14th Century Singapore: The Temasek Paradigm

14TH CENTURY SINGAPORE: THE TEMASEK PARADIGM LIM TSE SIANG (B.A. HONS., NUS) A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF ARTS DEPARTMENT OF HISTORY NATIONAL UNIVERSITY OF SINGAPORE 2012 ACKNOWLEDGEMENTS This thesis is not just the fruit of one’s labour, but that of many whose help I am grateful to have received. I would like to thank first and foremost everyone in the Department of History, my alma mater, for their willingness in accepting me as a graduate student, generosity in granting me a research scholarship, understanding in my decision to switch from a history to a historical-archaeology thesis-topic as well as kindness in supporting my request for a tuition fee waiver at the end of the scholarship. It is under the aegis of this department that I have been able to continue my wilful passion for history and often reckless endeavours in archaeology. I owe my greatest debt to my supervisors – Dr. Mark V. Emmanuel, Associate Professor John Norman Miksic and Associate Professor Peter Borschberg – for the successful completion of my thesis and candidature in the M.A. programme. Both Dr. Emmanuel and Professor Borschberg have graciously taken me under their wings despite the fact that my research falls outside their respective interests and specializations. They have never ceased to extend their patience and support despite my encountering of countless setbacks and failures in the course of my work. It is their professionalism as supervisors and scholars to which I look up and seek to emulate during my candidature. The same can no less be said of Professor Miksic, undoubtedly one of the greatest archaeologists of Southeast Asia, whose contributions to the i understanding of pre-colonial Singapore inspired this thesis in the first place. I am extremely grateful for this privileged opportunity to learn from the breadth of his knowledge and depth of his wisdom; it is an experience and honour which I will forever cherish. I wish to thank in particular my mentor and friend, Associate Professor Bruce Lockhart, for his personal guidance throughout my entire candidature. His unwavering and selfless support helped me overcome many bouts of uncertainty and doubts, whereas his strictness and impartiality kept me on the right track in my work. I will remember these lessons well as I move on towards the next stage in life. Several other scholars and individuals have also been the guiding beacons in my foray through the challenging path of an academic life. Their influence and support is invaluable for the completion of this thesis. I wish to thank Professor Merle Calvin Ricklefs for his instruction on academic rigour and thoroughness in research; Associate Professor Maurizio Peleggi for his enlightening lessons on critical analysis, subject-matter appreciation as well as sophistication and flair in one’s work; Associate Professor Huang Jianli for his keen interest, advice and encouraging words; Dr. Quek Ser Hwee, Dr. Sai Siew Min and Dr. Loh Kah Seng for their support and care whilst in their service as a tutor; Mr. Kwa Chong Guan and Dr. Derek Heng Thiam Soon for their time in advising an inexperienced and naïve graduate student; Dr. Goh Geok Yian and Mr. Lim Chen Sian for the opportunities to study the artefacts stored at No. 6 Kent Ridge Road and participate in various excavations in Singapore; Carrick Ang for his humour and help in digitalizing the soil profiles for posterity; and not least my senior-inarchaeology Xin Guang Can and volunteer Aoki Ryoko for their assistance and company in laboratorial analysis as well as patience with my temperaments and antics. Finally, I wish to thank Peter, Sally, Chris, Tim and my wife, Cherylyn, of the Wong family in Brisbane for their welcome, kindness and love over the past year, as well as my parents in Singapore for their trust and indulgence of freedom to pursue my interests. Any success in this thesis is attributed to all mentioned; any fault is mine alone. ii TABLE OF CONTENTS Acknowledgements ……………………………………………………...i Table of Contents …………………………………………………….iii Summary ……………………………………………………………...iv Abbreviations ………………………………………………………………v List of Tables ……………………………………………………………viii List of Figures ……………………………………………………………...ix I. Introduction ………....……………………………………………………1 II. Temasek-Singapura: The Paradigm and its Sources III. Artefact Society IV. Conclusion ……………………………………………22 …………………………………………………….51 ……………………………………………………………116 Bibliography ……………………………………………………………120 Glossary ……………………………………………………………150 Appendices A. STA Artefact Samples: Analyses and Deductions …………………………………………..153 B. STA Soil Stratigraphic Profiles: Sampled Excavation Units …………………………………………..209 C. STA Raw Artefact Data: Pre-colonial Samples …………………………………………………...251 iii SUMMARY Over two centuries have passed since the revelation of the existence of an ancient settlement in Singapore, but knowledge of this pre-colonial polity has not advanced much with time. Despite the archaeological recovery of material cultural remains of this settlement in recent years, historical discourse on the subject remains largely confined to either the epistemological significance of this settlement in the colonial and post-colonial histories of the Malay-speaking people in the Malay Archipelago, or the general importance of this landmark within the context of maritime trade in Southeast Asia. More often than not, artefacts are used only to highlight these narratives but are themselves seldom addressed. This thesis is hence dedicated to the study of these artefacts as primary sources of 14th century Singapore. It seeks to address a fundamental question which underlies all narratives of the settlement but has hitherto been inadequately addressed by conjunctures and hypotheses: How complex was this pre-colonial Singapore society? In doing so, this thesis will first address the historiographical issues underlying this question by reviewing existing literature and primary sources concerning this pre-colonial entity. Deductions from the analysis of material cultural remains sampled from the archaeological site at the churchyard of the Saint Andrew’s Cathedral (STA) will next be made. Finally, statistically-measured variations in their site distribution will reflect the nature of the settlement’s spatial organization from which the degree of social complexity in pre-colonial Singapore can eventually be discerned. iv ABBREVIATIONS AA American Antiquity, Journal AAE Artibus Asiae, Journal AP Asian Perspectives, Journal BMJ Brunei Museum Journal CA Current Anthropology, Journal CC Colombo Court, Archaeological Site cmbd centimetres below datum CODA The Concise Oxford Dictionary of Archaeology, Publication EDCC Economic Development and Cultural Change, Journal EFEO École Française d'Extrême-Orient EMP Empress Place, Archaeological Site FTC Fort Canning Hill, Archaeological Site IILJ Institute of International Law and Justice IPPA Indo-Pacific Prehistory Association ISEAS Institute of Southeast Asian Studies, Singapore v JAA Journal of Anthropological Archaeology JAMT Journal of Archaeological Method and Theory JASB Journal of the Asiatic Society of Bengal JFA Journal of Field Archaeology JIAEA Journal of the Indian Archipelago and Eastern Asia JMBRAS Journal of the Malayan Branch of the Royal Asiatic Society; later renamed as the Journal of the Malaysian Branch of the Royal Asiatic Society (also JMBRAS) in 1964 1 JSBRAS Journal of the Straits Branch of the Royal Asiatic Society; later renamed as the Journal of the Malayan Branch of the Royal Asiatic Society in 1923 2 JSEAS Journal of Southeast Asian Studies LAA Latin American Antiquity, Journal MBRAS The Malaysian Branch of the Royal Asiatic Society MS Manuscript NAG National Art Gallery, Archaeological Site NMTN Nusantao Maritime Trading Network NUS National University of Singapore OPH Old Parliament House, Archaeological Site 1 “Annual Report of the Malayan Branch Royal Asiatic Society for 1963,” in Journal of the Malayan Branch of the Royal Asiatic Society (JMBRAS) 37,1 (July 1964), p. 171-2. 2 “Proceedings of the Annual General Meeting,” in Journal of the Straits Branch of the Royal Asiatic Society (JSBRAS) 85 (1922), p. iv. vi PHC Parliament House Complex, Archaeological Site RIMA Review of Indonesian and Malayan Affairs, Journal SEAMO Southeast Asian Ministers of Education Organization SCC Singapore Cricket Club, Archaeological Site SM Sejarah Melayu, Publication SO Suma Oriental, Publication SOAS School of Oriental and African Studies, University of London SPAFA SEAMO Project in Archaeology and Fine Arts SMJ The Sarawak Museum Journal STA Saint Andrew’s Cathedral Churchyard, Archaeological Site TOCS Transactions of the Oriental Ceramic Society, Journal TP T’oung Pao, Journal VCH Victoria Concert Hall, Archaeological Site WA World Archaeology, Journal vii LIST OF TABLES I. Quantitative Distribution of Sample STA Artefacts …………………………..65 II. Quantitative Distribution of Earthenware Vessels in Sample STA Artefacts based on Rim Sherds ……………………………………………………………..66 III. Quantitative Distribution of Stoneware Vessels in Sample STA Artefacts based on Rim Sherds ……………………………………………………………………...67 IV. Quantitative Distribution of Porcelain vessels in Sample STA Artefacts based on Rim Sherds ……………………………………………………………………...68 V. Quantitative Distribution of Metals in Sample STA Artefacts ………………….69 VI. Totals, Percentages, and the Number of Pieces Per Gram of Classified Ceramic Sherds ……………………………………………………………………………....70 VII. Totals, Percentages, and the Number of Pieces Per Gram of Classified Ceramic Sherds by Area ……………………………………………………………..70 VIII. Totals, Percentages, and the Number of Pieces Per Gram of Classified Ceramic Rim Sherds by Area ……………………………………………………………..71 IX. Chi-Squared Test of Significance Between Ceramic Variety and Spatial Distribution …………………………………………………………………….102 X. Chi-Squared Value of Calculation …………………………………………..103 viii LIST OF FIGURES 1. Map indicating existing archaeological sites, landmarks and known boundaries of pre-colonial Singapore ……………………………………………………...2 2. Three fragments of the Singapore Stone 3. Map of reconstructed 14th century settlement in Singapore superimposed on 1836 map of Singapore town drawn by J. B. Tassin ………..............................25 4. Abstract model for Exchange between a Drainage Basin Centre and an Overseas Power ………………………………………………………………………………29 5. The principal [orthogenetic] urban hierarchies in Southeast Asia in the second half of the 14th century ……………………………………………………………..33 6. Kerimun Islands to Pedra Branca 7. STA Site Map (2003-4) indicating the thesis’s division of the site into four quadrants for comparative analysis ....................................................................57 8. Coarse-tempered earthenware brick or tile sherd, STA 5 9. Coarse-tempered earthenware ‘trivet’ or structural ornamentation sherds, STA ………………………………………………………………………………72 10. Coarse-tempered ‘Saw-Tooth with Flame’ impressed earthenware pottery body sherds, STA 13A ……………………………………………………………..73 11. Medium-tempered earthenware pottery rim sherds, STA 9 12. Medium-tempered earthenware pottery knobbed lid sherds, STA 50 ……………………………………………………………………………………….74 13. Medium-tempered decorated (paddle-impressed) earthenware pottery body sherds, STA 9 ...……………………………………………………………………74 14. Medium-tempered slipped earthenware pottery rim and body sherds, STA 16 ……………………………………………………………………………………….75 …………………………………...11 ……………………………………………45 ix …...……………..72 ………………….73 15. High-fired black-burnished earthenware pottery rim sherds, STA 59 ………………..……………………………………………………………………...75 16. High-fired black-burnished carved earthenware pottery body sherds with ‘Circleand-Ring’ motif, STA 5 ……………………………………………76 17. High-fired mercury-jar type earthenware base sherd, STA 5 ……………….....76 18. Fine-paste earthenware pottery rim sherds, STA 58 …………………………..77 19. Fine-paste earthenware pottery rim sherds with ‘apple-green’ glaze, STA 55 ……………………………………………………………………………………….77 20. Fine-paste decorated (Punctated ‘Dot-and-Ring’ motif bordered with double rib lines) earthenware pottery body sherds, STA 7 …………………………..78 21. Fine-paste earthenware kendi spout sherds, one with ‘rifled’ interior, STA 58 ……………………………………………………………………………………….78 22. Type A, B and C stoneware pottery inclusion patterns, STA 5………………….79 23. Buff versus brittle stoneware mercury-jar rim sherds, STA 5 ……………...…..79 24. Buff versus brittle stoneware mercury-jar base sherds, STA 51 25. Brittle versus buff stoneware pottery rim sherds with luted lugs, STA 5 .....................................................................................................................................80 26. Brittle versus buff stoneware pottery body sherds, STA 5 27. Buff stoneware basin rim sherds, STA 50 28. Buff stoneware basin rim sherd with folded ‘Pie-Crust’ fringe, STA 6 .....................................................................................................................................82 29. Brittle stoneware pottery body sherds with glaze, STA 7 30. Brittle stoneware pottery base sherds, STA 5 …………………………………...83 31. Buff stoneware pottery base sherds, STA 50 …………………………………...83 32. Buff purple-ware pottery rim sherd, STA 5 33. Brittle purple-ware pottery rim sherd with luted lug, STA 7 ………………….84 34. Assorted green-ware porcelain pottery rim sherds, one with an exterior lotus leaf relief, STA 5 ……………………………...…………..………………….85 35. Green-ware porcelain bowl sherd with a foliated rim and interior motif, STA 7 ……………………………….………………………………………...…………….85 36. Green-ware porcelain bowl base sherd, STA 51 37. Assorted green-ware porcelain pottery base sherds, STA 50 ………………….86 x ………....80 .............................81 …...………………………………81 ………………….82 ………………………………...…84 …………………………..86 38. Assorted green-ware porcelain pottery body sherds, one with ‘stylized onion-skin medallion,’ STA 7 …………………………………………………….87 39. Green-ware porcelain ‘gacuk’, STA 76 40. White-ware porcelain cup sherd with foliated rim, STA 5 ……………….....88 41. White-ware porcelain covered box base sherd, STA 26 ………………….88 42. Iron-spotted white-ware porcelain pottery body sherds, STA 9 43. Blue-and-white porcelain cup rim sherds with exterior chrysanthemum and foliage motifs and interior classic scrolls and key-fret motifs, STA 50 ……………………………………………………………………………………….89 44. Blue-and-white porcelain bowl rim sherd with an exterior foliage motif and interior classic scrolls motif, STA 9 ……...…………………………………….90 45. Blue-and-white porcelain bowl rim sherds with an exterior stylised lotus petals panelled motif and interior chrysanthemum flower with foliage motif, STA 16. ……………………………………………………………………………………….90 46. Octagonal blue-and-white vase neck or body sherds with an exterior panelled classic scrolls motif and luted interior, STA 7 …………………………………...91 47. Blue-and-white porcelain cup rim and base sherds with an exterior foliated motif, interior classic scrolls motif and calligraphic illustration of ‘longevity’ on the interior centre medallion, STA 50 ………………………………………........91 48. Misfired or degraded blue-and-white porcelain pottery body sherds, STA 59 ……………………………………………………………………………………….92 49. Misfired or degraded porcelain pottery base sherd, STA 18A 50. Metal slag, STA 7 and 26 51. Metal nails, fishing hooks and other objects, STA 26 …………………………..93 52. Bronze bangle fragment, STA 18A 53. ‘明道元宝’ (Míng Dào Yuán Băo), Northern Song dynasty coin, 1032-3 CE, STA 42 ……………………………………………………………………...94 54. Either ‘崇宁通宝’ (Chóng Níng Tōng Băo), ‘崇宁元宝’ (Chóng Níng Yuán Băo), or ‘崇宁重宝’ (Chóng Níng Zhòng Băo), Northern Song dynasty coin, 1102-1106 CE, STA 50 …………………………………………………….95 55. Sri Lankan coin, circa 13th-14th centuries CE, STA 16 56. Crucible-like stone with a rectangular depression on one side, STA 58 ……………………………………………………………………………………….96 …………………………………...87 …………89 …………92 …………………………………………………….93 ……………………………………………94 xi ………………….95 57. Stone structural ornamentation fragments with ornate carved motif on surface, STA 13A ……………………………………………………………………...96 58. Sand Grain Sizing Folder 59. Charts for Estimating Proportions of Mottles and Coarse Fragments…………97 60. Mohs’ Hardness Scale and Substitutes 61. Approximate age of STA Site derived from a selection of sampled artefacts ……………………………………………………………………………………….98 …………………………………………………….97 xii …………………………………...98 I. INTRODUCTION To write a history of the old Singapura would be something like the task imposed upon the children of Israel by Pharaoh: for where should one seek the straw to make those bricks with? Charles Otto Blagden, 1921. 3 Over the last 26 years, large amounts of material cultural remains or ‘artefacts’ approximately dating to the 14th century CE have been recovered from 10 archaeological sites on Singapore Island. 4 These sites are all located within an area bound by Fort Canning Hill (FTC) to the north, the Singapore River to the west Stamford Road to the east and the Straits of Singapore to the south. 5 Coincidentally, Malay oral tradition and written records, most notably the Malay Annals (Sejarah Melayu (SM) or Sulalat us-Salatin), claim that a polity known as ‘Singapura’ was established on the island of ‘Temasek’ by a semi-divine prince around the same region and period. 6 From as early as 1819, these artefacts have been associated by historians and scholars alike with a hypothetical settlement and polity derived from inferences drawn from the SM about ‘Temasek-Singapura.’ It is the present consensus that these artefacts belong to a complex port-city which once existed on Singapore Island in precolonial times. Despite almost two centuries of historical research however, very few of the political, cultural and socio-economic aspects of this settlement, its relationship with neighbouring polities and its significance within Southeast Asian pre-colonial history can be confirmed by verifiable historical evidence. Scholarship on pre-colonial Singapore history has 3 Charles Otto Blagden, “Singapore prior to 1819,” in One Hundred Years of Singapore. Volume 1, ed. Walter Makepeace, Gilbert E. Brooke & Roland St. John Braddell (Singapore: Oxford University Press, 1991), p. 1. 4 Fort Canning Hill (FTC), Parliament House Complex (PHC), Old Parliament House (OPH), Empress Place (EMP), Colombo Court (CC), Singapore Cricket Club (SCC), Saint Andrew’s Cathedral (STA), National Art Gallery (NAG), Victoria Concert Hall (VCH) and the Padang. 5 The settlement was also surrounded by salt marshes to the north-east and north-west beyond FTC and the old wall until 1822; see John Crawfurd, Journal of an Embassy from the Governor General of India to the Courts of Siam and Cochin China. Volume I (London: Henry Colburn and Richard Bentley, 1830), p. 68-70. 6 C. C. Brown, Malay Annals (Selangor Darul Ehsan: MBRAS, 2009), p. 30-1. 1 N C C 1 B 7 C 10 8 5 2 3 6 9 4 A Pre-colonial Archaeological Sites: 1. Fort Canning Hill 2. Parliament House Complex 3. Old Parliament House 4. Empress Place 5. Colombo Court 6. Singapore Cricket Club 7. Saint Andrew’s Cathedral 8. National Art Gallery 9. Victoria Concert Hall 10. Padang Pre-colonial Landmarks: A. Singapore Stone B. Old Wall and Moat C. Terraces and Brick Foundations FIGURE 1. Map indicating existing archaeological sites, landmarks and known boundaries of pre-colonial Singapore. Source: Google Earth, Imagery Date: 4 June 2011, altitude 7668ft. 2 only recently started to advance from the results of early- to mid-20th century colonial research, while archaeology as an academic field in Singapore remains in its infancy despite its promising inception in the 1980s. HYPOTHESIS This thesis aims to fill this lacuna of knowledge by determining the level of social organization within this pre-colonial settlement through the historical and archaeological analysis of a sample of pre-colonial artefacts recovered from excavations at the Saint Andrew’s Cathedral Churchyard (STA) between 2004 and 2005. Specifically, the present hypothesis assumes that a correlation exists between artefacts and social status in a socially-complex precolonial Singapore polity. The basic assumption on which this hypothesis rests is the general observation that elite members of societies have the desire and means to purchase and consume significantly more high-value goods than their non-elite counterparts as a result of ‘unequal social exchanges.’ 7 Similar research has also demonstrated that greater socio-economic and political complexity in pre-colonial Southeast Asian societies is typically accompanied by the development of a ‘political economy’ where certain goods function as a ‘material fund of power’ which is used to ‘reinforce social inequality and political authority, with hereditary ‘elites’ controlling the specialized production and exchange of such goods. 8 Within the context of the STA archaeological site, this would manifest itself in a distinct and non-random pattern within the spatial distribution of the artefact sample pool selected for this research. Therefore, certain areas within the STA site which functioned as the former elite residences should have a higher proportion of such goods – which are usually of a higher economic cost – than non-elite residential zones. A random distribution pattern of artefacts within the site will therefore illustrate the null hypothesis (H0) instead, where there is no correlation between social status 7 Charles Maisels, The Archaeology of Politics and Power. Where, When and Why the First States Formed (Oxford: Oxbrow Books, 2010), p. 343-7. 8 Laura Lee Junker, “The Development of Centralized Craft Production Systems in A.D. 500-1600 Philippine Chiefdoms,” Journal of Southeast Asian Studies (JSEAS) 25,1 (March 1994), p. 3. 3 and material culture and hence, suggest that the settlement did not have a complex, stratified society. This hypothesis predominantly relies on the abundance of ceramics or pottery sherds recovered from the STA site, as this class of artefacts accounts for more than 90 percent of the total amount of artefacts sampled by weight and count. Significantly, not only do ceramics encode social information concerning important social distinctions within a society, they are also useful in the illustration of intra- and inter-site social variation. 9 Since the study of social organization in complex societies is largely the study of social ranking, the disparity between the rich and poor should be visible in terms of ownership, access to resources, facilities and status. 10 In the absence of grandiose building and structural remains which usually define this disparity in other archaeological sites, the study of archaeological ceramics, their distribution patterns on the site as well as their relationship with social organization and structure rises to the fore. PRIMARY SOURCES Prior to the availability of archaeological sources however, the reconstruction of precolonial Singapore in history was indeed – as Blagden correctly observed – fraught with difficulty. To quote historical geographer Paul Wheatley, it is ‘doubly unfortunate that there should be a complete lack of indigenous written sources [on the Malay Peninsula] before the sixteenth century.’ 11 Hence, if the history of this island before the arrival of the British is the brick historians seek to mould, there are scarcely any reliable primary textual sources which scholars can use as straw. No indigenous records are known to have survived until the compilation of the SM (Raffles Manuscript (MS) No. 18) in the 17th century and its recensions by various Malay chroniclers over the next few centuries. 12 Significantly, no other source 9 Carla M. Sinopoli, Approaches to Archaeological Ceramics (New York: Plenum Press, 1991), p. 119-42. Colin Renfrew & Paul Bahn, Archaeology. Theories, Methods and Practice. Fourth Edition (London: Thames & Hudson, 2004), p. 213. 11 Paul Wheatley, The Golden Khersonese. Studies in the Historical Geography of the Malay Peninsula Before A.D. 1500 (Kuala Lumpur: University of Malaya Press, 1961), p. v. 12 Richard Olaf Winstedt, “A History of Malay Literature,” JMBRAS 17,3 (January 1940), p. 106-9. 10 4 besides the SM explains the relationship between the toponyms of ‘Temasek’ and ‘Singapura,’ where it clearly states that the first king of the settlement, ‘Sang Utama’ or ‘Sri Tri Buana,’ ‘established a city at Temasek, giving it the name of Singapura’ sometime in 1299. 13 Nonetheless, the SM’s narrative does not reveal much about the polity’s social organization; as a genealogy of the later Melakan and Johor Sultans, the SM elaborates instead on the founding myth of Temasek-Singapura, the court intrigue between its five rulers and their respective courtiers during its supposedly brief 91-year lifespan, and the polity’s demise under a Majapahit invasion at the end of the 13th century. 14 A number of contemporaneous Chinese accounts are claimed by scholars to have depicted Temasek-Singapura as well. The first comes in the form of a list of foreign lands known to the Chinese between 1230 to 1349 entitled Description of the Barbarians of the Isles (島夷梽畧 Dăo Yí Zhì Lüe), compiled by Wāng Dà Yüan (汪大淵) in the mid-14th century, under two Chinese toponyms, ‘Lóng Yá Mén’ (龍牙門 or ‘Dragon’s Teeth Gate’) and ‘Bān Zú’ (班卒), as well as the name of their residents, the ‘Dān Mǎ Xī’ (單馬錫) barbarians. 15 Wāng added under the toponym of ‘Xiān’ ( 暹 or ‘Siam’) that the ‘Siamese’ had launched an amphibious assault on ‘Dān Mǎ Xī’ in the mid-14th century. The attack grew into a month-long siege of the city wall and moat (城池 Chéng Chí) which was only dispersed with the sighting of a passing Javanese envoy. 16 According to the History of the Yuan Dynasty (元史 Yuán Shǐ), Mongol imperial envoys were even sent to ‘Lóng Yá Mén’ asking for ‘tamed elephants.’ 17 Presumably in response to this request, ‘Lóng Yá Mén’ dispatched a mission to China with a memorial and tribute. 18 While the accounts provide a detailed insight to the settlements and 13 W. Linehan, “The Kings of 14th Century Singapore,” JMBRAS 20,2 (December 1947), p. 60 Brown, Malay Annals p. 31, 23-52; Linehan, “The Kings of 14th Century Singapore,” p. 60, 117-27. 15 Wheatley, The Golden Khersonese, p. 82-6. 16 苏继倾 (Sū Jì Qīng), 《島夷梽畧校释》 (Dăo Yí Zhì Lüe Jiào Shì or ‘An explanation of the Description of the Barbarians of the Isles’) (北京: 中华书局, 1981), p. 154-5. 17 Hsü Yün-Ts’iao, “Singapore in the Remote Past,” JMBRAS 45,1 (January 1973), p. 3-4. 18 Oliver William Wolters, The fall of Śrīvijaya in Malay History (London: Lund Humphries Publishers Limited, 1970), p. 78. 14 5 inhabitants listed under these toponyms, the accuracy of ascribing these toponyms to precolonial Singapore however remains disputed to this day. 19 Various 14th century Southeast Asian texts also appear to have mentioned the polity in passing. A Javanese chronicle, the Pararaton, lists ‘Tumasik’ as one of the lands proposed to be conquered by Gajah Mada, the famed Prime Minister of the Majapahit kingdom centred in east Java; 20 ‘Tumasik’ is next declared in another Javanese text of the same period, the Deśawarṇana (or Nāgarakṛtāgama), as one of many vassals that ‘seek refuge’ and placed ‘under the protection’ of the Majapahit ruler by 1365. 21 It has also been advanced that the toponym ‘Sach-ma-tich’ (册馬錫 Sách Mã Tích), found in the Vietnamese Annals of 1330, was a transliteration of ‘Temasek.’ According to the Annals, a Vietnamese Prince, Tran Nhat Duat, could have served as an interpreter for the envoys from that land. 22 The earliest European sources on pre-colonial Singapore are Tomé Pires’ Suma Oriental (SO) which was written between 1512 and 1515, and the writings of Afonso d’Albuquerque, the Portuguese conqueror of Melaka in 1511, compiled by his son Braz d’Albuquerque as Commentarios do grande Afonso Dalboquerque in 1557. However, Pires and the d’Albuquerques gave a different account of ‘simgapura’ from that of the SM. According to them, the settlement’s last ruler, Parameswara, fled Palembang from the Javanese, murdered and usurped Singapore Island and its strait from its ruler, only to flee again from the wrath of both the rulers of Siam and Patani. Pires referred to simgapura primarily as a strait (‘sam agy simgapura que era Rey daquelle canal’ or Sam Agy Simgapura who was king of that channel) with islands and towns. This is repeated by the d’Albuquerques, who added that the settlement 19 Lin Wo Ling, Long-ya-men Re-identified (Singapore: 新加坡南洋学会出版, 1999), p. 104-10, 123; Paul Michel Munoz, Early Kingdoms of the Indonesian Archipelago and the Malay Peninsula (Singapore: Editions Didier Millet, 2006), p. 188. 20 I Gusti Putu Phalgunadi, The Pararaton. A Study of the Southeast Asian Chronicle (New Dehli: Sundeep Prakashan, 1996), p. 125. 21 Stuart Robson, Deśawarṇana (Nāgarakṛtāgama) by Mpu Prapañca (Leiden: KITLV Press, 1995), p. 33-4; the present thesis follows Robson’s orthography of this old Javanese poem. 22 Wolters, History, Culture and Region in Southeast Asian Perspectives (Singapore: Institute of Southeast Asian Studies (ISEAS), 1982), p. 48, cited in John Norman Miksic, Archaeological Research on the “Forbidden Hill” of Singapore: Excavations at Fort Canning, 1984 (Singapore: National Museum Singapore, 1985), p. 17. 6 ‘was a very large and very populous city – as is witnessed by its great ruins which still appear to this very day [in the early 16th century].’ 23 Mention of Temasek-Singapura’s existence, decline and fall were found in subsequent Arabic, Chinese and European maps, texts and other maritime records of the Melakan Strait in the later centuries, but they were either repeating the narratives found in the SM, Description, SO and Comentarios, or extremely scant in detail. Not surprisingly, very little is known about the pre-colonial settlement itself. This paucity of detailed and verifiable information, however, did not diminish efforts to reconstruct the island’s pre-colonial past of which three historiographical phases can be distinguished. LITERATURE REVIEW The first historiographical phase is characterised by the discovery and historicising of Temasek-Singapura by early colonial scholarship between the 18th and early-20th centuries. Knowledge of the SM, as well as another Malay text, the Taju assalatin or Makuta segala rajaraja (‘The Descent of all [Malayan] Kings’), were already communicated to the West by Dutchmen Petrus van der Worm, François Valentijn, George Hendrik Werndly, as well as Englishman John Leyden between the late-17th to early-18th centuries. 24 But it was not till the late-18th century that the SM’s narrative of Temasek-Singapura was first historicized by William Marsden within three editions of his book, The History of Sumatra, published between 1774 and 1811. Marsden was regarded by early colonial scholars to be ‘the first literary and scientific Englishman, with the advantage of local knowledge, who wrote about the Malay countries, with laborious care and scrupulous fidelity.’ 25 This is reflected most in his condensation of the first six chapters of the SM, covering the descent of the Malays from Raja 23 Armando Cortesão, The Suma Oriental of Tomé Pires. An Account of the East, From the Red Sea to Japan, written in Malacca and India in 1512-1515 and The Book of Francisco Rodrigues. Rutter of a Voyage in the Red Sea, Nautical Rules, Almanack and Maps, written and Drawn in the East before 1515. Volume II (London: Hakluyt Society, 1944), p. 229-33; Walter De Gray Birch, The Commentaries of the Great Afonso Dalboquerque. Second Viceroy of India. Volume III (New York: Burt Franklin, 1970), p. 71-4. 24 William Marsden, The History of Sumatra (Singapore: Oxford University Press, 1986), p. 326; Winstedt “Malay Works known by Werndly in 1736 A.D.,” JSBRAS 82 (September 1920), p. 163-5. 25 Charles Burton Buckley, An Anecdotal History of Old Times In Singapore (Singapore: Oxford University Press, 1984), p. 2. 7 Iskandar the ‘Two-Horned’ (Alexander the Great) to the fall of Temasek-Singapura, into two brief paragraphs so as to omit ‘mythological fable’ in order to retrieve ‘historical fact.’ 26 The same was done in Thomas Stamford Raffles’ The History of Java in 1816, where he outlined within a lengthy footnote the genealogical descent of the ‘Malayus’ from Alexander to the founding of the ‘city of Singa pura’ sometime in 1160. 27 John Crawfurd took the study of Temasek-Singapura a step further when his book, History of the Indian Archipelago, was published in 1820. Like Marsden before him, Crawfurd believed that the ‘parent country’ of the Malays lay in ‘the country of Menangkabao in Sumatra.’ He suggested that the establishment of ‘Singahpura’ on the Malay Peninsula was a result of a wave of migration of the ‘Malayu tribe,’ a sub-division of the people of Menangkabao, in a time when ‘a rapid and unusual start in civilization and population’ lead to a scarcity of land on the ‘great and fertile plain’ in Sumatra. Crawfurd however was unwilling ‘to attempt any narrative of their affairs’ due to the absence of a ‘full and connecting narrative of the history of any of the Malay states,’ as well as some ‘suspicious circumstances in the detail of events.’ Nonetheless, he considered Temasek-Singapura as a historical primary colony from which the Malays later dispersed to colonize the rest of the Malay Peninsula, the Riau-Lingga Islands, Borneo and eventually returning to Sumatra and established Kampar and Aru. 28 While Marsden, Raffles and Crawfurd generally agreed that Temasek-Singapura was located somewhere in the southern extremity of the Malay Peninsula, it was Raffles who advocated Singapore Island as its precise site. On 12 December 1818, Raffles wrote to Marsden revealing his intent to appropriate for the East India Company the ‘site of the ancient city of Singapura.’ 29 In a letter addressed to Colonel Addenbrooke on 10 June 1819, Raffles wrote that he would not have known about the existence of Singapore if not for his ‘Malay Studies.’30 Some scholars have dismissed this claim as ludicrous since the island’s toponym was found in 26 Marsden, The History of Sumatra, p. 327. Thomas Stamford Raffles, The History of Java (Singapore: Oxford University Press, 1988), p. 108-10. 28 Crawfurd, History of the Indian Archipelago. Volume II. (Edinburgh: Archibald Constable and Co., 1820), p. 371-8. 29 Blagden, “The Foundation of the Settlement,” in One Hundred Years of Singapore. Volume 1, p. 7. 30 Raffles, “The Founding of Singapore,” JSBRAS 2 (December 1878), p. 175, 178. 27 8 almost every contemporary map of the region in Raffles’ time. 31 They have in fact misunderstood Raffles, as he referred not just to the existence of the island in the 19th century, but also to its association with the narrative according to the centuries-old Malay oral and written tradition canonised in the SM. In his instructions to William Farquhar who was appointed as the first Resident of Singapore on 6 February 1819, Raffles reiterated that ‘the latter [Malay rulers] will hail with satisfaction the foundation and the site of a British establishment, in the centrical and commanding situation once occupied by the capital of the most powerful Malayan empire then existing in the East.’ 32 Just one week after acquiring a lease on the south coast of the island for the British East India Company, Raffles wrote another letter dated 13 February 1819 to the Supreme Government in Calcutta that, ‘a British Station, commanding the Southern entrance of the Straits of Melaka, and combining extraordinary local advantages with a peculiarly admirable Geographical position, has been established at Singapore the ancient Capital of the Kings of Johor.’ 33 Four months later, Raffles wrote in the letter to Addenbrooke that he had ‘planted the British Flag’ on the ruins of the ancient Capital of ‘Singapura,’ or ‘City of the Lion’ at the ‘Naval of the Malay countries.’ 34 In another letter to Marsden dated 21 January 1823, Raffles announced the construction of a bungalow on FTC where ‘the tombs of the Malay Kings are, however, close at hand.’ 35 Through these letters written in both official and private capacities, Raffles had implanted the idea of TemasekSingapura as a lost Malay capital and kingdom on Singapore Island firmly in colonial scholarship. The credibility of Raffles’ claim was bolstered by other accounts of physical and material cultural remains of a pre-colonial settlement in the early years of the British settlement on Singapore Island. According to the Hikayat Abdullah, the autobiography of Raffles’ 31 Eunice Thio, “Introduction,” in Singapore. 150 years, ed. Tan Sri Dato’ Mubin Sheppard (Singapore: Times Books International), p. iii. 32 Thomas Braddell, “Notices of Singapore,” Journal of the Indian Archipelago and Eastern Asia (JIAEA) 7 (1853), p. 326. 33 John Bastin, Letters and Books of Sir Stamford Raffles and Lady Raffles (Singapore: Editions Didier Millet Pte Ltd, 2009), p. 258. 34 Raffles, “The Founding of Singapore,” p. 178. 35 Sophia Raffles, Memoir of the Life and Public Services of Sir Thomas Stamford Raffles (Singapore: Oxford University Press, 1991), p. 535. 9 associate Abdullah bin Abdul Kadir, the Temenggong once told Farquhar that FTC was known as Bukit Larangan or ‘Forbidden Hill,’ the site of ‘palaces’ and a royal bath (Pancur Larangan or Forbidden Spring) belonging to the ‘kings of ancient times.’ Abdullah also noted that fruit trees of ‘great age’ were found when the land around the hill was cleared. 36 Crawfurd, who visited the island in 1822, also saw the fruit trees mentioned by Abdullah, suggesting that they were ‘cultivated by the ancient inhabitants of Singapore’ and found it remarkable that they still existed ‘after a supposed lapse of near six hundred years.’ Crawfurd also observed that the northern and western sides of FTC were ‘covered with the remains of the foundations of buildings, some composed of baked bricks of good quality.’ Near the summit of the hill was a square terrace of approximately 40 feet to a side which held ‘fourteen large blocks of sandstone… with a hole in each,’ thought by Crawfurd to be pedestals of wooden posts that supported a structure, as well as a ‘circular inclosure, formed of rough sandstone’ which he speculated once contained a Buddhist image and thus, conjectured that the hill was a site of ancient Buddhist monasteries. Crawfurd also encountered pottery shards found ‘in great abundance’ as well as Chinese brass coins minted between the ‘tenth and eleventh centuries’ among the ruins. Another terrace was communicated to Crawfurd as the ‘burying-place (or Keramat) of Iskandar Shah, King of Singapore.’ 37 Besides the ruins on the hill, Raffles was able to trace in 1819 the remains of the ‘lines of the old city, and of its defences’ within which the new British station was to be founded; 38 the old rampart was mapped and published as the ‘Old Walls of the City’ by Captain Daniel Ross in the same year. 39 Crawfurd added that it was a wall approximately ‘sixteen feet’ wide and ‘eight or nine feet’ high which ‘runs very near a mile from the sea-coast to the base of the hill, until it meets a salt marsh.’ A little rivulet also ran along the northern face of the wall which formed ‘a kind of moat’ up to the elevated side of 36 A. H. Hill, Hikayat Abdullah (Kuala Lumpur: MBRAS, 2009), p. 142, 168. Crawfurd, Journal of an Embassy, p. 68-73. Raffles, Memoir, p. 376. 39 Daniel Ross, “Plan of Singapore Harbour February 1819,” illustrated in Bastin, Letters and Books, p. 254. 37 38 10 FIGURE 2. Three fragments of the Singapore Stone. Source: J. W. Laidlay, Journal of the Asiatic Society of Bengal (JASB) 17,2 (1848), pl. 3, opp. p. 68, illustrated in Miksic, Archaeological Research, p. 41. 11 the hill where it became a dry ditch. 40 These observations led Crawfurd to conclude that the settlement was located at the foot of FTC and ‘bounded to the east by the sea, to the north by a wall, and to the west by a salt creek or inlet of the sea’ which he noted to be like ‘a kind of triangle.’ 41 The moat, wall and ruins on the hill were not the only pre-colonial remains which supported the historicity of pre-colonial Singapore as Temasek-Singapura. As the jungles were cleared and knolls levelled to fill the swamps for the development of the new British factory, Abdullah recorded that a large, inscribed rock was found at the mouth of the Singapore River. 42 Presently known as the ‘Singapore Stone,’ Crawfurd described it as ‘a rude mass’ of stone which was split into ‘two nearly equal parts by artificial means.’ Crawfurd observed also that the inscriptions had characters similar to ‘Pali, or religious character used by the followers of Buddha,’ which he noted to be found abundantly in Java and Sumatra. 43 Originally a sandstone boulder approximately three metres tall and three metres wide, the inscription had covered the inner surface of one-half of the boulder. 50 or 52 lines of script were inscribed on an area 2.1 metres wide and 1.5 metres high, with approximately 40 lines that were discernible while the first 12 lines at the beginning had been effaced. 44 The inscription was however badly weathered either by the rain or the action of the tides which made it difficult to read and hence hard to decipher. In an unfortunate series of events, the stone was blown up and chiselled away by colonial engineers and labourers in 1843. Drawings of three inscription fragments were made before they were sent to Calcutta for analysis, but only one found its way back to the National Museum in Singapore while the other two have yet to be found. 45 Scholars J. W. Laidlay and Hans Kern later identified the characters to be Kawi, a mainstream script used in East Java by the 13th century. 46 On this basis, four disparate Sanskrit 40 Crawfurd, Journal of an Embassy. Volume I, p. 68-73. Ibid., p. 68-70. 42 Hill, Hikayat Abdullah, p. 167. 43 Crawfurd, Journal of an Embassy. Volume I, p. 70 44 Miksic, Archaeological Research, p. 40. 45 Carl Alexander Gibson-Hill, “Singapore Old Strait & New Harbour (1300-1870),” in Memoirs of the Raffles Museum. No. 3, ed. Carl Alexander Gibson-Hill (Singapore: Government Printing Office, 1956), p. 24. 46 Johannes Gijsbertus de Casparis, Indonesian Paleography: A history of writing in Indonesia from the beginnings to c. A.D. 1500 (Leiden: E. J. Brill, 1975), p. 5. 41 12 words of ‘salâgalalasayanara,’ ‘ya-âmânavana,’ ‘kesarabharala’ and ‘yadalama’ were deciphered by Kern, but their meanings remain unknown. 47 Supported predominantly by the SM and pre-colonial material cultural remains, early colonial scholarship created the historical polity of Temasek-Singapura and established it on Singapore Island. To quote John Turnbull Thomson in 1849, ‘the site of the present town of Singapore… on the same spot we were led to believe from the perusal of Malayan history, was occupied by the ancient one…’ 48 Between 1847 and 1859, members of the British colonial service and public in Singapore such as Thomson, Edmund Augustus Blundell and Thomas Braddell published articles in the Journal of the Indian Archipelago and Eastern Asia (JIAEA, popularly known as ‘Logan’s Journal’) which continued to associate the island with TemasekSingapura. 49 Much of this work was later compiled within what was to be the first unofficial history of the British colony, Charles Burton Buckley’s An Anecdotal History of Old Times in Singapore in 1902, where the familiar narrative of Temasek as ‘ancient Singapore’ was faithfully repeated. 50 This trend of scholarship was continued in the Journal of the Straits Branch of the Royal Asiatic Society (JSBRAS), later renamed the Journal of the Malayan Branch of the Royal Asiatic Society (JMBRAS) and after Malaysia’s independence, the Journal of the Malaysian Branch of the Royal Asiatic Society (also JMBRAS). As the journals’ chief aim was to ‘investigate subjects connected with the Straits of Melaka and its neighbouring countries,’ various Chinese, European and Southeast Asian sources were increasingly cited in historical articles which affirmed the antiquity of the region. 51 This indirectly substantiated the historicity of the SM and consequently, TemasekSingapura, which were frequently cited in these articles; Blagden was able to declare that Temasek ‘is certainly Singapore;’ Richard James Wilkinson called Singapore ‘Palembang’s 47 Hans Kern, “Concerning some old Sanskrit Inscriptions in the Malay Peninsula,” JSBRAS 49 (December 1907), p. 100-1. John Turnbull Thomson, “General report on the Residency of Singapore, drawn up principally with a view of illustrating its Agricultural Statistics,” JIAEA 3 (1849), p. 618. 49 Edmund Augustus Blundell, “Notices of the History and Present Condition of Malacca,” JIAEA 2 (1848), p. 727; Thomas Braddell, “Abstract of the Sijarah Malayu or Malay Annals, with Notes,” JIAEA 5 (1851), p. 244. 50 Buckley, An Anecdotal History, p. ix, 18-25. 51 George Frederick Hose, M. A., “Inaugural Address by the President,” JSBRAS 1 (July 1878), p. 1. 48 13 daughter;’ Richard Olaf Winstedt even accused Dutchman van der Tuuk to have ‘robbed Singapore of its legendary founder’ by identifying Nila Utama (Sri Tri Buana) with Tillottama, a nymph of Indra’s heaven. 52 Blagden later enshrined the Temasek-Singapura narrative in the first chapter of One Hundred Years of Singapore, a historical volume celebrating the centenary of the island in 1921. 53 The discovery of the cache of pre-colonial gold ornaments at FTC in 1926 further reinforced Winstedt’s belief in the Javanese conquest of Temasek-Singapura in 1360, as armlets bearing decorations in the form of kala heads as well as a ring with a goose design suggested a Hindu-Javanese provenance from the Majapahit period sometime in the 14th century. 54 Wilkinson, Winstedt, Anker Rentse, W. Linehan and Mervyn Cecil Franck Sheppard subsequently grafted the narrative of Temasek-Singapura as well as other references found in the SM as the prologue to the history of almost every Malay state in British Malaya. 55 Notable Southeast Asian scholars of that generation such as Daniel George Edward Hall and George Coedès also unquestioningly accepted the assumption that Tumasik (in the Deśawarṇana) and Temasek (in the SM) was located on Singapore Island, whereas Paul Pelliot called Singapore (Tumasik or Tĕmasik) ‘Palembang’s colony.’ 56 Despite his contentions on the historicity of the SM, Oliver William Wolters nevertheless alluded that Temasek-Singapura was an actual polity, albeit as a less significant ‘trading and piratical kingdom’ than a prosperous cosmopolitan port of trade. 