The Archaeology of Hong Kong

The Archaeology ofHong Kong

Hong Kong University Press thanks Xu Bing for"而 iting the Press's name in his Square Word Calligraphy [or Lhεcovers o[ iLS books. For [unher in[orma Li on , see p. iv

Th能 Arch說eolo勸r of 詞。直至gKo豆豆露

William Meacham

香港式學出版社

HONG KONG UNIVERSITY PRESS

Hong Kong University Press

141F Hing Wai Centre 7 Tin Wan Praya Road Aberdeen Hong Kong

<9 William Meacham 2009

Hardback ISBN 978-962-209-924-1 Paperback ISBN

978-962-209月 925-8

All rights reserved. No portion of this publication may be reproduced or transmitted in any form or by any means , electronic or mechanic泣， including photocopy, recording , or any information storage or retrieval syste風 without

permission in writing from the publisher

Secure On-line Ordering http://\叭叭成 hkupress.org

British Library Cataloguing吐吐-Publication Data A catalogue copy for this book is available from the British Library

Printed and bound by Li ang Yu Printing

品

FactorγCo.

Ltd. in Hong Kong , China

Hong Kong University Press is honoured that Xu Bin且， whose art explores the complex themes of language across cultures , has written the Press's name in his Square Word Calligraph芋 This signals our commitment to cross-cultural thinking and the distinctive nature of our English-language books published in Chin且 “Al 且 rsl glance , Square Word Calligraphy appears lo be nOlhing more unusual than Chinese characte時， but in fact it is a new way of rendering English words in the format of a square so they resemble Chinese characters Chinese viewers expecl lo be able Lo read Square Word Calligraphy bm canno t. Western viewe呵， however are surprised to fìnd they can read it Del阱t erupts when meani月 is unexpectedly revealed." Britta Erickson , The Art of Xu Bing

Contεnts

Prεface

Vll

Abbreviations and Usage

lX

Main Dynastic Periods

Xl

Archaeological

Sitεs

in Hong Kong

Xll

E且司

Part 1 Archaεology in Hong Kong

Jnu

Early Discov叮les

3.

Fr. Finn and

4.

A Stratifìεd Site at Shek Pik

5.

Hong Kong and Hoifung

6.

The Fate of thεEarly Collections

7.

Post-war Developments , 1945-78

8.

Sham Wan , Lamma: A NεwEra

9.

Post-Sham Wan. 1978-90

thεTai

Wan Site

八U411ι

斗 55

『/

SalvagεArchaeology，

Bεgms

八U

2.

10.

thεPast

戶j 、1ifO

7,000 YIεars into

lIZ233

1.

1990-2007

Part 11 Prehistory of thεHong Kong Region

67

11.

The Palaeolithic: Early Humans in South China

69

12.

ThεAncient

74

Environment and Human

Movεmεnts

vi

Content

Earliεst

13.

Hong Kong's

14.

Middle Neolithic: The First lnhabitants

86

15.

Middle Neolithic Burial Sites on Chek Lap Kok

92

16.

A Surprisingly Early Phase of the

LatεNeolithic

95

17.

Classic Late Neolithi c: Culmination of the Stone Age 103

18.

Bronze Age: On the Margins of Civilization?

19.

Two

20.

The Rock Carvings

21.

A Missing Link in Local

BronzεAge

Site

81

Burial Sites

Prεhistory

22

ThεHistorical

23.

A Han Tomb at Lei Cheng Uk

24.

Hong

25

ThεSung

26.

A UniquεKiln Complex

27.

Later Historical Archaεology

Yueh

lndustry

137

143 149 -Ei EI 可

7. 勻且 戶J 「 rhu

Era

169

179

Present Knowledge and Future Prospects

Appεndix﹒ ThεUnivεrsity

130

139

Part IV Summary 28.

117 123

Part III Historical Archaeology

Kong這First

110

of Hong Kong and

181 189

Local Archaeology Rεferεnces

197

lndεx

201

Prεfacε

This book is a m句 or revision and expansion of Archaωlogy in Hong Kong (1 980). It is intended as a broad overview of the data from Hong Kong archaeology, with reference to the larger region of China and Southeast Asia. 1 have retained the basic division of the book into two main parts: the fì.rst ten chapters describe the progress of archaeology from the early pioneer work in the 1920s up to the present Quly 2008) , while the remaining chapters review the evidence from the Palaeolithic to the historical periods. A huge amount of new data has been obtained in the years since 1980 , and the text has been extensively rewritten to provide a current summa可 and synthesis of local archaeology The new information has naturally required reconsideration of theories and interpretations , and some ideas proposed in the earlier edition have been dropped. Many others have been amply borne out by the new data , and , of cour哎， there remain numerous grey areas where healthy disagreement contmues The last 16 years in particular have witnessed a geometric expansion of archaeological data , as large-scale excavations lasting several months have become common and full-time professional contract archaeologists have established practice in the territory This intensive pace of fì.eldwork and research has yielded a wealth of information on the material culture of the early inhabitants of the area. It would require severallarge monographs to properly describe the sites , features , potte可 and stone tools that have been recovered in just the last ten years and another large volume to fully describe the entire history of Hong Kong archaeology My aim in this book is simply to present a very general and concise review of the most salient data that will serve as an introduction to the su旬的

viii

Preface

lnevitably in a book of this nature , the emphasis on certain issues and the selection of examples reflects the personal views and experiences of the author, including a number of anecdotes about some of the main fìgures in local archaeology since 1970.1 have also included and commented on certain controversial aspects of local archaeology Other people might well have a different view，的 well as whether or not to comment at all. Personal views are clearly labelled as such , and it is my belief that these help to pro\吋e the reader with a more realistic overall impression of local archaeology 1 am grateful to colleagues in the Archaeological Socie旬; the Antiquities and Monuments Offìce and the universities who offered comments and constructive criticism of the earlier book and assistance in preparing this one.

Abbreviations and Usagε

Antiquities Office

cm kg

The Antiquities and Monuments Office of the Hong Kong Government The Hong Kong Archaeological Society All carbon dates are calibrated to calendar years BC or AD. centlmetre kilogram

立1

虹letre

mPD

Metres above Principal Datum. Lowest annual tides are about +0.2 mPD , highest around +2.7 mPD , and mean sea level +1. 2 mPD. Storm beaches generally are found at +2.5 to +斗 mPD and most archaeological sand bar sites between +斗 and +11 mPD. In standard archaeological writing , Neolithic refers to the period and neolithic

Archaeological Society C-14 dates

N eolithic/neolithic

Tang , Sung , etc

T-L

to the culture , stone tools , etc The Tang , Sung and other dynastic periods of Chinese history (see chronology below) Thermoluminescence dating of pottery or soils

Main Dynastic

Pεriods

Hsia (Xia) Clegendary) 2205-1766 BC Shang 1766-1121 BC Chou (Zhou) 1121-255 BC 訊社 stern Chou 1121-771 BC Spring and Autumn Period 770-255 BC 480-222 BC Warring States Period Ch'in (Qin) Han Chin (J in) Six Dynasties Sui Tang Sung (Song) Yuan Ming Ch'ing (Qing)

221-207 BC 206 BC-AD 220 AD 265-420 AD 420-589 AD 589-618 AD 618-907 AD 960-1279 AD 1271-1368 AD 1368-1644 AD 1644-1911

盾已〉意 @ [ßb叮

List of Sites 1.

2.

3 斗，

5. 6. 7. 8. 9 10. 11 12 13

14. 15. 16 17 18 19 20. 21. 22 23 24 25 26. 27 28 29 30 3 1. 32. 33 34 35

Big Wave Bay Cheung Chau Chung Hom Wan Diamond Hill Fan Lau Fan Ling Fat Tong Mun Fu Tei (Fu Tei Wan) Ha LawWan Ha Tsuen Hai Dei Wan High Island (see Sha Tsui Tau) Ho Chung Hung Shing Yee Island House Junk Island Kau Sai Chau Han site Kau Sai Chau rock carving Kau Sai Chau Welch site Kellett Island Kowloon Walled City Kwo Lo Wan Lei Cheng Uk Lo So Shing LungHa Wan Lung Kwu Sheung Tan Lung Kwu Tan Ma Wan (see Tung Wan Tsai) Mai PO Man Kok Tsui Mo Tat Wan Mong Tseng Wai Ngau Hom Shek Nim Shu Wan North Lamma School site Pa Tau Ku Pak Mong

36 37 38 39 斗 O. 斗1.

42 43 44 斗5

斗 6. 斗 7.

48 49 50 51 52. 53. 54 55 56 57 58. 59. 60 61 62 63 64. 65. 66 67 68 69 70. 71. 72

Peng Chau Penny's Bay PO Toi PO Yue Wan (sand bar and headland) Pui 0 Sai Wan Sek Kong Sha Chau Sha Ha Sha Lo Wan (sand bar and headland) Sha PO Tsuen Sha Tau Kok Sha Tsui Tau Sham Wan Sham Wan Tsuen Shek Kok Tsui Shek Pik Shek Pik upper carving Siu A Chau So Kun Wat Tai A Chau Tai Kwai Wan Tai Long (Tai Long Wan) Tai Wan Tit Sha Long Tung Chung fort Tung Chung rock carving TungKwu Tung Lung fort Tung Lung rock carving Tung Wan Tsai Wong Chuk Hang 1九Tong Tei Tung Wun Yiu Yi Long Yim Tin Tsai Yung Long Yung Shu Wan (see North Lamma School site)

Part 1 Archaeology in Hong Kong

7 ,000 yiεars into

thεPast

The history of Hong Kong as a port , British colony and city is generally well recorded. When it was ceded to Britain in 1841 , the island was described in the famous (and now highly ironic) phrase as "a rock with hardly a house upon it." The neighbouring islands supported a few scattered fishing villag白， while Kowloon peninsula and the New Territories to the north were settled by rice farmers of several large and many small clans It was not until the early twentieth century that scholars begin to examine the pre-British period of Hong Kong's history. Reliable documentary evidence extends back to the last years of the Ming dynasty (early seventeenth century). Genealogies of the m句 or New Territori白， clans indicate their movement into the territorγduring the Ming and Sung periods (tenth to sixteenth centuries). A few tantalizing earlier references exist that may correlate with places in Hong Kong and describe salt industries and pearl fisheries that may have been practised here after the area came under Chinese rule in the second century BC. Archaeological investigation began in the 1920s and showed that the territory had a much longer span of human occupation , now known to extend back at least 7000 years. Sites abound on outlying islands and along the coastline of the New Territories. More than 200 sites of the New Stone Age (Neolithic) and Bronze Age have been recorded , and many have been systematically surveyed and excavated. The results of this research have been

1.1 Excavatíng anímal bones and shells at a síte on Lamma Island

4

Archaeology in Hong Kong

1.2

Settlement on the tombolo island 01 Cheung Chau began several thousand years ago

well published in recent decad白， and scientific studies of excavated materials have thrown much light on prehistoric life in the area A large brick chamber tomb of the Han period (206 BC - AD 220) , discovered in 1955 during construction of a public housing estate , marks the beginning of the historical era. Much new data has been obtained on the historical period as well As archaeological work has advanced over the years and techniques have improved , the earliest date of human occupation in the Hong Kong region has been pushed far back from the Ch'in-Han expansion recorded in historical texts , and from the 1000 BC estimate of one pre-war archaeologis t. Recent radiocarbon dates on charcoal samples from one local site has confirmed that humans were present in this area of the South China coast by at least 5000 BC. Without caves to preserve even earlier sites , it is quite possible that the lower limit of archaeologically detectable human penetration of the area has been reached. Open sites of the Palaeolithic may likely exist under the seabed , and it would be an extraordinary circumstance if one were to be discovered

7 ,000 Years into the Past

In addition to the dramatic extension of Hong Kong's prehistory back several thousands of years , fìeldwork in recent decades suggests that there was a general continuity of occupation and population from the earliest period down to recent historical times. With the discovery of several well-stratifìed prehistoric sites and with long sequences of occupation and dozens of early historical sites , almost all of the major gaps in Hong Kong's 7 ,OOO-year human past have been closed. There remains one m月j or "missing link" (the Late Bronze Age and Early lron Age) and , of course , many, many unanswered questions about the prehistoric inhabitants , for example , • What fìrst brought humans into this area?

1.3

A fine抄 chipped stone axehead from Chek Lap Kok

1.4 Th e typical Hong Kong sand bar site: on a sandy beach with low-lying valley (formerly lagoon) behind.

5

6

Archaeology in Hong Kong

•

•

•

Who were the early inhabitants , where did they come from , and what trade and cultural ties did they have with a其jacent regions? What kind of life did they lead , and how did their culture and economic system change over the centuries? Which changes were of local or regional orig凹， and which were related to ideas or people coming from distant areas of development and civilization?

Excavation in progress at Sha PO Tsuen , Lamma

.... 1.5

Deep trench revealing more than 7, 000 years 01 deposit at Chung Hom 札旬n on Hong Kong island

... 1.6

7 ,000 Years 凶。 lhe PaSl

1.8 1.7

Slotted polished stone ring

Cord-marhed pot 砂plωI of the local Neolithíc

1.9 1, 500-yearold sheleton at Sha PO Tsuen , Lamma

1.10 Stratigraphy of deposits dating bach 3 , 000 years at a site near Tuen Mun

7

8

Archaeology in Hong Kong

Archaeologists in the past have paid much , often too much , attention to the last of these questions. Naturally, considerable energy has been devoted to the question of northem Chinese influenc亡， which has often been seen as the principal force for change and progress acting upon an othen九rise backward and stagnant tropical population The Han historian Ssu Ma-ch 'ien and many Chinese writers after him described the indigenous peoples of the Yangtze River Basin and southeast coast as disease-ridden , indolent barbarians , subsisting in the most miserable fashion on readily available tropical fruits and the most primitive type ofhorticulture. Among archaeologists , it was once believed that , during the prehistoric period , South China w倡 “ rather quiet and had a simple culture." Discoveries of early Chinese civilization in the Central Plains ofNorth China reinforced traditional notions that “ culture only travels south" (Watt 1971) and that material progress in agricultur亡， bronze metallurgy, art , etc. must be derived ultimately from the crucible of Chinese civilization. The relative paucity of archaeological work in South China tended for many years to corroborate such notions , since it appeared that these regions were “ sparsely populat吐， if at all" by primitive hunter-gatherers until the arrival of neolithic farmers from the north This world view began to change in the 1960s when important discoveries of early agriculture and bronze in Southeast Asia were made. In the 1970s , at a site near Shanghai , the discovery of a well-developed neolithic culture and rice cultivation (as early as 5000 BC) laid to rest forever the notions that the coastal areas of Southeast China had been a backwater in relation to the Central Plains of North China The evidence from Hong Kong archaeology provides an important part of the reconstruction of prehistoric life in the southem coastal regio凹， which , in tum , constitutes an important ecological and adaptational zone in the broader picture of East Asia. And while great movements of people or culture across the continental land mass are rarely postulated today, it remains important to examine the possible inter-regionallinks through trad亡， diffusion and small-scale mlgratlO n The detailed and relatively intensive archaeological work that has been done in the Hong Kong archipelago makes it probably the best-studied area of the southeast coas t. It is also an ideallaboratory to test new ideas and

7 ,000 Years into the Past

that were formerly assumed to have been imported from the north are now considered to have resulted from a widespread technological evolution , with little or perhaps no outside stimulus. It is certainly true that much has been accomplished in the last 80 years and that the questions now being discussed are the right ones. It is undoubtedly true that much more exciting data still awaits discovery，的 large-scale engineering projects transform the face of Hong Kong in the twenty-first century But it is equally and depressingly true that the full answers to the questions set out above will forever elude us. Such is , alas , the nature of archaeology

9

Early Discovεrles

The story of archaeology in Hong Kong begins one day in 1925 or 1926 when C. M. Heanley was hiking in the New Territories , in the vicinity of Castle Peak. Heanley was head of the government's Vaccine and Bacteriological Department but spent much ofhis leisure investigating the geology of Hong Kong. As he wrote later (Heanley 1928) A few years ago the writer was examining the outcrop of a quartz vein for any material of intere哎， and placed in his pocket a loose piece of rock because it contained needle-shaped miner祉， with the intention of breaking it at a more convenient time; later when balancing it in his hand to roughly estimate its specifìc gravity he was struck by its peculiarly smooth feel and on further examination noticed that it had a cutting edge and two shoulders It was immediately apparent to him that they were artefacts that

had been polished into a shape suitable for use as axe heads known as "adzes." His colleagues were less impressed; one remarked "you archaeologists have some imagination!" Others asked "how can you tell it from a water-worn stone 7" B叫， in fact , Heanley had recognized the fìrst prehistoric artefact to be found in Hong Kong Since he was undoubtedly aware of the dramatic archaeological discoveries being made in N orth Chi間， Heanley recognized the importance that these artefacts might have in providing information about the Stone Age south of the Yangtze River. On subsequent fìeld tn阱， many more similar artefacts were collected , and notes were taken on the locales that seemed to produce them. By 1928 , Heanley had gathered enough infonnation to publish a brief article on "Hong Kong Celts" in the Bulletin of the Geological Socie矽 of China. He was soon j oined in his research by Professor J. L. Shellshear of the Department of Anatomy at the University of Hong Kong. The vigorous pursuit of Hong Kong's prehistoric inhabitants had well and truly begun!

Early Discoveries

2.1

Stone adzesfound by Heanley in 1926-28

In the next few years , the new field of study grew rapidly Heanley and Shellshear located dozens of archaeological sites , noting especially the crests and spurs of eroding granite hills near the sea as the most productive areas. They were said to be “ superblyenergetic and covered tremendous distances in a day at great speed. Only fit and enthusiastic walkers could hope to last a whole day with them." Heanley sometimes found fragments of pottery associated with the stone tools , and the discoverγof such pottery alone was taken to be evidence of a prehistoric 乳白， which was then investigated more intensely for stone tools. It was estimated by Heanley that on granite outcrops in Hong Kong , an average of 30-40 adzes could be found per square mile within 600 yards of the coas t. No sites further inland were located , and the early occupation pattern seemed to have been confined to areas near the sea , usually around bays or at the mouths of small valleys

11

12

Archaeology in Hong Kong

As the surface prospecting continued , the variety and quality of pottery and polished stone tool industries that had flourished in Hong Kong's prehistoric periods soon became apparen t. There were several varieties of coar司se and fìne pottery; beautifully polished stone adzes , fragments of stone knives and rings , quartz discs and beads were also recovered The fìrst archaeological site in Hong Kong to be studied and reported was at 50 Kun Wat near Castle Peak in the area where the fìrst fìnds had been made. No excavation was undertaken at the site; indeed , there was little need to do so. Erosion had brought to the surface an extraordinary number of artefacts , mainly on the slopes of a small rounded hill , situated in a valley some 400 m from the sea The pottery and stone debris extended over an area of more than 100 m 2 lncluded among the fìnds were several complete adzes and rings; however, most of the artefacts were either broken , unfìnished implements or waste flakes created in the manufacture of the tools 50 great was the quantity of what Heanley called “ factory refuse" that the site was deemed to be a stone workshop; the presence of potte可 also indicated some cooking activities as well. Before the sand banks began to yield their wealth of material , Heanley believed that prehistoric habitation was predominately centred on hillside locales , such as 50 Kun Wa t. It was “ natural" erosion that brought to light the objects that Heanley and 5hellshear collected. This erosion of Hong Kong's hillslopes had probably begun in the T' ang and 5ung eras when deforestation had increased slgm且cantly in the area However, the attention of Heanley and 5hellshear was soon to focus on the beach deposits that were being dislodged directly by human agency ln the 1920s , Hong Kong was prospering , and there was a “ boom" in construction. Commercial sand companies were ranging out to the more remote islands in their search for exploitable sand deposits Wh at 5hellshear saw when he visited one such area must have been a marvellous - and at the same time horrifying - sight , even to someone accustomed to fìnding archaeological remains in the state of nature. The sand had been screened before being loaded on a junk. 5tones , large fragments of potter只 stone tools and bronze objects had been dumped as unwanted debris on either side of the area of digging. 5hellshear made several visits to the site Cprobably Tai Wan on Lamma); a nicely decorated sword , and an axe head and a dagger axe were amo月 t

Early Discoveries

deposited in the British Museum. The sword is a unique 五nd ， the only such specimen known to have been recovered in Hong Kong. Fortunatel)月 in post-war years , sand operations in Hong Kong moved offshore , but such is not the case in at least some areas of the coas t. In 1978 , 1 had the opportunity to observe prehistoric pottery in beach sand piles on several construction sites in Macau. The sand had been supplied by a raw-materials wholesaler in the Mainland Heanley and Shellshear both tentatively assigned the stone tools and associated pottery to the Neolith眩， although no estimate was made of a more precise date. It was cle缸， however , that the prehistory of South China had ended with the Han dynasty expansion to the south in 120 BC and that the local Neolithic and Bronze Age had begun somewhat prior to that time.

VJ hF HmH -u ，λunM

jF

們川 R

Profíεssor]. L.

川 m kJ 心

p、Jjι

TH

附m

PU

例

-yγf

Z

戶、 J

... 2.3

c3

口UL川

74

2E nu r IH

Shellshear

Heanley left Hong Kong in 1930 and handed over his collection to Shellshear. He also contributed to the draft of a much longer description of the territory's archaeology Their paper “A Contribution to the Prehistory of Hong Kong and the New Territories" was presented by Shellshear at the First Congress of Prehistorians of the Far East , held in Hanoi in 1932 (Heanley and Shellshear 1932) They described in detail their findings and made the first tentative

13

14

Archaeology in Hong Kong

2.4

Shellshear (second from righ t) at meeting of Prehistoηans of the Far East , in Hanoi , 1932

speculations about the relationship of this material with archaeological discoveries in China. Shellshear carried on the work for a few more years and made another great contribution to local archaeology by stimulating an ir也rest in the subject among othe芯， particularly Walter Schofield and Fr. Daniel]. Finn , whose investigations took up where he and Heanley left off

Fr. Finn and

thεTai

Wan

Sitε

After the discovery of polished stone tools near Castle Peak in 1926 ,

interest in the archaeology of Hong Kong grew rapidly. By the 1930s , dozens of sites were knoWll, mainly on lower hillslopes near the coast or in sand dune deposits behind beaches. Heanley and Shellshear made their fìrst discoveries on eroding hillsid白， until , as Shellshear wrote in a letter many years later,“we discovered the richness of the sand banks." As mentioned above , Hong Kong , at the time , was experiencing a rapid development in constructio立， and commercial sand diggers were exploiting these same sand deposits. Without doubt , many valuable sites were destroyed in this manner, but , as fate would have it , probably the most important Bronze Age site in Hong Kong was at least partially salvaged and recorded through the efforts of Fr. Daniel J. Finn Finn had been interested in the early discoveries and was invited in 1932 by Shellshear to become more active in the fìeldwork.“Afew days later," Finn recalled , while 1 was still regarding any active involvement as remote , luck seemed to confìrm the vocation. As 1 walked past a sand heap on a jetty in Aberdeen , 1 almost crushed underfoot a large fragment of prehistoric pottery Returning to the spot the next day, Finn recovered from the sand a piece of bronz亡， a stone spearhead , and potte可 fragments with stamped patterns. Years later, after much fìeldwork and research , Finn would remark that “ the connection between those fìrst fìnds remains the chief question for solution."

16

Archaeology in Hong Kong

\月是1n

3.1

The Tai

site in 1934

3.2

Excavations at Tai Wan , Lamma , 1934

17

Fr. Finn and the Tai Wan Site

Making inquiries at the jetty, Finn was informed that the sand had come from a beach on Lamma Is1and near the village ofYung Shu Wan. The site had been known to Shellshear, who had “reaped a good harvest" of bronze and stone artefacts the兒. Finn made severa1 visits to the si仗， called Tai Wan , and was struck by the great destruction of materia1 and 10ss of information that was taking p1ace. On one such visit , before Finn cou1d say a word , a woman worker pulled a soft pot from the sand and threw it asid亡， whereupon the vesse1 was smashed to pieces. Over the next few weeks , Finn quick1y organized a sa1vage excavation. In 1933-34 , government funds were made availab1e and a full-sca1e excavation was conducted This was the first excavation of an archaeo1ogica1 site to be conducted in Hong Kong , and the methods emp10yed were , quite predictab1其 rather primitive. Neverthe1ess , some information was obtained from the sit亡，的 well as an impressive collection of bronze weapons , pottery, stone weapons and too1s. A portion of this collection has survived the decades and is held by the Hong Kong Museum of History and the Fung Ping Shan Museu m.

WTU句

FS

ρζ

h

司aAP 心

t

心

的心 fetf

主

汁

們。-

Y'9

ou

EWC

G-GM 叭 m

一

可SL

'H

內毛一'的川叫{ nlp 仙

nDP-ob

。

mM'的

勻有 JAUtMHe

叫 戶戶的vh

Z

, dhgH mhmMM rnulJGHt

18

Archaeology in Hong Kong

The number of stone and bronze weapons was so great as to suggest to Finn that the site had been the scene of a battle between Han Chinese troops and southern Yueh “barbarians." He cited historical texts that mention a naval attack as part of the campaign waged against the Yueh in 120-111 BC. A less fanciful view would be that the site was used as a burial ground , and this interpretation is strengthened by the presence of complete pottery vessels and polished stone ornaments associated with the bronzes. But the large number of metal items at the site is still diffìcult to explain in light of more recently excavated burial sites of the same period that yielded only one or two small artefacts of bronze Most of the bronzes were of types quite common in South China during the Shang and Chou dynasties in the north (around 1600-256 BC) , and it now seems probable that the main occupation of the site occurred around 1000 BC. Other Bronze Age sites on Lamma Island with similar types of pottery have recently been dated by the C-14 method to around 1200-800 BC. Finn also noted that softer pottery and stone tools were found at generally deeper levels than was bron丘， and he proposed that an earlier occupation of the site had taken place towards the end of the Neolithi c. Salvage excavations conducted in 1979 and 1991 established that Tai Wan was inhabited even earlier, in the Middle Neolithic phase , dating to around 4000-3000 BC. One of Finn's most intriguing discoveries at T:扎扎1an was a type of Bronze Age pottery with an elaborate spiral design of much greater complexity than the rather simple geometric patterns found on most of the pot悅耳 Finn made a special contribution to local archaeology in the naming of this extraordinary design. “ From a desire not to prejudice the interpretation , 1 have chosen to call itιdouble-F'." The design does seem to resemble two F's back-to-back with one of them upside down and reversed. Chinese art historians have assigned the design to the “Kuei-dragon" class , a vaguely similar pattern found on Chou period bronze vessels. However, the direct deriv日tion of double-F from Kuei has not been established , and both the origin and the symbolism (if any) of the double-F design remain unknown. Finn himself speculated that the pattern represented a sea dragon or crocodile , since the Yueh people were seafarers. Others have suggested quite simply that the design is evolved from the spiral decoration of Late Neolithic potte ry, and may not have any dragon connotatlO ns a

Fr. Finn and the Tai Wan Site

3.4

Bronze Age pottery from the Tai Wan site

3.5

油門eties

of the double-F pattern

19

20

Archaeology in Hong Kong

ln 1933-36 , Finn published the results of his excavation and research in a series of articles in The Hong Kong Naturalist . These were republished in book form in 1958 as Archaeological Finds on Lamma Island near Hong Kong. Shortly before his death , Finn attended the Second Congress of Prehistorians of the Far East , held in Manila in 1936 , and presented a paper there describing his Tai Wan material He wrote extensively about the possible origins of the people and culture in evidence at Tai Wan; he tri吐， without much success , to relate various artefact types to the “ three great waves of migration" that some scholars then believed to have swept through this area. Much of his writing is thus out-of-date today, and his site report reflected the faults in his digging techniques We will , nonetheless , be forever in Finn's de恤， for he left us an exhaustive description of the material he found , and he stimulated a great deal of public and government interest in the archaeology of Hong Kong. lt is both ironical and appropriate that the most distinctive potte可 style in this area - discover吐， studied and named by Fr. Finn - should bear his initials.

A Stratifiεd

Sitεat Shεk

Pik

Shek Pik on Lantau lsland is one of the richest and most important archaeological sites in Hong Kong. Bya fortunate set of circumstances , it came to be investigated by Walter Scho fì. eld , a district offì. cer and amateur geologist who took up archaeology with a passion and a methodical approach From 1933 to 1937 , Scho fì. eld discovered or surveyed more than 100 prehistoric sites ,“mainly in the sandbars of the islands and coasts of the New Territories." Some of these sites , notably the dumb-bell shaped island (tombolos) of Tung Kwu and Siu A Chau , did not yield bronze artefacts but yielded many polished stone tools and a softer pottery than the well- fì. red Bronze Age type. From this data , Scho fì. eld concluded that an earlier,“ Neolithic" peopl亡， living mainly on the sea , had preceded the Bronze Age inhabitants of the sites on larger islands , such as at Tai Wan on Lamma , excavated by Finn in 1933 However , Schofield's major contribution to Hong Kong archaeology was in his careful excavation and exhaustive report on Shek Pik. The site was discovered in 1937 in the same manner as many other major sites along the coast - erosion and sand digging had cut back the sand bank, revealing a rich cultural deposi t. The excavation showed that a lengthy occupation of the site had taken place from the Late Neolithic period to the Bronze A詐， An ancient floor littered with broken pottery and other debris was found some 60-80 cm below the surface of the sand bank. He also noted earlier material in the levels below this floor , to depths of 180 cm. Well below the main activity floor , Scho且εld found the remains of six individuals , some of whom had been buried with funeral offerings of pottery, stone tools , shells and animal bones - the latter items probably serving as food and\or ornamental sacri fì. ces

4.1

玩1alter Schofield at Tung Kwu

in 1937

22

Archaeology in Hong Kong

OJ

句J/

旬，i

n KWU I n

勻 J

址

fjTi

rtL

σb

τd

1 dn n 、υ

月U

4.3

TJI t nu m'pnul ρι

,

弓4

4

The Shek Pik site prior to excavation

All of the burials were oriented with head to south , a fact that suggested to Schofìeld the existence of some primitive notions of feng shui (geomancy). It is also noteworthy that one individual was represented by skull fragments only, while another skull had a neat hole drilled into it. (Both of these features were also seen more recently at another prehistoric burial site at Sham Wan. Lamma Island discussed below). From the relationship of the burials to the old living surface , and from the type of objects placed in the burials. Schofìeld was able to date them to an early phase of the Bronze Age , which he put fairly accurately at around 1500-1000 BC Schofìeld presented a very detailed report on his excavations at Shek Pik to

23

A Stratifìed Site at Shek Pik

l

pinvelnt

c、 d

。

p-ν

，但

芯， R

r

e

白

、PL

升

，但

你JDA

5

4

4a HU

4.5

t t3 mm PK EoE5

Bronze λge pott的 from 51叫

d n JU

24

Archaeology in Hong Kong

ιpd

γ'

。、

，

d

pu

L? 心

刊，仙 d

6 甘心仰 4ιιHJ

dvp lvn4M nbgh tz: ST)5 md I'L

tr nfJ I l

the Third Congress of Prehistorians of the Far East , held in Singapore in 1938. This was a monographic description of the excavation , the material and Schofield's interpretations. It was republished by the Hong Kong Archaeological Society in 1975 , along with all of Schofield's published archaeological writings ，的 well as some unpublished work CScho且eld 1975). After his death in 1968 , his papers , including field notebooks , original photographs of Shek Pik and other sites were given by his widow to James Hay白， a local historian , who later donated them to the Archaeological Society Because of Schofield's meticulous work. the site at Shek Pik is important for a number of reasons: The excavation revealed clear str atigraphy with various periods of activity and different material cultures represented at different depths The site was the first in Hong Kong to be established beyond reasonable doubt as undisturbed since the time it was abandoned. Schofield concluded that it had thus been protected from typhoons , storms , erosion and redeposition over a period of 300 years or more It was the first site in Hong Kong to yield intact prehistoric burials with skeletal remains and associated offerings It demonstrated tha t. under certain conditions. human and animal bones would survive in the moist sandy conditions of the coastal sites

A Stratifìed Site at Shek Pik

•

The work at Shek Pik was extremely well recorded and reported Although all of the artefacts and human remains were lost during the war, the data available from Schofield's site report continues to be of great value.

Investigations after Schofield have added to our knowledge of the site. A rock carving and nearby site , also mainly of the Bronze Age , were reported by a Chinese scholar/c ollector, Chen Kung寸址， lna 1957 report. He conducted large-scale , but poorly recorded , digging on both sites in 1939 , a year after Schofield's work. Excavations conducted in 1979 at Shek Pik by the Hong Kong Archaeological Society showed that , in spite of considerable disturbances to the site from construction in the 1950s , some areas of the original deposit remained. A m句 or two-year excavation of the remaining deposit was conducted by The Chinese University ofHong Kong in 1986-87. An earlier, Middle Neolithic occupation was discovered , along with many interesting featur白， anda “ pre-ceramic culturallayer" was identi且ed at the base of the Middle Neolithic deposi t. Schofield wrote at length about the possible Chinese influence on the prehistoric Yueh peoples of this area and noted that many of the decorations on potte可 and bronze do bear some similarity to those of the Shang and Chou periods in N orth China. Others seemed to him to be more related to Southeast Asia. His general conclusion about Hong Kong's prehistory still seems relevant today - that this area was in ancient times "a crossroads of cultures and probably also of races . . . in some degree foreshadowi月 its present function as a meeting ground of nations."

