such they were carefully regulated by Ha Tsuen's water patrol. Tube-nets measured two meters wide at the mouth and up to 200 meters in length, sufficient to capture entire schools of fish. Teams of 12 to 15 men banded together to set the nets and haul in the catch during low tide; Ha Tsuen elders claimed that, in the 1930s, it was not uncommon to capture up to 500 pounds of high-quality fish on a good day.Footnote 75 Seventeen sites, each with its own place name (see Map 4), were set aside specifically for tube-netting in Ha Tsuen District. In 1978 several Teng elders could recite the entire list, even though they had not visited many of the sites for over two decades. Questions about tube-nets sparked a series of nostalgic stories, most of which focused on male camaraderie among teams of young men who spent long summer nights drinking together under the stars, watching the nets, and dreaming of the big catch that would make them rich. Three men, interviewed in their seventies and eighties, confided that these were among the happiest moments of their lives, even though they had earned barely enough to cover their expenses.

Map 4 Tube-Net Fishing Sites, Laufaushan Coast, ca. 1938

As with all categories of shore land and tidal flats in Ha Tsuen District, the rights to collect fees for use of the seventeen tube-net sites were controlled by the Teng ancestral hall (Yaugungtang 友恭堂). Managers of the hall held an annual auction, restricted to members of the lineage, for rights to exploit these sites. This was an informal, entirely local arrangement. Unlike rights to land and oyster beds, the colonial administration (and before 1898, the Qing imperial government) was neither involved with nor aware of these financial transactions. Several of the sites adjoined land owned by non-Teng farmers who lived in villages miles from Ha Tsuen. The reputation of the Teng water guard was enough to ensure that they were never challenged.

In 1936, according to one of the participants, a team of twelve men (all Teng) paid HK$500 for control of the eight most productive sites.Footnote 76 Interest in long-net fishing declined in the 1950s, due largely to the rising attractions of wage labor. It became increasingly difficult to put together teams of young men who had enough time to manage the nets. By the early 1960s no one was bothering to repair the nets and most of the sites had been abandoned or turned over to small-scale fish trappers (using rattan baskets) who paid a nominal rent to the Teng ancestral hall.

All resources taken from tidal flats bordering Ha Tsuen District were thus subject to fees collected by the Teng water patrol. This policy led to frequent confrontations with tenant farmersFootnote 77 who lived in small villages near the water and relied on shoreline foraging to supplement their diets. Shrimp, crab, and small fish constituted an important source of animal protein for these farmers, whose main staple was sweet potato gruel. In the early 1950s Teng patrolmen discovered a group of women from a tenant village collecting shrimp near one of the long-net sites. As they had not paid fees to the water patrol for fishing rights, their catch was confiscated and their baskets burned. Male residents of the village in question were so incensed by this action that they marched on the water patrol headquarters and threatened to burn it down. According to witnesses who recounted the event, Teng patrolmen brandished rifles to keep the crowd at bay until police officers from the nearby Laufaushan station arrived to take charge of the situation.Footnote 78

Fishing and Social Stigma: Drawing Social Boundaries

The discovery that Teng males regularly participated in shoreline fishing activities came as a surprise when I first learned of it in 1977. Earlier research in San Tin had led me to conclude that Cantonese farmers drew a clear social boundary between people who “plow fields” (Cantonese gaang-tihn 耕田) and those who “catch fish” (juk-yu 捉魚). The residents of San Tin were always keen to separate themselves from the fisherpeople who lived in an encampment of boats and ramshackle mat-sheds (茅屋) near Ha Wan Tsuen, near the mouth of the Shenzhen River—on the outer edge of San Tin District. No self-respecting member of the Man lineage (I was told repeatedly) would engage in full- or even part-time fishing. It was acceptable to manage fresh-water ponds, but this was treated as a subcategory of farming.Footnote 79 The Man spurned all other forms of coastal fishing and hired residents of Ha Wan Tsuen to operate the stake-nets that lined the reclamation dikes in San Tin District.Footnote 80

When I first raised this question with residents of Ha Tsuen District in 1977, many older people claimed that there was a wide social chasm between themselves (land-based farmers) and the fisherpeople who lived on boats anchored near Laufaushan. As my research progressed, however, it became evident that they drew a finer set of distinctions than their counterparts in San Tin and, hence, were less rigid in their definition of social boundaries relating to the exploitation of coastal resources.

In both districts there was a fundamental division between those who lived in brick and stone-built houses on land and those who lived on boats (floating, beached, or permanently anchored), or in temporary mat-sheds. The Teng, however, further distinguished between specialists who treated fishing as a full-time occupation and farmers who engaged in part-time shoreline fishing on an occasional basis to supplement their income.Footnote 81 Until the 1980s many land-based farmers in the Hong Kong New Territories and Guangdong Province socially ostracized full-time fisherpeople and assigned them to the ethnic category daahn-ga yahn 蛋家人, a Cantonese slur that is difficult to translate but means, literally, “egg people” (often Romanized as Tanka). Eugene Anderson notes that the full-time fisherpeople he lived among in the mid-1960s did not use this term and deeply resented it; they referred to themselves as “people of the water” (Cantonese shui-sheung yahn, 水上人).Footnote 82

Many farmers in the New Territories claimed that fisherpeople were descendants of non-Han peoples who had inhabited the delta before the Tang dynastic era (fifth century), when founders of several local lineages first began emigrating to coastal Guangdong from Jiangxi Province.Footnote 83 But genetic evidence does not support such claims. After years of research in the Pearl River Delta, Huang Xinmei, a medical anthropologist at Zhongshan University, concluded that the boat dwelling people of Panyu and Zhongshan Xian, Guangdong Province (identified in her research as 水上居民 “water-dwelling people”), were not significantly different from their land-based, farmer neighbors.Footnote 84 Furthermore, the main thrust of Barbara Ward's path-breaking study of the Hong Kong “boatpeople” is that, culturally and linguistically, they are little different from Cantonese-speaking land people.Footnote 85

The ever-changing nature of social/ethnic identity in the Pearl River Delta is evident in the work of Helen Siu, who has studied historical transformations in Xinhui County, Guangdong Province, 60 miles west of Laufaushan.Footnote 86 Siu notes that one of the dominant lineages in this region may well have started—three centuries earlier—as “hired hands” for land reclamation projects and were likely to have been of Dan (fisherpeople) origin. Local landlords referred to such workers as sha-min 沙民, literally “sand people.”Footnote 87 Siu's ethnic transformation argument is also supported by David Faure's study of land development in the Pearl River Delta.Footnote 88

Nonetheless, the social stigma of occupational origin was still very much alive in the New Territories well into the 1970s. People with fishing origins who wished to “pass” into the untainted status of landed Cantonese had to cut all ties with their past, lest it affect their job chances and marriage prospects.Footnote 89 This pattern of social conservatism may well be a consequence of the Pax Britannia that prevailed in the New Territories, as opposed to the dramatic revolutionary transformations that convulsed Xinhui County (and other parts of the Pearl River Delta) under Communist Party rule. The colonial administrators in rural Hong Kong did not attack and totally transform the preexisting systems of land ownership and social hierarchy they encountered in Ha Tsuen and San Tin Districts.Footnote 90 There were, as many observers have noted,Footnote 91 major changes in the economic, educational, and tenancy systems that underlay New Territories social life, but the Teng and Man lineages (along with their counterparts in British territory) continued to dominate social life in their respective districts until at least the 1990s. More will be said about these recent transformations in the final section of this article.

Oyster Workers and Social Marginality

In the 1960s and 1970s, many older villagers in the New Territories were suspicious of any “outcomers” (外來人) who were not generations-long residents of local communities. One such group was the community of oyster workers who lived in temporary, makeshift housesFootnote 92 along the Laufaushan coast. Theirs was a dangerous occupation that required diving (without oxygen tanks or sealed masks) and the underwater manipulation of stones and large shells that supported the oyster spat as they grew to maturity.Footnote 93 After a few hours of coaching from former water patrolmen, I learned to spot oyster workers by their scarred feet, caused by decades of treading on hidden debris in knee-deep water. The stones had to be reset after each storm to prevent mud from suffocating the oysters. Skill in operating mud-scooters was an essential feature of the tenders’ craft; these wooden sleds, propelled with one foot much like skateboards, were still in use at Laufaushan during the mid-1970s.Footnote 94

According to Liu Tik-sang, who has conducted ethnographic research on coastal ecosystems in the Pearl River Delta since the mid-1970s, Teng water patrolmen often referred to oyster workers by a term best translated as “oyster guys” (蠔佬), a designation that the specialists themselves resented.Footnote 95 Oyster workers, most of whom were surnamed Chan 陳, addressed each other by personal name or nickname—none of which local farmers professed to know or remember. Almost all of these oyster specialists were migrants from the Shajing 沙井 District in Dongguan County, Guangdong—25 miles north along the coast. Most had permanent (brick and tile) homes in sea-side villages in Shajing District and were registered as Shajing Commune members by the Chinese government.Footnote 96 As the Communist land reform campaigns (and associated political movements) proceeded in Shajing, however, many oyster workers began to build temporary homes in unregistered communities along the northern shore of the Laufaushan coast, where they resided throughout the year. This began to change, again, in the 1970s as the Cultural Revolution ended and many older oyster workers resumed the practice of returning to Shajing after retirement.Footnote 97

Farmers in Ha Tsuen District never fully accepted oyster workers as trustworthy neighbors and sometimes referred to them as “West Roaders” (Cantonese sai-louh yahn 西路人), a local term that has derogatory overtones given that “west” is depicted in popular religious iconography as the territory of darkness and death.Footnote 98 During interviews, older people also used “West Roader” as a synonym for bandits who raided New Territories villages during the 1920s and 1930s. By implication, oyster workers were equated with untrustworthy drifters no matter how many generations they had worked in the Laufaushan oyster beds.