57 In keeping with this scholarly tradition, Constance Mary 52 D. F. A. Hervey, “Valentyn’s Description of Malacca,” JSBRAS 13 (June 1884), p. 62-6; E. Koek, “Portuguese History of Malacca,” JSBRAS 17 (June 1886), p. 117-8; Blagden, “Notes on Malay History,” JSBRAS 53 (September 1909), p. 143-62; William George Maxwell, “Barretto de Resende’s Account of Malacca,” JSBRAS 60 (December 1911), p. 18-9; Richard James Wilkinson, “The Malaccan Sultanate,” JSBRAS 61 (June 1912), p. 67; Winstedt, “The Founder of Old Singapore,” JSBRAS 82 (September 1920), p. 127; A. Caldecott, “The Malay Peninsula in the XVIIth & XVIIIth Centuries,” JSBRAS 82 (September 1920), p. 129-32; Winstedt, “The Early History of Singapore, Johore & Malacca; an outline of a paper by Gerrit Pieter Rouffaer,” JSBRAS 86 (November 1922), p. 258-60; Wilkinson, “ Early Indian Influence in Malaysia,” JMBRAS 13,2 (October 1935), p. 8-9, 15; Wilkinson, “Old Singapore,” JMBRAS 13,2 (October 1935), p. 17-21; Linehan, “The Kings of 14th Century Singapore,” p. 117-27. 53 Blagden, “Singapore prior to 1819,” p. 1-5. 54 Winstedt, “Gold Ornaments dug up at Fort Canning, Singapore,” JMBRAS 6,4 (November 1928), p. 1-4; Miksic, Archaeological Research, p. 42-3. 55 Wilkinson, A History of the Peninsular Malays with Chapters of Perak and Selangor (Singapore: Kelly & Walsh, Limited, 1920), p. 18-31; Winstedt, “A History of Johore (1365-1895),” JMBRAS 10,3 (December 1932), p. 6; Winstedt & Wilkinson, “A History of Perak,” JMBRAS 12,1 (December 1932), p. 6; Anker Rentse, “History of Kelantan. I.,” JMBRAS 12, 2 (August 1934), p. 44-8; Winstedt, “A History of Selangor,” JMBRAS 12,3 (October 1934), p. 1-2; Winstedt, “Negri Sembilan. The History, Polity and Beliefs of the Nine States,” JMBRAS 12,3 (October 1934), p. 42; Winstedt, “A History of Malaya,” JMBRAS 13,1 (March 1935), p. 31-44; Linehan, “History of Pahang,” JMBRAS 14,2 (May 1936), p. 5-11; Mervyn Cecil Franck Sheppard, “A Short History of Trengganu,” JMBRAS 22,3 (June 1949), p. 3-6. 56 Daniel George Edward Hall, A History of South-East Asia. Fourth Edition (Hampshire: Palgrave Macmillan, 1995), p. 224; Paul Pelliot, Notes on Marco Polo. Volume II (Paris: Imprimerie Nationale, 1963), p. 772, 839; George Coedès, The Indianized States of Southeast Asia (Kuala Lumpur: University of Malaya Press, 1968), p. 241. 57 Wolters, The fall of Śrīvijaya in Malay History, p. 11, 78-80. 14 Turnbull’s A History of Modern Singapore begins with the citation of various references to precolonial Singapore as well as the SM’s account of Temasek-Singapura. 58 Having supposedly established the historicity of Temasek-Singapura, the second historiographical phase in turn saw the elaboration of Temasek-Singapura as a port-polity within a wider context of maritime history in the longue durée. Scholars began to corroborate European and Southeast Asian texts – well studied by this time – with ancient Chinese and Arabic records in their study of pre-colonial Southeast Asia. A general method for ‘the reconstruction of the ancient pictures of Malaya’ was hence inaugurated by the identification of ‘a number of ancient toponyms, Greek, Indian, Chinese, Arabic and so forth,’ through a coinciding of ‘the geographical, historical, and etymological data.’ 59 Regardless, the SM’s narrative of Temasek-Singapura remained the dominant narrative in the study of Singapore’s pre-colonial history. Consequently, research shifted to the search for ‘Temasek-Singapura’ within historical references to the Malayan Peninsula as well as the Melakan and Singapore Straits. In his study of Ptolemy’s trans-Gangetic geography, Gerolamo Emilio Gerini cites the SM as evidence for locating ‘Tamasak, or Ujong Tanah of the Malays’ on Singapore Island, as well as phonetically associating it with the toponyms of ‘Bêtumah’ in Arabic, Dān Mă Xī (淡馬錫) in Chinese, ‘Tamus’ or ‘Tamarus’ in French, and Tumasik in the Deśawarṇana. 60 At the same time, Warren D. Barnes directed attention to the maritime use of the Singapore Strait(s) in pre-colonial times. Citing Chinese and European sources from 1436, Barnes argued that the ‘old Straits of Singapore are none other than the present Keppel Harbour,’ a waterway that passes between the south-western coast of Singapore and its islands Pulau Brani and Pulau Blakang Mati (also known as Sentosa Island). 61 Having translated and studied Chinese maritime records, William 58 Constance Mary Turnbull, A History of Modern Singapore. 1819-2005 (Singapore: NUS Press, 2009), p. 19-23. Roland Braddell, “Notes on Ancient Times in Malaya,” JMBRAS 20,1 (June 1947), p. 162; see also Wolters, Early Indonesian Commerce. A study of the origins of Śrīvijaya (Ithaca: Cornell University Press, 1967), p. 170-1. 60 Gerolamo Emilio Gerini, Researches on Ptolemy’s Geography of Eastern Asia (London: Royal Asiatic Society & Royal Geographical Society, 1909), p. 199-200, 533, 775, 821. 61 Warren D. Barnes, “Singapore old Straits and New Harbour,” JSBRAS 60 (December 1911), p. 25-31. 59 15 Woodville Rockhill believed that the Chinese toponym of ‘Dān Mă Xī’ referred to Temasek and therefore, its adjacent waterway – Lóng Yá Mén – must refer to the Singapore Strait. 62 In his study of studied various Chinese sailing directions across the centuries, John Vivian Gottlieb Mills concurred in the identification of Singapore Island and its strait with the toponyms of Dān Mă Xī and Dragon’s Teeth Gate respectively, but argued instead that the ‘old strait’ lay along the present Strait of Singapore, delineated by its southern shores and the northern coasts of Batam and Bintan Islands in Indonesia. 63 Like his father Thomas Braddell, Roland St. John Braddell advocated that ‘Tan-ma-hsi is a transliteration of Temasek and, beyond doubt, is Singapore.’ Braddell supported Barnes’ idea that the old Strait was along Keppel Harbour and identified by the toponym of Dragon’s Teeth Gate. 64 Reiterating the views of Barnes and Braddell, both Carl Alexander Gibson-Hill and Hsü Yün-Ts’iao staunchly supported the location of Temasek Singapore and the Dragon’s Teeth Gate at Singapore and Keppel Harbour respectively. 65 Citing the work of Barnes and Gibson-Hill, Wheatley supports the association of Chinese toponyms Dān Mă Xī and the Dragon’s Teeth Gate with Singapore and Keppel Harbour as well. In addition, Wheatley brought ‘Bān Zú,’ a locality was mentioned by Wāng to be situated behind the Dragon’s Teeth Gate, to the discussion, arguing that it ‘can only be the eminence which dominates Singapore City, namely Fort Canning Hill.’ 66 Significantly, Coedès’ ‘rediscovery’ of Srivijaya in 1918, believed at that time to be a great pre-colonial Southeast Asian maritime thalassocracy, provided a firm historical context for the interpretation of Temasek-Singapura as a polity built upon the foundation of maritime trade. 67 This led Blagden, who had worked with Coedès on the transliteration of Malay inscriptions of Srivijaya, to suggest that Temasek-Singapura and ‘a number of “Straits Settlements” ’ on the Malay Peninsula came under the influence of Srivijaya for several 62 William Woodville Rockhill, “Notes on the Relations and Trade of China with the Eastern Archipelago and the Coasts of the Indian Ocean during the Fourteenth Century. Part II,” T’oung Pao (TP), Second Series 16,1 (March 1915), p. 61, 129-32. 63 John Vivian Gottlieb Mills, “Malaya in the Wu-Pei-Chih Charts,” JMBRAS 15,3 (December 1937), p. 21-8; Mills, “Arab and Chinese Navigators in Malaysian Waters in about A.D. 1500,” JMBRAS 47,2 (December 1974), p. 25-32. 64 Roland Braddell, “Lung-Ya-Men and Tan-Ma-Hsi,” JMBRAS 23,1 (February 1950), p. 37-51. 65 Gibson-Hill, “Singapore: Notes on the History of the Old Strait, 1580-1850,” JMBRAS 27,1 (May 1954), p. 163-214; GibsonHill, “Singapore Old Strait & New Harbour (1300-1870),” p. 11-115; Hsü, “Singapore in the Remote Past,” p. 1-9. 66 Wheatley, The Golden Khersonese, p. 82-6. 67 George Coedès, “The Kingdom of Sriwijaya,” in Sriwijaya. History, Religion & Language of an Early Malay Polity, ed. PierreYves Manguin (Kuala Lumpur: MBRAS, 1992), p. 1-40. 16 centuries. 68 With the decline of Srivijaya by the end of the 13th century, Blagden argued that Singapore, with kings ‘descended from the royal family of Palembang,’ became independent and capitalized on its ‘unique position’ as a ‘short cut’ for trade between the East and West. 69 Wheatley proceeded to label the period between the 13th to 14th centuries on the Malay Peninsula as the ‘Century of Singhapura,’ which was preceded by the decline of Srivijaya and succeeded by the next ‘Century of Mĕlaka.’ 70 In recent years, Michel Jacq-Hergoualc’h argued that Temasek-Singapore was ‘favoured by the irreversible weakening of the political entity that was Śrīvijaya-Palembang-Jambi’ effected by Javanese incursions, and stated that the ‘rapid development of Temasek-Singapore is undeniable’ in the 14th century. 71 Paul Michel Munoz also lists Temasek-Singapura as a ‘post-Srivijayan emporium’ in the Malay Archipelago during the ‘late Classical period.’ 72 In his seminal study of the Malays, Anthony Crothers Milner assumed ‘Tumasek’ to be Singapore and the Siamese and Javanese attacks on it in the 14th century as historical facts. 73 While the archaeological record had clearly been the motivation behind much research on pre-colonial Singapore in the last two historiographical phases, it was merely considered as a physical indicator of the island’s antiquity which supported existence of Temasek-Singapura as a historical polity; the artefacts themselves were never analysed to elicit information about the settlement itself. In the words of Wong Lin Ken, the antiquity of Singapore ‘does not bear serious investigation.’ Nonetheless, he believed that the trade which came to Singapore in the 19th century ‘was not merely the result of the transfer of an existing trade into new channels,’ but that of Raffles putting into operation ‘the principal of the ancient Emporia,’ which he thought was the ‘basis of commerce in the Malay Archipelago.’ 74 As the old wall and ruins on FTC, the Singapore Stone and other traces of material cultural remains still visible in 68 Coedès, “The Malay Inscriptions of Sriwijaya,” in Sriwijaya, p. 43-5. Blagden, “The Empire of the Maharaja, King of the Mountains and Lord of the Isles,” JSBRAS 81 (March 1920), p. 25-8. 70 Wheatley, Impressions of the Malay Peninsula in Ancient Times (Singapore: Eastern Universities Press LTD, 1964), p. 101-118. 71 Michel Jacq-Hergoualc’h, The Malay Peninsula. Crossroads of the Maritime Silk Road (100 BC – 1300 AD) (Leiden: Brill, 2002), p. 491. 72 Munoz, Early Kingdoms, p. 185. 73 Anthony Crothers Milner, The Malays (Oxford: Wiley-Blackwell, 2008), p. 36. 74 Wong Lin Ken, The Trade of Singapore (Bandar Puchong Jaya: MBRAS, 2003), p. 159, 194-204. 69 17 Crawfurd’s days fell to urban redevelopment, textual sources came to supplant the archaeological record in the research on pre-colonial Singapore for most of the 20th century. The reconstruction of Temasek-Singapura from archaeological and historical sources hence constitutes the third and present historiographical phase of Singapore’s pre-colonial history. Through the use of artefacts and archaeological approaches in historical research, a semblance of the pre-colonial settlement was finally developed and supported by concrete evidence. This was initiated with the advent of archaeological excavation and research in Singapore by John Norman Miksic in 1984. Despite the loss of pre-colonial structural and building foundation remains, Miksic managed to identify specialized activities that were conducted within two separate localities – FTC and Parliament House Complex (PHC) – through the meticulous and tedious study of thousands of artefacts recovered from these sites. In line with Crawfurd’s suggestion that FTC was a site of Buddhist worship and monasteries, Miksic inferred from the results of his excavations that ‘a centre of ceremonial activity… or perhaps a monastery’ was located in the vicinity of FTC. 75 The recovery of misshapen glass globules, shards and beads on FTC in 1988 led Miksic to suggest also the presence of an artisan’s glass-recycling workshop as well. 76 In consideration of Crawfurd’s observations as well as the Malay tradition of the hill as the site of ancient kings as recounted by the Temenggong and canonized in the SM, Miksic concluded that ‘FTC can be interpreted as a craftsmen’s quarter within a palace and temple precinct.’ 77 Miksic’s work inspired further archaeological research on artefacts recovered from other pre-colonial archaeological sites in Singapore. Having studied the collection of bronze, iron and gold artefacts, coins as well as huge quantities of metal slag recovered from excavations at PHC, Shah Alam Mohd. Zaini argued in his M.A. thesis that the PHC was a site 75 Miksic, Archaeological Research, p. 90. Miksic, “Beyond the Grave: Excavations North of the Keramat Iskandar Shah, 1988,” in Heritage, ed. Lee Chor Lin (Singapore: National Museum, Singapore, 1989), p. 55-6. 77 Miksic, “14th-Century Singapore. A Port of Trade,” in Early Singapore. 1300s-1819. Evidence in Maps, Texts and Artefacts, ed. Miksic & Cheryl-Ann Low Mei Gek (Singapore: Singapore History Museum, 2004), p. 52. 76 18 of a metal-working sector in pre-colonial Singapore. 78 The presence of numerous coins also spurred Miksic to infer that commercial activities were conducted in the vicinity of PHC as well. 79 Following Miksic’s suggestion that earthenware excavated in Singapore were ‘made by local inhabitants,’ earthenware ceramics from the PHC site formed the subject of another M.A. thesis by Omar Chen Hong Liang. 80 Similarities in appearance and manufacture with earthenware shards from other regional archaeological sites such as Johor Lama in Malaysia, Tanjong Kupang in Brunei and Kota Cina in Sumatra had suggested to Chen the possibility of a pre-colonial ‘Malay’ pottery tradition and to study these artefacts within this context. Compositional and morphological analysis of earthenware shards allowed him to identify the various types and functions of earthenware pottery, as well as reconstruct their approximate production processes. Chen concluded that PHC earthenware, specifically ‘paddle-stamped’ pottery, was manufactured in around the 12th century and suggested that pottery manufacture ceased with the demise of the settlement sometime around the 15th century. 81 A third M.A. thesis by Derek Heng Thiam Soon presents a hypothetical model of the pre-colonial settlement on Singapore Island to which he, taking the lead from Wheatley, ascribed the toponyms of both Bān Zú and Temasek. Drawing heavily from Miksic’s research, Heng utilized ‘historical sources, colonial accounts and results derived from the data of the archaeological excavations and surveys’ to ‘establish a plausible working model against which further studies may be carried out.’ Heng argued that the settlement was a ‘classical period Malay trading port polity’ with an economic hinterland at the Riau Archipelago. This was in turn derived from Miksic’s suggestion of close economic ties between Singapore and Riau on the basis of a similar distribution of Chinese green-ware ceramics in the two areas. 82 An international player exporting a wide range of commodities such as hornbill casques and 78 Shah Alam Mohd. Zaini. Metal Finds and Metal-working at the Parliament House Complex, Singapore. M.A. Thesis. University of Michigan. 1997. 79 Miksic, “14th-Century Singapore,” p. 52. 80 Miksic, Archaeological Research, p. 55-88. 81 Omar Chen Hong Liang. Earth to Earth: An Investigation into the Occurrence of Earthenware Artefacts at the Parliament House Complex Site. M. A. Thesis. National University of Singapore (NUS), 2001. 82 Miksic, “Recently Discovered Chinese Green Glazed Wares of the Thirteenth and Fourteenth Centuries in Singapore and the Riau Islands,” in New Light on Chinese Yue and Longquan Wares. Archaeological Ceramics Found in Eastern and Southern Asia, A.D. 800-1400, ed. Ho Chuimei (Hong Kong: The University of Hong Kong, 1994), p. 229-50. 19 lakawood, the settlement also functioned as an ‘entrepȏt’ which was supposedly heavily involved in economic ‘exchanges at the sub-regional level’ encompassing Sumatra, the lower portion of the Malay Peninsula and Java. Heng also consolidated the primary features of Bān Zú’s urban layout from these sources, namely a ‘Royal Residency,’ ‘Ritual District,’ ‘Royal Garden,’ ‘Servants and Artisans District,’ ‘Orang Laut (sea nomads or gypsies) Settlement Area’ and a ‘Foothill Plain/Commonalties District,’ which he then mapped on a reconstructed plan of pre-colonial Singapore. 83 While the archaeological record was consulted, it was used primarily as a visual illustration of Bān Zú’s material culture, but not in his formulation of the abovementioned settlement’s different districts which was inferred entirely from historical textual sources. Most recently, Heng together with Kwa Chong Guan, and Tan Tai Yong presented a 700-year history of Singapore by claiming Temasek-Singapura as the natural predecessor of the present nation-state of Singapore in the longue durée, thereby drawing close parallels between the island’s past as a ‘thriving emporium’ and its present status as a ‘global city in the post-Cold War cycle of globalization.’ 84 CHAPTER SUMMARIES Although much ink was spilled in the first two historiographical phases of Singapore’s pre-colonial history, a very disproportionately small amount of knowledge about the settlement has been attained. Nonetheless, an amalgamation of hypotheses primarily centred on disparate textual sources – a ‘Temasek Paradigm’ – has emerged as the current framework of analysis governing the study and understanding of pre-colonial Singapore in the present historiographical phase. More importantly, the recovery of new material cultural evidence of pre-colonial Singapore from recent archaeological excavations now provides an invaluable opportunity to verify the underlying narrative set in this paradigm: was pre-colonial Singapore a complex port-city in the 14th century as suggested in various historical sources? What more 83 Derek Heng Thiam Soon. Temasik: Reconstruction of a Classical Period Malay Trading Port Polity. M.A. Thesis. University of London, 1997; see also Heng, “Temasik as an International and Regional Trading Port in the Thirteenth and Fourteenth Centuries: A Reconstruction based on Recent Archaeological Data,” JMBRAS 72,1 (June 1999), p. 113-24; Heng, “Reconstructing Banzu, A Fourteenth Century Port Settlement in Singapore,” JMBRAS 75,1 (June 2002), p. 69-90. 84 Kwa Chong Guan, Heng, Tan Tai Yong, Singapore. A 700-Year History (Singapore: National Archives of Singapore, 2009), p. vii, 9-52. 20 can we learn about this settlement and its inhabitants, and what implications will the results of archaeological analysis have on this paradigm? To do anything otherwise risks rendering the results and arguments of further historical research meaningless if and when contradicting archaeological finds are recovered in the future. Hence, Chapter 2 will summarize and present the Temasek Paradigm as well as various elements of settlement complexity according to textual sources contemporaneous to the settlement. A review of key interpretations of these sources by various scholars will also illustrate their limits in providing a conclusive picture of pre-colonial Singapore and hence, highlight the need for incorporating more archaeological data and analyses in the study and reconstruction of this settlement. Chapter 3 presents the wealth of information derived from archaeological data. Besides showing that STA was a pre-colonial archaeological site contemporaneous to the polity of Temasek-Singapura as mentioned in various historical sources, the non-random variation in the distribution of sampled artefacts and evidence of craftspecialization at the site will also demonstrate the presence of complex organization in precolonial Singapore. This intra-site analysis will provide historians and archaeologists alike the first time an insight to the nature of pre-colonial Singapore’s spatial organization, which will be a first for any Southeast Asian settlement site in this period. In conclusion, the level of settlement complexity derived from archaeological sources will be evaluated against that already established in the paradigm to create a much more comprehensive and plausible picture of 14th century Singapore. 21 II. TEMASEK-SINGAPURA: THE PARADIGM AND ITS SOURCES This, then, is not an indictment of evidence but of methodology: of the way data has been assessed and used to conform to a preconceived theory. Michael A. Aung-Thwin, 2005. 1 This chapter begins by asking a simple yet fundamental question: What was precolonial Singapore? As seen in the previous chapter, various scholars have directly or indirectly answered this question in their attempts to historicize the island’s misty past. Consequently, their collective work has coalesced into the present conceptualization of pre-colonial Singapore as the site of the elusive polity of Temasek-Singapura. The present thesis refers to this conceptualization as the ‘Temasek Paradigm,’ one which serves as a theoretical and philosophical framework for the study of pre-colonial Singapore. In many respects, the paradigm and its historical sources suggest that pre-colonial Singapore – as Temasek-Singapura – was a highly complex, hierarchical and socially-stratified society built on the economic foundation of both regional and international maritime trade. However, to paraphrase Michael A. Aung-Thwin’s quote, it is not the validity of these sources that is interrogated here, but rather, that of their interpretations by respective scholars which have led to the formulation of this conclusion within the paradigm. THE PARADIGM The paradigm’s conceptualization of pre-colonial Singapore is best enunciated by the most recent works of local scholars seeking to shift the writing of a histoire événementielle of 1 Michael A. Aung-Thwin, The Mists of Rāmañña. The Legend That Was Lower Burma (Honolulu: University of Hawai‘i Press, 2005), p. 3. 22 Singapore, conventionally beginning with the arrival of the British in 1819, to that of the island’s settlement history in la longue durée. Seeing ‘the past in Singapore’s present,’ Heng identified ‘Temasik’ as a ‘Malacca Straits region port-polity’ and the ‘first documented settlement to exist on Singapore Island.’ Temasek-Singapura is hence situated as the first of five key phases in the island’s settlement history; followed by Singapore as part of the Melaka and Johor Sultanates between the 15th and 17th centuries; next as an East India Company factory in the 19th century; then as a Crown Colony and administrative centre of British Malaya until 1963; then part of the Malaysia which ended in 1965; and finally as the present independent nation-state of Singapore. In the process, Heng highlighted features such as ‘political autonomy,’ ‘existence of core and non-core groups,’ ‘having to make local products available for export,’ ‘absence of a geographical hinterland,’ ‘economic sphere limited to Johor and Riau’ and ‘extent of economic sphere dictated by larger/regional contexts’ as recurring patterns of Singapore’s settlement history governed by the ‘historical experiences and constraints’ of its geographical location in the Melakan Straits region. 2 Following Heng’s call for the integration of Temasek-Singapura into the island’s precolonial past, a textbook entitled Singapore. A 700-Year History. From Early Emporium to World City was published by the National Archives of Singapore to spearhead ‘a comprehensive re-telling of the Singapore story from a local perspective.’ Co-authored by Heng, Temasek-Singapura is advanced again as the natural predecessor of the present nation-state of Singapore in the longue durée, thereby drawing close parallels between the island’s past as a ‘thriving emporium’ and its present status as a ‘global city in the post-Cold War cycle of globalization.’ Calling for a ‘long-sighted view of Singapore’s past,’ the settlement is posited to be both a thriving port of trade and the capital of a Malay polity in a post-Srivijayan world that struggled to maintain its autonomy vis-à-vis Majapahit and ‘Siamese’ hegemony. At the same time, the emergence of Temasek-Singapura as a port of trade is now situated within the 2 Heng, “Indigenising Singapore’s Past: An Approach towards Internalising Singapore’s Settlement History from the Late Thirteenth to Twenty-First Centuries,” in New Perspectives and Sources on the History of Singapore, ed. Heng (Singapore: National Library Board, 2006), p. 15-27. 23 context of the rise of multiple autonomous Southeast Asian ports – a result of the increase in Chinese trading activity and the decline of Srivijayan hegemony along the Straits of Melaka – that participated in the Asian maritime economy between the 11th and 14th centuries. Its primary function, the authors argued, was to serve as a ‘collection centre’ and ‘export gateway’ for indigenous products between ‘the immediate hinterland of the Riau Archipelago and South Johor’ and ‘the wider Asian maritime economy,’ specifically that of the South China Sea. ‘Unique’ commodities such as lakawood, hornbill casques, cotton and elephants were quoted as the primary exports from its port. Consequently, the settlement is postulated to exist sometime between the end of the 13th century, ‘emerged as a prosperous emporium that catered mainly to the Chinese market’ by the mid-14th century and met its demise as a significant trade centre with the establishment of Melaka in the 15th century. 3 As a result of the supposed Palembang origin of Temasek-Singapura’s first ruler as well as the possible cultural links suggested by previous scholars between Srivijaya and the settlement, it is also proposed by the book’s authors that Srivijaya-Palembang characterized the physical attributes of Temasek-Singapura and the other Malay polities in the region. Specifically, Temasek-Singapura is described as a ‘classical Malay port-city,’ defined by the authors to mean a settlement model in the tradition of the ‘first Malay emporium’ of SrivijayaPalembang from the 7th to 13th centuries, which was subsequently inherited by that of the Melaka Sultanate in the 15th century. According to the authors, later polities in the Melaka Straits region sought to replicate in themselves the likeness of the former Srivijayan Empire in order to emulate its economic and political success and dominance. Succeeding port-cities in the south Sumatran-Riau region are believed to be generally modelled after the former Srivijayan capital of Palembang. A ‘remarkable sense of continuity’ from SrivijayaPalembang to Temasek-Singapura was apparently manifested in urban layouts that were 3 Kwa et. al., Singapore. A 700-Year History, p. v, vii-iii, 19-32. 24 FIGURE 3. Map of reconstructed 14th century settlement in Singapore superimposed on 1836 map of Singapore town drawn by J. B. Tassin. The North arrow is added by the present author to indicate the cardinal directions. Source: Kwa et. al., Singapore. A 700-Year History, p. 36. 25 broadly divided between a ‘ritual and political centre,’ and the ‘main settlement area.’ These spaces are in turn defined by key geographical features such as a hill, a foothill plain and a river drainage basin which ‘facilitated the full functioning of the raison d'être of a Malay port-polity.’ This theory is also apparently supported by ‘a high level of consistency in the material culture’ of other ‘port-cities’ such as Palembang, Kota Cina and Melaka. 4 In Singapore, these key features are hypothesized to be represented by FTC, the land adjacent to the north bank of the Singapore River and its basin respectively as seen in Figure 3. According to the authors, the hill symbolized Mount Meru, the centre of Hindu-Buddhist cosmology, and is also believed to be the ‘ritual and political centre’ or ‘palace precinct’ within which the ruling elite and their servants dwelled. This precinct is sub-divided into three different spaces – ‘free access spaces, restricted access spaces and prohibited access spaces.’ The artisan quarter, servants’ quarter and ritual grounds for the population were supposed to be within the free access space; those that were of restricted access consisted of the palace grounds and royal garden, whereas the royal residency and bath would make up the prohibited access spaces. Finally, the ‘main settlement area’ of residential and commercial zones occupied the remaining land. 5 SETTLEMENT MODELS AND COMPLEXITY From the paradigm’s conceptualization of the urban layout of pre-colonial Singapore as Temasek-Singapura, it appears self-evident that a complex society once thrived on the island in the 14th century. The fundamental problem however lies in the fact that no concrete settlement model has been established for Srivijaya-Palembang, the purported archetypal ‘classical Malay port-city,’ which is itself viewed as a model for Temasek-Singapura. As Wolters noted, ‘little more besides an impression of the power of the kingdom’ has been established despite the existence of substantial documentation in the form of several translated 4 5 Ibid., p. 33-52. Ibid. 26 inscriptions and Chinese records of the polity. 6 Despite the fact that these records illustrate ‘some form of hierarchical organization’ within the polity, the lack of archaeological data has prevented the study of its internal economic structure or organization, let alone fitting it into the larger typologies of ‘state,’ ‘kingdom,’ or ‘empire.’ 7 Even less is known about the ‘morphological structure of Sriwijaya-town’ at Palembang, as buildings were likely constructed of perishable materials of which almost nothing has survived. 8 Although recent surveys and excavations by French and Indonesian archaeologists at Palembang from 1988 to 1991 have uncovered several Srivijayan sites and large numbers of accompanying material cultural remains in the vicinity of Palembang, no more of its organization can be said beyond a generalized description as a ‘Malay World harbour-city.’ In the words of Pierre-Yves Manguin and his contemporaries, what can be deduced of Srivijayan-Palembang is only a sense of … a relatively dense population gathered around centres of political and economic power within a loosely knit, semi-rural urban environment, dissolving into the countryside at the peripheries, with wood providing the essential part of, if not all building materials… it appears that the sites in Palembang fit in nicely within the [abovementioned] pattern defined [by Anthony Reid] 9 for later times. 10 The ‘lack of sufficient archaeological data’ and analyses still plagues the understanding of this settlement to this day. Far from a well-defined idea of a ‘classical Malay port-city,’ little is known about the social organization or complexity at Palembang besides establishing it as the former capital of Srivijaya ‘beyond doubt,’ and the reiteration of the presence – not the analysis – of ‘archaeological evidence for settlement, manufacturing, commercial, religious and political activity.’ 11 This issue is further exacerbated by the textbook’s unsubstantiated claim 6 Wolters, Early Indonesian Commerce, p. 16. Miksic, “Srivijaya: Political, Economic, and Artistic Frameworks for Analysis,” in SEAMO Project in Archaeology and Fine Arts (SPAFA). Final Report. Consultative Workshop on Archaeological and Environmental Studies on Srivijaya (T-W3). Bangkok and South Thailand. March 29-April 11, 1983, ed. Suchitra Vuthisathira (Bangkok: SPAFA Co-ordinating Unit, 1983), p. 195-8, 2012. 8 Peter J. M. Nas, “The Early Indonesian Town. Rise and Decline of the City-State and its Capital,” in The Indonesian City. Studies in Urban Development and Planning, ed. Peter J. M. Nas (Dordrecht: Foris Publicaions, 1986), p. 25-6; see also Bronson & Jan Wisseman-Christie, “Palembang as Śrīvijaya. The Lateness of Early Cities in Southern Southeast Asia,” AP 19,2 (1976), p. 234-7. 9 Anthony Reid, “The Structure of Cities in Southeast Asia, Fifteenth to Seventeenth Centuries,” JSEAS 11,2 (September 1980), p. 235-50. 10 Manguin, “Palembang and Sriwijaya: An early Malay harbour-city rediscovered,” JMBRAS 66,1 (1993), p. 33-4. 11 Manguin, “The Archaeology of Early Maritime Polities of Southeast Asia,” in Southeast Asia. From prehistory to history, ed. Ian Glover and Peter Bellwood (Oxfordshire: RoutledgeCurzon, 2004), p. 305-8; see also Merle Calvin Ricklefs, Bruce Lockhart, 7 27 that archaeological data suggests the ‘settlement pattern of the Malay port-city’ persisted in the region from at least the second half of the 10th century to the 16th century. 12 If a definite, historical settlement layout of Srivijaya-Palemabang remains to be developed, the alternative appears to be the use of hypothetical settlement models as analytical frameworks to make sense of existing textual and archaeological sources on pre-colonial Singapore, through which an idea of its social organization can be derived. One such model was developed by Bennet Bronson in 1977 as seen in Figure 4. It was ‘designed to improve’ the understanding of the history of the Southeast Asian sub-region of ‘the thinly-populated coastlines of the large insular and peninsular land masses of Malaysia, the Philippines, and western Indonesia,’ to which pre-colonial Singapore and Palembang both belong. 13 Bronson’s model illustrates an exchange network along a drainage basin opening to the sea, with a firstorder (A) population centre at the river mouth, second- (B) and third- (C) order centres located upstream at primary and secondary river junctions, the most distant upstream centre (D) participating in the ‘A’-based system of market exchange, the ultimate producers of these products (E and F) and possibly centres of a separate exchange system based on non-market institutions, an overseas centre (X) serving as the main consumer of goods exported from ‘A’ and the principal supplier of its imports and another river-mouth centre ‘A*’ some distance along the coast, controlling a hinterland similar to that of ‘A.’ 14 Bronson’s model is useful in hypothesizing the possible ‘political and economic functioning’ of a single settlement site within a large system incorporating numerous other settlement sites in a wider geographical context. 15 According to this model, both SrivijayaPalembang and Temasek-Singapura can be interpreted as first-order population centres of two exchange networks along the Musi and Johor Rivers respectively; the identification of Albert Lau, Portia Reyes & Maitrii Aung-Thwin, A New History of Southeast Asia (Hampshire: Palgrave Macmillan, 2010), p. 2930, 61-2. 12 Kwa et. al., Singapore. A 700-Year History, p. 38. 13 Bennet Bronson, “Exchange at the Upstream and Downstream Ends: Notes toward a Functional Model of the Coastal State in Southeast Asia,” in Economic Exchange and Social Interaction in Southeast Asia: Perspectives from Prehistory, History and Ethnography, ed. Karl L. Hutterer (Ann Arbor: The University of Michigan, 1977), p. 39. 14 Ibid., p. 42-3. 15 Ibid., p. 44. 28 FIGURE 4. Abstract model for Exchange between a Drainage Basin Centre and an Overseas Power. Source: Bronson, “Exchange at the Upstream and Downstream Ends: Notes toward a Functional Model of the Coastal State in Southeast Asia,” p. 42. 29 Temasek-Singapura as a ‘collection centre’ and ‘export gateway’ as well as Riau and Johor as its ‘immediate hinterland’ by Heng and his contemporaries clearly demonstrates the application of Bronson’s model in the study of pre-colonial Singapore. Although the model does not elaborate on the social organization at any single settlement site along this exchange network, the degree of complexity can still be indirectly inferred from its position within it. As Bronson suggests, ‘quasi-kinship institutions combined with clientship and trade-partner relationships’ may have characterized ‘D’ centres; ‘B’ and ‘C’ centres may also be either subordinates or competitors of ‘A’ centres, which suggests the presence of bureaucratic institutions either asserting or competing for political control over the riverine exchange network. 16 This perspective of complexity however is entirely contingent upon comparisons between at least two well-documented settlement sites with verifiable historical and cultural connections. While a number of coastal sites in Sumatra and the Malay Peninsula bear evidence of Srivijayan cultural influence as suggested by several inscriptions as well as HinduBuddhist images and architecture on both sides of the Melakan Straits, no settlement site sharing a similar degree of cultural affinity with Temasek-Singapura has been found in either Riau or Johor to date. 17 It is hence premature to apply Bronson’s model in the study of precolonial Singapore. Therefore, one has to consider other alternative models instead. Wolter’s maṇḍala, the leading analytical framework for the structure of a Southeast Asian state, comes straight to mind in the study of organization in pre-colonial Southeast Asia. It conceptualizes a multicentred political system encompassing expanding and contracting ‘circles of kings’ governing through patrimonial bureaucracies. 18 Another similar model is represented by the ‘negara,’ a realm focused on a single capital city exerting a ‘centripetal pull’ which gradually diminishes in strength as it approaches a neighboring sovereign. The city is therefore characterized by a 16 Ibid., p. 44-9. Wolters, “Śrīvijaya Expansion in the Seventh Century,” Artibus Asiae (AAE) 24,3/4 (1961), p. 417-24; see also Nik Hassan Shuhaimi bin Nik Abd. Rahman. Art, Archaeology and the Early Kingdoms in the Malay Peninsula and Sumatra: C.400-1400 A. D. Ph.D Thesis. University of London, 1984. 18 Wolters, History, Culture and Region in Southeast Asian Perspectives (Ithaca: Cornell University, 1999), p. 27-40. 17 30 centralized administrative structure which is ‘formally hierarchical’ but effectively composed of ‘stratified clusters of patron-client relationships.’ 19 However, both the maṇḍala and ‘negara’ face the same problems as Bronson’s model if they are applied in the study of social organization in pre-colonial Singapore. Besides the fact that it is unclear if Temasek-Singapura functioned historically as a city-state given the paucity of knowledge, it is also impossible to tell if the settlement was either the centre of a Temasek-Singapura maṇḍala, which would affirm the settlement’s complexity by default, or a minor population centre at the periphery of another. There is also no verifiable evidence as yet suggesting that Temasek-Singapura had functioned as a negara which drew populations outside the settlement under its control. Another attempt to define the relationship between social organization and settlement models can also be found in Wheatley’s Nāgara and Commandery, a seminal study of the origins of Southeast Asian urban traditions. Synthesizing extensive archaeological, epigraphical and textual sources, Wheatley observed that the transition from an ‘undifferentiated tribal’ to ‘complex urban or state’ society is typically marked by a change from one distinct level of socio-cultural integration to another; 20 this is induced in turn at the intermediate phase of ‘chiefdoms’ by acculturation processes from the establishment of crosscultural contacts with India and China. Central to this process of change were the chiefs or ‘men of prowess,’ who begin to ‘consolidate and elaborate the structure of authority relationships,’ which led to the emergence of divine kingships from which the city and the state began to manifest. 21 In the process, Wheatley redefined the ‘chiefdom’ from a dichotomy between the ruler and the ruled into ‘a quadripartite division in which cultural-religious and politico-military elites are opposed to lower-status rural and urban groups. 22 19 Benedict Richard O’Gorman Anderson, “The Idea of Power in Javanese Culture,” in Culture and Politics in Indonesia, ed. Claire Holt (Ithaca: Cornell University Press, 1972), p. 1-69; Wisseman-Christie, “Negara, Mandala, and Despotic State: Images of Early Java,” in Southeast Asia in the 9th to 14th Centuries,” ed. David G. Marr and Milner (Singapore: ISEAS, 1986), p. 67-9. 20 Wheatley, Nāgara and Commandery: Origins of the Southeast Asian Urban Traditions (Illinois: The University of Chicago, 1983), p. 420. 21 See also Wolters, History, Culture and Region, p. 15-26, 93-5, 226-8; Ricklefs et. al., A New History of Southeast Asia, p. 3840. 22 Wheatley, Nāgara and Commandery, p. 20. 31 Wheatley’s model of social complexity revolves around his idea of urban genesis which depicts a three-level urban hierarchy (state capital – regional capital – centre of local importance) in Southeast Asia within an urban-political system where ‘provincial, district, and lesser seats were often constituted as reduced models of their paramount capitals’ as seen in Figure 5. The paradigm’s advocation of ‘continuity’ from Srivijaya-Palembang to Temasek-Singapura clearly reflects this idea. However, Wheatley’s model is only relevant to strictly orthogenetic (ceremonial) settlement sites with distinct features such as walls or ramparts, palace compounds, towers and gateways, sacred mountains and an urban layout orientated towards cardinal compass directions or axis mundis of cosmological significance – in other words, archaeological sites with significant monumental or structural remains. 23 Moreover, Wheatley excluded the heterogenetic (commercial) ‘pasisir’ or port-city settlement-type, which characterizes pre-colonial Singapore in the paradigm, outside his discussion of Southeast Asian urban genesis, as heterogenetic settlements do not conform to his idea of gradual evolution from one distinct level of socio-cultural integration to another as described in orthogenetic settlements. 24 Wheatley’s model is therefore unable to explain the development of social complexity at heterogenetic sites like pre-colonial Singapore. As an alternative to Wheatley’s approach, Miksic proposed to study socio-cultural organization in Southeast Asia through settlement pattern studies. This analytical framework builds upon the theory that settlement patterns ‘could serve to indicate the level which political development had reached in ancient societies.’ These patterns are usually defined by a hierarchy of settlement sites – typically on the basis of gross size – which are then used to reflect socio-economic relations between their inhabitants. 25 As Miksic noted however, settlement pattern studies is still in its infancy; significantly, they require a comprehensive 23 In ‘premodern’ Southeast Asia, ‘orthogenetic’ cities generally refer to settlement sites which, among other things, primarily function as a ceremonial centre of the state or polity and usually characterized with large monumental remains, whereas ‘heterogenetic’ cities generally refer to settlement sites which primarily function as centres of foreign trade and commerce instead of monument-building activities, hence leaving very little architectural traces of its physical existence. See Robert Redfield & Milton B. Singer, “The Cultural Role of Cities,” Economic Development and Cultural Change (EDCC) 3,1 (October 1954), p. 5373; Miksic, “Heterogenetic cities in premodern Southeast Asia,” World Archaeology (WA) 32,1 (June 2000), p. 106-20. 24 Wheatley, Nāgara and Commandery, p. 425-33. 25 Miksic, “Settlement Patterns and Sub-regions in Southeast Asian History,” Review of Indonesian and Malayan Affairs (RIMA) 31,1 (1991), p. 86-144. 32 FIGURE 5. The principal [orthogenetic] urban hierarchies in Southeast Asia in the second half of the 14th century. Source: Wheatley, Nāgara and Commandery, p. 426; see also Sandhu & Wheatley, “The Historical Context,” in Melaka, p. 24. 33 analysis of the material cultural remains of any group of settlement sites before variables can be used to establish any form of hierarchy between these sites. As such, only a general difference – that of hierarchies of settlements forming on mainland Southeast Asia a millennium before island Southeast Asia – has been elicited from the abundant archaeological data. 26 Another problem noted by Miksic is that the boundaries of any settlement system must first be specified before studying the patterns within it. 27 In other words, the relationships between individual settlement sites must be clearly established and justified which again, necessitates a thorough analysis of the artefact assemblages of every site involved. Finally, it might be convenient and logical to infer the organization of pre-colonial Singapore by extrapolating historical elements and characteristics of social organization in Southeast Asia from its well-documented ‘modern’ period into Temasek-Singapura. An erudite study of social organization in the region has already been provided by Anthony Reid in a chapter within his study of Southeast Asia in the Age of Commerce between 1450 and 1680. 28 This heuristic method may be useful to the historian as a historiographical ‘last resort’ in the event of the complete absence or loss of primary sources of pre-colonial Singapore. In light of the recovery of significant amounts of material cultural remains from this period however, it remains more appropriate and accurate to elicit elements of settlement organization directly from the archaeological analysis of these artefacts which this pre-colonial society – the subject of the present thesis – left behind. TEXTS AND INTERPRETATIONS The second issue lies in the interpretation of historical texts mentioned in the previous chapter as primary sources of pre-colonial Singapore. While it can be inferred from these sources that Temasek-Singapura was a major pre-colonial city of political significance, only two sources have described the social organization of the settlement at length: the SM compiled 26 Ibid., p. 101. Ibid., p. 130-1. Reid, Southeast Asia in the Age of Commerce. 1450-1680. Volume One. The Land below the Winds (New Haven: Yale University Press), p. 120-72. 27 28 34 in the 17th century and Wāng’s Description written in the 14th century. Details have been extracted from these sources by various scholars to support the Temasek Paradigm and used to reconstruct the urban environs of pre-colonial Singapore. I. The Sejarah Melayu It is clear that the narrative of Temasek-Singapura in the SM forms the basis of the paradigm, as it is claimed as ‘one of the most important indigenous sources of information for the reconstruction of Temasek’s urban features during the late-13th and 14th centuries.’ It is also advanced that the SM is a consolidation and reconstruction of Malay ‘collective memories’ and is no more inaccurate than the accounts of European sources which were also derived from the social memories of the indigenous population. 