25

Hong Kong and Hoifung

The 且eldwork

5.1

F r. Rajael Maglíoni

and research conducted by Fr. Rafael Maglioni in the Hoifung district played an important part in the development of local archaeology It is thus appropriate to consider a broad outline of this work , which was the 且rst attempt to relate Hong Kong archaeologically to another area of Guangdong province. Maglioni was an Italian priest working at Swabue , Hoifung CShanwei , Haifeng) district , some 140 km northeast along the coast from Hong Kong. His interest in archaeology was initiated in 1934 through a visit to the area by Fr. Finn. Maglioni made extensive surveys and collections in that district unti11939 and around Wuping during the war years. In 1946 , Maglioni came to Hong Kong where he would reside and write until his death in 1953 The coastal region of Hoifung was richly endowed with archaeological sit白， many of which had been extensively exposed by erosion. Maglioni surveyed these open sites and made collections from the enormous amount of material lying on the surface. Quite frequently, Maglioni found objects in the or屯inal position in which they were left on the surface several thousand years ago , as erosion had gently removed the sand or soil matrix but had not yet displaced the heavier stone or pottery artefacts. In other circumstances , artefacts that had been moved by water action were concentrated in rainwash gullies. Most of the Hoifung sites were in sand banks , dunes or low hillslopes near the sea , but , unlike Hong Kong , the beaches and sand banks were often several kilometres long , with massiv巴， rolling dunes behind the beach itsel f. In many of these landforms , Maglioni found clusters of occupation areas , with each site representing a single cultural phase. Bya close comparison of the material from these phase sites , Maglioni was able to draw up what has been shown to be an accurate cultural sequence for the region. This feat was accomplished

27

Hong Ko月 and Hoifung

without the aid of excavation and , with the exception of one minor si伐， without any observed stratigraphy in the sites under stud乎 The earliest cu1tur呵e was represented at on1y one site that Maglioni named SON - a code derived from the name of the nearest village. The pottery of SON consisted of a fìne paste ware decorated with painted

Y'

n

fL

nrnut

f#'tJ

fL

MG

-- n nu

I

d 9「、

THPLt

PLAU

σ凸

f4

tJ

FJA K

pz 紗

,H d, P1

月UM

PI

σb

t nu

八ν '的 H

P 、u

n 、υ

lt

ρι

H此

H

σ凸

付汁 HJ

可SL

ρι

,l

/L

σ凸

5

陀

勻 JJH

tu

'H HAU

fL

1

M

CJG

j叫VU

04 EUt 3H T

74/

門

戶J 、' 打ue

and incised designs , and a1so a coarse ware with cord impressions. The stone adzes were all of one simp1e , tapering rectangu1ar form The second culture was code-named SAK and was characterized by fine paste wares of 10w-tomedium fìring , with geometric or basketry patterns Some potsherds that Maglioni collected in the interior of Guangdong during the war years 1940-46 were very well fìred , but still classifìed by him as SAK. The classifìcation proved to be correct as this type of pottery is now well known (Yung Long phase) and dated to the Late Neolithic. Finally, at a m句 or village site on a 10w hill codenamed PAT, an advanced Neolithic and Ear1y Bronze Age cu1ture w的 discovered with high-fìred ceramics , sophisticated stone too1s and ornaments , and a few meta1 objects. While there are signifìcant differences between the SAK and PAT materia1 and their rough1y equiva1ent cultures in Hong Kong , Maglio凹's scheme did accord well with the sequence postu1ated by Finn and Schofìe1d for the Hong Kong area. An important contribution made by Maglioni was in the identifìcation of the earliest culture with painted pottery, and in the observation that some of the pottery excavated in Tai Wan by Fr. Finn close1y paralleled this culture The amount of materia1 collected by Maglioni at the 26 sites he discovered in Hoifung was enormous At the PAT site a1one , 250 comp1ete stone adzes , dozens of rings , arrowhea缸， beads and other ornaments were recovered. ln one small area of the sit亡， Maglioni collected every sherd on the surfa仗， weighed the bu1k and estimated the tota1 surface materia1 for the entire site at 600 kg. His study of this materia1 contributed much to an understanding of the typology and technology of the early ceramic and stone industry

28

Archaeology in Hong Kong

5.4

Stonε pr，句 ectile

points collected by Maglioni

5.5

Maglion i's magn昕cent double-F pot

Hong Ko月 and Hoifung

On the other hand , Maglioni followed Finn and others of that era in the practice of identi年1月 archaeological cultures with distinctive peoples. He thus referred to the “Sakians" and “Patians" as new tribes penetrating the area. His reconstruction of Hoifung's prehistory thus overlooked the very strong probability of a local evolution of the material cultures and of an ethnic continuity through time Maglioni was the fi. rst archaeologist working in China to make use of the C-14 dating method. Two of the sites he discovered Cboth assigned to PAT culture) had deposits of shells , bones and blackened soil , and samples of these organic materials were sent for dating. ln spite of several important differences in the potte可 from the two sit白， Maglioni stood by his PAT classi且cation of both , and C-14 proved him correc t. The dates obtained were around 1175 BC and 1000 BC These were the fi.rst C-14 dates obtained on samples from China. Sadly, Maglioni died rather suddenly at the age of 61 in 1953 , just months before the results were obtained. He also had completed a lengthy manuscript about his archaeological work, and this , along with his other writing , w扭曲1ally published in 1975 by the Archaeological Society as Archaeological Discovery in Eastεrn Kwangtung (Maglioni 1975) Nearly 40 years after his prehistoric chronology was devised , it has been shown that his estimate of the age of SON at 4000-3000 BC is also substantially correc t. While his reconstructions of ethnohistorγare today completely obsole仗， Maglio凹， nevertheless , made an important contribution , in general , to the development of archaeology in Hong Kong and South China. In 1967 , his extensive collection was donated by the Catholic Diocese to the Hong Kong governmen t.

29

The Fate of the Early Collections

With the enactment of the Antiquities and Monuments Ordinance , all archaeological “relics" (defìned as artefacts pre-dating the year 1800) discovered in Hong Kong after 1976 became , by law, the property of the Hong Kong governmen t. However, since the 1950s , most archaeological 五nds had been deposited with a public museum , and only private collectors (of which , fortunate峙， there were very few) retained personal possession of the objects found. Li ttle of the pre-war collections have survived the vicissitudes of time. The only human remains from the prehistoric period , for example , h日ving lain tranquil for approximately 3000 years in the sand at Shek Pik, survived only eight years of the turbulent twentieth centur乎叭Thole collections , such as Chen Kung-jit's , were lost through neglect or the chaos of w缸， while Finn's survived the war intact , only to be whittled away by years of attrition for want of safe-keeping The fìrst collection of artefacts , made by Heanley and Shellshear from 1926 to 1934 , was offered to the Hong Kong governrτlent by Shellshear on his retirement in the late 1930s. According to a letter written in 19 泣，“1 offered the collection to the Government provided that its 閻長ty could be guaranteed. As there was some dif五culty about this 1 took the collection home and presented it to the British Museum." These materials , consisting mainly of stone adzes but including also a number of bronze weapons , are still held by the museum; ironically, the guarantees sought by Shellshear would have proved of no value due to the] apanese invasion , and had the collection been taken by the Hong Kong government in 1938 , it would almost certainly be lost today Finn's collection was fortunate enough to survive the war years , but only because Fr. Finn was Irish! During the]apanese occupatio且， all government property, including all facilities at the University of Hong Kong , were declared to be the property of the Imperial] apanese

The Fate of the Early Collectio悶

Governmen t. Accordingly, an offìcial of the Education Department inspected the archaeological collection at Ricci Hall (where Finn had resided) and posted a notice (that survives today as part of the Finn collection) to the effect that the collection had been confìscated. He also took away a number of specimens. The matter was immediately taken up with the authorities by the Ricci Hall warden , who pointed out that all property in Ricci Hall belonged to the Irish ]esuits , not the universi可 and ， hence , should be protected since Ireland was a neutral party in the conflict. The ]apanese agreed to this. After 1945 , the material was moved to the regional semina門， then back to Ricci Hall and , fìnally, donated in two parts to the university's Fung Ping Shan Museum and the City Hall Museum. In recent years , some of Finn's writings and material has been brought to ligh t. During the editing of the Maglioni papers in 197斗， 1 found a copy of Finn's report to the governme凹， the original of which was lost during the war. In 1975 , during a visit to the Ashmolean Museum at Oxford , 1 found a copy of Finn's catalogue among the Schofìeld papers. This document gives numbe芯， descriptions and provenance data to the most important objects excavated by Finn at Tai Wan Two numbers on objects in the Finn collection at the Museum of Hist。可 matched the numbers and descriptions of the objects in the catalogue. Finally, in 1974 , three bronze axes were purchased at the Cat Street “ flea market" by a member of the Archaeological Socie句， Bob Maher, who two years later discovered that they were from Tai Wan and generously donated them to the Archaeological Society Two of the axes bore numbers corresponding to those in the newly discovered catalogue.

6.1 AnentryinFinn 's catalogue describing one of the bronze axeheads recovered ma叮 years later from Cat Street

31

32

Archaeology in Hong Kon呂

6.2

Th e three bro院e axeheads from Cat Street

As noted above , the Shek Pik materials excavated by Schofield were lost , but many of his original site and studio photographs and 五εld notebooks have been recovered. A number of stone adzes and bronze artefacts were donated by Schofield to the British Museum , but none of these are from Shek Pik There is another interesting tale of recovery concerning some of the artefacts excavated by Chen Kung寸 it. In a 12-page pamphlet Chen (1 952) described how most of his collection was lost during the ]apanese occupation of Hong Kong. He had hidden most of his valuable objects with friends , but somehow the ]apanese found and seized this material. He later heard that it had been put on display at the HongKong and Shanghai Bank building. After 1945 , he tried to find out what happened to the objects but could not locate them In 1972 , a member of the Archaeological Society and employee of the HongKong and Shanghai Bank was informed that an old trunk containing antique vessels had been found in one of the bank's godowns. As there was no indication to whom the trunk belonged , it was sent to the Archaeological Socie作 We opened the trunk to find several complete coarse corded pots and stoneware stem cups mSl缸， and 1 noticed a torn label still adhering to the side with the Chinese characters kung and jit. A comparison of the vessels with the photographs in Chen's (1 95 7) Shek Pik report revealed that at least three , and probably all , were excavated there. Apart from this one incide凹， however , the whereabouts of the large collection made by Chen remains a mystery

The Fate of the Early Collectio悶

Maglioni's collection was certainly the largest of any individual up to the present , and it also had an eventful history A disastrous typhoon in 1937 wrecked his mission house in Swabue , and “ much material was los t." A government official from Canton later visited and “ confiscated eight chests" of selected items that have also disappeared A small collection of Hoifung materi泣， along with a large quantity of artefacts collected in the Wuping region , were brought to Hong Kong by Maglioni in 1946. This collection was donated to the Hong Kong government by the Catholic Diocese after Maglioni's death in 1953. The early po哎-war years did not see much progress towards a systematic and centralized collection. Some of the m日terial found or excavated by Professor S. G. Davis in the 1950s was deposited in the Department of Geography and Geology at the University of Hong Kong. Material from Hung Shing Yee and other sites was taken back to England by one of Davis' colleagues and donated to the British Museum. Other material went into private collections. Fortunate旬， the prompt action of the Public Works Department prevented the looting of the Lei Cheng Uk tomb , and its contents have been held ever since in a small museum there. With the formation of the University Archaeological Team in 1956 , a territory-wide collection began to be maintained , first at the university's Fung Ping Shan Museum , then at the City Hall after the founding of the Archaeological Society in 1967. Finally, in 1979 , a central archaeological repository was established in the Hong Kong Museum of HistOl水 and all newly discovered material is placed in this collection , now maintained by the Antiquities Office There is cause for cautious optimism that archaeologists of the future will have a much more complete collection and better documented record of the work carried out in the late twentieth and twenty-first centuries

33

Post-war Developments , 1945-78

The first years after the end of the Second World War saw very little archaeological activity in Hong Kong. Mr. Chen Kung-jit , a Chinese scholar and collector who had taken an interest in the subject in the 1930s , apparently continued his searches and extensive collector digging at various sites. He published a small pamphlet (Chen 1952) describing his discoveries and providing a few illustrations of stone and bronze artifacts. Eventually, in 1957 , he published an article on his findings in the Chinese journal , KaoguXuebao (Chen 195 7). From 1947 , a small group , led by Professor S. G. Davis of the Department of Geography and Geology at the University of Hong Kong , carried out limited surveys and excavations. He wrote that he and several frien血， sometimes accompanied by stud凹的， visited Lamma eve可 Sunday to collect material. Davis told me that this was mostly scraping down exposed faces and , occasionally, digging into them , rather than in regular excavation pits 日ttle of this work seems to have been recorded or published , apart from a brief report (Weinberger 1948-49) on the Bronze Age site at Hung Shing Yee , Lamma. Davis also represented Hong Kong in the regional association of archaeologists when it was reorganized after the war as The Far Eastern Prehistory Association (FEPA). Professor Solheim of Hawaii initiated the journal Asian Perspectives that was published by Hong Kong University Press from 1958 until196斗， when it was taken over by the University ofHawaii. For a few years , Davis was the Hong Kong correspondent and wrote short reports on local archaeology In August 1955 , construction work for the Lei Cheng Uk es囚犯， on the western side of Kowloon , exposed an ancient tomb of considerable size and historical importance. Fortunately, little damage had been done to the structur乏， and prompt action by the Public Works Department prevented its contents from being looted. A team from the universi你 led by Professor F S. Drake of the Department

Post-warDεvelopme帖， 1945-78

7.1 Discovery 01 the Lei Cheng Uk tomb in 1955

of Chir哎哎， took charge of the excavation of the tomb and the study of its material. The structure was found to be a brick vaulted fourchamber tomb of Han dynasty type , containing a number of ceramic vessels , house models and bronze artefacts. This discovery evoked considerable public interest , as it suggested a possible Chinese settlement in this area as early as 100 BC - some 1,000 years earlier than an inscription at Fat Tong Mun The discovery at Lei Cheng Uk stimulated interest in local archaeology, and , in March 1956 , a University Archaeological Team was formed , under the direction of Davis. It had 25 members , approximately half were from the university and the remainder from the public. Due to some dispu仗， Drake refused to join and his involvement in local archaeology ended with his report on the Lei Cheng Uk tomb. The University Archaeological Team carried out a wide-ranging programme of surveys and exploratory digging during 1956-65 , covering sites noted by pre-war archaeologists as well as areas not previously inv臼tigated. A map of all known sites in the territ。可 was prepar吐， based on the earlier maps of sites drawn up by Shellshear and Schofìeld. This work was completed by the Archaeological Society and published by the Hong Kong Government Printer in 1971 An important Bronze Age site at Man Kok Tsui near Silvermine Bay on Lantau was discovered in 1958 by Dr呵. S. M. Bard , a member of the team. The site was extremely rich in artefacts , especially in the large quantity of pottery exposed by recent cultivation of the area A major excavation was mounted with government assistance and resulted in the discovery of a collection of complete pottery vessels and

35

36

Archaeology in Hong Kon呂

Activities of the University Archaeological Team. Upper left: Mary Tregear leading a βeld walk; upper right middle left: S. M. Bard; middle right: large pot revealed in a tank ground nεar Fanlir怒; bottom row: excavations at sites on Lantau. 7.2

James 札Tatt;

Post-warDεvelopme帖， 1945-78

polished stone artefacts. The site (discussed below) was spread over a large ar凹， and it was suggested that possible “ working ,"“ dwelling" and “ burial" areas could be distinguished. A report on the site was published by Davis and another team member, Mary Trege缸， inAsian Perspectives in 1961 , and was the first archaeological site report from Hong Kong to appear in an international journal since the pre-war era (Davis and Tregear 196 1) In March 1962 , another important historical find was made during construction. Bulldozers clearing for the Shek Pik reservoir brought to light a large cache of Sung dynasty coins and high quality glazed pottery (celadon). Many of the coins were taken away by workers , but most of the celadon was recovered. This material was obviously from an elaborate burial and was believed by some to be quite possibly related to the travelling court of Sung (Sung Wong Toi). Others held that it could have been the grave of a wealthy local merchant or landowner In 1967 , the team was transformed into the Hong Kong Archaeological Society in order to stimulate greater public interest and involvement in local archaeology Professor Davis was its first chairman. 1 joined the executive committee in 1971 at a time when lt was expenencmg a m句 or conflict between Davis (the old colonial) and the secretary James Watt Chi-ying , an ang可 anti-coloni泣， young man. Watt promptly resigned as secretary, and Davis stepped down as chairman a year later and was succeeded by Dr. Bard. 1 went on to serve as editor from 1974 to 1985 and chairman from 1985 to 1996. With a stable membership at about 200 , the Archaeological Society, since its inception , has organized excavations , exhibits and lectures. It publishes a regular journal that achieved international recognition in the years after excavations at Sham Wan. Professor W G. Solheim of the University ofHawaii , who has actively followed the Archaeological Society's work through the years , said that the society had two “ golden ages" - the work at Sham Wan in the 1970s and the project at Chek Lap Kok in the early 1990s Within a few months of its inauguration , the Archaeological Society was confronted with a very important discover手 Workmen digging a pipeline trench at Chung Hom Wan on Hong Kong Island uncovered several complete painted bowls and stem cups , apparently from the Neolithic Age. There seemed to be no cultural stratigraphy and no other artefacts in association. The painted ware was correctl

37

38

Archaeology in Hong Kon呂

7.3 Excavation at Tai Long, Lantau. in 1969

interpreted to be earlier than the well-known Late Neolithic geometric pottery, but wrongly believed to be evidence of a “ Lungshanoid Expansion ," originating in North China and reaching this far south , just prior to the rise of geometric potter手 Other sites excavated during the society's early years (1 967-72) were Tai Long on Lantau , the North Lamma School si仗， Shek Kok Tsui near Tuen Mun and Hac Sa Wan in Macau. The excavations were largely weekend affairs that provided some fieldwork for Bard's extramural classes in archaeology, and many new members of the society were recruited through these classes. Not much progress was made , however, in understanding the prehistory of the area. Members were encouraged , however, by the frequent visits of Professor Wilhelm Solheim of the University of Hawaii and one of the leading experts in Southeast Asian prehistor手 His first visit to Hong Kong in 1971 was particularly memorable: he gave a talk to the society about the exciting new discoveries that he had made in Thailand , and a field trip was also arranged for him to join. This was a walking survey of High lsland where construction of a reservoir was scheduled to begin soon. lt was later written up in the society's journal (vol. lll , 1972) and included this passage

Post-warDεvelopme帖， 1945-78

. . . a survey ofHigh Island had been organized [and got underway, but] a persistent he且可 rain began to fall mid-way through the afternoon. As the party arrived back at the launch , wear其 soaking wet and spattered with mud [with two dubious artifacts as reward for all their effort址， Dr. Solheim was heard to remark that he had quite enjoyed the walk and “ 1 wish 1 could spend every day like tha t." A major evolution in Hong Kong archaeology came with the excavations at Sham Wan. Lamma Island in 1972-76. The site was unique in providing evidence for most major phases in Hong Kong prehistory, including an earlier Middle Neolithic phase not previously seen stratigraphically Five seasons of excavation at Sham Wan were undertaken by the Archaeological Socie句" with government and Urban Council support , and brought to light a wealth of data on each cultural phase. It was the first site in this area to be subjected to a wide range of scienti五cs閃出es. A monograph was published on the site in English and Chinese versions in 1978 (Meacham 1978). The new knowledge derived from Sham Wan led to reinvestigation and reinterpretation of known sites. Several , such as Tung Kwu and Chung Hom Wa日， were found to have cultural deposits below the deepest levels reached in previous excavations. The Middle Neolithic phase was confirmed at both sites to be stratigraphically below the soft geometric pottery layer. Carbon dating at Chung Hom Wan showed that the earliest human activity on the site took place between 4500 and 3500 BC. No additional examples of painted pottery were found there but this earliest cultural phase was believed to have provided the painted wares found in 1967. By the end of the 1970s , an important Bronze Age site at Hai Dei Wan , Lantau (now part of Discovery Bay) , was excavated and published in detail. Several other new sites had been identified as promising - notably PO Yue Wan and Sai Wan on Cheung Chau , Lo So Shing and Sha PO Tsuen on Lamma. Sites from the pre-war period were re-excavated and found to still offer potential for further research , notably Shek Pik and Tai Long on Lant訓， and Tai Wan on Lamma (Fr. Finn's site). An important new avenue of research had opened up with the discovery of the widespread lime industry of the first millennium AD in Hong Kong Finally, mention must be made of the (generally) positive steps taken during this period by the government in archaeology Over the last 50 years , the development of local archaeology had been rather uneven. 叭7hile some of the early work was well done , many of the 且eld

39

40

Archaeology in Hong Kon呂

techniques used were not systemat眩， notes and maps were inadequate and photographs rarely made. As noted abo間， many of the valuable materials discovered went into private hands and were eventually los t. When the City Museum and Art Gallerγopened in 1962 , from all the extensive work of the pre-war period , only a part of Fr. Finn's collection remained in Hong Kong. Together with the Lei Cheng Uk and University Archaeological Team materi址， it formed the basis of the fì. rst centralized archaeological collection. The City Museum and Art Gallery in City Hall (later the Museum of History in Kowloon Park) provided a workshop for the Archaeological Socie哼" while the Urban Council (later the government through the Antiquities and Monuments Offì. ce) provided a small subvention. All material found

G泣

l,

，乎比ny

pι'

tu 5 tn pl

H

此

內關

卅M

TLM L9

叫

Z

7/ 戶、 d

刁峙

G

by the Archaeological Society was deposited in the central collection , then maintained by the Hong Kong Museum of Hist。可(later by the Antiquities Offì. ce) Pressure for antiquities legislation had started in the 1960s by members of the University Archaeological Team , and the baton was passed on to the Archaeological Society Finally, in 1976 , the Antiquities and Monuments Ordinance was enacted. All "relics" (de fì.ned as artefacts dating earlier than 1800) found in Hong Kong became government propert手 A licensing procedure for excavation

Post-warDεvelopme帖， 1945-78

7. 5

Openíng of the provísíonal Museum of Hístory at Star House ín 1975

was set up to ensure that proper standards were me t. And the government secretariat established an Antiquities and Monuments Unit for the documentation of antiquities; this unit also began to conduct its own surveys and excavations of sites threatened by development

41

必u

﹒司EA

m 郎

、EA

歐

倪志

L

Iε

叫 N A

到

WE aI nu IW

Certainly the most important archaeological site excavated thus far in Hong Kong is at Sham Wan , south of Picnic Bay on Lamma Island. Solomon Bard , director of the fi. rst three seasons of work there in 1972-75 and an active local archaeologist since the 1950s , noted in his introduction to the Sham Wan monograph (Meacham 1978):

8.1 The site at Sham Wan , Lamma, in 1972

Sham Wan was the turning point (in the development of archaeology here) not merely because of improved methodology and techniques , but also because it opened the door to a systematic approach to the problems of local cultural sequence and chronology. Here at last was one single site which seemed to hold the answers to a number of questions.

趴 S hamWa 叩 叭， n

Lamma: A Ne凹 wEra Be 咚嚕叩 ♂伊1T刊 E 1

Although the site at Sham Wan was known to both Finn and Schofield , it was forgotten after the war. ln April of 1971 , 1 began visiting known and potential sites on Lamma and rediscovered the site at Sham Wan. The high sand bank behind the beach seemed , from a distan白， very likely to be a rich archaeological slt亡; other similar sand deposits had yielded valuable material from Hong Kong's prehistory, and the one at Sham Wan was higher than most others. lnitial surface inspections and sub-surface probes yielded no evidence , but , finally, a probe to a depth of 1 m yielded prehistoric potsherds and stone flakes. Test pits showed the site to be indeed of great potenti泣， with more than 2 m of deposi t. The second test pit yielded a very dramatic discovery: a pair of Six Dynasties glazed jars and covering bowls under a small boulder. This generated much excitement among members and was taken as an indication that a ve可 rich harvest of early historical burial goods could be expected. As it turned out , this was the only such burial discovered during the entire excavation of the site. Five seasons of excavation by the Hong Kong Archaeological Society followed from 1971-75; funds were provided by the government and the Urban Council , and assistance was obtained from both universities ，也e Marine Departme凹， the District Offic亡， and the Museum ofHistory Bard directed the fìrst three phases of excavation , and 1 directed the 五nal two , as well as the preparation of the monographic site repor t. Professor Davis was actively involved in the first phase of excavation; the third phase was co-directed by American archaeologist Martha Joukowski and art historian James Wat t. Important visitors to the site were the governor of Hong Kong , Sir Murray MacLehose , and one of the best-known archaeologists in the region , Professor W G. Solheim of the University of Hawaii. Most of the work was done by members of the Archaeological Society as volunteers , plus very good-natured and diligent villagers from nearby Tung 0 who were employed.

8.2

訕。如 Dynasties pots bei啥叫ωvated at SÌU1m Wan

44

Archaεology in Ho月 Kong

8.3

First phase of excavation at Sham 札切1， 1972

The society purchased the small stone hut on the si伐， and , without running water or electrici句， it nonetheless became the hub of the projec t. Camaraderie on the site was as satisfying as the archaeology; each phase was marked by several camping sessions of two to three weeks , with considerable discussion of the day's 且ndings by kerosene lantern over dinner and beer The most important feature of the 乳白， as noted by Bard , was its extensive cultural deposi的， attaining a depth of 3.3 m in some areas. This feature distinguished Sham Wan from most other sites in Hong Kong that , at the tim亡， appeared to have only a singl亡， rather shallow cultural layer belonging to one period of activity or occupation. Prior to Sham Wan , almost all excavations in Hong Kong were less than 1 m deep and stopped when a sterile layer was reached , in the belief that the end of cultural deposit had been reached. Fortunately, it was decided to extend one test pit at Sham Wan down to bedrock or water, and , after a sterile layer of only 30 cm , the second prehistoric culturallayer was encountered. The sequence oflayers at Sham Wan spans at least 5 ,700 years and added a new phase to Hong Kongs prehistory

趴 S hamWa 叩 叭， n

Lamma: A Ne凹 wEra Be 咚嚕叩 ♂伊1T刊 E 1

8.4 Visit 可 the Governo月 Sir Murray MacLehose , to the excavation at Sham Wan in 1975; ProJessor W G. Solheim standing 1可t oJ centre

46

Archaeology in Hong Kon呂

Coins and potte可 types from the fìrst two layers date them clearly to the late historical (Ch'ing dynasty) and early historical (fìrst

UFLIU

Gωotnfv

陀 mgm

叮Ardnμ~

叫=立一一一­

到

-0 -

叩門叫 w VW

LMm= 的叫 4 划 AO5

ru

川少地咒

』們

M廿悅

Z

imm 心叫忱此

millennium 的 Ch'in to Tang dynasties) , respectivel手 Scientifìc dating tests , such as radiocarbon (C-14) and thermoluminescence (下L) on samples from Sham Wa口， have indicated dates of around 1000-400 BC for the Bronze Age level , 2400-1400 BC for the Late Neolith芯， and 3500-2800 BC for the Middle Neolithic Sham Wan was the 五rst site in Hong Kong to provide such a dated sequence and the fìrst to yield an identifìable Middle Neolithic culturallayer, pushing the known period ofhuman habitation in the region back some 1500 years. As luck would have it , this earliest cultural deposit at the site was also the richest in artefacts , providing a wealth of data on the newly identifìed phas亡. In addition , the valley adjacent to the sand bank provided considerable information on the formation of the site and its environs. Oyster shells found in the JJpyn 8BdqAMd , FLE--ZIR HPAEd thUULM valley (former lagoon) clay deposits were dated to around 4400 BC t=EE - indicating that the sea had already reached its present level by that timε. By 2400 BC , the lagoon had evolved into a brackish water swamp.

趴 S hamWa 叩 叭， n

The many months of excavation revealed burials in each cultural laye r: a skeleton , said by villagers to have been a fisherman killed by ]apanese strafing in 1942; a coffin accompanied by two teapots from the late Ch'ing era; two burial jars from the Six Dynasties period; a stone slab structure (almost certainly related to a burial) of the Bronze Age; and cremation and inhumation burials accompanied by funerary objects of the Neolithic phases. Interestingly, cremation remains were also found in the early historicallayer and in the same general area of the site as those of 4000 years earlier. Detailed examination of the human remains yielded clear evidences for the Middle Neolithic practice of intention剖， high-temperature cremation (the first such evidence from South China) , for the number of individuals and their age , and for malnutrition and dental abscesses A total of 10-15 i日dividuals were represented in the remains of the Middle Neolithic layer; most were young adults. It was later claimed that the material constituted evidence of cannibalism , but this notion was refuted by the overwhelming evidence that deliberate cremation had been carried ou t. The society was fortunate in having the services of Professor Peter Lisowski of the Department of Anatomy at the University of Hong Kong. He had experience in identifying human remains post-cremation. He pointed out that the cracking and splitting

Lamma: A Ne凹 wEra Be 咚嚕叩 ♂伊1T刊 E 1

8.6 Photograph 0] the “ H-sqLLare" stone slab structLLre partially excavated

8.7 Drilled hLLman jaw bones ]rom the Middle Neolithic at Sham Wan

48

Archaeology in Hong Kon呂

flb

u

-r

的叫 仰甘心

p 、ν

TILPZU

「-lfL

--圳

Er

，的 UIHH

nω

d4A AL o04LWj ew}~ til EJHd ji'H dd HJU MN

of the skull fragments from Sham Wan indicated a temperature of at least 800 0 C , a level that would have reduced all flesh to ash Professor Lisowski was also one of several people who discovered how treacherous the surf at Sham Wan could be. The dinghy carrying him , his 五εld kit and his picnic lunch was swamped by a large wave just as the boatboywas attempting to land on the beach. The professor recovered his kit , others shared their lunch with him , and he was later able to inspect some of the human remains in situ. True food remains were also found at Sham Wan , consisting mainly of fìsh bones and shells. Analyses of these materials yielded valuable new data on the types of marine resources that the early people of this region were exploiting. Two species of fìsh predominated: the headgrunt and the marine catfìsh. The average size of the fìsh was 1 m in length , much bigger than ever found inshore toda手 Remains of deer and pig were also found. Again , Sham Wan was the fìrst site to yield any information of this type , although recent investigations have indicated that other sites , such as PO Yue Wan on Cheung Chau and Sha PO Tsuen on Lamm且， also have food remains in certain ar亡的­ This topic is now one of the most important research priorities in Hong Kong archaeolo gy:

pu

趴 S hamWa 叩 叭， n

Among the other studies carried out during or after the excavations at Sham Wan were: an ethnographic (folklore) study of the village at Tung 0 near the si仗， a geological study of the sand bank , analyses of the pottery from different layers to determine composition and fìring temperatur亡， and analyses of stone and bronze tools. A detailed study of the soil characteristics relevant to T-L dating was conducted by Martin Aitken of the Oxford Laboratorγfor Archaeolo gy, and the site gained a certain notoriety in theτL world because of the diffìculties its ground water radon posed for dating. The Sham Wan site study was the fìrst major multi-disciplinary project devoted to an archaeological site in this region , employing many of the most advanced scientifìc techniques to analyze the material. After much internal debate and dispu仗， am呵。rmonograph (Meacham 1978) containing the results of this study was published by the Archaeological Society in English , and a Chinese version was suggested and funded by the Urban Council. Interest in and favourable comment on the study was expressed internationally The site thus made a small but not insignifìcant contribution to the broad study of the cultural , social and economic development of this area over the last 6 ,000 years

Lamma: A Ne凹 wEra Be 咚嚕叩 ♂伊1T刊 E 1

Post-Sham Wan , 1978-90

9.1 Excavatíon at Chung Hom Wan ín 1975

After the full publication of the Sham Wan study in 1978 , Hong Kong archaeology entered a more focused phase. Excavations at other sites were 五nely tuned to specifìc qu臼tions of chronology and typology raised by the stratigraphy at Sham Wan; several sites were re-examined and provided corroboration and amplifìcation of the basic cultural sequence. Some sites were even found to have one or more cultural laye的 below the limits of the previous excavations , which usually stopped when a signifìcant sterile zone was reached. One such site was Tung Kwu Island , where a complete pot was found in the bottom of an old test p旺， about 30 cm below the level at which "sterile" had supposedly been reached.

Post-Sham W，訓， 1978-90

9.2

Ex cavation at Lo So Shing , Lamma， 的 1978

PO Yue Wan on Cheung Chau (excavated 1980-81 and 1985) became only the second site in Hong Kong to yield significant quantities of fish remains and shells in a prehistoric contex t. The predominant species were once again the headgrunt and marine catfish. The site also had polished bone points and chipped shell scrapers - two artefact types not previously recorded in Hong Kong. Three series of C-1 4 dates on shell remains from the site gave conflicting dates clustered around 2100 , 1600 and 1200 BC. A major research project came to fruition in the early 1980s , with the absolute dating and identification of the lime kilns. After the initial work at Yi Long (now 5ea Ranch) on Lantau and Lo 50 5hing on Lamma , several other kiln sites were investigated. The major site at 5ham Wan Tsuen on Chek Lap Kok was excavated over four呵 seasons from 1979 to 1984 , in response to initial plans to build Hong Kong's new airport there. The work provided conclusive evidence of the 5ix Dynasties-Tang dati月 of the kil凹， with abundant coins , potterγand C-14 samples all indicating activity spread over the fìfth to ninth centuries. Another kiln site at Pui 0 on Lantau produced a large structure believed to be a slaking pit for lime; carbon from clay

51

52

Archaeology in Hong Kon呂

deposited in the bottom of the pit dated around AD 400. This result overlapped with the first dates obtained on samples from kilns at Yi Long , and together they indicate the probable beginning of the lime industry The work on the kilns served to fill what had been a virtual missing link in Hong Kong archaeology - the 五rst millennium AD. As the kiln furniture began to be understood , it provided an important marker for the interpretation of stratigraphy at many other sites where kiln debris is common. For years previously, this debris (and the kilns themselves) were thought to be of recent origin and were not properly recorded archaeologicall手 The resounding success of this project was due to the persistence of Hugh Cameron , chairman of the Archaeological Society, who had an interest in the kilns as structures and how they functioned , regardless of their age. Less enthusiastic , but nonetheless supporti前， were some stalwart members of the Archaeological Socie句" myself included The impressive progress of the late 1970s and early 1980s was dampened from 1982 by increasingly ill-conceived government policies , and serious conflicts arose between the authorities and the Archaeological Society A moratorium on all archaeological research excavation was declared , supposedly to prevent "a piecemeal approach ," while the government commissioned a three-year , territory-wide survey of archaeological sites. A challenge to this moratorium was launched by me when an application to investigate the site ofTai Long on Lantau was rejected. An appeal to the governor W佑 , quite extraordinarily, allow吐 ， 仰 0 v疋耐叫 臼叮r叫 E t of the principal officials. The governor, Sir (l ater Lord) Murray MacLeho仗， was noted for his no-nonsense style , and the story goes that , instead of merely rubber-stamping the proposed rejection of the appeal , he wrote a short memo questioning the reasons behind it and inquiring whether there were personality conflicts involved A sub-committee was hastily set up to examine the matter, and , by a close vo仗， it was decided to recommend in favour of the appeal. The moratorium did , nonetheless , eventually take effe 叫， limiting local archaeology to salvage work, while the much-vaunted “ territory-wide survey" was carried ou t. Poorly designed from the outset , the survey did not focus on priority development areas , had no topographic search strategy, made no serious attempt to assimilate the work of the previous decad白， failed to discover sites that wo

Post-Sham W，訓，

with tiny squares and reached conclu日ons totally unsupported by the evidence. To cite one example among many, the survey report (Peacock and Nixon 1988) had this 的兒ssment of the hill site behind Yung Shu Wan on Lamma: Limited excavation [a 1 x 1. 5 m trench] on the proposed site of a private residence indicated that . . . this hilltop site no longer contained any valuable in-situ archaeological deposits Later excavation at the site of another house plot on the same hill provided clear evidence of in-situ Bronze Age deposi t. When the survey report was finally prepared , it was circulated to local and overseas scholars for comme凹， and the reactions were overwhelmingly scathing. While not admitting it publicly, government officials recognized that this survey had not been done well One important site that did emerge from the survey was at Yung Long near Castle Peak, although the site had been previously identified as promising by the Archaeological Society. Grandiose claims were made by the survey director Brian Peacock concerning the “ discovery of a Neolithic pottery kiln" with great potential “ to shed light on China's developing technolo gy." Minor damage to the site by a contractor led to a larger excavation in 1986 , conducted with m句 or government funding by Peacock. A serious protest was made by the Archaeological Socie句，的 the discoverer of the sit亡， for not being consulted , and the issue was investigated by the Legislative Council. The excavation showed the “ kiln" to have been little more than an ash-filled depression , containing fired-clay fragm凹的 of what is generally regarded by Guangdong archaeologists as kitchen stove debris. The site did yield a good assemblage of Late Neolithic material and appeared to have been occupied only at that timεltwa站5 th 立 es 叩 u句 ecαt of a mas 臼凹 5曰lVe Society and (怕 di已scussed

出 t he

Antiquities and fu 凹1ft 吐 t由 he 叮r below).