My 1978 interviews in Laufaushan revealed a strikingly different picture: oyster workers perceived themselves as members of a skilled occupation with its own secrets and complex history. Senior specialists were addressed by their colleagues as “master” (師傅), much like the Cantonese woodcarvers described by Eugene Cooper.Footnote 99 Specific patrilines maintained a monopoly on esoteric skills that were essential to oyster production. Perhaps the most interesting was the ability to judge water salinity by its taste. Oysters are susceptible to rapid changes in salt content, which can kill an entire crop overnight.Footnote 100 During droughts, and surges of fresh water following typhoons, oyster masters had to judge when it was time to pull the shells, even if they had not reached full maturity.

The skill of water tasting (done every day at high tide) was passed from father to son. Not all oyster workers had such skills because, as one master put it, “you have to learn while your taste-sense (Cantonese hau-meih 口味) is still young.” Teng water patrolmen respected the skills of the Laufaushan taste-masters and allowed them to recruit their own teams of Shajing oyster workers. Jakob Eyferth's study of “skill reproduction” among Chinese paper makersFootnote 101 presents interesting comparisons to the complex demands of oyster farming. Both communities of specialists depended upon the transfer of techniques and knowledge within restricted lines of descent.

Salt and Salt Workers

The industries along the Pearl River Delta consumed vast quantities of low-grade sea salt which was dried in shallow pans adjoining the coast. Salt was used as a preservative for oysters, shrimp, crabs, and other sea foods that were shipped to markets throughout South China and Southeast Asia.Footnote 102 Allied industries—including food preservation (pickling), peanut oil processing, basket weaving, and clothing manufacture—needed salt in large quantities.

Several elders in San Tin and Ha Tsuen noted that small-scale saltpans existed in their districts until the mid-nineteenth century.Footnote 103 One of the wealthiest members of the Man lineage built an elegant, multi-chambered mansion for his extended family on the outskirts of San Tin and purchased a low-level imperial title in the 1860s—reputedly from the proceeds of salt trafficking.Footnote 104 An eighteenth-century TengFootnote 105 is also said to have become wealthy on the proceeds of saltpans located at a long-abandoned site near Ha Tsuen, known to local people as Yim Chong (鹽廠 lit. “salt-yard”). Ha Tsuen's walled market (厦村市) housed several refineries that boiled brine for the manufacture of block-salt that was shipped to urban markets in Hong Kong and Guangzhou.Footnote 106 By the early twentieth century, however, physical evidence of local saltworks had disappeared as a consequence of extensive land reclamation projects that transformed the coastal flats in both districts.Footnote 107

The Tou (陶) lineage, settled in a district adjoining Ha Tsuen, maintained a specialized “salt patrol” (鹽巡), which operated until the mid-nineteenth century, to protect their drying facilities along the Tun Mun Coast.Footnote 108 According to Tou elders,Footnote 109 salt was a tempting target for thieves because it was always in short supply in nearby markets, where it was used to preserve fish, vegetables, and pickles. The leader of the salt patrol was chosen during an annual auction in the Tou ancestral hall. Patrolmen (all of whom had to be members of the Tou lineage) received a share of the finished salt every year, which they could sell in local markets. As in Ha Tsuen District, the duties and territorial purview of this specialized organization did not overlap with the activities of the Tou village patrol (巡丁). Local salt workers were described (by Tou elders) as fisherpeoples and other marginalized “outcomers” (外來人) who lived along the nearby coast.Footnote 110

In San Tin District these specialists were known by a Cantonese term, yim-tin lo 鹽田佬, which translates as “salt-field guys.”Footnote 111 Little is known about the social organization and the division of labor that governed salt production in Xin'an County, even though saltpans existed along the coast during the Ming dynastic era (1368–1644) and probably as early as the tenth century.Footnote 112 The county gazetteer includes a detailed section on the bureaucratic system devised by state authorities to monitor and tax salt production, starting in 1369.Footnote 113 Approximately 3,000 salt makers were registered (for tax purposes) with the Xin'an salt bureau but, as Peter Ng and Hugh Baker note,Footnote 114 there were probably many more unregistered producers in the delta—including the small-scale enterprises that emerged in San Tin and Ha Tsuen Districts.

Marginal Specialists on the Delta's Edge

By the mid-eighteenth-century communities of specialists emerged to exploit resource microniches in the delta ecosystem. One such resource was the expanse of reeds (鹹水草 lit. “saltwater grass”) that flourished in the brackish-water marshes near San Tin and the village of Mong Jeng in Ha Tsuen District. The finest quality reeds were woven into sleeping mats that had the unique quality of remaining cool during the hottest of summer nights; not surprisingly there was always a strong demand for these items in local markets.Footnote 115 Reeds were essential for the cottage weaving enterprises that produced baskets, hats, fish traps, salt bags, and dozens of other essential items sold throughout south China. Marsh grasses were also used as packing materials (for pottery and bottled products), fuel (for domestic and industrial purposes), and thatching for roofs and hut construction.Footnote 116 Water hyacinth (Eichhornia crassipes), a succulent used as pig feed,Footnote 117 was a major source of income for coastline foragers. Semi-permanent communities of reed gatherersFootnote 118 and basket weavers could still be found in the San Tin District until the early 1970s. During my 1969 interviews, these specialists insisted that they were not related to the residents of nearby waterfront settlements and objected when older residents of San Tin sometimes referred to them as “drifters” (Cantonese sui lau chaai 水流柴 lit. “floating driftwood/debris”).Footnote 119

The reed gatherers resided in mat-sheds along the outer dikes of reclamations owned by San Tin's main ancestral hall and were allowed to stay only as long as they served the interests of the Man lineage. A similar community emerged along a major dike system that was built in Ha Tsuen District during the early twentieth century. Residents of these two communities used boats on a regular basis but they lived and slept on land—an important distinction from their point of view. In addition to reed work, they were employed by managers of ancestral halls and reclamation companies to maintain dikes, dig ponds, and repair sluice gates.

In San Tin District the right to harvest reeds was auctioned each year in a local ancestral hall that also served as headquarters for the village patrol; bidding was restricted to members of the Man lineage. Auction winners hired groups of reed gatherers to do the actual harvesting and weaving. Allotments of finished mats, hats, and baskets were then sold to wholesalers in regional markets. Designated reed gatherers were granted exclusive use of the flat surfaces along reclamation dikes where they sorted, soaked, and dried the harvested reeds. San Tin's security force patrolled the area to keep fisherpeople from using the dikes to dry fish and shrimp—space for which was always in short supply.

Large flocks of ducks that fed on insects and small crustaceans in the saline marshes were an important feature of the local economy. The sight of an accomplished duck herder, guiding his ravenous horde with a 30-foot bamboo pole as they swarmed along the reclamation dikes, is truly one of the wonders of delta life.Footnote 120 In spite of the obvious skills involved, local farmers considered duck herding to be a demeaning occupation.Footnote 121 Most households in Ha Tsuen and San Tin Districts kept a few chickensFootnote 122 or geese for personal consumption, but ducks were potentially more destructive of crops and required constant attention—something a busy farmer could ill afford. Duck herders, like reed gatherers, operated under the auspices of Man entrepreneurs who owned the herds.Footnote 123 The butchered and dried ducks ended up in meat shops and restaurants throughout the delta.

Social Boundaries in the Delta

The most significant cultural feature separating people who lived along the water margin of the delta was the practice of endogamy—people married within their own social category and avoided marriages across categories. In the Cantonese context, it is (usually) the woman who marries out of her parental household and moves into the home of her husband's parents (or her husband's household).Footnote 124 The Teng of Ha Tsuen had complex marriage alliances with other landed lineages in the delta region, including the Man of San Tin. These marriages were instrumental in building links between the two communities—reinforcing their power and influence. Fisherpeople and oyster workers also tended to marry within their own occupational category and were strictly proscribed as marriage mates for landed villagers—at least in the two districts under study.

In other parts of the Pearl River Delta, isolated cases of intermarriage between fisherpeople and farmers have been reported—but even these few examples were met by considerable resistance. James Hayes reports one such case that occurred on Lantau Island (in Hong Kong waters) during the 1940s. Residents of the village in question ostracized the in-marrying bride and her children; the family eventually moved to another community located on the New Territories mainland.Footnote 125 During four decades of field research in San Tin and Ha Tsuen Districts, I found no evidence to suggest that marriages of this nature ever existed in the Man or the Teng lineage.

Another important gauge of social interaction is commensality—the practice of dining together in public settings. In the 1960s and 1970s, residents of farming communities avoided teahouses or restaurants that served the fishing community and never (to my knowledge) invited fisherpeople—including those with whom they had business ties—to banquets or weddings. In the thirty or more banquets I have attended in New Territories lineage villages since my research began in 1969, I do not recall a single occasion when an oyster worker, a fisherperson, or a coastal itinerant appeared as a guest.