29 According to the SM, the settlement was established when Sri Tri Buana, a Sumatran prince with a semi-divine lineage, migrated with his followers from Palembang to Temasek sometime between the 12th and 13th centuries. A total of five kings had apparently reigned in this Temasek-Singapura before it was supposedly annexed by the soldiers of the Batara (ruler) of Majapahit towards the end of the 14th century. The last ruler, Sultan Iskandar Shah, fled and re-established his court in Melaka. Descriptions of social status such as ‘master,’ ‘servant’ and ‘slave,’ nobility titles of ‘Awang’ and ‘Dara’ as well as major (‘chief minister,’ ‘Bendahara,’ ‘Penghulu Bendahari,’ ‘Temenggong,’ etc) and minor (‘ministers,’ ‘knights,’ ‘courtiers,’ ‘heralds,’ ‘war-chiefs,’ etc) political ranks, said by the SM to be ‘in accordance with the custom dating from ancient times,’ suggest at least a two-tier social stratification – commoners and the political elite – below the third and highest tier of the royal family. 30 As Wolters noted, this was a ‘fully-fledged court hierarchy’ almost identical to that of the later Melakan Sultanate. 31 Besides providing a genealogy of its successive kings, the SM also records anecdotes of various socio-political events which explain Temasek-Singapura’s political and economic relations 29 Kwa et. al., Singapore. A 700-Year History, p. 11-5, 33. Brown, Malay Annals, p. 31. 31 Wolters, The fall of Śrīvijaya in Malay History, p. 120. 30 35 with its neighbours such as Kalinga, Pasai, Perlak and Majapahit. 32 In doing so, the SM implicitly alludes to Temasek-Singapura as the successor of Palembang, the old capital of the once formidable Srivijayan thalassocracy, as well as the predecessor of Melaka, the capital and famed emporium of the Melakan sultanate. The fact that the SM was accepted ‘everywhere in Southern Malaya,’ especially by the royal families of Perak, Johor, Lingga, Terengganu and Pahang who believed in their descent from Alexander the Great, appear to validate its authenticity as a historical source. 33 As late as the 19th century, the Tuhfat al-Nafis (The Precious Gift) still outlines the ‘history of the Bugis in the Malay world’ since the ‘original Malay king’ from Palembang, ‘Raja Seri Teri Buana’ (Sri Tri Buana). 34 The SM’s stature among Malay oral tradition and culture is reinforced further by the fact that it is one of the oldest surviving indigenous Malay histories, besides the Hikayat Raja-Raja Pasai (Chronicles of Pasai) in both the Malay Peninsula and Sumatra. 35 Moreover, the SM provides a relatively accurate account of the Melaka and Johor Sultanates which can be broadly corroborated with early Portuguese sources. 36 Fantastical stories which characterize the SM’s narrative of Temasek-Singapura, such as a semi-divine lineage from Alexander the Great, a submarine voyage to an underwater city and an invasion of a horde of killer swordfish, are notably reduced or absent in the accounts of these two sultanates as well. Although it was noted by Wilkinson that ‘the miraculous incidents [within the SM] fail to commend themselves to the sceptical historians of the present day,’ he argued that the text ‘gave a very life-like picture of the times… which is of very real importance to the scientific historian who cares more about the condition of the people…’ 37 Similarly, Winstedt believed that the SM can provide ‘a vivid picture’ and life-like vignettes’ of the times, even though he 32 Brown, Malay Annals, p. 31-51. Wilkinson, A History of the Peninsular Malays, p. 20-4. 34 Virginia Matheson & Barbara Watson Andaya, The Precious Gift (Tuhfat al- Nafis) (Kuala Lumpur: Oxford University Press, 1982), p. 5, 12. 35 Winstedt, “A History of Malay Literature,” p. 105-9. 36 C. H. Wake, “Melaka in the Fifteenth Century: Malay Historical Traditions and the Politics of Islamization,” in Melaka. The Transformation of A Malay Capital c. 1400-1980. Volume One, ed. Kernial Singh Sandhu & Wheatley (Kuala Lumpur: Oxford University Press, 1983), p. 128-40. 37 Wilkinson, A History of the Peninsular Malays, p. 19; Wilkinson, Papers on Malay Subjects. Malay Literature, Part 1. Romance. History. Poetry (Kuala Lumpur: Federated Malay States Government Press, 1907), p. 33-4. 33 36 called the SM an anachronistic ‘olla-podrida’ and ‘hotch-potch of myths and traditions.’ 38 These beliefs were then inherited by the scholars of the post-colonial period. J. C. Bottoms argued that the SM, together with other early Malay texts, can be used to write ‘A Social History of the Malays,’ 39 whereas A. H. Johns labelled the SM as ‘an important cultural document’ for an insight into the ‘ethnic Malay [world] shape, Indo-Persian [world] shape, and an AraboIslamic [world] shape’ which pervaded pre-colonial Malay society. 40 The present thesis does not dispute the value of the SM as a source of social history, but rather, questions the claim that factual details can be elicited from it such that an accurate reconstruction of the physical urban environment of pre-colonial Singapore can be achieved. 41 At the forefront of this criticism was Wolters, who argued that SM was a literary device meant to justify the legitimacy of the Sultans of Johor, as well as their right to rule the Malays as descendants of the kings of the Srivijaya-Palembang-Melaka lineage. In particular, Wolters stated that the ‘glorious history of Singapore’ was likely an account intended to mask the ‘passing of Palembang’s hegemony in the 11th century to Melayu-Jambi,’ an ‘alternative rendering of a period of history obviously unacceptable from a Palembang point of view.’ 42 In other words, there is no factual basis for Temasek-Singapura’s image as a historical polity in the accounts of the SM. This might still be a valid presumption, had the SM been written by an actual inhabitant of Temasek-Singapura when it supposedly existed between the 13th to 14th centuries. However, the earliest known version of the SM only came into commission under the patronage of the Johor court sometime between 1612 and 1613; two centuries after the events recorded in the 38 Winstedt, “Malay Chronicles from Sumatra and Malaya,” in Historians of South East Asia, ed. D. G. E. Hall (Oxford: Oxford University Press, 1961), p. 25-6; Winstedt, “A History of Malaya,” p. 34. 39 J. C. Bottom, “Some Malay Historical Sources: A Bibliographical Note,” in An Introduction to Indonesian Historiography, ed. Soedjatmoko, Mohammad Ali, G. J. Resink & G. McT. Kahin (Ithaca: Cornell University Press, 1965), p. 188-90. 40 A. H. Johns, “The Turning Image: Myth & Reality in Malay Perceptions of the Past,” in Perceptions of the Past in Southeast Asia, ed. Reid and David Marr (Singapore: Asian Studies Association of Australia, 1979), p. 46. 41 Kwa et. al., Singapore. A 700-Year History, p. 33-4. 42 Wolters, The fall of Śrīvijaya in Malay history, p. 80-1, 90-1, 145-53, 176-7. 37 SM supposedly took place; 43 Gerrit Pieter Rouffaer placed the exact date as 12 May 1612. 44 Winstedt even postulated that Tun Bambang, the author of the SM, was a former inhabitant of Melaka; it is now believed that he began writing the SM in Johor and probably completed his work while in captivity in Aceh when Johor was overrun by the Acehnese in 1613. 45 Even if there was some factual basis in the SM’s narrative of pre-colonial Singapore, it remains impossible to distinguish between myths and historical facts within the SM without knowing the epistemological context of its author in the first place. The authors of Singapore. A 700-Year History, however, have chosen to identify information ‘on various urban features of Temasek’ as historical facts, thereby leading to the hypothetical reconstruction of the physical layout and usage of spaces at the pre-colonial settlement on Singapore Island. To quote the authors, … the value of the text [SM] as a source of information on various urban features of Temasek should not be too readily dismissed. These include descriptions of markets, the royal gardens, spaces that reflect the social hierarchy of the population and the defence features of the city. Almost all the instances of the physical features mentioned in the text are accurate, even if the means by which these features were supposed to have come about are mythical. 46 The authors then argued that these urban features can be corroborated with the 19th century observations of Crawfurd and Abdullah on the pre-colonial ruins in the vicinity of FTC. 47 However, the corroboration of the physical existence of Temasek-Singapura with these early colonial accounts is actually a rhetorical tautology centred on the SM, based entirely on the mutually-reinforcing but unsubstantiated belief of Raffles and the Temenggong in locating Temasek-Singapura on Singapore Island. How Raffles came to this conclusion has already been addressed in the previous chapter; in the case of the Temenggong and his followers, they were themselves only recent migrants from Riau to Singapore Island in 1811, a mere eight years 43 Brown, Malay Annals, p. ii. Peter Borschberg, “The Johor-VOC Alliance and the Twelve Years’ Truce: Factionalism, Intrigue and International Diplomacy. 1606-13,” Institute of International Law and Justice (IILJ) Working Papers 2009/8. (History and Theory of International Law Series). Finalized 12/17/2009. (www.iilj.org), p. 33, footnote 120. 45 Winstedt, “A History of Malay Literature,” p. 106-9; 46 Kwa et. al., Singapore. A 700-Year History, p. 34. 47 Ibid., p. 34, 39, 43, 48-9. 44 38 before the British arrived in 1819. 48 Just as the SM led Raffles to identify Singapore Island as Temasek-Singapura, the Temenggong’s account of the ruins of a pre-colonial settlement to Farquhar can only be, again, presumptions similarly informed by the SM; as opposed to knowledge inherited through successive generations of descendants who lived on the island since the original inhabitants of pre-colonial Singapore. The ‘story dating from the kings of ancient times’ is clearly a reference of the opening chapters of the SM, which recounts the semidivine genealogy of the Malay royal families from the time of Alexander the Great. 49 Moreover, as the earliest version of the SM is thought to have been complied around 1436, Miksic suggested that ‘a number of embellishments’ have been added over the centuries in view of the ‘various longer versions’ of the SM such as the Raffles MS 18. 50 This view is supported by Milner who identified the cause of such ‘embellishments’ as that of succeeding copyists who ‘are well known to have made changes of various types’ to the SM. 51 Hence, the relationship between the SM’s accounts of these urban features and the early colonial observations of precolonial material cultural remains on Singapore Island remain tenuous. II. Description of the Barbarians of the Isles The second crucial source of the Temasek Paradigm is that of Wāng’s Description. As mentioned in the previous chapter, Wāng’s toponyms of Lóng Yá Mén and Bān Zú in particular are believed by many scholars to refer to Keppel Harbour and FTC respectively. 52 Wāng’s account is hence crucial to the paradigm, as not only did these references appear to depict precolonial Singapore as Temasek-Singapura, it provides an insight into the social organization of the polity as well. According to Wāng, the ‘Dān Mǎ Xī barbarians’ who lived at Lóng Yá Mén practiced wet-rice cultivation, but the fields were noted to be ‘barren.’ They had a chief who wore a ‘jewelled head-dress’ as well as ‘ceremonial dress,’ and practiced a custom where the people 48 Crawfurd, Journal of an Embassy. Volume II, p. 403. Hill, Hikayat Abdullah, p. 142. 50 Miksic, Archaeological Research, p. 19. 51 Milner, The Malays, p. 90. 52 Wheatley, The Golden Khersonese, p. 82-6. 49 39 ceremonially congratulated the chief in his full splendour at the start of the new year. Wāng also noted that the ‘natives and the Chinese dwell side by side.’ The ‘Dān Mǎ Xī barbarians’ also appeared to be familiar with metal-working, ship-building and sailing, as they were known to have engaged in piracy dressed in ‘armour’ and on ‘some two or three hundred pirate prahus.’ Trade was an important feature in this settlement, as the locals traded indigenous products such as ‘coarse lakawood and tin’ for imports which include ‘red, gold, blue satin, cotton prints, Ch’u [Chinese] porcelain, iron cauldrons.’ The settlement also had a city wall and moat as noted in the previous account of the Siamese conflict with the ‘Dān Mǎ Xī barbarians.’ In another entry under the toponym of Bān Zú, this settlement was described to be located on a hill behind Lóng Yá Mén which ‘resembled a truncated coil’ that ‘rises to a hollow summit [surrounded by] interconnected terraces, so that the people’s dwellings encircle it.’ Its inhabitants were also organized under a chieftain and they had occupations which involved salt production from drying sea water, as well as fermenting rice for alcohol. Bān Zú’s inhabitants produced ‘very fine hornbill casques, lakawood of moderate quality, cotton’ and ‘cotton prints of local manufacture’ which they exchanged for ‘green cottons, lengths of iron, [Chinese] half-tael coins, porcelain-ware, iron pots’ in trade. 53 Wāng’s account, as an independent primary source of pre-colonial Singapore, makes a compelling case for the historical existence of Temasek-Singapura. The problem however is not the source itself, like that of the SM, but in the interpretations of this source within the Temasek Paradigm. It must first be noted that successive scholars have interpreted Lóng Yá Mén as present-day Keppel Harbour of Singapore Island, primarily because ‘Dān Mǎ Xī’ is widely believed to be the Chinese transliteration of ‘Temasek.’ Consequently, it was concluded that Bān Zú – as a hill ‘behind Lóng Yá Mén’ – can only be FTC, the ‘eminence which dominates Singapore City.’ 54 In other words, the interpretation of these two toponyms under the paradigm and hence, the accompanying information of social organization and urban layout See also 苏继倾 (Sū Jì Qīng) 《島夷梽畧校释》 (Dăo Yí Zhì Lüe Jiào Shì or ‘An explanation of the Description of the Barbarians of the Isles’), p. 154-9, 196-9, 213-8; Rockhill, “Notes on the Relations and Trade of China,” p. 62, 99-100, 129-32. 54 Wheatley, The Golden Khersonese, p. 82-3. 53 40 of the respective localities, is entirely contingent on the identification of Lóng Yá Mén with Keppel Harbour. This identification is based on the method which considers a toponym to be validly identified if there is a coincidence of ‘geographical, historical, and etymological data’ of the locality in question. Implicit in this method of association is the assumption that Chinese toponyms are products of direct transliteration from indigenous names, if a coincidence between the two can be made. The identification of Sān Fó Qí (三佛齊) as Śrīvijaya is a classic example; apparently in use by Chinese texts produced during the Song period from the second half of the 10th century, the origins of this transliteration ‘have never been definitely explained.’ 55 Another example would be Láng Yá Xī Jĭao (狼牙犀角) as Langkasuka, believed by scholars to have existed as a kingdom on the northern most tracts of the Malay Peninsula as early as the 2nd century. 56 However, the association of Wāng’s ‘Dān Mǎ Xī’ with the SM’s ‘Temasek’ only has the presumption of an accurate transliteration between the two terms; no other data has been presented to support this transliteration. Moreover, Wāng clearly indicated that the toponym of Lóng Yá Mén was a description of the physical appearance of the locality itself instead of a transliteration of the indigenous name of the locality. Such etymological explanations are rarely provided for other Chinese toponyms in Chinese records, but nonetheless prove that not all Chinese toponyms are simple transliterations of an indigenous name. The arbitrary nature in the identification of Chinese toponyms of Southeast Asia is most aptly demonstrated by the glaring incongruity between Wáng’s description of the gate and modern interpretations of the Chinese source. To quote Wāng’s description of Lóng Yá Mén, 門以單馬錫番兩山相交若龍牙狀中有水道以間 55 56 Wolters, “Landfall on the Palembang Coast in Medieval Times,” Indonesia 20 (October 1975), p. 3. Wheatley, The Golden Khersonese, p. 252-67. 41 Translated, it means that the ‘gate’ (門 Mén) consisted of the ‘intersection’ (相交 Xiāng Jiāo) of ‘two hills’ (兩山 Liăng Shān) of the ‘Temasek barbarians’ (單馬錫番 Dān Mă Xī Fān) in ‘the shape of dragon’s teeth’ (龍牙狀 Lóng Yá Zhuàng) where ‘a waterway ran in the middle’ (中有水道以間 Zhōng Yŏu Shuĭ Dào Yĭ Jiān). Clearly, the presence of two hills is important to Wáng in his description of Lóng Yá Mén. The impression that Wāng may have wanted to render seems to be that of sailing into the open jaw of a dragon on the surface of the sea. It stands to reason that the search for the ‘two hills’ which matches this description would be central to the identification of Lóng Yá Mén at Singapore Island. However, the importance of this crucial feature apparently went unnoticed in almost every research pertaining to the whereabouts of the gate, even when they cited Wáng’s description in their various analyses. Incredibly, the ‘two hills’ were displaced arbitrarily in favour of another geological feature – rocks (石 Shí). Wheatley stated in 1961 that, ‘Dragon’s teeth’ was the name for two vertical pegs at the bow of a junk through which was carried the anchor cable, and hence was a term applied by Chinese sailors to natural features such as twin peaks or, in the present instance, to prominent rocks weathered out from strong, jointed granite, which formerly overlooked the western entrance to Keppel Harbour. According to Wheatley, two such ‘prominent rocks’ were described in John Thornton’s Oriental Navigation, published in London in 1703, as ‘a Bluff-Rock and smooth aloft, with Trees, and Grass,’ on the ‘South-side of the Straits,’ whereas the other was ‘a scragged Rock, shewing like a ruinated Wall’ that lay on the ‘North-side.’ Presumably, a 19th century rock feature known as ‘Batu Bĕlayar’ (Sail Rock), also known to the British as Lot’s Wife, was Thorton’s ‘scragged Rock’ in 1703 as it sat on the northern side of Keppel Harbour. Existing up till 1848, Batu Bĕlayar was apparently later demolished together with the southern ‘BluffRock’ to widen the Straits. 57 This rock is also known and documented by the Portuguese and in later European accounts as the ‘varela’ or ‘varella.’ 57 Ibid. 42 Wheatley was not the original proponent of this hypothesis, but rather, derived this idea from the work of earlier scholars such as Braddell and Hsü. They had in turn built upon the argument of Barnes, a former Chinese Protector in Singapore who wrote in 1911 that The words Lêng-gê-mûy (Amoy dialect) translated by [W. P.] Groeneveldt ‘the straits of Lingga’ mean dragon-teeth gate, strait or passage, and in the Amoy dialect ‘dragon-teeth’ is the name given to the two upright pegs in the bows of a ship through which the cable runs. 58 However, Braddell had stated that should Barnes’ explanation hold true, the name ‘Dragon’s teeth’ could then be applied to more than one straits as it was ‘the kind of name which Amoy sailors would give to any straits which passed close between two hills or prominences that resembled cable pegs,’ and not exclusively to Keppel Harbour. 59 More importantly, it is most unlikely that Wáng could not differentiate between geological features such as pillars of rock and hills, or sailed on the same type of vessel observed by the British to have ‘two vertical pegs’ at the bow five hundred years later. There is also no evidence which can prove that Wáng was himself an Amoy sailor who was thinking and recording in a southern Chinese dialect. Nonetheless, Braddell was the first of two scholars who sought to identify the ‘dragon’s teeth’ with two hills, citing Linschoten’s Reys Gheschrift Vande Navigatiën der Portugaloysers in Orienten account of the Keppel harbour in 1598. A more recent and accurate translation of this passage from the original text by Peter Borschberg differs from Braddell’s and is presented as follows, The mouth, or entrance, of this Strait lies between two high hills about one stone’s throw in breadth, and extends towards the east with the length of about one shot of a 12-pounder… Next to the entrance, at the foot of the mountain on the northern side, is a stone cliff which in its appearance resembles a pillar. This is generally named by all nations who sail this passage the Varella of the Chinese… 60 58 Barnes, “Singapore old Straits and New Harbour,” p. 25-31. Roland Braddell, “Lung-Ya-Men and Tan-Ma-Hsi,” p. 37-51; Hsü, “Singapore in the Remote Past,” p. 1-9. Borschberg, The Singapore and Melaka Straits. Violence, Security and Diplomacy in the 17th Century (Singapore: NUS Press, 2010), p. 32; see also Roland Braddell, “Lung-Ya-Men and Tan-Ma-Hsi, p. 25. 59 60 43 Linschoten’s late-16th century account comes closest to that of Wáng’s description of Lóng Yá Mén in the mid-14th century, thereby providing the best evidence that supports the location of gate the along Keppel harbour. Unfortunately, the credibility of Linschoten’s account is undermined by the fact that he did not sail past the straits himself, but merely compiled what Gibson-Hill describes as ‘the considered opinions of experienced men over the fifteen to twenty more years before Linschoten left Goa,’ which was as far as he ever went in the East. 61 Interestingly, the best known early European account which merged the two distinct geological features of mountains and rocks together when describing the gate came from Linschoten, who was evidently compiling different strands of information from disparate sources while residing in Goa. Lin Wo Ling was the only other scholar who searched for two hills that fit Wáng’s account of the gate in recent years. However, Lin reads Wáng’s description of the gate not as consisting of the ‘intersection of two hills,’ but as of a single hill with twin peaks which is aptly represented by Gunung (Mount) Lingga on Lingga Island. 62 Consequently, Gunung Sepintjan near Gunung Lingga is claimed by Lin to be the site of Bān Zú, the community which resided on a hill behind the gate as observed by Wáng. 63 Even though the physical characteristics of both Gunung Lingga and Sepintjan may appear to match Wáng’s account of the gate and Bān Zú respectively, his argument falls short on several counts according to Chung Chee Kit who criticizes Lin’s methodology and interpretation of sources. 64 Above all, both individuals did not realise that Lin’s reading of Wáng’s description of the gate was incomplete. While the twin peaks of Gunung Lingga is alluring to the seekers of the gate,no waterway could ever have run up the slopes through the peaks. Although Lin asserted that no archaeological evidence has been found thus far to support the existence of a 14th century settlement on Singapore Island, 61 Gibson-Hill, “Singapore Old Strait & New Harbour (1300-1870),” p. 11. Another possible site is identified by Borschberg and Huang Jianli to be located between Batam and Bintan (Borschberg, personal communication, 13 January 2012). 63 Lin, Long-ya-men Re-identified, p. 104-10, 123. 64 Chung Chee Kit, “Longyamen is Singapore: The Final Proof?” in Admiral Zheng He & Southeast Asia, ed. Leo Suryadinata (Singapore: ISEAS and International Zheng He Society Singapore, 2005), p. 161-6. 62 44 FIGURE 6. Kerimun Islands to Pedra Branca. Source: Mills, “Malaya in the Wu-Pei-Chih Charts,” p. 20-21, Plate II. 45 he did not present any similar evidence for a same settlement on Lingga Island either. 65 To date, no archaeological excavation on the scale of those in Singapore has been conducted on Lingga; neither have a similar amount or density of 14th century artefacts within an undisturbed soil stratum been discovered. Other individuals such as Gibson-Hill and Mills have chosen instead to locate Lóng Yá Mén by determining the location of the strait instead of what and where Wáng’s ‘Dragon’s Teeth’ were. While Gibson-Hill locates it along Keppel Harbour based on his reading of Chinese and European sources, Mills argued that the strait lie along the present-day Singapore Main Straits (south of Pulau Satumu) instead after studying 17th century Chinese maritime navigational instructions, especially those provided in the map of the Treatise on Armament Technology (武備志 Wǔ Bèi Zhì). 66 As such, Mills believed that this course through the Singapore Main Straits is ‘older’ than the ‘old strait’ (Keppel Harbour) of the Portuguese, Dutch and English, since the map relates to sailing directions given as early as 1433, the year of Eunuch Zhèng He’s last voyage. 67 But just like the others, Mills dispensed with the need to accommodate Wáng’s ‘two hills’ in the search for the gate. This confusion is further cofounded by the fact that Lóng Yá Mén as a toponym is problematic in itself. A century before Wāng wrote the Description, another toponym called Líng Yá Mén (凌牙門), was recorded within a Chinese gazetteer entitled Records of Foreign Peoples (諸蕃志 or Zhū Fān Zhì) written by Zhăo Rŭ Gua (趙汝适) in the 13th century. Zhăo instructed his readers to sail south to Líng Yá Mén from the Chinese port of Quanzhou (泉州 Quán Zhōu) for a little over a month following the winds during winter, ‘where one-third of the passing merchants (put in) before entering Sān Fó Qí (三佛齊),’ presumably Śrīvijaya. 68 Beginning with Willem Pieter Groeneveldt in 1876, some scholars have equated Zhăo’s Líng 65 Lin, Long-ya-men Re-identified, p. 97-127, 150. Mills, Ma Huan. Ying-Yai Sheng-Lan. ‘The Overall Survey of the Ocean’s Shores.’ 1433, ed. Feng Ch’eng-Chün (Cambridge: Cambridge University Press for the Haklyut Society, 1970), p. 236-302, 311-28. 67 Ibid., p. 1-34, 327-8. 68 Friedrich Hirth & Rockhill, Chau Ju-Kua: His work on the Chinese and Arab Trade in the twelfth and thirteenth Centuries, entitled Chu-fan-chi (St. Petersburg: Printing Office of the Imperial Academy of Sciences, 1911), p. 60. 66 46 Yá Mén to Wáng’s Lóng Yá Mén. 69 By studying the accompanying navigational instruction that used this particular toponym, Hsü also argued that Líng Yá Mén and Lóng Yá Mén are one and the same locality, a conclusion derived also from similarities in the phonetics and script of both toponyms. 70 There is, however, no other credible justification for this association besides similarities in phonetics and their approximate geographical location, which can be anywhere between the tip of the Malay Peninsula, Borneo and Sumatra; moreover the meanings of the two Chinese characters of Líng (凌) and Lóng (龍) are completely different; Lóng specifically means ‘dragon’ whereas depending on the context, Líng has many meanings, some of which are ‘overbearing,’ ‘messy,’ ‘rising,’ ‘passing’ or ‘approaching.’ On the other hand, since ‘Líng Yá’ sounds similar to ‘Lingga’ as well, some advocates of this equation argued that Líng Yá Mén (and possibly Lóng Yá Mén) lies nowhere in the immediate vicinity of Singapore Island but instead, refers to a passage either between Lingga and Singkep Islands (presently known as Selat Lima) or the Berhala Strait between Berhala Island and Sumatra (south of Singkep Island). 71 In recent years, it was even advanced that that there were two ‘Dragon’s Teeth Gates’ in this region: one at the southeast cape of Lingga Island referred to by the Song and Yuan texts (10th to 14th centuries), and another at the northern entrance of Selat Riau (between Batam and Bintan) according to the Ming and Qing texts (14th to 19th centuries). 72 THE ARCHAEOLOGICAL RECORD From the discussion above, it is evident that no definite conclusion can ever be reached concerning the accuracy of representing historical pre-colonial Singapore in either the SM’s account of Temasek-Singapura or that of Wāng’s accounts of Lóng Yá Mén and Bān Zú in his Description. This begs the question if scholars, without access to the epistemological context 69 Willem Pieter Groeneveldt, Notes on the Malay Archipelago and Malacca: Compiled from Chinese Sources (Batavia: -, 1876), p. 79-80. 70 Hsü, “Singapore in the Remote Past,” p. 3. 71 Rockhill, “Notes on the Relations and Trade of China,” p. 129-30; Roland Braddell, “Lung-Ya-Men and Tan-Ma-Hsi,” p. 42; Gibson-Hill, “Singapore Old Strait & New Harbour (1300-1870),” p. 37; Miksic, Archaeological Research, p.15. 72 Lin. Long-ya-men Re-identified. Singapore: 新加坡南洋学会出版, 1999; Munoz, Early Kingdoms, p. 188. 47 that governed the creation of Southeast Asian narratives or Chinese toponyms, can ever successfully identify these historical landmarks and corroborate them with localities and polities recovered on the archaeological record. Until an inscription bearing the name of ‘Temasek’ is found, it would be difficult to prove conclusively that pre-colonial Singapore was the site of Temasek-Singapura in the 14th century. Nonetheless, the heuristic issue involved in the study of pre-colonial Singapore as Temasek-Singapura can be phrased in terms of level of certainty. While it is impossible to attain a 100% certainty in any inference, it remains the duty of the scholar to weigh competing hypotheses and render an educated opinion, based on a set of explicit criteria, on the likelihood that pre-colonial Singapore may or may not have been Temasek-Singapura. 73 Therefore, one can compare the SM’s or Description’s accounts of pre-colonial Singapore – if indeed they were – with the existing archaeological record, and determine if the material cultural remains from the island are in fact consistent with what one would expect to find from Temasek-Singapura, Lóng Yá Mén and Bān Zú instead. From his survey of artefacts retrieved from the FTC, PHC and EMP sites, Miksic was able to conclude that ‘Singapore was not a major ceremonial centre’ which defines an orthogenetic city, as economic pursuits largely determined ‘its existence and character’ which was ‘closer to the heterogenetic end of the scale.’ 74 This can be seen from the remains of numerous imported goods such as trade ceramics, a distribution pattern which is shared, among other common features such as a dense population and a lack of monumental remains, with other heterogenetic settlements in pre-colonial Southeast Asia. 75 With the vast majority of recovered artefacts still awaiting analysis however, Miksic was more cautious in any attempt to reconstruct the urban layout of pre-colonial Singapore. Regardless, the analysis of archaeological evidence thus far has managed to show that the FTC, PHC and Empress Place (EMP) sites were ‘sites of very different activities.’ 76 73 Miksic, personal communication, 21 March 2012. Miksic, “14th-Century Singapore,” p. 52. Miksic, “Heterogenetic cities in premodern Southeast Asia,” p. 109-18. 76 Miksic, “14th-Century Singapore,” p. 41-54. 74 75 48 Nonetheless, the archaeological record provides direct evidence to support the existence of pre-colonial Singapore as a complex port-polity for the first time in the island’s historiography. Judging from the type and variety of material cultural remains, Miksic argues that pre-colonial Singapore appears to fit largely into the ‘port of trade’ model, a concept attributed to economic historian Karl Polanyi in 1957. The ‘port of trade’ is a ‘clearly defined complex of institutions and personnel,’ and ‘a geopolitical entity in which trade was an affair of the state and not a function of the economy.’ It was also usually i. an autonomous, specialized town, city or small state intended by policy for trade; ii. a transshipment point between different ecological regions; iii. a deliberately neutral, buffer zone; and iv. a site where often port officials instead of an indigenous group were involved in the trading and groups of foreign merchants resided in the port; institutional organization involved strict laws for the maintenance of peace, prosperity and security. The ‘port of trade’ would also be administered by native officials and professional organizations, and traders seldom or never had the freedom of the city or markets. Miksic concluded that the island fulfilled the first three conditions of the ‘port of trade’ model, while the paucity of information limited any assessment of the fourth criteria. 77 A TEST OF COMPLEXITY Clearly, the validity of the Temasek Paradigm can only be established on the basis of quantifiable and verifiable archaeological sources. The problem then lies not in the lack thereof, but the paucity of archaeological analyses from which meaningful data can be generated to compare with existing models and theories. It has already been observed that ‘sharp contrasts or affinities’ in the range and variety of ceramic artefacts in archaeological sites may potentially hold crucial information of settlement organization and inter-site relationships. 78 As part of this effort to process raw archaeological evidence into useful knowledge, the following chapter will attempt to elicit evidence of social complexity from the material cultural remains recovered from STA, the largest pre-colonial site in Singapore Island to date. 77 78 Ibid., p. 41-3, 53-4. Manguin, “Palembang and Sriwijaya,” p. 36-7. 49 III. ARTEFACT SOCIETY Objects are reluctant witnesses to the past; they have to be questioned carefully and closely if they are to provide accurate information. Chris Caple, 2006 79 This chapter presents the study of artefacts sampled from the STA site. Archaeological artefacts embedded literally in the landscape are the most obvious and important clues to be ‘read’ in the historical reconstruction of pre-colonial Singapore. Artefacts possess both a physical and metaphysical existence which can reveal information on the people and society which manufactured them. 80 As goods once produced, exchanged and consumed, these artefacts were ‘needed for making visible and stable the categories of culture.’ 81 Therefore, it can be said that artefacts once possessed a ‘social life’ in their tenure as ‘commodities’ since they were used in a variety of ‘social arenas.’ 82 Not only do they yield mute testimonies of social change, artefacts also serve as realms of social action through which social relationships are reproduced, negotiated, and transformed. 83 This information can be derived from data collected when artefacts are excavated, studied and interpreted through archaeological approaches and methods. The data collected thereafter can subsequently be used as the ‘straw’ needed to mould the bricks to reconstruct Singapore’s pre-colonial past. The importance of the archaeological record is ever more accentuated given that none of the pre-colonial monumental or structural remains, such as the Singapore Stone, the old wall, 79 Chris Caple, Objects. Reluctant Witnesses to the Past (London: Routledge, 2006), p. xv. James M. Skibo, “Pottery and People,” in Pottery and People. A Dynamic Interaction, ed. James M. Skibo & Gary M. Feinman (Salt Lake City: The University of Utah Press, 1999), p. 2. 81 Mary Douglas & Baron Isherwood, The World of Goods (London: Routledge, 1996), p. 38. 82 Arjun Appadurai, “Introduction: commodities and the politics of value,” in The social life of things. Commodities in cultural perspective, ed. Arjun Appadurai (Cambridge: Cambridge University Press, 1986), p. 3-15. 83 Li Min, “Fragments of Globalization: Chinese Blue-and-white Porcelain in Early Colonial Philippines.” Conference Paper. 18th Indo-Pacific Prehistory Association (IPPA) Congress. University of the Philippines, Manila. 20-26 March 2006, p. 1. 80 50 moat and ruins on the hill have survived colonial and post-colonial waves of urbanization. Clearly, the pre-colonial material cultural remains that were excavated and salvaged in recent years are now even more crucial primary sources for the study of pre-colonial Singapore. OBJECTIVES The analysis of a sample of STA artefacts in this chapter is motivated by the lack of knowledge concerning material culture and socio-economic aspects of society in pre-colonial Southeast Asia. As highlighted by Miksic, archaeological research in this region remains largely oriented towards the study of major orthogenetic pre-colonial urban centres largely defined by the presence of architectural, statuary and epigraphic remains. 84 Conversely, relatively few archaeological studies of household architecture and status goods which might reveal ‘evolving wealth differentials and social stratification in prehistoric and early historic period societies’ have been conducted. 85 Because of the paucity of knowledge on pre-colonial Southeast Asian material culture, the first and natural goal of this study is to provide a well-documented and categorized record of the sampled artefacts. This includes the identification of their material, forms and dimensions from which basic typologies of artefacts can be constructed. It is the author’s hope that this data will constitute a core body of information from which a library of pre-colonial Southeast Asian archaeological data can be built and shared freely for future research. The second goal of this study is to develop interpretations of the samples within a framework of analysis established by the earlier chapters. While Chapter 2 illustrates the problems and limitations of conventional historical sources in the study of pre-colonial Singapore, existing archaeological research coupled with ethnographic and ecological data is presented and consolidated within a historical 84 Miksic, “Classical Archaeology in Sumatra,” Indonesia 30 (October 1980), p. 50, 64-5; Miksic, “Heterogenetic cities in premodern Southeast Asia,” p. 106-9; Miksic, “Evolving Archaeological Perspectives on Southeast Asia, 1970-95,” JSEAS 26,1 (March 1995), p. 55. 85 Junker, Raiding, Trading and Feasting. The Political Economy of Philippine Chiefdoms (Honolulu: University of Hawai’i Press, 1999), p. 151. 51 context to inform the interpretations and deductions made in this study. Due to the limitations imposed on this thesis however, these sections are now enclosed in Appendices A, B and C. The reconstruction of the social-economic organization of pre-colonial Singapore society is the focus of this chapter and the third goal of this study. The significance of the abovementioned hypothesis in the study of pre-colonial Singapore should not be downplayed despite the seemingly obvious or self-evident nature of social stratification pre-modern societies in general; on the other hand, it would be more fallacious to write the history of the settlement’s existence and characterize its nature as a historical trading port-polity without any credible knowledge or evidence of its socio-economic and political organization represented by its material cultural remains. Research has demonstrated that pre-modern exchange and distribution of elite goods are linked to political exchange spheres. 86 The desire for goods and the distinction between ‘wealth’ (luxury) and ‘subsistence’ (utilitarian) goods formed the impetus for the organization of economic activities in early complex societies. 87 It is further advanced that once the elite established the mechanisms of trade, they will ‘inexplicably [and inevitably] expand to accommodate the demands of a broader sector of population.’ 88 Therefore, this thesis is a basic, but important step towards building that knowledge and supplanting abstract theoretical models of pre-colonial Singapore with concrete data based on verifiable archaeological evidence. METHODOLOGY The STA artefact assemblage is used in this thesis because excavation at that site is the ‘most extensive, longest-lasting, as well as the most systematically designed and executed archaeological project in the history of Singapore’ to date. 89 As the west lawn of the churchyard 86 Monica L. Smith, “The Role of Ordinary Goods in Premodern Exchange,” Journal of Archaeological Method and Theory (JAMT) 6,2 (June 1999), p. 112-3. 87 Elizabeth M. Brumfiel & Timothy K. Earle, “Specialization, exchange and complex societies: an introduction,” in Specialization, Exchange and Complex Societies, ed. Elizabeth M. Brumfiel & Timothy K. Earle (Cambridge: Cambridge University Press, 1987), p. 1-9. 88 Smith, “The Role of Ordinary Goods in Premodern Exchange,” p. 113. 89 Miksic & Lim Chen Shien, Archaeological Research on The Padang and in the St. Andrew’s Cathedral Churchyard: St. Andrew’s Cathedral Archaeological Research Project Progress Report Summary September 2003 – June 2004. ARI Working Paper, p. 3. 52 was due to be redeveloped into a two-level basement extension of the cathedral in December 2003, a team of archaeologists led by Professor Miksic of the National University of Singapore initiated an archaeological project to recover any artefacts at that site before they were lost in the redevelopment process. 90 The entire site occupied an area of approximately 2,400 square feet. (223m2) – northwest of the Cathedral, along North Bridge Road. Over a period of 28 weeks between September 2003 and June 2004, the team surveyed and excavated the site within 3 consecutive phases: augering, test-pit excavation and salvage excavation. A total of 138 archaeological units (excluding 5 additional units at the south lawn) had been excavated at the time of writing of the site’s only progress report in June 2004, but 190 units worth of artefacts were found to be stored at No. 6 Kent Ridge Road in January 2010. Each unit, measuring 2m x 1m, was excavated by stratigraphic layer and level either down to the point of artefact sterility or when the water table was reached at approximately 180 centimetres below the datum point (cmbd) located 15cm above ground level. The scale of the project yielded proportionally significant results: over 330,000 pieces of artefacts from the pre-colonial to colonial periods, estimated to weigh around 1 ton (0.91 metric ton), were recovered from 1009 metric tons (or 636 m3) of excavated soil. This amount is estimated to equal ‘the amount of artefacts recovered in all previous archaeological research in Singapore.’ 91 In other words, the STA site had an average artefact density of 519 artefacts per cubic meter; of this amount at least half can be ascribed to the Temasek-Singapura period. No other Singaporean site thus far has yielded an equivalent artefact density to date. From an archaeological perspective alone, the preliminary results from STA already present irrefutable evidence for a pre-colonial site of human habitation or activity on Singapore Island which, in consideration of the PHC, EMP and FTC sites, ‘extended all the way from the Singapore River to Stamford Road,’ where the old earthen walls of the city once ran. 92 Significantly, it was inferred from the results that a large reservoir of pre-colonial artefacts still lie deposited within 90 Pamphlet, St. Andrew’s Cathedral’s Quiet Places Project. A project to construct an extension to the existing church building. Singapore: The QPP Building Committee, St Andrew’s Cathedral, 2003. 91 Miksic, Archaeological Research on the Padang, p. 3. 92 Ibid. 53 the remaining unexcavated churchyard lawns and arguably, beneath the entire former colonial ‘civic district.’ The preservation of these colonial monuments from post-colonial urban development since 1965 has fortuitously led to the survival of the pre-colonial material cultural stratigraphic layer that lies beneath them. It is clear from the soil stratigraphic profiles, artefact yield as well as historical sources that three phases of human occupation exist on the STA site. The last two phases – colonial and post-colonial – can be seen as a single layer that starts from the top soil to approximately around 65 to 95 cmbd. This layer itself is made up of several layers of soil ranging between ‘dark brown clayey’ sand and ‘yellowish-red sandy’ clay. Artefacts recovered from this layer include colonial ceramic and glass vessels, bricks and pipe remains, rusting pieces of metal, colonial and post-colonial coinage, and even expended small arms cartridges from the Second World War. Directly beneath this stratum lies the pre-colonial layer approximately between 95 and 175 cmbd which generally consists of a single layer of very dark grey or black sand. Artefacts found within this layer are predominantly from the pre-colonial period and mainly consists of ceramic sherds: imported Chinese porcelain and stoneware as well as locally-made and imported earthenware. Chinese coins minted from the Song and Yuan dynasties, glass beads and bangles, worked stone, shellfish and bone remains as well as charcoal are included in this pre-colonial artefact assemblage as well. 93 Beneath this pre-colonial stratum lies a layer of coarse to fine white sand approximately between 175 and 180 cmbd. This stratum is artefactsterile and represents sand which one can expect of the beaches of prehistoric Singapore. The water-table, which also represents the adjacent sea-level, lies below 180 cmbd and therefore forms the lower limit of excavation on the STA site. 94 The primary unit of analysis on the STA site is the archaeological unit, but there is no clear basis for spatial comparison between units. They were excavated with the purpose of recovering as much pre-colonial material as possible and hence, did not represent any apparent 93 94 Ibid. See Appendix B. 54 form of spatial order. It is also impossible to determine the exact locations of any pre-colonial dwellings because no associated building foundations or structures were recovered; therefore comparisons on the household level cannot be made. For the purpose of this research, the site was hence divided along North-South and East-West axes into four arbitrary quadrants to compare the land areas and their associated artefacts instead: South-East (SE), South-West (SW), North-East (NE) and North-West (NW). A total of 20 units – five from each quadrant – were then selected at random from each quadrant. In order to maintain the integrity of archaeological interpretation from the pre-colonial artefact sample data, the units selected for analysis should meet two criteria: they must first possess an undisturbed pre-colonial stratum. This is best represented on the units’ respective soil profiles which illustrate uniform dark grey or black sandy strata without any sudden or deep intrusions from a different stratum. The second criterion requires at least 90 percent of the total artefacts recovered in situ within the unit must be of pre-colonial provenance for it to be considered sufficiently undisturbed for selection. Precolonial artefacts found in the colonial and post-colonial strata were therefore discounted; so were units which contain such pre-colonial artefacts but had a disturbed pre-colonial soil stratum. However, two units in the NW quadrant, STA 21 and STA 13, did not contain any precolonial artefacts at all despite having undisturbed pre-colonial soil strata. Of the 13 existing units in this quadrant, only three units met the two criteria. Among the remaining 10 units, three did not have an undisturbed pre-colonial soil layer; four units either did not have any excavation records or artefacts and no pre-colonial artefacts were recovered from the last three units. Consequently, STA 21 and STA 13 were randomly selected from these three units to fulfil the 20 units needed for a quantitatively viable comparison for this research. 