Monument臼s Of伍五C 臼e

in

1992一93

Also in 1985 , a week-long excavation by the Archaeological Society at Hac Sa Wan in Macau , sponsored by the Macau City Council , provided one verγimportant piece of information. Painted pottery of the type found in 1968 during construction at Chung Hom Wan in Hong Kong was found below a layer containing chalky incised potte可 similar to that of the Middle Neolithic layer at Sham Wan. Although not completely equitable with the Middle Neolithic at Sham Wan , the upper layer at Hac Sa Wan was clearly pre-geometric and had several strong links to the Sham Wan assemblage. For the

1978-90

53

54

Archaeology in Hong Kon呂

五rst t1m亡，

it was shown that painted pottery was earli白， eventhough it has generally similar ceramic shapes. It was thus established that there were two phases of the 10ca1 Midd1e Neolithic , the earlier one marked by painted ware Yet another dispute with government offìcia1s arose in the 1980s over the status and importance of the site at Penny's Bay, where 1arge amounts of Ming b1ue-and-white porcelain had been found on the surface. The governmen t's position was that the site was not immediately threatened and , therefo肥， shou1d not be excavated unti1 the territory survey was comp1ete and the site assessed. The Archaeo1ogica1 Society believed the site was eroding and shou1d be investigated at 1east to determine its potential. After severa1 rounds of discussion and application rejectio悶， the site was fìnally deemed by the survey to have on1y "a superfìcia1 deposit." Test excavations by the

m 洲的

戶的 m

roσ

pιv

凸

、

αILmu

Arω

巴vρL

r--

有UNHO

td

利 UM叮叮

dJμ41

BVτdl

。 JTD'HrL

3ym 肘

--IEh fno

society showed , to the contrary, that it had over a metre of deposit , much of it under the water tab1e. The site was unique in having such 1arge quantities of Ming porce1ain , and it posed something of a daumystery as it was clear1y not an ordinary farming village site. M句 or excavations were conducted there in the 1990s (discussed further be1ow) In 1986-88 , a m句 or excavation was undertaken at Shek Pik by a new1y formed team from The Chinese University of Hong Kong , headed by a]apanese-trained archaeo1ogist , Tang Chung , and assisted by the Archaeo1ogica1 Societ手 Test pits in 1979 had indicated that some are的 of this important site still had undisturbed deposits. The 10wer levels had a well-defìned Midd1e Neolithic cu1tur茫， but without the incised and painted pottery of other known sites. An extreme1y rare type of coarse incised pottery was quite frequent in this deposit Moreover, at the 10west 1evel a non-ceramic horizon with chipped stone too1s and flakes was identifìed. It remains unclear whether this cultura11ayer constitutes a distinct cu1tura1 phase apart from the Midd1e Neolithic with which it merges stratigraphically A breakthrough of sorts occurred in 1980 when the Archaeo10gica1 Society was invited , after informa1 “ fee1ers" had been put out , to send a de1egation to visit Canton (Guangzhou). Six members , 1ed by the chairman Hugh Cameron , made the trip. Four ta1ks were given by members to staff of Guangdong Provincia1 Museum and to staff and students of Sun Yat Sen Univ叮叮叮 ln 1981 , a delegation from Guangzhou made a trip to Hong Kong as guests of the

Post-Sham W，訓， 1978-90

9.4

Excavation team of Lung Kwu Tan villagers and Archaeological Society members in 1990

9.5

Members of the Archaeological Society met the delegation from Guangdong at the train station in 1981

55

56

Archaeology in Hong Kon呂

A tribute should be paid to the villagers who assisted in the major excavations. Beginning with Sham Wan , a very good relationship W的自 tablished with the villagers employed as labourers on the excavations. These Punti and Hakka were hard-working farming people and proved to be excellent excavators , learning ve可 quickly the basic techniques and artefact types. In addition , they were ho ∞ ug 拉 h some cωor凹d 世 ler眨 ed 趴the 叫 ta 址lka 瓜盯 t仙llV good-natured and talkative. Alt由 char呵acteristic a nui芯sance ， for those of us anxious to practise Cantonese even that tinged with a strong country accent- it was a gift Particularly notable were the older village時， some in their seventies who were even tougher than the younger ones and who could endure the mid-day sun with an amazing capacity. Special mention should also be made of the villagers at Lung Kwu Tan and Chek Lap Kok who assisted most ably in numerous excavations later in the 1990s.

Salvage Archaeology, 1990-2007

The early 1990s marked the beginning of another new era in Hong Kong archaeology, characterized by numerous survey projects and large-scale salvage excavations funded by government or private developers wanting to acquire the land containing archaeological sit臼 Government policy 且nally took a positive turn , requiring salvage excavation prior to destruction of the sites. But the bureaucracy and government control through funding also began to intrude on the freedom of archaeologists to operate normally. In the period 1990-95 , contract archaeology took root dramatically in Hong Kong , with 35 rescue pr吋 ects involving full-time staff and weeks of fieldwork; over half of these pr叮 ects were m句 or surveys or excavations. The construction of a large infrastructure project for the new airport and highway on Chek Lap Kok and Lantau was one of the main triggers

10.1 Field survey just ahead of the bulldozers at Kau Sai Chau in 1994

58

Archaeology in Ho月 Kon

10.2

Monitoring ωnstruction

projects revealed many new sites

10.3 Snatching a pot intact from the bulldozer treads

for this new policy, along with pressure from the Archaeological Socie可 The increasing involvement of mainland archaeological teams in Hong Kong was another new feature of this period , initiated by The Chinese University of Hong Kong team and later taken up with vigour by the Antiquities Office as 1997 approached The plan for the new airport involved the complete levelling of the island of Chek Lap Kok and construction of a four-lane highway and railroad along the north shore of Lantau to connect the airport to the cit芋 Several sites were already known on the island and along the Lantau shoreline . an intensive survey brought to light many mo白， and one year was allotted for the salvage excavations. The work at Chek Lap Kok was conducted by the Archaeological Society, while the Lantau coast was investigated by The Chinese University of Hong Kong team. Chek Lap Kok provided a number of new features to local archaeology The complete excavation of the Tang lime-kiln site on the sand bar at Sham Wan Tsuen revealed a total of 12 kilns and 3 clay structures , probably slaking pi伍， along with a large quantity of Tang pottery and more than 100 coins. Sandy deposits at the base of the hill at Sham Wan Tsuen provided clear evidence that painted potte可 pre-dated the incised potte叮 of the Middle Neolithic ，的 the Macau site had already indicated.

Salvage Archaeology; 1990-2007

Middle Neolithic “ Sham Wan phase" burial grounds at Kwo Lo Wan and Fu Tei provided a large amount of the incised pottery of that period , in complete forms never seen before. A Bronze Age burial area nearby at Kwo Lo Wan yielded three pairs ofbi-valve moulds for casting bronze axes , and a piece of fìne textile stuck to and preserved by a corroded bronze objec t. The site of Ha Law Wan yielded a surprising collection of 13 kilns constructed ve可 differently from the Tang lime kil悶， burrowed into the hillslope and probably used for smelting iron-rich sands. This site was dated by C-14 to around AD 1250-1400 , roughly correspondi月 to the Yuan dynasty A 300-page monograph and a colour booklet describing the work were published by the Archaeological Society (Meacham 1994且， 1994b). The most important site on the north Lantau coast was that of Pak Mong , which had cultural deposits from the Late Neolithic , Han , Six Dynasties-Sui and Tang. The Han occupation was marked by some pottery dati月 to the Western Han , although most is of the later Eastern Han; it is the only site with pottery of the early Han known in Hong Kong. A report on the site was published in the national archaeological journal Kao Gu. Another important site on

10.4 ArchaeologicalSociety members visited the Chinese University excavation team at Pak 瓦1ong， Lantau , in 1993

59

60

Archaeology in Ho月 Kon

10 .5 Hong Kong archaeologists at the beginning of excavation at Lung Kwu Tan in 1989. From 1可t: Chau Hing-wah , Alex Yip Cho-hong, William Meach的n， Chiu Siu-tsun , Tang Chung, S. M. Bard , Au Ka Fat. 10.6 Recording an activi妙 jloor at Lung Kwu Sheung Tan , 1990

Salvage Archaeology; 1990-2007

the Lantau coast was discovered by members of the Antiquities Office and excavated by a team from London in 1993. The si仗， ShaLoWan ， was on a headland just opposite Chek Lap Ko k. lt had occupation debris and burials with a number of complete pots dateable to the Late Neolith眩; it also had irregular postholes with stones inside , much like those at the two Middle Neolithic sites across the water. A report on this site was published by the Antiquities Office (Drewitt 1995). A site on Ma Wan lsland , Tung Wan Tsai , was also investigated , 且rst by the Chinese University ofHong Kong team , and then partly, but incompletely, salvaged in 1994 by a group of contract archaeologists focusing on the Han deposit and shell midden at the southern end of the sand bar. The site yielded valuable information on the Han inhabitants , including their subsistence strategies. It also had Bronze Age material as well. However, in 1997 , the northern end of the site (that had been ignored by the previous groups) was investigated and found to have an important Late Neolithic burial ground with somε20 graves with pottery, stone , bone and shell artefacts; 15 of them contained skeletal remai悶. This sit亡， though important locally, was given the rather dubious but politically meaningful title of “ one of the year's top ten archaeological discoveries in China" in 1997 (which , coincidentally, was the year of the handover of Hong Kong to China). Another headland site was investigated by the Chinese University of Hong Kong team on the Lantau coast at Pa Tau Ku. On the basis of one test pit of 1 m x 1 m , the site had been dismissed as unlikely to be significant by the 1982一切 su叮叮 The team chose to excavate the entire surface of the headland down to only 20 cm to reveal the overalllayout of the site. They claimed to have found an alignment of village houses. The method was highly controversial , as it left the entire site exposed to erosion; the claim was also much disputed ，的 there were no floors , no artefact concentrations and no clear alignment of stones or postholes to indicate houses. The site w缸， however , clearly an important Late Neolithic site and was spared destruction due to efforts by the Antiquities Office ln 1991 , the office carried out an excavation on the Penny's Bay Sl峙， and , in addition to the usuallarge haul of Ming blue-and-white ceramics , it recovered fine pottery of Thai and Khmer origin as well. The following year, in near-constant rain , the Archaeological Society conducted a final salvage ex

61

62

Archaeology in Hong Kon

Malay" earthenware resembling geometric potte可

In

addition , a large

collection of nut shells , seeds and wood fragments was obtained by wet -sieving clays from the valley floor; these plant remains yielded important environmental data and were carbon-dated to around 5000 BC. The excavation also revealed a long wooden pole dating to the Sung era. Despite such extensive excavation , it remained unclear what function the site had served , but several interpretations were ruled ou t. It was not a shipwreck , as porcelain sherds were found well up the surrounding hillsides and at the back of the valley; also , it was not a farming village. There was a connection with overseas tra缸， possibly smuggling or even piracy In 1993 , one of the largest archaeological sites in Hong Kong , Yung Long at Black Point on Deep Bay, was slated for destruction to make way for a giant power plan t. As mentioned abov亡， the site had previously been excavated on a small scale in 1985. Resurvey showed that the site was massiv亡， and to salvage it was a m月jor undertaking The task was divided between the Antiquities Office that excavated the northern half and the Archaeological Society The site revealed a

附 A

r、d r、以

y

叫N

σb

U

UHU 月

fL

ρL

-l

quHU 7m Iid VG TT tnls n lr

new phase of the Late Neolithic , yielding a large quantity of potte可 and stone artefacts , many complete specimens coming from burials. One of the most important features of the site was the “ ovens" or firedclay structures presumably used for cooking. Charcoal was relatively plentiful on the site and 15 C-14 dates were eventually obtained on the Late Neolithic phase; the dates were remarkably consistent and indicated an occupation in the period 2700-2400 BC. Underlying this layer was a cultural deposit of Middle Neolithic belonging to the painted pottery phase , and it dated to 4500-3700 BC. Although only a few fragments of painted ware were found , the deposit provided potterγand stone types belonging to that phas巳 One of the most important was a stone bark cloth beater with grooves. Through the mid-1990s , the number of archaeological salvage projects increased along with the pace of construction in the territo月正 In 1997 , the Antiquities Of且ce decided that another intensive survey of the whole of Hong Kong was needed and divided the territory into 12 parcels. These were assigned to various local archaeological teams and several from the Mainland. There was no coordination between the 12 teams , although strict (and not well-enforced) guidelines were set regarding methodol

Salvage Archaeology; 1990-2007

archaeological sites in Hong Kong. And just as in the 1982-85 survey, sites that certainly were exposed and should have been discovered went unnoticed , in particul缸， the massive quarry site at Wong Tei Tung on Tolo Channel (discussed below) One of the missing links in local archaeology was at least partly removed by discoveries in 1999 and 2000. A large-scale salvage excavation by a mainland team at Ho Chung in Sai Kung found abundant quantities of Sung and Ming ceramics , believed to represent a small coastal village or "port" - more likely simply a fishing village , rather than one heavily involved in trade. Another large-scale excavation by the Antiquities 0且ce at So Kun Wat near Tuen Mun uncovered a m句 or Ming dynasty burial ground , consisting of some 15 pairs of burial jars with celadon-glazed coveri月 bowls. One jar in each pair had rice or other grain , the other jar was empty and pr吋ably contained wine. Previous句'， Mi月 blue-and-white porcelain had been reported from the are且， but no village was recorded there in historical documents. Another Ming settlement with house floors and drainage system was discovered by the Archaeological Society at the site of Mong Tseng Wai near Lau Fau Shan. These discoveries demonstrate that evidence of villages from the Ming period are indeed to be found in the territory, although their relative scarcity and lack of correlation with historical records remains something of a myste叩 A major Late Neolithic site at Sha Ha in Sai Kung was discovered in 1995 by a member of the Archaeological Socie句" photographer Jerry Sousa , while he was walking in the area. A few test pits during the 1997-98 survey indicated that , like Yung Long , the site extended over a very large area estimated at 20 ,000 m 2 When the site was threatened by road construction in 2001 , the Antiquities Of五ce commissioned no less than four teams of mainland archaeologists to conduct the salvage excavations. It proved to be the largest , most expensive and most publicized archaeological excavation ever carried out in Hong Kong. At least ten Neolithic and Bronze Age burials were discovered , along with stone-working areas and irregular patterns of postholes. Complete stone objects , bowls and complete or restorable pots were retrieved , and , quite literally, tons of potsherds and other artifactual material were unearthed. New information was obtained , including the first examples of rice phytoliths (pollen-like crystals) from a prehistoric context

63

64

Archaeology in Ho月 Kon

10. 8 Aerial view of the exω1叫ion at Sha Ha , Sai Kung and vicini砂 and Monuments Office , Lεisure and Cultural Services Department)

10.9 The excavation grid at Sha Ha in 2002

φ hotograph

by permission of the Antiquities

Salvage Archaeology; 1990-2007

ultimate importance of this m句 or endeavour , and the question remains as to whether the enormous expense for the unprecedented scale of excavation was indeed justifìed. Sadly, during the course of the project , the two most senior offìcials of the Antiquities 0伍ce were arrested for corruption; one was never charged , and the other was convicted but acquitted on appeal Finally, politics had to intrude on the local archaeological scene , and in the 1990s there were occasional references t。“the matriarchal stag丸"“ the primitive commune" and other terms from classic Marxism. As the 1997 return of Hong Kong to China approached , there were strong assertions that Hong Kong was culturally part of China from the earliest times 個 notion never disputed by anyone) , and the local prehistoric population w的“ Chinese" (akin to claiming that the Gauls o f] ulius Caesar法 time were “ Frenchmen"). Amazingly, the rather ordinary Late Neolithic burial site at Tung Wan Tsai was declared one of the ten most important archaeological discoveries in all of China in 1997. A coup d'etat in the Archaeological Society resulted in its “ localization" and virtual disappearance from the international scene. Silly claims seemed to garner more publicity than usual: a circular potsherd and a broken , crescent-shaped piece were said to be evidence of "primitive sun-moon worship"; numerous lime kilns all over the territory were stated to have been potte可 kilns (and an astounding grant of HK$4S0 ,000 was made for research into this absurd notion); and an ordinary neolithic quartz ring workshop was said to indicate that the people “ were already well integrated into Chinese culture in their admiration for jade and precious stones." Despite the ups and downs of Hong Kong archaeology in the last two decad白， the reader will surely have gleaned from the foregoing brief summary that an enormous amount of material and data has been acquired during this time. Much of it remains unpublished and undigested , as often happens when salvage takes precedence due to availability of funding. However, the basic framework of Hong Kong's prehistory and early history is well established , and the related body of knowledge is being constantly expanded - at times exponentially A chronological outline and discussion of local pre- and proto-history is the subject of Parts II and III.

65

Part 11 Prehistory of the Hong Kong Region

ThεPalaεolithic:

Early Humans in South China

The story of the 且已t habitation in the Hong Kong area must be prefaced by a consideration of the early peopling of East Asia , as a whole , and of South China , in particula r. When humans 五rst established temporary settlements in Hong Kong and other regions along the coast some 6 ,000 to 7 ,000 years ago , they brought with them an advanced neolithic culture and way of life that had evolved over thousands of years before their arrival. We shall examine briefly to what extent cultures (and the human species itselD had evolved in the East Asian region by 5000 BC , and what circumstances might have led to the first human occupation of this area Fossil evidence found during the 1920s and 1930s in North China andJava had indicated that early humans (Homo erectus) was present in East Asia for several hundred thousand years. More recent evidence suggests that the earliest humans existed in Asia between one and two million years ago , at the beginning of the Ice Ages (the Pleistocene epoch). Human and great ape fossils of this period have been found in caves in Yunnan and Guangsi provinces in South China Some of these fossils found their way into the Chinese chemis的， shops in Hong Kong in the 1930s as “ dragon bones" believed to have medicinal properties. The great palaeontologist von Koenigswald made regular visits to Hong Kong to inspect these shops , and he found a new species of hominoid among the fossils. In a lecture given to the Archaeological Society in 1971 , he descr咐ed how he visited the chemists' shops and used a Chinese medical prescription to induce the stores to show what fossils they had on hand A somewhat later (Middle Pleistocene) fossil was discovered in 1959 in a small cave near the village ofMa-pa in northern Guangdong It was the skull-cap of a more advanced species of man than the earlier

70

p吋11Storγofthe Ho月 Ko月 Region

Homoerect肘， possibly akin to , but distinct from , Neanderthal. The age of this “Ma Pa Man" has been estimated at 200 ,000 to 100 ,000 years Still more advanced fossil species of humans (Homo sapiens sapiens or modern humans) have been found in Guangsi and estimated to

date from 到， 000 to 20 , 000 years Throughout these hundreds of thousands of years , the economic life had presumably changed very little , based as it was on hunting and food collecting. Similar峙， the stone tools used by th臼e early people were rather crudely made pebble tools chipped into a rough shape , or sharp flakes struck off from such pebbles. In this part of the world , bamboo tools may also have been important At the very end of Pleistocene , some 15 ,000 to 10 ,000 years ago , profound changes had already begun to take place in both the economic patterns and the stone tools of the inhabitants of South China. In the coastal and river areas , it is probable that more reliance began to be placed on collection of shell foods: fishing techniques were developed , and voyages by raft and boat were made. Knowledge of pottery-making was either discovered or learned from a neighbouring people at this time. Thus far , the earliest known pottery in the world is from ]apan , Siber間， and South China , dating to around 15 ,000 years ago. Stone tools began to be ground and polished as well as chipped , and were made into more advanced shapes. Finally, and most importantly, certain types of plants and animals were being "manipulated" by humans; that is , they were being captured and tended in addition to being killed or collected in the wild. This process led to the appearance of agriculture and animal husbandry by at least 8000-6000 BC in various parts of China and Southeast Asia. Grains such as rice and wheat began to be cultivated at this time , laying the foundations for eventual civilization. This new way of life is generally termed "Neolithic" in contrast with the long “Palaeolithic" era. In archaeology the main material trai的 distinguishing neolithic technology are pottery and polished stone tools , but these are not always associated with an agricultural way of life. The transition from terminal Palaeolithic to Early Neolithic was first recognized in the cultural deposits of cave sites in South China and northern Vietnam. Cultures named “ Hoabinhian" and 包acsonian" (after sites near Hanoi) indicate a gradual evolution in technology that may have accompanied the first efforts at plant manipul

The

11.1 Palaeolithic tool5 from

50uthw的 t

Palaeoli山c:

Early Humans in South China

China

polished stone tool was manufactured , but , at the important early Hoabinhian site of Spirit Cave in northwestern Thailand , a number of possibly “ manipulated" plant remains were recovered. The Bacsonian has the same pebble tools as its predecessor, but is classi且ed as Early Neolithic because of the presence of pottery and partially polished stone axes. Both cultures are securely dated by C-14 to the early Holocene (1 1 ,000-8 ,000 BC) and suggest that the emergence of the Neolithic was more evolution than revolution. Similar Early Neolithic sites have been found in South China , particularly at Tsangpiyen in Guangsi and Hsienyentung in ]iangsi. The tradition of Hoabinhian pebble tools is reflected in the roughly chipped pebbles found on most prehistoric sites in Hong Kong , alongside much more advanced stone tool types Most of the Early Neolithic sites are in caves m mountamous reglO凹， but it is highly probable that the lowlands of the Pearl River and Red River valleys and the coastal plains were also inhabited at this time , although

11.2 Chipped pebble tool5 from Hong Kong

71

72

p跎historγof the Hong Kong Region

11.3 Cave with Early Neolithic deposits in Vietnam

the population would have been very small and thinly spread. It lS certa1吋 possible that , on occasion , bands did pass through the territory of Hong Kong , but no sites of their activities have yet been found. There may be geological reasons for this lack of sites and will be considered below. At any ra仗， it now seems virtually certain that the 趾st semi-permanent inhabitants of Hong Kong were a seafaring people brought into the area by the rising sea level In200斗， a site claimed to be Palaeoli由此， dating to about 35 ,000 years , was discovered on the southern coast ofTolo Harbour at Wong Tei Tung. This site was stumbled upon by archaeologist Steven Ng and a friend while fìshing. A large amount of stone debitage and some partially complete stone tools on the surface of the beach at fìrst suggested a prehistoric workshop , but further investigation of the adjacent hillslope revealed a 凹的sive amount of worked stone , and several rock outcrops further up the slope seemed to be the source. Dating of the soil by Optical Luminescence (OSL) yielded an age range of 39 ,000 to 35 ,000 years , although other dates on the

The Palaeolithic: Early Humans in South China

site were only around 7 ,000 years. When these very early dates were obtained , the site was announced to the world with much fanfare as “ a Palaeolithic quarry site." A press release described the discovery as a very slgm且cant finding that “ rewrites Hong Kong's history" (Xinhua N ews Agency 2006). According to Professor Zhang Senshui , a prominent expert on early stone tools and advisor to China's State Bureau of Cultural Relics , the site opened new horizons in the study of the Palaeolithic in the whole Southeast China region There are serious doubts about this claim , however , most notably arising from the fact that quarrying sites of this magnitude are invariably from the Neolithic. The OSL dating of soils is not very reliable for hillslope environmen心， especially when the rain forest canopy effect (protecting the soil and preventing resetting of its radioactivity) is not well known. Furthermore , it seems highlyunlikely that two separate phases of quarrying took place at the si仗， almost 30 ,000 years apart , with no apparent difference in the stone-working techniques. It is much easier to imagine the massive stone industry at Wong Tei Tung in the Epi-Palaeolithic (palaeolithic culture surviving into the Neolithic period) to Early-Middle Neolithic (around 10 ,000 to 6 ,000 BC) when sites and population were increasing in the area and stone artefacts were in great demand. A few of the stone artefacts from the site are definitely of neolithic type (rough-outs for adze and disc). But even if this Neolithic dating tur日s out to be the case , the site is still of great importance and poses a number of intriguing questions. On the other hand , if the site is truly 30 ,000 or more years old , then it is one of the most important Palaeolithic sites in Asia. Clearly，也lS issue needs to be clarified; at present , it is the subject of negotiations between the authorities and the Archaeological Society

73

The Ancient Environment and Human Movεmεnts

Most of the evidence relating to the Early Neolithic era has come from caves in the interior of South China and Indo-China. However, it is virtually certain that the low-lying river valleys and coastal plains were also inhabited during this time. The first people in Hong Kong - known from archaeological sites dated back to 5000 BC - must have been derived from such a coastal populatio日， and their first forays into what is now Hong Kong probably began a few thousand years earlier.

12.1 Deep excavation at Fu Tei Wan on Chek Lap Kok produced charcoal from 11 ,000 years ago

The Ancient Environment and Human Movements

There is no direct evidence for these early forays , but a possible indication was found during the salvage excavations on Chek Lap Ko k. At the hillslope site ofFu Tei 叭站立， bits of charcoal were found in natural soil deposits below the Middle Neolithic culturallayer. C-14 dates of around 11 ,000 and 9 ,000 were obtained on the charcoal. It is tempting to see this charcoal as evidence of humans in the general area ofNorth Lantau. While no de且nite evidence ofhuman occupation earlier than 5000 BC has been found in Hong Kong , there were almost certainly small bands of hunter-gatherers making use of the territ。可 on occasion before that time. Their use of 五re may have accidentally or even intentionally created massive forest fìres. If not due to human agency, the fìres on Chek Lap Kok 9 ,000 or more years ago must be attributed to regular lightning strikes during the dry seasons when the forest was susceptible to burning. While extremely rare in rain forests , such fìres do happen , so this question remains unresolved. 12.2 View 011 the southern tip 01 Cheung Chau; enti陀 area would have been lowlying land 10,000 years ago

12 .3 Estimated coastline in this region 10,000 years ago.

75

76

Preh帥ry o[ LhεHong Kong Region

12 .4 Extensive mud flats would have been characteristic of the area before the sea reached i的 present level

What is quite clear is the association of the earliest inhabitants with the rise of sea level. During the last Ice Ag亡， the sea level was 100-120 m below its present level , and the continental shelf of South China was a vast plain. As recently as 15 ,000 years ago , Taiwan was linked to the Mainland , and the present territory of Hong Kong was at the inner edge of this vast coastal pla凹， where it met higher ground and the topography became more rugged. Hong Kong was probably some 200 km inland from the coast at that time , although the coast was probably less of a “ line" and more a series of marsh白， islets and sand or mudflats. The edge of the coastal plain bordering a foothill reglO n rτlight have been attractive to early humans in their movement from one ecological zone to another in search of food. However, Hong Kong has no caves to preserve archaeological deposits , and open sites of this period , if any survive , probably lie buried deep under the seabed and may never be found The coastal region itself would certainly have been an attractive food source for the early people. Indeed , with its rich inter-tidal zon亡， with shell foods easily accessible on rocks in shallow water, and with fìsh and other marine life trapped in small pools at low ti世， the shore was probably one of the richest environments for food-gatherers. As the sea level rose and shoreline ecology became more diverse , even greater food resources would have been available. Travel , however, would have been rendered more diffìcult as the coastline became more sinuous and indented. Necessity led to the development of primitive boats , and the long process of adaptation to the coastal and archipelagic environment began.

The Ancient Environment and Human Movements

As the glaciers melted in the period from 15 ,000 to 4000 BC , a rapid rise of sea occurred , finally reaching its present level. During this tirr質， the continental plain Cof which Hong Kong was a part) was flood吐， the coastline moved inland , islands were created where once were hills or higher ground , river deltas began to form , and many other topographic and ecological changes took place. Evidence has been found in Hong Kong itself to substantiate the general scenario of sea-land change in this period. From Sha Tsui near High Island , wood from an alluvial (stream) clay deposit 20 m below sea level and several metres below the seabed has been dated to more than 40 ,000 years. In Lai Chi Kok Ba)月 a tree trunk from alluvium under the seabed at -14 mPD was dated to around 6800 BC , indicating that even at that late time , most ofHong Kong was still dry land. By 5800 BC , however, the sea level had reached at least -11 mPD , as shown by dates on marine shells obtained during the construction of the Mass Transit Railway Finally, shells from the valley floor at Sham Wan , Lamma Island , dated to 4400 BC indicate that the sea had by then reached to within 2 m of its present level. The importance of this geological data for archaeology is that we have a clearer picture of the events leading up to the first appearance of humans in this area. The present evidence indicates that the first inhabitants of this region were a marine-oriented population , living near and on the sea , and relying on boats for transportation. As the coast advanced slowly (in human time , but rapidly geologically) over the continental shelf, this coastal population would have moved with it and adapted themselves to the new environment created by the more broken and rugged coastline. Any inhabitants of the coastal plain itself would have been pushed back as it decreased in size. Eventually, they would have to adapt to either coastal or lowland river valley environments once the plains had been entirely submerged As the sea moved into the Hong Kong area in the period 8000-4000 BC , the coastal population probably came with it. But it is only from about 4000 缸， when the sea level had stabilized and the submergence of living sites had ceased , that we 五nd signi五cant remains of their presence. Sandy beaches began to form at many of the present day beach sites , and , in the more stable and protected back-beach deposits , the earliest archaeological evidence in Hong Kong was preserved The role of sea level rise in the initial pOpl

77

78

Preh巳tory of the Hong Ko月 Region

this idea at a meeting of the society, there was a great deal of head-scratching and dubious expressions

12 .5 D1i llíng for samples in swampy valley near Sham Wa咒 ， Lamma

12.6 Log in lagoonal clay at Chung Hom Wan dated to circa 4000 BC

It now seems well established , even though quite a number of inland sites have been discovered in recent years. Apart from the controversial quarry site discussed above , all the known inland sites are later than around 4500 BC - the earliest date for the painted pottery phase. The fact that no earlier sites have been discovered , after such intense archaeological surveying of the territory, strongly supports the hypothesis that the fìrst humans to occupy the territory of Hong Kong were brought in by the rise of sea level. A side issue that arose early on in the evaluation of archaeological sand bar sites was the claim that these were “ raised beaches" formed at a time when the mean sea level was 4-6 m higher than today This was much debated in the 1970s , as it was the fìrm view of some geologists , but archaeological sites had depths of deposit that did not seem to correlate with the claimed higher sea levels. To investigate this question , Ray Frost , an engineer with the Mass Transit Railway, Dr. T. N. Chiu , of the Department of Geography and Geology at the University of Hong Kong , and 1 set out to survey the surface elevation of several well-known sites and compare the elevation of cultural layers belonging to the same phase. What we found was that different sites had different rates of sand deposition at different tim白， du亡， undoubtedly, to local factors and not to sea level rise or fall. The same cultural phase was found at different levels on different sites , depending on the nature of each site. It was this data that fìnally laid to rest the notion of archaeological sites as "raised beaches" formed at a higher sea level. Indeed , we found considerable evidence for a stable sea level (va可ing by no more than 1 m or so) since around 4000 缸，的 each cultural phase from Middle Neolithic to Bronze Age to Early Historical is represented by occupation levels in certain sites

The Ancient Environment and Human Movements

down to +3.5 mPD , the present storm beach level. If there had been a higher sea level at any time during the last 6 ,000 years , one would expect that fact to be reflected in the elevation of cultural deposits from that period. Sand bars above +3.5 mPD (the usual storm beach level) were not in normal times subject to wave action , and cultural materialleft on the surface was preserved intact and in situ. At a few sites , basal deposits below +3.5 mPD contain gravel and cobbles in sand , and potsherds rounded by water-action , probably representing wave-reworking of the initial surface deposit Study of shells in the clay deposits of the valley floor near Sham Wan provides further evidence of the absence of a higher sea level at the end ofthe Holocene transgression. At -1. 0 to -2.2 mPD , an inter-tidal community of shells was dated to 4550-3970 BC , indicating the exact sea level at that time. This indication of slightly lower sea level around 4000 BC was strongly corroborated by the carbon dating of a log found in alluvial clay at + 1. 60 mPD beneath the culturallayers at Chung Hom Wan (and excavated in shockingly dangerous conditions that 1shudder to recall!). The log was dated to 4490-4040 BC , confirming that the mean sea level was lower than present at that time. Furthermore , shells from a low salinity marsh environment in the Sham Wan valley at +0.6 to + 1. 4 mPD dated to 2640-2120 BC , indicating that the sea level was still somewhat below its present level at that time. However, in this same period , from at least 4000 to 2000 BC , sand was being deposited on some sand banks , such as Sham Wan , as high as +8 mPD. Similarly, the inner sand bar at Pui 0 built up from +5.5 to +7.9 mPD in the last 1,000 years as a sand dune , during which time it is clear that no higher stand of sea has occurred. The sand bodies themselves were probably stabilized by vegetation , unintentional human agency, and topography, in the case of tombolos and sand bars forming against or on a saddle or low spur Some of us maintained that these “ dunes" were built up mainly, if not entirely, by wind action; others claimed that the occasional monster typhoon would have played a role in the deposition of sand. The elevated sand bodi凹， in most cases , are not clearly separated from the present beach deposits and were never actual beaches themselves In a few instances , an inner and presumed older sand bar stands on the valley floor , isolated from the present beach and storm beach , but sedimentation , drift an

79

80

Preh巳tory of the Hong Ko月 Region

mouth of the Pearl River at Chek Lap Kok the same cultural phase is found on an inner sand bar at +3.5 mPD. Early historical occupation levels on inner sand bars vary in a similar fashio立， with a + 6.0 mPD level at Pui 0 and +4.3 mPD at Chek Lap Ko k. The so-called inner bars were most probably part of an extensive back-beach sand accumulation and appear to be remnants surviving from extensive stream-cutting and erosion of these deposits. On many sites , the rate of deposition decreased sharply after reaching the maxlmum ra仗， and shifts in wind direction or velocity, the volume of stream discharge , and offshore currents may have initiated the destruction of most of the dunes. It is evident that the cultural deposit survived in the present sites , not on甘 because oft血 he 追阻 臼1盯 le E 叮 va 泣t仙l昀on 叫叫 above the wave zone but also because of the stability of the dunes themselves. Many of the present sites appear, however, to be small remnants of extensive deposits at the time of occupation. 5evere erosion of these sites due to modern human agency has also been recorded since at least the 1920s , with a host of possible or known cau間， including commercial sand diggir嚕， clearing of vegetation , cultivation and subsequent abandonment , and land reclamation. Although most stabilized sand dune sites in Hong Kong have similar topographic and vegetational features , rates of sand deposition have varied considerably through time and from site to site. With the higher sand bars of +8 to +10 mPD , no more rapid sand deposit can be demonstrated through archaeological time , while the more common +5 to +6 mPD sites also vary dramatically in the depth and nature of their cultural deposits. In the case of 5ham Wan Tsuen on Chek Lap Kok , the sand bar h的 prograded over time at virtually the same elevation , while other sites , such as Lo 50 5hing , built up rapidly in one period , with almost 1. 5 m of sand deposited in the Late Neolithic At Pui 0 , however, some 2 m of sand has been deposited since the abandonment of the site at around AD 900 , compared with 0.6 m of deposit at 5ham Wan in the same period. The survival of the archaeological sites in the elevated sand bodies is obviously of considerable importance in Hong Kong archaeology The evidence shows that , just as in modern tim白， the lives of these early occupants were inextricably bound up with the sea - in settlement patterns , travel , fìshing and probably trade as well.