Nor did the processes of segregation between land and sea end at death. Like all land-owning lineages along the delta, the Teng and Man guarded their burial grounds with special vigilance. One of the most important tasks of the village patrol organizations in Ha Tsuen and San Tin was the regular monitoring of hills and shorelines in their districts. All legitimate graves had to be approved, in advance of burial, by the leader of the local patrol.Footnote 126 Members of dominant lineages, and their long-settled tenant farmers, had rights to local burial sites. Problems arose when strangers attempted to bury their dead in lineage territory. The interment of corpses or bones on land claimed by lineages was treated as a dangerous act—with serious, long-term consequences (including future claims to land or resources).

Unauthorized corpses discovered in Teng or Man territory were exhumed and dumped—without ceremony—into the sea.Footnote 127 Fisherpeople who frequented Laufaushan buried their dead in the hills of isolated (and uninhabited) islands in the delta.Footnote 128 Oyster specialists who worked in Laufaushan solved the problem by repatriating their dead for burial in their home district of Shajing, 25 miles to the north.Footnote 129 The delta has long been a dumping ground for unwanted, unclaimed, or inconvenient corpses. During the Communist land reform campaigns of the 1950s and the Cultural Revolution disruptions in the 1960s, dozens of corpses washed ashore along Laufaushan—a gruesome reminder of political upheavals only a few miles to the north.Footnote 130

The workers who made Laufaushan industries possible had no lasting memorials to commemorate their efforts or even their residence in the area. They were, by the standards of Teng farmers, “people without history.”Footnote 131 Fisherpeople, oyster specialists, salt workers, and reed gatherers left no written genealogies, no inscribed ancestor tablets, and no impressive ancestral halls adorned with stone inscriptions and signed paintings (all of which are on permanent display in Ha Tsuen and San Tin).Footnote 132

By contrast, members of landowning lineages sometimes confided (to this outsider) that they felt overburdened with history—so much, in fact, that the legacy of the past restricted their ability to adapt to changing circumstances in the late twentieth century. “We can't do anything without thinking of the ancestors,” said one Teng entrepreneur who was struggling to start a new business in 1978. “We have to plan for the future, but our history Footnote 133 cannot be forgotten. Look around [gesturing to the Teng ancestral hall]. It is everywhere.”

The End of a Social Order: Laufaushan Transformations

The social, economic, and political underpinnings of village social life have been completely transformed since my initial field investigations of the 1960s and 1970s. By the time I returned to the New Territories for a third round of fieldwork in the 1980s, the oyster beds, reed fields, lime smelteries, and brackish-water paddy systems had all ceased operation. The Laufaushan oyster beds had been hit by a series of pollution crises that effectively destroyed the industry for the next two decades.Footnote 134 Production rebounded somewhat in the 2010s, but traditional cultivation (in shoreline beds) has been largely replaced by floating barges that nurture oysters on ropes suspended in deep water. It is also significant that most of these oysters are not sold, or consumed, in Hong Kong.Footnote 135

Meanwhile the physical landscape of Ha Tsuen District (which includes the Laufaushan coastline) has been transformed by unplanned, chaotic development of the type sometimes referred to as “Desakota” sprawl; others have called it “despoliation.”Footnote 136 Paddy fields and vegetable plots have morphed into a haphazard collection of industrial sites, warehouses, lorry parks, and concrete platforms that store thousands of shipping containers for Hong Kong's busy freighter trade. In 2007, one such platform (with 800 rusty containers stacked six to eight high) arose two hundred yards from the entrance to Ha Tsuen's 350-year-old ancestral hall—a sight many Teng elders found particularly disturbing.Footnote 137

In the course of these disruptions, Hong Kong's newest “New Town” emerged three miles north of Ha Tsuen. Named for the brackish-water paddy system discussed earlier in this article, Tin Shui Wai arose in record time—eight years from start to finish.Footnote 138 Today its thirty- and forty-story apartment blocks house over 300,000 people—all but a tiny fraction of whom have no previous connection to the original inhabitants of the New Territories. Tin Shui Wai is a bedroom community served by a rapid transit system that shuttles residents to and from Hong Kong's major metropolitan center, 20 miles to the south.

Similar transformations have occurred in San Tin District, where one of the world's busiest border stations (Lok Ma Chau) was constructed on paddy fields that had sustained the Man lineage during the six previous centuries.Footnote 139 In 2015, over 28 million people passed through this checkpoint, most of whom departed on fast trains that rocket through the New Territories on their way to and from Kowloon.Footnote 140 Members of the Man lineage began emigrating in large numbers to Europe and Canada during the 1960s and 1970s; the descendants of these pioneer migrants constitute an international diaspora linked by social media and (increasingly infrequent) visits to San Tin.Footnote 141

Today, in the third decade of the twenty-first century, ecotourism and seafood dining are the dominant industries in Laufaushan.Footnote 142 The local scene is dominated by a gleaming, ultra-modern bridge that links the New Territories to new cities on the western shores of Deep Bay (the bridge looms high above the old fishing stations 12 and 13 on Map 4). Laufaushan's drying racks for salt fish and the dilapidated shacks of oyster workers are long gone. Nothing of the complex social past remains, save in the memories of a handful of aging villagers—and in the fieldnotes of a retired anthropologist.

Footnotes

The author wishes to thank Teng Tim-sing, Liu Tik-sang, David Faure, and Hsu Cho-yun for assistance in clarifying many of the Chinese technical terms cited in this article. Patrick Hase and Liu Tik-sang helped locate critical documents regarding the delta's oyster industry and commented on key ethnographic details. Denise Ho kindly shared insights from her current research on oyster workers in Guangdong. Rubie Watson added many insights from her fieldwork in Ha Tsuen District. Thanks also to Angela Collins for preparing the maps.

References

1 Robert B. Marks argues that the Pearl River Delta 珠江三角洲 is not a “true or pure delta but rather an embayment,” a unique structure resulting from the actions of three rivers gradually filling a large coastal bay; see his Tigers, Rice, Silk, and Silt: Environment and Economy in Late Imperial South China (Cambridge: Cambridge University Press, 1998), 32.

2 流浮山, lit. “Flowing Floating Mountain.”

3 James C. Scott, The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia (New Haven: Yale University Press, 2009).

4 Cynthia Chou, The Orang Suku Laut of Riau, Indonesia (Oxford: Routledge, 2010); “The Water World of the Orang Suku Laut in Southeast Asia,” Trans-Regional and Trans-National Studies of Southeast Asia 4 (2016), 265–82.

5 See, for example, Antony, Robert J., “Peasants, Heroes, and Brigands: The Problems of Social Banditry in Early Nineteenth-Century South China,” Modern China 15 (1989), 123–48CrossRefGoogle Scholar and Unruly People: Crime, Community, and State in Late Imperial South China (Hong Kong: Hong Kong University Press, 2016); Paul A. Van Dyke, The Canton Trade: Life and Enterprise on the China Coast, 1700–1845 (Hong Kong: Hong Kong University Press, 2005).

6 Dian H. Murray, Pirates of the South China Coast, 1790–1810 (Stanford: Stanford University Press, 1987), 14. The term “inner Asian frontier” refers to Owen Lattimore's classic study of north China, Inner Asian Frontiers of China (Boston: Beacon Press, 1940).

7 Patrick H. Hase, The Six-Day War in 1899: Hong Kong in the Age of Imperialism (Hong Kong: Hong Kong University Press, 2008).

8 Report on the Census of the Colony for 1911, Hong Kong Sessional Papers, 1911/17.

9 Watson, James L., “Chattel Slavery in Chinese Peasant Society: A Comparative Analysis,” Ethnology 15 (1976), 361–75CrossRefGoogle Scholar.

10 Maurice Freedman, Chinese Lineage and Society: Fukien and Kwangtung (London: Athlone Press, 1966), 20–21; Watson, James L., “Chinese Kinship Reconsidered: Anthropological Perspectives on Historical Research,” China Quarterly 92 (1982), 589–22CrossRefGoogle Scholar.

11 The history of San Tin's reengagement with Man lineage communities across the border following the 1997 “return” is discussed in Watson, James L., “Virtual Kinship, Real Estate, and Diaspora Formation: The Man Lineage Revisited,” Journal of Asian Studies 63 (2004), 893910CrossRefGoogle Scholar.

12 Stuart Lockhart, a leading colonial official during the transition to British rule, notes in his 1899 Report on the Extension of the Colony of Hongkong: “There are no roads in the ordinary acceptance of that term in the territory.” Hong Kong Sessional Papers, No 9/99 (177), January 6, 1899, p. 8.

13 While living in San Tin (1969–1970) I never left my house without a heavy walking cane and, like other residents, I had to be constantly on guard against canine challengers.

14 James L. Watson, “Self-Defense Corps, Violence, and the Bachelor Sub-Culture in South China: Two Case Studies,” in Village Life in Hong Kong: Politics, Gender, and Ritual in the New Territories, edited by James L. Watson and Rubie S. Watson (Hong Kong: Chinese University of Hong Kong Press), 251–65; more will be said about this organization below.