95 As the STA assemblage consists mostly of numerous small fragments, all artefacts measuring at least 1cm x 1cm were selected for the present study. They were then tabulated to create a set of quantifiable statistical data from which subsequent analysis could be conducted. These artefacts were first grouped into general material classes: ceramics, metals, charcoal, 95 See Appendices B & C. 55 South East West North FIGURE 7. STA Site Map (2003-4) indicating the thesis’s division of the site into four quadrants for comparative analysis. 96 Source: Miksic & Lim, Archaeological Research on The Padang and in the St. Andrew’s Cathedral Churchyard, p. 6-7. 96 Not all STA units are illustrated on this map. 56 stone, wood, resin and bone. As the largest class of artefacts in the sample pool, ceramic sherds were divided further into categories of earthenware, stoneware and porcelain (Figures 8-49). These sherd categories were in turn subdivided into several distinct varieties each and identified by the vessel parts they represent. Metal artefacts were also subdivided into distinct varieties (Figures 50-5), whereas the remaining artefact classes (Figures 56-57) were not subdivided as little or no meaningful variation exists due to their small numbers. Besides the count and weight, physical characteristics of the artefacts such as their thickness, width and diameter were measured with a pair of vernier callipers to the accuracy of two decimal places. Other aspects of manufacture such as surface treatment, decorations and any other observable features were recorded, sometimes with the aid of a magnifying glass at 2x magnification. Only the count, weight and observable features were recorded for non-ceramic artefacts. 97 Digital images of all artefacts were taken and stored in a DVD-ROM as a pictorial archive and, together with the artefact data tables, included in this thesis for reference. The integrity of the respective layer, unit and quadrant is maintained for every single artefact to allow spatial distribution to be analysed. It is only through this laborious but necessary process of analysis that accurate identification, provenance and dating of these artefacts can subsequently be deduced. Since these artefacts were not found within well-defined structural boundaries such as walls or building foundations, one can only assume that they were deposited as primary or secondary domestic refuse in middens. The fragmentary form of most ceramics found together with the assortment of ashes, bones and shells across all pre-colonial Singapore sites have led Miksic to conclude that artefacts in most of these sites were probably thrown away as rubbish. 98 The study of midden artefacts, instead of those deposited within clearly-defined residential compounds, does not discount their utility as indicators of patterned human behaviour. The analysis of both household and neighbourhood middens have been found to be more ‘economical, meaningful and representative’ than items on floor structures which tend to 97 See Appendix C. Miksic, “Research on Ceramic Trade within Southeast Asia and between Southeast Asia and China,” in Southeast Asian Ceramics. New Light on Old Pottery, ed. Miksic (Singapore: Southeast Asian Ceramic Society, 2009), p. 99. 98 57 provide skewed examples. 99 Intra-site comparisons of secondary refuse have also been used to evaluate differences in ‘consumption, status, ethnicity, or activities between different areas of the community.’ 100 Ethnoarchaeological studies of middens in the modern village of Dalupa in the Philippines have even determined the transport distances of household, local and communal middens of a densely populated village to be within the range of 6, 11 and 35 metres respectively. 101 Based on this ethnoarchaeological data, it is reasonable to assume that the STA site is large enough to accommodate all of the three abovementioned midden levels. This allows us to interpret the STA site as a village-size residential area within the entire pre-colonial settlement in Singapore and hence, capture variations within the artefact distribution pattern which may highlight evidence of social stratification. It should be noted that this interpretative analysis relies on the assumption that a single village (or kampong), neighbourhood, ward or district of a city would contain households of different status groups. On the other hand, the alternative residential pattern would resemble a modern city like Singapore where different status groups tend to live isolated from each other. As Miksic has shown however, this was not the case in ancient Sumatra and hence, unlikely in pre-colonial Singapore as well. 102 It would, of course, be preferable to prove the validity of this assumption by excavating a larger area, but this option is not presently feasible due to technical and financial constraints. As ceramics form the bulk of the artefact sample, the analysis conducted here lies primarily on the study of their production, use and eventual deposition in middens across the STA site. Margaret E. Beck has highlighted three potential applications for midden ceramics which are useful for the archaeological interpretation of the STA site. Firstly, as many more 99 Brian Hayden & Aubrey Cannon, “Where the garbage goes: Refuse disposal in the Maya Highlands,” Journal of Anthropological Archaeology (JAA) 2,2 (June 1983), p. 117, 130. 100 James M. Bayman, “Shell ornament consumption in a Classic Hohokam Platform Mound Community Center,” Journal of Field Archaeology (JFA) 23,4 (Winter 1996), p. 403-20; Eric Blinman, “Potluck in the Protokiva: Ceramics and ceremonialism in Pueblo I villages,” in The Architecture of Social Integration in Prehistoric Pueblos, ed. William D. Lipe & Michelle Hegmon (Cortez: Crow Canyon Archaeological Center, 1989), p. 113-24, cited in Margaret E. Beck & Matthew E. Hill Jr, “Rubbish, Relatives, and Residence: The Family Use of Middens,” JAMT 11,3 (September 2004), p. 298. 101 Beck & Hill, “Rubbish, Relatives, and Residence,” p. 327. 102 Miksic, “Urbanization and Social Change in Sumatra,” Archipel 37 (1989), p. 1-29. 58 vessels are broken and discarded than preserved intact in archaeological sites, midden ceramics may be a broader and more representative of the complete ceramic assemblage. Secondly, midden ceramics can be used to reconstruct activities as they were deposited relatively near the activity areas which generated them. Thirdly, the occupation span and population of the site can be estimated from the accumulation of frequently used and discarded ceramics such as cooking and storage vessels. 103 Archaeological ceramics are also an important source in the study of socio-economic complexity of human societies, which is the subject of this thesis. One theory on the origins of pottery takes into consideration the role of decorative motifs in various media in the social and symbolic elaboration of social organization in early human societies. Prudence M. Rice has suggested that decorated pottery in these societies could have been ‘prestigious vessels’ for displaying and communicating messages of social identity. 104 The use of clay is viewed by Brian D. Hayden as a ‘putative software horizon’ and part of an emergence of prestige technologies in early cultures, where pottery was used primarily in prestige display contexts such as competitive and reciprocal feastings. 105 When pottery became more commonplace, specific vessel shapes and type became reserved only for high-status households to reinforce social distinctions. 106 As human societies become more complex over time, so too did the relationship between ceramics and social structure. Dean E. Arnold has demonstrated that the social status of both the profession and their products became dependent on population pressure, the sexual division of labour and demand for pottery. Potters who are men and forced from agriculture into craft-specialization for their livelihood generally occupy a low social position until they make high-value items, whereas women potters tend to enjoy higher social status because their income supplements primary subsistence activities. However, potters may enjoy high social status if demand for pottery involves ‘mythical, religious or social structural symbols.’ 107 103 Margaret E. Beck, “Midden Ceramic Assemblage Formation: A Case Study from Kalinga, Philippines,” American Antiquity (AA) 71,1 (January 2006), p. 27. 104 Prudence M. Rice, “On the Origins of Pottery,” JAMT 6,1 (March 1999), p.13, 37. 105 Brian Hayden, “Practical and prestige technologies: the evolution of material systems,” JAMT 5,1 (March 1998), p. 26, 28-30. 106 Kevin J. Vaughn, “Households, Crafts, and Feasting in the Ancient Andes: The Village Context of Early Nasca Craft Consumption,” Latin American Antiquity (LAA) 15,1 (March 2004), p. 61-88. 107 Dean E. Arnold, Ceramic Theory and Cultural Process (Cambridge: Cambridge University Press, 1985), p. 196-8. 59 Ceramics were also used in pre-colonial Southeast Asia as a measure and indicator of social rank distinction. Stylistic, technological and distribution analyses of decorative earthenware pottery found in the Philippines have suggested that some decorative styles could have been iconographic representations of inter-chiefly elite alliances or shared identity in the late prehistoric and protohistoric period. 108 A Khmer inscription dated to 674 compares the ruler to the potter’s wheel which is constantly in motion and a source of creation, a metaphor used to express ‘the majesty of the elite.’ 109 As trade routes were gradually established between Southeast Asian and Chinese markets over the centuries, Southeast Asian elites found a new source of high-quality and beautiful ceramics which they did not have the technology to produce. 110 Imported Chinese ceramics, such as small vessels with lids, came to be considered a luxury suitable for use only by the Khmer nobility in the 11th century. 111 According to Barbara Harrisson, Chinese porcelain was also the first choice of ‘sultans and datos’ in Borneo who ‘endeavoured to imitate the prevailing attitude in China of collecting the best porcelain as a status symbol’ among many other export goods dispersed into the region around 1500. 112 Chinese porcelains, among other exotic luxury goods, also became ‘key symbols of social status and political power’ together with indigenously manufactured prestige goods for the Philippine chiefly elite. 113 As late as the 19th century, it was observed in Borneo that some expensive Chinese jars decorated with dragon motifs could only be owned by Dayak chiefs who also protected the jars with expensive textiles. 114 Hence, it is hoped that the analysis of gathered archaeological data will illustrate this similar trend of socio-economic complexity in pre- 108 Elizabeth A. Bacus, “Styles of Alliance?: Decorated Earthenwares in Late Prehistoric and Protohistoric Philippines Polities,” in Earthenware in Southeast Asia, ed. Miksic (Singapore: Singapore University Press, 2003), p. 39-51. 109 Roxanna Maude Brown, The Ceramics of South-East Asia. Their Dating and Identification (Singapore: Oxford University Press, 1988), p. 41, footnote 1, cited in Miksic, “Research on Ceramic Trade,” p. 96. 110 Derek Heng Thiam Soon, Sino-Malay Trade and Diplomacy from the Tenth through the Fourteenth Century (Athens: Ohio University Press, 2009), p. 149. 111 Miksic, “Research on Ceramic Trade,” p. 97. 112 Barbara Harrisson, “The Ceramic Trade across the South China Sea,” in Southeast Asia-China Interactions, compiled by Geoff Wade (Kuala Lumpur: MBRAS, 2007), p. 494. 113 Junker, Raiding, Trading and Feasting, p. 183. 114 Hugh Low, Sarawak: Its Inhabitants and Productions. Being notes during a residence in that country with His Excellency Mr. Brooke (London: Frank Cass & Co., 1968), p. 271-2, cited in Miksic, “Research on Ceramic Trade,” p. 99; see also photo of a Balawit aristocrat in Kalimantan posing with his prized Chinese ‘dragon jars’ taken by Tom Harrisson circa 1945, in John S. Guy, Oriental Trade Ceramics in South-East Asia. Ninth to Sixteenth Centuries. With a Catalogue of Chinese, Vietnamese and Thai Wares in Australian Collections. (Singapore: Oxford University Press, 1990), p. vii. 60 colonial Singapore from the distribution pattern of deposited midden ceramics across the STA site. LIMITATIONS The results of this analysis are limited by several factors. To quote Carla M. Sinopoli, ‘any analysis is only as good as the quality of the data used in it.’ 115 A significant portion of excavation data however was lost during the conduct of the project and the six-year administrative lapse that followed until the start of the present thesis in 2009. The project only had a small amount of time to survey a large plot of land, and had to recruit untrained volunteers from the public for manpower which inevitably involved a high turn-over rate in personnel. As a result, the excavation data and soil profiles for a large proportion of units – including the ones selected for research – were neither recorded consistently nor in significant detail; artefacts recovered from the salvage phase had no excavation data at all and hence could not be used for this study. Besides the brief preliminary report compiled in the final stages of the project in 2004, no detailed excavation report was ever published. As a result, very little or no information on the matrix of the artefacts, associated archaeological features or ecofacts, important in the identification of natural or cultural formation processes on the STA site, were recorded. The analysis and subsequent interpretation of the artefact samples are therefore limited to the finds themselves. It should also be noted that the stratification of sandy soil is difficult to interpret. Since sand particles do not adhere to each other, sandy strata are easily disturbed by natural processes such as root growth and disturbances from soil erosion. Organic material, particularly small particles of charcoal which usually act as key indicators of human activity and chronology, often migrate downwards through percolation of rainwater, blurring the boundaries between strata in the process. 116 115 Sinopoli, Pots and Palaces. The Earthenware Ceramics of the Noblemen’s Quarter of Vijayanagara (New Delhi: American Institute of Indian Studies, 1993), p. 31. 116 Miksic, personal communication, 23 December 2011. 61 The analysis of the artefact sample is also limited by the nature and conditions of the finds themselves; for a start, most, if not all sampled artefacts were recovered in fragments. Although some complete ceramic vessels were recovered from the site, none were represented in the 20 units selected for research. As such, the study is hampered by the need for analysis at the level of the sample pool rather than the individual vessel. It is difficult, if not impossible, to identify different vessel forms present, let alone measure or quantify certain variables which can be used to infer the varieties and possible typological sequence of ceramic vessels that were present during the lifespan of the pre-colonial settlement. Moreover, artefacts like ceramics are adaptable and may not be used for what they were originally made for, therefore not all vessels in the sample can be assumed to have identical functions. A ceramic vessel may even be a skeuomorph which possess physical features ‘which are irrelevant or even detrimental to its purpose or manufacture, but are present because they were relevant to prototypes made in a different material’ such as metal. 117 However, vessel forms and functions can still be estimated from distinctive sherd features and corroborated with the existing local and regional archaeological and ethnographic record to infer the possible intended use. The present study is also limited by the absence of historical, ethnographical and ethnoarchaeological data concerning material culture of pre-colonial Singapore. As such, the present thesis draws analogies and derives its interpretations from comparisons with existing data from the rest of the region. While these comparisons are not claimed to depict an exact image of the past, they serve as a useful estimate from which a historically- and archaeologically-accurate model of pre-colonial Singapore society can be reconstructed than left literally buried in the ash heap of history. RESULTS In total, 8681 individual pieces of artefacts weighing 53,362.5g were sampled from the undisturbed pre-colonial strata from 20 archaeological units at the STA site. The quantitative 117 Clive Orton, Paul Tyers & Alan Vince, Pottery in Archaeology (Cambridge: Cambridge University Press, 1994), p. 28-9. 62 distribution of all sampled artefacts with respect to their units and quadrants of provenance is presented in Table I. The quantitative distribution of the different types and varieties of ceramic vessels, estimated from vessel rim sherds identified and matched within the artefact sample pool, with respect to their units and quadrants of provenance is presented next in Tables II, III and IV. The quantitative distribution of the different varieties of metals is presented in Table V. Table VI gives the total counts and weights of classified ceramic sherds for the STA site as a whole. Table VII illustrates the distribution of the general ceramic types represented in percentages by weight and count of all sample ceramic sherds between the 4 quadrants, whereas Table VIII shows the same data but uses only ceramic rim sherds and includes also vessel count and proportion as well. Tables VI, VII and VIII represent tabulating methods adopted from William G. Solheim II’s 1959 research on pottery sherds. 118 SETTLEMENT DATING AND LIFESPAN Figure 61 presents the approximate ages of only well-documented STA artefacts as elaborated in Appendix A. With the exception of Chinese copper cash coins (Figures 53-4), most of the artefact dates suggest that the STA site was occupied primarily between the 13th and 14th centuries, with the site reaching its peak of settlement towards the end of the 13th century as highlighted in red. No earlier material cultural strata dating to the previous centuries were found in excavations on Singapore Island to date as well. The dating of pre-colonial settlement in Singapore via material cultural means therefore reaffirms that provided by various Southeast Asian, Chinese and European historical sources mentioned in the previous chapters. The absence of artefacts typically found at Southeast Asian archaeological sites dated before and after the 13th and 14th centuries also supports the relatively brief existence of the STA settlement within this period. These artefacts include Chinese ceramics such as the earlier ‘Changsha’ (长沙 Cháng Shā) or Yue (越 Yuè) ceramics of the Tang and Northern Song dynasties, as well as that of later blue-and-white ceramics from the Ming and Qing dynasties. 118 William G. Solheim II, “The Use of Sherd Weights and Counts in the Handling of Archaeological Data,” Current Anthropology (CA) 1,4 (July 1960), p. 325-9. 63 Table I: Quantitative Distribution of Sample STA Artefacts Earthenware NE Quadrant STA 26 STA 56 STA 59 STA 69 STA 76 Total (Quadrant) NW Quadrant STA 9 STA 13 STA 13A STA 18A STA 21 Total (Quadrant) SE Quadrant STA 5 STA 7 STA 51 STA 55 STA 58 Total (Quadrant) SW Quadrant STA 6 STA 6A STA 16 STA 42 STA 50 Total (Quadrant) Stoneware Porcelain Metal Stone Charcoal Wood Resin Bone Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight 188 343 202 51 197 454g 505g 407g 85.5g 369g 49 43 110 48 113 488g 424.5g 1810g 303g 898g 17 21 60 29 26 85.5g 66.5g 465g 165g 123g 36 9 22 11 153g 8g 75g 65g - - - 3g 14.5g 2g 1g - - - - 1 - 1g - 981 1820.5g 363 3923.5g 153 905g 78 301g - - - 20.5g - - - - 1 1g Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight 620 77 57 - 1476.5g 171g 112g - 280 43 128 - 3726g 626g 944g - 95 4 24 - 1255.5g 17g 179g - 49 39 3 - 391g 160g 18g - 4 2 - 558g 35g - - 33g 2g - - - - - 1 - 2g - 754 1759.5g 451 5296g 123 1451.5g 91 569g 6 593g - 35g - - - - 1 2g Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight 1377 3951.5g 705 1213g 8 12.5g 50 160g 531 647g 384 96 49 31 32 6448g 1193g 1232g 518g 168g 106 1610.5g 55 260g 9 177g 11 64g 1 0.5g 77 85 1 825g 422g 6g 1 2 1 49g 357g 89g - 45.5g 3g - 19 2 - 2g 4g - - - - - 2671 592 9559g 182 178 1270g 4 495g - 48.5g 21 6g - - - - 5984g 2112g Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight Count Weight 91 7 348 19 189 116.5g 9g 604g 24.5g 565g 47 58 108 46 851 249g 449g 865g 636g 11120g 13 59.5g 22 211.5g 36 320.5g 20 129.5g 128 1338.5g 1 25 1 7 1g 142g 1g 15g 2 337g - 1g 2g 2g 2g - - 3 3 2g 5g 6 - 3g - 654 1319g 1110 13,319g 219 2059.5g 34 159g 2 337g - 7g - - 6 7g 6 3g 1425g - 111g 21 6g 6 7g 8 6g Summary Total (Artefact) Total (Count) Total (Weight) 5060 10,883g 2516 32,097.5g 677 6528g 381 2299g 12 8681 53,362.5g 64 Table II: Quantitative Distribution of Earthenware Vessels in Sample STA Artefacts based on Rim Sherds Earthenware Coarse-Tempered Medium-Tempered Fine-Paste NE Sherd Vessel Sherd Vessel Shed Vessel Weight Weight Weight Quadrant Count Count Count Count Count Count STA 26 14 8 120g 3 3 4g STA 56 24 21 68g STA 59 1 1 2g 7 7 20g 6 6 13.5g STA 69 5 5 13g 2 2 2g STA 76 4 3 52g 7 7 19g 2 2 3.5g Total 5 4 54g 57 48 240g 13 13 23g (Quadrant) NW Sherd Vessel Sherd Vessel Shed Vessel Weight Weight Weight Quadrant Count Count Count Count Count Count STA 9 74 51 434g 7 6 16g STA 13 STA 13A 13 5 45g 3 1 8g STA 18A 5 5 19g 2 1 3g STA 21 Total 92 61 498g 12 8 27g (Quadrant) SE Sherd Vessel Sherd Vessel Shed Vessel Weight Weight Weight Quadrant Count Count Count Count Count Count STA 5 2 2 16g 70 47 619g 35 29 70g STA 7 31 29 172g 36 36 48g STA 51 1 1 2 STA 55 3 3 41g 2 1 0.5g STA 58 1 1 36g 7 6 62g 56 28 71g Total 3 3 52g 112 86 896g 129 94 189.5g (Quadrant) SW Sherd Vessel Sherd Vessel Shed Vessel Weight Weight Weight Quadrant Count Count Count Count Count Count STA 6 13 10 29g 2 2 2g STA 6A 1 1 2g STA 16 2 2 25g 20 16 144g 5 5 4g STA 42 4 3 3g STA 50 12 7 38g Total 2 2 25g 50 37 216g 7 7 6g (Quadrant) Summary Total 10 9 131g 311 232 1850g 161 122 245.5g (Variety) 65 Total (Unit) High-Fired Sherd Vessel Sherd Vessel Weight Weight Count Count Count Count 17 11 124g 24 21 68g 2 2 4g 16 16 39.5g 7 7 15g 13 12 74.5g 2 2 4g 77 67 321g Sherd Vessel Sherd Vessel Weight Weight Count Count Count Count 81 57 450g 16 6 53g 7 6 22g - - - 104 69 525g Sherd Vessel Sherd Vessel Weight Weight Count Count Count Count 107 78 705g 67 65 220g 1 1 2g 5 4 41.5g 64 35 169g - - - 244 183 1137.5g Sherd Vessel Sherd Vessel Weight Weight Count Count Count Count 15 12 31g 1 1 2g 27 23 173g 4 3 3g 12 7 38g - - - 59 46 247g Total (Samples) 2 2 4g 484 365 2230.5g Table III: Quantitative Distribution of Stoneware Vessels in Sample STA Artefacts based on Rim Sherds Brittle "Mercury Jar" Buff "Mercury Jar" NE Sherd Vessel Sherd Vessel Weight Weight Quadrant Count Count Count Count STA 26 STA 56 STA 59 STA 69 1 1 5g STA 76 1 1 2g Total 2 2 7g (Quadrant) NW Sherd Vessel Sherd Vessel Weight Weight Quadrant Count Count Count Count STA 9 2 1 2g 3 3 126g STA 13 STA 13A 3 1 50g STA 18A 4 2 32g STA 21 Total 9 4 84g 3 3 126g (Quadrant) SE Sherd Vessel Sherd Vessel Weight Weight Quadrant Count Count Count Count STA 5 10 9 174g 2 2 23g STA 7 2 2 28g 3 3 8g STA 51 STA 55 STA 58 Total 12 11 202g 5 5 31g (Quadrant) SW Sherd Vessel Sherd Vessel Weight Weight Quadrant Count Count Count Count STA 6 1 1 4g STA 6A STA 16 2 1 104g STA 42 STA 50 7 4 14g Total 8 5 18g 2 1 104g (Quadrant) Stoneware Total (Unit) Brittle (General) Buff (General) Purple-ware Shed Vessel Sherd Vessel Sherd Vessel Sherd Vessel Weight Weight Weight Weight Count Count Count Count Count Count Count Count 4g 1 1 4g 1 1 4g 6 6 320g 7 7 324g 1 1 1 1 3g 1 1 8g 3 3 16g 2 2 86g 4 4 92g 7 7 180g 4 4 93g 12 12 424g - - - 3 2 47g 8 7 449g - - - 31 22 311g 10 9 261g 18 524g 23 16 706g Shed Vessel Sherd Vessel Sherd Vessel Sherd Vessel Weight Weight Weight Weight Count Count Count Count Count Count Count Count 1 1 100g 7 7 336g 1 1 8g 21 20 641g 4 4 88g 1 1 58g 10 10 182g 50g 2 2 50g 2 2 1 1 21g 1 1 21g 4 4 171g 11 11 424g 2 2 66g 34 33 894g Shed Vessel Sherd Vessel Sherd Vessel Sherd Vessel Weight Weight Weight Weight Count Count Count Count Count Count Count Count 1 1 2g 1 1 45g 3 3 51g 5g 1 1 5g 1 1 1 1 6g 3 3 33g 6 5 143g 1 1 1g 1 1 1g 5 3 61g 23 10 1180g 35 17 1255g 8 6 70g 28 15 1263g - - - 46 381g 59 45 2560g 2 2 66g 121 19 16 27 1455g Total (Samples) Summary Total (Variety) 18 Shed Vessel Sherd Vessel Sherd Vessel Sherd Vessel Weight Weight Weight Weight Count Count Count Count Count Count Count Count 1 1 30g 8 7 449g 14 12 607g 50g 3 1 2 1 17g 6 3 49g - 66 94 3579g Table IV: Quantitative Distribution of Porcelain Vessels in Sample STA Artefacts based on Rim Sherds Porcelain Green-ware White-ware Blue-and-White-ware NE Sherd Vessel Sherd Vessel Shed Vessel Weight Weight Weight Quadrant Count Count Count Count Count Count STA 26 11 5 30g STA 56 4 4 25g 2 1 5g 1 1 0.5g STA 59 5 5 17g 8 6 26g 5 1 19g STA 69 2 2 9g 2 2 7g 3 3 34g STA 76 5 5 28g 2 2 19g 1 1 5g Total 27 21 109g 14 11 57g 10 6 58.5g (Quadrant) NW Sherd Vessel Sherd Vessel Shed Vessel Weight Weight Weight Quadrant Count Count Count Count Count Count STA 9 17 11 302g 12 9 38.5 STA 13 STA 13A STA 18A 5 5 70g 1 1 1g STA 21 Total 22 16 372g 13 10 39.5g (Quadrant) SE Sherd Vessel Sherd Vessel Shed Vessel Weight Weight Weight Quadrant Count Count Count Count Count Count STA 5 20 14 207.5g 4 4 21g 1 1 6g STA 7 7 7 62g 5 5 18.5g STA 51 1 1 6g 4 1 17g STA 55 1 1 2g 1 1 15g STA 58 1 1 0.5g Total 29 23 276g 14 11 58.5g 2 2 21g (Quadrant) SW Sherd Vessel Sherd Vessel Shed Vessel Weight Weight Weight Quadrant Count Count Count Count Count Count STA 6 3 3 30g 1 1 6g STA 6A 4 4 18g 2 2 4.5g STA 16 2 2 28g 3 3 17g 3 1 8g STA 42 1 1 46g 1 1 6g 2 2 6g STA 50 9 8 290g 19 15 79.5g 7 3 12g Total 19 18 412g 26 22 124g 12 6 26g (Quadrant) Summary Total 97 78 1169g 67 54 279g 24 14 105.5g (Variety) 67 Total (Unit) Badly-Fired-ware Sherd Vessel Sherd Vessel Weight Weight Count Count Count Count 11 5 30g 7 6 30.5g 9 7 100g 27 19 162g 3 3 3g 10 10 53g 8 8 52g 12 10 103g 63 48 327.5g Sherd Vessel Sherd Vessel Weight Weight Count Count Count Count 8 8 38g 37 28 378.5g 1 1 13g 1 1 13g 3 3 7g 9 9 78g 12 12 58g 47 38 469.5g Sherd Vessel Sherd Vessel Weight Weight Count Count Count Count 25 19 234.5g 10 7 23g 22 19 103.5g 5 2 23g 2 2 17g 1 1 0.5g 10 7 23g 55 43 378.5g Sherd Vessel Sherd Vessel Weight Weight Count Count Count Count 2 2 4.5g 6 6 40.5g 1 1 3g 7 7 25.5g 9 7 52.5g 17 13 105.5g 4 2 14g 8 6 72g 13 12 48g 48 38 440.5g 29 24 122g 86 70 684g Total (Samples) 63 53 306g 251 199 1859.5g Table V: Quantitative Distribution of Metals in Sample STA Artefacts Metals Objects Slag NE Quadrant STA 26 STA 56 STA 59 STA 69 STA 76 Total (Quadrant) NW Quadrant STA 9 STA 13 STA 13A STA 18A STA 21 Total (Quadrant) SE Quadrant STA 5 STA 7 STA 51 STA 55 STA 58 Total (Quadrant) SW Quadrant STA 6 STA 6A STA 16 STA 42 STA 50 Total (Quadrant) Coinage Count Weight Count Weight Count Weight Count Weight 2 3 1 - 12g 2g 4g - 34 6 21 11 141g 6g 71g 65g - - 36 9 22 11 153g 8g 75g 65g 6 18g 72 283g - - 78 301g Count Weight Count Weight Count Weight Count Weight 9 - 9g - 40 38 2 - 382g 158g 17g - 1 1 - 2g 1g - 49 39 3 - 391g 160g 18g - 9 9g 80 557g 2 3g 91 569g Count Weight Count Weight Count Weight Count Weight 54 51 - 696g 241g - 23 49 1 129g 198g 6g - - 77 100 1 825g 439g 6g 105 937g 73 333g - - 178 1270g Count Weight Count Weight Count Weight Count Weight 6 - 31g - 1 17 6 1g 108g 13g 2 1 1 3g 1g 2g 1 25 1 7 1g 142g 1g 15g 6 31g 24 122g 4 6g 34 159g Total (Samples) Summary Total (Variety) Total (Unit) 126 995g 249 1295g 68 6 9g 381 2299g Table VI: Totals, Percentages, and the Number of Pieces Per Gram of Classified Ceramic Sherds Weight % Number % Pc/g Earthenware Stoneware 10,883g 32,097.5g 22 64.8 5060 2516 61.3 30.5 0.46 0.08 Porcelain 6528g 13.2 677 8.2 0.1 Total 49,508.5g 100 8253 100 0.17 Table VII: Totals, Percentages, and the Number of Pieces Per Gram of Classified Ceramic Sherds by Area NE Quadrant NW Quadrant SE Quadrant SW Quadrant Weight % Number % Pc/g Weight % Number % Pc/g Weight % Number % Pc/g Weight % Number % Pc/g Earthenware Stoneware 1820.5g 3923.5g 27.4 59 981 363 65.5 24.2 0.54 0.09 1759.5g 5296g 20.7 62.3 754 451 56.8 34 0.43 0.08 5984g 9559g 33.9 54.1 2671 592 77.5 17.2 0.45 0.06 1319g 13,319g 7.9 79.8 654 1110 33 56 0.5 0.08 69 Porcelain 905g 13.7 153 10.2 0.17 1451.5g 17.1 123 9.3 0.08 2112g 12 182 5.3 0.09 2059.5g 12.3 219 11 0.11 Total 6649g 13.4 1497 18.1 0.23 8507g 17.1 1328 16.1 0.16 17,655g 35.7 3445 41.7 0.2 16,697.5g 33.6 1983 24 0.12 Table VIII: Totals, Percentages, and the Number of Pieces Per Gram of Classified Ceramic Rim Sherds by Area NE Quadrant NW Quadrant SE Quadrant SW Quadrant Weight % Number % Pc/g Vessels % Weight % Number % Pc/g Vessels % Weight % Number % Pc/g Vessels % Weight % Number % Pc/g Vessels % Earthenware Stoneware 321g 524g 27.4 44.7 77 18 48.7 11.4 0.24 0.03 67 18 50.4 13.5 525g 706g 30.9 41.5 104 23 59.8 13.2 0.2 0.03 69 16 56.1 13 1137.5g 894g 47.2 37.1 244 34 73.3 10.2 0.21 0.04 183 33 70.7 12.7 247g 1455g 10.4 61 59 46 30.9 24.1 0.24 0.03 46 27 32.2 18.9 70 Porcelain 327.5g 27.9 63 39.9 0.19 48 36.1 469.5g 27.6 47 27 0.1 38 30.9 378.5g 15.7 55 16.5 0.15 43 16.6 684g 28.7 86 45 0.13 70 49 Total 1172.5g 15.3 158 18.5 0.13 133 20.2 1700.5g 22.2 174 20.3 0.1 123 18.7 2410g 31.4 333 38.9 0.13 259 39.4 2386g 31.1 191 22.3 0.08 143 21.7 FIGURE 8. Coarse-tempered earthenware brick or tile sherd, STA 5. FIGURE 9. Coarse-tempered earthenware ‘trivet’ or structural ornamentation sherds, STA 5. 71 FIGURE 10. Coarse-tempered ‘Saw-Tooth with Flame’ impressed earthenware pottery body sherds, STA 13A. FIGURE 11. Medium-tempered earthenware pottery rim sherds, STA 9. 72 FIGURE 12. Medium-tempered earthenware pottery knobbed lid sherds, STA 50. FIGURE 13. Medium-tempered decorated (paddleimpressed) earthenware pottery body sherds, STA 9. 73 FIGURE 14. Medium-tempered slipped earthenware pottery rim and body sherds, STA 16. FIGURE 15. High-fired black-burnished earthenware pottery rim sherds, STA 59. 74 FIGURE 16. High-fired black-burnished carved earthenware pottery body sherds with ‘Circle-and-Ring’ motif, STA 5. FIGURE 17. High-fired mercury-jar type earthenware base sherd, STA 5. 75 FIGURE 18. Fine-paste earthenware pottery rim sherds, STA 58. FIGURE 19. Fine-paste earthenware pottery rim sherds with ‘apple-green’ glaze, STA 55. 76 FIGURE 20. Fine-paste decorated (Punctated ‘Dot-and-Ring’ motif bordered with double rib lines) earthenware pottery body sherds, STA7. FIGURE 21. Fine-paste earthenware kendi spout sherds, one with ‘rifled’ interior, STA 58. 77 FIGURE 22. [Left to Right] Type A, B and C stoneware pottery inclusion patterns, STA 5. FIGURE 23. [Left to Right] Buff versus brittle stoneware mercury-jar rim sherds, STA 5. 78 FIGURE 24. [Left to Right] Buff versus brittle stoneware mercury-jar base sherds, STA 51. FIGURE 25. [Left to Right] Brittle versus buff stoneware pottery rim sherds with luted lugs, STA 5. 79 FIGURE 26. [Left to Right] Brittle versus buff stoneware pottery body sherds, STA 5. FIGURE 27. Buff stoneware basin rim sherds, STA 50. 80 FIGURE 28. Buff stoneware basin rim sherd with folded ‘PieCrust’ fringe, STA 6. FIGURE 29. Brittle stoneware pottery body sherds with glaze, STA 7. 81 FIGURE 30. Brittle stoneware pottery base sherd, STA 5. FIGURE 31. Buff stoneware pottery base sherds, STA 50. 82 FIGURE 32. Buff purple-ware pottery rim sherd, STA 5. FIGURE 33. Brittle purple-ware pottery rim sherd with luted lug, STA 7. 83 FIGURE 34. Assorted green-ware porcelain pottery rim sherds, one with an exterior lotus leaf relief [left], STA 5. FIGURE 35. Green-ware porcelain bowl sherd with a foliated rim and interior motif, STA 7. 84 FIGURE 36. Green-ware porcelain bowl base sherd, STA 51. FIGURE 37. Assorted green-ware porcelain pottery base sherds, STA 50. 85 FIGURE 38. Assorted green-ware porcelain pottery body sherds, one with ‘stylized onion-skin medallion’ [bottom-left], STA 7. FIGURE 39. Green-ware porcelain ‘gacuk’, STA 76. 86 FIGURE 40. White-ware porcelain cup sherd with foliated rim, STA 5. FIGURE 41. White-ware porcelain covered box base sherd, STA 26. 87 FIGURE 42. Iron-spotted white-ware porcelain pottery body sherds, STA 9. FIGURE 43. Blue-and-white porcelain cup rim sherds with exterior chrysanthemum and foliage motifs and interior classic scrolls and key-fret motifs, STA 50. 88 FIGURE 44. Blue-and-white porcelain bowl rim sherd with an exterior foliage motif and interior classic scrolls motif, STA 9. FIGURE 45. Blue-and-white porcelain bowl rim sherds with an exterior stylised lotus petals panelled motif and interior chrysanthemum flower with foliage motif, STA 16. 89 FIGURE 46. Octagonal blue-and-white vase neck or body sherds with an exterior panelled classic scrolls motif and luted interior, STA 7. FIGURE 47. Blue-and-white porcelain cup rim and base sherds with an exterior foliated motif, interior classic scrolls motif and calligraphic illustration of ‘longevity’ on the interior centre medallion, STA 50. 90 FIGURE 48. Misfired or degraded blue-and-white porcelain pottery body sherds, STA 59. FIGURE 49. Misfired or degraded porcelain pottery base sherd, STA 18A. 91 FIGURE 50. Metal slag, STA 7 and 26. FIGURE 51. Metal nails, fishing hooks and other objects, STA 26. 92 FIGURE 52. Bronze bangle fragment, STA 18A. FIGURE 53. ‘明道元宝’ (Míng Dào Yuán Băo), Northern Song dynasty coin, 1032-3 CE, STA 42. 93 FIGURE 54. Either ‘崇宁通宝’ (Chóng Níng Tōng Băo), ‘崇 宁元宝’ (Chóng Níng Yuán Băo), or ‘崇宁重宝’ (Chóng Níng Zhòng Băo), Northern Song dynasty coin, 1102-1106 CE, STA 50. FIGURE 55. Sri Lankan coin, circa 13th-14th centuries CE, STA 16. 94 FIGURE 56. Crucible-like stone with a rectangular depression on one side, STA 58. FIGURE 57. Stone structural ornamentation fragments with ornate carved motif on surface, STA 13A. 95 FIGURE 58. Sand Grain Sizing Folder. Source: Forestry Suppliers Incorporated, 2009. FIGURE 59. Charts for Estimating Proportions of Mottles and Coarse Fragments. Source: Munsell Soil-Color Charts, 2009 Year Revised/2010 Production. 96 FIGURE 60. Mohs’ Hardness Scale and Substitutes. Source: Rice, Pottery Analysis. A Sourcebook (Chicago: The University of Chicago Press, 1987), p. 356. Approximate Age of Sampled STA Artefacts Glazed 'Majapahit' Fine-paste Earthenware Mercury Jars Longquan Greenware Ceramics Iron-spotted Qingbai Ceramics Dehua Whiteware Ceramics Blue-and-white Ceramics Chinese Copper Cash Coins Sri Lankan Copper Coin 10 11 12 13 14 15 16 Century (CE) Approximate Lifespan Approximate Material Cultural Peak FIGURE 61. Approximate age of STA Site derived from a selection of sampled STA artefacts. 97 export ceramics of the 15th century are also missing from the samples as well. This absence clearly indicates that this pre-colonial settlement had existed before the ‘Ming gap,’ a period of Chinese ceramics shortage, especially that of blue-and-white ceramics, but abundance of Southeast Asian ceramics observed across maritime archaeological sites and shipwrecks in the region during the early Ming dynasty (1368-1488). 119 The apparent anomaly of 11th-12th century Chinese copper cash coins (Figures 53-4) in a 13th-14th century settlement can be explained by the longevity of coin circulation in trade and exchange due to their intrinsic value as a trade metal, as well as their propensity to be removed from this circulation through accidental loss. This is no better illustrated by the cargo of the 13th century Quanzhou wreck off the coast of Fujian (福建 Fú Jiàn) province which held 504 Chinese copper coins and five iron coins of both Tang and Song dynasties. 120 Hence, coins on their own can neither be used to accurately date archaeological sites nor indicate any geographical pattern of commercial exchange. 121 Nonetheless, since they are the oldest datable artefacts among the samples, their presence suggests that the STA site cannot be significantly older than the dates they were minted: 1032-1106. It should be noted that this distribution pattern of Northern Song and Sri Lankan coins without any native Southeast Asian coin issues at the STA site is mirrored at Kota Cina, a mid-12th-14th centuries cosmopolitan port-settlement site in Sumatra, where Northern Song coins were again found with ‘copper coins of late twelfth and early thirteenth century from Srī Laṅka.’ 122 STA: A COMPLEX PRE-COLONIAL SETTLEMENT SITE Based on artefact analyses and deductions as well as the dates provided in Appendix A, it is evident that pre-colonial STA was a single-phase habitation site which was wellestablished by the start of the 13th century as seen in Figure 61. Together with the rest of the 119 Brown, The Ming Gap and Shipwreck Ceramics of Southeast Asia. Towards a Chronology of Thai Trade Ware (Bangkok: The Siam Society under Royal Patronage, 2009), p. 23-31, 69-70. 120 Heng, Sino-Malay Trade and Diplomacy, p. 167. 121 Robert S. Wicks, Money, Markets, and Trade in Early Southeast Asia. The Development of Indigenous Monetary Systems to AD 1400 (Ithaca: Cornell Southeast Asia Program, 1992), p. 15-6. 122 Ibid., p. 234. 98 settlement, the STA site attained its material cultural peak in terms of artefact variety by that century’s end and went into decline from the mid- to late-14th century. One can either infer that the site was abandoned or at least, ceased to conduct foreign trade, by the start of the 15th century. There is no material cultural or soil stratigraphic evidence which indicates any human activity before or after this phase of settlement at the STA site until the early colonial period in the 19th century. The absence of a significant amount of coinage in terms of artefact proportion as compared to the PHC site also rules out the possibility that large-scale commercial activities occurred at STA. 123 As seen in Table I, only earthenware, stoneware and porcelain sherds are consistently found at the archaeological unit, quadrant and site levels; no unit or quadrant contains any artefacts which radically distinguish them from each other. Neither do they contain any archaeological features akin to high concentration of glass artefacts or slag at FTC, nor metal objects and slag at PHC which suggests the conduct of specialized glass-recycling and metal-working activities at those sites as discussed earlier. 124 This is clear evidence of site homogeneity and, given the largely domestic functions of ceramic artefacts, use of the STA site as a residential quarter by the inhabitants of pre-colonial Singapore. I. Ceramics and Complexity The investigation of socio-economic complexity at the STA site formally begins with the conduct of Solheim’s test of site homogeneity. 125 This involves a test of difference between total ceramic count and weight percentages, as presented in Tables VII and VIII, of all four quadrants. According to Solheim, both percentages will be the same if the spatial distribution of ceramics at the STA site is completely random, as any random sample should be representative of the whole which has the same total count and weight percentages (100%) as illustrated in Table VI. Therefore, the consistent difference between the two percentages at all 123 Miksic, “14th-Century Singapore,” p. 52. Miksic, “Intrasite Analysis of 14th-Century Singapore,” in Uncovering Southeast Asia’s Past. Selected Papers from the 10th International Conference of the European Association of Southeast Asian Archaeologists, ed. Elizabeth A. Bacus, Ian C. Glover & Vincent C. Pigott (Singapore: NUS Press, 2006), p. 337-41. 125 Solheim, “The Use of Sherd Weights and Counts in the Handling of Archaeological Data,” p. 327-8. 124 99 four quadrants presented in Table VII clearly indicates that the site is not spatially homogenous. In other words, there is a non-random variation in the spatial distribution of ceramics within the site. This variation is again repeated in the spatial distribution of rim-sherds (and hence, estimated vessel counts) in Table VIII, thereby confirming the non-random scatter of ceramics at each of the four quadrants and invalidating the null hypothesis. Evidence of this non-random variation is reinforced by another statistical test commonly used in archaeology known as the ‘Chi-Squared Test.’ This test measures the difference between the observed frequency of a quantifiable category and its subsequent expected frequency in the context of pure chance. The greater the difference, the higher the probability that the observed frequency and hence, variation in spatial distribution, is a nonrandom occurrence. 126 In this instance, the estimated numbers (observed frequencies) of vessels from the rim sherds of the three ceramic categories are tested for the probability of a relationship between the four quadrants of the STA site. The total number of ceramic sherds cannot be used because of its inaccuracy in revealing relative proportions of ceramic varieties at the STA sites; for instance, the presence of 100 earthenware sherds will not equate to 100 earthenware vessels, nor should it suggest that there are necessarily more vessels within 100 earthenware sherds than 50 stoneware or 10 porcelain sherds. As illustrated in Table IX, if 133, 123, 259 and 143 ceramic vessels is estimated from the artefact sample pool in the NE, NW, SE and SW quadrants respectively, one would expect (in a completely random situation) that the estimated 365 earthenware, 94 stoneware and 199 porcelain ceramic vessels should be divided among the NE, NW, SE and SW quadrants according to the ratio of 133:123:259:143. Therefore, the respective expected frequencies of estimated vessels for each ceramic variety should be calculated according to this ratio. To reiterate the null hypothesis, H0: There is no difference in the distribution of estimated vessels for all three ceramic categories across the four quadrants. 126 Stephen Shennan, Quantifying Archaeology (Edinburgh: Edinburgh University Press, 1988), p. 65-76. 100 Table IX: Chi-Squared Test of Significance Between Ceramic Variety and Spatial Distribution Observed Frequency NE Quadrant Expected Frequency Observed Frequency NW Quadrant Expected Frequency Observed Frequency SE Quadrant Expected Frequency Observed Frequency SW Quadrant Expected Frequency Total Estimated Earthenware Vessels Estimated Stoneware Vessels Estimated Porcelain Vessels 67 18 48 Total 133 73.57 19 40.22 69 16 38 68.23 17.57 37.2 183 33 43 143.67 37 78.33 46 27 70 79.32 20.43 43.25 365 94 199 123 259 143 101 658 Table X: Chi-Squared Value Calculation Category 1 2 3 4 5 6 7 8 9 10 11 12 Oi 67 18 48 69 16 38 183 33 43 46 27 70 Ei 73.57 19 40.22 68.23 17.57 37.2 143.67 37 78.33 79.32 20.43 43.25 (O i -E i ) -6.57 -1 7.78 0.77 -1.57 0.8 39.33 -4 -35.33 -33.32 6.57 26.75 (O i -E i )2 43.16 1 60.53 0.59 2.46 0.64 1,546.85 16 1,248.21 1,110.22 43.16 715.56 (O i - E i )2/E i 0.59 0.05 1.5 0.01 0.14 0.02 10.77 0.43 15.94 14.00 2.11 16.54 *χ2 = 62.1 12 *�2 = � � (�� −�� )2 �� , where k is the number of categories, Oi is the number of cases in category i, Ei is the expected number of cases in category i, and χ2 is the symbol representing ‘chi’-squared. 127 127 Ibid., p. 67. 102 For the purpose of comparison to H0, the alternative hypothesis H1 is simply as follows, H1: There is a difference in the distribution of estimated vessels for all three ceramic categories across the four quadrants. By archaeological convention, the significance level (α) selected for this test will be 0.05, hence α = 0.05. 128 In other words, there can only be a one in 20 chance or less for the observed results (i.e. Tables IX) to occur such that the difference in the distribution of the three ceramic categories (i.e. H1) is significant. Given that Table IX has four rows and three columns of crossclassified data, the number of degrees of freedom (v) is given by, v = (the number of rows in table - 1)(the number of columns in table - 1) = (4-1)(3-1) = 3 x 2 = 6. Given that α = 0.05 and v = 6, the tabulated Chi-Squared value (��2 ) is 12.5916. 129 According 2 ≥ ��2 ; since the Chi-Squared value calculated at Table to the test, H0 must be rejected if ����� 2 2 > ��2 , which means that H0 is clearly rejected. Even if the significance ) is 62.1, ����� X (����� level is taken at its most conservative level, where α = 0.001 (one in 1000 chance), ��2 is still 2 . In other words, there is a high only 22.458 (where v = 6) and significantly less than ����� probability that a definite relationship exists between ceramic variety and the difference in spatial distribution of their respective estimated vessels at the STA site. One reason behind this relationship appears to be the location of ceramic artefacts at the STA site as illustrated by Tables I to IV. As seen in Table I, the majority of earthenware, stoneware and porcelain ceramics are concentrated in the SE and SW quadrants which are closer to the Singapore River, as opposed to the NE and NW quadrants. This pattern is expected, as the NE and NW quadrants are directly adjacent to Stamford Road where the old wall demarcating the maximum extent of the pre-colonial settlement once stood. Hence, one would expect a greater level of human activity in areas further away from settlement boundaries. The 128 129 Ibid., p. 52. Ibid., p. 336-7. 103 difference in the amount of ceramics between the northern and southern quadrants however is not too great; there is still a significant amount of estimated vessels of all ceramic varieties at both the NE and NW quadrants as shown in Tables II to IV. This suggests in turn that the site was likely to be densely occupied as all available space within the confines of the old wall was maximized. Another probable reason which can explain this relationship lies in the relative values between the varieties of earthenware, stoneware and porcelain ceramics. 130 This differentiation in the relative values between artefacts and its correlation to spatial distribution was observed by Miksic in his preliminary analysis of the STA site. Although he thought that some parts of the STA site corresponded to a ‘low-energy-level area on the fringe of the settlement into which refuse was sometimes deposited,’ Miksic also acknowledged that this hypothesis was ‘somewhat contradicted by the discovery of some objects of higher-than-normal value’ among the artefact assemblage. 