Hong

Kong這 Earliest

Site

Ironically, the earliest securely dated site in Hong Kong is not in a

back-beach deposit but on a small hillock overlooking a narrow inlet on the island of Kau Sai Chau. This 曰 s lte seems tωor跎 eprese日tthe 且r臼st “干 pl昀 oneer explor跎旬毛巳甘 位叮r呵3γ5 E

approach its

presεntlεvel..

In early 1993 , plans to construct a public golf course on the northern half of Kau Sai Chau began to take shape and receive publicit乎 In April , the Hong Kong Archaeological Society wrote to the Golf Association to inform them of the existence of archaeological sites on the island and the need for a survey of the impacted area prior to construction. In October, the golf course plan was approved by the Executive Council , and , in December, the Antiquities and Monuments Offìce commissioned the Archaeological Society to conduct a twomonth survey and test excavation. Two important sites were ide目前ed during this projec t: a Han site rich in pottery on a low hillslope and a stone working site on top of a small rounded hill. The latter site was to prove quite important in spite of the paucity of its material The fìrst task of the survey was to review previous archaeological discoveries on the island. The only published reference to any archaeological work on the island was an article published in the Journal of the Hong Kong Branch of the Royal Asiatic Soci的 in 1962 by Mrs M. W Welch It was noted that two of the three sites described in the 1962 report were outside the golf course are且. The location of the other, numbered 3 in her report , was not adequately described in the article , but notes in the Archaeological Society's fìle had the site plotted in two locatio凹， one with a question mark beside it. The wntten comm凹的， however, seemed to suggest a third location to

82

p吋1以 ory o[ LhεHong Ko月 Region

13.1 Excavation at the “ Welch site" on Kau Sai

Chau in 1993

自治'

的HYH5

，仗的， ε

td，叫 nm 叫叫

陀的

的“

PL

ωEW

5

勻 JVut

2

--MG

-ft

l:1 91-

the southwest of some house ruins. This site was later found during the su叮叮 bya combination of luck and diligent surface search , in a different location altogether from any of the three mentioned above. The initial evidence alerting us to the site was in the form of a few very small (2-3 cm in length) stone flakes of a chert-like stone. The site became known as the “Welch site." The island had formerly been used by the British forces (army and na可T) 的 a firing range , and unexploded shells had been reported. This posed a threat to archaeological works , and detailed inquiries were made before the fieldwork began. Conflicting advice was received regarding the degree of hazard this might pose to excavation; one expert from a government department advised to stop excavation if a metallic “ ping" was heard; an army explosives expert warned that such a sound might be the last thing we ever heard! It was decided to err on the side of caution - a metal detector was borrowed and sweeps over the excavation floor were made every 10 cm of depth. During the walking survey of the possible locations recorded for this si仗， a few flakes of alien stone suitable for artefact manufacture were found in an erosion gully near the top of one of the hills. Two test squares were excavated , and each yielded similar cultural material small flakes of a chert-like stone that was clearly non-local. Square B also yielded a roughly chipped pebble tool similar to specimens from Neolithic sites in Hong Kong. Square A yielded two larger cores from which flakes had been struck; one core seemed to have a rough shape and roughly squared side. No pottery was found in either square nor on the surface in spite of careful searches and considerable erosion scars in the area. The absence of pottery also had been noted by Mrs. Welch

Ho月 Kong's E叫 est

Site

Square A also provided a very clear stratigraphic context for the cultural remains A grey topsoil extended to a depth of 20-25 cm , followed by a yellowish brown clayey soil to 50-60 cm , grading to reddish brown and more clayey below 60 cm. The flakes and cores were found in the third layer, which on other sites would have been considered basal , sterile 吐ecomposed granite" or residual soil. This context might be taken to suggest a considerable a阱， but it is not conclusive. A similar sequence was also observed in square B, but the first two layers were much thinner (1 0-15 cm each). The total number of artefacts recovered from this site is small , and the significant feature of the site is the absence of potter手 The absence of pottery is the major factor , as all other known Neolithic sites in the territory have potsher缸， even if only a small number. Another factor is that the site is not a source of stone material for tool-making , but rather the stone has been brought to the site from elsewhere. Tools were then made an d!or sharpened on the si仗; there ,

must have been other activities on the site besides stone working. The assemblage is precisely what one might expect of a simple , open Paleolithic site and not at all what one would expect of the well-known Neolithic inhabitants of this region. At the end of the test excavatlo丘， the dating of the site was thus uncle缸， though Neolithic seemed more likely It was decided that further excavation was needed Previous reports mention polished stone artefacts found there in the 1960s , and one partially shaped stone artefact from a test pit seemed to be intended as a stone adze

13. 3 D昕cult digging and paltry rewards!

13. 4 Polished adze and chipped pebble tool from the “Welch site"

83

84

Preh巳tory of the Hong Ko月 Region

Returning to the area for further excavation , it was found that a had burned off all the scrub cover, making surface inspection much easier. A large sandstone polishing stone , an adze roughout and a fully polished stone adze were recovered from the surfa白， and the "neolithic" nature of the occupation was con五rmed at least to the extent that polished stone tools were presen t. The excavation was expected to yield pottery that would enable the site to be assig日ed with confidence to one of the known phases of the Neolithic in Hong Kong. Several squares of different sizes were excavated , although the drier weather made it very tough digging in the hard ground. The mater凶 wasmau勾引one flakes , only a handful or so for each square A few more stone artefacts were also unearthed , notably another polished adze , chipped pebble tools and a flat , notched pebble. All of these tool types are found in the local Neolithic. However, no pottery was discovered. The stratigraphy of the site was consistent with the phase 1 test pits; stone tools and a very small quantity of flakes were found in the uppermost portion of a reddish yellow clayey soil , below grey topsoil and yellowish clayey soil. The number of artefacts was extremely small , indicating that the site was not used for a very long occupation. There were no hearths or other kinds of evidence relating to site use. Small bits of charcoal were found , but only in the areas where the number of flakes was highes t. What emerged as unique and highly important about the site is the early dating obtained on two of the charcoal samples: around 5200-4800 BC. Doubt has been cast on whether these dates really hill 五re

indicate the age of the site occupation , since the polished adzes and chipped pebble tools are indistinguishable from Middle Neolithic types. These doubts are unfounded: the pieces of charcoal were found scattered about only where the number of stone flakes was highe哎， and no charcoal was found in areas with no flak白， indicating that the charcoal was associated with the human activity and did not occur naturally in the clay Samples of charcoal from two different squares gave C-14 results that overlap each other. There is no backward contamination that could cause these samples to give results older than they really ar亡， and other scenarios (such as very old driftwood being used as firewood) that might produce old ages for two charcoal samples from one site are so highly unlikely they can be discounted The site does not prove tha

Ho月 Kong's E叫 est

was present The site was probably a temporary work-station , as the cultural deposit was very sparse. The ar凹的 not a source of stone material for tool-making , so the stone , probably already fashioned into tools , was brought to the site from elsewhere. Tools were then sharpened on the site. One would imagine there must have been other activities on the site besides stone working , and it is difficult to imagine why the ubiquitous pottery was not brought to this site if it is indeed Neolithi c. The combination of the earliest dates on a culturallayer and the absence of pottery therein seem too strong to be a mere coincidence Another similar assemblage of chipped and polished stone tools without pottery has been found at the base of a Middle Neolithic cultural layer at Shek Pik. This layer is found in a beach deposit , indicating that the sea had reached its present level and that the deposit mu鈍， therefo攻， date to 4000 BC or perhaps a few centuries earlier at most Another si仗， at Sha Tau Kok near the border, has also been claimed to have a pre-ceramic culturallevel , but the evidence is less convincing than Shek Pik. But if the earliest material at either of these sites does indeed represent the same cultural phase as Kau Sai Chau , then one would have to conclude that the 五的t inhabitants of Hong Kong , from around 5200-4800 BC , did not have potter手 The evidence is still not strong enough to make this a firm conclusion , but if it does eventually prove true , then several theories (including some of mine) will need to be revised. Because of the still tentative nature of this earliest phas亡， we will refer to it as a local phase of the “Ear甘 Neolithic" but with the quotation marks to indicate that it is not yet con五rmed as a distinct cultural phase. This small hillock is the first archaeological site in Hong Kong , to our knowledge , that had no pottery at all , except for the perplexing Wong Tei Tung site discussed in a previous chapter. Although its stone tools are similar to those of Neolithic sites , the radiocarbon dating has shown that the 訊社lch site is earlier than other known sites in Hong Kong. The human occupation of the Hong Kong area has been pushed back by the C-14 dating of this site to at least 7000 years BP - almost 1,000 years earlier than previously established. The evidence from the carbon dating was quite defini仗， but , strangely, neither the Museum of History nor the Antiquities Office seemed to recognize it , continuing to refer to Hong Kong's "six thousand years of history

Site

85

Middle Nεolithic: ThεFirst Inhabitants

The first inhabitants for whom we have quite clear and widespread evidence were a coastal population , sheltering in the bays and lagoons of the islands with anchora阱， and making extensive use of the back-beach areas. Most of the Middle Neolithic sites are in beach or stabilized dune deposits 4-6 m above mean sea level - well above the storm and monsoon tidal zones. More than 20 C-14 dates indicate an age of 4500-2700 BC for the Middle Neolithi c. The main sites are Sham Wan and Tai Wan on Lamma , Fu Tei and Kwo Lo Wan on Chek Lap Kok , Shek Pik on Lantau , Chung Hom Wan on Hong Kong Island , Yung Long on Deep Bay and Sha Ha in Sai Kung From the pattern of sites on islands , it is apparent that the earliest population was adept at travel by boa t. The excavated food remains suggest that they may have already developed sophisticated and possibly deep-water 五shing techniques. At Sham Wa立， bones of large fish and remains of stingrays and sharks were uncovered in the cultural layer. Bones of wild or domesticated pigs were also found An unusual site at Sai Wan on Cheung Chau is on a 20-m ridge well above the beach and may have been a small , partly agricultural settlement. Unfortunate垠的 in most ofHong Kong's prehistoric sites , no house floors or structural remains , apart from irregular posthol白， have been found , nor have any plant remains survived to clarify the subsistence base. The thickness of deposits sugg白白， however , that the inhabitants had established themselves in the area on a semipermanent or at least seasonal basis. The first inhabitants of our region had a very good knowledge of pottery-making and produced vessels of various shap白， types of clay and decoration. Generally, the pottery consists of two categories: a coarse-tempered ware with stamped cord impressio凹， usually roundbottomed and often blackened from firing and cooking; secondly, a more sophisticated ware consisting of dishes and bowls made from

M創 le

Neolithic: The First

a fÌne paste clay, usually incised around the foot-rim. This duality in the pottery, consisting of a coarse basic type and a fÌ ne paste (more re fÌ ned variety) , continued to the end of the Bronze Age. There is a similar persistent duality in stone tools: simple , chipped pebble tools co-existed with fÌnely polished tools of high quality stone. A few sites , notably Chung Hom Wan , also yielded very fÌnely painted potte可 with the same forms and fÌ ne paste as the other Middle Neolithic sites. It was believed at fÌ rst that this rare class of pottery might have been reserved for ritual or funerary use. But the third phase of excavations at the site ofHac Sa Wan in Macau showed clearly, and somewhat surprisi日gly， that the pai日ted ware phase was earlier than the typical “ Sham Wan type" of incised chalky war宅， This was later con fÌ rmed by excavations at Chek Lap Kok , Yung Long and Cheung Chau in Hong Kong where this phase was seen in isolation. C-14 dates from several sites indicate that the painted pottery phase dates from about 4500 to 3500 BC , whereas the Sham Wan phase dates around 3500-2700 BC. This dating is supported by a thermo-luminescence date on incised pottery samples from Sham Wan that gave 2900 缸， and one on similar ware from Sai Wan that gave 2700 BC The excavations at Shek Pik in 1986 further complicated the picture. A cultural layer, clearly pre-dating soft geometric pottery (thus Middle Neolithic in local terms) , was found to have hardly any fÌ ne paste pottery, but a variety of incised patterns on coarse wa哎， of a type virtually unknown at Sham Wan and all other Middle Neolithic sites. C-14 dating of two charcoal samples averaging around 3400 BC , however, placed this layer in the same period as the others. It is unclear whether this Shek Pik culture represents a transitional phase or regional variant not observed at the other sites , or perhaps a separate ethnic group Stone tools made in this period included a variety of types: rectangular polished stone adzes (asymmetrical axes) , slotted rings and other ornaments , chipped pebble tools and flattened pebbles used as grinding stones. While the pebble tools are rather primitive and amorphous , the polished tools exhibit a high degree of craftsmanship Other tools and ornaments were made of shell , bone and probably also of wood and bamboo. Perforated fÌ sh vertebrae were used as beads: a double perforated shark tooth resembling a face was recovered from Sham Wan. Drilled , grooved and notched net sinkers were also found

lnl吋

88

p吋11Storγofthe Ho月 Ko月 Region

14.1 Painted pottely typical 0] the early phase 0] the Middle Neolithic

14.2 Incised pottery characterizes the ~ later phase 0] the Middle Neolithic ...

M創 le

Neolithic: The First

lnl吋

14.3 Shouldered adze ]rom PO Toi 仙。 tograph courtesy

0] the British Museum)

Interestingl其 in both the stone and pottery industri白， a pnmlt1ve and an advanced element can be distinguished. The coarse cordmarked potte可 and crude chipped stone tools were little advanced from the Early Neolithic cultures dated in other parts of South China to 10 ,000-6000 BC. The fìne paste pottery and polished stone adzes and rings a況， on the other hand , typical of the more developed Middle Neolithic of the southeastern coas t. One tool type that is highly characteristic of this general region - the “ shouldered" adze 一- has been found at many Hong Kong sites Some of the religious customs of the early people have been revealed by archaeology Sham Wan yielded two human skulls and a few associated long bones of ordinary burials , one of which was accompanied by a polished stone adze. Nearby were dozens of skull fragments of man and animal (deer): study of the fragmentary rerτlains established that they represented 10-15 human individuals and had been intentionally cremated (along with some parts of a deer) at temperatures of 800 C or more. Therefo白， both ordinary burial and cremation of the dead were practised at this si伐， apparently at the same time , possibly indicating differential burial according to social rank , type of death , etc. Some archaeologists mistakenly argued that this constituted e\吋ence of cannibalism; in fact , the evidence for high temperature cremation was overwhelming (cracking , curling , colour change , etc.) , meaning that no flesh would have been left on the bones. Human sacrifìce by cremation remains a possibility 0

90

p吋11Storγofthe Ho月 Ko月 Region

14.4 Pits in the basallayer at Fu Tei Wa 仇 ， Chek Lap Kok

Placement of whole pots and stone tools in ve可 small pits may indicate that secondary burial was also practised , but no human remains have been found in these pits. Burial grounds have been identified from this period at Fu Tei Wan and Kwo Lo Wan on Chek Lap Kok and Sha Ha in Sai Kung , all producing quite a number of complete vessels of 五ne paste incised ware. Shek Pik is probably also a burial ground , but a variant of the Sham Wan phase. For the earlier painted pottery phase , burial sites are believed to be represented by the complete vessels found at Tai Wan by Fr. Finn and at Chu日gHom Wan Another very unusual feature found among the Sham Wan human remains was several jawbones with a neat hole drilled through the血， the pieces may have been worn as “ trophies" of battle. A similar practice was observed until recently in Taiwan and the northern Philippin白， where aboriginal people drilled holes in jawbones taken from enemies killed in battle. These objects were then used in rituals as gong handles. As we shall see in later periods , the original inhabitants of South China were closely akin in many ways to the peoples of Southeast Asia - an ethnological and geographical region that many consider to have extended in ancient times as far north as the Yangtze River Bas凹. The early coastal settlement as Homutu in Zhejiang provin凹， which yielded rice dated to 5000 BC , had houses built on stilts - a type of dwelling still found in much of Southeast Asia. Another characteristic of the local “ painted pottery phase" is the presence of bark cloth beate店， used to make clothing out of bark. This cultural

M創 le

91

Neolithic: The First lnhabitants

d可 m 改

U

JH 『pν

p

仇H

門UUn 叮

19

叮戶

yedLT

、心，

ω 泊 ，

siptuy

，

4

trait also widely occurs later in Southeast Asia and probably originated on the southeast coast of China. However, by the next Sham Wan phase of the local Middle Neolithic , spindle whorls are found and bark cloth beaters are not , indicating that spun fibre had replaced bark as a source for cloth. While the Homutu and later Neolithic cultures of the Yangtze Delta are different in many ways from those of Southeast Asi日， there are many shared featur白， such as the cultivation of wet-field rice , stilt-dwellings and the shouldered adze. It has become increasingly clear, since the 1970s , that Southeast China had its own independent development during the Neolithic and was not influenced by the early Lungshan cultures that were thought , at one time , to have originated in the Central Plains of North China and spread southward in a masslVe “ Lungshanoid expansion."

ZUGFJ 5dn EILUfL zl qud mub GF :tz

Middle Nεolithic Burial SitεS on Chεk Lap Kok

In 1960 , a church worker found a complete Six Dynasties glazed jar at Fu Tei Wan in a low sand bank behind the beach. It appeared that the site would be another typical back-beach sand bank site. Two weeks of excavation in 1992 provided a large quantity of Middle and Late Neolithic , Tang and Sung pottery, but yielded little information since the material had been redeposited either by water or human actlvl你 One interesting artefact recovered was a bark beater, used during the Middle Neolithic to make cloth from tree bark Surface searches on a 15-m plateau overlooking the beach yielded more material , and this site proved to be very importan t. It had a well-preserved Middle Neolithic occupation area (a seasonal camp or village). Many irregularly shaped holes or pits were found in the subsoil , giving the floor a pock-marked appearance. These holes generated much debate among the excavation team membe芯， and various theories were proposed to account for them - from large tree roots to burrowing animals or insec伍， to digging for clay, 15.1 The site at Fu Tei Wan

Chek Lap Kok

M吋心

to postholes. Eventually, it became clear that most of these pits were related to human habitation , the m句 ority as postholes but others as shallow hearths and even for secondary burial N 0 particular pattern of postholes could be discerned , but many of them had large stones inside and were clearly intended to support posts. Areas of cooking and stone working were identifìed , but they did not seem to correlate in any meaningful way with the postholes. No compacted floors were found even though the soil seemed ideal to pr臼erve such evidence of intensive use , as , for example , would be the case of the floor inside a small hut The most important discovery at the site was the burial or ritual are且， which had a number of complete pots and stone tools placed in ten small pits. No human remains were found , probably due to the acidity of the ground water. It is believed that the complete artefacts were placed as offerings with human remains in secondary burials (reburi日1 of the bones a few years after death). The burial pits a兒， m every ca仗， too small for an adult human body, and the idea that secondary burial was practised is by no means certain. Purely ritual offering is another

15.2 A pair of corded pots in situ at Fu

possibilit乎

Some of the pits contained an unusual assortment of goods. One very shallow pit yielded two complete corded pots , one set in the mouth of the other Another pit , affectionately known on site as the "adze-hole ," had two pots , a chalky bowl with foot rim , and nine polished adzes. Two sherd-lined pits had no objects apart from an adze and a solid pottery leg , but these might have been part of the backfìll and not intentionally placed. At the end of the excavatio日， a bulldozer was engaged to knock down the small buildings on the site and remove the topsoil. Several more buriaVritual objects were found , including a very well made highpedestalled cup with a complex incised pattern. This piece is one of the fìnest examples ofMiddle Neolithic ceramics from the South China coastal region

15.3 The two pots after cleaning

Iìα 訊而ln

94

p吋11Storγofthe Ho月 Ko月 Region

Carbon dating of several charcoal samples from the hearths gave a general dating of around 3800-3400 BC for this sit亡， but these dates may have been slightly skewed by the presence of older natural charcoal in the soil KwoLoWan On the other side of the island , about 20 minutes by foot , was the small bay ofKwo Lo Wan. In the 198此， neolithic pottery and stone tools were observed in cultivated 五elds on the lower hillslope near the beach. When excavation started at the site in 1992 , mosquitoes were a m句 or nuisance. To counter them , the villagers lit fires and then placed green leaves on the fire to create great plumes of smoke 一 the cure being almost as bad as the disease! Excavations revealed a similar cultural deposit to that of the plateau site at Fu Tei Wan. The pottery was clearly of the same period , namely Middle Neolith眩， and posthol白， hearths and stoneworking areas were also found. The site is much smaller than at 15.4 Excavating a large corded pot at Kwo Lo Wan

15.5 The ω rded pot after restoration

Fu Tei Wan , and the deposit is shallower, only 20-30 cm deep. Once again , a buriaVritual area was also discovered , and the same type ofburial seems to have been practised. Complete coarse pots , fine paste bowls and polished stone adzes were placed in seven small , shallow pits , probably as offerings to accompany secondary burials. One pit also contained a polished stone ring. Four of the pits had a more regular shap亡， between 1.2 and 1. 4 m in length. Unlike the burials at Fu Tei Wa口， these pits seem to have been for prima可 buri泣， although they are generally not long enough for a full-length adult burial. The bodies may have been flexed. Other pits were devoid of any artefacts , but had similar shapes and may have been burials that were not accompanied by grave goods Carbon-14 dating of two charcoal samples gave an age of around 3400-3000 BC , slightly later than Fu Tei Wan. But the pottery is virtually identical to that of Fu Tei Wa立， so the two sites must have been inhabited at almost the same time. The two sites together provided a 五ne and unique assemblage of complete pots and bowls for this phase of the Middle Neolith眩， normally represented on other sites only by potsherds or small portions of the complete vessels

A Surprisingly Early Phasεof thεLatεNεolithic

Archaeological remains were fìrst discovered at Yung Long in 1974 by members of the Hong Kong Archaeological Society during a general survey of the Black Point and Lung Kwu Tan area. Several coarse and soft geometric sherds were found ，的 well as fragments of polished stone tools. During the government-commissioned survey of Hong Kong in 1982-85 , two small test pi的 were excavated in the central part of the si仗， and it was claimed (rather rashly) that a cluster of fìred-clay slabs represented a Late Neolithic pottery kiln Minor damage to the site by a contractor led to a larger excavation in 1986 that was conducted with government funding by the same archaeologist (Brian Peacock) who had conducted the survey The excavation failed to provide any conclusive evidence as to the function of the fìring areas and fìred-clay remains. In early 1992 , it was announced that the site was required by China Light and Power Company for a mammoth new power station In spite of the ve可 limited testing and excavation previously carried out on the si仗， it was deemed to be of major importance , primarily because of the large area of potential deposi t. A major rescue operation was conducted with two independent teams from the Hong Kong Archaeological Society and the Antiquities and Monuments Offìce As this operation unfolded , it became increasingly apparent that the site was one of considerable signifìcanc亡， even greater than previously believed , and a further phase of excavation was conducted in the sweltering heat of June and July At the tim亡， this project was the largest archaeological project ever conducted in Hong Kong The site was divided into northern and southern halv白， with the Antiquities Of五ce and the Archaeological Society taking responsibility for investigating each. The fìrst few weeks of excavation on the site

96

p吋11Storγofthe Ho月 Ko月 Region

16.1 One of the initial test grids at Yung Long in 1994

16.2 Excavation in progress at Yung

L口 ng

A Surprisi月lyEa向 Ph忱。 f the Late Neolithic

were devoted to the systematic testing of the sand bar and areas behind it. A rigorous surface search of the lower hillslopes and valley floor was also conducted ，的 well as augering at selected locations in the valley Results from these testing activities revealed that most of the sand body and lower hillslope in the northern sector had an extensive cultural deposit , whereas the southern sector seemed rather poor. However, it would later turn out that the southern sector had a richer deposit ，的 well as an underlying Middle Neolithic cultural layer not present to the north. One of the test pits yielded a Late Neolithic buri址， and soon others were being uncovered. A large excavation plot was then opened up on both sides of a house. The owner, Mr. Chan , very kindly allowed , in places , the excavation to within 1-2 m of the house A total of six probable burials were found in a tight cluster. There were other pits that had the exact size and shape of burial cuts , but contained no artefacts. Later, after Mr. Chan moved out , the house was demolished , and a collection of fine stone artefacts was found beneath it. Another area to the north was believed likely to contain burials , also likely to be clustered together. The challenge was to locate the

16.3 Final stages ofexωvatlOn in the foreground , with land works already in progress on the slope above

97

98

p吋11Storγofthe Ho月 Ko月 Region

16.4芳pical burial o.fferings consisted o.f a corded po. t and a pair o.f sl o. tted sto. ne rings

burial clusters with the time available. Several random and systematic testing strategies were considered; it was fìnally decided to excavate alternating squares in every other ro w. None of the initial test pits revealed any burials. A backhoe was then engaged to take several large trenches down to the buriallevel at the bottom of the Late Neolithic layer. As a result , another fìve burials were discovered. A few other burials were discovered in the northern sector. The contents of all the burials were similar, almost predictable. The offerings consisted of pottery, both coarse and chalky war白， and polished stone rings; half of the burials had rings , and fìve had a pair of rings. Only six burials did not have potter乎 Strangely， none of the burials had stone projectile points or spindle whorls , though these artefacts were quite common in the activity layer. Their presence in the graves might have provided a clue as to the sex of the interred A total of 20 probable burials were discovered , but , in many cases , determination of the exact number is clouded by the lack of clear burial cuts and , occasionally, by the proximity of complete artefacts. Only a few burial pits could be discerned in whole or par t. Other obvious grave-shaped cuts of around 2 m x 0.6 m had no grave goods but were probably burials without survivable offerings in pottery or stone. Two

16.5 The burial po. t after cleaning

graves were side-by-side , parallel and equally endowed (one pot in each) , possibly representing a husband and wife Considering the estimated 20 burials in terms of 吼叫us in society, they would seem to be closer to egalitarian rather than ranked , except , of course , for those that had no goods at all. These , however, may have had wooden or other items made of organic material that did not survive Some burials were accompanied by a single pot , others by a single ring , while the "riche缸"

A Surprisingly

Ea向 Phase

of the Late Neolithic

individuals had three pots and two rings in one case , or one pot , two rings , one bracelet and one adze in another. There were no clearly dominant or highly ranked individuals evident in these burials , judging from grave goods alone. There W缸， however , one cluster of beautifully made large polished stone yueh axes that was either a burial or a cache. If a buri泣， it may well have been a chieftain. A similar 五nd at Fu Tei Ngau , not far from Yung Long , yielded two very large (even more impressive) yueh stone axes , each greater than 30 cm in length.

16.6 Large ， β nely polished "yueh" stone axes from Yung Long 酬的 tograph by permission of the Antiquities and Monuments Offic亡， Leisure and CulturalSε川ices Department)

16.7 Layout of burial goods from one section of the site

99

100

Prehistorγof the Hong Kong Region

16.8

Reζ ording

a burial pot

The stratigraphy of the Yung Long site was relatively simple , and the two Neolithic phases of occupation were easily identifìed and studied in isolation. The Late Neolithic culturallayer was generally 50-100 cm thick and marked by black or greyish brown sand and the extensive occurrence of the fìred-clay oven remains. Firing areas and “ ovens" were freque旺， and some were still partly intact with upright clay pillars (formerly called “potstands") , clay grills and portions of intact sides and bases. These seemed to have been related to ordina可 domestic activiti白， and no evidence whatever was found to support the notion that the site had been used for fìring potter手 These clay features constituted the most obvious special trait of Yung Long The excavations in 1992-93 at Yung Long showed very clearly that these features were widespread over a huge area; were rather insubstantial structur白， which , in all cas凹， had collapsed shortly after use (except occasionally for sides and base); and had occurred through the Late Neolithic layer with other types of typical occupation debris All of the clusters of fìred clay were rather compact , and , judging from the amount of material in each , it seems probable that there was no “ superstructure" at all , and the features are simply small clay-lined pits or troughs with props (“potstands") and stones built into the walls as supports. Unfortunately, no decisive clues were unearthed to reveal the function of these ovens. The excavations yielded a number of charcoal sampl白， and C-14 dating of 12 samples from the Archaeological Society's excavation gave ve可 consistent results , indicating that the site was inhabited 五rst at around 4100-3600 BC (the Middle Neolithic layer; six samples) and again around 2700-2400 BC (the Late Neolithic layer; six samples). Three samples from the 1986 excavation gave similar results. The results from the Middle Neolithic are somewhat less preci仗， owmg probably to the smaller sample size and greater age. Nevertheless , Yung Long now stands as the best -dated prehistoric site in Hong Kong. Perhaps the most unexpected aspect of the new dates is the verγ early position of what may be termed the “Yung Long phase" of the Late Neolithic. The site yielded a very well-fìred geometric pottery, just slightly lower fìred than Bronze Age stoneware , and Yung Long was at fìrst believed to be a borderline Late Neolithic/Bronze Age site because of the advanced nature of its potter于 this no

The

dates clearly refute

A Surprisi月lyEa向 Phase of the Late Neolithic

gave a weighted average of 2130-1770 BC , and 13 shell samples gave a weighted average of 1910-1740 BC. Clearly, these sites are later than the Yung Long phase , and they are characterized by a great variety of more elaborate geometric patterns The Late Neolithic at Yung Long is marked by the same twin pottery traditions as the Middle Neolithic , name峙， a well-fired , fine paste ware and a heavier, cruder coarse ware. The burials have provided many complete examples of the potte可 from this phase. Most of the fine ware has basketrγ-type decoration and verγlow， non-functional foot-rims. Some of the vessels are ribbed or have sharp carinations at the shoulder; there is 日ot much variety in either shape or decoration. Some of the fine ware is remarkably high fired , approaching stoneware in quality. Testing of three of the hardest pieces revealed that two were fired above 1l 00o C, while the third o W晶晶red at around 1080 C. Clearly, the kiln technology had already advanced to the point that high temperatures could be reached and proto-stonewares could be mad巳 This information , together with the dating , is perhaps the most important new evidence from the Yung Long site , since previously the Late Neolithic “ soft geometric" phase had been distinguished from the Bronze Age by its softer fine paste potter其 generally of the consistency of chalk The common polished stone artefacts in the Late Neolithic phase at Yung Long were the leaf-shaped projectile poi旺， the slotted rings , and quadrangular, stepped and shouldered adzes. Interestingly, although many specimens of the projectile point were unearthed , none were found in a burial. The slotted rings of the burials were of very fine quality quartz or cryst泣， and well made. Adzes were of types well known in the Late Neolithic. Ceramic spindle whorls were also common in the Late Neolithic layer: dozens of complete or near-complete specimens were found , but , agai丘， none were placed in a burial. Notched pebbles presumed to be net weights were also found in significant quantity. Flakes were very rare , considering the rich deposi t. Unfortunately，的 is the case for most Hong Kong sites , virtually no direct evidence was obtai日ed regardi日g subsistence. No food remains were unearthed , apart from two carbonized seeds of a local variety of plum. There is thus no data to address the question of subsistence base , especially the existence of agriculture during the period. \九lhether or not the occupants ofYung Long practised farming themse

10 1

1 02

Prehistorγof the Hong Kong Region

16.9 Spindle whorls from the Late Neolithκ

specialists (probably specialist clans or villages by the Late Neolithic) for certain industries such as potter乎 Yung Long proved to be a remarkable si仗， one of the most important in the territ。可Its importance resided not only in the dating , well-fired potte ry, burials and oven featur凹， but especially because each phase of the Neolithic present on the site had not previously been seen in isolation. Yung Long , therefore , added two new phases to the prehistoric chronology ofHong Kong. The Middle Neolithic was well known from the Archaeological Society's work at Sham 叭lan in the 1970s , and painted pottery was known from a number of sites. But Yung Long was the first site where the painted potte可 culture was isolated and dated. It was also the first site where the early phase of the Late Neolithic was so dramatically represented. After Yung Long , several other sites were discovered with Yung Longphase materials , notably the headland site at Sha Lo Wan and the m月jor site at Sha Ha

Classic LatεNεolithic: Culmination of thεStonεAge

In the Middle Neolithic , the population of this area was seafaring , living along the coast on the beaches and sand banks in sheltered ba戶，

fìshing and perhaps also tending small gardens and domesticated animals. The culture was fully neolithic - pottery and polished stone tools were made with considerable craftsmanship , alongside simpler types. The fìnal period of the Neolithic in this area is marked by a new type of potte可; new stone tools and ornaments , and an increase in the number of coastal sites. In most important respec俗， however, life seems to have continued as before. The characteristic pottery of the La te Neolithic in Hong Kong and much of South China is a grey ware made of fìne clay and stamped with n哎， bask缸， herringbone , d凹nond ， trellis and other geometric designs. This geometric pottery replaced the fìne painted and incised ware of the earlier phases and probably evolved directly from it , as is suggested by parallels in type of clay, method of manufactu肥， fìring and some elements of style. The elaborate decoration was applied by either carved paddle or roller. Coarse potte叮 with cord-mark impressions continued as befo況， with only minor changes. In addition to the different decoration , the geometric potte叮 has new shapes not previously seen: large round vessels with small ring feet or indented bottoms , spouted vessels with loop handles and sharp angular forms. As we have seen from the pottery at Yung Long , the fìring temperatures of the pottery had increased to more than lO OOOC , indicating that some form of elementary kiln had begun to be used; the Middle Neolithic pottery was probably fìred in the open at a range of 600-800 o C. These elements of the pottery technology - higher 五ring temperatures , the use of kilns , and stamping with carved or moulded decoration - would assume great importance

1 04

Prehistorγof the Hong Kong Region

17.1 Typical soft geometric pot of the classic Late Neolithic

A

17. 3 50ft patterns

geometri 正，

17.2 50ft geometric pot with loop handle from Hoifung

1 05

Classic Late Neolithi c: C山nination of the Stone Age

in the development of bronze metallurgy at the end of the Neolithic period. Puzzlingly, however, in the terminal phase of the Late Neolithic in Hong Kong , there was a marked decrease in the fÌ ring temperatures , and the fÌ ne paste potte可 is much softer and friable. The decoration is much more elabora仗， with many fancy geometric patterns. This is the classic Late Neolith止， as represented by the pottery from Tung Kwu described by Scho fÌ eld in the 1930s. New forms of polished stone artefacts were also being made: the stone ho dagger, large stone axes with perforation , fluted points and T-section “ flanged" stone ring are the most characteris山. Polished stone adzes , knives and projectile points were made in great numbers , and rare types of stone were obtained from areas quite distan t. This trade in raw material is another aspect of the economy that later played an important role in the rise of bronze metallurgy, relying as it does on tin and copper that usually are found in different regions At the same time , more primitive types of artefacts continued to be made as in the earlier periods: the simple coarse cord-marked pottery was still used for cooking , and the rough pebble tools were still used , probably for a wide variety of scraping and chopping functions. Most of the sites that had been occupied during the Middle Neolithic continued to be used in the next phase. One of the preferred types of site during this phas巴， as Scho fÌ eld fÌ rst noted , was the tombolo island , formed by a sand bar linking two smallland masses.