15 Cantonese: Sam-dim jung, mouh mahn-tai 三點鐘冇問題, sung in a rhythmic manner (which I can still hear in my mind's ear, more than fifty years later).

16 A handful of outsider shopkeepers and, in Ha Tsuen's case, the household of a resident priest (喃嘸佬), were exceptions to this rule. Rubie Watson and I were granted temporary residence in San Tin (1969–70, 16 months) and Ha Tsuen (1977–78, 12 months), but only after lengthy vetting by lineage leaders, with help from Assistant District Officers (Cantonese), Yuen Long District Office.

17 Hong Kong Government land records dating from 1905, kept at the Yuen Long District Office and reviewed by author in 1970 and 1977.

18 James L. Watson, Emigration and the Chinese Lineage: The Mans in Hong Kong and London (Berkeley: University of California Press, 1975).

19 For a detailed ethnography of Ha Tsuen and the Teng lineage see Rubie S. Watson, Inequality Among Brothers: Class and Kinship in South China (Cambridge: Cambridge University Press, 1975); for the early history of Ha Tsuen, see Patrick H. Hase, “Notes on the History of Ha Tsuen,” in Village Studies: Settlement, Life and Politics in Traditional New Territories Communities (Hong Kong: City University Press, 2020), 45–143.

20 Author's (rather liberal) translation from a key passage in the stone tablet in Yaugungtong 友恭堂, Ha Tsuen: 尤精地學,見厦村地盤廣濶,漁鹽之利,甲於海邦,故由錦田而遷居厦村. For the full Chinese text, see David Faure 科大衛, Bernard H. K. Luk 陸鴻基, and Alice N. H. Lun Ng 吳倫霓霞, 香港碑銘彙編,第一 (Hong Kong Stone Inscriptions, Volume 1). 香港市政局出版 (Hong Kong: Urban Council, 1976), 33–36. For a detailed English translation, see Hase, Village Studies, 46–52. The history of this hall, and its role in local politics, is discussed in Watson, Rubie S., “The Creation of a Chinese Lineage: The Teng of Ha Tsuen, 1669–1751,” Modern Asian Studies 16 (1982), 69100CrossRefGoogle Scholar.

21 One of these lineages was the Teng of Kam Tin 錦田, the original home of the Ha Tsuen Teng, and is widely recognized as the wealthiest community in the region. See Baker, Hugh D. R., “The Five Great Clans of the New Territories,” Journal of the Royal Asiatic Society Hong Kong 5 (1966), 2548Google Scholar, for a survey of lineage-communities in the New Territories.

22 The English name “Deep Bay,” appears on British colonial maps created during the mid-nineteenth century and appears, for instance, on an 1841 “Chart of the Canton River,” by James Wyld, Geographer to the Queen; see Hal Empson, Mapping Hong Kong: A Historical Atlas (Hong Kong: Government Printer, 1992), 27. Stuart Lockhart notes: “Deep Bay … is extremely shallow and at low tide miles of mud can be seen”; Papers Relating to the Extension of the Colony of Hongkong, Sessional Papers, No. 9/99 [177] (January 6, 1899), 2. According to local villagers, the original Chinese name for this bay, 后灣 or 后海灣 (lit. “Empress Bay” or “Empress Sea Bay”), draws on the popular local deity 天后 (usually translated as “Empress of Heaven”). Patrick Hase notes that 后 is also the simplified form of 後 (behind) and speculates (in a personal communication) that the bay's name originated as a geographical counterpoise to the small estuary 前海灣 that was, literary, “in front” 前 of Nantou 南頭—an important town on the west side of Deep Bay. By this reasoning, the larger body of water beyond 前海灣 would have been known as 後/后海灣—literally the “bay behind.” On Chinese language maps produced in the People's Republic the term used is 深圳湾 (Shenzhen Bay), named for the market town (turned city) nearby. The use of this name is a clear statement of Chinese political suzerainty and a rejection of British imperial ambitions.

23 R. Watson. “Creation of a Chinese Lineage,” 61–72.

24 Duck ports/jetties (yabu 鴨埠) are fenced enclosures that front on a river, a pond, or a bay; the ducks are herded in and out of the enclosure via a wood-plank wharf (Liu Tik-sang, personal communication). In local documents the term for duck port was 鴨埗.

25 Territorial rights to collect grass and brush were strictly maintained and enforced by the Teng lineage. Unauthorized fuel-gathering expeditions were a frequent a source of intercommunity conflict, as discussed in the work of Patrick H. Hase, Custom, Land and Livelihood in Rural South China: The Traditional Land Law of Hong Kong's New Territories, 1750–1950 (Hong Kong: Hong Kong University Press, 2013), 61–64; see also Elizabeth L. Johnson and Graham E. Johnson, A Chinese Melting Pot: Original People and Immigrants in Hong Kong's First ‘New’ City (Hong Kong: Hong Kong University Press, 2019), 20ff. Dried grass was also an important industrial resource, used in lime smelteries along the Laufaushan coast; see Wong Tak-yan, “Lime-Making in Tsing Yi,” Journal of the Royal Asiatic Society Hong Kong, 24 (1984), 295.

26 Fees were charged for the privilege of drying nets along the beach; see Liu Tik-sang, “Home on the Water: Livelihood and Society of the Fishing Community in Tai Po,” in Traditions and Heritage in Tai Po, edited by Liu Tik-sang (Hong Kong: Tai Po District Council, 2008), 109, on linen nets, and Ward, Barbara E., “Kau Sai, An Unfinished Manuscript,” Journal of the Royal Asiatic Society Hong Kong 25 (1985), 3233Google Scholar on ramie nets.

27 In 1947 the Hong Kong fishing industry produced 11,293 long tons of salt/dried fish: see Hong Kong Statistics, 1947–1969 (Hong Kong Government Press: Census and Statistics Department, 1969), 78. A “long ton” equals 2,240 pounds.

28 1978 interview with the hall's financial trustee. The bulk of the ancestral hall's income derived from leases on ancestral estate lands and rent from shops in Ha Tsuen's market.

29 Xiacun xiangyue jiayin nian jiao tekan 厦村鄉約甲寅年建醮特刊 (Ha Tsuen Jiao Celebration Brochure), published locally in Yuen Long, Hong Kong New Territories, p. 28 (copies available from author). The term used in this source for Imperial deed is hongqi (紅契 “red deed”).

30 Hong Kong Administrative Report, New Territories, 1909, 13.

31 See original map as depicted in Richard Irving and Brian Morton, A Geography of the Mai Po Marches (Hong Kong: Hong Kong University Press. 1988), 22, which draws on H.K.R.S. 58–1-24 (17), C.S.O. no 517/1904, “Oyster Cultivation in Deep Bay,” Land Registry Office, no. 601, Aug. 27, 1908, Public Records Office, London. Similar stretches of oyster and razor clam beds were developed along the Fujian coast; see Michael Szonyi, The Art of Being Governed: Everyday Politics in Late Imperial China (Princeton: Princeton University Press, 2018), 115, 120, 244n56. Large-scale scallop harvesting was a major industry in the Taipo estuary, on the eastern coast of the New Territories (Liu, Home on the Water, 104–106).

32 CSO Ext. 586/03, Oyster Bed at Ping Shan, Petition of Pang Wai-leung. Ag. C.S.P. (1903).

33 In 1966, this bed (No. 1 on Map 3) produced 30,582 catties of fresh oyster meat, 9,331 catties of dried oyster meat, and 4,138 catties of oyster sauce (1 catty 斤 = 1.3 English pounds). Source: Yuen Long District Office, 1984 memo, “Background Notes on the Oyster Industry of Deep Bay,” archives of James Hayes, copy courtesy of Patrick Hase (original deposited in the Hong Kong Collection, Hong Kong University Library).

34 The Cantonese term for individual oyster beds is hou-tin 蠔田, literally “oyster field”; the term 塘 (Cantonese Tong) is used in imperial records and covers a larger section of tidal flats, as illustrated in Map 3. There were several individual “fields” (田) in every Tong 塘. Until the late 1980s, local people wrote the character for oyster as 虫+毫, a Cantonese variation of the standard dictionary term 蠔 (see, for example, oyster harvesting pass in Figure 1). Other variations on this character also exist in local documents. The dominant species in the Pearl River Delta is the Pacific cup-oyster, Crassostrea gigas; see Brian Morton and John Morton, The Sea Shore Ecology of Hong Kong (Hong Kong: Hong Kong University Press, 1983), 242.

35 The oyster spat (幼苗) matured for one or two years in breeding beds (繁殖區) until they were relocated to maturing beds (寄肥區) for an additional year—or sometimes two years—prior to harvest; see Morton, Brian and Wong, P. S., “The Pacific Oyster Industry in Hong Kong,” Journal of the Royal Asiatic Society Hong Kong 15 (1975), 139–49Google Scholar.

36 Buddhists in south China believe that oysters grow from “seeds” and, hence, are edible; see Cheung, Sidney, “Floating Mountain in Pearl River: A Study of Oyster Cultivation and Food Heritage in Hong Kong,” Asian Education and Development Studies 8 (2019), 433–42CrossRefGoogle Scholar.