131 While not explicitly mentioned, these high-value goods were implied to be primarily imported Chinese high-fired stoneware and porcelain ceramics. As mentioned earlier in the chapter, porcelain ceramics were highly-demanded by the various ruling elites of Southeast Asian polities. This association between high social status and imported Chinese ceramics in pre-colonial Singapore was first suggested from their recovery at the FTC site, which is generally considered to be the ceremonial and elite precinct of the settlement. This was supported by the recovery of a large number of ‘rare and unique’ items including the gold ornaments found in 1926, a rare porcelain pillow and compass as well as glass beads and bangles on the site. 132 At the same time, the absence of green-ware porcelain ceramics with the raised double-fish motif, ‘typical of Song-Yuan collections from other sites in the region’ from the FTC site is perceived by Miksic to be an anomaly which had a ‘measure 130 The term ‘value’ is used in this thesis to mean the monetary or material worth of a commodity assigned in the market. In the absence of detailed information on exchanges and equivalencies, it is difficult, if not impossible to determine the exact cost of such exchanges to both the merchant and consumer in pre-colonial Southeast Asia. This problem is exacerbated by the absence of a standardized monetary system of exchange in pre-colonial Southeast Asia as well as the fact that different societies may often assign different values to the same objects based on localized cultural factors and idiosyncratic preferences of both the merchant and consumer. One can however have a fair idea of relative costs between artefacts within the same category by comparing estimaties of production cost between the respective sub-categories, i.e. earthenware, stoneware and porcelain ceramics. 131 Miksic, “Intrasite Analysis of 14th-Century Singapore,” p. 344. 132 Ibid., p. 337-9. 104 of social significance.’ 133 On the other hand, numerous fragments of this ceramic variety were recovered at the other Singaporean sites on the ‘plain’ area by the river, thereby suggesting that ‘some kind of intentional cultural selection process’ had occurred which this ceramic variety was ‘not used by the elite, but were available to the other categories of inhabitants.’ 134 Building upon this dichotomy of elite and non-elite ceramic consumption pattern, Heng argued that the larger proportions of blue-and-white ceramics on FTC than the other plain area sites also indicates ‘clear disparities in the ceramics that the residents could afford,’ and subsequently concluded that the FTC residents were the most affluent in the pre-colonial settlement. Ceramics of ‘low-unit-value’ were also thought to be redistributed to its ‘economic dependencies’ believed to be the Riau Islands. 135 As various varieties of porcelain were recovered across all pre-colonial sites in Singapore however, there is a need for the refinement of porcelain as a measure of relative value and social status in order to elicit meaning in the non-random variation observed in the spatial distribution of ceramic artefacts at the STA site. This can be achieved through an estimation of their prices as trade goods which has been documented to some degree in historical Chinese records. It should be noted first and foremost that before the Ming period (1368-1644), porcelain was ‘rarely used as state gifts made in exchange between the Chinese emperors and Southeast Asian rulers’ as indicated by records of imperial tributary exchange during the 10th-12th centuries in the 宋会要 (Sòng Huì Yào or ‘Manuscript of the Song Dynasty’) and 宋史 (Sòng Shǐ or ‘History of the Song Dynasty). 136 Documentation of private Chinese maritime enterprise in the same records however record that ‘fine porcelain’ was one of the goods used to ‘trade and exchange for aromatic products, rhinoceros horns, coral, amber, pearl, iron, turtle shells, cornelian, giant clam, foreign clothes, ebony, sappanwood and the like.’ 137 133 Miksic, “14th-Century Singapore,” p. 49. Miksic, “Recently Discovered Chinese Green Glazed Wares of the Thirteenth and Fourteenth Centuries in Singapore and the Riau Islands,” p. 231. 135 Heng, Sino-Malay Trade and Diplomacy, p. 182. 136 Grace Wong, “A Comment on the Tributary Trade between China and Southeast Asia, and the Place of Porcelain in this Trade, During the Period of the Song Dynasty in China,” in Chinese Celadons and other Related Wares in Southeast Asia, p. 74-88. 137 Ibid., p. 94. 134 105 Porcelain was also ‘given the same emphasis’ as other trade items such as silk, cloths and brocades in terms of revenue during that period. 138 Zhào Rǔ Kuò (趙汝适), the Inspector of Foreign Trade at the port of Quanzhou in Fujian province, wrote in his Description of the Barbarous Peoples (諸蕃志 Zhū Fán Zhì) sometimes in the mid-13th century that while ‘porcelain-ware’ (瓷器 Cí Qì) in general was traded in Champa, Cambodia, Sumatra, the Malay Peninsula and the Philippines, white-ware ceramics were specifically traded in the eastern Indonesian islands such as the Celebes and Moluccas, green-ware ceramics in Borneo and both green- and white-ware ceramics in Java. 139 Wheatley stated that green-ware ceramics were especially in demand as they were ‘popularly believed to change colour’ upon contact with poisoned food. 140 By the mid-14th century however, Wāng recorded in his Description that blue-and-white porcelain ceramics had become a popular import product in almost half of all identified Southeast Asian polities. In her comparison of both Zhào’s and Wāng’s records, Grace Wong observed that blue-and-white porcelain had become a far more valuable commodity than the erstwhile reputable green-ware ceramics within the span of a century. Polities which were known to import blue-and-white ceramics were ‘all big and comparatively rich,’ and never appeared to consume ‘crude bowls’ (粗碗 Cū Wǎn) at the same time. 141 Conversely, ‘crude bowls’ were often imported together with green-ware ceramics (粗碗青器 Cū Wǎn Qīng Qì) instead. 142 Although the exact prices of the specific porcelain varieties were not recorded by either Zhào or Wāng, chapter 179 of the Collected Statutes of the Ming Dynasty (明會典 Míng Huì Diǎn) apparently provides an estimated price index of various articles – including porcelain ceramics – which was fixed during the first year of the Hong Wu (洪武 Hóng Wǔ) reign of the Ming dynasty in 1368; the index valued a ‘big porcelain vase’ at one ‘guan’ (貫 Guàn, official 138 Wong, “Chinese Blue-and-White Porcelain and Its Place in the Maritime Trade of China,” in Chinese Blue & White Ceramics, p. 51. 139 Hirth & Rockhill, Chu-fan-chi, p. 49, 53, 61, 69, 78, 156-8, 160-162, 140 Wheatley, “Geographical Notes on some Commodities involved in Sung Maritime trade,” JMBRAS 32,2 (June 1959), p. 83. 141 Wong, “Chinese Blue-and-White Porcelain and Its Place in the Maritime Trade of China,” p. 56-60. 142 Ibid., p. 59. 106 paper money) and two guan for every 10 pieces of porcelain plates and bowls. 143 Another list in Chapter 113 of the same volume apparently provides a more precise breakdown of prices of different porcelain varieties during the Hong Zhi (弘治 Hóng Zhì) reign from 1433-1505. Due to an inflation of 20 times the prices in 1368 during this period, a blue-and-white ‘wine jar’ (玉 壶春瓶 Yù Hú Chūn Píng or ‘Jade-like wine-bottle vase,’ also known as ‘yuhuchun’) was valued at 1500 guan, blue-and-white vase and plate at 500 guan each and a blue-and-white bowl at 300 guan. On the other hand, a celadon plate was worth 150 guan, the same as a ‘bottle’ which was presumably a small stoneware storage vessel akin to mercury or purple-ware jars, whereas a celadon bowl was priced at 100 guan. 144 Hence, blue-and-white ceramics are clearly much more expensive than green-ware ceramics by the 15th century. It is now possible to reconstruct a hierarchy of ceramic value at the STA site based on the relative costs of imported Chinese ceramics estimated from the historical data listed above. Since blue-and-white ceramics were popular trade goods during the 13th and 14th centuries, it is reasonable to assume that blue-and-white ceramic imports (i.e. Figures 43-7) to Southeast Asia would have cost the most during that period, to be followed by white-ware (i.e. Figures 40-2) and green-ware (i.e. Figures 34-9) ceramics. ‘Crude’ ceramics, most likely referring to misfired porcelain and stoneware vessels (i.e. Figures 22-33, 48-9) in general, would have cost the least among imported ceramics. Given the disparities in technology and economics in production between high-fired Chinese imports and low-fired Southeast Asian earthenware ceramics as discussed in Appendix A, it is also fair to assume that earthenware ceramics would cost the least among the three ceramic categories. The same disparity between imported fine-paste ceramics (i.e. Figures 18-21) and locally-made coarse- and medium-tempered ceramics (i.e. 143 One ‘Guan’ usually refers to one string of ‘cash’ which normally consists of 1000 coins and equivalent to one tael of pure silver. The unusually high prices for ceramics reflected the hyperinflation of paper currency in China during the transitionary period from the Yuan to the Ming dynasty in 1368. Fiscal mismanagement under the Yuan government resulted in a series of inflations between 1260-1276, 1280-87 and 1299 made paper currency worthless and taken off circulation by 1356. Reinstated under the early Ming government to defray expenditures, it was devalued and made worthless again not long after circulation in the late 14th century. 144 Ibid., p. 61-2. 107 Figures 8-14) was likely to have rendered the former much more expensive than the latter as well. If the locations and relative values of the different ceramic categories and varieties are indeed the reasons behind the non-random variation in the spatial distribution of ceramics artefacts, one should expect to find proportionately more artefacts of lower value than those with a higher value. This can be seen in Table I which illustrates that earthenware, stoneware and porcelain sherds generally have the largest to smallest proportion of ceramic artefacts respectively not only for the entire site, but at the quadrant and unit level as well. One would also expect to find a large amount of low-value artefacts widely distributed across the entire site, but a much lower amount of high-value ceramics concentrated in smaller and distinct areas as well. This is certainly the case as presented in Tables II to IV. The NE and SW quadrants appear to be residential areas of relatively high affluence, as the estimated few blue-and-white porcelain vessels are mostly concentrated in these two quadrants, along with the majority of estimated porcelain and stoneware vessels in general. On the other hand, most of the mediumtempered and fine-paste earthenware vessels are concentrated in the NW and SE quadrants, but a relatively low proportion of the estimated stoneware porcelain vessels. Significantly, not a single blue-and-white vessel was found in the NW quadrant. As the affluence of an individual household at the STA site should correspond directly to the amount and value of ceramics it consumes and subsequently discards, the non-random variation in spatial distribution of artefacts should also reflect the disparity in the abilities of each quadrant in general to consume ceramics of differing values. The more high-valued ceramics the households of one quadrant consumes, the more wealth or goods it has to purchase or exchange for them. Therefore, this correlation between artefacts and social status would have divided the STA site into stratified residential quarters of differing levels of affluence during the pre-colonial period. Given that the quadrants of higher affluence (NE and SW) are diagonally-staggered with quadrants of low affluence (NW and SE), boundaries between 108 residential zones of differing affluence at the STA site appear to be less of socio-economic segregation than more along neighbourly lines in pre-colonial times. It is also interesting to note that despite the recovery of 292 green-ware sherds from which 78 vessels could be reconstituted, not a single sherd bears the raised double-fish motif. If this motif is truly an indicator of non-elite consumption, then it may suggest that the STA site, as a whole, was an elite residential quarter contiguous to FTC. This indicates in turn some form of distancing between the affluent households whom resided closer to the ruling classes at FTC in the northern sector of the pre-colonial settlement, and the craftsmen, merchants and quite possibly foreign traders who conducted their professions and commercial activities by the Singapore River to the south. II. Craft-Specialization and Complexity Social complexity in pre-colonial Singapore is not only illustrated by the disparity in material-cultural wealth, but also by evidence of craft-specialization and exchange represented by ceramic, metal and stone artefacts. 145 By definition, specialization is ‘the existence of individuals who produce goods or services for a broader consumer population’ and hence, involves economic differentiation and interdependence. Hence, craft-specialization should also be an identifable feature of stratified societies besides economic, social and political differentiation as well as a measure of heterarchical and hierarchical power structures. 146 As a continuum, specialization ranges between the ideal ‘Domestic Mode of Production,’ where the division of labour is limited to age and sex differences within the family, and the modern industrial economy with an extremely complex division of labour. 147 However, neither the location of the pre-colonial Singaporean economy in this continuum, nor the variables of specialization, such as the affiliation of craft-specialists, the nature of the products, the intensity of specialization and the scale of the production unit in this economy has ever been 145 See Appendix A. Lisa Kealhofer & Peter Grave, “Land Use, Political Complexity, and Urbanism in Mainland Southeast Asia,” AA 73,2 (April 2008), p. 202. 147 Brumfiel & Earle, “Specialization, exchange and complex societies: an introduction,” p. 5. 146 109 clearly identified. Some scholars believe that a ‘part-time level of intensity’ is consistent with ethnographic evidence from Southeast Asia. The seasonal demands of agricultural activity, especially wet-rice cultivation during the six-month rainy season, generally confines Southeast Asian craft-production to the dry season as low humidity-levels makes ‘crafts involving pyrotechnology’ – such as pottery-making and metal-working – more convenient and successful. 148 The ecological context of pre-colonial Singapore however does not accommodate this version of craft-specialization. Unlike the intermediate tropical zone of mainland Southeast Asia and Java which has alternating rainy and dry seasons, the island and most of the Malay Archipelago – southern and eastern Sumatra, Malaya, Borneo, west Java and eastern Philippines – is located within the equatorial zone characterized by the absence of prolonged dry periods and consistent rainfall. 149 Moreover, the ground of this equatorial belt primarily consists of clays of poor fertility. 150 Usually found as yellow to red leached lateritic formations (latosols), they are rich in iron and aluminum but generally low in plant nutrients and organic matter. 151 While the soil does support a seemingly evergreen forest, nutrients falling as leaves are more quickly broken down and recycled through forest biomass rather than building up topsoil suitable for agriculture. 152 The island’s soil would be as infertile as the adjacent Malay Peninsula which consisted of a thin layer of humus covering a thick substratum of laterite and is ‘almost always mediocre.’ Large alluvial deposits are present along valleys, the foot of hills and along the coasts but their primarily contribution to the fertility of the area is noted to be ‘less in their composition than in the fact that the network of waterways flowing through them can be tamed and used by man.’ 153 This is in stark contrast to the fertile soils found in the intermediate tropical zone. An arc of volcanoes, formed between Aceh in northern Sumatra and 148 Joyce C. White & Pigott, “From Community Craft to Regional Specialization: Intensification of Copper Production in Pre-state Thailand,” in Craft Specialization and Social Evolution: In Memory of V. Gordon Childe, ed. Bernard Wailes (Philadelphia: The University Museum of Archaeology and Anthropology, University of Pennsylvania, 1996), p. 159. 149 Peter Boomgaard, Southeast Asia. An Environmental History (Santa Barbara: ABC-CLIO, 2007), p.18. 150 Reid, “Humans and Forests in Pre-Colonial Southeast Asia,” in Nature and the Orient. The Environmental History of South and Southeast Asia, ed. Richard H. Grove, Vinita Damodaran, Satpal Sangwan (Delhi: Oxford University Press, 1998), p. 108. 151 Bellwood, Prehistory of the Indo-Malaysian Archipelago (Honolulu: University of Hawai’i Press, 1997), p. 11-2. 152 Reid, “Humans and Forests,” p. 108. 153 Jacq-Hergoualc’h, The Malay Peninsula, p. 7. 110 Manado in northeast Sulawesi, give rise to a fertile crescent of alkaline volcanic soils in the islands; 154 on the mainland, alluvial soil in river deltas annually enriched by flooding and the wet season, such as those of the Mekong, Chaophraya, Salween and Irrawaddy rivers, are the most fertile. 155 Hence Singapore Island, like many other places within the equatorial zone, cannot support a dense, agriculture-based population. According to the records of the Lóng Yá Mén and Bān Zú toponyms in Wāng’s Description which were associated to Singapore Island by Wheatley, the fields of the island are ‘barren and there is little padi,’ whereas the soil ‘is poor and grain scarce.’ 156 The inability of the island to sustain agriculture is again demonstrated when the British arrived in 1819. Food was apparently scarce despite the fact that island was settled by the Temenggong ‘some years before with a few score of followers.’ 157 The first colonial soldiers who arrived with Raffles and Farquhar could neither find food, nor buy any from its inhabitants who lived in ‘four or five small huts with six or seven coconut trees planted beside them.’ The locals only ate ‘young shoots, dried fish, fried sago and occasionally rice,’ while the Orang Laut brought fish they speared for the Temenggong. It was noted however that shell-fish ‘abounded on the edge of the sand, floating in pools, and could be collected in basketfuls,’ but that did not appear to be the staple diet of the inhabitants either. 158 In the absence of a seasonal agrarian economy, craft-specialization in pre-colonial Singapore thus appears to be conducted at a full-time intensity and most likely sustained the entire settlement through the exchange and trade of their respective products. As Miksic suggests, the population of 14th century Singapore was likely ‘divided among various occupations involving the import of raw materials and the production and possibly export of finished products.’ 159 The viability of this economic model is no better illustrated than by Melaka in the 15th century, a socially-complex historical polity which ‘depended almost entirely 154 Boomgard, Southeast Asia, p. 18-20. Reid, “Humans and Forests,” p. 109. 156 Wheatley, The Golden Khersonese, p. 82-3. 157 Blagden, “The Foundation of the Settlement,” p. 8. 158 Hill, Hikayat Abdullah, p. 141, 144-5. 159 Miksic, “14th-Century Singapore,” p. 52. 155 111 for its subsistence on its long-distance maritime network.’ 160 Given the suggested relationships between the two polities as highlighted by numerous scholars, it is entirely possible that economics of pre-colonial Singapore functioned in the same modus operandi as that of Melaka in the succeeding century. While the precise dynamics of this exchange and trade may never be clearly discerned beyond their distribution patterns as elicited from archaeological data, it is evident that various social institutions would have been developed to manage and coordinate these processes for the smooth functioning of its society. The position of ‘syahbandar’ in the 16th-17th centuries is perhaps the closest early modern equivalent of these social institutions. A term derived from Persian, a syahbandar was responsible for the collection of duties from foreign merchants entering the port and held authority over the quarter where they resided, as well as most aspects of foreign trade conducted ‘including transport and warehousing.’ The syahbandar also acted as the ‘intermediary in negotiations’ between these foreign merchants and the ruler of the polity. 161 It is interesting to note that a ‘Xabandaria,’ the realm of the syahbandar, was placed at the mouth of the Singapore River – on the same location as the pre-colonial settlement two centuries earlier – in a Portuguese map of the island drawn in 1604. 162 Social institutions or notable elites like the syahbandar would have been especially important in the import and trade of goods unavailable in the settlement for various crafts, especially metal ore for the metalworking industries. As metals were generally valuable commodities in the pre-industrial world due to their relative intrinsic scarcity and labour-intensive production processes, obtaining these resources would often involve ‘big capital’ and significant organization during this ‘early historical period.’ 163 As Shah Alam suggests in his thesis, the ‘affluent class’ would have been 160 Luis Filipe Ferreira Reis Thomaz, “The Malay Sultanate of Melaka,” in Southeast Asia in the Early Modern Era. Trade, Power, and Belief, ed. Reid (Ithaca: Cornell University Press, 1993), p. 71-2. 161 Malcom H. Murfett, Miksic, Brian P. Farrell & Chiang Ming Shun, Between Two Oceans. A Military History of Singapore from First Settlement to Final British Withdrawal (Singapore: Marshall Cavendish Academic, 2005), p. 33. 162 Mills, “Eredia’s Description of Malacca, Meridional India, and Cathay. Translated from the Portuguese, with Notes,” JMBRAS 8,1 (September 1930), p. 224-5, Plate VI. 163 Bronson, “Patterns in the Early Southeast Asian Metals Trade,” in Early Metallurgy, Trade and Urban Centres in Thailand and Southeast Asia, ed. Ian Glover et. al. (Bangkok: White Lotus, 1992), p. 64-73, 96-104. 112 involved, or even be responsible, in sustaining this supply of metals as well as various other common and prestige goods for the consumption of pre-colonial Singapore society. SUMMARY AND IMPLICATIONS The wealth and variety of material cultural remains, their correlation with social status as illustrated by their non-random distribution at the STA site, as well as the presence of craftspecialization is conclusive evidence for a socially-complex pre-colonial Singapore polity. More than the confirmation of a hypothesis, it is also the successful test of an inter-disciplinary method of analysis which has provided a new insight on the organization of this settlement based on firm archaeological data and historical sources. Although the STA site is large enough to have contained a diverse population, the relative homogeneity in material culture suggests that its inhabitants probably do not represent more than one stratum of an elite, commoner or perhaps even foreigner social group in precolonial Singapore society. As stated earlier in the methodology, this test of complexity measures only the relative differentiation within the size of a neighbourhood as opposed to the entirety of the island’s pre-colonial population. Nonetheless, if the settlement on the whole whole had a high degree of social-complexity, there would still be a nuanced diversity of wealth even within a relatively homogenous neighbourhood containing a population drawn mostly from one of these social groups. In light of the heterogenetic nature of pre-colonial Singapore as established by Miksic, the correlation between material cultural remains with differing social status also seem to suggest the existence of a political structure constructed around alliance networks ‘maintained through the charismatic attraction of individuals’ or possibly groups of elites, as opposed to a strongly centralized bureaucratic government found in orthogenetic polities. According to Laura Lee Junker, institutionalized forms of exchange formed the core of ‘Southeast Asian political economies’ and were the ‘very foundations of political power.’ The social elites are hence required to constantly establish and reinforce various ‘political relationships and 113 hierarchies of authority’ through the ‘strategic disbursement of wealth to cronies and clients;’ 164 the control of the trade in imported prestige goods, especially that of Chinese ceramics as well as valuable metals, are known to have pre-occupied the ruling elites of various pre-colonial mainland and island Southeast Asian polities including Srivijaya and Angkor. 165 It is therefore likely that the differentiation in the spatial distribution of various ceramic categories at the STA site is a physical reflection of the political economy of pre-colonial Singapore. The inherently volatile and conflictive nature of this heterogentic political system however may have eventually been the reason behind its brief existence and unceremonious demise. 164 Junker, “Political Economy in the Historic Period Chiefdoms and States of Southeast Asia,” in Archaeological Perspectives on Political Economies, ed. Gary M. Feinman and Linda M. Nicholas (Salt Lake City: The University of Utah Press, 2004), p. 22351. 165 Kenneth R. Hall, A History of Early Southeast Asia. Maritime Trade and Societal Develoment, 100-1500 (Lanham: Rowman & Littlefield Publishers, Inc.), p. 1-36; Ricklefs et. al., A New History of Southeast Asia, p.7-9. 114 IV. CONCLUSION This thesis has been an attempt to conduct a thorough study of a single pre-colonial archaeological site located in Singapore though the analysis of its material cultural remains. Not only has it confirmed the dating of these pre-colonial remains to the 14th century, the thesis has also identified the location of the first neighbourhood-size residential site of the pre-colonial polity in Singapore. In the process, the study was able to demonstrate the complexity of the site through the statistical study of spatial distribution of its artefacts which was revealed to be of a non-random, patterned order. This supported the interpretation of differentiation in the amount and variety of high-value ceramic artefacts between the four quadrants as a reflection of differentiation in material wealth and social status. That social stratification exists at a site-level is indicative of the likelihood that the wider settlement, to which the STA site belongs to, was a highly-complex pre-colonial polity. The veracity of this conclusion should be tested against the artefact assemblages of other pre-colonial Singaporean sites, chief of which is FTC, as it is currently the only site on the island hypothesized as the ceremonial centre of the settlement and the residential quarter of the ruling elites. The recovery of artefacts such as glass and gold ornaments to porcelain compass and other rare Chinese ceramics unique to the site already suggests a similar distribution pattern reflecting a high-degree of social complexity. 1 The conduct of the same test of complexity at the other well-studied PHC site, already identified to be the location of metal- 1 Low, “Singapore from the 14th to 19th Century,” in Early Singapore, p. 14-21. 115 working and commercial activities, will also illuminate a greater level of distinction between differing pre-colonial social groups who had operated in that vicinity. A finer level of comparative analysis in the spatial distribution of its material cultural remains may even allow for the identification of the precise location of these social groups and the identification of specific activities in which they conducted. Naturally, the same approach can be adopted for all other pre-colonial archaeological sites in Singapore and arguably, in the region as well. The significance of the study of social organization at STA lies not on the illumination of social complexity at a pre-colonial settlement site, but also in providing the historian a picture of a heterogenetic Southeast Asian settlement site organization just before the arrival of the Europeans. The hundred-odd year long period of its existence coincides exactly with the watershed between a millennia-long continuum of localized Southeast Asian settlement development on the brink of European imperialism and colonialism in the region. Following the fall of Melaka to the Portuguese in 1511, European governors and soldiers, followed by surveyors, engineers and architects began to build colonial settlements and fortifications across the region, a process which would introduce far-reaching changes for Southeast Asian settlement-planning in the peak of colonialism during the 19th and 20th centuries. Two quintessential examples of this transition were Portuguese-Melaka and its fort A’Famosa, as well as Banten (present-day Banten Lama) and its fortress-palace Keraton Surosowan. Fort A’Famosa was built over the ruins of the former Sultan’s palace and fort which were destroyed during Afonso d’Albuquerque’s attack in 1511, whereas Banten’s Keraton Surosowan, commissioned by Sultan Haji in 1680, was built in a style reminiscent of the European star forts (trace italienne) by the Dutch overseer Hendrik Lucasz Cardeel. 2 The study of pre-colonial Singapore is therefore the study of an indigenous settlement before this transitional period in Southeast Asian history. 2 See Graham W. Irwin, “Melaka Fort,” in Melaka, p. 783; Claude Guillot, Hasan M. Ambary & Jacques Dumarcay, The Sultanate of Banten (Jakarta: Gramedia Book Publishing Division, 1990), p. 64. 116 This is fortuitous, as the wealth of information enclosed within the appendices demonstrates the potential of the archaeological record in providing verifiable and accurate data much needed for scholars to understand the socio-economic complexity of a pre-colonial settlement and polity in this period. Scholars of pre-colonial Southeast Asia have generally relied heavily on the attempt to corroborate indigenous texts and epigraphic data with Chinese records, but this approach present its own set of problems. Chinese records in particular were usually compiled from disparate strands of information, which were ‘filtered through their own cultural prejudices and general worldview.’ 3 Most inscriptions, if found largely complete, legible and translatable, present more of ‘scattered morsels of knowledge about rulers, their claims to power and glory and their deeds.’ 4 There are of course exceptions to the rule, as Antoinette M. Barrett illustrated from her study of 10th century Javanese inscriptions. 5 But while these inscriptions contain ‘unusually rich in social and economic information about a wide spectrum of the population,’ such epigraphic data are few and far between. 6 Other indigenous texts, especially court chronicles and genealogies, reveal more about how the authors ‘saw their past than about factual details of that past’ itself, and hence are poor sources of information to reconstruct any dependable historical narrative or urban environment of the polities in question. 7 As a result, there is a paucity of historical data which can illuminate the level of socio-economic complexity for any Southeast Asian polity in the 14th century. Hence, the study of the artefact assemblage from pre-colonial Singapore looms large in the study of maritime Southeast Asia before the transitions of the 19th and 20th centuries. In conclusion, this thesis is among the first steps taken towards the study of not just the historical, but archaeological dimensions of traditional pre-colonial settlements. It is hoped that the deductions made in this thesis will contribute to the existing knowledge of Temasek- 3 Ricklefs et. al., A New History of Southeast Asia, p. 18. Ibid. 5 Antoinette M. Barrett Jones, Early tenth century Java from the inscriptions: a study of economic, social, and administrative conditions in the first quarter of the century. Dordrecht: Foris Publications, 1984. 6 Jan Wisseman Christie, “Reviewed work(s): Early Tenth Century Java from the Inscriptions: A Study of Economic, Social and Administrative Conditions in the First Quarter of the Century by Antoinette M. Barrett Jones,” JSEAS 17,2 (September 1986), p. 371. 7 Ricklefs et. al., A New History of Southeast Asia, p. 37. 4 117 Singapura and serve as a platform for future generations of students to continue the pursuit of this elusive pre-colonial settlement on Singapore Island. By its investigation of social organization and complexity at the STA site, it is also hoped that the thesis has demonstrated the potential of archaeological data in the creation of a more comprehensive and plausible historical narrative in Singapore’s pre-colonial history. Through the study of both material cultural remains and textual sources, this thesis also represents the confluence of two different approaches, archaeological and historical, in the third historiographical phase of pre-colonial Singapore. With the limits in interpretations of historical sources reached, the verification of the Temasek Paradigm now rests upon the archaeological recovery and analysis of more material cultural remains which still lies buried underneath our feet. 118 BIBLIOGRAPHY PRIMARY SOURCES I. Chinese 苏继倾 (Sū, Jì Qīng). 《島夷梽畧校释》 (Dăo Yí Zhì Lüe Jiào Shì or ‘An explanation of the Description of the Barbarians of the Isles’). 北京: 中华书局, 1981. 朱彧 (Zhū, Yù). “《萍州可談》 (Píng Zhōu Kě Tán, Discourse on the Floating Islands),” circa 1116 CE. In Friedrich Hirth & William Woodville Rockhill, Chau Ju-Kua: His Work on the Chinese and Arab Trade in the twelfth and thirteenth Centuries, entitled Chu-fan-chi, p. 29-33. St. Petersburg: Printing Office of the Imperial Academy of Sciences, 1911. II. European Birch, Walter De Gray. The Commentaries of the Great Afonso Dalboquerque. Second Viceroy of India. Volume III. New York: Burt Franklin, 1970. Cortesão, Armando. The Suma Oriental of Tomé Pires. An Account of the East, from the Red Sea to Japan, written in Malacca and India in 1512-1515. And the Book of Francisco Rodrigues. Rutter of a Voyage in the Red Sea, Nautical Rules, Almanack and Maps, written and drawn in the East before 1515. Volume II. London: Hakluyt Society, 1944. III. Southeast Asian Brown, C. C. Malay Annals. Selangor Darul Ehsan: MBRAS, 2009. 119 Phalgunadi, I Gusti Putu. The Pararaton. A Study of the Southeast Asian Chronicle. New Dehli: Sundeep Prakashan, 1996. Robson, Stuart. Deśawarṇana (Nāgarakṛtāgama) by Mpu Prapañca. Leiden: KITLV Press, 1995. CATALOGUES & INDICES Barnes, Amy. “Catalogue entries.” In Qingbai Ware: Chinese Porcelain of the Song and Yuan Dynasties, p. 27-233. Edited by Stacey Pierson. London: Percival David Foundation of Chinese Art, 2002. Bautista, Edwin R. et. al. “Guangdong Wares. 10th to 12th century,” “Zhejiang Wares. 9th to 11th century,” “Fujian Wares. 12th to 14th century” & “Thai Kendi (Satingphra-type). 10th to 12th century.” In Guangdong Ceramics from Butuan and other Philippine sites, p. 87-112, 113-7, 118-25, 126-127. Edited by Roxanna Maude Brown. Manila: The Oriental Ceramic Society of the Philippines, Inc., 1989. Brown, Roxanna Maude & Sjostrand, Sten. “Artifacts,” p. 25-39. In Roxanna Maude Brown & Sten Sjostrand. Turiang: A Fourteenth-Century Shipwreck in Southeast Asian Waters. Los Angeles: Pacific Asia Museum, 2000. Brown, Roxanna Maude. “Ceramics Provenance and Categories: An Inventory,” p. 79-187. In Roxanna Maude Brown. The Ming Gap and Shipwreck Ceramics of Southeast Asia. Towards a Chronology of Thai Trade Ware. Bangkok: The Siam Society under Royal Patronage, 2009. Eng-Lee Seok Chee. “Catalogue,” p. 55-83. In Eng-Lee Seok Chee. Kendis. Singapore: National Museum Singapore, 1984. Gotuaco, Larry. “Coloured Plates. Chinese Blue and White Ceramics. Yuan Period. 14th century.” In Chinese and Vietnamese Blue and White Wares Found in the Philippines, p. 3274. Edited by Larry Gotuaco et. al. Makati City: Bookmark, Inc, 1997. 120 Guy, John S. “Chinese Ceramics,” “Vietnamese Ceramics” & “Thai Ceramics.” In John S. Guy. Oriental Trade Ceramics in South-East Asia Ninth to Sixteenth Centuries. With a Catalogue of Chinese, Vietnamese and Thai Wares in Australian Collections. Singapore: Oxford University Press, 1990. Hiromu, Honda & Noriki, Shimazu. “The Collection.” In Hiromu, Honda & Noriki, Shimazu. The Beauty of Fired Clay. Ceramics from Burma, Cambodia, Laos and Thailand, p. 53-267. Kuala Lumpur: Oxford University Press, 1997. Khoo Joo Ee. “The Collection,” p. 30-113. In Khoo Joo Ee. Kendi. Pouring Vessels in the University of Malaya Collection. Singapore: Oxford University Press, 1991. Kwan, K. K. & Lam, Peter Y. K. “Guangdong Wares. Song Dynasty 11th-12th century,” “Fujian Wares. Song and Yuan Dynasties 12th-14th century,” “Jiangsu Wares. 12th-14th century and Guangdong Wares and other kiln sites,” “Zhejiang Wares. Song and Yuan Dynasties 11th-14th century,” “Earthenware Pottery. Southeast Asia” & “Ceramics found at Kampong Juara in the Muzium Sultan Abu Bakar, Pekan 12th-19th century.” In A Ceramic Legacy of Asia’s Maritime Trade. Song Dynasty Guangdong Wares and other 11th to 19th century Trade Ceramics found on Tioman Island, Malaysia, p. 83-107, 108-17, 118-120, 121-124, 125-126, 138-45. Compiled by K. K. Kwan et. al. Petaling Jaya: Southeast Asian Ceramic Society, West Malaysia Chapter, 1985. 劉巨成 (Liú, Jù Chéng). 《中國古錢譜》(Zhōng Guó Gǔ Qián Pǔ or Illustrative Plates of Chinese Ancient Coins). 北京: 文物出版社, 1989. Medley, Margaret & Pierson, Stacey. “Catalogue.” In Illustrated Catalogue of Celadon Wares in the Percival David Foundation of Chinese Art, p. 19-60. London: University of London, SOAS, 1997. 121 Miksic, John Norman & Ong, Natalie SY. “Catalogue.” In Southeast Asian Ceramics. New Light on Old Pottery, p. 101-65. Edited by John Norman Miksic. Singapore: Southeast Asian Ceramic Society, 2009. Quie, Malcom V. et. al. “Illustrated Catalogue of Exhibits. Chinese Celadons” & “Illustrated Catalogue of Exhibits. Other Related Wares.” In Chinese Celadons and other Related Wares in Southeast Asia, p. 103-274, 275-306. Compiled by Southeast Asian Ceramic Society, Singapore. Singapore: Arts Orientalis, 1979. Tan, Rita C. “Coloured Plates. Early Ming Period. Late 14th-Mid 15th Centuries.” In Chinese and Vietnamese Blue and White Wares Found in the Philippines, p. 109-79. Edited by Larry Gotuaco et. al. Makati City: Bookmark, Inc, 1997. Yeo, S. T. & Martin, Jean. “Catalogue of the Exhibits. Yuan Dynasty 1280-1368,” “Early Ming 1400-1487,” “Middle Ming 1488-1566,” “Late Ming 1566-1643” & Transitional Ming/Ching 1630-1660.” In Chinese Blue & White Ceramics, p. 95-101, 103-21, 123-61, 163-215 & 21723. Compiled by S. T. Yeo & Jean Martin. Singapore: Arts Orientalis, 1978. Yeo, S. T. & Martin, Jean. “Index of Decorations.” In Chinese Blue & White Ceramics, p. 299314. Compiled by S. T. Yeo & Jean Martin. Singapore: Arts Orientalis, 1978. ACADEMIC THESES Chen, Hong Liang Omar. Earth to Earth: An Investigation into the Occurance of Earthenware Artefacts at the Parliament House Complex Site. M.A. Thesis. National University of Singapore, 2001. Heng, Thiam Soon Derek. Temasik: Reconstruction of a Classical Period Malay Trading Port Polity. M.A. Thesis. University of London, 1997. Miksic, John Norman. Archaeology, Trade and Society in Northeast Sumatra. Ph.D. Thesis. Cornell University, 1979. 122 Nik Hassan Shuhaimi bin Nik Abd. Rahman. Art, Archaeology and the Early Kingdoms in the Malay Peninsula and Sumatra: C.400-1400 A. D. Ph.D. Thesis. University of London, 1984. Shah Alam Mohd. Zaini. Metal Finds and Metal-working at the Parliament House Complex, Singapore. M.A. Thesis. University of Michigan. 1997. Wong, Wai Yee. A Preliminary Study of Some Economic Activities of Khmer Empire: Examining the Relationship between the Khmer and Guangdong Ceramic Industries during the 9th to 14th centuries. Ph.D. Thesis. National University of Singapore, 2010. CONFERENCE PAPERS Miksic, John Norman, Yap Choon Teck, Fong, Sam Yau Li & Wan Kebao. “EDXRF Analyses of Some Yuan Dynasty Artifacts Excavated in Singapore.” Fifth Chinese Symposium on Archaeometry. University of Science and Technology of China, Hefei. October 14-19, 1998. Li, Min. “Fragments of Globalization: Chinese Blue-and-white Porcelain in Early Colonial Philippines.” 18th Indo-Pacific Prehistory Association (IPPA) Congress. University of the Philippines, Manila. 20-26 March 2006. UNPUBLISHED WORK Foo, Soo Ling. A Study of 14th Century Blue and White Ceramics Excavated in Singapore. Unpublished ISM paper. Southeast Asian Studies Programme, Faculty of Arts & Social Sciences, National University of Singapore, 2005. PUBLICATIONS Amara Srisuchat. “Earthenware from Archaeological Sites in Southern Thailand: The First Century BC to the Twelfth Century AD.” In Earthenware in Southeast Asia, p. 249-60. Edited by John Norman Miksic. Singapore: Singapore University Press, 2003. Anderson, Benedict Richard O’Gorman. “The Idea of Power in Javanese Culture.” In Culture and Politics in Indonesia, p. 1-69. Edited by Claire Holt. Ithaca: Cornell University Press, 1972. 123 Anonymous. “Miscellaneous. Agriculture in Singapore.” JIAEA 3 (1849): p. 508-11. Appadurai, Arjun. “Introduction: commodities and the politics of value.” In The social life of things. Commodities in cultural perspective, p. 3-63. Edited by Arjun Appadurai. Cambridge: Cambridge University Press, 1986. Arnold, Dean E. Ceramic Theory and Cultural Process. Cambridge: Cambridge University Press, 1985. Atmosudiro, Sumijati. “Notes on the Tradition of Pottery Making in the Region of Kasongan, Regency of Bantul, Yogyakarta.” In Studies on Ceramics, p. 121-6. Edited by Departemen P dan K. Jakarta: Proyek Penelitian Purbakala Jakarta, 1984. Aung-Thwin, Michael A. The Mists of Rāmañña. The Legend That Was Lower Burma. Honolulu: University of Hawai‘i Press, 2005. Ayers, John. “A Note on Celadon Wares from Sinan,” in Chinese Celadons and other Related Wares in Southeast Asia. In Chinese Celadons and other Related Wares in Southeast Asia, p. 49-55. Compiled by Southeast Asian Ceramic Society, Singapore. Singapore: Arts Orientalis, 1979. Bacus, Elizabeth A. “Styles of Alliance?: Decorated Earthenwares in Late Prehistoric and Protohistoric Philippines Polities.” In Earthenware in Southeast Asia, p. 39-51. Edited by John Norman Miksic. Singapore: Singapore University Press, 2003. Barnes, Warren D. “Singapore old Straits and New Harbour.” JSBRAS 60 (December 1911): p. 25-36. Bastin, John. Letters and Books of Sir Stamford Raffles and Lady Raffles. Singapore: Editions Didier Millet Pte Ltd, 2009. Bayman, James M. “Shell ornament consumption in a Classic Hohokam Platform Mound Community Center.” JFA 23,4 (Winter 1996): p. 403-20. 124 Beck, Margaret E. “Midden Ceramic Assemblage Formation: A Case Study from Kalinga, Philippines.” AA 71,1 (January 2006): 27-51. Beck, Margaret E. & Hill Jr, Matthew E. “Rubbish, Relatives, and Residence: The Family Use of Middens.” JAMT 11,3 (September 2004): p. 297-333. Bellwood, Peter. Prehistory of the Indo-Malaysian Archipelago. Honolulu: University of Hawai’i Press, 1997. Bellwood, Peter & Omar, Matussin. “Trade Patterns and Political Developments in Brunei and Adjacent Areas, A.D. 700-1500.” BMJ 4,4 (1980): p. 155-79. Blagden, Charles Otto. “Notes on Malay History.” JSBRAS 53 (September 1909): p. 139-62. __________________. “Singapore prior to 1819.” In One Hundred Years of Singapore. Volume 1, p. 1-5. Edited by Walter Makepeace, Gilbert E. Brooke & Roland St. John Braddell. Singapore: Oxford University Press, 1991. __________________. “The Foundation of the Settlement.” In One Hundred Years of Singapore. Volume 1, p. 6-12. Edited by Walter Makepeace, Gilbert E. Brooke & Roland St. John Braddell. Singapore: Oxford University Press, 1991. __________________. “The Empire of the Maharaja, King of the Mountains and Lord of the Isles.” JSBRAS 81 (March 1920): p. 23-8. Blundell, Edmund Augustus. “Notices of the History and Present Condition of Malacca.” JIAEA 2 (1848): p. 726-54. Boomgaard, Peter. Southeast Asia. An Environmental History. Santa Barbara: ABC-CLIO, 2007. Borschberg, Peter. “The Johor-VOC Alliance and the Twelve Years’ Truce: Factionalism, Intrigue and International Diplomacy. 1606-13.” IILJ Working Papers 2009/8. (History and Theory of International Law Series), p. 1-64. Finalized 12/17/2009. (www.iilj.org). 125 ______________. The Singapore and Melaka Straits. Violence, Security and Diplomacy in the 17th Century. Singapore: NUS Press, 2010. Bottom, J. C. “Some Malay Historical Sources: A Bibliographical Note.” In An Introduction to Indonesian Historiography, p. 156-93. Edited by Soedjatmoko, Mohammad Ali, G. J. Resink & G. McT. Kahin. Ithaca: Cornell University Press, 1965. Braddell, Roland St. John. “Notes on Ancient Times in Malaya.” JMBRAS 20,1 (June 1947): p. 161-86. _____________________. “Lung-Ya-Men and Tan-Ma-Hsi.” JMBRAS 23,1 (February 1950): p. 37-51. Braddell, Thomas. “Notices of Singapore.” JIAEA 7 (1853): p. 325-57. ______________. “Abstract of the Sijarah Malayu or Malay Annals, with Notes.” JIAEA 5 (1851): p. 125-34, 173-9, 244-9, 312-22, 451-8, 543-8, 642-9, 729-40. Bronson, Bennett. “Exchange at the Upstream and Downstream Ends: Notes toward a Functional Model of the Coastal State in Southeast Asia.” In Economic Exchange and Social Interaction in Southeast Asia: Perspectives from Prehistory, History and Ethnography, p. 3952. Edited by Karl L. Hutterer. Ann Arbor: The University of Michigan, 1977. _______________. “Patterns in the Early Southeast Asian Metals Trade.” In Early Metallurgy, Trade and Urban Centres in Thailand and Southeast Asia, p. 63-114. Edited by Ian Glover et. al. Bangkok: White Lotus, 1992. Bronson, Bennet & Wisseman-Christie, Jan. “Palembang as Śrīvijaya. The Lateness of Early Cities in Southern Southeast Asia.” AP 19,2 (1976): p. 220-39. Brown, Roxanna Maude. The Ceramics of South-East Asia. Their Dating and Identification. Singapore: Oxford University Press, 1988. 