17. 4 Fluted stone projectile pomt

17.5 The Chau with

sit ε at

Sha m

ex正 avatlO n

prog陀俗， 1976

1 06

Prehistorγof the Hong Kong Region

Islands of this ty阱， such as Tung Kwu and Sha Chau off Castle Peak , and Cheung Chau and Siu A Chau off Lantau , have two sheltered bays on either side of the tombolo and provide good refu阱， regardless of the prevailing wind. This protection was strikingly demonstrated during the excavation at Tung Kwu in ]une 1975. Our launch had anchored on the western side of the tombolo , and we went ashore in calm water in the morning. By mid-afternoon , a violent thunderstorm came roaring across the Pearl River estuary, and the sea on that side got verγrough. The coxswain brought the launch around to the eastern side , which by contrast was quite calm , and we were able to board the small dingy with no difficulty The tombolo islands were also used for burial on occasion. Both Tung Kwu and Sha Chau had artefact clusters indicative ofburials. Sha Chau had the higher quality material; several complete soft geometric pots and a beautiful white stone ko ceremonial dagger were retrieved from the sandbank. The ko was picked up from a smallledge where it had been exposed by erosion but had not yet fallen into the sea below. It was probably saved for posterity by only a few days. A new type of site that appears in this period is the promont。可 or headland si仗， jutting out from the land mass into the sea. These sites have a relatively flat surface and are ideal for looking out to sea in all directions. The main ones are Pa Tau Ku and Sha Lo Wan on Lantau and PO Yue Wan on Cheung Chau. The deposit is not thick , but finds are quite numerous. At the PO Yue Wan headland when first discovered in 1985 , hundreds of sherds were exposed on the surface in rather thick vegetation , but visible through the leaves and branches. A possible predecessor is found on the opposite arm of the bay at PO Yue Wan - the Middle Neolithic site on the ridgetop at Sai Wan. It remains unclear what use these sites served 1 7. 6 The stone ko from Sha Chau

Classic Late Neolithic: Culm

1 7. 7 The headland site at PO Yue Wan , Cheung Chau 6心附 dfrom a正叩ss

the bay)

17.8 Grooved polishing stones believed to be "arrowshaft polishers"

As in the previous period , many other Late Neolithic sites are found at the mouths of small bays or lagoons. The main ones are Shek Pik and Pak Mong on Lantau; Tai Kwai Wan and PO Yue Wan beach on Cheung Chau; Sham Wan , Lo So Shing and Sha PO Tsuen on Lamm日; Sha Ha in Sai Kung; and Tung Wan Tsai on Ma Wan. These sites are also well sheltered but were quite possibly selected , m some cas白， for agricultural use. Brackish water rice could have been cultivated in the tidal marshes and mudflats of these lagoons; this practice was observed by the 五rst Han settlers in South China at a later period. Two sites (Sha PO Tsuen and PO Yue Wan beach) have yielded bones of immature pigs - the 五rst firm evidence for the presence of domesticated animals in the area

108

p跎historγof the Hong Kong Region

17. 9 Polished bone points are very rare in the local Neolithic

However, it is quite probable that the sea continued to be the source of food. At Sham Wan , thousands ofbones were found of the marine catfish and another large fish known as the headgrunt , some weighing as much as 5 kg and measuring up to 1 m in length. Remains of shark , stingr耳其 dolphin ， crocodile and sea turtle were also found in this horizon. Side-notched and drilled-through pebbles were used as net weights. By this tim亡， it is clear that the people were capable of extended sea voyages , and they seem to have had a deepwater fishing technology Excavations at PO Yue Wan on Cheung Chau revealed similarly large quantities of catfish and headgrunt remains , prompting one archaeologist to suggest that an early form of commercial fishing may have been practised. The site also yielded two new artefact types polished bone points (possibly used in fishing) and chipped shell scrapers that were also found later at Tung Wan Tsai. Three series of C-14 dates on shell samples from PO Yue Wan gave conflicting results , but the average date was around 1800 BC All of the sites described above have the classic Late Neolithic potte可 with a variety of geometric patterns , some rather complex As discussed in the previous chapter, an earlier phase was identified in 1993-94 at the Yung Long site on Deep Bay These two phases of the Late Neolithic have been quite securely dated: the Yung Long phase at 2700-2400 BC and the classic geometric phase at roughly 2400-1500 BC. Rather surprisingly, while the later phase has much greater variety and sophistication in the geometric patterns , its pottery is less well fired. The Yung Long phase has a range of basketry-type m月jor

Classic Late Ne口lit 叫岫 出血 tlhi此 c:

Cu 叫 1址lmina 前t仙l間叮 lOn

patterns and a few simple geometric typ白， but its pottery is much harder and better fìred. Yung Long and Sha Ha had burials yielding complete pots and stone rings , but rarely had stone adzes or other artefacts. A cache (probably burial offering) of beautiful ceremonial jadeite axes was also found at Yung Long. Artefact caches representing presumed burials from the classic soft geometric period have been found at Shek Pik , Sha Chau , Sha Ha and Sham Wan. Human remains with grave goods were found at Tung Wan Tsai; seven of the skeletons were relatively well preserved N othing is known of the political or social organization of these Late Neolithic people , but , in northern China , the Shang dynasty appeared towards the end of this period , and tribal confederations with some degree of central authority probably had been formed in northern Vietnam at the same time. It is virtually certain that the Late Neolithic inhabitants of the coastal regions were the ancestors of the Yueh tribes mentioned by later historical texts , but archaeology sheds more light on their geometric pottery cultur呵e than on their soci祉， political or ethnic features.

of the 5t口 neAge

109

Bronze Age: On thεMargins of Civilization?

The 1芯e of Bronze Age cultures (around 1500-500 BC) in South China has traditionally been seen in the light of the impact of Shang and Chou civilizations on the more primitive southern “ barbarians. " In recent decades , however, dramatic evidence has been brought to light of equally early bronze-working cultures in northern Vietnam , northern Thailand , Sichuan province and the 10wer Yangtze River Basin. Therefo肥， it is not at all clear to what extent (if any) the civilization of North China influenced the development of metallurgy and of more advanced politica1 and socia1 organization in the south. There is no evidence that the know1edge ofbronze was closely linked with the spread of the other attributes of civilization , such as cities , writing , distinct economic classes , kingsh中， priesthood , etc From the Late Neolithic to the Bronze Age in the Hong Kong 訂閱， life appears to have continued in much the same way as in earlier times. Most of the same sites were occupied; beaches on small she1tered bays and sand bars on the is1ands continued to be the preferred type of site. For reasons not yet known , the importance of the tombo1o is1and and the head1and as a shelter or activity area seems to have declined. Both simp1e and refìned pottery and stone too1 types continued to be made Fishing remained a major subsistence activit手 At Sham Wan , severa1 thousand bones of the headgrunt were virtually the on1y type found; the marine catfìsh seems to have disappeared from the diet or from the 10ca1 environmen t. Fishing and mass netting of spawning stocks may have been quite intense. Bronze fìsh hooks and pebb1e net weights have been discovered on many sites. Shell collecting was another m句 or source of food , and shell remains from Sham Wan indicate that sandy, muddy and rocky shores were all being exp10ited for shell food resources. Deer, p嗯， dog and crocodi1e remains have been identifìed from this phase. Finally, the Sha PO Tsuen site on

Bror

Lamma yielded a rhinoceros tooth - evidence that the rhinoceros was still present in 50uth China then , although it became extinct by historical times. This , undoubtedly, was a trophy piece brought to the island , as opposed to a lone rhinoceros hunted down on Lamma 凹， as someone suggested in all serious白白， brought across from the Mainland alive In addition to the coastal sites , lower hillslopes near small valleys were new activity areas in this period. Hills near Yung 5hu Wan and Mo Tat Wan on Lamma , Man Kok Tsui on Lantau , and 50 Kun Wat near Castle Peak (the first archaeological site investigated by Heanley and 5hellshear in the 1920s) have yielded thousands of stone and pottery artefacts datable to the Bronze Age. These sites were probably related to agricultural activities , such as rice cultivation in the valleys , rather than to 且shing or shell collecting. Animal husbandry was also almost certainly practised , as indicated by the discovery of ox figurines made of fired clay The local Bronze Age is characterized , of course , by bronze artefacts , but , at most sites , these are few and often fragmentary Fish hooks , projectile poi立的， knives and axe heads are the most commonly found typ白， but moulds for the manufacture of vessels , bells and hairpins have been found. In Guangdong , there was a debate for many years concerning whether the province had a “ true" Bronze A詐，的 distinct from the Early lron Age beginning around 500 BC. Although numerous bronze objects have been found , it had been argued by some "traditionalist" archaeologists that these were imported into the area. The excavation in 1989 of the site of 5ha PO Tsuen on Lamma provided important data on these questions and permanently laid to rest the debate (such as it was). A clearly defined Bronze Age cultural

的 1 Bronze dagger found on a hillside above Sheh Pih

悔M

cu mg

m 肌

明古

削叫

m

正lJ

GL 山

1i

2m 80

nu7

mm rd

112

Prehistorγof the Hong Kong Region

layer was identified , with several moulds for casting bronze axes in intimate association with high-fired geometric pottery Also found were traces of bronze slag on the inside of large potsherds that had been prepared with special surface treatment. Several “ pot burials"

18. 3 Bronze fish hook and projectile point from Sham Lamma

叭lan.

18. 4 Bronze weapons from Ho嗯 Ko愕 sl1es φhotograph of the British Musωm)

ωurtesy

were also discovered at the si峙， consisting of large pots deliberately broken into pieces and buried as a cluster for unknown reasons. Two charcoal samples from the casting area were dated to around 1200-800 BC. The site was the first bronze-casting site to be discovered in Guangdong. In addition , two instances of bivalve moulds in casting position were found in burials on Lantau and Chek Lap Ko k. There is no doubt that this was a bronze-working culture , and its dating is certainly earlier than 500 BC. Other aspects of the material culture in the Bronze Age continued , little changed from the earlier periods: geometric designs were the most prominent pottery motif; coarse stamped potte可 was still the main utilitarian ware; chipped pebble tools were still used for a variety of tasks; and polished stone adzes and rings were still made in great profusion. A m句 or change , however, did take place in the manufacture of the better-made potte可 The design of kilns had improved to the stage that very high temperatures could be reach吐， and much of the geometric pottery was fired at 1200-1300 C and is extremely hard. A variety of circle , spiral , double- F, diamond and other geometric patterns decorated the potte可 While no local kiln site has been found for this ware , it is assumed that only a few highly specialized kilns run by specialized clans produced this stoneware for a large area. One such kiln that manufactured double-F pottery has been excavated in Guangdong 0

Bronze Age: On the Margins

tl n

c3

σb

、/

l《叫 pu

nu

的山

Other aspects of the improvement in ceramic technology should also be noted , especially the use of the wheel and glazes . The rims of larger vessels were thrown on a fast wheel , after the body was built by hand , perhaps with the aid of a slow wheel. Smaller stem cups were entirely wheel-throw丘， and the better specimens indicate a mastery of fast wheel techniques. Glaze is rarely found on the larger pots with geometric patterns , but frequently occurs on the stem cups that do not have impressed patterns M句 or burial sites have been found at Man Kok Tsui , Shek Pik and Hai Dei Wan on Lantau; Tai Wan on Lamma; and Kwo Lo Wan on Chek Lap K仗， but only Tai Wan yielded a substantial number of bronzes. An excavation there in 1994 by the Chinese University of Hong Kong team yielded the first example of a jadeite yazhang sceptre , a ritual artefact found on several Mainland sites of the Shang dynasty era. It was found in a cache or burial along with beads and other elaborate grave goods in what was most likely the burial of a chieftain. A C-14 date possibly associated with that culturallayer at Tai Wan gave around 1700 BC , probably too early for the beginning of the local Bronze Age. At Sham Wan , a collection of several dozen very well-made polished stone kniv白， rings and ornaments were found 凹， around and under a pavement of large stone slabs , probably the burial of a highly ranked individual or an

YS fL

月u

ι 叫 戶

5

j1 8m MM uz (}

viLU

18.6 Double-F pot from Kwo Lo Wan , Chek Lap Kok

114

p吋1以 ory o[ LhεHong Ko月 Region

18.7 Patterns on hard sherds

geome削C

altar of some kind. This may have been a stone slab buri泣，

and , if so , was the fìrst and only one so far discovered in the territor芋 In contrast to the rather meagre burial grounds found in Hong Kong , a very large burial ground of the Bronze Age was excavated in Guangdong's Boluo county in 2000 (Guangdong Provincial Institute 2005). A total of 302 burials dating from late Shang to the Spring and Autumn period (approximately 1200-476 BC) were unearthed; a bronze bell and a bronze tripod vessel , both of defìnite northern af五mtl白， were included in two of the tombs. These items were dated by typology to the Western Chou 18.8 Stone rings oJ Bronze Age type

period 0121-771) , probably a good estimate for the dating of double-F potte可 as well , although some archaeologists argue that it continued to around 500 BC. My view is that double-F began in the late Shang , El ourished during the Western Chou , and went into decline by 700 BC. The late Shang dating comes from a musical instrument 1 saw in the archaeological collections of the Academia Sinica in Taipei - a stone ching from one of the Shang royal tombs. It is engraved with a design indistinguishable from doubleF. Researching the matter further , 1was pleased to 五nd

Bronze Age: On the Margins

18.9 Stone ching musical instrument with double-F pattern 制lO tograph court的

ofAωdemiaSini叫，

that a prominent local scholar of Chinese art , Jao Tsung-yi , held a similar view. He wrote “ The so-called double-F pattern . . . appears to be a local expression of Hsiao-tun [late Shang] art" Oao 1954). The dating of the Bronze Age has implications for the elaborate rock carvings found at six locations along the coast in Hong Kong , since they are related by style to that era. Most of the carvings are in remote locations on the coas t. The only carving closely associated with an archaeological site is the one at Shek Pik. The most recently discovered can吐ng at 叭Tong Chuk Hang is in a completely differe凹， upland setting on rock faces next to a small stream. Some of the rock art designs are very similar to the geometric decoration on the potter手 notably spirals , and one can confidently date the rock carvings to the Bronze Age. They probably had a ceremonial or religious significance The carvings are discussed further below The Bronze Age inhabitants , just as those of earlier periods , do not seem to have had permanent settlements in the coastal regions , unlike , for example , those at the Boluo site that seems to represent a fair-sized village. In spite of the dozens of sites belonging to this period , and , in spite of several debatable claims , no definite house structures or settled village sites have been found in Hong Kong Postholes are often found , sometimes with large rocks inside them , but no indisputable pattern forming a square , rectangle or circle has yet been demonstrated. N or has any hardened or sunken floor been recorded , nor has any signi五cant difference in artefact density inside and outside the claimed “ house structures" been ShOWll. The postholes

Taipei)

116

Prehistorγof the Hong Kong Region

18.10 Cluster of sherds at Sha PO Tsu凹， Lamma

18. 11 Pots restoredfrom the sherds

were probably employed to support tents , lean-tos or simple huts easy to erect and dismantle. It would seem that this area was still only 駝的onally or occasionally occupied; where the permanent bases (if any) were located remains a myste可﹒ Local archaeology would seem to provide corroborating negative evidence to an ancient text which reported that the Yueh of South China “ made their homes upon the water."

Two

BronzεAgεBurial Sitεs

Man Kok Tsui was discovered in April1958 by S. M. Bard , a member of the University Archaeological Team. Excavations were conducted later that year by the team , and the site proved to be of considerable importance. A report was published in Asian Perspectives (Davis and Tregear 196 1) Like many other archaeological sites in Hong Kong , Man Kok Tsui at the mouth of Silvermine Bay exhibits the topographic features that appealed to the prehistoric population 一 sheltered bay with good anchorage and a sand bar enclosing a shallow lagoon (now a small valley). What proved to be unusual about this site was the enormous quantity ofbroken pottery throughout the area and the discovery of a number of complete pots on the small hillock headland to the north It was recognized that most , if not all , of the material was datable to the Bronze Ag巴， even though the amount of bronze found was ve可 small: a fìsh hook , three small arrow points and a fragment of a vessel The most striking feature of the site at the time of discovery, and even today, is the large amount of pottery that may be found on the surface in the fìelds and on the hillsides. Bard remarked that “ in 1958 it was di伍cult to walk over certain areas of the valley without stepping on a fragment of prehistoric potter手" Years of cultivation in the valley and erosion on the lower hillslopes had exposed thousands of potsherds and stone fragments ，的 well as dozens of complete stone artefacts Recent terracing and cultivation of the hillslopes had also revealed several complete pots. As the story goes , one of these pots was being used by a village woman as a flower jar on her windowsill. On passing by and spying this vessel , members of the team asked to be taken to the

118

Prehistorγof the Hong Kong Region

19.1 Excavation at Man Kok Tsui in 1958

spot where it was found , and a brief probe brought up two more Bronze Age vessels in perfect condition. Excavations at this hillock to the north of the main site produced 12 complete vessels , ranging in height up to 40 cm. All of the pots have a similar net -impressed decoration. From the same hillock , a number of complete stone adzes and rings were also found. In other areas of Man Kok Tsui , a wide variety of decorated pottery was recovered , most of the patterns being of spiral , diamond and other geometric types. One question still unresolved is whether the complete net -stamped pots from the burial area are of a slightly later age than the main occupation since none of the elaborate geometric pottery of typical Bronze Age type (for exampl亡， double-F or spiral decoration) was found among the complete pots

19.2 Bronze Age pots from Man Kok Tsui

Two Bronze Age Burial Sites

In the valley and on hillsides to the south , a wide variety of stone implements was found , including polished adzes and axes , chipped pebble tools , pebble grinders and slate knives. Evidence was uncovered of the production of polished stone rings on site: rings broken at various stages during their manufactu眩， cores bored out from the centre of the rin軒， and smoothed round pebbles used for polishing the interior of the rings. However, four of the six complete rings discovered at the site were from the northern hillock , in association with the complete pots. The various parts of the site were thus interpreted as “ working ,"“habitation" and “ burial" areas. Another possible interpretation for the survival of so many complete pots was offered: they could have been storage vessels placed on flat ledges cut into the hillside and were eventually abandoned or forgotten. However, burial is the most likely explanation , for several reasons: the presence of associated complete stone artefacts; the likelihood that the larger pots would not have remained intact if not intentionally buried; the small size of some of the pots; and the dif五culty of access to the area for ordinary storage It is interesting to note that the burial ground at Man Kok Ts凹的 outside the main habitation area in a hillside to the north of the valley This contrasts with the Bronze Age burials at Shek P虫， Tai Wan and Hai Dei Wan that were in the same sand bank also used for living activities It is , however, in line with the early historical practice of hillside burial By the end of the local Bronze Age , around 700 BC , Man Kok Tsui was abandoned , though later material is found on the site dating from the Tang dynasty era. A grave dated t。“ the sixteenth year of Guangxsu" (1 891) was also recorded. Apparentl民 the sense of feng shui a吋 spiritual response to the natural setting had changed little in the intervening 2 ,000 years - the grave was located in the same northern hillock where the Bronze Age occupants of Man Kok Tsui had also buried their dead

19.3 Large pot being removed at Man Kok Tsui

11 9

12 0

Prehistorγof the Hong Kong Region

Hai Dei Wan The small bay at Hai Dei Wan , north along the coast from Man Kok Ts凹， was fìrst visited by archaeologists in 1958 , and numerous artefacts were found in the low eroding sand bank near the beach It was immediately apparent that the site would merit further investigation , as it yielded not only a quantity of prehistoric potsherds , but complete objects as wel1. Among the initial fìnds were several adzes , a flanged stone ring and a bronze fìsh hoo k. Four test squares were excavated in 1968 , and one square yielded several complete pots , stone adzes , rings and a spear poin t. The report noted that , although no trace of human remains was unearth付，“the special fìnds were grouped in such a way as to suggest at least the possibility of a single buri日1." In 1976 , the land was included in plans for the DiscoverγBay developmer立， and four seasons of excavation were organized by the Archaeological Society, ending in April1979. A wealth of Bronze Age material was recovered and published in a detailed report (Williams 1980). In 1982 , the site was destroyed in landscaping for Discovery Ba手 α

川

山

叫

1i

陀

叮心

UVP ‘

G

叫吟

DW

Hm7 npn lG Mm

flJmh GE

正， h)

W

4

ρι

MU

。Jan

-TLpi

Two Bronze Age Burial Sites

More than a dozen clusters of complete

0句 ects

were found ,

but , just as in the original 1968 test pits , no human remains were discovered. It is virtually certain , however, that the site is a burial ground , possibly beginning in the late Neolithic owing to the presence of soft chalky geometric and coarse corded vessels. But the main and most impressive materials are from the Bronze Ag亡， including 17 stem cu阱， some with green glaze and pot此時'marks. Severallarge double-F vessels were unearthed; other pots had alternating bands of lozenge and net impressions. Other finds included a bronze knife , a bronze axe with woven straw matting adhering to its underside (probably from a mat on which the burial objects were laid) , several flanged stone rings , adzes and projectile points. A nearly complete ceramic figurine of an ox was also recovered. Clearly, the site proved to be a rich one in terms of artefacts. The extent of the deposits was not confined to the sand bar, but continued into the next agricultural terrace cut from the hillslope colluvium This significant area was verγnearly missed , as a single test pit failed to yield any material of note; auger prob白， however, soon came upon a small complete double-F po t. Unfortunately, the site did not provide final confirmation of its use as a burial ground , nor was the stratigraphy sufficiently clear to detect any possible phases of burial activit手 A C-14 date of around 1700 BC is difficult to relate precisely to any of the grave goods. The burials at this site seem to be slightly earlier than those at Man Kok Tsui , judging from the prevalence of round-bottomed vessels and the

19.5 Eχcavatíon ín progress at Haí Deí Wan

121

122

p跎historγof the Hong Kong Region

19.6 Bronze axe head and being removed

stone 叫ngs

common occurrence of double-F and other geometric patterns at Hai Dei Wan. The material from the two sites have many similarities and probably are situated squarely in the Bronze Age , roughly 1500-700 BC , though neither site has provided any conclusive evidence on the question of dating The objects from the two sites attest to the sophistication of the Bronze Age craftsman , potters and metallurgists. Together with the dozens of other known sites , we have a clear picture of the local Bronze Age material culture

ThεRock

Carvings

Of all archaeological remains found in Hong Kong , the rock carvings are the most impressive visible evidence of an ancient human occupation of this area. Generally, at living sites of the same period (Bronze Age) , no structures are found , and no alterations have been made to the landforms. Apart from the broken pottery and other occupation debris , the people left little mark of their presence at these sites However, they went to con日derable effort to carve designs into the volcanic rock along the coast , sometimes at remote points and barely above sea level While there are still many unanswered questions about the meaning and function of the carvings , they are of great importance as examples of the art and religion of Bronze Age people in this area. This importance has been recognized by the Hong Kong government , which has declared all the rock carvin軒 的 monuments under the Antiquities Ordinance and has taken steps (sometimes poorly considered) for their permanent preservation and display. The existence of the carvings has been known to the local people for a long time , especially the fisherfolk in settlements near them. The first known record of a carving (Tung Lung) is a one-sentence entry in the Sun On Country Gazetteer in 1819 stating that “ at Fat Tong Mun , there is a dragon design carved in the rock face." However, until1939 , this record and the general popular knowledge of the rock carvings escaped the attention of the early archaeologists here. As mentioned above , in that year, the scholar-collector Chen Kung-jit was digging at the site at Shek Pik on La ntau when he was told by the villagers that there was an engraved rock nearby “where people of old used to bu可 their treasure." He was taken to a small rock outcrop at the

124

p跎historγof the Hong Kong Region

20.1 Tung Lung

20.2 Shek Pik

20.3 PO Toi

20.4 Wong Chuk Hang

The Rock Carvings

20 .5 Lung Ha Wan

20.8 Kau Sai Chau

20.6 Big Wave Bay

σb

可/

叫

月U 弓4

u

C3Hd u

125

126

Prehistorγof the Hong Kong Region

base of the hill Qust below the present reservoir) and there observed a series of square spirals and circles cut into the rock just above ground level. The villagers claimed that another carving had formerly existed across the valley, but it had long ago been destroyed when lightning struck and split the rock Chen proceeded to dig up the whole area around the carving and found , what he termed ,“ a good harve仗" of Bronze Age material. Unfortunately, his notes are not clear enough to indicate exactly where he dug or whether the area had been used as a living site or burial area. At any r缸， Shek Pik is the only carving to be directly associated with an archaeological site The carvings at Tung Lung and PO Toi were next to be archaeologically 跎corded， in the 1950s or 1960s, although it is notclearp記cisely when or by whom they were 吐iscovered" for archaeology There were reports and photographs of the PO Toi carving in the 1960s in Chinese newspapers; the first scholar to report on it appears to have been Professor Jao Tsung-yi of the University of Hong Kong. A team from the Hong Kong Archaeological Society visited the PO Toi carving in 1970 , and it was then reported in the English newspapers. This led to the discovery of another carving at Tai Long Wan (Big Wave Bay) on Hong Kong Island A police officer, Mr. B. Hai阱， sent a photograph of the carving to the Archaeological Society after seeing press reports about PO Toi. Also in 1970 , a new can吐ng in abstract style was reported on Cheung Chau by Dr. Charles Peng , a geologist at the University of Hong Kong. Then , in 1976 , after reviewing my old notes regarding the western coast of the remote island of Kau Sai Chau as an ideal place to look for carvings , followed by a few hours of search , 1 discovered a small emblem-like carving there. In 1983 , a group of carvings at Wong Chuk Hang near Aberdeen came to light when it casually was mentioned by villagers at a district committee meeting Also composed of spirals and abstract elements , these carvings are in a totally different setting - in a narrow ravine just above a small stream. Unlike all the other Bronze Age carvings , these are in upland terrain , quite distant from the sea 吋ings in an upland setti日g consist of squa 缸r尚它 E Another group of carv 5叩 o-ca 址 alled “冶 gameboa 盯rd" patte 叮rns. These carvings may be lron A詐， as there are similarities to patterns on the pottery of that period. Some are located in almost inaccessible locations , for example , two carvings on a very steep slope high above Shek Pik rese

12 7

The Rock Carvings

similar patterns in Europe and Asia have been dated to various early historical periods; in Egypt , they appear as early as 1800 BC. The issue was settled by the discovery in 1985 in Macau of several “ gameboard" carvings in close association with a more complex and certainly ancient carved pattern. The early lron Age dati月 seems likely

之 0.9

20.10 Macau

20.11 Negative Macau carving

Finally, in 1978 , an apparent carvi月 was reported by a hiker on a coastal boulder at Lung Ha Wan near Sai Kung. The hiker brought photos to me , and , at fir哎， 1was quite excited. An on-site visit the next day, however, raised serious doubts in my mind , and , later, a number of geologists concurred that it was probably the result of purely natural eroslO n pro白白白， though highly unusual. Controversy followed when this “ pseudo-carving" was declared an ancient monument , displayed and protected by the Hong Kong government

01 a

Tung Chung

rubbing

01 the

128

Prehistorγof the Hong Kong Region

All six of the coastal carving sites are only 3-8 m above mean sea leve l. Shek Pik is unusual in that the carving is now somewhat removed (1 00 m) from the sea , but it formerly was behind a beach and sand bar that may have had a small Bronze Age settlement. The Cheung Chau and Big Wave Bay carvings are on rock outcrops near a beach , while the others are at remote and inhospitable points along the coas t. Clearly, these carvings indicate an intimate relationship with the se且， while those at Wong Chuk Hang may have marked the source of a perennial stream The dating of the petroglyphs rests mainly on the association of the geometric style with the ceramic and bronze decoration of the Bronze Age. All probably belong to the same period since they have ve叮 close similarities in their use of geometric and zoomorphic elemen怡， as well as in their manner of execution. While the soft potte可 of the Late Neolithic does have simple spirals and square meande芯， the closest relation of the rock art is with the more elaborate spirals of Bronze Age ceramics. There is a complex , multiple-spiral pattern on pottery that is very close to the PO Toi design; another pattern of square spirals on coarse pottery of the Bronze Age is identical to one of the Shek Pik designs. Finally, on a bronze dagger found near Shek Pik is a central face or mask above a series of spirals; a similar design may be seen at Big Wave Bay, where spirals make up part of the body (viscera?) below the head ，的 well as ornamentation to the left ln summary, while a Late Neolithic dating cannot be entirely ruled out , the evidence points strongly towards a Bronze Age date (around 1500-700 BC) for the carvings. The coastal group are interrelated , but it is quite possible that they were made at different times and represent changing artistic trends within a few centuries. For the gameboard carvings , a date of 300-100 BC has been suggested , but that date is much less supported than the earlier dating of the main group of carvings. The most intriguing question about the carvings is also the most diffìcult to answe r: What was their signi五cance to the people who made them? Rock paintings and carvings are known in many parts of the world , and China has tens of thousands. They are universally associated with ritual and religious beliefs. There is no evidence nor reason to believe that these designs were executed merely “ for art's sake." It has been argued elsewhere that some petroglyphs record epic events , but even the

The Rock Carvings

beliefs may have been held by the Bronze Age inhabitants of Hong Kong , there are interesting para11els in the use of similar geometric and animal symbols - for clearly ritual purposes - on the early drums of southwestern China and Vietnam Furthermore , we have noted that most of the carvings are in remote are缸， others are near possible vi11age or spring sites , and a11 (except Wong Chuk Hang) are near月 the sea and just above sea level. It is interesting to note that today, at many remote points of this same coastli眩， the only evidence of human presence are the Tin Hau temples and sma11er nameless shrines , frequented by fisherfolk seeking (or giving thanks for) good catches and a safe journe芋 The same temples and shrines are found , of course , in the base areas as we11. There are temples in the vicinity of the rock carvings themselv白， in or near the main areas of shelter and fisherfolk settleme凹， for exampl亡， the Tin Hau temples at PO Toi and Cheung Chau , and the Hung Shing temple at Shek Pik. Others are found in settings distant from any population centre at points of entry or departure for the fishing fle仗， for exampl巴， the Tin Hau temple at Fat Tong Mun , opposite Tung Lung , or on islands situated in fishing grounds , for example , the Tin Hau temple at Sha Chau off Castle Peak coas t. The setting of the Water Spirit shrine at Stanley and the Lin Fa Kung temple in Causeway Bay is uncannily similar to that of the Cheung Chau carving , that is , on a rock outcrop separating two beaches. It seems highly unlikely that these similarities in setting between the ancient rock carvings and the temples of the present-day fisherfolk are simply coincidental. It would be straining the evidence at hand to propose that the surviving stone cults or use of stones in shrines are directly descended from the rituals that surrounded the rock carvings There is probably only a [;刮目， far-removed relationsh甲， just as there may be in the siting of temples a remnant of the religious impulses that once dictated where rock carvings should be done. The ancient religion may have involved totemism , animism and other belief systems long abandoned. The 且gures in the carvings may represent protective or harmful deities to be placat吐， tribal emblems or totems , or gods of the sea on whom the people depended for sustenance. The folk religious practices we see today have undoubtedly been so transformed through time , adapted to circumstances , and influenced by and intermingled with other traditions that only

129

A Missing Link in Local

Prεhistory

It is useful , at this point , to review the dating of each major phase (except for the later dynastic periods). The table below is derived from C-14 dates on charcoal , the 凹的t reliable material for dating The number of samples is given in square brackets. There are also approximately 40 additional dates on shell , most of which support the chronology given belo w. The boundary dates should be considered as estimates only, since the margin of error for carbon dating is around 100 years on either side Palaeolithic -

no known occupation; one site proposed but much

disputed Neolithic" -

吧arly

5200-4900

BC

[2]

Middle Neolithic - 4500-2700 BC painted pottery phase 4500-3500 BC [10] incised pottery phase 3500-2700 BC [12] Late Neolithic - 2700-1500 BC Yung Long phase 2700-2400 BC [15] classic soft geometric phase 2400-1500 BC [8] Bronze Age - 1500-700 BC [16] Early Historical periods - 220 BC-AD 907 Han dynasty 220 BC-AD 206 [4] Chin-Tang dynasties AD 265-907 [25] This chronology clearly illustrates that there appears to be a m司jor gap between prehistoric and historical phases. There is continuity between each pr前lÌstoric pha哎， both in the C-14 chronology and in the material culture. The major gap from 700 BC to about 100 BC does not seem to be a distortion caused by too few or randomly skewed C-14 dates. A total of 21 samples from Bronze Age contexts (16 charcoal and 5 shell) 5沮lVI玖 of have been dated. Four samples are aberrant and probably intru芯凹