37 On jiao rites see Choi Chi-cheung, “Reinforcing Ethnicity: The Jiao Festival in Cheung Chau,” in Down to Earth: The Territorial Bond in South China, edited by David Faure and Helen Siu (Stanford: Stanford University Press, 1995), 104–22; Liu Tik-sang 廖迪生, Xianggang tianhou congbai, 香港天后崇拜 (Tianhou Worship in Hong Kong) (Hong Kong: Joint Publishing); and James L. Watson, “Standardizing the Gods: The Promotion of T'ien Hou (‘Empress of Heaven’) along the South China Coast, 960–1960,” in Popular Culture in Late Imperial China, edited by David Johnson, Andrew Nathan, and Evelyn Rawski (Berkeley: University of California Press, 1985), 292–324.

38 Mammal flesh (rou 肉) and seafood (海鮮/海味, general terms for fresh/dried fish, crab, shrimp, lobsters, and other creatures that move or swim) are proscribed during the jiao.

39 Known as “Yu Wo Tong Oyster Products” (裕和塘產品).

40 The mid-1980s death of the principal oyster-sauce specialist (a member of the Teng lineage) also hastened the demise. A Ha Tsuen elder summarized the problem during a 2009 interview: “No one knew all the secret techniques of making good oyster sauce and young people were not interested in learning. They could get better jobs in Yuen Long or Kowloon.”

41 Many Ha Tsuen Teng 鄧 prefer to Romanize their surname as “Tang,” the version used on Hong Kong identification cards. The Cantonese pronunciation, however, is close to the Mandarin surname “Deng,” as in Deng Xiaoping. “Teng” is the Romanized form of this surname that was used by Ha Tsuen villagers until the 1990s.

42 Watson, James L., “From the Common Pot: Feasting with Equals in Chinese Society,” Anthropos 82 (1987), 389401Google Scholar.

43 This changed in the 1990s when “common pot” chefs became television celebrities and were featured in Hong Kong “repatriation” (回歸) banquets; see James L. Watson, “Feasting and the Pursuit of National Unity: American Thanksgiving and Cantonese Common-Pot Dining,” in Culinary Nationalism in Asia, edited by Michelle T. King (London: Bloomsbury Academic, 1997), 252–63; Chan, Selina, “Food, Memories, and Identities in Hong Kong,” Identities: Global Studies in Culture and Power 17 (2010), 204–27CrossRefGoogle Scholar.

44 Irving and Morton, A Geography of Mai Po Marshes, 42–43.

45 Discussed in Rudolf P. Hommel, China at Work: An Illustrated Record of the Primitive Industries of Chinas Masses, Whose Life is Toil, and Thus an Account of Chinese Civilization (New York: John Day, 1937), 277–78. Oyster lime was used as a caulking material for wooden boats, see Sung Ying-hsing, Chinese Technology in the Seventeenth Century: T'ien-kung K'ai-wu, translated by E-Tu Zen Sun (New York: Dover Publications, 1966), 202. Lime was also the key ingredient for caulking material for wooden boats (Sung, Chinese Technology in the Seventeenth Century, 202). The Cantonese term for lime is 石灰, lit. “stone ash.” Irving and Morton, A Geography of Mai Po Marshes, 42–43, note that there were four lime kilns in Laufaushan, one of which was still operating in the 1980s.

46 Mak Shui-hung, “The Fish Ponds and Oyster Beds in Wang Chau Area, Hong Kong,” in Land Use Problems in Hong Kong, edited by S. G. Davis (Hong Kong: Hong Kong University Press, 1964), 150.

47 Watson, James L., “Saltwater Margin: A Common Fields System in South China,” Past and Present 224 (2014), 271CrossRefGoogle Scholar; see also Armando da Silva, Tai Yu Shan: Traditional Ethnological Adaptation in a South Chinese Island (Taibei: Orient Cultural Service, 1972), 51.

48 Zhuangli consists of lime, fine sand, granite pebbles, liquefied sugar, and water mixed together and pounded into layers. A comparable form of lime-based concrete was used along the Fujian coast during the seventeenth century (Sung Ying-hsing, Chinese Technology in the Seventeenth Century, 202).

49 Irving and Morton, A Geography of Mai Po Marshes, 41. Lockhart, Extracts, 543–44, mentions this kiln system in his 1899 report on the New Territories. See also Naquin, Susan, “The Material Manifestations of Regional Culture,” Journal of Chinese History 3 (2019), 369CrossRefGoogle Scholar, on brickmaking in north China. Note that one of the seventeen fishing stations shown in Map 3 is called “Brick Kiln” (磚窰).

50 To cite only one example, in 1912 a gang of twenty bandits from the eastern banks of the Pearl River Delta attacked a shop in Ha Tsuen and made off with goods worth HK$4,000—a very large sum for that era; see London, Public Records Office, CO1290/400, Hong Kong Dispatches 1913, item 14223, 302–7: “Armed Robbery in the New Territories, April 7, 1913”).

51 A wide variety of security forces guarded Hong Kong commercial enterprises in the nineteenth and early twentieth centuries; see Sheila E. Hamilton, Watching Over Hong Kong: Private Policing, 1841–1941 (Hong Kong: Hong Kong University Press, 2008). See also Patrick H. Hase, “The Alliance of Ten: Settlement and Politics in the Sha Tau Kok Area,” in Faure and Siu, eds., Down to Earth, 156–57, which discusses the “town watch” that guarded Shataukok Market (in the northeast New Territories) during the nineteenth century.

52 C. K. Yang, A Chinese Village in Early Communist Transition (Cambridge: Massachusetts Institute of Technology Press, 1959), 110.

53 During my 1978 interviews with Teng elders, several noted that the term “water patrol” (水巡), and the institution it represented, was not unique to Ha Tsuen. They cited other examples in the oyster beds of Shajing 沙井 District, Guangdong Province.

54 “Elder” (Cantonese fu lo 父老 or, in formal documents, 耆老) is a local term of address for lineage males aged 61 or older (it was not used for women). Elders, many of whom were illiterate, carried stone seals (圖章) with their names carved in ornate characters.

55 According to Ha Tsuen elders, author's interviews 1978.

56 This is often referred to as the dual-ownership system, which was common throughout the Pearl River Delta; see Kamm, John T., “Two Essays on the Ch'ing Economy in Hsin-An, Kwangtung (Perpetual Tenancy),” Journal of the Royal Asiatic Society Hong Kong 17 (1977), 5960Google Scholar; Palmer, Michael J., “The Surface-Subsoil Form of Divided Ownership in Late Imperial China: Some Examples from the New Territories of Hong Kong,” Modern Asian Studies 21 (1987), 1119CrossRefGoogle Scholar; Wickberg, Edgar, “Another Look at Land and Lineage in the New Territories, ca. 1900,” Journal of the Royal Asiatic Society Hong Kong 21 (1981), 2542Google Scholar.

57 The term 巡丁 literally means “patrol men,” (in village discourse 丁 refers to able-bodied males in their physical prime).

58 By the mid-1970s there were eighteen restaurants and tea houses at Laufaushan, employing nearly 100 people; see Yuen Long District Office Files, 1984 memo, “Background Notes on the Oyster Industry of Deep Bay,” 7.

59 Oyster harvesting permits (採蠔證) were printed in the nearby town of Yuen Long and issued by the manager of the water patrol, a post that required a relatively high level of literacy. Patrolmen referred to these permits as 出紙, “ticket” or “pass.”

60 This man had to be literate and trained in basic accounting; his primary task was to keep records, issue harvesting permits (see Figure 1), and deal with local businessmen.

61 Producing too many sons was considered a problem even among the wealthy elite in premodern China; see Patricia B. Ebrey, Family and Property in Sung China: Yüan Ts'ai's Precepts for Social Life (Princeton: Princeton University Press, 1984), 106–7. On the social problems caused by bachelors in Chinese society, see Quanbao, Jiang and Sanchez-Barricarte, Jesús, “Bare Branches and Social Stability: A Historical Perspective on China,” Frontiers of History in China 6 (2011): 538–56Google Scholar; David Ownby, “Approximations of Chinese Bandits: Perverse Rebels, Romantic Heroes, or Frustrated Bachelors,” in Chinese Femininities/Chinese Masculinities, edited by Susan Brownell and Jeffrey Wasserstrom (Berkeley: University of California Press, 2002).

62 Bride-wealth (禮金) is cash paid by the family of the groom to the family of the bride. In Ha Tsuen during the mid-twentieth century these payments constituted (approximately) two years average income for an ordinary farm family—necessitating years of savings (see R. Watson, Inequality Among Brothers, 120–22).

63 The role of bachelors in Chinese society is discussed in Antony, Robert J., “Peasants, Heroes, and Brigands: The Problems of Social Banditry in Early Nineteenth-Century South China,” Modern China 15 (1989): 123–48CrossRefGoogle Scholar; see also Hudson, Valerie and Boer, Andrea, “A Surplus of Men, A Deficit of Peace,” International Security 26.4 (2001), 538CrossRefGoogle Scholar.

64 This view is confirmed by Morton, Brian and Wong, P. S., “The Pacific Oyster Industry in Hong Kong,” Journal of the Royal Asiatic Society of Hong Kong 15 (1975), 146Google Scholar.