126 ____________________. “Guangdong. A Missing Link to Southeast Asia.” In Guangdong Ceramics from Butuan and other Philippine sites, p. 81-5. Edited by Roxanna Maude Brown. Manila: The Oriental Ceramic Society of the Philippines, Inc, 1989. ____________________. The Ming Gap and Shipwreck Ceramics of Southeast Asia. Towards a Chronology of Thai Trade Ware. Bangkok: The Siam Society under Royal Patronage, 2009. Brown, Roxanna Maude & Sjostrand, Sten. Turiang: A Fourteenth-Century Shipwreck in Southeast Asian Waters. Los Angeles: Pacific Asia Museum, 2000. _____________________________________. Maritime Archaeology and Shipwreck Ceramics in Malaysia. Kuala Lumpur: Department of Museums & Antiquities, 2004. Brumfiel, Elizabeth M. & Earle, Timothy K. “Specialization, exchange and complex societies: an introduction.” In Specialization, Exchange and Complex Societies, p. 1-9. Edited by Elizabeth M. Brumfiel & Timothy K. Earle. Cambridge: Cambridge University Press, 1987. Buckley, Charles Burton. An Anecdotal History of Old Times In Singapore. Singapore: Oxford University Press, 1984. Caple, Chris. Objects. Reluctant Witnesses to the Past. London: Routledge, 2006. Caldecott, A. “The Malay Peninsula in the XVIIth & XVIIIth Centuries.” JSBRAS 82 (September 1920): p.129-32. de Casparis, Johannes Gijsbertus. Indonesian Paleography: A history of writing in Indonesia from the beginnings to c. A.D. 1500. Leiden: E. J. Brill, 1975. Chen, Hong Liang Omar. “Ancient Singapore Earthenware Pottery.” In Early Singapore. 1300s-1819. Evidence in Maps, Texts and Artefacts, p. 55-72. Edited by John Norman Miksic & Cheryl-Ann Low Mei Gek. Singapore: Singapore History Museum, 2004. 127 Chung, Chee Kit, “Longyamen is Singapore: The Final Proof?” In Admiral Zheng He & Southeast Asia, p. 142-68. Edited by Leo Suryadinata. Singapore: ISEAS and International Zheng He Society Singapore, 2005. Coedès, George. “The Kingdom of Sriwijaya.” In Sriwijaya. History, Religion & Language of an Early Malay Polity, p. 1-40. Edited by Pierre-Yves Manguin. Kuala Lumpur: MBRAS, 1992. _____________. The Indianized States of Southeast Asia. Kuala Lumpur: University of Malaya Press, 1968. _____________. “The Malay Inscriptions of Sriwijaya.” In Sriwijaya. History, Religion & Language of an Early Malay Polity, p. 41-92. Edited by Pierre-Yves Manguin. Kuala Lumpur: MBRAS, 1992. Crawfurd, John. History of the Indian Archipelago. Volume II. Edinburgh: Archibald Constable and Co., 1820. _____________. Journal of an Embassy from the Governor General of India to the Courts of Siam and Cochin China. Volumes I & II. London: Henry Colburn and Richard Bentley, 1830. Daigoro, Chihara. Hindu-Buddhist Architecture in Southeast Asia. Leiden: E.J. Brill, 1996. Darvill, Timothy. The Concise Oxford Dictionary of Archaeology. Second Edition. Oxford: Oxford University Press, 2008. Douglas, Mary & Isherwood, Baron. The World of Goods. London: Routledge, 1996. Eng-Lee Seok Chee. Kendis. Singapore: National Museum Singapore, 1984. Eska Asih Putrina Taim. “The Batujaya Pottery: Early Hindu-Buddhist Pottery in West Java.” In Archaeology: Indonesian Perspective. R. P. Soejono’s Festschrift, p. 334-42. Edited by Truman Simanjuntak et. al. Jakarta: Indonesian Institute of Sciences, International Center for Prehistoric and Austronesian Studies, 2006. 128 Evans, I. H. N. “On the Persistence of an Old Type of Water-Vessel.” JMBRAS 1 (April 1923): p. 248-50. Flecker, Michael. “A Ninth-Century Arab Shipwreck in Indonesia. The First Archaeological Evidence of Direct Trade with China.” In Shipwrecked. Tang Treasures and Monsoon Winds, p. 100-19. Edited by Regina Krahl et. al. Washington D. C.: Arthur M. Sackler Gallery, Smithsonian Institution, 2010. ______________. “A Ninth-Century AD Arab or Indian Shipwreck in Indonesia: First Evidence for Direct Trade with China,” WA 32,3 (February 2011): p. 335-54. Frédéric, Louis. The Temples and Sculpture of Southeast Asia. London: Thames and Hudson, 1965. Garnsey, Wanda & Alley, Rewi. China. Ancient kilns and modern ceramics. A guide to the potteries. Canberra: Australian National University Press, 1983. Groeneveldt, Willem Pieter. Notes on the Malay Archipelago and Malacca: Compiled from Chinese Sources. Batavia: -, 1876. Gerini, Gerolamo Emilio. Researches on Ptolemy’s Geography of Eastern Asia. London: Royal Asiatic Society & Royal Geographical Society, 1909. Gibson-Hill, Carl Alexander. “Singapore: Notes on the History of the Old Strait, 1580-1850.” JMBRAS 27,1 (May 1954): p. 163-214. _______________________. “Singapore Old Strait & New Harbour (1300-1870).” In Memoirs of the Raffles Museum. No. 3, p. 10-115. Edited by Carl Alexander Gibson-Hill. Singapore: Government Printing Office, 1956. Guillot, Claude, Ambary, Hasan M. & Dumarcay, Jacques. The Sultanate of Banten. Jakarta: Gramedia Book Publishing Division, 1990. Guy, John S. Ceramic Traditions of South-East Asia. Singapore: Oxford University Press, 1990. 129 __________. Oriental Trade Ceramics in South-East Asia Ninth to Sixteenth Centuries. With a Catalogue of Chinese, Vietnamese and Thai Wares in Australian Collections. Singapore: Oxford University Press, 1990. Hall, Daniel George Edward. A History of South-East Asia. Fourth Edition. Hampshire: Palgrave Macmillan, 1995. Hall, Kenneth R. A History of Early Southeast Asia. Maritime Trade and Societal Develoment, 100-1500. Lanham: Rowman & Littlefield Publishers, Inc. Hallett, Jessica. “Pearl Cups Like the Moon. The Abbasid Reception of Chinese Ceramics.” In Shipwrecked. Tang Treasures and Monsoon Winds, p. 74-81. Edited by Regina Krahl et. al. Washington D.C & Singapore.: Arthur M. Sackler Gallery, Smithsonian Institution, Singapore Tourism Board in association with National Heritage Board, 2010. Harrisson, Barbara. “The Ceramic Trade across the South China Sea.” In Southeast Asia-China Interactions, p. 489-505. Compiled by Geoff Wade. Kuala Lumpur: MBRAS, 2007. Hayden, Brian. “Practical and prestige technologies: the evolution of material systems.” JAMT 5,1 (March 1998): p. 1-55. Hayden, Brian & Cannon, Aubrey. “Where the garbage goes: Refuse disposal in the Maya Highlands.” JAA 2,2 (June 1983): p. 117-63. Heng, Thiam Soon Derek. “Indigenising Singapore’s Past: An Approach towards Internalising Singapore’s Settlement History from the Late Thirteenth to Twenty-First Centuries.” In New Perspectives and Sources on the History of Singapore, p. 1-29. Edited by Derek Heng Thiam Soon. Singapore: National Library Board, 2006. ______________________. “Temasik as an International and Regional Trading Port in the Thirteenth and Fourteenth Centuries: A Reconstruction based on Recent Archaeological Data.” JMBRAS 72,1 (June 1999): p. 113-24. 130 ______________________. “Reconstructing Banzu, A Fourteenth Century Port Settlement in Singapore.” JMBRAS 75,1 (June 2002): p. 69-90. ______________________. Sino-Malay Trade and Diplomacy from the Tenth through the Fourteenth Century. Athens: Ohio University Press, 2009. Hervey, D. F. A. “Valentyn’s Description of Malacca.” JSBRAS 13 (June 1884): p. 49-74. Hill, A. H. Hikayat Abdullah. Kuala Lumpur: MBRAS, 2009. Hirth, Friedrich & Rockhill, William Woodville. Chau Ju-Kua: His work on the Chinese and Arab Trade in the twelfth and thirteenth Centuries, entitled Chu-fan-chi. St. Petersburg: Printing Office of the Imperial Academy of Sciences, 1911. Ho, Chuimei. “Problems in the Study of Zhejiang Green Glazed Wares.” In New Light on Chinese Yue and Longquan Wares. Archaeological Ceramics Found in Eastern and Southern Asia, A.D. 800-1400, p. 187-212. Edited by Ho Chuimei. Hong Kong: The University of Hong Kong, 1994. ___________. “Yue-Type and Longquan-Type Green Glazed Wares made Outside Zhejiang Province.” In New Light on Chinese Yue and Longquan Wares. Archaeological Ceramics Found in Eastern and Southern Asia, A.D. 800-1400, p. 103-19. Edited by Ho Chuimei. Hong Kong: The University of Hong Kong, 1994. Hose, George Frederick, M. A. “Inaugural Address by the President.” JSBRAS 1 (July 1878): p. 1-12. Hsü, Yün-Ts’iao. “Singapore in the Remote Past.” JMBRAS 45,1 (January 1973): p. 1-9. Hutchison, Charles S. & Tan, Denis N. K. Geology of Peninsular Malaysia. Kuala Lumpur: University of Malaya and Geological Society of Malaysia, 2009. 131 Irwin, Graham W.“Melaka Fort.” In Melaka. The Transformation of A Malay Capital c. 14001980. Volume One, p. 782-805. Edited by Kernial Singh Sandhu & Paul Wheatley. Kuala Lumpur: Oxford University Press, 1983. Jacq-Hergoualc’h, Michel. The Malay Peninsula. Crossroads of the Maritime Silk Road (100 BC – 1300 AD). Leiden: Brill, 2002. Johns, A. H. “The Turning Image: Myth & Reality in Malay Perceptions of the Past.” In Perceptions of the Past in Southeast Asia, p. 43-67. Edited by Anthony Reid and David Marr. Singapore: Asian Studies Association of Australia, 1979. Jones, Antoinette M. Barrett. Early tenth century Java from the inscriptions: a study of economic, social, and administrative conditions in the first quarter of the century. Dordrecht: Foris Publications, 1984. Junker, Laura Lee. “The Development of Centralized Craft Production Systems in A.D. 5001600 Philippine Chiefdoms.” JSEAS 25,1 (March 1994): p. 1-30. ______________. Raiding, Trading and Feasting. The Political Economy of Philippine Chiefdoms. Honolulu: University of Hawai’i Press, 1999. ______________. “Political Economy in the Historic Period Chiefdoms and States of Southeast Asia.” In Archaeological Perspectives on Political Economies, p. 223-51. Edited by Gary M. Feinman and Linda M. Nicholas. Salt Lake City: The University of Utah Press, 2004. Kamei, Meitoku. “Chronology of Longquan Wares of the Song and Yuan Periods.” In New Light on Chinese Yue and Longquan Wares. Archaeological Ceramics Found in Eastern and Southern Asia, A.D. 800-1400, p. 46-84. Edited by Ho Chuimei. Hong Kong: The University of Hong Kong, 1994. Kealhofer, Lisa & Grave, Peter. “Land Use, Political Complexity, and Urbanism in Mainland Southeast Asia.” AA 73,2 (April 2008): p. 200-25. 132 Kern, Hans. “Concerning some old Sanskrit Inscriptions in the Malay Peninsula.” JSBRAS 49 (December 1907): p. 95-101. Kerr, Rose. Song Dynasty Ceramics. London: V&A Publications, 2004. ________. The World in Blue and White. An exhibition of Blue and White ceramics, dating between 1320 and 1820, from members of the Oriental Ceramic Society. London: The Oriental Ceramic Society, 2003. Khoo, Joo Ee. Kendi. Pouring Vessels in the University of Malaya Collection. Singapore: Oxford University Press, 1991. Koek, E. “Portuguese History of Malacca.” JSBRAS 17 (June 1886): p. 117-49. Kwa, Chong Guan, Heng, Thiam Soon Derek & Tan, Tai Yong. Singapore. A 700-Year History. From Early Emporium to World City. Singapore: National Archives of Singapore, 2009. Kwan, K. K. & Martin, Jean. “Introduction to the finds from Pulau Tioman.” In A Ceramic Legacy of Asia’s Maritime Trade. Song Dynasty Guangdong Wares and other 11th to 19th century Trade Ceramics found on Tioman Island, Malaysia, p. 69-82. Compiled by K. K. Kwan et. al. Petaling Jaya: Southeast Asian Ceramic Society, West Malaysia Chapter, 1985. Lam, Peter Y. K. “Northern Song Guangdong Wares.” In A Ceramic Legacy of Asia’s Maritime Trade. Song Dynasty Guangdong Wares and other 11th to 19th century Trade Ceramics found on Tioman Island, Malaysia, p. 1-29. Compiled by K. K. Kwan et. al. Petaling Jaya: Southeast Asian Ceramic Society, West Malaysia Chapter, 1985. ______________. “Decorative Techniques and Motifs in Guangdong Trade Wares of the Song Dynasty.” In Guangdong Ceramics from Butuan and other Philippine sites, p. 46-59. Edited by Roxanna Maude Brown. Manila: The Oriental Ceramic Society of the Philippines, Inc, 1989. 133 Lee Kim Woon & Zhou Yingxin. Geology of Singapore. Singapore: Defence Science and Technology Agency in collaboration with Nanyang Technological University, Building and Construction Authority, 2009 Lefferts, Leedom & Cort, Louise Allison. “A Preliminary Cultural Geography of Contemporary Village-bassed Earthenware Production in Mainland Southeast Asia.” In Earthenware in Southeast Asia, p. 300-10. Edited by John Norman Miksic. Singapore: Singapore University Press, 2003. Li, Dejin. “Technology of Longquan Ware Manufacturing.” In New Light on Chinese Yue and Longquan Wares. Archaeological Ceramics Found in Eastern and Southern Asia, A.D. 8001400, p. 85-99. Edited by Ho Chuimei. Hong Kong: The University of Hong Kong, 1994. Lin, Shimin. “Zhejiang Export Green Glazed Wares: Ningbo Data.” In New Light on Chinese Yue and Longquan Wares. Archaeological Ceramics Found in Eastern and Southern Asia, A.D. 800-1400, p. 141-68. Edited by Ho Chuimei. Hong Kong: The University of Hong Kong, 1994. Lin, Wo Ling, Long-ya-men Re-identified. Singapore: 新加坡南洋学会出版, 1999. Linehan, W. “The Kings of 14th Century Singapore.” JMBRAS 20,2 (December 1947): p. 11727. _________. “History of Pahang.” JMBRAS 14,2 (May 1936): p. 1-256. Loewenstein, Prince John. “The Origin of the Malayan Metal Age.” JMBRAS 29,2 (1962): p. 5-78. Low, Mei Gek Cheryl-Ann. “Singapore from the 14th to 19th Century.” In Early Singapore. 1300s-1819. Evidence in Maps, Texts and Artefacts, p. 14-40. Edited by John Norman Miksic & Cheryl-Ann Low Mei Gek. Singapore: Singapore History Museum, 2004. 134 Lu, Yaw. “Introduction.” In Chinese Celadons and other Related Wares in Southeast Asia, p. 13-31. Compiled by Southeast Asian Ceramic Society, Singapore. Singapore: Arts Orientalis, 1979. Maisels, Charles Keith. The Archaeology of Politics and Power. Where, When and Why the First States Formed. Oxford: Oxbrow Books, 2010. Manguin, Pierre-Yves. “Palembang and Sriwijaya: An early Malay harbour-city rediscovered.” JMBRAS 66,1 (1993): p. 23-46. __________________. “The Archaeology of Early Maritime Polities of Southeast Asia.” In Southeast Asia. From prehistory to history, p. 282-313. Edited by Ian Glover and Peter Bellwood. Oxfordshire: RoutledgeCurzon, 2004. Marsden, William. The History of Sumatra. Singapore: Oxford University Press, 1986. Matheson, Virginia & Andaya, Barbara Watson. The Precious Gift (Tuhfat al- Nafis). Kuala Lumpur: Oxford University Press, 1982. Maxwell, William George. “Barretto de Resende’s Account of Malacca.” JSBRAS 60 (December 1911): p. 1-24. McKinnon, Edmund P. Edwards. “Oriental Ceramics Excavated in North Sumatra.” TOCS 41 (1975-7): p. 58-118. ___________________________. “Yue and Longquan Wares in Sumatra.” In New Light on Chinese Yue and Longquan Wares. Archaeological Ceramics Found in Eastern and Southern Asia, A.D. 800-1400, p. 284-98. Edited by Ho Chuimei. Hong Kong: The University of Hong Kong, 1994. ___________________________. “Historic Period Earthenware from the Island of Sumatra.” In Earthenware in Southeast Asia, p. 311-21. Edited by John Norman Miksic. Singapore: Singapore University Press, 2003. 135 Medley, Margaret. The Chinese Potter. A Practical History of Chinese Ceramics. Oxford: Phaidon Press Limited, 1976. Medley, Margaret & Pierson, Stacey. Illustrated Catalogue of Celadon Wares in the Percival David Foundation of Chinese Art. London: University of London, SOAS, 1997. Miksic, John Norman. “Classical Archaeology in Sumatra.” Indonesia 30 (October 1980: p. 42-66. __________________. “Srivijaya: Political, Economic, and Artistic Frameworks for Analysis.” In SPAFA. Final Report. Consultative Workshop on Archaeological and Environmental Studies on Srivijaya (T-W3). Bangkok and South Thailand. March 29-April 11, 1983, p. 195-206. Edited by Suchitra Vuthisathira. Bangkok: SPAFA Co-ordinating Unit, 1983. __________________. Archaeological Research on the “Forbidden Hill” of Singapore: Excavations at Fort Canning, 1984. Singapore: National Museum Singapore, 1985. __________________. “Urbanization and Social Change in Sumatra.” Archipel 37 (1989): p. 1-29. __________________. “Beyond the Grave: Excavations North of the Keramat Iskandar Shah, 1988.” In Heritage, p. 34-56. Edited by Lee Chor Lin. Singapore: National Museum, Singapore, 1989. __________________. “Settlement Patterns and Sub-regions in Southeast Asian History.” RIMA 31,1 (1991): p. 86-144. __________________. “Recently Discovered Chinese Green Glazed Wares of the Thirteenth and Fourteenth Centuries in Singapore and the Riau Islands.” In New Light on Chinese Yue and Longquan Wares. Archaeological Ceramics Found in Eastern and Southern Asia, A.D. 8001400, p. 229-50. Edited by Ho Chuimei. Hong Kong: The University of Hong Kong, 1994. 136 __________________. “Evolving Archaeological Perspectives on Southeast Asia, 1970-95.” JSEAS 26,1 (March 1995): p. 46-62. __________________. “Heterogenetic Cities in premodern Southeast Asia.” WA 32,1 (June 2000): p. 106-19. __________________. “Preface.” In Earthenware in Southeast Asia, p. xvii. Edited by John Norman Miksic. Singapore: Singapore University Press, 2003. __________________. “14th-Century Singapore: A Port of Trade.” In Early Singapore. 1300s1819. Evidence in Maps, Texts and Artefacts, p. 41-54. Edited by John Norman Miksic & Cheryl-Ann Low Mei Gek. Singapore: Singapore History Museum, 2004. __________________. “Intrasite Analysis of 14th-Century Singapore.” In Uncovering Southeast Asia’s Past. Selected Papers from the 10th International Conference of the European Association of Southeast Asian Archaeologists, p. 335-46. Edited by Elizabeth A. Bacus, Ian C. Glover & Vincent C. Pigott. Singapore: NUS Press, 2006. __________________. “Kilns of Southeast Asia.” In Southeast Asian Ceramics. New Light on Old Pottery, p. 48-69. Edited by John Norman Miksic. Singapore: Southeast Asian Ceramic Society, 2009. __________________. “Research on Ceramic Trade within Southeast Asia and between Southeast Asia and China.” In Southeast Asian Ceramics. New Light on Old Pottery, p. 69-99. Edited by John Norman Miksic. Singapore: Southeast Asian Ceramic Society, 2009. Miksic, John Norman & Lim Chen Shien. Archaeological Research on The Padang and in the St. Andrew’s Cathedral Churchyard: St. Andrew’s Cathedral Archaeological Research Project Progress Report Summary September 2003 – June 2004. ARI Working Paper. Miksic, John Norman & Yap Choon Teck. “Fine-Bodied White Earthenwares of Southeast Asia: Some X-Ray Fluorescence Tests.” AP 28,1 (1990): p. 45-60. 137 ___________________________________. “Compositional Analysis of Pottery from Kota Cina, North Sumatra: Implications for Regional Trade during the Twelfth to Fourteenth Centuries A.D.” AP 31,1 (Spring 1992): p. 57-76. Mills, John Vivian Gottlieb. “Eredia’s Description of Malacca, Meridional India, and Cathay. Translated from the Portuguese, with Notes.” JMBRAS 8,1 (September 1930): p. 1-288. ______________________. “Malaya in the Wu-Pei-Chih Charts.” JMBRAS 15,3 (December 1937): p. 1-48. ______________________. “Arab and Chinese Navigators in Malaysian Waters in about A.D. 1500.” JMBRAS 47,2 (December 1974): p. 1-82. ______________________. Ma Huan. Ying-Yai Sheng-Lan. ‘The Overall Survey of the Ocean’s Shores.’ [1433]. Edited by Feng Ch’eng-Chün. Cambridge: Cambridge University Press for the Haklyut Society, 1970. Milner, Anthony Crothers. The Malays. Oxford: Wiley-Blackwell, 2008. Moore, Eine. “A Suggested Classification of Stonewares of Martabani type.” SMJ 18,36-7 (July-December 1970): p. 1-78. Munoz, Paul Michel. Early Kingdoms of the Indonesian Archipelago and the Malay Peninsula. Singapore: Editions Didier Millet, 2006. Murfett, Malcom H., Miksic, John Norman, Farrell, Brian P. & Chiang, Ming Shun Between Two Oceans. A Military History of Singapore from First Settlement to Final British Withdrawal. Singapore: Marshall Cavendish Academic, 2005. Naniek Harkantiningsih. “Yue and Longquan Green Glazed Wares from Archaeological Sites in Java and East Indonesia.” In New Light on Chinese Yue and Longquan Wares. Archaeological Ceramics Found in Eastern and Southern Asia, A.D. 800-1400, p. 273-83. Edited by Ho Chuimei. Hong Kong: The University of Hong Kong, 1994. 138 Nas, Peter J. M. “The Early Indonesian Town. Rise and Decline of the City-State and its Capital.” In The Indonesian City. Studies in Urban Development and Planning, p. 18-36. Edited by Peter J. M. Nas. Dordrecht: Foris Publicaions, 1986. Orton, Clive, Tyers, Paul & Vince, Alan. Pottery in Archaeology. Cambridge: Cambridge University Press, 1994. Parker, S. R. “Chinese Celadons and Other Wares Excavated in Sarawak.” In Chinese Celadons and other Related Wares in Southeast Asia, p. 57-64. Compiled by Southeast Asian Ceramic Society, Singapore. Singapore: Arts Orientalis, 1979. Pamphlet. St. Andrew’s Cathedral’s Quiet Places Project. A project to construct an extension to the existing church building. Singapore: The QPP Building Committee, St Andrew’s Cathedral, 2003. Pelliot, Paul. Notes on Marco Polo. Volume II. Paris: Imprimerie Nationale, 1963. Pierson, Stacey. “Qingbai Porcelain: Technology, Forms and Decoration.” In Qingbai Ware: Chinese Porcelain of the Song and Yuan Dynasties, p. 15-23. Edited by Stacey Pierson. London: Percival David Foundation of Chinese Art, 2002. Quie, Malcom V. “A Chronological Study of Ceramics Excavated off the Sinan Sea Shore in 1976 and 1977.” In Chinese Celadons and other Related Wares in Southeast Asia, p. 43-7. Compiled by Southeast Asian Ceramic Society, Singapore. Singapore: Arts Orientalis, 1979. Raffles, Sophia. Memoir of the Life and Public Services of Sir Thomas Stamford Raffles. Singapore: Oxford University Press, 1991. Raffles, Thomas Stamford. The History of Java. Singapore: Oxford University Press, 1988. _____________________. “The Founding of Singapore.” JSBRAS 2 (December 1878): p. 17582. 139 Redfield, Robert & Singer, Milton B. “The Cultural Role of Cities.” EDCC 3,1 (October 1954): p. 53-73. Reid, Anthony. “The Structure of Cities in Southeast Asia, Fifteenth to Seventeenth Centuries.” JSEAS 11,2 (September 1980): p. 235-50. ___________. Southeast Asia in the Age of Commerce. 1450-1680. Volume One. The Land below the Winds. New Haven: Yale University Press. ___________. “Humans and Forests in Pre-Colonial Southeast Asia.” In Nature and the Orient. The Environmental History of South and Southeast Asia, p. 106-26. Edited by Richard H. Grove, Vinita Damodaran, Satpal Sangwan. Delhi: Oxford University Press, 1998. Reith, Charlotte. “A Comparison of Ground-Firing Techniques in Contemporary Myanmar Villages.” In Earthenware in Southeast Asia, p. 311-21. Edited by John Norman Miksic. Singapore: Singapore University Press, 2003. Renfrew, Colin & Bahn, Paul. Archaeology. Theories, Methods and Practice. Fourth Edition. London: Thames & Hudson, 2004. Rentse, Anker. “History of Kelantan. I.” JMBRAS 12, 2 (August 1934): p. 44-62. Rice, Prudence M. Pottery Analysis. A Sourcebook. Chicago: The University of Chicago Press, 1987. ______________. “On the Origins of Pottery.” JAMT 6,1 (March 1999): p. 1-54. Ricklefs, Merle Calvin, Lockhart, Bruce, Lau, Albert, Reyes, Portia & Aung-Thwin, Maitrii. A New History of Southeast Asia. Hampshire: Palgrave Macmillan, 2010. Rockhill, William Woodville. “Notes on the Relations and Trade of China with the Eastern Archipelago and the Coasts of the Indian Ocean during the Fourteenth Century. Part II.” TP Second Series 16,1 (March 1915): 61-159. 140 Rooney, Dawn F. Khmer Ceramics. Singapore: Oxford University Press, 1984. _____________. Folk Pottery in South-East Asia. Singapore: Oxford University Press, 1987. Sandhu, Kernial Singh & Wheatley, Paul. “The Historical Context.” In Melaka. The Transformation of A Malay Capital c. 1400-1980. Volume One, p. 3-69. Edited by Kernial Singh Sandhu & Paul Wheatley. Kuala Lumpur: Oxford University Press, 1983. de Silva, G. P. S. H. History of Coins and Currency in Sri Lanka. Colombo: Central Bank of Sri Lanka. Sri Soejatmi Satari. “A Preliminary Study on the Local Ceramics of Trowulan.” In Studies on Ceramics, p. 41-7. Edited by Departemen P dan K. Jakarta: Proyek Penelitian Purbakala Jakarta, 1984. Scott, Rosemary. “Introduction: Qingbai Porcelain and its Place in Chinese Ceramic History.” In Qingbai Ware: Chinese Porcelain of the Song and Yuan Dynasties, p. 6-12. Edited by Stacey Pierson. London: Percival David Foundation of Chinese Art, 2002. Shennan, Stephen. Quantifying Archaeology. Edinburgh: Edinburgh University Press, 1988. Shepard, Anna O. Ceramics for the Archaeologist. Washington D.C.: Carnegie Institution of Washington, 1954. Sheppard, Mervyn Cecil Franck. “A Short History of Trengganu.” JMBRAS 22,3 (June 1949): p. 1-74. Shippen, Mick. The Traditional Ceramics of Southeast Asia. Honolulu: University of Hawai’i Press, 2005. Sinopoli, Carla M. Approaches to Archaeological Ceramics. New York: Plenum Press, 1991. _______________. Pots and Palaces. The Earthenware Ceramics of the Noblemen’s Quarter of Vijayanagara. New Delhi: American Institute of Indian Studies, 1993. 141 Skibo, James M. “Pottery and People.” In Pottery and People. A Dynamic Interaction, p. 1-8. Edited by James M. Skibo & Gary M. Feinman. Salt Lake City: The University of Utah Press, 1999. Smith, Monica L. “The Role of Ordinary Goods in Premodern Exchange.” JAMT 6,2 (June 1999): p. 109-35. Solheim II, Wilhelm G. “The Use of Sherd Weights and Counts in the Handling of Archaeological Data.” CA 1,4 (July 1960): p. 325-9. _____________________. “Earthenware Pottery, the T’ai and the Malay.” AP 29,1 (1990): p. 25-36. _____________________. “Notes on “Malay Pottery” in East Malaysia and neighboring areas.” SMJ 29,50 (December 1981): p. 3-16. _____________________. “Southeast Asian Earthenware Pottery and Its Spread.” In Earthenware in Southeast Asia, p. 1-21. Edited by John Norman Miksic. Singapore: Singapore University Press, 2003. _____________________. Archaeology and Culture in Southeast Asia: Unraveling the Nusantao. Quezon City: The University of the Philippines Press, Diliman, 2006. Stargardt, Janice. “Kendi Production at Kok Moh, Songkhla Province, and Srivijayan Trade in the 11th Century.” In SPAFA. Final Report. Consultative Workshop on Archaeological and Environmental Studies on Srivijaya (T-W3). Bangkok and South Thailand. March 29-April 11, 1983, p. 181-9. Edited by Suchitra Vuthisathira. Bangkok: SPAFA Co-ordinating Unit, 1983. Stark, Miriam T. “The Chronology, Technology and Contexts of Earthenware Ceramics in Cambodia.” In Earthenware in Southeast Asia, p. 208-29. Edited by John Norman Miksic. Singapore: Singapore University Press, 2003. 142 Sullivan, Michael. “Notes on Chinese export wares in Southeast Asia,” TOCS 33 (1961-1962): 61-77. 孫仲匯 (Sūn, Zhòng Huì). 《中國古錢》 (Zhōng Guó Gǔ Qián or Ancient Chinese Coins). 台 北 : 藝術圖書公司, 1994. Teo, Catherine. “Qingbai Ware for Export.” In Qingbai Ware: Chinese Porcelain of the Song and Yuan Dynasties, p. 244-51. Edited by Stacey Pierson. London: Percival David Foundation of Chinese Art, 2002. Thio, Eunice. “Introduction.” In Singapore. 150 years, p. i-iv. Edited by Tan Sri Dato’ Mubin Sheppard. Singapore: Times Books International. Thomaz, Luis Filipe Ferreira Reis. “The Malay Sultanate of Melaka.” In Southeast Asia in the Early Modern Era. Trade, Power, and Belief, p. 69-90. Edited by Anthony Reid. Ithaca: Cornell University Press, 1993. Thomson, John Turnbull. “General report on the Residency of Singapore, drawn up principally with a view of illustrating its Agricultural Statistics.” JIAEA 3 (1849): p. 618-28, 744-55. Treloar, Francis Edward. “Stoneware Bottles in the Sarawak Museum: Vessels for Mercury Trade?” SMJ 20,40-1 (January-December 1972): p. 377-84. Turnbull, Constance Mary. A History of Modern Singapore. 1819-2005. Singapore: NUS Press, 2009. Vaughn, Kevin J. “Households, Crafts, and Feasting in the Ancient Andes: The Village Context of Early Nasca Craft Consumption.” Latin American Antiquity (LAA) 15,1 (March 2004): p. 61-88. Wake, C. H. “Melaka in the Fifteenth Century: Malay Historical Traditions and the Politics of Islamization.” In Melaka. The Transformation of A Malay Capital c. 1400-1980. Volume One, 143 p. 128-61. Edited by Kernial Singh Sandhu & Paul Wheatley. Kuala Lumpur: Oxford University Press, 1983. Wang, Gungwu. “The Nanhai trade: A Study of the Early History of Chinese Trade in the South China Sea.” In Southeast Asia-China Interactions, p. 51-166. Compiled by Geoff Wade. Kuala Lumpur: MBRAS, 2007. Wang, Qingzheng. A Dictionary of Chinese Ceramics. Singapore: Sun Tree Publishing Limited, 2002. Watt, James C. Y. “The Dating of Chinese Ceramics and Archaeological Sites in South-East Asia. A Preliminary Draft.” In Studies on Ceramics, p. 187-99. Edited by Departemen P dan K. Jakarta: Proyek Penelitian Purbakala Jakarta, 1984. Wicks, Robert S. Money, Markets, and Trade in Early Southeast Asia. The Development of Indigenous Monetary Systems to AD 1400. Ithaca: Cornell Southeast Asia Program, 1992. Wheatley, Paul. “Geographical Notes on some Commodities involved in Sung Maritime trade.” JMBRAS 32,2 (June 1959): p. 1-140. ____________. The Golden Khersonese. Studies in the Historical Geography of the Malay Peninsula Before A.D. 1500. Kuala Lumpur: University of Malaya Press, 1961. ____________. Impressions of the Malay Peninsula in Ancient Times. Singapore: Eastern Universities Press LTD, 1964. ____________. Nāgara and Commandery: Origins of the Southeast Asian Urban Traditions. Illinois: The University of Chicago, 1983. White, Joyce C. & Pigott, Vincent C. “From Community Craft to Regional Specialization: Intensification of Copper Production in Pre-state Thailand.” In Craft Specialization and Social Evolution: In Memory of V. Gordon Childe, p. 151-75. Edited by Bernard Wailes. Philadelphia: The University Museum of Archaeology and Anthropology, University of Pennsylvania, 1996. 144 Wicks, Robert S. Money, Markets, and Trade in Early Southeast Asia. The Development of Indigenous Monetary Systems to AD 1400. Ithaca: Cornell Southeast Asia Program, 1992. Wilkinson, Richard James. Papers on Malay Subjects. Malay Literature, Part 1. Romance. History. Poetry. Kuala Lumpur: Federated Malay States Government Press, 1907. _____________________. “The Malaccan Sultanate.” JSBRAS 61 (June 1912): p. 67-71. _____________________. A History of the Peninsular Malays with Chapters of Perak and Selangor. Singapore: Kelly & Walsh, Limited, 1920. _____________________. “Early Indian Influence in Malaysia.” JMBRAS 13,2 (October 1935): p. 1-16. _____________________. “Old Singapore.” JMBRAS 13,2 (October 1935): p. 17-21. Winstedt, Richard Olaf. “The Founder of Old Singapore.” JSBRAS 82 (September 1920): p. 127. ___________________. “Malay Works known by Werndly in 1736 A.D.” JSBRAS 82 (September 1920): p. 163-5. ___________________. “The Early History of Singapore, Johore & Malacca; an outline of a paper by Gerrit Pieter Rouffaer.” JSBRAS 86 (November 1922): p. 257-60. ___________________. “Gold Ornaments dug up at Fort Canning, Singapore.” JMBRAS 6,4 (November 1928): p. 1-4. ___________________. “A History of Johore (1365-1895).” JMBRAS 10,3 (December 1932): p. 1-167. ___________________. “A History of Selangor.” JMBRAS 12, 3 (October 1934): p. 1-34. ___________________. “Negri Sembilan. The History, Polity and Beliefs of the Nine States.” JMBRAS 12, 3 (October 1934): p. 41-111. 145 ___________________. “A History of Malaya.” JMBRAS 13,1 (March 1935): p. 1-270. ___________________. “A History of Malay Literature.” JMBRAS 17,3 (January 1940): p. 1243. ___________________. “Malay Chronicles from Sumatra and Malaya.” In Historians of South East Asia, p. 24-8. Edited by Daniel George Edward Hall. Oxford: Oxford University Press, 1961. Winstedt, Richard Olaf & Wilkinson, Richard James. “A History of Perak.” JMBRAS 12,1 (December 1932): p. 1-180. Wisseman-Christie, Jan. “Reviewed work(s): Early Tenth Century Java from the Inscriptions: A Study of Economic, Social and Administrative Conditions in the First Quarter of the Century by Antoinette M. Barrett Jones.” JSEAS 17,2 (September 1986): p. 370-2. ___________________. “Negara, Mandala, and Despotic State: Images of Early Java.” In Southeast Asia in the 9th to 14th Centuries,” p. 65-93. Edited by David G. Marr and Anthony Crothers Milner. Singapore: ISEAS, 1986. Wolters, Oliver William. “Śrīvijaya Expansion in the Seventh Century.” AAE 24,3/4 (1961): p. 417-24. ____________________. Early Indonesian Commerce. A study of the origins of Śrīvijaya Ithaca: Cornell University Press, 1967. ____________________. The fall of Śrīvijaya in Malay History. London: Lund Humphries Publishers Limited, 1970. ____________________. “Landfall on the Palembang Coast in Medieval Times,” Indonesia 20 (October 1975): p. 1-57. ____________________. History, Culture and Region in Southeast Asian Perspectives. Ithaca: Cornell University, 1999. 146 Wong, Grace. “Chinese Blue-and-White Porcelain and Its Place in the Maritime Trade of China.” In Chinese Blue & White Ceramics, p. 51-75. Compiled by S. T. Yeo & Jean Martin. Singapore: Arts Orientalis, 1978. ___________. “A Comment on the Tributary Trade between China and Southeast Asia, and the Place of Porcelain in this Trade, During the Period of the Song Dynasty in China.” In Chinese Celadons and other Related Wares in Southeast Asia, p. 73-94. Compiled by Southeast Asian Ceramic Society, Singapore. Singapore: Arts Orientalis, 1979. Wong, Lin Ken. The Trade of Singapore. Bandar Puchong Jaya: MBRAS, 2003. 楊勝德 (Yáng, Shèng Dé) & 馬紀恩 (Mǎ, Jì Ēn). “中國的青花瓷 (Zhōng Guó De Qīng Huā Cí or Blue-and-white Ceramics of China).” In Chinese Blue & White Ceramics, p. 76-83. Compiled by S. T. Yeo & Jean Martin. Singapore: Arts Orientalis, 1978. Ye, Wencheng, “A Preliminary Discussion of Fujian Wares Made In Imitation of Zhejiang Green Glazed Wares.” In New Light on Chinese Yue and Longquan Wares. Archaeological Ceramics Found in Eastern and Southern Asia, A.D. 800-1400, p. 120-8. Edited by Ho Chuimei. Hong Kong: The University of Hong Kong, 1994. Yeo, S. T. & Martin, Jean. “Chinese Blue and White Ceramics – A Brief Introduction.” In Chinese Blue & White Ceramics, p. 14-42. Compiled by S. T. Yeo & Jean Martin. Singapore: Arts Orientalis, 1978. 147 GLOSSARY The following is a selection of technical terms which may require some explanation for the non-specialist reader. Their definitions are either obtained from The Concise Oxford Dictionary of Archaeology (CODA), or seminal works by various scholars as listed in the bibliography (Author, Year) respectively. 8 ARTEFACT Any object which has been modified, fashioned, or manufactured according to a set of humanly imposed attributes, including tools, weapons, ornaments, utensils, houses, buildings, etc. Artefacts are the basic components of material culture (CODA). ECOFACT Strictly, natural materials that have been used by humans, for example the remains of plants and animals that were eatern by a given community. More generally taken as material recovered from archaeological sites, or other sealed deposits, which is relevant to the study of ancient environments and ecology. Examples include animal bones, seeds, snail shells, waterlogged wood, and pollen (CODA). 8 Timothy Darvill. The Concise Oxford Dictionary of Archaeology. Second Edition. Oxford: Oxford University Press, 2008. 148 FEATURE Recurent patterned arrangements of archaeological contexts forming an interpretative category recorded during an excavation. Positive features are defined in terms of constructions or accumulations, such as walls, floors, buildings, middens, banks, etc. Negative features are defined in terms of holes left by the removal of material such as pits, ditches, wells, postholes, etc (CODA). FORMATION PROCESSES Cultural formation processes involve the deliberate or accidental activities of human beings as they make or use artefacts, build or abandon buildings, plow their fields and so on. Natural formation processes are natural events that govern both the burial and the survival of the archaeological record (CODA). MATRIX The material or sediment in which cultural debris is contained; the surrounding deposit in which archaeological finds are situated (CODA). PARADIGM A term used and popularized by Thomas Kuhn to refer to a common set of philosophies and methods shared by a scientific [or academic] community within which basic assumptions and orientations are left uncriticized. When competing approaches successfully challenge and replace an existing position there is said to be a ‘paradigm shift,’ as for example between Newtonian and Einsteinian physics (CODA). PROVENANCE The place where an object was found or recovered in modern times; the findspot (CODA). [Also used in this thesis to refer to an object’s place or source of origin] POLITY Generally, a complex of decision-making roles that relate a society to the goals of its individual members through collective decisions. 149 Colin Renfrew used the term in a particular archaeological context to refer to small-scale politically autonomous early states such as can be seen in Mycenaean Greece or Etruscan Italy (CODA). STRATIGRAPHY The study and validation of stratification; the analysis in the vertical time dimension, of a series of layers in the horizontal, space dimension. It is often used as a relative dating technique to access the temporal sequence of artefact deposition (Renfrew & Bahn, 2004). UNIT A basic category used in describing the components in a stratigraphic sequence or for grouping comparable artefacts together when describing and analysing them (CODA). 150 APPENDIX A. STA ARTEFACT SAMPLES: ANALYSIS AND DEDUCTIONS ANALYSIS I. Earthenware As can be expected in a pre-colonial Southeast Asian settlement site, earthenware sherds occupy the largest proportion of the sampled STA artefacts in terms of count (5060 sherds; 58.3%), but second in terms of weight (10,883g; 20.4%). The average earthenware sherd size is 0.46Pc/g (or 46 pieces per 100g), the smallest of three ceramic categories. Most of the earthenware sherds are classified into three varieties – ‘coarse-tempered,’ ‘mediumtempered’ and ‘fine-paste’ – on the basis of the presence (or absence) and size of observable temper within the sherd fabric as well as the sherd surface texture, whereas the remainder fall into a fourth ‘high-fired’ variety on the basis of sherd surface texture and hardness. Coarse-tempered sherds are defined by the presence of very coarse quartz and/or sand temper with particle diameters over 2mm embedded within the sherd fabric, hence a very rough surface texture. 9 This variety is similar to ‘Type A4’ and ‘Type D’ earthenware as recorded by Miksic in his research on artefacts recovered at FTC as well as ‘coarse’ sherds classified by Chen in his study of earthenware found at the PHC site. 10 This variety of sherds has a hardness rating of 3-5 on the Moh’s scale. 11 A total of 124 sherds weighing 991g are of this variety, 9 See Figure 58. Miksic, Archaeological Research, p. 57, 61, 63-4; Miksic, “Beyond the Grave,” p. 40-2; Chen, Earth to Earth, p. 79-93. 11 See Figure 60. 10 151 second in average sherd size (0.13Pc/g) and third in proportion (2.5% in count; 9.1% in weight) of the four varieties. Surface colours vary mostly within the hues of brown, red and yellow, the remaining few in gray. 12 However, cross-section colours are either the same as the surface, which indicates the presence of an oxidizing atmosphere in the firing process, or range between dark gray (10YR 4/1) to gray (10YR 5/1), which indicates an incomplete oxidizing or reducing atmosphere instead. 13 A total of 99 sherds (464g; 0.21Pc/g) consist of pottery fragments in the form of rim, carination, body and base sherds. Sherd thickness varies from 0.23-1.4cm and most have plain surfaces save for a few decorated with reliefs (rib-lines), carvings (grooves and intersecting-combed lines) and carved-paddle impressions. Only two varieties of impressions, ‘Saw-Tooth’ and ‘Saw-Tooth with Flame’ pattern, are distinct whereas the rest are too worn for clear description and therefore listed as ‘general.’ Rims have ‘round,’ ‘bulbed’ and ‘square’ edges, external diameters of 14-30cm and are generally everted in shape. As seen in Table II, only nine vessels of this variety (2.5% of 365 earthenware vessels) were estimated from the rim sherds. No coarse-tempered vessels were recovered from the NW quadrant; whereas distribution variation of this vessel variety among the NE (4), SE (3) and SW (2) quadrants is little. Only one vessel-base with a thickness of 0.59cm, external diameter of 22cm and flat in shape was recovered, which suggests that the majority of coarse-tempered vessels had round bases. No lid or spout sherds of this variety were recovered. Eight sherds (240g; 0.03Pc/g) appear to be brick or tile fragments, as they are exceptionally large and heavy pieces with sherd thickness from 0.49-2.44cm and most have a smoothened surface on one side. More complete fragments were found on FTC which Miksic originally suggested were eavesboard tiles (tuiles de chanlatte), thus evidence for the existence of tile-roof structures in 14th century Singapore. 14 Seventeen sherds (287g; 0.06Pc/g) remain indistinct but probably belong to either of the previous two categories. Two of these sherds however bear striking resemblance to an oven 12 Light red (2.5YR 7/8), red (2.5YR 4/8 & 5/8), reddish-brown (2.5YR 5/4), dark reddish-brown (5YR 3/4), yellowish-red (5YR 4/6, 5/6 & 5/8), reddish-yellow (5YR 6/6, 6/8 & 7/6, 7.5YR 8/6), brown (7.5YR 5/4, 10YR 5/3), very pale brown (10YR 7/4), light yellowish-brown (10YR 6/4) & dark gray (10YR 4/1). 13 Anna O. Shepard, Ceramics for the Archaeologist (Washington D.C.: Carnegie Institution of Washington, 1954), p.104-6; Rice, Pottery Analysis, p. 343-6. 14 Miksic, Archaeological Research, p. 63-4; he is now less certain of the accuracy of this identification (Miksic, personal communication, 23 December 2011). 152 trivet used to support pottery over the traditional fuels of wood and charcoal. The only noticeable decoration is an incised line running along the borders of the sherds which outlines the shape of the ‘trivet.’ This relatively intricate and unique design vis-à-vis other coarsetempered sherds also raises the possibility that it functioned as a building or structural ornamentation. Medium-tempered sherds are also defined by the presence of quartz and/or sand temper, but with particle diameters between 1 16 -2mm in the sherd fabric. As the temper particles are much smaller, they seldom protrude out of the sherd body which results in a slightly smoother surface texture than coarse-tempered sherds but abrasive to touch nonetheless. This variety corresponds to Miksic’s ‘Type A1, ‘Type B’ and ‘Type C’ earthenware at FTC and Chen’s ‘Medium Coarse’ sherds at PHC. 15 Medium-tempered sherds also have a hardness rating of 35 on the Moh’s scale. A total of 3454 sherds weighing 7813.5g are of this variety, third in average sherd size (0.44Pc/g) but largest in proportion (68.3% in count; 71.8% in weight) of all four varieties. Surface colours vary mostly within the hues of brown, yellow and red, with a few in gray and black. 16 Cross-section colours are again either the same as the surface or range between dark gray to gray due to either oxidizing or incomplete oxidizing and/or reducing atmospheres in the firing process respectively. In the case of gray and black sherds however, the similarity in their surface and cross-section colours suggests the presence of highly carbonaceous clay in the sherd fabric and/or the reduction of ferric oxide within the fabric in a reducing environment. 17 All medium-tempered sherds consist of pottery fragments primarily in the form of rim, carination and body fragments, while a few are lid (with knob) and spout fragments. Sherd thickness varies from 0.14-2.09cm. Most of these sherds are decorated with reliefs (rib-lines and foliated), carvings (grooves, lines, ring, linear and intersecting-combed lines) and carved-paddle impressions with a large variety of geometrical and morphological 15 Ibid., p. 56, 60; Chen, Earth to Earth, p. 79-93. Very pale brown (10YR 7/4), light brown (7.5YR 6/4), brown (7.5YR 4/2 & 5/2), yellowish-red (5YR 4/2), reddish-yellow (5YR 7/6), red (2.5YR 5/6), very dark gray (2.5Y 3/1) & dark gray (2.5Y 4/1). 17 Shepard, Ceramics for the Archaeologist, p.104-6; Rice, Pottery Analysis, p. 343-6. 16 153 patterns (‘Chevron,’ ‘Linear,’ ‘Curvilinear,’ ‘Rectangle,’ ‘Box,’ ‘Box-and-Cross,’ ‘Box-andCross with Triangular motif,’ ‘Box-Crosses,’ ‘Box-Diagonal,’ ‘Box-and-Square,’ ‘Box-andDot,’ ‘Herring-Bone,’ ‘Palm-Frond,’ ‘Wreath,’ ‘Petals,’ ‘Sun-Flower,’ ‘Saw-Tooth,’ ‘Bulb’ and ‘Mat’). Carved-paddle impressions are labelled ‘general’ when the precise pattern cannot be accurately discerned. Rim sherds have edges which are ‘round,’ ‘bulbed,’ ‘inverse-bulbed,’ ‘even-bulbed,’ ‘square,’ ‘Y-shaped’ and ‘hooked’ in shape, have external diameters from 540cm and are generally everted in shape. From Table II, 232 vessels of this variety (63.6% of 365 earthenware vessels) were estimated from the rim sherds. The SE quadrant has the largest number of medium-tempered vessels (86), followed by the NW (61), NE (48) and SW (37) quadrants. Only two vessel-bases with thickness of 0.3-0.83cm, a flat or concave shape and base diameters of 10cm were estimated from base sherds, thereby suggesting again that most medium-tempered vessels had round bases which are virtually indistinguishable from the rest of the body sherds. Seven lids, some with knobs attached, with external diameters of 7-8cm were recovered from the SE (3), SW (3), and NW (1) quadrants. Given the narrow range of lid external diameters and the fact that only vessels with the same rim measurements can wear these lids, it stands to reason that this particular range of lidded medium-tempered vessels was specifically made and used as storage vessels. Only one straight-tapered spout sherd, of this variety was recovered on the site in the SE quadrant, evidently used as part of a generally spouted, high-necked water-pouring vessel known as a ‘kendi,’ an Indonesian word derived from the Sanskrit term, ‘kundika.’ 18 A set of 241 sherds (1137g; 0.21Pc/g) within the medium-tempered variety have an approximate 1mm layer of dark grayish-brown (10YR 4/2) slip coated on either or both sides of a red (2.5YR 5/6) sherd body, resulting in a smoother surface texture than other mediumtempered sherds. This ‘slipped’ variety of medium-tempered earthenware is also much harder and scratch-resistant as a result; it has a rating of 5-5.5 on the Moh’s scale, much harder than the other medium tempered earthenware. Sherd thickness ranges from 0.19-0.87cm and 18 Khoo Joo Ee, Kendi. Pouring Vessels in the University of Malaya Collection (Singapore: Oxford University Press, 1991), p. 4. 154 decorated with horizontal rib-line reliefs and groove carvings only. Its average sherd size of 0.21Pc/g is also twice as large as that of all medium-tempered sherds combined. Rims have edges ‘round,’ ‘bulbed’ and ‘square’ in shape and have particularly flanged and everted bodies. Rim external diameters, from 20-40cm, are also relatively larger than the rest of the mediumtempered variety. A total of six such vessels among the 232 medium-tempered vessels can be discerned from rim sherds and were distributed between the NW (1), SE (4) and SW (1) quadrants. No similar base, lid or spout sherds were found; therefore it is safe to assume that they are round vessels, and possibly some of the largest on site as well. These distinctive features, coupled with its small numbers, distinguish this sub-variety from the rest of the medium-tempered vessels in form and therefore, suggest the possibility of a unique function as well. Conversely, fine-paste sherds are defined by the presence of very fine temper of diameter under 1 16 mm to the relative absence of visible temper within the sherd fabric. This suggests the use of a higher grade or better clay composition which can withstand thermal stresses during and after the firing process without the aid of temper. 