AMissi可Link

the remaining 17 dates , 14 cluster at around 1600-800 BC. The two samples with late dates are 780-380 BC and 368 BC-AD 24. They are particularly suspect since the cultural context is precisely the same as the other samples and is characterized by high-fìred pottery of a very specialized style (double- f, inter-locking spirals , lozenges and other elaborate geometric designs). Two interpretations are possible: the two late dates are due to contamination; or, they represent the lingering of this Bronze Age culture on a few sites , perhaps with some minor stylistic evolution not yet isolated from the main phase. The former interpretation seems much more likely It would appear that there was a signi且cantchange after 斗， 000 or more years of near-continuous and thriving occupation of the territory, marked by strong continuities from one phase to the nex t. From about 700 缸， there was a rather sudden near-abandonment of the area. Even if one takes the dating estimate by some Guangdong archaeologists of 500 BC as the end of the double-F phase , there is still quite an obvious gap. This hiatus seems so clear from the C-14 chronolo gy, and it is also revealed by the virtual absence of so-called “ Union Jack" or "rice character pattern" pottery in the Hong Kong region. This Early lron Age pottery (also called “ asterisk pattern") is dated by Guangdong archaeologists to roughly fìfth to second centuries BC. Only a few isolated sherds of asterisk ware have been found in Hong Kong , usually associated with Han pottery, particularly the "seal-impressed-over-net" variety and the very short rim style. In a recent article summarizing the archaeological work in the territory since 1997 , the curator for archaeology at the Antiquities Of五ce ， Kevin Sun (2007) , wrote Up to the present , there has not been any discovery of definite Warring States or Western Han sit白， so the upper limit of the historical period should be Ch'in . . . There is no evidence relating to how the prehistoric culture disappeared or how it relates to the Han culture

21.1 “UnionJac眩，

in Local P吋listory

or asteηsk pattemed pot

131

13 2

Prehistorγof the Hong Kong Region

21.2 Han wu-chu coins

21.3 Han pot with seal impressions resembling wuchu coins

However, for most of central and eastern Guangdong there is no such gap or decline in archaeological sites , and the Early lron Age is quite vigorous ther丸 with hundreds of sites of asterisk pottery Up the coast from Hong Kong , however, Maglioni found in Haifeng district a pattern of sites similar to Hong Kong's during his surveying in the 1930s: dozens of Neolithic and Bronze Age sites , no asterisk

AMissi可Link in Local P吋listory

21.4 Excavating a sherd cluster at the Han site on Kau Sai Chau , 1994

pottery sites , and one Han site. Just north of Hong Kong , the in1and Shenzhen region a1so bears witness to a similar sltuatlO口， although it does have at 1east one site with a substantia1 Early lron Age cultura1 deposit interming1ed with Bronze Age materi泣; it a1so has severa1 modest Han sites and Han buri日ls The evidence indicates that the 1arge popu1ation suggested by the number and richness of Bronze Age sites in this coasta1 region had dispersed. The reasons for such a virtua1 abandonment of what had previous1y been prime rea1 estate supporting a thriving popu1ation through millennia can on1y be specu1ated: eco1ogica1 change creating adverse conditions (perhaps ma1aria) , shifting triba1 confederation boundaries creating a "no man's 1and ," or simp1y the erratic and random changes in demographics. None of these specu1ations seems very appealing , and there is , at present , no satisfactory exp1anation for the significant depopu1ation that is indicated by the evidence Furthermore , the inhabitation of the Hong Kong region does not suddenly reappear in force during the Han era. Twenty years ago , 1 wrote that the Han w的 am句 or missing link in Hong Kong archaeo1ogy for呵， despite the 1arge chamber tomb at Lei Cheng Uk , there were simp1y no sites with more than a few sherds of Han potte可 This situation has now changed somewhat ,

之 1 .5

Han pot resto陀dfrom the sherd cluster

133

134

Prehistorγof the Hong Kong Region

and there are a few more substanti祉， though still rather minor, Han sites. Pak Mong on Lantau probably dates to late Western and early Eastern Han (益的t centunes BC and AD). A few other sites date to Eastern Han (五rst two centuries AD): Kau 5au Chau off 5ai Kung , probably one farm homestead lasting a few decades; Tung Wan Tsai on Ma Wan , with a thin shell midden layer, probably representing a fìsherfolk site; and a cache of about 100 Han coins at 50 Kun Wa t. There are other verγminor sites with a mere handful of Han sherds. While these sites have provided evidence of sparse human presence during the period concerned , they have hardly fìlled the void. \入7hen compared with the wealth and sheer numbers of Bronze Age sites in Hong Kong , they are very few and meagre Concerning the Tung Wan Tsai site , the excavators wrote “ At Tung Wan Tsai we have what appears to be a transition from prehistory to history, a record of combined continuity and gradual evolutionary chang亡" (Rogers et al. 1995). This is absolutely not what we have: the published report clearly shows that there are two main cultural components to the site. One is Bronze Age and dated by three C-14 dates to between 1700 and 900 BC , while the second is Han , dated by coins , ceramics and two C- 14 dates that fall within the second century BC to fìrst century AD There is no occupation and a m司jor chasm between these two phases. A thriving local population reappears rather suddenly in abundance during the fourth and 且fth centuries AD with the rise of the lime industry, and almost all the former beach sites of Neolithic or Bronze Age are , mirabile dictu , scenes of activity and life once again. These folk were probably simple boat people who took up lime-making as a more attractive economic activity than fìshing and marine foraging , or possibly as a seasonal sideline to their usual subsistence by marine exploitation. The mystery here is not so much where these people came from as where they lived. No village or house remains have been found , and strangely, there are very few burials. They appear to have the same settlement patte凹， or rather lack of settleme血， as the prehistoric inhabitants - probably seasonal or very temporary encampm凹的 in the back-beach areas.

AMissi可Link in Local P吋listory

21.6 Junhs beachedfor repairs at low tide

The first millennium lime-kiln workers may have lived , just as perhaps the vast m只jority of prehistoric inhabitants had do前， in simple thatched huts on stilts over tidal mudflats or marshes. What evidence that might have survived of these settlements may eventually be recovered , with regular testing of likely low-lying sites It might be argued that a stilt village over tidal mudflat scenario could also be used to explain how a large lron Age and Han population might have escaped the notice of archaeology, except for one inconvenient consideration. Boat people living on tidal mudflats and exploiting the sea in this archipelago for their main livelihood could hardly have avoided using the sand banks and low dunes behind almost every beach in Hong Kong. The evidence of their activities should be the況， just as it is in ample proportions for their precursors and successors. The tendency of the terrain to focus people onto those landforms is very strong indeed ，的 witnessed by the nearly 100 Neolithic/Bronze Age and first millennium AD sand bank sites in the small territory ofHong Kong alone. It is also well attested by ethnographic studies of the nineteenth and early twentieth centuries that show how important the back-beach areas were to the boat people. It requires too great a leap of fantasy to co叮 ure up anyalternativ臼 to the obvious , but extraordinary, conclusion that presents itself - this area was only very sparsely inhabited from about 700 BC to AD 300.

135

Part 111 Historical Archaeology

ThεHistorical Yuεh

Twom司jor

questions dominate the later archaeology of ancient South China: Who were the aboriginal inhabitants in each area before the

coming of the Chines亡， and what happened to these peoples? Were they absorbed , driven away or killed o ff? The answer for each region is probably different and often a mixture of the various possibilities Relations with the Han-Tang Chinese were sometimes harmonious , but often no t. The aboriginal population was generally viewed by Chinese writers as inferior and broadly referred to as “Man" or “Yi" (barbarians). They were universally regarded by the Chinese as in desperate need of acculturation , forcibly if necessar乎 According to Chinese writers who generally considered them sub-human , the natives engaged in all sorts of despicable practices , such as tattooing , running around naked , squawking like chicke凹， drinking through their n。此， cutting their hair ve叮 short ， and so on. The southeastern coastal inhabitants were known also by the name some of them apparently called themselves: Yueh or Yuet or Viet The name was extended southward as the Chinese expanded their empl咒. These Yueh people were noted for their skills in navigation on the water and their savagery in battle; the population of the early state of Yueh (五fth to fourth centuries BC) , centred in the Lower Yangt丘， practised wet rice cultivation and were engaged in trade along the coas t. There is evidence that even as early as the Shang dynasty some items from the Yangtze region were reaching N orth China as tribute or trade goods. The Yueh were also known for their metal-working skills , and , in southwest China , very sophisticated bronze drums were produced as early as the sixth century BC. Modern ethnographic and linguistic researches point to a general Austroasiatic and Tai linguistic affìliation for most of these peopl白， although there

140

Historical Archaεology

22.1 Bronze drum from southw 的 t China (p hotograph courtesy of the Camoes Museum , Macau)

22.2 The “ spirit boat" decoration on a bronze drumfrom G叫

is much debate on this issue. The Vietnamese retain the name Yueh and are the direct descendants of the southernmost branch of this population - the Nan Yueh. Similarly, the Cantonese are also still called Yuet and are derived in no small measure from the aboriginal population. Cantonese is believed by some to derive from Tai (the linguistic family including Thai and related languages) and by others from an Austroasiatic language similar to proto-Vietnamese , thus emerging as a distinct sinitic language during the first millennium AD. The Yueh , later called the "Hundred Yueh" , were certainly a diverse population and probably included tribes speaking many unrelated languages and with markedly different customs. The Tai-speaking Zhuang of Guangsi province today have oral traditions of an earlier occupation of coastal areas and may have been included in the Yueh. Similarly, the Tai-speaking Li tribes of Hainan Island are almost certainly descended from the Yueh. The aboriginals of Taiwan , however, speak Austronesian and are related to the peoples of the Philippines and islands of Southeast Asia. A recent hypothesis that has gained much support (unwarranted in my view) derives all Austronesian speakers from the migration across the

The Historical '{I

22.3

Depκ tion

of a hunting party on a bror

Taiwan Strait of a few small groups , bringing neolithic technology to the island that supposedly was only sparsely inhabited by huntergatherers. We have already noted that there may have been some contacts between the northern Yueh tribes of the Lower Yangtze and Shang civilization to their north as early as 1500 BC. But it is not until the sixth to fìfth centuries BC that the Yueh state of that region begins to come into clear focus ，的 it was involved increasingly in the interstate warfare of the Chou dynasty Furthermore , it was not until the military expansion of the Ch 'in empire (255-206 BC) that the Nan Yueh of the Pearl River Basin and the a句 acent coastal areas were bro時ht 叫o direct contact with Chinese civilization. A process of gradual , though errat眩， cultural assimilation of the Yueh began from that time and , by the Han and Six Dynasties eras , had brought a large number of the Yueh people into the sphere of Chinese culture. Han historical texts provide ample evidence of the acceptance of Yueh chiefs and warno的 into the Han administrative system , even into the army itself. According to one accou旺， when the rebel Nan Yueh king in the present-day Guangdong was fìnally defeated in 111 BC , four of the fìve generals in the campaign waged against him were actually Yueh chieftains themselv白， commanding Yueh troops under the Han banner. This brief historical outline helps in the interpretation of the archaeological evidence from this area. Numerous large chamber tombs and the remains of a shipyard in the Canton area provide evidence of the Han presence and of the spread of Han material

142

Historical Archaεology

culture. At the same tim巴， evidence is also seen of continuities in the way ofli長， in the type of sites occupied , in the pottery, etc. Through most of Guangdong provin白， the gradual assimilation of the Yueh began in earnest and continued throughout the first millennium AD , though not without resistance. Rebellions of indigenous peoples were quite common in Guangdong throughout the first millennium AD. But there remained many remote regions where the aboriginals continued their way of life unsinicized until the Sung or even later. Some archaeologists have argued that , in some remote areas such as Hong Kong undoubtedly was at that time , it is possible that a much looser control and influence on the local population was exercised , perhaps by a small Han garrison or administrative pos t. As mentioned already, a few minor Han sites have been found in Hong Kong , in striking contrast to the large number of Bronze Age sites. The direct continuity between the prehistoric and early historical populations of this specific area is thus in some doubt. As discussed above , Hong Kong seems to have been virtually uninhabited during the Warring States er且， due possibly to upheavals and conflicts generated by the enormous changes taking place during that time in South China

A Han Tomb at Lei

ChεngUk

A very important discovery for Hong Kong's early history occurred in early August 1955. Workmen levelling the ground for construction of a resettlement estate at Lei Cheng Uk, Sham Shui Po , cut into the entrance of an ancient tomb of considerable size. Prompt action by the Public Works Department (PWD) prevented further damage of the tomb or tampering with its contents , and the tomb is now preserved as a museum. The discovery was reported to Professor F. S. Drake , of the Department of Chinese at the University of Hong Kong , who immediately made an inspection of the site. An excavation of the interior of the tomb was then undertaken by staff and students from the universi可 assisted by PWD workmen and surveyors. Professor Lo Hsiang-l凹， a prominent historia日， also participated in the pr可 ect and in the evaluation of the tomb Clearing revealed the tomb to be a brick structure consisting of three chamb亡的， each roughly 4 m x 1. 6 m x 1. 6 血， and a central chamber with dome ceiling 3 m high. A layer of 五ne soil between 50 and 120 cm deep had been deposited on the floor of the tomb , and careful excavation yielded a number of burial objects that had been placed inside the tomb. A total of 61 potte可 objects was found , including storage po的， bowls , cups and house models. These generally are known 的“spirit vessels" and are of a type frequently found in Han burials. In addition , eight pieces ofbronze in a heavily corroded condition were uncovered. Among these were a bronze bell , handle , mirror and bowl. Many of the bricks had moulded geometric decoration and a few stamped inscriptions that read “ Peace to Pan-yu" (Canton) , “Great Fortune to Pan-yu" and “Master Hsueh." Unfortunately, these inscriptions are of a generalized nature and , apparently, do not relate

144

Historical Archaεology

23.1 Examining the entrance to the Han tomb at Lei Cheng Uk

23.2 View from inside the tomb during excavation

A Han Tomb at Lei Che月 Uk

之3.3

Cataloguing the

們叫

TQ

PL

昕

nd, 5

后HI

勻、 J 勻， h

并峙，

QPL PLO G ]

0句 ects from

-G

PL

the tomb

145

146

Historical Archaεology

23 .5 Excavating a pot in one of the chambers

directly to the tomb or the deceased. No trace of a coffìn or human remains was found , and , although it is beyond doubt that the structur毛 is indeed a tomb , nothing is known about the person or persons for whom it was buil t. The dating of the tomb is better know孔， and the best estimate , based on comparison with similar tombs in Canton , is that it was built around AD 100-150. A slightly later date cannot , however, be entirely ruled ou t. The tomb and its contents are fully Han in styl亡， although there are a number of local characteristics. One of the most interesting of these is the pottery with stamped net patterns found among the sophisticated glazed Han wares. The tradition of geometric pottery antedates the Han dynasty by almost 3 ,000 years in this area. While the tomb established quite dramatically that Han material culture and , in certain respects , Han way of life had penetrated into this region by around AD 100-200 , it brings more questions than answers about the Han occupation of this area. The question that 1 raised 20 years ago still poses itself today, but in a slightly modifìed form: Where is the population and the support base for the wealthy or highly ranked individual who was buried in the Lei Cheng Uk tomb? With all the intensive archaeological work of recent decades , there is still a glaring anomaly between the few Han minor sites on the one hand and the administrative/military rank or degree of wealth implied by the tomb on the other. It is possible

A Han Tomb at Lei Che月 Uk

that the evidence for the immediate community or enclave or outpost on Kowloon peninsula where this individual resided was lost to urban developmen t. ln 2004 , a group of plain pots , probably dating to Han , were uncovered during the laying of a pipeline beneath a road in Shamsuipo , not far from Lei Cheng Uk. But no other evidence has come to light in that area , and no pre-war archaeologist found any Han materials of note in the whole of northern Kowloon peninsula , which was rural at the time lt is scarcely imaginable that the network of settlements and the population normally needed to support or justify such a rich and powerful individual has been missed by archaeology And it is nonsensical to claim (Watt 1970: 9) that the contents ofthe tomb give an idea of what “ the daily life of ordina可 people" in Han times in this region was like. The grave goods are stylized artefacts , almost certainly imported from Canton (Panyu) or copied from such imports. The issue of who was buried at Lei Cheng Uk has been debated from time to time , with discussion centring on whether he was a Han administrative official or garrison chief, or even possibly a local chi記 efta 缸ma 址lreadyas臼slml迅lated intωo Han culture (typi五ed by the rulε位r司 ofthε “ Kingdom of Nan Yueh" decεased wer跎呵毛 e 由 t he cωommand 由 er of a mi血 litary garrison placed there for strategic reasons (to protect the sea approach to Panyu) , it must have been one of the loneliest of postings and one that was alimented largely by supply ship. Perhaps it was the tomb of a Han Chinese 0伍cial assigned to administer this area , heading a small administrative encla間， or a feudal lord , ruling a largely uninhabited fiefdom. Another possibility is that local Yueh tribal chiefs had begun to imitate Han styles and were themselves in the process of becoming Chinese culturally But for all of these scenan肘， it is significant that no major sites of Han garrison or village have been found in Hong Kong. One final possibility should be mentioned - that the tomb was the grave of a Han military leader involved in a campaign against the Yueh tribes. This

23.6 A large glazed vessε1 typical of the Han

23.7 A hardgεometric pot from the tomb

147

148

Historical Archaεology

would explain the apparent absence of major contemporaneous sites and perhaps the missing human remains as well. Professor Drak亡， noting the total absence of bones and teeth , jade bodily ornaments and metal belt hooks , commented that the tomb could be a cenotaph (memorial),“perhaps of a military offìcer whose body could not be recovered from the battlefìeld." It is also of interest to note that in Hoifung , Maglioni found only one Han site that he believed to be a military outpos t. This sort of question can ultimately be answered only by new evidence , but it does pose interesting problems for the archaeologist to ponder. The Han presence in the coastal area and the fate of the aboriginal Yueh population remains very much a mystery The tomb and its contents were the subject of an illustrated , bilingual booklet published by the City Museum and Art Gallery in 1970. A more complete bilingual report has recently been published by the Hong Kong Museum of History (2005)

Hong Kong's First Industry

Until 1976 , virtually nothing was known about the occupation of Hong Kong in the entire first millennium AD. There had been scattered finds in the colony of burial vessels attributable to the Six Dynasties-Tang era on stylistic grounds , for example , the discovery of a Tang jar on the beach at Fu Tei Wan in the 1960s. Excavations of the last 30 years have brought to light several new burials of the first millennium. The pair ofburial jars with covering bowls at Sham Wanh的 been mentioned already A similar jar was found at Pui 0 , along with a cluster of small bowls dati月 to the Six Dynasties or Chin (third to 五fth centuries AD). Another cluster of similarly dated bowls was found at Pak Mong; a group of small dishes at Tung Kwu; and four plain pots in Mong Kok (during trenching under a road). A very well-preserved skeleton of a young adult female was uncovered at Sha PO Tsuen with no burial objects except a silver ring on her finger and a long U-shaped pin in her hair. Generally, however, the early historical archaeology of this area had failed to yield village sites , burial grounds or activity areas of the thousand years from Han to the end of Tang. Historians cite a number of references in the Tang literature that could possibly (albeit tenuously) be linked with places in Hong Kong , but local archaeology had not provided any evidence whatever to support these claims. (The dif且culty of reconciling the results of archaeology with historical accounts is discussed further in Chapter 25.) Most of the first millennium AD thus appeared to be another missing link in local archaeology The evidence , however, had been there all along and was not recognized by most archaeologists who , quite literally, were tripping over it. A number of kiln sites had been noted on or near the beaches

150

Historical Archaεology

24.1 The site of Yi Long in 1974 (now Sea Ranch Resort)

24.2 Recording a kiln on Cheung Chau

24 .3 Debris from a kiln at Sha PO Tsuen , Lamma

Ho月 Kong's

151

TR H t Mvm iad mm 月U

Wσ 計

before 1975 , and there was a general impression that these structures were related to the present-day villages and were not more than a century or two old. Only Heanley ventured to date them to the fìrst millennium , correctly as it turned out , on the basis of a Tang coin and glazed pottery he found inside a kiln. The subject was not pursued by other pre-war archaeologis的， and his interpretation was soon forgotten. The University Archaeological Team dismissed the kilns as being of fairly recent origin; the team hardly even recorded the kiln structures or debris when they were found. It was not until 1975-78 that a systematic study was undertaken of the kiln sites at Yi Long , Lantau , and Lo 50 5hing , Lamma , under the direction of Hugh Cameron , then chairman of the Archaeological 50ciet于 His admirable and persistent interest in the structures opened a new chapter in local archaeology. Despite being informed by one of the Lo 50 5hing villagers that the kilns had been in operation in his great-grandfather's time , the excavation pushed ahead. The fìrst carbon dates from the kilns were dramat止; they dated from AD 400 to 500. Investigations since then at dozens of additional sites have shown quite clearly that the kilns were operating from the 5ix Dynasties to the end of Tang and that they were used to fìre coral and shell to make lime. A study of the best-preserved kiln at 5ham Wan Tsuen on Chek Lap Kok showed that , contrary to previous theori白， the kilns did not have any dome or superstructure. The age of the kilns has now been established conclusively by 15 C-14 and T-L dat白， and by the presence of numerous coins and datable

EEJ fLvt xh 11 {r

山 L

八υ

n 斗， CJ

MUU

First lndus盯

152

Historical Archaεology

glazed ceramics. The site at 5ham Wan Tsuen on Chek Lap Kok was particularly rich in coins and pottery of the late Tang. The evidence at present also shows very clearly that the lime industry had ceased to function by the 5ung period , as no 5ung coin or celadon has been found in the kiln layers , and the latest C-14 date on any kiln-related deposit is around AD 850 The kiln industry of the fìrst millennium AD is thus fìrmly established and was quite widespread and intensive. 50me sites , such as Lo 50 5hing , Yi Long , 5ham Wan Tsuen and Ma Wan , had a large number of kilns (1 5 or more) and may have been producing lime over several generations. There are over a hu日dred sites where kiln debris has been discovered in the territory. This fact alone would be suffìcient to dismiss the ridiculous recent proposal that the kilns were actually for producing pot悅耳 There is other compelling evidence pottery kilns of that era are well known and were totally different in structure; there is hardly any misfìred pottery refuse at the sites; and there are several tons of lime in the bottom of some of the kilns. 24.5 Aerialviewofthesandbm after complete excavation at Sham Wan Tsuen , Cheh Lap Koh (now under the airport terminal buildiniJ

Ho月 Kong's

First lndus盯

153

The vast quantity of lime that would have been produced in these kilns leaves little doubt that it was made for export. It is not entirely clear what the main use of the lime was; some possibilities that have been proposed are that the lime was a building material or it was used for agricultural purposes. Historical texts also mention that lime was used to caulk the hulls of ships. The market for lime must have been attractive enough to lure the local fìsherfolk (and perhaps their neighbours and distant relatives) into the industry as labourers , if not entrepren凹的It is reasonable to assume that the lime was transported to urban areas such as Canton , probably by middlemen who acquired the product directly from the kiln workers The evidence is clear that the lime-kiln industry died out rather suddenly in this area in the tenth century Possibly the industry shifted to other areas of the coast during Sung and Ming times. It reappeared dramatically in Hong Kong in the eighteenth and nineteenth centuries; a 1905 survey found several thousand men and several hundred boats engaged in the collection of shells and coral for lime production in the New Territories. By the 1930s , this modern lime industry was in deep trouble , perhaps similar, in some respects , to the demise of its precursor 1,000 years ago. The environment strained under the weight of exploitation: shell and coral resources were dwindling (leading to occasional , violent disput臼)， fuel was increasingly diffìcult to obtain , and a new product (cement in the 1930s , a better source of lime from limestone after the tenth century) was reducing the value of the shell lime.

PLfL

nd' a 『

b

的

hκG

tγtyf

，

月U 仆υnu

，

rysσb

bdl

《、

叫

仙

4L'

山M ov4GEI

f

1l

1

勻Jhdhvpit

r， 刁峙

--qlp hkdI irsT > hqHE 63c tpJHE dtvtn mour tuln nnvn5

154

Historical Archaεology

A linguistic confusion resulted in a rather serious dispute over this feature of 10ca1 archaeo1ogy when a de1egation from the Hong Kong Archaeo1ogica1 Society made its first visit to the Main1and in 1978. During a 1engthy post-dinner discussion with our hosts in Guangzhou , Chairman Hugh Cameron described the work on the kilns and concluded with their dating and function. The Chinese archaeo1ogists listened polite1y to the trans1ation , but they firm1y disagreed , insisting that the historica1 record was very clear on the subject , and lime did not come into use unti1 the Sung era. Our de1egation respond吐， equally politely and firm1y, that the evidence was absolutely irrefutable and the dating unshakeable. It went back and forth severa1 times. The立， someone from our side added that on some sites there were shells and coral , and some kilns had tons of lime in the bottom from the 1ast firing , so there was no doubt about what the ki1ns were producing. At that point , there were huge sighs from the Chinese side as they exclaimed: “ Ah , shell lim亡， not rock lime!" They said that shelllime goes back even to the Neolithic. Our translator had been using the term "rock lim亡"的 it is the common word for lime in Cantonese. Perhaps the most intriguing question posed by the discovery of the early kiln comp1ex in Hong Kong is the suggestion of continuity through time suggested by the occupation patterns of the prehistoric and early historical inhabitants. Virtually every kiln site is on a prehistoric si仗， with ki1n debris often direct1y over1ying the Late Neolithic or Bronze Age cultura1 deposits. The converse is a1so beginning to appear tru亡， that is , every coasta1 prehistoric site a1so yie1ds ki1n debris when thorough1y searched. This corre1ation can hard1y be attributed to chance or to topographic factors a10ne A sense of continuity is suggested , not on1y by site choice , but a1so by buria1 traditions. At the Sham Wan si仗， two Tang buria1 jars were found in the same area of the site used for buria1 during the BronzeA阱. A1so , rather remarkab1y, in this s日me area cremated bone fragments were uncovered in both Tang and in Midd1e Neolithic 1evels Finally, at Hai Dei Wan a 1arge early historica1 vesse1 was found in the midst of a number of Bronze Age vessels , indicating that the same area was re-emp1oyed for buria1 purposes in the Tang era. We shou1d a1so take note of the continuity of negative e\吐ence: in both prehistoric and ear1y historica1 period , no p

Ho月 Kong's

First lndus盯

24.7 Kilns from various sites

155

156

Historical Archaεology

24.8 The slaking pit at Pui 0 , Lantau

elsewhere) that did not leave any archaeological traces on the sites of their activities But overriding all these obvious continuities is the m吋 or break of around 700 BC to AD 300 , notwithstanding the minor Han sites. The direct continuity from prehistoric to early historical populations cannot be established for this reason. It seems quite like旬" however that the people who worked the kilns in the Tang period remained in the territory after the demise of the lime industry, perhaps reverting back to exploiting marine resources for subsisten凹. They would be the aboriginal “ barbarian" population recorded in Hong Kong when the territory began to be settled by rice farming clans from around AD 1000.

ThεSung

Era

It is claimed by some that Hong Kong's recorded history begins in the Sung dynasty era or even earlier , when written records refer specifìcally to places believed to be in the territory Previously, we have discussed the Han to Tang proto-historical periods , for which archaeology is still the major source of knowledge about the life of the local population , even though the larger region of Guangdong province had a recorded history from early Han. Reconciling the information from archaeology with that of history and clan genealogies presents a number of major problems. From archaeology alone we have a reasonably clear picture of how Hong Kong developed in early historical times , although there are still many unanswered questions and a major gap in the record. As discussed above , one of the most important discoveries of local archaeology in recent years was the thriving lime-production indust可 that existed during the 五rst millennium AD It is unclear what economic activity

25.1 Test excavation at Diamond Hill , north Kowloon , in 2005

158

Historical Archaεology

之5.2

Sung celadon s拾你 from Nim Shu Wan , Lantau

月υ

K n I n 1inyfo 月υ

PL

σb

F 、d

J

』

un-d npudy

fIJ qj

H 司

3

戶、 J

d

γ'

戶、 υ

勻、 υ

7J 』

戶J 、

A 'n

the kiln workers took up as the indust叮 faded. Fishing and shell collecting could certainly have reabsorbed some of them And ifthem叮叮 local clan genealogies are to be believed , the clan founders settled in the New Territories during the Sung era. These clans have traditionally practised rice agriculture Thus far , however , there is little archaeological evidence bearing directly on the early clan villages. Sung ceramics have been found in large quantity at several sites , notably Nim Shu Wan and Shek Pik on Lantau , Diamond Hill in Kowloon , and Lung Kwu Sheung Tan near Tuen Mun , as well as many other minor sites; at latest count , there were 斗o in total (Wong 2006) At Lung Kwu Sheung Tan , there were literally hundreds of pounds of potsherds , mostly plain , black , domestic-type storage jars , but also a large amount of celadon bowl fragments Caches of Sung coins and high-quality celadon ceramics were found during the construction of the reservoir at Shek Pik. These were on the hillslope above the reservoir and most probably derived from an aristocrat's burial , although , at the time , there was much excitement over the (rather far-fetched) speculation that this was the grave of the Sung boy emperor or someone in his entourage as they fled south. Other Sung coin caches were found at Fan Ling , Sheung Shui , Mai Po , Tai Po , San Tin near Yuen Long and in a pot on Kellet Island in Causeway Ba乎 Most involve several hundred coins , more than 500 were recorded at Fan Ling , but the discovery at San Tin took place around 1912 , and few details are available These coin caches may represent burials ,

The Sun呂 Era

offerings , or simply money hidden away in a sock (so to speak) A different sort of coin cache (a handful) plus a teapot and several bowls were found at Peng Chau , and this group of artefacts is more suggestive ofburials. Sham Wan Tsuen also had a ve可 similar group , with 14 coins , 2 bowls and a ewer. In addition , the latter site also had another group of two bowls , one dish , one brown jar and a long iron knife. Although no human remains were found with these artefact clusters , they are typical of burial offerings. Elaborate "ash jars" have also been found in the territol水 such as those found at Sek Kong in 1960. Two ash jars were also found in fragments during the excavation at Lung Kwu Sheung Tan in 1990. Finally, cremated human bone fragments were found in one of a pair of jars with covering bowls also found at Sham Wan Tsuen. Oddly, these jars were buried in the mouth of a lime kiln from the earlier activity on the site. One wonders if it were a faint folk memory of the previous generations who worked there From historical sources it is known that salt production and pearl collection were other major industries practised along the coast of Guangdong as ear甘 as Han tim白， but , in each ca仗， the first mention of the Hong Kong area specifically occurs in the Sung literature. The exploitation of the “ TaiPO Sea" (possibly Tolo Harbour or Deep Bay) for pearls was administered by a military post employing native divers from around AD 900. The collection of pearls from the Tai-Po sea was revived briefly in the twelfth century Salt pans were operated in this area under government monopoly in the Sung period , and the visit

25.5 Burial pots with cremated remains at Sham Wan Tsuen

之 5.4

Tang and Sung coins from

\叫lOUS sit的

159

160

Historical Archaεology

25.6 The Sung inscription at Fat Tong Mun

of a “ salt fìeld offìcial" is recorded in an inscription at Fat Tong Mun dated 1274. This industry of salt pan evaporation continued into more recent tlm忱的 a number of local place names with “Yim Tin" (salt fìeld) indica仗， and one such fìeld was still in operation in nearby Daya Bay as recentlyas 1980. However, there is no archaeological evidence for either salt production or pearl collection , nor are there important sites with Sung ceramics in the places where such activities would likely have been carried ou t. While one might not expect to fìnd pearls , and the process of obtaining salt from sea water might not have required tools , nonetheless , there should be sites where the people lived The Sung inscription at Fat Tong Mun also mentions the renovation of two Tin Hau temples there by people originally from Fujian and Zhejiar嗯. The present temple at Fat Tong Mun dates from later periods but is believed to be on the site of the earlier structure. There are a couple of dozen temples , ancestral halls , monasteries or tombs in Hong Kong that are believed to have been rebuilt on the site of an earlier pre-Ch'ing structure , but there is 日o building or structure above ground today in

The Sun呂 Era

Hong Kong that dates from the Sung or Ming periods 一- nor have ruins or foundations from those periods been securely identified , apart from several house floors. The difficulty for the archaeologist is that none of these temples or ancestral ha11s have be11s , plaques or inscriptions dating earlier than Ch'ing. It is hard to imagine that a11 such items were destroyed or looted during the coastal evacuation Two of the five m句 or clans of the New Territories have genealogical records indicating that their early settlements go back to Sung times. According to the genealogy, the 且rst member of the prominent Tang clan came fromjiangsi and settled in Kam Tin around 1050-1100. Other settlements followed as the clan expanded and 1的 wealth increased. The Ho clan came from Canton and settled in the Sheung Shui area about the same time. Settlement of other m吋 or clans , such as the Liu , Man and Pong , date from the Ming era. These genealogies are difficult to prove or disprove because hard evidence is lacking. There are a number of legendary associations of places with Sung dynasty figures , such as the Che Kung Temple in Shatin , said to have been built after Che Kung (a general in the Sung army) accompanied the last Sung emperor on his exile in South Chi凹， according to some accounts , he later settled in the area. According to historical references , Tung Chung is an area that should provide evidence of Sung settlement and activi作 Yet an extensive survey of the entire Tung Chung valley that 1 directed in 1992 with several assistants yielded , in total , onlya few dozen sma11 pieces of Sung ceramics. The va11ey and lower hillslopes have been heavily terraced for cultivation in the last two centuries , and there are several deep erosion faces as we11. These conditions should have revealed cultural deposits if any were present The area of Castle Peak , Tuen Mun and Ha Tsuen is another that is believed to be of considerable antiquit芋 The Ling Tu monastery and the fort at Tuen Mun (“ fortified gate") are recorded even earlier than Tang , but extensive archaeological surveying of the area several times over has not brought to light a single item of military equipme凹， nor any significant Sung sites. Especia11y perplexing is the Ling Tu monaste叮 south of Ha Tsu凹， for which there is specific historical evidence of an origin in the Chin era (AD 265 -4 20). The present building and the bell inside date from the middle Ch'ing period. The monaste可 is at

p

the end of a long , narrow va11ey that has road cuts ，五sh

161

162

Historical Archaεology

It is difficult to account for this complete absence of archaeological material 丘，

indeed , the historical record is accurate and the location of the monaste叮叮ld of “ Tuen Mun" itself has not changed over time. Historical texts also mention the indigenous population of the area in Sung times. On Lantau , there was a m句 or settlement of “ Tan" fisherfolk who resisted efforts to enforce the salt monopoly Other sources mentlO n “Yao" who also lived on Lantau. Chinese sources describe several efforts to bring these folk to heel and , finally, a campaign to annihilate them. A thirteenth century source states that “ the local Tan and Man ["barbarians"]live near the pearl-oyster beds [presumably in Tolo Harbour or Deep Bay] but they are more wretched and starved than the poorest people . . ." Later sources refer to the Tanka boat people as “Yao" or “ barbarian ," and for centuries they were shunned and not allowed to settle on land. Even as late as 1729 , the Sun On county gazetteer recorded that “ in Guangdong there is a tribe of Yao barbarians called the Tanka , who have boats for homes and live by fishing." These presumed remnants of the Yueh and their traditional way of life were looked down upon by Han Chinese through the centuries , even (or perhaps especially) by literati and scholars. The poet , author and traveller Qu Dajun wrote ofthe “ Tan" in Guangdong in the mid-1700s All Tan women are known to eat raw fish and swim under water In the past they were seen as belonging to the family of dragons. It was because they dived into the water with tattooed bodies in order to look like dragons. They could move in water for thirty, forty li without difficulty. They were seen as sea otters . . . they are really non-human. Ccited in Crossely et al. 2006) In spite of the continued use of the pejorative barbarian labels Yueh , Yao , Tan and Man for the local folk , the population in most of Guangdong , by Sung tim白， was at least partly assimilated into Chinese culture , language and society; pristin亡， unsinicized aboriginal villages survived only in more remote areas of the province. Influenc巴， trade and intermarriage (plus doses of force) since Han tim凹， ln most areas , had replaced previous languages and cultur亡， exce阱， of course , those Cmany) cultural items that had become accepted as "Chinese." Even the “ wretched" Tan and Man barbarians living near the oyster beds were , no doubt , on the road to assimilation. Further “ acculturation" and economic interdependence would soon obliterate the last traces of any distinct aboriginal heritage in this coastal region