65 The Hong Kong Government gave formal permission for indigenous security forces to carry arms in 1946 (Annual Report of the District Office, New Territories, 1946–47, 2). Ha Tsuen's water patrol was formally registered under the Hong Kong government's “Watchmen's Ordinance, Chap. 299, issued by authority of the Commissioner of Police for Yau Kung Tong Oyster Beds,” dated September 30, 1977. This signed document was kept in a locked safe at the Laufaushan water patrol headquarters (author's personal observation).

66 Several San Tin men, acting for their local security force (巡丁), purchased Enfield rifles in the market town of Shenzhen 1930s (author's 1969 interviews). In 1913 Hong Kong's Governor Francis Henry May gave personal permission for a shopkeeper in Ha Tsuen to purchase “modern rifles” (including a bolt-action Mauser) for protection following a bandit raid cited in note 50 above.

67 See Laufaushan Police Post,” HKRS 478–2-12, Hong Kong Public Records Office. A much larger, fortress-like police station was built on a hill overlooking the Laufaushan coast in 1962—a date that corresponds to an upsurge of illegal migration from China.

68 1904 Hong Kong Government regulations required that oyster beds be clearly marked with bamboo poles, two feet above high-water mark, to protect shipping (Hong Kong Government Gazette, Notification no. 544, “Regulation of Oyster Fisheries at Deep Bay,” July 27, 1904).

69 These Teng patrilines owned no land of any consequence and relied on specialized services to survive in Ha Tsuen. Men from these lines also became “minders” and bodyguards for wealthy Teng landlords; see J. Watson, “Saltwater Margin,” 273–74).

70 This method differed in interesting ways from that employed in Maine lobster fisheries to demarcate territory. Concepts of territoriality in Maine lobster grounds are discussed by James M. Acheson, The Lobster Gangs in Maine (Hanover, NH: University Press of New England, 1988), 71–83); New Zealand crayfish systems were also governed by indigenous notions of territoriality; see Levine, H. B., “Controlling Access: Forms of ‘Territoriality’ in Three New Zealand Crayfishing Villages,” Ethnology 32/2 (1984), 8999CrossRefGoogle Scholar. Cantonese oyster production was distinct from sea fishing, which implies catching prey in open waters. Laufaushan oyster production is best treated as a form of farming, which implies unambiguous notions of property ownership and leasing.

71 This system was still operative during my 1977–78 field research, although many local people felt that it was time to introduce scientific survey procedures.

72 The colloquial Cantonese terms (1977–78 usage) were: to-yu 拖魚, towing nets behind ocean-going boats; saat-yu 撒魚, casting nets to catch fish; mohng-yu 網魚, trapping fish in long, tubular nets; and jam yu-luhng 浸魚籠, catching fish in trap-baskets. For an outline of terminology used by Cantonese fisherpeople in the 1960s, see Eugene N. Anderson, The Floating World of Castle Peak Bay (Washington, DC: American Anthropological Association, 1970), 42–52.

73 Cantonese aau-yu 拗魚; in other parts of the Pearl River Delta, these stationary dip-nets were called jang-paahng 罾棚. The nets were raised and lowered in tandem with the tides.

74 Several fish weirs were located near the tube-net sites shown in Map 4.

75 1978 interviews with several Teng elders who had participated in tube-net fishing (during the 1930s and early 1940s).

76 To put this sum in perspective, S. G. Davis notes that the average skilled worker (carpenter, bricklayer, plumber) earned from HK$30 to HK$40 per month in 1940. In the same year, rice sold for an average of HK$0.84 per catty see S. G. Davis, Hong Kong in its Geographical Setting (London: Collins, 1949), 155–56.

77 Watson, James L., “Hereditary Tenancy and Corporate Landlordism in Traditional China: A Case Study,” Modern Asian Studies 11 (1977), 161–82CrossRefGoogle Scholar.

78 During my 1978 interviews, elders in this tenant village cited the incident as a major factor in their decision to withdraw from the Ha Tsuen District Rural Committee (a local government organization) in 1952 and join the committee established by the nearby village of Ping Shan. (It was no coincidence that Ping Shan was Ha Tsuen's chief political rival.)

79 On pond fishing in the New Territories, see Sidney C. H. Cheung, “Fish in the Marsh: A Case Study of Freshwater Pond Fishing in Hong Kong,” in Food and Foodways in Asia, edited by Sidney Cheung and Tan Chee-beng (London: Routledge, 2007), 37–50.

80 Described in J. Watson, “Saltwater Margin.” Helen Liu refers to specialized groups of “boat-dwelling laborers” who worked on major reclamation dikes in the Pearl River Delta; see her essay “Recycling Tradition: Culture, History, and Political Economy in the Chrysanthemum Festivals in South China,” Comparative Studies in Society and History 32 (1990), 770.

81 Given the stereotypes that attached to full-time fishing, it was not surprising that land people in Ha Tsuen District were defensive about their use of boats. Teng patrolmen, for instance, were always careful to emphasize that they used “Western-style” rowboats, which have two oars mounted on opposite sides (until the 1980s full-time fisherpeople used sampans with a single oar mounted at the rear).

82 Anderson, Eugene N., “Prejudice and Ethnic Stereotypes in Rural Hong Kong,” Kroeber Anthropological Society Papers 37 (1967), 90106Google Scholar, and The Floating World of Castle Peak Bay. See also Ward, Barbara E., “A Hong Kong Fishing Village,” Journal of Oriental Studies 1 (1954), 195214Google Scholar and “Floating Villages: Chinese Fishermen in Hong Kong,” Man 59 (1959), 44–45. It is also interesting to note that, further north along the Chinese coast in Zhejiang province, fisherpeople did not suffer such a high level of social discrimination—perhaps because they also engaged in farming and other agricultural pursuits; see Micah S. Muscolino, Fishing Wars and Environmental Change in Late Imperial and Modern China (Cambridge: Harvard University Press, 2009), 24 ff.

83 During the 1960s, people in San Tin and Ha Tsuen referred to themselves as (Cantonese) tong-yahn 唐人 (people of the Tang [Dynasty]). This term was used to distinguish themselves from “outcomers” (Cantonese, ngoi-loih yahn 外來人), such as the Hakka and Chaozhou farmers who also lived in the New Territories.

84 See Huang Xinmei 黃新美, Zhujiang kou shui shangshui humin 珠江口水上居民(疍家)的研究. 中山大學出版社 (Guangzhou: Zhongshan University Press, 1989). For a survey of speculations regarding the non-Han ethnic origins of south China's fisherpeople, starting as early as the second century CE, see Su-ching, Chen, “The Origin of the Tanka,” Nankai Social and Economic Quarterly 8 (1935), 250–73Google Scholar.

85 Barbara E. Ward, “Varieties of the Conscious Model: The Fishermen of South China,” in The Relevance of Models for Social Anthropology, edited by Michael Banton (London: Tavistock, 1965); see also Liu Tik-sang, Shuishang tange 水上嘆歌 (Hong Kong: South China Research Center, Hong Kong University of Science and Technology, 2018) for a discussion of women's subculture among Hong Kong fisherpeople.

86 Helen F. Siu, Agents and Victims in South China: Accomplices in Rural Revolution (New Haven: Yale University Press, 1989), 51–63; see also Helen Siu and Liu Zhiwei, “Lineage, Market, Pirate, and Dan: Ethnicity in the Pearl River Delta of South China,” in Empire at the Margins: Culture, Ethnicity, and Frontier in Early Modern China, edited by Pamela Crossley, Helen Siu, and Donald Sutton (Berkeley: University of California Press, 2006), 285–92.

87 This term derives from 沙田 (lit. “sand fields”), the Cantonese term for polder reclamations.

88 David Faure, Emperor and Ancestor: State and Lineage in South China (Stanford: Stanford University Press, 2007), 41–44 ff; see also He Xi, “Gods Adrift: Religious Ritual and Local Society in Naozhou Island,” in The Fisherfolk of Late Imperial and Modern China, edited by He Xi and David Faure (Abingdon: Routledge, 2016), 99–100; and Ho-fung, Hung, “Thousand-Year Opposition and Thousand-Year Resistance: The Tanka Fisherfolks in Tai O Before and After Colonialism,” Chinese Sociology and Anthropology 30.3 (1998), 7599CrossRefGoogle Scholar.

89 The parallels to other marginal groups are striking, as illustrated in ethnographic studies of fishing subcastes in south India (see Ajantha Subramanian, Shorelines: Space and Rights in South India (Stanford: Stanford University Press, 2009)); burakumin leatherworkers in Japan (see George DeVos and Hiroshi Wagatsuma, Japans Invisible Race (Berkeley: University of California Press, 1966) and Joseph D. Harkins, Working Skin: Making Leather, Making a Multicultural Japan (Berkeley: University of California Press, 2014)); and coastal fishing communities in nineteenth century Scotland (see Nadel, Jane H., “Stigma and Separation: Pariah Status and Community Persistence in a Scottish Fishing Village,” Ethnology 23 (1984), 101–15CrossRefGoogle Scholar).

90 For a comparative perspective on the social transformations that occurred just across the New Territories border, see Anita Chan, Richard Madsen, and Jonathan Unger, Chen Village: Revolution to Globalization (Berkeley: University of California Press, 2009).

91 See for example, Chun, Allen, “Policing Society: The ‘Rational’ Practice of British Colonial Land Administration in the New Territories of Hong Kong, c. 1900,” Journal of Historical Sociology 3 (1990), 401–22CrossRefGoogle Scholar.