19 As a result, they have an extremely smooth and chalky surface texture. With a Moh’s rating of 2-3, this variety is also relatively softer than coarse- and medium-tempered sherds. Fine-paste sherds correspond to Miksic’s ‘Type A2’ and ‘Type A3’ earthenware sherds at FTC as well as ‘Fine Paste’ sherds in Chen’s research at PHC. 20A total of 1450 sherds weighing 1649.5g are of this variety, the smallest in average sherd size (0.88Pc/g) but second in proportion (28.7% in count, 15.2% in weight) of the four varieties. Surface colours vary between the hues of brown, yellow and red only. 21 It should be noted that Mikisc’s fine-paste category excludes sherds with a red clay body, but this group of sherds have been included in the fine-paste category in here on the basis of its definition in this present thesis as stated above. Cross-section colours are also either the same as the surface or range between dark gray to gray due to either oxidizing or incomplete 19 Rice, Pottery Analysis, p. 72-5, 229, 406-7. Miksic, Archaeological Research, p. 56-7; Chen, Earth to Earth, p. 79-93. 21 Light brown (7.5YR 5/6), pink (7.5YR 7/4), yellowish-red (5YR 5/6), reddish-yellow (5YR 7/6) & red (2.5YR 5/6). 20 155 oxidizing and/or reducing atmospheres in the firing process respectively. All fine-paste sherds are pottery sherds mainly in the form of rim, body and base fragments, while the remainder consist of a few spout sherds. Sherd thickness varies from 0.10-0.97cm, most of which are undecorated; the few sherds which are have a limited variety of patterns: reliefs (rib-lines), punctuations (small circular depressions in a line), carvings (grooves and lines), incisions (lines) and a unique combination of punctations and carvings which produces a ‘Dot-and-Ring’ motif; two sherds also have a layer of very dark brown (7.5YR 2.5/2) paint or glaze whereas nine sherds have a dark grayish-green (5GY 3/2) or ‘apple-green’ glaze on it. Rim edges are ‘round,’ ‘bulbed,’ ‘inverse-bulbed,’ ‘even-bulbed,’ ‘square,’ ‘hooked’ and ‘flat-angular’ with external diameters of 4-24cm and everted bodies. As presented in Table II, 122 vessels of this variety (33.4% of 365 earthenware vessels) were estimated from the rim sherds. Most of these vessels were concentrated in the SE Quadrant (94), while there is little vessel variation in the remaining quadrants (NE: 13; NW: 8; SW: 7). There are 37 vessel-bases of this variety with thickness from 0.17-0.86cm, a flat or concave shape and external diameters of 2-8cm estimated from base sherds. This suggests that a significant proportion of fine-paste vessels have horizontal bases. The interior surfaces of some body and base sherds have ‘rilling’ – rhythmic ridges and grooves or striations that spiral around the vessel wall and base – which indicate the wheel-throwing technique of producing these vessels. 22 Ten kendi-spouts, all straight-tapered in shape, were also present among the sherds, with the majority from the SE quadrant (8) and the remainder from the NE Quadrant (2). One spout even appears to have a ‘rifled’ interior. High-fired sherds, the fourth variety of earthenware recovered, are defined primarily by their hardness. With a Moh’s rating greater than five, it is the hardest of all four earthenware varieties. This indicates that this variety of sherds were produced at a firing temperature significantly higher than the rest of the earthenware sherds, possibly at temperatures close to 1200-1350ºC, which leads to partial fusion or vitrification of the clay body that usually results 22 Rice, Pottery Analysis, p. 128-32. 156 in the production of stoneware ceramics. 23 Firing temperatures within this range are usually only attainable in a kiln, a technology so far identified only in limited regions and societies of pre-colonial Southeast Asia. 24 Just like the fine-paste variety, no visible temper is found within the fabric as well. No such sherds were reportedly found on either the FTC or PHC sites, a rarity attested to also by their few numbers on the STA site. Only 32 sherds weighing 429g are of this variety, the largest in average sherd size (0.07Pc/g) but the smallest in proportion (0.6% in count, 3.9% in weight) among all four earthenware varieties. Even so, there are two distinct high-fired sub-varieties: ‘black-burnished’ (10 sherds; 25g; 0.4Pc/g) and ‘mercury-jar type’ (22 sherds; 404g; 0.05Pc/g). The only commonality between the two sub-varieties is that the interior body and base sherd surfaces of both have rilling which indicate again the wheelthrowing technique of producing these vessels. Black-burnished sherds have surface and crosssection colours of very dark gray (2.5Y 3/1) and a very smooth, fine and lustrous surface texture which is usually a result of burnishing, a surface-finishing technique for pottery. 25 Again, these colours suggest the presence of highly carbonaceous clay in the sherd fabric and/or the reduction of ferric oxide within the fabric in a reducing environment. All black-burnished sherds are pottery fragments with sherd thickness of 0.26-0.55cm and decorated with reliefs (rib-lines and), groove carvings, incisions (‘Circle-and-Ring’) and carved-paddle impressions. Rims have a ‘round’ edge shape, external diameters of 10-12cm and everted bodies. No base, lid or spout sherds were recovered of this sub-variety. Only two ‘black burnished’ vessels were estimated from rim sherds in the NE quadrant; they are also the only vessels representing ‘highfired’ sherds on Table II. On the other hand, ‘mercury-jar type’ sherds are so defined because of their close resemblance to a distinct variety of stoneware vessels that bears the same name, especially the distinctively defined rilling which protrude sharply from the interior surfaces of the body sherds as well as their thick and heavy base. They usually have surface colours within the hues of brown or gray on the exterior, and red on the interior as well as the cross-section. 26 23 Ibid., p. 6, 103. See below section, “Earthenware: Locally-made Vessels, Regionally-imported Goods.” Ibid., p. 138. 26 Exterior: brown 7.5YR 5/2 & dark gray 2.5Y 4/1; Interior and cross-section: light reddish-brown (2.5YR 7/4) & pink (5YR 7/4). 24 25 157 This suggests an incomplete to relatively well oxidizing atmosphere during the firing process. 27 This sub-variety is also distinguished by a relatively heavier (four sherds; 162g; 0.02Pc/g) and thicker (0.79-1.07cm) base, flat in shape and with external diameters of 6-8cm. Sherd thickness range from 0.39-1.48cm and are all undecorated. No ‘mercury-jar type’ rim sherds were recovered, hence they are unrepresented on Table II. However, a total of three such vesselbases were estimated from the existing base sherds and were located at the NE (2) and SE (1) quadrants. No lid or spout sherds of this sub-variety were recovered as well. II. Stoneware Stoneware sherds form the second largest number of sampled STA artefacts in count (2516 sherds; 29.0%), but occupies the largest proportion in terms of weight (32,097.5g; 60.1%). All stoneware sherds are pottery fragments with an average sherd size of 0.08Pc/g (or 8 pieces per 100g), the largest among the three ceramic categories, and have a hardness rating greater than five. These sherds are mainly classified into 2 varieties – brittle and buff – primarily on the basis of the presence and density by proportion of observable inclusions, ‘grit’ or ‘grog’ as Miksic calls it, in the sherd fabric. 28 Sherds are considered brittle when these inclusions are over 2mm in diameter, typically black in colour (but can also be brown or white) and occupy approximately 10-15% in proportion of the sherd surface. This pattern of inclusion is given the label ‘Type A’ in the artefact data tables. Consequently, brittle sherds have an extremely coarse and abrasive surface texture. On the other hand, buff sherds have inclusions, usually brown, white or clear in colour (but can also be black) that are less than 2mm in diameter and occupy approximately 10-15% of the sherd surface, or otherwise noticeably absent from the sherd fabric. These inclusion patterns, labelled ‘Type B’ and ‘Type C’ respectively, give buff sherds a smoother and often chalky surface texture. 29 While all stoneware sherds have a Moh’s rating greater than five, brittle sherds are generally harder and sharper than their buff counterparts. Brittle sherds here correspond to those labelled similarly by Miksic at the FTC site, whereas 27 Shepard, Ceramics for the Archaeologist, p.104-6; Rice, Pottery Analysis, p. 343-6. Miksic, Archaeological Research, p. 68. 29 See Figure 59. 28 158 buff sherds are similar to those designated by Miksic as ‘Kwantung Ware.’ Miksic’s categories were themselves adopted from Eine Moore’s seminal study of stoneware ceramic categories. 30 Brittle sherds are also generally thinner and lighter than their buff counterparts. Regardless, rilling marks are commonly found on the interior surface of body and base sherds of both stoneware varieties. Majority of the stoneware sherds (70.8% in count; 69.5% in weight) are general pottery fragments consisting of 738 brittle sherds (7156g; 0.10Pc/g) and 1043 buff sherds (15,159.5g; 0.07Pc/g) with Type A, B and C inclusion patterns. Most of the stoneware here are of the buff variety. Surface and cross-section colours are uniform and limited to the hues of brown and gray, with the occasional sherd having a contrasting cross-section colour of weak red (2.5YR 5/2) or very pale brown (10YR 7/3). 31 This suggests the presence of an oxidizing atmosphere in the firing process typically achieved in a kiln. Sherd forms consist of rim, body and base fragments with thickness varying from 0.19-1.40cm. Most of these sherds are plain, save for a few with reliefs (rib-lines and zoomorphic motifs), carvings (grooves and lines) and incisions (lines). Some sherds have complete or partial ‘lugs’ which were moulded separately and then attached or ‘luted’ onto the vessel body, while another handle-piece had a hollow, cylindrical body which resembles that commonly found on medicinal clay pots. One sherd has a circular depression and hole in the middle which appears to be an opening into a spout while another has a flat and circular knob. Despite the paucity of decorative features, a large number of sherds have green, olive, brown, gray and black hues of glaze as well as brown (7.5YR 4/2) paint. 32 Three body sherds and two rim sherds, all of the buff variety, are also adorned with stamped impressed Chinese characters on the exterior surface. The characters are unfortunately too worn to be read, but could possibly depict the name of the specific kiln or product originally stored within it. Rims have edges which are ‘bulbed,’ ‘hooked’ or ‘flat-angular’ in shape with some 30 Ibid., p. 67-75. Very pale brown (10YR 7/3), light brownish gray (2.5Y 6/2) & gray (2.5Y 5/1 & 6/1). 32 Glaze: greenish black (10Y 2.5/1), very dark grayish olive (5GY 3/2 & 10Y 3/2), pale olive (5Y 6/4), olive (5Y 4/4), dark olive gray (5Y 3/2), olive gray (5Y 4/2 & 5/2), olive brown (2.5Y 4/6), dark olive brown (2.5Y 3/3), very dark brown (7.5YR 2.5/2), dark brown (7.5YR 3/4), brown (10YR 4/3), very dark grayish brown (2.5Y 3/2, 10YR 3/2), very dark gray (2.5Y 3/1 & 5YR 3/1), reddish-black (2.5YR 2.5/1) & black (5Y 2.5/1). 31 159 variations, have external diameters from 8-36cm as well as everted bodies. Specifically, the ‘bulbed-rectangular’ and ‘flat-bulbed’ varieties have larger external diameters (28-42cm) than the rest of the rim sherds (6-24cm). 33 From Table III, 61 general stoneware vessels (64.9% of 94 stoneware vessels) were estimated from the rim sherds. The SW quadrant has the most stoneware vessels (21), followed by the NE (16), SE (15) and NW (9) quadrants. A total of 58 vessel-bases with thickness of 0.23-1.15cm, flat or concave base shapes and base diameters of 4-18cm were estimated from base sherds. The preponderance of such bases suggests that most, if not all, general stoneware vessels have horizontal bases. ‘Mercury-jar’ sherds form the next variety of stoneware artefacts within the STA artefact sample population. Commonly found on FTC as well as other sites along the Singapore River, they are defined by their distinct vessel shape. 34 As Miksic describes, mercury-jars are ‘distinctive and easily recognized’ as they have ‘heavy, flat, narrow bases, and small round neckless mouths with perfunctory rims.’ 35 Mercury-jar sherds comprises of 643 brittle sherds (8167g; 0.08Pc/g) and 81 buff sherds (1426g; 0.06Pc/g), making up the second largest variety of stoneware sherds (28.8% in count, 29.9% in weight). These sherds exhibit only Type A and B inclusion patterns with surface colours similar to general stoneware sherds, but almost all mercury-jar sherds are of the brittle variety. Cross-section colours are also uniform with the exception of a few sherds carrying a contrasting pale brown (10YR 7/3) or light brownish gray (2.5Y 6/2) colour. Sherd forms consist of rim, body and base sherds with thickness varying from 0.21-1.16cm. Like general stoneware sherds, most of the mercury-jar sherds are undecorated but a few carry spots of olive glaze on the exterior surface. 36 Another distinctive feature of mercury-jar sherds is that they have extremely defined rilling which protrude sharply from the interior surfaces of the body sherds. Rim edges are consistently ‘flat-angular’ with external diameters within a narrow range of 1.72-2.58cm. As seen on Table III, 31 mercuryjars (33.0% of 94 stoneware vessels) were estimated from the rim sherds. The SE quadrant has 33 ‘Bulbed-Rectangular,’ ‘Bulbed-Rectangular with folded ‘Pie-Crust’ fringe,’ ‘Hooked Bulbed-Rectangular,’ & ‘Flat-Bulbed.’ Low, “Singapore from the 14th to 19th Century,” in Early Singapore, p. 20. Miksic, Archaeological Research, p. 68. 36 Olive brown (2.5Y 3/3 & 4/4) & olive (5Y 4/4). 34 35 160 the largest concentration of mercury-jars (16), followed by the NW (7), SW (6) and NE (2) quadrants. A set of 24 vessel-bases with thickness of 0.3-1.3cm, base diameters of 4.81-6.26cm and flat in shape were estimated from base sherds. One base sherd has a stray stoneware fragment, most likely a kiln waster, and bits of olive (10Y 5/4) glaze attached to its exterior. ‘Purple-ware’ sherds, defined by the purplish colour of their surface, are the last and smallest variety of stoneware artefacts (0.4% in count, 0.6% in weight). 37 Only three buff sherds (44g; 0.07Pc/g) and eight brittle sherds (145g; 0.06Pc/g) make up all purple-ware sherds. Despite having Type A, B and C inclusion patterns, their surface texture regardless is the roughest, almost sandpaper-like, among all three stoneware varieties. Sherd forms consist of rim and body fragments with thickness of 0.3-0.71cm. Most purple-ware sherds are undecorated, save for a couple of sherds with a single rib relief and moulded handle piece. Some sherds also have a layer of glaze within the hues of olive and brown. 38 Besides their surface colour, purple-ware sherds are also distinguished by the shape of their rims which are ‘round,’ ‘inversed-tapered’ and everted with external diameters of 8-10cm. As seen on Table III, only two purple-ware vessels (2.1% of 94 stoneware vessels) were estimated from rim sherds and both located at the SE quadrant. As there are only a small number of purple-ware sherds and no base sherds were found in the STA artefact sample population, it is not possible to determine the approximate shape of purple-ware vessels. III. Porcelain Porcelain sherds form the third largest proportion of the STA artefact sample pool in both count (677 sherds; 7.8%) and weight (6528g; 12.2%). Like stoneware sherds, all porcelain sherds are also pottery fragments but with an average sherd size of 0.10Pc/g (or 10 pieces per 100g), the second largest among the three ceramic categories, with a hardness rating greater than five as well. Three varieties of porcelain – green-ware, white-ware and blue-and-whiteware – can generally be discerned according to the colour of their glaze, a standard method for 37 38 Ranging between weak red (2.5YR 5/2), dark reddish-brown (5YR 3/2) & gray (2.5Y 6/1). Very dark grayish olive (5GY 3/2 and 10Y 3/2) & dark brown (7.5YR 3/4). 161 basic porcelain identification. The remaining sherds are either misfired or degraded porcelain fragments; but as it is difficult to differentiate between the two, they are grouped together under a fourth variety labelled ‘misfired or degraded.’ Their glaze colour generally lies between the hues of brown and gray which does not match those of the previous three porcelain varieties. The sherds are also blemished with pin-holes, cracks and multiple crease-like grayish streaks across the glaze surface. Some sherds of this variety also have ‘flaking’ glaze which loses adhesiveness and falls off the sherd body, whereupon fragmentation and burial over the centuries likely expedited this degradation. Consequently, these sherds generally have a rougher glaze surface than the other porcelain varieties. The sherd body cross-section colours of all four porcelain varieties vary between the hues of white and gray, and unlike earthenware and stoneware ceramics, do not contain any visible inclusions. However, stray sand granules, probably accidentally attached during the firing process, are sometimes found embedded between the sherd surface and glaze layer. A total of 292 green-ware sherds (3495g; 0.08Pc/g) occupy the largest proportion of total porcelain sherds (43.1% in count, 53.5% in weight) within the STA artefact sample pool. This variety corresponds to Miksic’s ‘Green Wares’ or ‘Celadons,’ a term which ‘refers to a colour varying from brownish green to greenish blue’ at the FTC site, some of which were associated with the Longquan (龙泉 Lóng Quán) kilns of Zhejiang (浙江 Zhè Jiāng) province. 39 Sherd forms consist of rim, body and base fragments with thickness varying from 0.2-1.83cm. Decorations on the sherd surface consist of reliefs (fluted, foliated, lotus petal (莲瓣纹 Lián Bàn Wén), bulbed and floral), incised lines and carvings (linear, curvilinear, grooves and swirl). Relief decorations are labelled ‘general’ when the pattern itself cannot be clearly discerned. Rim edges are ‘round,’ ‘bulbed’ and ‘foliated’ and have external diameters from 3-44cm. As seen on Table IV, 78 green-ware vessels (39.2% of 199 porcelain vessels) were estimated from the rim sherds. These vessels are distributed among the NE (21), NW (16), SE (23) and SW (18) quadrants without much variation. Vessel forms, where discernible, include dishes, plates, 39 Miksic, Archaeological Research, p. 76-81. 162 cups, bowls as well as jarlets. 25 vessel-bases with thickness of 0.25-1.66cm were also estimated from the base sherds. With the exception of two small vessels with flat bases of 36cm in diameter, all green-ware vessel-bases have an attached foot-ring each which measure 0.2-1.19cm in height, 3-20cm in diameter and are either ‘flat angular’ or ‘recessed’ in shape. Regardless, the exterior surface of the bases and foot-rings are completely glazed, unglazed or partially glazed. The exterior medallion surface, together with the unglazed parts of the footring, is sometimes painted maroon or light brown in colour. On the other hand, the interior medallion surface of seven vessel-bases has an unglazed circular band known as the ‘stacking ring,’ while the rest are completely glazed. A circular green-ware sherd weighing 15g, 0.62 cm thick and approximately 3-3.5cm in diameter stands out from the rest of the vessel sherds. Decorated with underglaze lines and foliated reliefs on the surface, it was clearly once part of a bowl or dish which was deliberately hacked to produce this shape. Porcelain tokens like this are known as ‘gacuk’ in Indonesian and found across pre-colonial Southeast Asia settlement sites. White-ware sherds are the third largest porcelain variety in the STA artefact sample pool (24.5% in count, 15.8% in weight). This variety corresponds to Miksic’s ‘White Wares’ which covers the ‘Qingbai,’ ‘Shufu’, ‘Dehua’ and ‘Ding-type’ sub-varieties. 40 There are a total of 166 sherds (1034.5g; 0.16Pc/g) measuring 0.14-1.27cm in rim, body and base forms. Decorations on the sherd surface consist of reliefs (rib-lines, double rib-line with dotted protrusions, foliated and floral) and carvings (grooves and curvilinear lines). Eight sherds are also decorated dark olive brown (2.5Y 3/3) or black (5Y 2.5/2) underglaze spots. Relief decorations are again labelled ‘general’ when the pattern itself cannot be clearly discerned. Rim edges are ‘round,’ ‘flat-angular’ or ‘foliated’ with smaller external diameters of 8-22cm. From Table IV, 54 white-ware vessels (27.1% of 199 porcelain vessels) were estimated from the rim sherds. A large proportion of white-ware vessels are concentrated at the SW (22) quadrant, 40 Ibid., p. 81-4. 163 while the remainder are distributed relatively evenly between the NE (11), NW (10), SE (11) quadrants. Discernible vessel forms include cups, bowls and small covered boxes which are round in shape and shallow in depth. A thin unglazed strip usually occurs on the rim edges of the boxes and lids. Only 10 vessel-bases with thickness of 0.22-0.94cm were estimated from the base sherds. All covered boxes and some miniature vessels have flat bases with external base diameters of 4-8cm. The exterior surface of these flat bases is usually unglazed. The remaining white-ware vessel-bases have attached foot-rings measuring 0.18-1.79cm in height, 6-8cm in external diameter and ‘round’ or ‘flat-angular’ in shape. The exterior surface of these foot-rings and their respective exterior base surfaces are typically completely glazed, unglazed or partially glazed. One foot-ringed vessel-base also has a ‘stacking ring’ on the interior medallion surface, while the rest are completely glazed. One such base sherd also appears to be a gacuk as its edges appears to be neatly cleaved around the circular foot ring while leaving the centre medallion intact. While the fragmented nature of the white-ware sherds makes it almost impossible to distinguish the different sub-varieties present, five pieces (36g; 0.14Pc/g) measuring 0.16-0.37cm in thickness could be positively identified as ‘Dehua (德化 Dé Huà) type’ ware similar to those recovered from FTC, as they have a ‘creamy offwhite tending to yellow or beige’ glaze as well. 41 Two were rim sherds with external diameters of 12-18cm and ‘flat angular’ in shape, whereas one was a base sherd with a flat base 0.22cm thick, 7cm in external diameter and a foliated relief on the exterior surface. Like the Dehua sherds which Miksic found on FTC, these sherds appear to be the same kind of bowls and covered boxes with ‘small flat-bottomed bowls, with moulded decoration on the exterior of the cavettoes in the form of long triangular foliage conventionally called plantain leaves (焦叶纹 Jiāo Yè Wén).’ 42 Two such vessels estimated from the rim sherds were located at the NE quadrant, while one vessel estimated from the base sherd was located at the SE quadrant. 41 ‘Dehua’ refers to a known category of porcelain made by kilns in 德化 (Dé Huà), a county in Quanzhou province, China; here with a very pale yellow (2.5Y 9.5/2 & 9/2) glaze. 42 Miksic, Archaeological Research, p. 84. 164 Blue-and-white sherds form the smallest proportion of porcelain sherds sampled from the STA site (6.8% in count, 3.5% in weight) with only 46 sherds (228g; 0.20Pc/g) found within the sample artefacts. This variety corresponds to Miksic’s ‘White Wares with Underglaze Blue’ and is distinguished from the white-ware variety by the application of cobalt blue motifs beneath the glaze against a white background as a decorative feature. 43 Blue-and-white sherds are comprised of primarily rim, body, base sherds and one lid fragment with thickness of 0.180.9cm. Decorations on these sherds usually consist of a combination of underglaze cobalt blue flora (foliage, blackberries, chrysanthemum flowers (菊花 Jú Huā) and lotus petals), panels (lines, stylised lotus petals (变形莲瓣纹 Biàn Xíng Lián Bàn Wén) and hexagons), key-fret (回 纹 Huí Wén), classic scrolls (卷草汶 Juàn Cǎo Wèn) and calligraphic (壽, shòu or ‘longevity’) motifs on both the interior and exterior surfaces of the sherds. 44 Rim edges are ‘round,’ ‘bulbed’ and ‘flat-angular’ with external diameters of 7-26cm. From Table IV, there are only 14 blueand-white vessels (7.0% of 199 porcelain vessels) estimated from the rim sherds which are distributed between the NE (6), SE (2) and SW (6) quadrants. Nine sherds in the NE quadrant, estimated to be a single vessel, have misfired glaze but was nevertheless included in the count because of its visible and distinct underglaze cobalt blue motifs. Discernible vessel forms consist of either cups or bowls. Only two vessel-bases measuring 0.56-0.57cm in thickness, ‘flat-angular’ in shape, with foot-rings of 0.62-0.8cm in height and 2.8-3cm in external diameter were estimated from base sherds. The foot-rings and exterior medallion surface of both vesselbases are mostly unglazed and set against a glazed interior. Two body sherds have an angled, glazed exterior but curved, unglazed interior body shape and surface. The sherds also appear to be luted from sections which were moulded separately. A lid fragment, the only one among all sampled porcelain sherds, measuring 8cm in diameter with an unglazed interior surface was found among the blue-and-white sherds as well. 43 Ibid., p. 85-7. See S. T. Yeo & Jean Martin, “Index of Decorations,” in Chinese Blue & White Ceramics, ed. S. T. Yeo & Jean Martin (Singapore: Arts Orientalis, 1978), p. 299-314; Wang Qingzheng, A Dictionary of Chinese Ceramics (Singapore: Sun Tree Publishing Limited, 2002), p. 241-91; Foo Soo Ling. A Study of 14th Century Blue and White Ceramics Excavated in Singapore. Unpublished ISM paper. Southeast Asian Studies Programme, Faculty of Arts & Social Sciences, NUS, 2005. 44 165 The fourth variety of sampled porcelain sherds are made up of 173 misfired or degraded sherds (1770.5g; 0.10 Pc/g), the second largest in porcelain proportion (25.6% in count, 27.1% in weight). It is impossible to tell what the intended glaze colours of these sherds were, but they were probably meant to be any of the previous three porcelain varieties as demonstrated by the presence of blue-and-white sherds of this variety. No such porcelain was found on FTC, although Miksic did report ‘severe degradation’ experienced by some Dehua/Ding sherds recovered on that site. 45 Neither were they mentioned in any other survey of archaeological finds from pre-colonial Singapore, which could be due to an oversight resulting from the focus on ceramics of a more presentable quality. A greater distribution of this porcelain variety outside the STA site is still possible, as thousands of ceramics across all pre-colonial Singapore sites still remain in storage and await further research. These sherds comprise of rims, body and base fragments measuring 0.17-1.50cm. Besides underglaze cobalt blue motifs, other decorations include reliefs (rib-lines, foliated and swirls), groove carvings and incised lines as well. Once again, reliefs are labelled ‘general’ when the pattern itself cannot be clearly discerned. Rim edges are ‘round,’ ‘flat-angular’ or ‘foliated’ with external diameters of 3-22cm. As seen on Table IV, a total of 53 misfired or degraded vessels were estimated from rim sherds (26.6% of 199 porcelain vessels). A significant number of the estimated vessels are concentrated within the SW quadrant (24), whereas the remainder are distributed between the NE (10), NW (12) and SE (7) quadrants without much variation. Discernible vessel forms include bowls, covered boxes and jarlets. A thin strip along the rim edges of the boxes and lids is left unglazed as usual. A total of 13 vessel-bases with thickness of 0.34-1.68cm and external diameters of 4-12cm were estimated from the base sherds. All vessel-bases had foot-rings measuring 0.23-1.18cm in height and ‘round,’ ‘flat-angular’ or ‘recessed’ in shape. Both the foot-ring and exterior medallion surface are either glazed or unglazed, although two such exterior medallion surfaces were painted light brown and maroon. Two vessel bases also had stacking rings on the interior medallion surface as well. 45 Miksic, Archaeological Research, p. 84. 166 IV. Metals Metals are fourth in proportion in terms of both count (381; 4.4%) and weight (2299; 4.3%) of the total sampled STA artefacts with an average size of 0.17Pc/g (or 17 pieces per 100g). Three general metal varieties – slag, objects and coinage – can be discerned, but the precise identification of the different kinds of metals will require further chemical and compositional analyses which are beyond the scope and resources of the present thesis. Slag is a by-product of metal-working produced when impurities, separated from the smelting of metal ore, combine with refractories (such as sand and husks) added to lower the melting point of the ore. In the case of iron ore, both the production and hammering of wrought iron by the smith produce slag as well. 46 The slag identified in this thesis corresponds to ‘Type 1 – 7’ slag categories in Shah Alam’s study of metal finds at the PHC site. A total of 126 pieces of slag (995g; 0.13Pc/g) were identified, with majority of the pieces concentrated at the SE quadrant (105) and the rest scattered among the NE (6), NW (9) and SW (6) quadrants as illustrated in Table V. It is second in proportion of all metals in both count (33.1%) and weight (43.3%). Slag pieces are extremely hard and have a rough surface texture partly due to rusting, corrosion as well as the presence of sand particles adhered onto the surface. The colours range between the hues of red, gray and black. 47 The ‘reddish-brown discolouration’ on parts of the slag body may represent the high iron content of pre-industrial slag. 48 Slag is also distinguished by their distinctively random and irregular shapes as opposed to the relatively even and regular shapes of crafted metal objects and are often riddled with cavities. Objects here is a general variety referring to any crafted metal pieces which are neither slag fragments nor coinage. A total of 249 fragments (1295g; 0.19Pc/g) were present, with most of the objects distributed between the NE (72) and NW (80) quadrants, and the remainder in the SE (73) and SW (24) quadrants as seen in Table V. Most of these artefacts share a similar 46 Shah Alam, Metal Finds and Metal-working, p. 13-4. Dusky red (5R 3/3 & 7.5R 3/3),reddish black (5R 2.5/1 & 7.5YR 2.5/1), dark gray (GLEY 4/N) & black (5YR 2.5/1 & 7.5YR 2.5/1). 48 Hans-Gert Bachmann, Identification of Slags from Archaeological Sites. Occasional Publication No. 6 (London: Institute of Archaeology, 1982), p. 3, cited in Shah Alam, Metal Finds and Metal-working, p. 14. 47 167 surface texture and range of colours with slag, but are more distinct and varied in generally geometrical shapes. These fragments are divided further into flat metal pieces (137), rods (79), nails (14), hooks (6), points (4), coiled wires (4), small rings (2), a bead, a bangle fragment and a ‘button-like’ object. There are no visible decorative features present on these metal objects, although one flat metal piece has a carved-paddle impressed earthenware sherd attached on its surface. Corrosion has led to the adhering of sand particles on some surfaces as well. Four of the six hooks can be positively identified as fishing hooks because of the distinctive shape of their respective hooks and the presence of eyelets, though which a metal wire or line would run for hook-and-line fishing. 49 Two flat metal pieces, one coiled wire and the bangle fragment appears to be made of bronze because of the greenish corrosion colour on their surface. This indicates the presence of copper alloy which is the major component of bronze artefacts. 50 It is therefore likely that the majority of the metal objects, most of which carry the same ‘reddishbrown discolouration’ found on the surfaces of slag with high iron content, are likely to be made of iron. Some flat metal pieces may also be ‘hammer scales’ or ‘hammer slag’ which were ejected during forging as well. 51 Metal objects found at the PHC site similar to those in the STA sample pool include flat iron pieces, iron rods, iron fishhooks and bronze and copper hooks. 52 Coinage forms the last and smallest category of STA metals (1.6% in count; 0.4% in weight). A total of six coins weighing 9g were identified and found distributed between the NW (2) and SW (4) quadrants. All six coins are likely to be made of bronze and/or copper as inferred again from the greenish corrosion colour on their surfaces. V. Others This section contains the few remaining artefacts within the STA artefact sample pool. A total of 12 stones (1425g; 0.008Pc/g), 111g of charcoal, 21 wood pieces (6g; 3.5Pc/g), six 49 Shah Alam, Metal Finds and Metal-working, p. 26-30. Ibid., p. 23. Ibid., p. 22. 52 Ibid., p. 15, 22, 26-8. 50 51 168 pieces of resin (7g; 0.86Pc/g) and eight bone fragments (6g; 1.3Pc/g) were recovered. As illustrated in Table I, charcoal deposits were found mainly in the NE (20.5g), NW (35g) and SE (48.5g) quadrants, with the remainder at the SW (7g) quadrant. All 21 wood pieces were located in the SE quadrant, whereas the six resin pieces were found only at the SW quadrant. The 8 bone fragments were distributed between the NE (1), NW (1) and SW (6) quadrants. The amount of artefacts from these categories is unfortunately too little to be of any significance. No archaeological features comprising these artefacts were indicated in the excavation records of the STA units selected for this research as well. The lack of archaeological expertise or supervision among the volunteer excavators and tight excavation schedule may have led to the low amount of these artefacts collected. Preference for the collection of more visible and recognizable artefacts such as ceramic sherds and metals could have been another factor as well. Archaeobotanical and faunal analysis of charcoal, wood, resin and bone fragments may reveal more information but is again beyond the scope and resources of the present thesis. The physical characteristics of stone artefacts however warrant further mention. Of the 12 stone pieces, seven are round and irregular in shape while three have flat surfaces, two of which had rectangular profiles 1.22-2.69cm thick. All 10 pieces were fractured on the edges but had smoothed surfaces. They are distributed between the NW (4), SE (4) and SW (2) quadrants. Two more smoothened stone fragments are found in the NW quadrant; one had a carved motif on a surface and bordered with a circular, recessed edge on one side, whereas the other appears to be a fractured fragment of the previous piece. The last stone artefact, located in the SE quadrant, has a curved and partially-fractured body with a crucible-like rectangular depression on one side. DEDUCTIONS I. Earthenware: Locally-made Vessels, Regionally-imported Goods The study of Southeast Asian earthenware ceramics has largely been overshadowed by that of its more aesthetically attractive cousins: high-fired stoneware and porcelain ceramics. 169 As late as 1987, art historian Dawn F. Rooney noted that although earthenware sherds made up the majority of artefacts recovered from archaeological sites in the region, they still remain ‘neglected in reported finds, analyses, and published references.’ 53 Almost two decades later, the lack of information dissemination in Southeast Asian earthenware studies was still endemic in the region. 54 In an effort to remedy the situation and create a regional archaeological framework of earthenware analysis, the seminal study Earthenware in Southeast Asia encompassing the latest earthenware research across the major Southeast Asian nations was published. To date, it remains the only publication which focuses solely on earthenware ceramics in Southeast Asia. At the same time, ethnographic records of traditional or rural earthenware potters remain few in number because research is narrowly concentrated on selected Southeast Asian ceramics which are ‘recognized as having great cultural and monetary value, such as the famous prehistoric Ban Chiang pottery and the high-fired celadon wares of Sukhothai and Sawankhalok.’ 55 In the case of earthenware ceramics from island Southeast Asia, much emphasis has been placed on kendis by various publications while there has yet to be any study dedicated to other earthenware forms of a more utilitarian nature. 56 Consequently, any identification and analysis of earthenware ceramics can only rely on comparisons with the few existing published catalogues and be informed by the existing archaeological record and historical context which surrounds these artefacts. Earthenware ceramics have been around in Southeast Asia for a very long time. The earliest well-dated examples in both the region and the mainland comes from the Spirit Cave in north-western Thailand in the form of decorated (cord-marking, net-marking, burnishing, appliquéing and incising) Hoabinhian pottery sherds which are approximately 8400 years old. The earliest known earthenwares in island Southeast Asia on the other hand consist of plain and burnished pottery recovered in East Timor dating to a horizon of 4500 to 3700 years ago. 53 Dawn F. Rooney, Folk Pottery in South-East Asia (Singapore: Oxford University Press, 1987), p. 1. Miksic, “Preface,” in Earthenware in Southeast Asia, p. xvii. 55 Mick Shippen, The Traditional Ceramics of Southeast Asia (Honolulu: University of Hawai’i Press, 2005), p. 11. 56 Khoo. Kendi. Pouring Vessels in the University of Malaya Collection. Singapore: Oxford University Press, 1991; Eng-Lee Seok Chee. Kendis. Singapore: National Museum Singapore, 1984. 54 170 Paddle-impressions with basket- or cord-marked patterns, carved-paddle impressions, cord marking, incising, burnishing, slipping are among the oldest pottery decorative techniques found in Southeast Asia. 57 This early pottery tradition was also hypothesized by Wilhelm G. Solheim to be distributed across Southeast Asia by the Neolithic Nusantao Maritime Trading Network (NMTN). 58 The Lapita, as well as Solheim’s Sa-Huynh-Kalanay and the Bau-Malay Pottery Traditions are thought to be derived from this Neolithic pottery tradition, with the SaHuynh-Kalanay Pottery Tradition believed to be gradually replaced by that of the Bau-Malay during the first millennium CE in island Southeast Asia. 59 Solheim’s ‘Bau-Malay’ pottery rubric claims all carved-paddle-impressed earthenware produced around and after 700 CE in the southern Philippines, western Borneo, western Indonesian islands and the Malay Peninsula. 60 On this basis of a common pottery tradition, Solheim hypothesized further that the later ‘Malay-Nusantao coaster maritime traders’ were involved in the development of ‘Srivijaya, Madjapahit, Malayu and the other “Malay” states of western Indonesia and Malaysia.’ 61 Peter Bellwood and Matussin Omar’s ‘Tanjong Kubor (TK) ware’ is the only other proposed precolonial (700 to 1500 CE) major pottery style in the Malay archipelago which affiliate earthenware ceramics from sites in Brunei, south Philippines, Sarawak and Johor Lama, where ‘the distribution pattern is connected with the development of the Malay language as a trading language in those areas where it is most commonly spoken today.’ 62 Therefore, coarse- and medium-tempered earthenware sherds in the STA sample were probably made by pre-colonial Singaporean potters of a Malay-Nusantao or Malay-speaking people. There is no reason to believe otherwise, given the long history and distribution of ceramic technology in the region highlighted earlier. Abundant sources of clay along the beach and north bank of the Singapore River next to the pre-colonial settlement would have been 57 Solheim, “Southeast Asian Earthenware Pottery and Its Spread,” in Earthenware in Southeast Asia, p. 1-3. Solheim, Archaeology and Culture in Southeast Asia: Unraveling the Nusantao (Quezon City: The University of the Philippines Press, Diliman, 2006), p. 57-139. 59 Solheim, “Southeast Asian Earthenware Pottery and Its Spread,” p. 17; Solheim, “Earthenware Pottery, the T’ai and the Malay,” Asian Perspectives (AP) 29,1 (1990), p. 26-30. 60 Solheim, “Earthenware Pottery, the T’ai and the Malay,” AP 29,1 (1990), p. 28-9. 61 Solheim, “Notes on “Malay Pottery” in East Malaysia and neighboring areas,” SMJ 29,50 (December 1981), p. 13. 62 Bellwood & Matussin Omar, “Trade Patterns and Political Developments in Brunei and Adjacent Areas, A.D. 700-1500,” Brunei Museum Journal (BMJ )4,4 (1980), p. 168, 171-4. 58 171 easily obtainable to produce coarse- and medium-tempered earthenware vessels. It was recorded in 1849 that ‘fine kaolin’ clay was ‘abound’ in colonial Singapore and found ‘close to the beach.’ This clay was apparently ‘the best adapted of any in India for the manufacture of porcelain.’ Ceramics made from the ‘coarsest parts of this clay’ were ‘strong, hard, of a beautiful and rich crimson colour;’ ‘picked clay,’ where impurities had been removed or sieved away, gave a ‘snow-white biscuit;’ whereas in its raw form, ‘unpicked and worked up as it is dug,’ produces a ‘light yellow stoneware of the very best kind, as far as density, hardness, strength, lightness, and colour are concerned.’ It was noted as well that ‘blueish alluvial clays,’ presumably from the river banks, were used to make bricks and tiles ‘of which the town of Singapore is built.’ 63 However, compositional analyses of similar sherds found on FTC had shown that clay supplies in pre-colonial Singapore were ‘not homogenous’ and unsuitable for making ‘high-quality’ ceramics. 64 There are several possible reasons which can explain this conflict of information: despite the optimistic colonial report, there might only be a limited supply of ‘fine kaolin clay’ in reality. Local pre-colonial potters may also have found it too laborious or did not understand the need to sieve and extract kaolin clay from freshly-dug clay. Moreover, it will be established later that pre-colonial Singapore did not possess the necessary pottery manufacturing technology – the potter’s wheel and kiln – to produce ‘high-quality’ ceramics. Seen in this context, coarse- and medium-tempered earthenware are the more likely candidates for locally-made ceramics. This pre-colonial Singaporean craft is likely to be that of a wider pottery tradition which would also, given the close proximity of pre-colonial Singapore sites, include earthenware pottery recovered from FTC, believed by Miksic to be made by local inhabitants perhaps as a part-time occupation, and those from PHC which Chen suggested were likely made by ‘Malayspeaking’ women potters on the basis of comparisons with the regional ethnographic record. 65 Alluding to Solheim’s broader Bau-Malay and Bellwood and Omar’s TK-ware Pottery 63 Anonymous, “Miscellaneous. Agriculture in Singapore,” JIAEA 3 (1849), p. 511. Miksic & Yap Choon Teck, “Compositional Analysis of Pottery from Kota Cina, North Sumatra: Implications for Regional Trade during the Twelfth to Fourteenth Centuries A.D.,” AP 31,1 (Spring 1992), p. 58. 65 Miksic, Archaeological Research, p. 65; Chen, Earth to Earth, p. 117-9. 64 172 Tradition, Chen argues that the PHC earthenware sherds also represent a distinct ‘Malay Pottery Tradition’ distributed along the coastal sites of Sumatra, the Malay Peninsula, Kalimantan and the Philippines, a culturally-distinct area from the neighbouring Tai and Javanese cultures by the 13th century. 66 The absence of rilling on these local coarse- and medium-tempered earthenware ceramics suggests that these ceramics were hand-moulded instead of wheel-thrown. 67 From the study of ‘Iron Age’ (circa 9th century) pottery found on the Malay Peninsula, Prince John Loewenstein observed that the potter’s wheel ‘was unknown to the Iron Age people in Malaya.’ 68 The ethnographic record also highlights the use of either a paddle and anvil or a spatula to smoothen and shape the vessel into its final pre-firing form; a ‘slow’ wheel, intermittently spun by hand to shape the vessel body, may be used in the shaping process as well, but not in the same manner as the faster potter’s wheel. Of the six general earthenware pottery production techniques (Type ‘A’-‘F’) surveyed from contemporary villages in mainland Southeast Asia, only two (Type ‘D’ and ‘F’) are true wheel-throwing techniques involving the use of a rapidly turning wheel which generates a centrifugal force to ‘throw a hollow cylinder from a lump of clay and shape a mouth rim on the cylinder.’ 69 Only this process can produce the characteristic rilling observed on wheel-thrown ceramics. In both cases, the use of the ‘fast’ wheel is thought to be a technique originating outside the region in ‘Eastern and Southern India’ as well as China. 70 Moreover, the use of the fast wheel to throw and mould clay is typically associated with large-scale workshop-level production as it permits the rapid production of large number of vessels. 71 The high variability and irregularity in sherd and base thickness, external rim diameters and types of decorations of the sample STA earthenware sherds, however, does not reflect the uniformity and consistency one would expect of pottery manufactured by a large-scale workshop-level production. Hence, it is safe to conclude that 66 Chen, Earth to Earth, p. 125-131. Rice, Pottery Analysis, p. 129. 68 Prince John Loewenstein, “The Origin of the Malayan Metal Age,” JMBRAS 29,2 (1962), p. 55. 69 Leedom Lefferts & Louise Allison Cort, “A Preliminary Cultural Geography of Contemporary Village-bassed Earthenware Production in Mainland Southeast Asia,” in Earthenware in Southeast Asia, p. 301-5, 70 Ibid.,308. 71 Sinopoli, Approaches to Archaeological Ceramics, p. 21. 67 173 STA earthenware vessels were generally hand-moulded and possibly made at the householdlevel. While the ethnographic record states that earthenware vessels can either be fired in a kiln or in the open, open firing is still the more likely firing technique used to produce local coarse- and medium tempered earthenware vessels in pre-colonial Singapore society. Kiln firing requires the construction of either a subterranean or above-ground chamber within which pottery is fired at a closely monitored and controlled temperature, whereas open firing is a relatively simple process of stacking pottery over the fuel and fired in the open. Both methods would involve the use of natural fuel such as dried grass, paddy husks, straw, wood, bamboo, palm fronds, straw or even dung for firing. 