A UniquεKiln

Complεx

One of the most attractive beaches on Chek Lap Kok was at Ha Law Wan , located between two major archaeological sites. Although there was no sand bank formation behind the beach , there was a relatively flat isthmus at an elevation of nearly 15 m. During the initial survey of the island , the site was identi五ed as ofhigh potential for Neolithic or Bronze Age material , although no artefacts were recovered during surface search. The site has some natural advantages. Villagers reported that Ha Law Wan had a very reliable water supply, even when the rest of the island suffered from drought. We were able to confirm this situation during June 1990 when the island had its worst drought in living mem。可 An open well at the base of the ravine still had water Schofield searched the site in the 1930s but reported no finds. Our team spent several weeks in conducting surface survey, test pits and excavation. No prehistoric deposit was found , in spite of the marked similarity of the site to the others nearby (Fu Tei Wan and Kwo Lo Wan) that possessed a relatively flat area well above the beach Seven stone adzes of neolithic type were found at Ha Law Wan , either on the surface or in late contexts. This absence of prehistoric deposit is puzzling During the survey, it was noted that the beach was deposited against a hard , gritty, reddish brown hillslope soil. Most of the flat area above the beach was extensively terraced , and the terrace cuts showed clear stratigraphy down to 1 m below ground surface. No cultural deposits or features were seen , with the sole exception of part of the wall of a fired-clay structure exposed in one of the terrace walls. In spite of the lack of surface finds and observable deposit in the terrace cuts , it , nonetheless , was considered worthwhile to commit two to

164

Historical Archaεology

26.1 Kiln wall and vent as dis 正 overedα tHaLawWan

Chek Lap Kok , 1991

three weeks to the investigation of this site because of its setting and probable attractiveness to prehistoric people in the area Three weeks of rather disappointing test excavation did not yield any Neolithic or Bronze Age material , with the exception of several polished stone axes. The final task that remained was the excavation of the fired-clay structure partially exposed in one of the terrace walls It was almost dismissed as modern because it had a large iron pole embedded in one corner and seemed to be related to the terraced field system. Meanwhile , other terrace walls were shaved , but no other structures were found. A small patch of reddish clay just beyond the end of a cement footpath was noticed , and , although strange , it did not appear at first to be of archaeological significance. Upon excavation , the fired-clay wall exposed in the terrace cut turned out to be a rather elaborate kiln of considerable interes t. The structure was clearly a kiln or furnace; it had a thick layer of charcoal at the bottom under hard red clay that seemed to be part of a superstructure or interior furnitu仗; and it had three vents , one of them an L-shaped chimney.. Attention then turned to the red clay patch at the end of the footpath , and a female worker was assigned to clear away the soil but not remove any of the red clay. Fifteen minutes later, an assistant reported: “She's got a kiln there." Further excavation revealed that it too was a large kiln similar to the 五rst on亡， with channels in the floor that continued outside the mouth , which was bridged by a stone

165

A Unique Kiln Complex

f仁，

γ'

的

vh

八U

U

，但

月U?ru

d

門 1

仿制

structure. The soil inside this kiln had built up to a thickness of about 60 cm before the 五red-clay blocks had fallen in , and these seemed to be fragments of the roof. It was unclear what purpose these kilns served , but they were very different from the Tang lime kilns that are so common on Hong Kong sites. All sorts of possibilities were discussed as the excavation proceeded , including that they were an evolution and refìnement of the lime kilns. Several surprises were in store! Clearing of the area outside the second kiln revealed large quantities of charcoal , mainly in and around the channelleading out from the mouth. But the spread of charcoal seemed to be slightly askew to the east , well outside the general “ flow" out from the kiln. It appeared that further evidence of activity might be found in that direction , and a large 2 m x 5 m trench was opened A massive charcoal flow was immediately revealed , and , eventually, three more kilns were found by following charcoal trails right up to their mouths. At the same time , the walls of an old irrigation ditch were shaved. and another three kilns were discovered. A call of nature answered by the deputy director revealed the existence of another kiln. Further shaving of the terrace walls brought to light charcoal trails that led to four more kilns. It was now clear that this was a major si仗， but the dating was still unknoWll. Quite surprisingl其 no pottery or tools of any kind had been found. A sample of the charcoal was despatched for urgent carbon dating (at twice the usual cost). The result came back as 620 years , confìrming that the kilns were defìnitely later than Tang , but their

們 nvL

2

LH l? 2wew 315 Ett YA tel tm -t x'h nunc < u

166

Historical Archaεology

function remained unclear lt was considered a distinct possibility that this ki1n industry was a continuation of the Tang lime indust門， but this hypothesis was 1ater

26.3

Nem炒正omplete

kiln begins to emerge

26.4 The kiln after exωvatlO n

rejected. All 13 ki1ns had been constructed in a simi1ar fashion: a cavern was dug in the resid ua1 hills10pe so泣， it was p1as胎兒d with a clay mortar and fired. The over-1ying soil stratigraphy was undisturbed by the kiln construction and use , except in cases where the roof collapsed and brought the soi1 profile down into the ki1n cavern. Severa1 ki1ns had some or most of the roof intact , and two were on1y partially filled when discovered. One fine examp1e was comp1ete1y silted up , a condition that preserved its roof a1most intact All ki1ns had 1arge amounts of charcoa1 inside on the floor and outside near the mouth. Another C-14 date of 720 years ago was obtained on a samp1e from inside one of the ki1ns. The over1ap of the two C-14 dates at around AD 1260-1380 allowed a secure dating of the ki1n comp1ex to the 1ate Sung and Yuan eras. During the excavation , villagers produced a Sung coin dating to AD 1256. They reported that a jar full of coins had been found in the area of the ki1ns many years ago , but all but one were 10st Pottery from the ki1n comp1ex was remarkab1y scarce - on1y some undistinguished pieces of p1ain ware and a few Sung ce1adon fragments were found. Most surprising1y, no too1s of any kind were found in the kiln ar凹， nor were there any pieces of containers or ki1n furniture

A Unique Kiln Complex

In addition to excavating each kil丘， several exploratory trenches were excavated to provide stratigraphic sections and to search for further clues on what sort of industry was being carried out on the site. The first hint came from a solitary, tiny, grey metallic pellet resembling lead from the mouth of one of the kilns. This was followed by the discovery of a pit containing dozens of such pellets 4 m away from the kiln mouth. This material appeared to be some sort of slag; other pieces appeared to be clinker with pellets embedded within The possibility presented itself that the industry was a smelting or castmg operatlon. Another trench yielded a pellet deposit in a thin 2-3 cm layer at one end of the trench not far from , and obviously related to , a kiln nearby This material could have been what remained on the ground surfac亡，

ensnared in grass and roots , after the bulk of the pellets was collected and taken awa手 Other trenches in a small ravine between the two main groups of kilns revealed no more structures but several areas of pellets and charcoal deposit Chemical analysis of the slag revealed that it was composed almo位 entirely of iron. In view of this , it seems reasonable to assume that the pellets are not slag at all but represent the output of the kilns. Pelletization is one method of smelting iron , but exact1y how the kilns operated remains a mystery One might hypothesize that the iron ore or iron-rich sand was si日tered or only partially smelted , resulting in the formation of the pellets embedded in the slag matrix These slag clumps would then have been manually broken up , and the true waste slag pushed downslope. The steepness of the ravine would then , over time , have washed away the slag deposits Many questions remain If the kilns did indeed function in some manner for the smelting of iron ore or iron-rich sand , and the pellets are the end product , where is the slag? On most iron smelting sites , slag deposits are massive. Even if most of it went down the ravine , it is hard to understand how there could be no trace of it on the site. How did the kilns operate? There is no possibility that molten iron ever flowed in the channels since numerous crevices have no traces of metal. If the channels were not for the product of the kiln , what purpose did they serve? Why was this particular site chosen? There are only insignificant occurrences of iron in the ar凹， certainly not sufficient raw product to run an indust可 on the scale of the operation in evidence. Was

167

168

Historical Archaεology

Why are there no crucib1白， too1s and other articles necess且可 m a sme1ting centre? The kilns are unlike any described in historica1 sources or reported from China. The grooves in the floor suggest that something was meant to flow out of the ki1n; in some ways , the kilns seem more appropriate for the production of an organic substance , such as resin , charcoa1 or even lime. The fact that they were burrowed into the hills10pe and their shape suggest that a higher temperature was required than what wou1d be needed for charcoa1 or lime The kilns at Ha Law Wan are unique in Hong Kong archaeo1ogy Despite intensive archaeo1ogica1 investigation since their discoverγ， no similar kilns or sites have been found. The fÌ na1 word on the site must await future excavation. A 1ate modi五cation in the design for the new airport made it possib1e to preserve the ki1n comp1ex at Ha Law Wan for future disp1ay and research. This was a most fortunate deve10pme凹， and the site had great potentia1 for heritage purposes and cou1d have been a nice disp1ay on the airport grounds. However, the site has 1anguished since then , and poor monitoring during the construction of the airport resulted in severa1 of the kilns' roofs collapsing. Despite this 10ss , the site is the on1y one of its kind in Hong Kong and remains a mystery to be reso1ved in the future. The kilns are now properly buried , and stone walls have been bui1t around them. Hopefully, the key to the site's de五nitive interpretation is still in the ground around the kilns

Later Historical Archaεology

The Ming period , like the Han , is another period of relative scarcity in Hong Kong's archaeolo gy, but for reasons that can be more readily postulated. Many of the Ming settlements continued into the Ch'ing era and down to the prese凹， presumably leaving little obvious archaeological evidence of their existence. The coastal evacuation of 1662-69 to combat piracy and rebel activity resulted in many villages being abandoned and falling into ruin. One contemporary (cited in 5iu 1989) wro仗，“Their houses were demolished to provide materials for fortifìcatio肘， and their graves were dug up to make deep moats." It is stran阱， however, that the villages abandoned at that time have proven so di且cult to identi年 Them句 or Ming archaeological sites include a remote site (probably a trading or smuggling station) at Penny's Bay, Lantau , datable to 1480-1530; a cemetery at 50 Kun Wat; a coastal village at Ho Chung; house floors with drainage system at Mong Tseng Wai; and the remains of a junk found in the seabed 0旺 High Island , dated to the early 1400s The site at Penny's B可 now completely destroyed by development related to the Hong Kong Disneyland theme park , is a mystery The site was fìrst noticed by then district of五cer and local historian]ames Hayes on a visit to the adjacent Choy Lee 5hipyard in 1975. He recognized that the pottery was likely to be earlier than nineteenth century, when the area was fìrst farmed according to local records Hayes collected a few dozen representative sherds and handed them over to the Hong Kong Museum of History The Archaeological 50ciety tool歪扭 interest in the site in 1983 when it became apparent that the pottery was from the Ming dynasty During several surface collections organized by the Archaeological 50ciety, huge quantities of Ming blue-and-white porcelain were recovered , not only from the

1 70

Historical Archaεology

low

lying 五εlds ，

but also from the tidal mudflats below and

the hillsides above. After a period of rather absurd dispute with government bureaucrats over the status of the si仗， a permit was finally issued. Several excavations were mounted by the Archaeological Society in the period 1986-90 and by the Antiquities Office in 1992. The dating of the site was narrowed to the first three decades of the 五fteenth century based on the styles ofMing blue-and-white pottery Examples of Southeast Asian ceramics were also identi且ed ， namely Khmer, Vietnamese and Bau-Malay types Various theories have been bandied about to account for this unique si仗， including that it was involved in smuggling or piracy and supported by the fact that the site is not visible from the sea lanes off Lantau and se ems to have been chosen for this reason. Another possibility is that it was a base for 之 7. 1 Excavat何 Ming

pottery from the mud at Penny's Bay, Lantau ,

1994

ship provision and repair, fed by a perennial stream. None of these notions are entirely satisfactory, and certain facets of the evidence argue against each interpretation that has been offered. Claims that building foundations and structures on the site date to the Ming era are almost certainly incorrect Despite all the archaeological investigation , the matter still stands 且sJames Hayes (1 981) wrote in first reporting the si仗， “ There is clearly some mystery to be unravelled here."

斗， 5

，U 汁 VYJ

晶晶

f古ιy

74γL 可/。

l

pu

M 仇

-y dMU MP un -pw nn

2pi

uy

l1

1 71

Later Histo此al Archaeology

On 23 May 1974 , an article appeared in the Chinese-language newspaper Ching Dao reporting the discovery of what was described as a “ Sung dynasty old junk and pottery" The discovery had apparently been made a month earlier by a hike r. What was remarkable about the report was that the materials were found on the exposed seabed off Sha Tsui near High Island , an area that had been dammed off and pumped out for the construction of a reservoir. The report further stated that , when the hiker “ removed some sand , he found a number of rare vessels and bowls." The next day all m司jor newspapers carried the st。可 some with headlines reading “ Archaeologists excited by n.nd." The problem was that no active local archaeologist knew anything about the discovery By luck 1 was free that day and went straight to the si仗， arnvmg mid-morning. Several people were ther宅， poking around in the sand , and 1 persuaded them to stop and wait for the site to be investigated properly As luck would have 泣， a fellow member of the Archaeological Society, Ray Fro哎， was one of the resident engineers on the High Island Reservoir pr叮 ect ， and he came out to the site at noon. Things then began to happen: police arrived to exercise control , a fence went up around the area in an amazing two hours and arrangements were made for staff from the museum to visit the site. A grid was laid out over the area , and a proper excavation began the next day However, the n.ne Sung and Ming ceramics featured in the newspaper article were never recovered

Lκr

G UMU nu 』

1J%' ，

τd

γιThHi­

m

w pud, •

eTS

hmu月

5dt叩

，仙

古

RdM 山

戶、 u

hH 句、 JV仆U 可

FPIσ

Uα 印

WMG -mn tt'

1 72

Historical Archaεology

The junk had burned before it sank , and only isolated planks and other pieces survived. The potte可 associated with it was dated to the 5ung and early Ming period. In add 出ltlOntωot由 he Chines仗 ewa 盯re 臼s ， t出 he 叮re wa 站s a quantit句 y 吋 0 f Khme 叮r pott怯阻 臼r咱γy dating 仕 E fromt出 he 且缸fteent出 h1 臼 C ent凹 ur 可弘 Ba 叫 u

50ut出 he 臼as泣tAs釘ia ， pos臼Sl由 bl甘 yde 叮riving ultimately from

Lr

Lpu

「[

V切

間， drs

們可

vly 川，

L

nzd

公

吵吵

d

fmmdem 叫

TL

itνELt-c

ωn

古 rmCAU

khu

的

Mh↓川叫扒叫W 叫 W 心而山A 叫m 叫 m

們

Erw

引νAVURhdrL

刁寸寸ιotιnτdnr

India. The presence of 50utheast Asian artefacts at this site and at Penny's Bay sites showed that long-distance trade had begun to play an important part in the Hong Kong region by the 五fteenth century A few isolated burial jars dating to the Ming era have been found over the years. These isolated finds could have been related to a transient or floating population as opposed to a settlement. The Ming burial ground at 50 Kun Wat , however, seems to be derived from a small village and is the only one of its kind thus far discovered At least 20 burials were found the咒， most of them characterized by uoOHUE tzlil;r t t T [ ' ] d'hrct5 a pair of pots with covering bowls. One pot of each pair contained MPFjdnk FGLHOZ remains of rice or other grain; the other was empty and presumably 1:1 had once contained wine. In addition , coins , iron implements and a long iron knife were found in the burials

Later Historical Archaeology

27.5 Insα.ibed plaque found near the Ti n Hau temple at SoKwu 訊而1n， Lamma

The local historian Anthony K. K. 5iu (1989) has written: During the Ming Dynasty, because of the cultivation of incense trees , which gave great pro祉， the population [in the Hong Kong area] increased rapidly However, the Coastal Evacuation at the 1st year of the K'ang Hsi Reign [1661] obliged the people to retreat to the mainland. Fields were left barren , and houses were pulled down As mentioned previous旬， there is no archaeological evidence of an increase in population; indeed , quite the contrary is true , as there are fewer Ming sites than of the preceding 5ung era. It is not clear whether this lack of evidence relates to destruction of sites or real lack of settlement. From a list of villages recorded in late Ming tim白， onlya few can be linked at present to Ming materials excavated at or near those locations: Mong Tseng Wai , 50 Kun Wat , Ho Chung and Wun Yiu (discussed below). With the repopulation of this area after the coastal evacuation order was revoked in 1669 , villages still in existence today were founded. Many of these were walled and moated , indicative of interclan warfare and marauding piracy that is well known from history

1 73

1 74

Historical Archaεology

Rice agriculture intensi五ed and expanded into most of the valleys suitab1e for it. Temp1es were erected , and en1arged as the villages prospered , and 1arge fan-shaped graves began to dot the hillsides. A 1arge grave of this type was discovered in a 10w slope just above the sand bank at Lung Kwu Sheung Tan. 1t w晶晶rst believed that the grave might be earlier than Ch'ing , but , on thorough investigation , it too was dated from the textua1 sty1e of an inscription insi世 the tomb to the eighteenth centu可 or 1ater. After Hong Kong was ceded to the British in 1841 , customs stations were established at severa1 points , including Cheung Chau , Ma Wan , Kow1oon andJunk 1s1and. The 1atter has been the subject of severa1 excavations. Fragm凹的 of a 1arge stone tab1et were discovered on the site in 196斗， and the one missing piece was found in the 1978 excavation. 1t can be trans1ated as: “ Annam has gracious1y shown respect by sending tribute." From the use of certain terms , a secondary inscription suggests that the tab1et dates from Ch'ing times. It seems clear that , at some point , the customs station at] unk 1s1and received tribute from Vietnam , perhaps over many years , and there may have been an earlier station there during the Ming or Sung A text dating between 1796 and 1820 noted that “ at Ting Chi 叭lan [quite likely an old name for Junk Bayl the ruins of an old customs house can be seen" (cited in Siu and Meacham 1982). However, the most recent excavation there in 2006 found no materia1 pre-dating the Ch'ing era. 1n the eighteenth and nineteenth centuries , forts at Tung Chung and Fan Lau on Lantau 1s1and , and at Tung Lung 1s1and were built for the stationing of Chinese troops and the contro1 of waterways. These forts have all been excavated and restored by the Antiquities Offìce. A battery overlooking the passage between Tung Chung and Chek Lap Kok has a1so been fully excavated. The information gleaned from excavation proved usefu1 in understanding the 1ayout of the forts' interiors and how they functioned. Other forts or batteries are known from historica1 texts an cIJor nineteenth centu可 paintings to have been in Tsim Sha Tsui and Yau Ma Tei (still remembered in the street named “ Battery Street") , but no archaeo1ogica1 remains have ever been found of them. The administrative and military headquarters of this region was situated in the Kow1oon Walled City, 10cated in the northeastern corner of Kow1oon. It was bui1t in 1843-47 , shortly after Hong Kong 1s1

Later Historical Archaeology

2 7. 6 The Ch'ingfort at Fan Lau , Lantau , in the 19505

Ch'ing offìcials and soldiers depart吐， but the compound remained , and , by the 1950s , had become a slum. In 1994 , it was demolished , retaining only the remaining nineteenth century structures and paving stones. Archaeological excavations on the site recovered remnants of the old garrison wall and fragments of the original plaques that had been mounted on the main ga仗， one of them bearing the Chinese characters for "Kowloon Walled Cit手" The site is now a park with several historical items on display Am句 or pottery-producing village at Wun Yiu near Tai PO was believed to have been established in the early Ch'ing er且， although the clan genealogies indicate that the village and its pottery industry was in existence by the late Ming. The name Wun Yiu (bowl kiln) occurs on an early Ch'ing list of villages reoccupied after the coastal evacuation. A recent archaeological investigation of the site provided material that allowed for a fìrm dating of the early phase of the pottery manufacturing there to the last decades of the Ming dynasty (1 610-44). There are huge mounds of ceramic wastage at this si伐， mostly misfìred bowls and dish白. The clay source for this industry was a kaolin deposit uphill from the village. The archaeological investigation revealed some of the pits where clay was d1.:嗯， in addition to clay-processing metho血， such as watermill or animal powered grinders and clay-soaking tanks. In the early twentieth century, the kilns were still producing pottery, though at a much-reduced rate , 的 competition from mass production of ceramics began to take its toll on the traditional industry The last fìring of the kilns at Wun Yiu took place around 1917 , according to an old potter interviewed by local historian James Hayes A much larger and more important industry of the nineteenth and early twentieth centuries in the N ew Territories was the burning

1 75

176

日iSLorical Archaeology

of shells and coral to make lime. In a Hong Kong Government Gazette notice of 1901 , a survey revealed that this industrγemployed around 2 , 000 m凹， involved more than 400 boats and collected shell and coral from 11 locales. Apart from rice cultivatio日， it was the largest economic activity in the New Territories at the beginning of the twentieth centuηz But it was in declin亡， and , as resources decreased , a number of disputes ending in violence were recorded in subsequent years. Like its predecessor in the Tang dynasty, this industry appears to have died out from over-exploitation and the appearance of a new and cheaper materi泣， in this instance cemen t. Regarding Ch'ing graves and tombstones , large fan-shaped graves dot the hillsides all over Hong Kong. These relate to the indigenous clans and thus are not investigated by archaeologists. However, they often need to be moved because of developme凹， and these occasions constitute huge missed opportunities for providing information on local histo ry, especially on the dating of the earlier clan founders believed to belong to Ming times. 0句 ects or inscriptions inside the tombs would probably allow for such a dating , but the grave removals are carried out by workers from municipal services and have not been supervised by the Antiquities Office. An example is the removal in the 1980s of the grave belonging to the Shatin Wai clan's founding ancestor, believed to date to around 159 1. Ch'ing burials are occasionally encountered during archaeological excavations. These are generally thought to belong to the boat people ,

27.7 Large Ch'ing grave at Lung Kwu Sheung Tan , Tuen Mun

Later Historical Archaeology

who often bury their dead in sand banks behind beaches. Some of these burials exhibit traditional customs , such as coins in the mouth or reburial of bones in a box. A burial at Pui 0 on Lantau had a well-worked piece of flint and beside it a rusty piece of iron. This was instantly recognized as a traditional "strike-a-light" device for creating a fire 一一位riking a spark from the flint. A few years prior to that excavation , 1 had made a trip into the interior of Mindanao , and , during a break, the local guide took out his strike-a-light kit and proceeded to make a small fire. 1 asked why he did not carry a lighter, and he replied that they were less reliable In addition to the main Punti or Cantonese clans described above , other important ethnic groups are the Hakka farme芯， and

2 7. 8 Secondary “boxburial" 01 late Ch'ing at Shek Pik, Lantau

27.9 Ch'ing burial on tile pillows at Pui 0 , Lantau

177

1 78

Historical Archaεology

Hoklo and Tanka boat people. The Hoklo and Tanka n.sherfolk have probably been in the territ。可 for centur肘， but there is no evidence as to their original arrival in these waters. The Hoklo speak a dialect of Fujian province. Previously, it was often said that the Tanka were descendants of the local aboriginal population , but this was something of an urban myth. They are simply Cantonese-speaking n.sherfolk , no more nor less descendants of the Yueh than are the Punti or Cantonese land-dwellers. Hakka farmers have been immigrating into the territory from the eighteenth century, if not earlier, and are noted for settling on and making a living from the least desirable land , the prime farmland already having been occupied by the main clans. The folklore of these ethnic groups has much to contribute to an understanding of life in Ch'ing times. The study of the development of Hong Kong's population and its later history, trade , growth of settlements , etc. is the task of ethnography and history, with archaeology of much-reduced importance , but it still provides useful data to complement historical studies.

Part IV Summary

Prεsεnt Knowledgεand FuturεProspεcts

Investigations into Hong Kong's earliest periods of occupation have made enormous progress since the 五rst archaeological work in the 1920s. The pre-war work of Heanl句; Shellshear, Schofìeld , Finn and Chen resulted in a large number of sites being record吐， a greater public awareness of the wealth of archaeological material in this region , and many good studies of the material recovered More recent work has thrown new light on the cultural sequenc亡， the dating of the different phases , and the technology and way of life of the early people. Fieldwork and publications have become much more systematic and scienti且c. Archaeological material is now held in a centralized collection , and measures are being sought to give important sites some form of protection. On the negative side , bureaucracy has had a suffocating effect on archaeologists , and the central collection is woefully under-catalogued and little more than a giant warehouse. A computerized database is still lacking Proper publication has almost ceased , and there is a huge backlog of unpublished and undigested reports gathering dust on the shelves of the Antiquities Offìce. Probably the most important single accomplishment of the fìrst 80 years of Hong Kong archaeology is the establishment of a welldefìned sequence of cultural phases , beginning at 5000 BC , which provides the framework for the study of local prehist。可 Work on the dating and detailed division of the sequence continues and will remain an important task for some years to come. However, with so much attention focused in recent years on salvage work , the research and collation of data has seriously lagged.

182

Summary

Identification of the main cultural phases represented at properly excavated sites is now a relatively simple matter, with the benefit of pottery seriation studies conducted in the 1970s. An exception is the Late Neolithic , since the characteristic pottery types of that phase appear to continue into the early part of the Bronze Age. Therefore , sites with hard , soft and coarse geometric and coarse corded types (as seen , for example , at Shek Kok Tsui , Lo So Shing , Tai Kwai Wan and Sha Lo Wan to cite a few among many examples) in a single cultural layer may represent either a single period of occupation during a transitional Early Bronze Age or a mixture of assemblages from two phases of occupation. Micro-stratigraphic studies may eventually resolve this issue Dεscription

of Sitεs

A large number of sites were recorded in the 1930s and 1950s on the basis of surface fin血， principally adzes. Unfortunately, in most cas白， it is not possible to classify these sites by cultural pha仗， Slnce many of the adze types are found in all three prehistoric periods. Schofield listed nearly 120 sites on his map in the 1930s , but only 87 sites could be documented for the Archaeological Map of Hong Kong published in 1972. At prese凹， there are well over 200 archaeological sites recorded by the Antiquities Offi白， though many have now been destroyed by developmen t. Hong Kong's known prehistoric sites have no architectural features and cannot properly be termed “ villages" or “ settlements. " The remains , of cours亡， do indicate human presen白， and the sites may thus be referred to simply 的“activity site丸" even though the nature of the activity is not always clear. Stone tool and ornament manufacture has definitely been established to have taken place at a number of sites , and cooking and wood-working can reasonably be inferred from the abundance of pottery and adzes. Only a few sites have anim泣，且sh or shell remains. Bronze casting should be added on the evidence of moulds and bronze slag found at Sha PO Tsuen , and similar bivalve moulds in casting position found in burials at Sha Lo 札1an and Kwo Lo Wan The only structural features securely associated with prehistoric cultural remains are the stone slab structure , probably a burial co伍n or platform , in a Bronze Age context at Sham Wan; and an arc alignment of stones at Shek Pik. Burial sites with skeletal material were found at Sham Wan (Middle Neolithic) , Tung Wan Tsai (Late

Present Knowledge and Future Prospects

Neolithic) and Shek Pik (Bronze Age). The existence ofburial grounds may be surmised also from the discovery of numerous complete pottery vessels and stone artefacts at many sites , notably Sha Chau (Late Neolithic) , Tai Wa口， Man Kok Tsui and Hai Dei Wan (Bronze Age) , and Sha Ha (Neolithic and Bronze Age) Shells and mammal and fish bones (food middens) have only been found in significant quantities at Sham Wan , PO Yue Wan , Sha PO Tsuen , Tung Wan Tsai and Sham Wan Tsuen. The survival ofbones is believed to be due to marine influence (sea spray or significant quantities of shell in the sand) preventing the formation ofhumic acids in the soil or, in the case of Sha PO Tsuen and Sham Wan Tsu凹， large quantities of lime in the overlying layer serving the same purpose. There has been discussion of the “ stone circles" as possible early monuments. Unfortunately, there is no evidence to link any of them with the ancient inhabitants; some of the circles may indeed be natural. The two main candidates (on Lamma and Lantau) are more likely to have been structures built as temporary graves or burial urn sites in late historical times. The Bronze Age rock carvings constitute the only visible feature from local prehistory It has been suggested that the narrow ledge in front of the carvings at PO Toi is also made by humans , but , at the other sites , there is no indication that the natural rock forms were altered in any way except for the carving itse旺， which may have been executed on a naturally flat rock face with little or no preparation of the surface The lime-kiln sites of the early historical era are the first architectural remains , apart from the Han chamber tomb at Lei Cheng Uk. The latter is the only tomb structure of certain pre-Ming age in the territ。可 More recent lime kilns and one potte可 kiln at Tai PO are known to date to Ch'ing dynasty times. Abandoned forts at Fan Lau , Tung Chung and Tung Lung also date to Ch 'ing. Similar甘， all temples and other buildings are Ch'ing or later in tim巴， as are all known grav白， boundary markers , etc Landforms in which prehistoric sites occur are mainly sand bars on tombolo islands or at the mouths of small valleys (former lagoons) , headlands and low hillslopes , or flat ridges near the sea. lnland sites near small streams have been reported in many areas of the New Territories , but these have thus far shown little significance Tombolo island sites are Tung Kwu , Sha Chau , Cheung Chau and Siu A Chau. The former two have been investigated in

183

184

Summa

Wan was a1so a tombo1o formation at an early stage , evo1ving into a unique sand bar 10cated at the back of the sheltered 1agoon rather than at its mouth. Lagoona1 sand bars are the most characteristic and certain1y the most numerous of archaeo1ogica1 sites in the Hong Kong area Marked variations have been observed in the extent and height of these sand deposits. At sites such as Chung Hom Wa立， Hai Dei Wan and Lo 50 5hing , on1y a small stump of sand bar remains of what probab1y were much more extensive deposits; at Chek Lap Kok , an iso1ated sand mound stands in the midd1e of the former 1agoon. On the other hand , at 5ham Wan , Tai Wan and Tai Long , sand bars of 100-200 m in 1ength are found. 5urface e1evations vary from around +4.5 mPD at 5hek Kok Tsui and Hai Dei Wan to +9 and +10 .5 mPD at 5ham Wan and Tai Long. The two 1argest sand bar sites by extent of deposit in Hong Kong are Yung Long and 5ha Ha 5ites on 10wer hills10pes or p1ateaus near the sea are a1most exclusive1y of the Bronze A詐 two exceptions are the sites at 5ai Rcconstructing the Past Wan (Midd1e Neolithic) and at the southern end of 5ha Chau (Late Neolithic) , both on flat ridges 10-20 m above the sea. The sites at 50 Kun Wat , Ngau Hom 5ha and the North Lamma 5choo1 are on 10w (20-25 m) hills over1ooking sheltered bays , while the 10wer slopes at Man Kok Ts凹， Tai Long and Mo Tat Wan have yie1ded materia1 up to 30 m elevation. In contrast , prominent head1ands that jut out into the sea with re1ative1y flat tops 20 to 25 m above sea 1eve1 , such as Pa Tau Ku and 5ha Lo Wan on Lantau , and PO Yue Wan on Cheung Chau , are

28.1 Sites are now surveyed more accurately

m句 or

Late Neolithic sites

Present Knowledge a吋 F肌

Cultural deposits vary widely among the tombolo and sand bar sites. At Shek Kok Tsui , for example , the Late Neolithic culturallayer beganjust 15-20 cm below surfa仗， while at Tai Long , a Bronze Age layer was found generally at depths of 80-90 cm and in one part of the site at 1. 6 to 1. 8 m. At Pui 0 , Tang kiln remains and a slaking pit were found 2.5 m below surface. Many sites previously thought to have onlya single , shallow layer of material have proved , upon reinvestigation after the work at Sham Wan , to have deposits at depths reaching 2 .5 to 4 m below ground surface (for exampl亡， Hai Dei Wan , Chung Hom Wan , Tai Long , Shek Pik). At Tung Kwu , an old test pit was reopened and excavated further , yielding a complete po t. The hill sites generally have been discovered after considerable erosion or cultivation. and most of the materi日1 found has been surface or slightly sub-surface in origin. However, excavations at Sai Wa口， Sha Chau (Southern Plateau) and Ngau Hom Sha revealed deeper cultural deposits Rεconstructing thεPast

As discussed above , excavation techniques have undergone vast improvement over the years. The methods of analysis of material have also improved markedly, and stone tools and pottery are now examined by shape , method of manufacture , function , fabric or rock typ亡， source , et c. Much has been learned from this analysis about the technology employed by the early potter and stone worker, and , eventually, these research methods and oth亡的， such as trace analysis and edge wear study, may also provide information on the precise functions of the various types of pottery and stone tools. Above all , there are two aspects of archaeology that have on甘 recently come to the fore: environment and adaptation. Studies of the ancient environment of the Hong Kong area generally has long been recognized as important , but studies of the former environment around individual archaeological sites only began with Sham Wan. It is highly important for the archaeologist to know about the environment in which the early inhabitants lived. New efforts in recent years to obtain pollen and phytolith evidence have already yielded results , but what is sorely lacking are good dated pollen pro五les from secure clay deposits. Such pro且les are needed as evidence for the environmental impact ofhumans in the area , especially with the appearance of cereal agriculture

186

Summa

28.2 Ti ny bits of charred material ω refully collected

28. 3 Seeds of the castor oil plant recovered from an excavation Geft) 的d modern reference samples (right) T

lntimately related to the reconstruction of the palaeo-environment itself is the question of subsistence patterns - how the human population interacted with the natural setting to extract a livelihood. There has been much speculation on this subject by archaeologists , but little research has been directed here specifìcally towards the identifìcation and analysis of food remains. The simple reason for this is that most of the sites investigated in the past did not have shells , animal bones or other food remains. But there are indications now that marginal areas near some of the main occupation sites do have such material.