92 These small, single-room dwellings were constructed of wooden planks (often floating debris), with tin roofs.

93 Eugene N. Anderson, “Traditional Aquaculture in Hong Kong,” in Mountains and Water: Essays on the Cultural Ecology of South Coastal China, edited by Eugene N. Anderson and Marja Anderson (Taibei: Orient Cultural Service, 1973), 59; see also Brian Morton and John Morton, The Sea Shore Ecology of Hong Kong (Hong Kong: Hong Kong University Press, 1983), 243.

94 Local residents refer to the mud-scooter as hua-pan 滑板. The dictionary term is 木橇; see Morton, Brian and Wong, P. S., “The Pacific Oyster Industry in Hong Kong,” Journal of the Royal Asiatic Society Hong Kong 15 (1975), 143Google Scholar (see also their photos F and G for a scooter in use). The mud-scooter allows oyster workers to “glide over the semi-liquid surface [of an oyster bed] with the ease and speed of a cyclist”; see Morton and Morton, The Sea Shore Ecology of Hong Kong, 243).

95 Liu Tik-sang. Becoming Marginal: A Fluid Community and Shamanism in the Pearl River Delta of South China (PhD diss., University of Pittsburgh, 1995), 49–51, 201–3.

96 Denise Ho (personal communication) who discusses Shajing oyster workers during the socialist era in a forthcoming study, Fields from the Sea: Cultivating Oysters in Shajing from Socialism to Postsocialism. On the pre-socialist era, see Jenyns, Soame, “Our Local Oyster,” Hong Kong Naturalist 2 (1931), 164–66Google Scholar.

97 A 1984 report by the Yuen Long District Office notes that oyster workers retired at or before age 40 “due to harsh working conditions, particularly diving for oysters”‘ see Yuen Long District Office, 1984 memo, “Background Notes on the Oyster Industry of Deep Bay,” 8, see note 33 above).

98 The “west” is also the land of the setting sun and the yin 陰 principle, characterized by female concerns; the “east,” by contrast, is the direction of the rising sun and the yang 陽 principle, associated with life, light, and male concerns. Patrick Hase (personal communication) speculates that the term might also refer to those who lived on the western shores of Deep Bay.

99 Eugene Cooper, The Wood Carvers of Hong Kong: Craft Production in the World Capitalist Periphery (Cambridge: Cambridge University Press, 1980), 49–50.

100 Bromhall, J. D., “On the Biology and Culture of the Native Oyster of Deep Bay, Hong Kong,” Hong Kong University Fisheries Journal 2 (1958), 93108Google Scholar; Jenyns, “Our Local Oyster,” 164, notes that oysters cannot survive in water with a salinity level lower than three percent. The danger month is July, when the Pearl River's discharge is highest; see Xu, Jie, “Long-Term and Seasonal Changes in Deep Bay, Hong Kong,” Estuaries and Coasts 33 (2010), 412, 414CrossRefGoogle Scholar.

101 Jacob Eyferth, Eating Rice from Bamboo Roots: The Social History of a Community of Handicraft Papermakers in Rural Sichuan, 1920–2000 (Cambridge: Harvard University Press, 2009), 6–11.

102 Trade networks for salt and related products in the upriver districts of south China are discussed by Steven B. Miles, Upriver Journeys: Diaspora and Empire in Southern China (Cambridge: Harvard University Press, 2017).

103 These claims are supported by the 1858 account by Rev. Rudolph Krone, who mentions salt pans in the Yuen Long area, which incorporates Ha Tsuen; see Krone, R., “A Notice on the Sanon District” (reprinted from 1858 original), Journal of the Hong Kong Branch of the Royal Asiatic Society 7 (1976), 119Google Scholar. By 1903, however, a British colonial map notes that these pans were “disused”; see Irving and Morton, A Geography of Mai Po Marshes, 9.

104 The mansion, known as 大夫第 (House of Imperial Title Holder), is now a government-recognized historical monument. It was in a dilapidated state in 1969 but was renovated, at Hong Kong Government expense, in 1988. The building, with its honor boards (功名牌) in Manchu script, has become a major tourist attraction. See P. D. W. Bouton, The Heritage of Hong Kong (Hong Kong: Antiquities and Monuments Office, Government Printer, 1992), 60–61; Victor Kwok and Dominic Lam, Rural Architecture in Hong Kong (Hong Kong: Government Information Service, 1979), 115–17.

105 Fung Chi-ming and Elizabeth Sinn, Yuen Long Historical Relics and Monuments (Hong Kong: Yuen Long District Board, 1996), 119, mention this case in their study of New Territories oral history. There was speculation among Teng elders in the 1970s that the name for their village, Ha Tsuen 厦村, derives from a mansion (厦/廈) built by this ancestor near the Yim Chong salt yard.

106 Interviews with Teng elders, including Mr. Teng Mun-kwong (1977). See also Hase, “Notes on the History of Ha Tsuen,” 85–86, on the history of salt making (brine boiling) in nineteenth-century Ha Tsuen.

107 These reclamations were associated with the original, early twentieth-century development of Tin Shui Wai 天水圍 by a consortium of overseas Chinese—primarily from California—that acquired land rights to the marshland north of Ha Tsuen. Armando da Silva discusses this development in his excellent study of land reclamation in the New Territories, Native Management of Coastal Wetlands in Hong Kong: A Case Study of Wetland Change at Tin Shui Wai Agricultural Lot, New Territories (PhD diss., University of Hawaii, Manoa, 1977). In the 1990s and early 2020s, Tin Shui Wai was transformed into one of Hong Kong's largest “New Towns.”

108 This area had salt fields dating from the thirteenth century; see Hase, Patrick H., “Eastern Peace: Sha Tau Kok Market in 1925,” Journal of Royal Asiatic Society Hong Kong 33 (1996), 147–85Google Scholar. Although the Tou village in question did not adjoin the coast, lineage elders claimed that their ancestral hall held Qing Imperial title to the subsoil rights (地骨, lit. “earth bones”) of the salt pans in question.

109 Interviews conducted by the author in Lam Tei San Tsuen, located on the boundary between Ha Tsuen and Tuen Mun Districts, in 1978.

110 According to Hase, “Eastern Peace,” 184, salt workers along the coast in the northeast New Territories were ranked at the “very bottom of the social scale.” The best descriptions of salt making in the Pearl River Delta are found in Armando da Silva, Tai Yu Shan, 30; see also Liu Tik-sang and Cheung Siu-woo 張兆和, Tai O Island 大澳 (Hong Kong: Joint Publishing Company, 2006), 43–48. See also Shu-yen, Lin, “Salt Manufacture in Hong Kong,” Hong Kong Naturalist 10 (1940), 3438Google Scholar.

111 The Cantonese suffix lo 佬 in this context is derogatory and perhaps best translated as “fellow” or “guy.” See James L. Watson, “Funeral Specialists in Cantonese Society,” in Death Ritual in Late Imperial and Modern China, edited by James L. Watson and Evelyn S. Rawski (Berkeley: University of California Press, 1988), 117–18 on uses of the term lo 佬 in village discourse (to be distinguished from lo 老 used in the term for elder, 父老).

112 Patrick H. Hase, “Land Hunger and Emigration from the New Territories Area in the Late Nineteenth Century,” in Studies in Hong Kong History and Society: Conference Proceedings 香港的歷史與社會研究 (Chinese University of Hong Kong, June 2017), 130–53.

113 Xin'an County (Guangdong) Gazetteer 新安縣志, 1819 edition (Taibei: Ch'eng-wen Reprints), 290–94. (Page numbers in text refer to Ch'eng-wen Publishers modern pagination.) The Deep Bay saltworks were quite small in comparison to the massive enterprises located in Jiangsu Province; see, for example, Ping-ti, Ho, “The Salt Merchants of Yang-Chou: A Study of Commercial Capitalism in Eighteenth Century China,” Harvard Journal of Asiatic Studies 17 (1954), 130–68Google Scholar and Thomas A. Metzger, “The Organization and Capabilities of the Ch'ing State in the Field of Commerce: The Liang-huai Salt Monopoly, 1740–1840,” in Economic Organization in Chinese Society, edited by W. E. Willmott (Stanford: Stanford University Press, 1972).

114 Peter Y. L. Ng and Hugh Baker, New Peace County: A Chinese Gazetteer of the Hong Kong Region (Hong Kong: Oxford University Press, 1983), 47.

115 Reed mats were placed on wooden beds (smooth planks suspended on two saw-horse stands) that most village adults slept on until the 1970s—when western-style mattresses became popular. The plank-beds were “firm” in the extreme; necks were supported by cloth pillows filled with unhusked rice.

116 The water resistant Phragmites communis was an excellent material for thatching, rain cloaks, bindings, foot ware, rope, salt bags, and matting; see Irving and Morton, A Geography of Mai Po Marshes, 39. On thatched huts (茅屋) in the Pearl River Delta, see Liu Tik-sang, “Zhujiang sanjiaozhou dongyong diqu ‘weikou’ shenghou bianqian” 珠江三角洲東涌地區“圍口”生活變遷 (Social Change in a Pearl River Delta “Enclosure” System), in Congcanghai shatian dao fengchingshui xiang: zhujiang sanjiaozhou dongbu shehuishang taibianqian yanjiu 從滄海沙田到風情水鄉:珠江三角洲東涌社會生態變遷研究, edited by He Lin 何霖 and Liu Tik-sang (Beijing: Zhongguo xiju chubanshe, 2013), 13–15.

117 See, for example, Hayes, James, “Deep Bay Marshes,” Journal of the Royal Asiatic Society Hong Kong 13 (1973), 168Google Scholar (see plate 15 for a photo of hyacinth gatherers in action).

118 Reed gatherers constituted a recognized social category and existed in many parts of China, including the extensive sandbars at the mouth of the Yangzi River; see Hilary J. Beattie 1979, Land and Lineage in China: A Study of T'ung-Ch'eng County, Anhwei, in the Ming and Ch'ing Dynasties (Cambridge: Cambridge University Press, 1979), 29, 34. “Grass cutters” who lived along riverbanks in south China were often mentioned in official sources as an occupational category prone to banditry; see Antony, Robert J., “Peasants, Heroes, and Brigands: The Problems of Social Banditry in Early Nineteenth-Century South China,” Modern China 15 (1989), 130CrossRefGoogle Scholar.

119 This term is also cited in Helen Siu and Liu Zhiwei, “Lineage, Market, Pirate, and Dan: Ethnicity in the Pearl River Delta of South China,” in Empire at the Margins, 290; see also Liu Tik-sang, “Home on the Water: Livelihood and Society of the Fishing Community in Tai Po,” in Traditions and Heritage in Tai Po, edited by Liu Tik-sang (Hong Kong: Tai Po District Council, 2008), 118. Perhaps the best English rendition of 水流柴 is “flotsam and jetsam.” Another San Tin term for delta itinerants was “marsh people” (Cantonese sap-dei yahn 濕地人, lit. “damp-place [marsh] people”), used for shrimp trappers and seaside scavengers.

120 Paul Van Dyke describes large sampans that transported thousands of ducks to rice paddies along the delta “to clear them of frogs and insects”; the herd was disciplined with a long switch (The Canton Trade: Life and Enterprise on the China Coast, 1700–1845 (Hong Kong: Hong Kong University Press, 2005), 60.

121 This seems to have been true in other parts of China. Speaking of a village in Yunnan, Fei Hsiao-t'ung and Chang Chih-li observe: “[Duck raising] is supposed to be a somewhat discreditable occupation, carried on only by the poor … Only those who do not care about face can engage in this occupation, it is said”; see their classic study, Earthbound China: A Field Study of Rural Economy in Yunnan (London: Routledge and Kegan Paul, 1948), 236. C. K. Yang likewise did not have good things to say about duck herding; see A Chinese Village in Early Communist Transition (Boston: M.I.T. Press, 1959), 65–66.

122 On chicken raising in Cantonese villages, see Liu Tik-sang, “Custom, Taste and Science: Raising Chickens in the Pearl River Delta Region, South China,” Anthropology and Medicine 15 (2008), 7–18.

123 In 1969 duck herders in San Tin District were paid 15 percent of the sale price.

124 A system referred to in the anthropological literature as patrilocal residence; see Watson, Rubie S., “Class Differences and Affinal Relations in South China,” Man 16 (1981), 593615CrossRefGoogle Scholar.

125 James Hase, “Chinese Customary Law: Family Cases from Shek Pik, Lantau, New Territories of Hong Kong,” Journal of the Royal Asiatic Society Hong Kong 57 (2017), 212. In his study of social discrimination in late imperial China, Anders Hansson concludes that “Dan people rarely intermarried with families of land[based] people.” See Anders Hansson, Chinese Outcasts: Discrimination and Emancipation in Late Imperial China (Leiden: Brill, 1996), 122.

126 The following public notice was posted in San Tin's market in 1977: “All residents should watch for new graves in our district. Only those graves authorized by the San Tin Village patrol are permitted in our territory. If you see any strange graves that do not belong to local people please report this to us immediately. Signed, Man—-, Trustee of Founding Ancestor's Estate.” On burial restrictions in New Territories lineage territory, see Chan Kwok-shing, “Hillside Burials: Indigenous Rights in the New Territories of Hong Kong,” Anthropology Today 19.6 (2003): 7–9.

127 Retired village patrolmen (巡丁) in Ha Tsuen reported (during 1978 interviews) that they did not do the actual exhumation work themselves; this was contracted-out to funeral specialists (仵/忤葬佬 or, alternatively, 仵作佬) associated with coffin shops in the nearby market town of Yuen Long (discussed in J. Watson, “Funeral Specialists,” 126). In the late 1970s patrolmen began reporting rogue burials to the Hong Kong Police; henceforth government authorities dealt with disposal (by cremation rather than dumping).

128 Eugene Anderson notes that fisherpeople who resided in Castle Peak Bay (12 miles south of Laufaushan) also buried their dead on delta islands, until the mid-twentieth century when the Hong Kong Government set aside a special cemetery for this purpose; see Anderson, “Floating World,” 195 and personal communication, June 22, 2020. Laufaushan did not have a cemetery of this type, primarily because it was not a harbor (offering protection from typhoons) and thus had no permanent “floating village” of the type studied by Anderson.

129 Based on interviews with Teng water patrolmen, 1978. Most of the Laufaushan oyster workers retired to Shajing in their late forties and fifties; only a handful died in the New Territories.

130 See for example, FC021/1420, file no. FEH 18/1, Part D, 1975, Public Records Office, London, an account of four dead young people found along the Laufaushan coast.

131 The colloquial Cantonese for this phrase is Kuih-deih mouh likh-sih 佢哋冇歷史, literally: “They do not have [any] history.” Teng and Man elders sometimes used it as a term of dismissal during my four decades of active field research. For parallels in European contexts, see the classic study by Eric R. Wolf, Europe and the People Without History (Berkeley: University of California Press, 1982).

132 Oyster workers represent an interesting variation on this theme: They did “have history” in their home district of Shajing, where they contributed to the construction of a major Tianhou 天后 temple and a museum celebrating the trials and tribulations of their trade (photographs and notes on Shajing, courtesy of Denise Ho). The Chan lineage of oyster specialists also maintained a written genealogy in their home district of Shajing (Liu Tik-sang, personal communication). In the New Territories, however, the only visible evidence of oyster workers’ contributions to local institutions is a large wooden model of an oyster boat presented by the local Oyster Union 蠔業聯會 to the local Tianhou temple on the coast near Laufaushan. Members of this organization also participated in annual flower-cannon (花炮會) competition at this temple.

133 This Cantonese phrase (ngoh-deih-ga lihk-si, 我哋嘅歷史) was always pronounced with deliberation and solemnity. Elders who used it in public pronouncements never failed to command attention.

134 The increase of fresh water being discharged into the Pearl River during the 1970s also led to a lowering of salinity in Deep Bay, seriously affecting oyster development; see Silvia Chang, “Saving Hong Kong Oysters from Disaster,” China Daily.com, August 4, 2016.

135 For a description of deep-water oyster production see Lee, Miriam and Cheung, Sidney, “The World is Your Oyster,” Hong Kong Discovery 98 (2017), 1443Google Scholar. New Territories residents still remember the pollution scares of the 1980s and few will eat oysters raised in Hong Kong waters.

136Desakotas [a term derived from the Bahasa Indonesian for ‘village and town’] are transformed areas that are no longer clearly urban or rural … but a blending of the two”; Gregory E. Guldin, “Desakotas and Beyond: Urbanization in Southern China,” Ethnology 35 (1996), 265. On the despoliation of the New Territories, see Malcolm Merry, The Unruly New Territories (Hong Kong: Hong Kong University Press, 2020), 269–70.

137 The container “mountain” (山), as one elder called it, blocked the good geomantic influences (風水) that are said to account for the success of the Teng lineage during the preceding seven centuries.

138 See, for example, Chi-kwong Law, A Study of Tin Shui Wai New Town: Final Report (Hong Kong: Department of Social Work and Social Administration, University of Hong Kong, 2009).

139 Discussed in J. Watson, “Saltwater Margin,” 282.

140 See Jeffrey Twu, “One Country, Two Systems Is Full of Contradictions: Just Look at the Hong Kong China Border,” Quartz, June 27, 2017, qz.com/1012174. (2015 is the last year when immigration figures for Lok Ma Chau station were posted on-line and available to the public.)

141 Watson, James L., “Virtual Kinship, Real Estate, and Diaspora Formation: The Man Lineage Revisited,” Journal of Asian Studies 63 (2004), 893910CrossRefGoogle Scholar.

142 Starting in 2010, Hong Kong's Grayline Bus Company has offered a “Hong Kong Wetlands Delight Tour,” featuring Laufaushan and the surrounding countryside.

Figure 0

MAP 1 Pearl River Delta and Hong Kong New Territories, 1898–1978 (Colonial domain shaded)

Figure 1

MAP 2 Deep Bay and Yuen Long District, 1969–1978

Figure 2

Map 3 Oyster Territory in Deep Bay, British Colonial Depictions, 1904

Figure 3

Figure 1 Oyster Harvesting Permit, 1977, Water Patrol, Laufaushan (Photo by J. L. Watson©)

Figure 4

Map 4 Tube-Net Fishing Sites, Laufaushan Coast, ca. 1938