Present Knowledge and Future Prospects

Hong Kong archaeology is especially handicapped in the study of food remains for another reason - the absence of hearths and carbonized deposit in which plant remains and animal bones might have survived. M句 or sltes that have yielded tens of thousands of potsherds have not produced a single clearly de五ned hearth area with substantial ash and charcoal deposi t. Of cour哎， this phenomenon has rendered attempts to obtain C-14 dates extremely difficult , and charcoal is a highly valued substance on local excavations. Sieving and flotation are now commonly employed when a site is suspected to have important micro-remai肘， and rice phytoliths from Sha Ha have been identi五edfrom prehistoric contexts Another missing feature of local archaeology is habitation structure; no site of prehistoric or even early historical age has yielded any house or hut foundations , definite evidence of a compacted or sunken 日oor， or postholes in a clear pattern to suggest a hu t. Rather, a confusing variety of postholes often presents itse旺， probably constituting the remains of many seasonal camps with tents or easily erected huts with one or a few supporting poles Finally; we may recall the gaps and missing links in Hong Kong archaeology There is a total absence of Palaeolithic and Early Neolithic sites , whereas about 20 sites of the Middle Neolithic are known. No Han village sites have been found , but the large Han chamber tomb at Lei Cheng Uk would seem to indicate there was at least one substantial settlement. Over a hundred sites have been recorded with lime-kiln debris , but not a single house foundation from that era This lime industry of Tang times disappears for 800-900 years , after which another m司jor limekiln complex re-emerges in this ar凹，

28.4 Soil is sieved to recover very small material

28 .5 Taking soil samples for micro-stratigraphy studies

187

188

Summa

Certainl;叭叭th more careful attention to areas not previously investigated and with better archaeological methods , information will eventually be obtained on many of these topics - but only if suffìcient archaeological sites remain for future study. lronical旬， the science and technology that gave birth to modern archaeology have also spurred the destruction of its resources. In the 1920s and 1930s , the building boom related to Hong Kong's industrial revolution prompted commercial sand digging that destroyed or severely damaged a large number of sites along the coast The urban development of the New Territories today is affecting even well-known sites at such a rate that salvage archaeology can hardly cope. And few detailed surveys have been undertaken of the prime areas slated for developmen t. At present (200 7), no archaeological site (except the rock carvings) is legally protected; at one time in the 1990s , there were moves afoot to have the more important sites gazetted as monuments under the Antiquities Ordinance , but it appears that this plan has been shelved. The preservation of sites is vital to the future of archaeolo gy, an extremely limited fìeld of study in any case 一 limited by the surviving maten日1 evidence and what can be inferred therefrom about past human behaviour. ln the small area in Hong Kong , many dozens or even hundreds of sites have been partially or totally destroyed by erosion or developments of the land since the deforestation of the region began 1,000 or more years ago Perhaps we should count ourselves fortunate to have learned as much as we have about early life here. But we should constantly strive to make better facilities available for the study of what little evidence has survived from the remote past , when our predecessors led an existence ve可 different from our own on these ve可 shores

Appεndix:

ThεUnivεrsity

of Hong Kong and Local Archaεology

The university has had a m句 or role in Hong Kong archaeolo gy, almost from the very beginning. News of the first discoveries of Neolithic stone adzes by Heanley in about 1926 very quickly reached Professor ]oseph L. Shellshear of the Anatomy Department. In a paper that Shellshear (932) wrote a few years later, he recorded: Some 仗凹 time afte 臼r司 I口1926] we became associat仗 ed in making a genera survey of the Colony. . Dr. Heanley left Hongkong in 1930 lea 趴 aVl 叭 盯7凡m 呵 1喀 gme 跎ew 叭lt出 ht由 he cωollection to do with as 1 pleased. We have to acknowledge our indebtedness to Sir William Peel , Governor of Hongkong and Chancellor of the University of Hongkong , for the interest which he has taken in this work

The archaeological work

w佑，

however a purely private endeavour ,

at that stage , and it was not until the 1950s that archaeology became somewhat more integrated into the university with the formation of the Department of Geography and Geology Shellshear was at pains to point out that Heanley deserved the sole credit for discovering Hong Kong's prehistory In a letter written in 1952 , Shellshear remarked: Please do not regard me as a father [of local archaeology] with Dr. Heanley He was the father and 1 should have been able to do nothing if it had not been for his great geological knowledge and previous discoveries However, he and Heanley are generally regarded together as the first investigators of local archaeolog乎 According to Sir Lindsey Ride Ccited in Davis 1965) , who knew both archaeologists quite well and “ often accompanied them on their field tri阱. . . they were superbly

190

Appe叫x

energetic and covered tremendous distances in a day at great speed." Shellshear left Hong Kong for Australia in 1935. By the early 1930s , the other major pioneers in Hong Kong archaeology were on board , namely Fr. Daniel]. Finn and Walter Scho五eld. Fr\Finn was a part-time lecturer in geography at the universi你 but it is not known how long he held this position and whether it was honorary or part-time. He passed away in 1936. As noted in Chapter 6 , most ofhis special 五nds were stored at Ricci Hall when the ]apanese invaded , and some material was taken away by a ]apanese of五cial. Finn's articles on the excavation he conducted on Lamma were re-published in 1958 in a book entitled Archaeological Finds on Lamma Island Near Hong Kong , edited by Fr. F. Ryan of Ricci Hall. Archaeology in the early post-war years was carried on by S. G. Davis (later professor and head of the Department of Geography and Geology) and several of his friends. ln 1953 , the Geographic祉， Geological and Archaeological Society was formed at the universi你Its membership consisted of internal , extern訓， graduate and associated students; it was chaired by Professor Davis. It is not known if any fìeldwork was done by this society prior to the formation of the University Archaeological Team three years later ln 1955 , the university was invited by the colonial secretariat to conduct the excavation of a Han tomb discovered during construction at Lei Cheng Uk in Sham Shui Po. Strangely, neither Davis (1965) nor Bard (1 995) mention the event in their accounts of the development oflocal archaeology, but it is clear that this was a pivotal moment for the discipline. The discovery was reported by a friend to Professor Frederick S. Drake , a noted British sinologist and head of the Department of Chinese. He made contact with the Public Works Department and quickly assembled an excavation team composed of students and lectur它的 Among the former was Michael Lau Wai Mai , who later became curator of the Fung Ping Shan Museum. One of the lecturers involved was the geologist Lynn Berry who would later be a key member in the Univ巴的ity Archaeological Team. Eminent visitors to the excavation included]ao Tsung-yi and Lo Hsiang-lin of the Department of Chinese Drake completed the excavation of the tomb in four days and brought the objects from the tomb back to the Fung Ping Shan Museum for stud乎 He later 五led a report to the governme凹， and a brief pamphlet was published. Drake reportedly p

Appendix

full report on the excavation and took his notes with him to England when he retired in 1964. His death in 1974 intervened , and attempts to locate his papers through his son Bernard (also a participant in the excavation) were fruitless. Professor Lo Hsiang-lin published a series of articles about the tomb and its contents. 1 wrote to him in 1976 to enquire about field notes and photographs , and he replied that “at present 1 am working on a third article about the Han tomb , i.e. the burial and matters relating to it. My final aim is to publish a book incorporating these three theses." Once again , death intervened in 1978. Professor Lo's papers were deposited in the Hong Kong collection of the university's library, and 1 inspected them in 1978 There were valuable photographs of the objects in situ in the tomb and of the excavation in progress. A note on these photographs and also several sketches found with them was published in the Archaeological Society's journal (Meacham 1980c) In March 1956 , a University Archaeological Team was formed lt was to consist of 25 members , approximately half from within the university. It began under the auspices of the Institute of Oriental Studies and was chaired by Professor Drake. Mary Trege缸， curator of the Fung Ping Shan Museum , was secretar手 lts objective was t。 “ further the discovery, excavation and preservation of prehistoric sites and relics in the Hong Kong region." An ambitious plan was drawn up to inspect and record all possible sites , to monitor any construction or other digging on beaches and likely sites , and ，已nally， to excavate and publish any promising sites. Areas of the territory were assigned to individuals or small groups , and potential sites were visited on weekends. A progress report dated November 1957 reveals that numerous sites on Lamma , Cheung Chau and Lantau had been visited and surface finds collected. The report also reveals an impressive number of informal talks given by team members on various subjects related to archaeolog于 Sometime in 1958 , the affiliation and leadership of the team shifted to the Department of Geography and Geology and its head , Professor Davis. Reportedly there was a dispute of some sort between Professor Drake and some members of the tea凹， and he ceased being actively involved , though continuing as an advis。可 member\ Also in 1958 , a member of the team , Dr. S. M. Bard (later director of the University Health Service) , discovered a site at Man Kok Tsu

19 1

192

Appe叫x

29.1 The Man Kok Tsui exhibit at the Fung Ping Shan Li bra吵 the University 01 Hong Kong , 1958

by Davis as the “ largest and most outstanding work" of the team. An exhibition of the fìndings at Man Kok Tsui along with information about other sites was mounted at the Fung Ping Shan Museum in December 1958; it was titled “ Stone Age Man on Lantao." A report on the site was later published by Davis and Tregear (1 961) in the newJo盯nalAs叩1 Perspec帥的， which formerly had been Bulletin of the Far East Prehistory Association. Hong Kong University Press became the publisher of the journal in 1958 , while the editorial management remained with the association. In 1964 , the University of Hawaii Press took over responsibility for the publication

Am句 or event in the history of the team was the fiftieth anniversary celebration of the university in 1961. A symposium was held on historical , archaeological and linguistic studies of southern China , Southeast Asia and the Hong Kong region. Bard presented a paper reviewing Hong Kong archaeolo gy, and contacts were made with a number of prominent scholars of China and Southeast Asia history and prehistory, including Professor W G. Solheim , editor of Asian Perspectives , and the sinologist Michael Loewe Another project of the team was to update the map of archaeological sites that had been started by Heanley, Shellshear and Schofìeld , particularly to accurately locate the pre-war sites and to discover new ones. The collection of data continued on into the late 1960s when the Hong Kong Archaeological Society was formed; the Archaeological Map of Hong Kong was eventually published in 1972 by the City Museum and Art Galle可 The driving force in the mapping project was T. N. Chiu of the Department of Geography and Geology

AF耶叫x

193

Both Bard and Davis mention the involvement of the team in pressing for antiquities legislation. Bard wrote in 1964: “We have also been responsible for promoting a new Bill for the protection of historic relics , soon to be passed into law." More than a decade later, the law was 五nally enacted in 1976 There is some disagreement over how and why the team came to be dissolved and the Hong Kong Archaeological 50ciety created in its place. According to Bard (1 995): Around 1964 , after severa1 years of productive activity . . . the team appeared to decline. By 1967 it became clear that the dep1eted team cou1d no 10月er function effectively a吋 that a w吐er public participation was needed. (3 85) However , Davis told me that the team was doing well , but increasingly reliant on people outside the universi句; it seemed that the logical thing to do would be to go public. A general meeting report dated 6 March 1964 , would seem to bear out Davis' accoun t: it lists 17 members from the university and 20 members from the public , with new applications from another 7 people (2 from the university) John Warner, curator of the City Museum and Art Gallery who was listed in attendance at that meeting , told me that the team had grown beyond its narrow limits. On the other hand , the papers available from the mid-1960s indicate that there was much less fieldwork being carried ou t. Another reason for the transformation cited by Bard (in an article published in the supplement to the University Gazette of 1 April 1964) was the existence of the City Museum and Art Gallerγ. from 1962 onwards i的恆的tence has become both a reason and an incentive for forming a [publicl society which can be based in the Museum and utilise its many facilities and expert guidance The Archaeological 50ciety was founded in 1967 and the team , having given birth to the new organization , disbanded. The founding members of the society and its first executive committee were made up largely of ex-team members: Davis was elected chairman , with Bard , T. N. Chiu and C. L. 50 (both of the Department of Geography and Geology) and Mary Visick, of the English Department , serving on the committ仗. Davis served as chairman until 1972 when he retired , and Bard was elected in his place. The society prospered in the 1970s with the major excavation at 5ham Wa口， Lamma Island Active university members were Chiu , 50 and Henry Talbot , all from the Department of Geography and Geolog乎 Bard served as chairman

29.2 d可r) β叫

Pr可essor S. G. Davis and Henry Talbot in the d仰 pit at Sham Wan ,

1972

194

Appe叫x

29.3 T. N. Chiu (above) and Shih Hsio-yen at an excavation in 1985

until1976 and stepped down to join the civil service and set up the Antiquities and Monuments Office. 1joined the society upon arrival in Hong Kong in 1970 , and one of the greatest first impressions of the colony was to listen to tales from Bard and Davis about the early days of local archaeolo gy, war and POW stories , and other assorted subjects , not to mention past and present personality and factional conflicts. ln 1980 , 1 became af且liated with the university as an Honorary Research Associate (later Honorary Research Fellow) in the Centre of Asian Studies , continuing as editor of the society's journal , a position 1 had held since 1972. ln 1984 , T. N. Chiu was elected chairman of the society, the first Chinese to hold the position. Tze-nang was a very personable man and had many friends both at the university and in the societ乎 He and 1 collaborated on a number of projec的， in particular, the survey of levels of prehistoric cultural deposits in various back beach sand formations. lt was a source of great sadness when he had to step down after just one year in office due to liver cancer; he died on 20 September 1985. lt was decided that the next issue of the journal would be dedicated to his mem。可 and it was my honour as editor to prepare this special issue , which included testimonials and papers from students and colleagues in the Department of Geography and Geology, as well as his friends in the society 1 followed Tze-nang as chairman and continued in that position until 1996. Other members of the university who were involved indirectly in archaeological work were Professor F P Lisowski and Dr. Nina Jablonski - both honorary advisors to the society on human and animal bones , respectively lt was a very fortunate coincidence that Lisowski had studied the effects of cremation on human bone earlier in his career, for the Sham Wan excavation yielded quite a number of Neolithic human bone fragments that he was able to identify definitively as cremated Many other university personnel from various departments have also contributed their expertise to various archaeological studies in recent decades , notably Professor Ron Hill and Dr. Richard lrving of Geography and GeologμD r. Richard Corlett of Botany, Professor Brian Morton of Zoolo gy, Dr. Frank Kendall of the Radioisotope Unit , Professor Shih Hsio-yen of Fine Arts , and Drs. David Workman and w W S. Yim of Earth Sciences. There have also been unfortunate outcomes: certain university members served on the Antiquities Adv

AF耶叫x

In 1995 , the University Museum and Art Gallery (successor to the Fung Ping Shan Museum) organized a conference on “ Archaeology in Southeast Asia" with funding from the Lord Wilson Heritage Trust and the Land Development Corporation. The museum executive director was Michael Lau Wai-mai who as a student at the university had participated in the excavation of the Lei Cheng Uk tomb in 1955 The conference brought together archaeologists from Hong Kong , the Mainland and several Southeast Asian countries plus Australia and the United States. Despite the rather formidable language barrier, the conference was a success , and there were “ frank and productive exchanges." Its proceedings were published in a large volume in the same year (Yeung and Li 1995). The university has had a long and distinguished participation in Hong Kong archaeology spanning 75 years. At present (2008) , however, there is no one at the university who is actively involved in local archaeological fieldwork - a regrettable situation that one hopes will change in the near future.

195

Rεfεrεncεs

Ant Kong:Ar， κζ chaεolog，♂ical Discωovεnεs at Sha Ha , Sai Kung. Hong Kong Leisure and Culture Services Department. Balfour, S. F 1970. Hong Kong before the British. Journal of the Hong Kong Branch of the Royal Asiatic Socie 紗， 10:134-79 (Reprinted from Ti en Hsia Month妙， vols. 11 and 12 , 1940-1941 , Shanghai.) Bard , S. M. 1967. A survey of prehistoric sites in the Hong Kong region. Symposium on Historiω1， Archaeological , and Linguisti正 Studies on Southern China , Southeast Asia, and Hong Kong. Hong Kong: The University of Hong Kong 一一-. 1988. In Search of the Past: A Guide to the Antiquities of Hong Kong. Hong Kong: Urban Council. 1995. Archaeology in Hong Kong: A review of achievement. In Archaeology in Southeast As間， ed. C. T. Yeung and Brenda W L. Li. Hong Kong: University Museum and Art Gallery, The University of Hong Kong Chen Kung-jit (Chen Gung-che). 1952. Hong Kong PrehistoricArtifacts. Hong Kong: Kowloon Commercial Association [pamphlet; in

Chinesel 一一一-. 1957. Archaeological surveys and excavations in Hong Kong Kaogu Xuebao. No. 4 [in Chinesel Chiu Siu-tsan. 1995. Major achievements of archaeological rescue pr叮 ects conducted in Hong Kong since 1990. In Archaeology in Southeast Asia , ed. c.T. Yeung and Brenda W L. Li. Hong Kong: University Museum and Art Gallery, The University of Hong Kong.

198

Refere肛es

s. Sutton. 2006. the Margins: Culture , Ethnicity， αnd Frontier in Early Modern China. Berkeley: University of California Press Davis , S. G. 1965. Archaeological discovery in and around Hong Kong. Journal of the Hong Kong Branch of the Royal Asiatic Society. Vol. 5 Davis , S. G. and M. Tregear. 1960. Man Kok Tsui Archaeological Site 30 , Lantau Island , Hong Kong. Asian Perspectives , 4: 183-212 Drewi缸， Peter. 1995. Neolithic Sha Lo 札'an. Occasional Paper no. 2 Hong Kong: Antiquities and Monuments Offi. ce. Finn , D.]. 1958. Archaeological Finds on Lamma Island near Hong Kong. Adapted from 13 articles appearing in the Hong Kong Naturali泣， vols. 4-7 , 1932-1936; ed. T. F Ryan. Hong Kong Ricci Publications Guangdong Provincial Institute of Cultural Relics and Archaeology 2005. Henglingshan in Boluo County: Eχcaì叫ion R叩ort of Shang and Zhou Period Cemete叮叮1 2000. Beijing: Science Press [in Chinese] Heanley, C. M. 1928. Hong Kong celts. Bulletin of the Geological Society 。fChina ， 7: 209-14. Be可 mg. Heanley, C. M. and]. L. Shellshea r. 1932. A contribution to the pr跎 εhistory of Hong Kong and thεNewT，εr叮ritories. Proζαεεd 品mgs5 of the 甘εF盯 t Con 時 gr陀ε5臼5 of Pr 陀 eh肘 istωor叩 pp. 63-77 , Publisher unknown. Hong Kong Archaeological Socie作 1968-2008. Journal of the Hong Kong Archaeological Socie妙， vols. 1-15. 一一一 1972. The Archaeological Map of Hong Kong. Hong Kong Government Printer Hong Kong Museum of Hist。可 2005. Lei Cheng uk Tomb. Hong Kong . Hong Kong Museum of Histor手 Hsu Sung-hsieh. 1939. History of the Peoples of the Pearl River Delta Shanghai: China Book Company [in Chinese] Jao Tsung-yi. 1954. Prehistoric remains in south China and the culture of Yin-hsu. Ta-lu Tsa-chih , vol. 8 [in Chinese]. Lam , Peter (editor and co-author). 1983. Archaeological Finds from Han Tombs at Guangzhou and Hong Kong. Hong Kong: Art Gallle門， The Chinese University of Hong Kong. 一一一-. 1984. Archaeological Finds from Pre-Qin Sites in Guangdong Hong Kong: Art Gallery, The Chinese University of Hong Kong Crossley, Pamela Kyle and Helen F Siu , Donald Empi陀 at

Refere此es

Maglio凹，Rafael.

1975. Archaeological Discovery in Eastern Kwangtung ]ournal Monograph IIl. Hong Kong: Hong Kong Archaeological Society Meacham , William. 1973. Notes on the early Neolithic in Hong Ko月 journal of the Hong Kong Archaeological Socie紗， vol. 4 一一一一-. 1976. Roch Carvings in Hong Kong: An Ill ustrated and Interpretive Study. Hong Kong: Christian Study Centre on Chinese Religion and Culture. 一一一一-. 1980a. Archaeology in Hong Kong. Hong Kong: Heinemann 一一一一-. 1980b. The Archaeology of Hong Kong. Archaeology 33(4) 16-23 一一一-. 1980 c. Photographs and a fìeld note of the 1955 excavation at Lei Cheng Uk. journal of the Hong Kong Archaeological Socie紗， vol. 8 一一一-. 1995. Middle and Late Neolithic at “ Yung Long South." In Archaeology in Southeast Asia , ed. C. T. Yeung and Brenda W L Li Hong Kong: University Museum and Art Gallery, The University of Hong Kong. Meacham , William (ed.) 1978. Sham Wan , Lamma Island: An Archaeological Site Study. ]ournal Monograph IIl. Hong Kong Hong Kong Archaeological Society .(吐.) 1994且. Archaeolo耳cal Investigations on Cheh Lap Koh Island. ]ournal Monograph IV Hong Kong: Hong Kong Archaeological Society 一一一一一 (ed.) 1994b. Archaeological Discovεry at Cheh Lap Koh. Hong Kong: Hong Kong Archaeological Society Peacock , B. A. V and Taryn]. P Nixon. 1988. The Hong Kong ArchaeologicalSurv句 : Subsurface Investigation Reports. Occasional Paper no. 1. Hong Kong: Antiquities and Monuments 0伍ce Rogers , Pamela Rumball and Nan W Leininger, Sophia Mirchandani , ]ulie Van Den Bergh , Ellen A. Widdowson. 1995. Tung ì凡是mTsai A Bronze Age and Han Pe叫od Coastal Site. Occasional Paper no. 3. Hong Kong: Antiquities and Monuments Offìce Schofìeld , Walter. 1975. An Archaeological Site at Sheh Pih. ]ournal Monograph 1. Hong Kong: Hong Kong Archaeological Societ手 Siu Kwok-kin , Anthony 1989. The history of Hong Kong: From a village to a city journal of thεHo嗯 Ko嗯 Branch of the Royal Asiatic Society 29: 390-4 Siu , K. K. and W Meacham. 1982. On the question of an early customs house at ]unk lsland. journal of the Hong Kong Archaeological Socie吵 9: 95-104.

199

200

Refere肛es

Sun De-rong (Kevin Sun Tak-wing). 2007. Recent archaeological discoveries and research in Hong Kong. Kao Gu , 6: 3-7 [in Chinese]. Tang Chung. 1991. A journey into Hong Kong's Archaeological Past Hong Kong: Regional Council [in Chinese and English]. Watt , James C. Y. (Qu Zhiren). 1970. A Han Tomb at Lei Cheng Uk Hong Kong: City Museum and Art Gallery, Urban Council. . 197 1. Hong Kong's prehistory Arts of Asia , 1( 6) :4 8-52. Weinberger, W 1948-49. Some notes on early pottery and stone artefacts excavated on Lamma Island , Hong Kong. Transactions London: Oriental Ceramic Society Welch , W M. 1962. A new archaeological site in Hong Kong.journal 。f the

Hong Kong Branch of the Royal Asiatic Socie秒 2: 109-114. WongWai-y哎， Sharon. 2006. A preliminary study of the distribution and consumption of ceramics in Hong Kong during the SongYuan period. Bulletin of the lndo-Pac昕c Prehistory Association , 26:140-6. Williams , Bernard. 1980. Hai Dei Wan. journal of the Hong Kong Archaeological Socie吵， 8: 27-51 Xinhua News Agency 2006. Discovery of ancient quarry rewrites Hong Kong's human histOlγ People's Dωly. January 15 Yeung Chun-tong and Brenda Li Wai-ling (ed.) 1995. Archaeology in Southeast Asia. Hong Kong: University Museum and Art Gallery, The University of Hong Kong

Indεx

Aitken , Martin 49 Antiquities and Monuments Office viii , 33 , 40 , 41 , 53 , 58 , 61 , 62 , 63 , 64 , 65 , 81 , 85 , 95 , 131, 170 , 172 , 174 , 176 , 181 , 182 , 194 Archaeological Society (see Hong KongArchaeological Society) Au Ka Fat 60 Bacsonian 70. 71 Bard , S. M. 35 ，沛， 37 ，誦， 42 ， 43 、斗斗， 191. 192. 193 . 194. 197 Big Wave Bay 125 , 126 , 128 Boluo 114 , 115 , 198

60 ， 117 ，

190 ,

Cameron , Hugh 52 ，弓 4 ， 15 1, 154 Chau Hing-wah 60 Chek Lap Kok 5 ， 37 ， 51 ， 56 ， 57 ， 58 ， 59 ， 61 ， 7斗 ， 75 ， 80 , 86 , 87 , 90 , 91 , 92 , 112 , 113 , 151 , 152 , 163 , 164 , 174 , 184 , 199 Chen Kung-jit 25 , 30 , 32 , 3 斗， 123 , 126 , 181 , 197 Cheung Chau 斗， 39 ， 48 ， 51 ， 75 ， 86 ， 87 ， 106 ， 107 ， 125 , 126 , 127 , 129 , 150 , 174 , 183 , 184 Chinese University of Hong Kong , The 25 , 5斗， 58 ， 59 , 61 , 113 , 198 Chiu Siu-tsun 60 Chiu Tze-nang 78 , 192 , 193 , 194 Ci ty Museum and Art Gallery (Ci ty Hall Museum) 31 , 33 , 40 , 148 , 192 , 200 Corlett , Richard 194 Chung Hom Wan 6 , 37 , 39 , 50 , 53 , 78 , 79 , 86 , 87 , 90 , 18斗， 185 Davis , S. G. 33 , 3斗， 35 , 37 , 43 , 190 , 191 , 193 , 194 Diamond Hill157 , 158 , 190 , 191

Drake , F S. 35 , 143 , 190-191 Fan Lau 174 Fan Ling 36 , 158 Fat Tong Mun 35 , 123 , 129 , 160 Finn , D. J 1斗 ， 15-20 , 26 , 27 , 29 , 30 , 3 1, 39 , 40 , 43 , 90 , 181 , 190 , 198 Frost , Ray 78 , 171 Fu Tei Wan 59 , 74 , 75 , 86 , 90 , 91 , 92-94 , 149 , 163 Fung Ping Shan Museum 17 , 31 , 33 , 190 , 191 , 192. 195 Ha Law Wan 59 , 163-168 Hac Sa Wan 38 , 40 , 53 , 79 , 87 Hai Dei Wan 39 , 11 3 , 119 , 120-122 , 154 , 183 , 184 , 185 , 200 Haigh , Brian 126 Hayes , James 2 斗， 169 ， 170 ， 175 Heanley, C. M. 10-1 斗， 30 , 111 , 150 , 181 , 189 , 189. 192 High Island 38 , 39 , 77 , 169 , 171 Hill , Ron 194 Ho Chung 63 , 169 , 173 Hoabinhian 70 , 71 Hoifung (Haifeng) 2ι凹，刃， 1 阱 ， 132 ， 14 8 Homutu 90 Hong Kong Archaeological Society viii , 214 , 29 , 31 , 32 ， 37 ， 39 ， 40 ， 43 ， 52 ， 53 ， 5斗， 55 ， 58 , 61, 62 , 63 , 65 , 77 , 8 1, 95 , 100 , 102 , 120 , 126 , 151 , 169 , 170 , 171 , 190 , 192 , 193-194 Hong Kong Museum of History 17 , 31 , 33 , 40 ， 的， 43 , 85 , 148 , 169 , 198 Hong Kong University (see University of Hong Kong , The)

202

lndex

Hong Kong University Press Hsienyentung 71 Hung Shing Yee 33 , 34

3 斗， 192

Penny's Bay 5斗， 61 ， 169 PO Toi 89 , 12斗， 126 , 127 , 128 , 183 PO Yue Wan 29 斗 8.51.106.107.108.183.18斗 Pui 0 5 1. 79. 80.149.156.177.185

Irving , Richard 194 Jablonski , Nina 194 Jao Tsung-yi , 115 , 126 , 190 , 198 Joukows缸， Martha 43 Junk Island 17斗， 199 Kau Sai Chau 日tes 57 , 81 , 82 , 85 , 131 , 132 , 133 Kau Sai Chau rock carving 125 , 126 Kellett Island 158 Kendall , Frank 194 Kowloon Walled City 174-175 Kwo Lo Wan 59 ， 86 ， 90 ， 9斗， 1 日， 163 ， 182 Lau Wai-mai , Michael195 Lei Cheng Uk 33 , 3斗， 35 ，斗0 ， 131 ， 133 ， 143-148 ， 183 , 187 , 190 , 19 弓， 198 ， 199 ， 200 Li sowski , F. P 47 , 48 , 194 Lo Hsiang-lin , 1 斗3 ， 190 , 191 Lo So Shing 39 , 51 , 80 , 107 , 150 , 151 , 182 , 184 Lung Ha Wan 125 , 127 Lung Kwu Sheung Tan 60 , 158 , 159 , 174 , 176 Lung Kwu Tan 55 , 56. 60 , 95 Ma-pa 69 , 70 Ma Wan 61 , 107134 , 151 , 1 泣， 174 Macau 13 , 38 ， 40 ，日，呵， 79 ， 87 ， 127 ， 140 MacLehos亡， Murray 43 ，峙，立 Maglioni , Rafael26-29 , 31 , 33 , 132 , 148 , 199 Maher, Bob 31 Mai PO 158 Man Kok Tsui 35 , 111 , 113 , 117-119 , 121 , 183 , 18斗， 191-192 ， 198

Mo Tat Wan 111 , 184 Mong Tseng Wai 日， 169 ， Morton , Brian 194

173

仁時到

n3

Z5

E4α

QU

Pa Tau Ku 61 , 106 , 184 Pak Mong 59 , 107 , 13斗， 1 斗9 Peacock. Bri且n 53 , 95 , 199 Peng , Charles 126 Peng Chau 159

口。

勻、 J

、

M

俊

門、J

叮叮8 叮叮h

叫 52m

qua--

叫 mHvm

arr

tρua

、

σbσbHO

Mhush NNNN u--zvc yufhuna uHhyL

-i

4l

Sai Wan 39 , 86 , 87 , 106 , 18斗， 185 Schofìe肘， Walter 1 斗， 21-25 ， 27 ， 31 ，咒，站，斗 3 ， 1 的， 163 , 181 , 182 , 190 , 192 , 199 Sek Kong 158 , 159 Sha Chau 21 , 10弓， 106 , 109 , 129 , 183 , 18斗， 185 Sha Ha 63 , 6斗， 86 , 90 , 102 , 107 , 109 , 183 , 184 , 187 , 197 Sha Lo Wan 61 , 102 , 106 , 182 , 184 , 198 Sha PO Tsuen 6 ， 7 ， 39 ，斗 8 ， 107 ， 109 ， 110 ， 111 ， 116 ， 149 , 150 , 182 , 183 Sha Tau Kok 85 Sha Tsui 77 , 171 Sham Wan 22 , 37 , 39 , 42一斗9 ， 50 , 51 , 53 , 56 , 58 , 59 , 77 , 78 , 79 , 80 , 86 , 87 , 89 , 90 , 91 , 102 , 107 , 108 ， 109 ， 110 ， 112 ， 113 ， 149 ， 15斗， 159 , 182 183 , 184 , 185 , 193 , 19斗， 199 Sham Wan Tsuen 51 , 58 , 80 , 151 , 152 , 153 , 159 183 Shek Kok Tsui 38. 182. 184. 185 Shek Pik site 21-25 , 30 , 32 , 37 , 39 , 54 , 85 , 86 , 87 , 90 , 107 , 109 , 111 , 113 , 119 , 128 , 129 , 158 17斗 177. 182. 183. 185. 199 Shek Pik lower carving 115 , 123 , 12斗， 126 ， 127 Shek Pik upper carving 126 Shellshear,]. L. 10-1 斗， 15 ， 17 ， 30 ， 35 ， 111 ， 181 ， 189 , 190 , 192 , 198 Shih Hsio-yen 194 Siu A Chau 21 , 106 , 183 Siu Kwok-kin , Anthony 173 So Chak-lam , Albert 193 So Kun Wat 12 , 63 ， 111 ， 13斗， 169 , 172 , 173 , 184 Solheim , WG. 3斗， 37 ， 38 ， 39 ， 43 ， 45 ， 192 Souza , Jerry 63 Ssu Ma-ch'ien 8 Sun , Kevin 131 Tai Kwai 叭lan 107 , 182 Tai Long Wan 38 , 39 , 52 , 126 , 18斗， 185 Tai Wan (Lamma) 口， 15-20 ，缸， 27 ， 31 ， 39 ， 86 ， 90 ， 113 , 119 , 183 , 184 Talbot , Henry 194 Tang Chung 5 斗， 60 ， 200 Tregear, Mary 36 , 37 , 117 , 191 , 192 , 198 Tsangpiyen 71 Tung Chung fort 17斗， 183 Tung Chung rock carving 126 , 127 , 128

I吋ex

Tung Kwu 21 , 22 , 39 , 50 , 105 , 106 , 149 , 183 , 185 Tung Lung fort 17斗， 183 Tung Lung rock carving 12斗， 126 ， 129 Tung Wan Tsai 61 , 65 , 107 , 108 , 109 , 13斗， 182 , 183 , 199 University Archaeological Team 33 , 35 , 36 , 40 , 117 , 119 , 134 , 151 , 190 , 191-193 University ofHong Kong , the 10 ， 30 ， 33 ， 3斗， 47 ， 78 ， 126 ， 1 斗3 ， 189-195 ， 197 ， 199 ， 200

Visick , Mary 193 Von Koenigswald 69

203

Warner, John 193 Watt Chi-ying , James 36 , 37 , 43 , 147 , 200 Welch site 81-85 Wong Chuk Hang 115 ， 12斗， 126 , 127 , 128 , 129 Wong Tei Tung 63 , 72-73 , 85 Workman , David 19斗 Wun Yiu 173 , 175 Yi Long 51 , 52 , 150 , 151 , 152 Yim ，嗎TWS.194

Yip Cho-hong , Alex 60 YungLong27 , 53 , 62 , 63 , 86 , 87 , 95-102 , 103 , 108 , 109 , 130 , 184 , 199 Yung Shu Wan 17 , 53 , 111 Zhang Senshui 73