72 However, variables such as wind daft, firing temperature and duration, atmosphere type and fuel combustion rate for the individual vessel in open firing are considerably more uncontrollable and varied than those in kiln firing. Exposure to the elements such as humidity and rain in open firing affects these conditions as well. 73 The nature of these conditions surrounding the manufacture of locally-made STA earthenware can be estimated from the sherd surface and cross-section colours. 74 As seen in the results, coarse- and medium-tempered earthenware sherds exhibit a wide range and irregularity of surface colours on the individual surfaces as well as between fractured sherds of the same vessel. At the same time, the majority of these sherds exhibited contrasting sherd surface and cross-section colours, evidence of incomplete oxidizing and/or reducing atmosphere during firing. Both situations are more common-place in open than kiln firing. Moreover, current research on pre-colonial kiln technology suggests that kilns were largely used to produce glazed or unglazed, high-fired earthenware ceramics made of ‘fine clay’ and of ‘high aesthetic value’ which were likely used for special or ceremonial purposes. 75 These pre-colonial kilns were all located within the intermediate tropical zone of mainland Southeast Asia and Java, but none 72 Shippen, The Traditional Ceramics of Southeast Asia, p. 18-220; Charlotte Reith, “A Comparison of Ground Firing Techniques in Contemporary Myanmar Villages,” in Earthenware in Southeast Asia, p. 314-20; Sinopoli, Approaches to Archaeological Ceramics, p. 31. 73 Rice, Pottery Analysis, p. 153-63. 74 Ibid., 343-5. 75 Miksic, “Kilns of Southeast Asia,” in Southeast Asian Ceramics, p. 48-69. 174 have been found or reported thus far from the equatorial belt where Sumatra, the Malay Peninsula, Singapore Island, Borneo and their adjacent islands are located. 76 It was observed that ‘generally pottery baking is done in the open air, without using a permanent kiln (tungku/tobong)’ in various parts of Java and Indonesia for ‘many generations. 77 High-fired ceramics were never produced in this area according to the historical and archaeological record as well. 78 To quote Rooney, ‘pottery technology in Indonesia, Malaysia, and the Philippines, did not advance until modern times beyond the production of unglazed, utilitarian wares.’ 79 Interestingly, three coarse-tempered and one medium-tempered earthenware sherds, adorned with the Saw-Tooth’ and ‘Saw-Tooth with Flame’ carved-paddle impressions, from the sampled STA earthenware sherds have similar decorative motifs to that of a Thai ‘paddlemade, unglazed, earthenware’ pot recovered from the 14th century Rang Kwien shipwreck in the Gulf of Thailand. It was estimated that half the cargo contained Thai earthenware ceramics, with the remainder comprising a mixture of Thai (Suphanburi, Sawankhalok and San Kamphaeng), Vietnamese and Chinese high-fired ceramics. 80 Some earthenware pottery fragments recovered from Tioman Island – off the east coast of Malaysia – also exhibit the same ‘Saw-Tooth’ carved-paddle impressions as well, but besides carrying a vague ‘Southeast Asia’ label, the exact dating and provenance of these sherds remains unknown. 81 The paucity of this particular type of carved-paddle impressions also appears to support a likely non-local origin of these STA sherds. Compositional analysis and comparisons in the future will determine if the sampled STA sherds share the same Thai provenance as the Rang Kwien 76 Pre-colonial Vietnam, Cambodia, Thailand, Laos, Myanmar on the mainland and Java in the islands; see also Miksic, “Kilns of Southeast Asia,” in Southeast Asian Ceramics, p. 48-69; Brown. The Ceramics of South-East Asia. Their Dating and Identification. Singapore: Oxford University Press, 1988; Rooney, Folk Pottery in South-East Asia, p. 25-31. 77 Sumijati Atmosudiro, “Notes on the Tradition of Pottery Making in the Region of Kasongan, Regency of Bantul, Yogyakarta,” in Studies on Ceramics, ed. Departemen P dan K (Jakarta: Proyek Penelitian Purbakala Jakarta, 1984), p. 123. 78 Naniek Harkantiningsih, “Yue and Longquan Green Glazed Wares from Archaeological Sites in Java and East Indonesia,” in New Light on Chinese Yue and Longquan Wares, p. 273. 79 Rooney, Folk Pottery in South-East Asia, p. 3. 80 Brown, The Ming Gap and Shipwreck Ceramics of Southeast Asia, p. 38, 87. 81 K. K. Kwan & Peter Y. K. Lam, “Earthenware Pottery. Southeast Asia,” in A Ceramic Legacy of Asia’s Maritime Trade. Song Dynasty Guangdong Wares and other 11th to 19th century Trade Ceramics found on Tioman Island, Malaysia, compiled by K. K. Kwan et. al. (Petaling Jaya: Southeast Asian Ceramic Society, West Malaysia Chapter, 1985), p. 126. 175 earthenware pot. If this conjecture is accurate, it may yet be another indicator of intra-regional trade between pre-colonial mainland Southeast Asia and Singapore. On the other hand, the sampled STA fine-paste earthenware ceramics are not locallymade but rather, a class of imported goods from the region. Miksic and other archaeologists had noted that similar sherds recovered from FTC possessed the same ‘chalky texture’ and ‘general mode of potting’ as those discovered in other pre-colonial sites at Oc-Eo (1st to 6th centuries), Angkor Borei and Long-dien (9th to 15th centuries) at the Mekong River and Delta, Kota Cina (12th to 14th centuries), Muara Jambi (11th to 14th centuries) and Palembang (8th to 9th centuries) across Sumatra, Pengkalan Bujang (12th to 14th centuries) at the northern end of the Malay Peninsula, Ban Pa O (Kok Moh) and Satingphra (12th to 14th centuries) on the Satingphra Peninsula, Tanjong Kubor (12th to 14th centuries) in Borneo, Banten Girang (12th to 15th centuries), Gatak (8th to 9th centuries), Gresik, Tuban and Trowulan (14th to 15th centuries) across Java and Butuan (9th to 10th centuries) in Mindanao. 82 Miriam T. Stark’s ‘fine orangeware’ (3rd to 4th centuries BCE to 1st century CE), a ‘distinctive collection of thin-walled, orange to buffcoloured containers’ with a ‘mottled surface,’ as well as ‘fine buffware’ (5th to 8th centuries) recovered from the Angkor Borei site in Cambodia are likely to be fine-paste earthenware ceramics as well. She also associates these sherds with those recovered by B. P. Groslier at Sambor Prei Kuk (7th to 9th centuries) by the eastern bank of the Tonlé Sap. 83 Further evidence supporting fine-paste earthenware, primarily kendis, as a intra-regional trade ceramic is given by their discovery on various shipwrecks in Southeast Asian waters, namely the Intan (10th century, Indonesia), the Cirebon (10th to 11th centuries, Indonesia) and the Java Sea (13th 82 Miksic, Archaeological Research, p. 56-7; Miksic & Yap, “Fine-Bodied White Earthenwares of Southeast Asia: Some X-Ray Fluorescence Tests,” in AP 28,1 (1990), p. 46-50; Manguin, “From Funan to Srivijaya: Cultural Continuities and Discontinuities in the Early Historical Maritime States of Southeast Asia,” in 25 Tahun Kerjasama Pusat Penelitian Arkeologi dan École Française d'Extrême-Orient (EFEO) (Jakarta: EFEO, 2002), p. 59-82, cited in Eska Asih Putrina Taim, “The Batujaya Pottery: Early HinduBuddhist Pottery in West Java,” in Archaeology: Indonesian Perspective. R. P. Soejono’s Festschrift, ed. Truman Simanjuntak et. al. (Jakarta: Indonesian Institute of Sciences, International Center for Prehistoric and Austronesian Studies, 2006), p. 337; Edmund P. Edwards McKinnon, “Historic Period Earthenware from the Island of Sumatra,” in Earthenware in Southeast Asia, p. 164, 1689; Amara Srisuchat, “Earthenware from Archaeological Sites in Southern Thailand: The First Century BC to the Twelfth Century AD,” in Earthenware in Southeast Asia, p. 255, 258-9; Janice Stargardt, “Kendi Production at Kok Moh, Songkhla Province, and Srivijayan Trade in the 11th Century,” SPAFA, p. 181-8. 83 Miriam T. Stark, “The Chronology, Technology and Contexts of Earthenware Ceramics in Cambodia,” in Earthenware in Southeast Asia, p. 217-20. 176 century, Indonesia) wrecks. 84 ‘X-Ray fluorescence tests had revealed that the composition of mineral content within these FTC sherds ‘resemble closely’ indeed to that of fine-paste sherds recovered at Satingphra, but not those from Sumatra or Java. 85 The south Thailand region is one of only two known Southeast Asian locales so far (the other being Java) where deposits of fine clay, a ‘relatively pure kaolin derived from primary weathering of feldspar,’ used to make fine-paste earthenware ceramics have been found. 86 This corroborates with the earlier observation made regarding the ability of fine-paste ceramics to cope with thermal stresses without the assistance of temper. Given the close proximity of the STA and FTC sites, it is therefore safe to assign a provisional south Thailand provenance to the STA fine-paste sherds. Differences in pottery manufacturing technology from the coarse- and mediumtempered earthenware ceramics also support the foreign provenance of fine-paste earthenware as well. The presence of rilling on the surfaces of the sampled STA fine-paste sherds suggest that they were wheel-made, whereas the limited variation and uniformity in colour suggest that these sherds were more likely to be kiln than open fired. This is affirmed by Janice Stargardt’s and Amara Srisuchat’s identification of a sophisticated fine-paste kendi complex at the Kok Moh kiln site in Ban Pak O (present-day Songkla province) between the 11th and 13th centuries. 87 Greater consistency in sherd and base thickness, external rim diameters and types of decorations than that of locally-made earthenware also suggests the possibility of manufacture on a large-scale workshop-level production. This ‘distinctive mode of potting’ was also noted by Miksic to characterize fine-paste sherds from FTC for ‘its exceptional level of skill, the symmetrical shapes produced, and the uniform surface colours free from firing clouds.’ He also believed that some were made with a medium-fast wheel. Differences are also reflected in the types of decorations which adorn the earthenware sherds as well. While carved-paddle impressions are the most common form of decoration on coarse- and medium-tempered 84 Miksic, “Research on Ceramic Trade,” p. 87-8. Miksic & Yap, “Fine-Bodied White Earthenware,” p. 54-56. 86 Ibid., p. 49; Miksic & Yap, “Compositional Analysis of Pottery from Kota Cina, North Sumatra,” p. 74. 87 Stargardt, “Kendi Production at Kok Moh, Songkhla Province, and Srivijayan Trade in the 11th Century,” in SPAFA, p. 181-9; Srisuchat, “Earthenware from Archaeological Sites in Southern Thailand,” p. 249-60. 85 177 earthenware sherds, none were found on the surfaces of fine-paste sherds which are largely undecorated. Where they are decorated, fine-paste sherds exhibit a unique punctated and carved ‘Dot-and-Ring’ motif not found on any other earthenware variety as well. Fine-paste sherds are also the only earthenware sherds to have glaze applied on their surface. In light of the provisional south Thailand provenance of the fine-paste sherds, this difference in decorative techniques and motifs also supports the theory of a pre-colonial cultural boundary or ‘fence,’ first alluded by Solheim’s distinction of the Bau-Malay Pottery Tradition and then explicitly suggested by Chen in his definition of the Malay-type Pottery Tradition, between the cultures of mainland and island Southeast Asia. 88 It should be noted also that sampled STA fine-paste sherds with surface colours within the hue of red may yet correspond to the ‘uniform dark red (2.5 YR, 5/4-6/6), probably wheel-made, almost highly burnished’ fine-paste sherds recovered in central and east Java as well, the only other known source of pre-colonial fine-paste ware. 89 Significantly, the nine sherds with a dark grayish-green (5GY 3/2) or ‘apple-green’ glaze are similar to lead-glaze fragments of the same colour found in Trowulan, the capital of the Majapahit kingdom at Java, hence the term ‘Majapahit Ware.’ 90 The difference between the local-made and regionally-imported earthenware ceramics appears to extend to their perceivable forms and functions as well. From the observations and measurements made in the results as well as comparisons made with the regional ethnographic and archaeological record, coarse- and medium-tempered pottery could be spherical, ellipsoid or ovaloid vessels with flared rims, narrow to wide orifices and globular bodies. 91 With the greatest average sherd size, they would have been also the largest earthenware vessels amongst the four identified STA earthenware varieties. While they do not appear to have ‘necks,’ they may have a ‘collared’ rim-orifice. Some of these vessels have carinated bodies as well. 92 Their 88 Chen, Earth to Earth, p. 128. Miksic & Yap, “Fine-Bodied White Earthenware,” p. 53. 90 Miksic, “Kilns of Southeast Asia,” p. 68; Miksic, Archaeological Research, p. 56. 91 Miksic & Natalie SY Ong, “Catalogue,” in Southeast Asian Ceramics, p. 109, 161; Shippen. The Traditional Ceramics of Southeast Asia. Honolulu: University of Hawai’i Press, 2005; Bellwood & Omar, “Trade Patterns and Political Developments in Brunei and Adjacent Areas, A.D. 700-1500,” p. 163-6; Omar, “Ancient Singapore Earthenware Pottery,” in Early Singapore, p. 57. 92 “Anatomy of a Vessel” in Rice, Pottery Analysis, p. 212-5; “Vessel Shape” in Shepard, Ceramics for the Archaeologist, p. 22455. 89 178 vessel forms therefore fall within the category of shapes used to fulfil basic household tasks such as cooking or storage. Round-based vessels are advantageous for cooking as they ‘transmit heat easily and are less susceptible to breakage from thermal stress than flat-based vessels.’ 93 Resistance to thermal stresses, normally from the repeated exposure to fire as cooking vessels often are, can also enhanced by the addition of non-plastic temper into the unfired clay or increasing the porosity of the clay fabric. It should be noted that black smudges or staining is found to occur only on earthenware ceramics, the majority of which are exhibited by coarseand medium-tempered sherds. This is likely the result of fire-clouding or sooting, a clearly indication of use in cooking or other activities involving fire, and supports the specialized use of such pottery as cooking vessels. 94 The porosity of coarse- and medium-tempered sherds, demonstrated by the large pores within the clay fabric visible to the naked eye, is a characteristic trait of earthenware ceramics. This allows the seepage and evaporation of stored water from the exterior such that the water is kept cool. 95 Evidence of their possible use as short-term water storage vessels or kendis is at least supported by the presence of one medium-tempered spout fragment, but this was evidently not a common vessel form for this earthenware variety. It is therefore likely that the medium-tempered kendi was a rare, local skeuomorph of the more common, regionally-imported fine-paste kendis. A single ‘coarse’ beige, burnished earthenware kendi (UM 81.15) dated to the 11th and 12th centuries from east Java, at the Muzium Seni Asia (Asian Art Museum) of the University of Malaya in Kuala Lumpur, as well as ‘thirty spouted vessels of the unslipped variety, including both small pots with wide-mouths and kendis’ recovered at Kota Cina are the only non-fine-paste kendis known to the author. 96 It is also significant that all seven lid fragments in the earthenware category are medium-tempered sherds, thereby supporting their use as general purpose storage vessels as well. Besides pottery, some coarse-tempered sherds may also be fragments of building or structural materials such as bricks, tiles or ornamentation as pointed out earlier in the results. According to Rooney, 93 Sinopoli, Approaches to Archaeological Ceramics, p. 84. Ibid., p. 235-6. 95 Rice, Pottery Analysis, p. 230. 96 Khoo, Kendi, p. 33; Miksic, Archaeology, Trade and Society in Northeast Sumatra, Ph.D Thesis, Cornell University, 1979, p. 196-7. 94 179 earthenware pottery was ‘a primary material used in ancient South-East Asia for fixtures on temples and other public buildings.’ 97 It should be noted that while exploring the remains of the ‘ancient town of Singapore’ on FTC in 1822, Crawfurd encountered the ‘remains of the foundations of buildings, some composed of baked brick of good quality.’ 98 Coarse- and medium-tempered earthenware recovered from FTC and PHC sites are also believed by Miksic and Chen respectively to be predominantly utilitarian and with similar functions as well. 99 Fine-paste ceramics however only appear to function strictly as pottery. Indeed, Miksic notes that the majority of the fine-paste ceramics found across sites in Sumatra, Java and south Thailand were kendis. 100 As an earthenware ceramic closely associated with Hindu-Buddhist divinities, the kendi – a term related to kundika, Sanskrit for ‘water vessel’ as well – has been in circulation across Southeast Asia long before high-fired stoneware and porcelain versions were made in mainland Southeast Asia and China from the 10th century onwards. 101 While most recorded fine-paste vessel forms belong to the spouted spherical kendi, there is evidence that there are other vessel forms – and hence, functions – that the fine-paste earthenware can have. 102 Stark records the existence of not just ‘fine buffware kendis,’ but also ‘pedestaled bowls, bottles, and small, narrow-necked flared-rimmed jars’ as well as ‘fine buffware ring-based jar, fine orangeware cup, and round-bottomed fine orangeware pot.’ 103 Eka Asih Putrina Taim also reports the finding of fine-paste ‘kitchen utensils’ such as ‘bowls, cups, and dishes’ at the Palembang site. 104 Fine-paste sherds at Oc-Eo are apparently fragments of ‘bowls, pedestal cups and jugs’ which served as ‘good quality household wares’ as early as the 2nd century. 105 Miksic notes the presence of forms which included ‘flat-bottomed simple spherical bowls and 97 Rooney, Folk Pottery in South-East Asia, p. 23. Crawfurd, Journal of an Embassy. Volume I, p. 71. 99 Miksic, Archaeological Research, p. 57-8, 61, 64, 67; Chen, Earth to Earth, p. 114-5. 100 Miksic, Archaeological Research, p. 57. 101 I. H. N. Evans, “On the Persistence of an Old Type of Water-Vessel,” JMBRAS 1 (April 1923), p. 248-50; Eng-Lee, Kendis, 119. 102 Khoo, Kendi, p. 30-43; Eng-Lee, Kendis, 20-1, 58-61, 72, 79-81; Edwin R. Bautista et. al., “Thai Kendi (Satingphra-type). 10th to 12th century,” in Guangdong Ceramics from Butuan and other Philippine sites, ed. Brown (Manila: The Oriental Ceramic Society of the Philippines, Inc., 1989), p. 126-7. 103 Stark, “The Chronology, Technology and Contexts of Earthenware Ceramics in Cambodia,” p. 219-20. 104 Taim, “The Batujaya Pottery: Early Hindu-Buddhist Pottery in West Java,” p. 337. 105 Manguin, “From Funan to Srivijaya: Cultural Continuities and Discontinuities in the Early Historical Maritime States of Southeast Asia,” p. 59-82, cited in Taim, “The Batujaya Pottery: Early Hindu-Buddhist Pottery in West Java,” p. 337. 98 180 jars’ and ‘tall vases with flanged necks’ among the Kota Cina fine-paste sherds as well. 106 These alternative fine-paste vessel forms can also be extrapolated from comparing the measured and observed dimensions in the results with that of coarse- and medium-tempered earthenware. These comparisons lead to the conclusion that some of the sampled STA fine-paste ceramics in the STA sample share similar vessel shapes with coarse- and medium-tempered vessels, albeit in smaller and thinner-walled variations as they have the smallest average sherd size among all sampled earthenware ceramics. Hence, it is entirely possible that some fine-paste vessels at STA were also used as storage vessels, but likely for other kinds of goods or materials which are correspondingly smaller in size or lesser in quantity than those within coarse- and mediumtempered earthenware storage vessels. It should be noted however that it is not clear if these alternative fine-paste vessel forms from Palembang and Oc-Eo were the same as those from Singapore or Kota Cina. Although there were some ceramics in Palembang made from clay very similar to fine paste ware in the recent period, it is not likely that this clay source was used in pre-colonial times. These alternative fine-paste vessel forms seem to have been rarely exported as only fine-paste kendis have been found on shipwrecks thus far. 107 In addition, no accompanying fine-paste lid fragments were found, although a few were found in Kota Cina, which appear to have functioned as stoppers for orifices approximately 1.5cm in diameter; 108 hence it would appear that most fine-paste vessels were uncovered vessels. Another possibility is that the covers for fine-paste vessels were made of perishable materials, as some large Chinese jars of this period are known to have wooden lids. Only a few sherd surfaces exhibit signs of fire-clouding or sooting, but it is unlikely that they were used as cooking pots. Perhaps they functioned as incense urns as suggested by Chen due to their relatively smaller size and thinner walls, or maybe even oil lamps. 109 The presence of 37 flat or concave vessel-bases suggests that a significant number of fine-paste vessels do not have the characteristic round-bodied vessel form 106 Miksic, “Srivijaya: Political, Economic, and Artistic Frameworks for Analysis,” p. 200. Miksic, personal communication, 23 December 2011. Ibid. 109 Chen, Earth to Earth, p. 115. 107 108 181 of cooking vessels as well. The existence of a group of 10 fine-paste spouts also point to the specialized use of fine-paste ceramics as kendis, as only one other medium-tempered spout was found in the entire sampled STA artefact assemblage. One medium-tempered spout was also excavated on Fort Canning. 110 In particular, the ‘apple-green’ glazing on only the few red-bodied fine-paste sherds suggests that glazed fine-paste vessels formed a select group of earthenware ceramics with a higher economic and aesthetic value than the other earthenware varieties. As glaze is technically a type of glass, it renders the ceramic body impermeable and thus, non-porous to liquids. 111 It is therefore unlikely that glazed fine-paste earthenware functioned as an ideal water storage vessel as described earlier. The provenance and function of high-fired sherds, the fourth and last earthenware variety in the sampled STA arefacts, remain mysterious as they are too few in number and too fragmented for any comparisons or conclusions to be made. Nonetheless, differences with locally-made earthenware ceramics in terms of physical characteristics and manufacturing technology suggest that they, like fine-paste ceramics, are foreign in origin as well. Both the black-burnished and mercury-jar sub-varieties appear to be wheel-thrown and kiln fired which is evident from the presence of rilling and their relatively higher hardness rating. The absence of visible temper suggests the use of fine clay likely to be similar to that of fine-paste ceramics as well. Black-burnished sherds in particular have decorations which are similar to locallymade earthenware ceramics, but their surfaces exhibit signs of burnishing, evident from the lustrous sherd surface texture, which is a pottery finishing technique not found on any other earthenware variety or sub-varieties. Both high-fired sub-varieties each seem to have a distinct vessel form. While the precise shape remains unknown, uniformity in the measured dimensions of the black-burnished sherds suggest that they are fragments of a standard small and thinwalled vessel. Black burnishing as a finishing technique has been utilized at various times in 110 Miksic, personal communication, 23 December 2011. Shepard, Ceramics for the Archaeologist, p.44; Margaret Medley, The Chinese Potter. A Practical History of Chinese Ceramics (Oxford: Phaidon Press Limited, 1976), p. 40. 111 182 Thailand and India, but more comparative data is needed to identify these particular wares. 112 At the same time, mercury-jar type sherds, as their name suggests, do not assume any other vessel shape besides that of actual stoneware mercury-jars. While both sub-varieties are likely to be storage vessels as well, their unique manufacture and shape coupled with their rarity in both sherd and vessel counts suggests that these high-fired earthenware vessels may, albeit in a storage capacity, have a special or function in pre-colonial Singapore. II. Stoneware and Porcelain: High-Fired Imports for a Local Market In the case of the sampled STA stoneware and porcelain sherds, the manufacturing technique used to produce these vessels is easy to discern. Features such as the presence of rilling marks and glaze on the sherd surfaces, uniform sherd body thickness, consistent surface and glaze colours, the high hardness rating of these sherds, illustrates clearly that they were all fragments of wheel-thrown and kiln-fired ceramics. The precise identification of these sherds however, down to the exact region or even kiln site of origin as well as period of manufacture, is entirely dependent on comparisons with the existing knowledge of high-fired ceramic production and distribution. One may use the scientific technique of thermoluminescence (TL) dating to determine the absolute date when the ceramics were produced, but previous results have show that the range of error lies between +/- 90 to 110 years. 113 While sufficiently accurate for a prehistorian or palaeontologist, this range of error is too great for the historicalarchaeological study of pre-colonial Singapore which appears to have a narrow lifespan of approximately three centuries. Therefore, ceramic identification and dating still has to rely on ‘comparative stylistic grounds.’ 114 In particular, relative dating through the presence of Chinese trade ceramics is quite accurate because of the susceptibility of their styles (decorations, motifs and glaze patterns) to change over unit time. 115 Moreover, ceramic finds from most Southeast Asian archaeological sites ‘usually fall within a century or so in date’ as few sites were ever 112 Miksic, personal communication, 23 December 2011. Ibid., p. 89. 114 Ibid. 115 Brown & Sten Sjostrand, Maritime Archaeology and Shipwreck Ceramics in Malaysia (Kuala Lumpur: Department of Museums & Antiquities, 2004), p.15. 113 183 occupied continuously for more than two centuries. 116 Existing knowledge on pre-colonial ceramics however is still far from accurate and complete. To quote Roxanna Maude Brown, ‘there is no detailed, certain chronology for a majority of Chinese – as well as Southeast Asian – trade ceramics.’ 117 John S. Guy also notes that ‘little reliable information has been amassed from Chinese literary sources’ despite the long history of connoisseurship and scholarly interest in Chinese ceramics. 118 Therefore, similarities in physical characteristics, forms, decorative styles as well as archaeological and historical contexts will provide the most conclusive evidence in the identification of sampled STA stoneware and porcelain sherds. Unlike their low-fired earthenware counterparts, high-fired stoneware and porcelaeous ceramics have had a relatively shorter lifespan in Southeast Asia. According to Miksic, it was not until the Khmers (circa 9th century) where ‘the technique of producing pottery fired at temperature hot enough to melt the surfaces of the clay particles,’ otherwise known as sintering, was first developed indigenously in Southeast Asia. 119 ‘Khmer ware,’ as these pioneering Southeast Asian stoneware ceramics are called, is believed to be a result of the diffusion of Chinese pottery manufacturing technology as they share many similarities with contemporaneous products from the Xicun (西村 Xī Cūn) and Chao-an (潮安 Cháo Ān) kilns in the southern Chinese province of Guangdong (广东 Guǎng Dōng). 120 The Vietnamese, having had a long history of political and cultural interactions with the Chinese since their conquest by the Han dynasty in 111 BCE, started to produce stoneware ceramics approximately 2000 years ago as well. 121 On the other hand, mainland Southeast Asian potters never achieved the required temperatures to produce true porcelain. Customarily fired at 1280 to 1400ºC, porcelain ceramics are made of ‘white-firing, highly refractory kaolin clay which is relatively 116 James C. Y. Watt, “The Dating of Chinese Ceramics and Archaeological Sites in South-East Asia. A Preliminary Draft,” in Studies on Ceramics, p. 187. 117 Brown & Sjostrand, Maritime Archaeology and Shipwreck Ceramics in Malaysia, p.15. 118 Guy, Oriental Trade Ceramics in South-East Asia, p. 75-6. 119 Miksic, “Kilns of Southeast Asia,” p. 50; see also Guy, Ceramic Traditions of South-East Asia (Singapore: Oxford University Press, 1990), p. 10, 19-22, 27-8, 32, 36-41, 48; Brown, “Guangdong: A Missing Link to Southeast Asia,” in Guangdong Ceramics from Butuan and other Philippine sites, p. 81-5. 120 Rooney, Khmer Ceramics (Singapore: Oxford University Press, 1984), p. 24; see also Wong Wai Yee. A Preliminary Study of Some Economic Activities of Khmer Empire: Examining the Relationship between the Khmer and Guangdong Ceramic Industries during the 9th to 14th centuries. Ph.D Thesis. NUS, 2010. 121 Miksic, “Kilns of Southeast Asia,” p. 58; Rooney, Folk Pottery in South-East Asia, p. 3; Brown, The Ceramics of South-East Asia, p. 17-9. 184 free of impurities and mixed with quartz and ground, partially decomposed feldspathic rocks which acts as a flux.’ The characteristic translucency, hardness, and melodious ring’ of porcelain is only attained from the melting of feldspar at such high temperatures. 122 The Chinese themselves had only developed such ‘true translucent porcelain’ during the Tang dynasty (618-906), but held the porcelain monopoly until the Japanese and Germans discovered the technique in the 17th and 18th centuries. 123 Porcelain was first recorded as a maritime trade item only during the Song dynasty, a period where foreign trade flourished under the incentive to secure additional revenue especially after the Song defeat by the Khitan nomads in 1126. 124 However, the Thai and Vietnamese potters were able to produce high-fired, glazed ceramics which often resemble their Chinese porcelain counterparts, such as Thai celadon and Vietnamese underglaze cobalt blue, and more importantly, export them across the region as trade items between the 14th and 16th centuries. 125 As discussed earlier, the existing archaeological record illustrates that the production of wheel-thrown and kiln-fired stoneware ceramics spread across mainland Southeast Asia but never crossed south of the Isthmus of Kra into the equatorial belt of island Southeast Asia. Despite the lack in production capabilities, the peoples in the islands nonetheless managed to acquire not just stoneware but porcelain ceramics as well through a long established maritime trading network with China which is sometimes referred to as the Nanhai (南海 Nán Hǎi or ‘South Sea’) trade. 126 Chinese ceramics were apparently imported into Southeast Asia as early as the 1st to 2nd centuries with the recovery of Western Han dynasty Chinese earthenware again in the south Thailand region. 127 The earliest physical evidence of such maritime trade comes from the Belitung shipwreck. 128 Discovered in 1998 by Indonesian fishermen, it held a large 122 Rice, Pottery Analysis, p. 6. S. T. Yeo & Jean Martin, “Chinese Blue and White Ceramics – A Brief Introduction,” in Chinese Blue & White Ceramics, p. 16; Rice, Pottery Analysis, p. 7. 124 Wong, “Chinese Blue-and-White Porcelain and Its Place in the Maritime Trade of China,” p. 51; Miksic, “Research on Ceramic Trade,” p. 73-4. 125 Guy, Oriental Trade Ceramics in South-East Asia, p. 45-67. 126 Wang Gungwu, “The Nanhai trade: A Study of the Early History of Chinese Trade in the South China Sea,” in Southeast AsiaChina Interactions, compiled by Geoff Wade (Kuala Lumpur: MBRAS, 2007), p. 51-166. 127 Miksic, “Research on Ceramic Trade,” p. 71. 128 Also known as the Batu Hitam or Tang wreck. 123 185 cargo of Chinese Changsha ceramics dating to the 9th century. 129 This period is generally believed by many scholars to be ‘the first successful stage of Chinese ceramic exports’ at a voluminous level. 130 For the ‘people of the Malay region’ in particular, China was a ‘vital source’ of manufactured products which they could not produce between the 10th to 14th centuries. According to Heng, the importing of stoneware and porcelain ceramics in this region occurred at 2 levels: High-fired ceramics of lower quality were produced and exported from the southern ‘provincial’ or ‘folk’ kilns (民窑 Mín Yáo) of Guangzhou (广州 Guǎng Zhōu) in Guangdong province and Quanzhou in Fujian province. On the other hand, ‘national’ or ‘imperial’ kilns (官窑 Guān Yáo) at Jingdezhen (景德镇 Jǐng Dé Zhèn) and Jizhou (吉州 Jí Zhōu) in Jiangxi (江西 Jiāng Xī) province, Longquan in Zhejiang province and Dehua in Fujian province produced renowned and higher quality ceramics. 131 The Turiang, a 14th century (circa 1305-1370) shipwreck found closest to pre-colonial Singapore (off the east coast of the Malay Peninsula) in both time period and proximity, was found to have set sail from Guangdong with a cargo containing monochrome, brown- and green-glazed stoneware vessels made in the provincial kilns of Guangdong as well as green-ware vessels from the imperial kilns in Zhejiang. High-fired Thai (Sukhothai, Swankhalok and Suphanburi) and Vietnamese ceramics, the bulk of the ship’s cargo, was found alongside as well. 132 This is followed by the Nanyang (circa 1380), Longquan (circa 1400), Royal Nanhai (circa 1460), Xuande (circa 1540) and Singtai (circa 1550) shipwrecks, all of which held an assortment of high-fired Chinese, Thai and Vietnamese ceramics. 133 It stands to reason, therefore, that the sampled STA stoneware and porcelain sherds can only be ceramics of either a mainland Southeast Asian or Chinese provenance. This preliminary conclusion is attested to by the provenance of similar sherds from the adjacent pre-colonial 129 Michael Flecker, “A Ninth-Century AD Arab or Indian Shipwreck in Indonesia: First Evidence for Direct Trade with China,” WA 32,3 (February 2011), p. 339, 344. 130 Ho, “Problems in the Study of Zhejiang Green Glazed Wares,” in New Light on Chinese Yue and Longquan Wares, p. 195. 131 Heng, Sino-Malay Trade and Diplomacy, p. 149, 171-90. 132 Brown & Sjostrand, Turiang: A Fourteenth-Century Shipwreck in Southeast Asian Waters (Los Angeles: Pacific Asia Museum, 2000), p. 8-39. 133 Brown & Sjostrand, Maritime Archaeology and Shipwreck Ceramics in Malaysia, p. 43-51, 55-7; Brown, The Ming Gap and Shipwreck Ceramics of Southeast Asia, p. 41-3, 113-9, 124-29. 186 archaeological sites in Singapore. As early as 1822, Crawfurd observed the presence of ‘… various descriptions of pottery, some of which is Chinese and native’ on FTC, fragments of which were ‘in great abundance.’ 134 More than 150 years later, Miksic would arrive with the same conclusion in his study of stoneware and porcelain ceramics recovered from FTC. Specifically, Miksic assigned to the brittle and buff stoneware sherds a Fujian provenance and a north Vietnam and Guangdong provenance respectively. Porcelain sherds on the other hand were all thought to have originated from China and dated to the period between the late-Song to early-Ming dynasties (1279-1368). 135 Most green-wares were generally associated with the Longquan kilns, whereas some white-wares were identified to be from the Dehua kilns; the remainder of the sherds could not be positively identified. Most ceramics found across all other pre-colonial sites in Singapore are apparently Chinese and ‘stylistically dateable to the 14th century’ as well, but no specific kiln or region of orign was given. A few Vietnamese green overglaze and blue underglaze sherds as well as Thai Sawankhalok celadons and underglaze ceramic boxes and lids ‘stylistically dated’ to the 15th century and late-14th century respectively were found at sites along the Singapore River as well. 136 Presumably on the basis of these archaeological finds, Heng concludes that ‘Temasik imported large quantities of low unit-value ceramics’ from the provincial Guangdong and Fujian kilns. Heng states further that the neighbouring Riau islands, assumed to be ‘peripheral settlements dependent on the main port of Temasik,’ obtained their green- and white-wares as well as utilitarian stonewares from the pre-colonial settlement in Singapore. 137 Specifically, the majority of the sampled STA stoneware sherds appear to be south Chinese in origin. They can be further divided broadly along the provincial lines of Guangdong and Fujian. Comparisons with existing catalogues show that buff general pottery sherds within the sampled STA artefacts share a greater affinity with Guangdong than Fujian stoneware ceramics. Features such as sherd surface texture and colours, glaze colours, presence of lugs, 134 Crawfurd, Journal of an Embassy. Volume I, p. 72. Miksic, Archaeological Research, p. 67-87. Low, “Singapore from the 14th to 19th Century,” p. 20-9. 137 Heng, Sino-Malay Trade and Diplomacy, p. 182. 135 136 187 relative paucity of decorations and rim-shapes match those of known 11th to 12th century Song dynasty Guangdong stoneware vessels and fragments recovered from Tioman. 138 The sampled STA buff general pottery sherds also match 14th century Guangdong stoneware pottery recovered from the Turiang shipwreck as well as Moore’s ‘Kwangtung Ware’ found across Phases I (10th to 11th centuries), II (12th to 13th centuries) and III (14th to 15th centuries) sites in Sarawak. 139 Four of the sampled buff general stoneware sherds bear the marks of stamp impressed Chinese characters which have also been found on Guangdong stoneware vessels. 140 On the other hand, the sampled STA brittle general pottery sherds are likely to originate from Quanzhou in Fujian during the Song-Yuan period between the 12th to 14th centuries. These sherds have similar characteristics such as being ‘thinly potted, light for the size of the vessel, with a ‘rough and gritty’ inner surface ‘black and white impurities’ in the body,’ with vessels believed to be made at the ‘Quanzhou kiln complex.’ 141 Moore also suggested a ‘Fukien District’ provenance for Phase I and II Sarawak ‘Brittle Ware’ ceramics. Distinguished from the buff Guandong stoneware by the ‘hard, granular grey’ surface texture and colour, Moore believed that this difference in clay body is the primarily indicator of manufacture by two separate kiln traditions. 142 Nonetheless, three buff stoneware sherds also have moulded dragon reliefs which are commonly found on both Guangdong and Fujian stoneware vessels. 143 Indeed, Moore stated that ‘many striking similarities’ in physical features exist between the ceramics from both Guangdong and Fujian kilns which often ‘imitate each other and develop on very similar lines.’ 144 Moore extends the Fujian provenance in her study to encompass what she called ‘tall narrow jars’ or mercury-jars as well. 145 Specifically, a Quanzhou provenance can also be assigned to the sampled STA brittle mercury-jar sherds, as Tioman ‘stoneware storage bottles’ 138 Kwan & Lam, “Guangdong Wares. Song Dynasty 11th-12th century,” in A Ceramic Legacy of Asia’s Maritime Trade, p. 99-107. Brown & Sjostrand, Turiang, p. 33-4; Eine Moore, “A Suggested Classification of Stonewares of Martabani Type,” The Sarawak Museum Journal (SMJ) 18,36-7 (July-December 1970), p. 4, 29-55, 140 Kwan & Lam, “Guangdong Wares,” p. 107. 141 Kwan & Jean Martin, “Introduction to the finds from Pulau Tioman,” in A Ceramic Legacy of Asia’s Maritime Trade, p. 73-4. 142 Moore, “A Suggested Classification of Stonewares of Martabani Type,” p. 75-6. 143 Kwan & Lam, “Fujian Wares. Song and Yuan Dynasties 12th-14th centuries,” in A Ceramic Legacy of Asia’s Maritime Trade, p. 99-107; Moore, “A Suggested Classification of Stonewares of Martabani Type,” p. 35-9, 44, 47-8, 50-52. 144 Moore, “A Suggested Classification of Stonewares of Martabani Type,” p. 55-6, 76. 145 Ibid., p. 8. 139 188 attributed to the ‘Quanzhou kiln complex’ have the exact same vessel form, ‘heavy base, small mouth rim and potting rings’ as well as a ‘coarse and flecked with black and white grog’ body. 146 These bottles are specifically labelled by Chinese archaeologists as 小口瓶 (Xiǎo Kǒu Píng or ‘small-mouth bottle’) made specifically at 曾竹山 (Céng Zhú Shān or Cengzhushan) kilns which was part of the Quanzhou kiln complex as well. 147 The few buff mercury jars however do not match this provenance as they do not have the characteristic coarse surface and black grog albeit having the same vessel form. This difference suggests a slightly altered pottery manufacturing technique that may have occurred in a different kiln complex in either Fujian or Guangdong, but there is no reason to doubt a south Chinese provenance as well. The few remaining purple-ware sherds are likely to be from Yixing (宜兴 Yí Xìng or Yixing) kilns in the central Chinese province of Jiangsu (江苏 Jiāng Sū or Jiangsu) produced between the 12th to 14th centuries, as the ‘small storage jars’ appear to share the same distinct rim shape, described in the catalogue as a ‘broad folded back mouthrim’ as well as an extremely rough surface texture. Other similar features include a relatively small orifice as well as dark glaze and body colours. 148 With the exception of the rim and orifice, purple-ware vessels share a strikingly similar vessel form with that of mercury jars: both appear to be relatively small but tall and narrow bottles. It is probably because of this similarity in vessel shape as well as their common occurrence in Phase I and II Sarawak sites that Moore initially gave purple-ware vessels, which she labelled as ‘Jars with a folded rim,’ a Fujian provenance. 149 Stoneware vessel characteristics and forms are intrinsically linked to their functions. As seen in the results, these sherds are much heavier, larger, thicker and coarser than their earthenware counterparts. Most rim sherds have, in proportion to their approximate vessel size, a relatively constricted orifice. In the case of mercury-jar vessels, this orifice is extremely small. 146 Kwan & Lam, “Guangdong Wares,” p. 113. Kwan & Martin, “Introduction to the finds from Pulau Tioman,” p. 74. 148 Ibid., p. 76; Kwan & Lam, “Jiangsu Wares. 12th-14th century and Guangdong Wares and other kiln sites,” in A Ceramic Legacy of Asia’s Maritime Trade, p. 119. 149 Moore, “A Suggested Classification of Stonewares of Martabani Type,” p. 8-11. 147 189 Some general stoneware and purple-ware sherds have lugs attached on their shoulder near the rims as well. Where the small orifice of the mercury-jars is easily covered, covers can also be tied on larger orifices through the lugs as well. These physical features are all clear indicators of specialized use as storage vessels. According to Moore, stoneware storage vessels are particularly strong enough to ‘withstand transportation, often over very difficult country, and capable of being closed, to keep the contents safe from insects and rats.’ 150 The veracity of the inference that stoneware ceramics were specialized long-distance storage vessels is no better illustrated than by their recovery in situ on various shipwrecks. The Ping Zhou Ke Tan (萍州 可談 Píng Zhōu Kě Tán or Discourse on the Floating Islands), written by Zu Yu (朱彧 Zhū Yù) in the 12th century, describes how ‘the greater part of the [ship’s] cargo consists of pottery (陶器 Táo Qì), the small pieces packed in the larger, till there is not a crevice left.’ 151 Stoneware vessels of various sizes onboard the Belitung wreck were used to store a wide range of trade goods including numerous stacks of Changsha bowls, lead ingots and even star anise. 152 Stoneware jars on the Turiang wreck are also recorded to have contained Chinese brown-glazed covered boxes as well as bones of fishes probably kept as food for the crew. 153 On land, a 10th century Kwangsi (广西 Guǎng Xī or Guangxi) ‘green glazed ‘martavaan’ (stoneware)’ vessel found at Pekalongan, Java contained a number of socketed iron tools or implements. 154 Clearly, the items stored within these vessels are not only limited to food or perishable goods, but a wide horizon of items depending on the needs of their respective owners. The vessel form of mercury jars in particular is ‘highly suited for carrying mercury,’ a dense metal in liquid form at room temperature but 13.6 times heavier than water; the thick but small bases allow the jars to support its ‘great density’ as well as divert the ‘pressure of mercury on diagonally sloping sides of the vessel’ respectively; their relatively tall height coupled with 150 Ibid.,”p. 1-2. Hirth & Rockhill, Chu-fan-chi, p. 31. 152 Flecker, “A Ninth-Century AD Arab or Indian Shipwreck in Indonesia,” p. 342, 349; Flecker, “A Ninth-Century Arab Shipwreck in Indonesia. The First Archaeological Evidence of Direct Trade with China,” in Shipwrecked. Tang Treasures and Monsoon Winds, ed. Regina Krahl et. al. (Washington D. C.: Arthur M. Sackler Gallery, Smithsonian Institution, 2010), p. 108-9. 153 Brown & Sjostrand, Turiang, p. 14. 154 Loewenstein, “The Origin of the Malayan Metal Age,” p.78 (Plate 6). 151 190 a small orifice reduces the amount of spillage during a long land or sea voyage. Moreover, the unglazed surface of mercury jars suit the storage of mercury as it, unlike other forms of liquids, ‘will not penetrate the most porous vessel.’ 155 It was also conjectured that liquid mercury was used in Tantric religious ceremonies, as medicine and a means of extracting gold from its ore or purifying it. 156 Nonetheless, mercury jars are known to have stored wine and sauces. 157 Chinese archaeologists have identified mercury jars as wine bottles, which Heng later extrapolated to contain southern Fujian rice wine in particular. 158 The widespread use of these jars in pre-colonial Southeast Asia is testified by their recovery not just in Singapore, but also in Kedah, Sarawak and Cambodia at archaeological sites dating between the 11th and 14th centuries. 159 Remarkably, they were reportedly still used by the Jakun tribe of Orang Asli in Padang Siam, Kuala Seria and Pahang as late as the 20th century. 160 It is also possible that purple-ware vessels were also used in the transportation and export of mercury and other liquids because of their similarity in vessel form with mercury-jars. Besides general storage jars and mercury jars, a third vessel form exists within the sampled STA stoneware assemblage. Buff stoneware sherds with ‘Bulbed-Rectangular,’ ‘Bulbed-Rectangular with folded ‘Pie-Crust’ fringe,’ ‘Hooked Bulbed-Rectangular’ and ‘FlatBulbed’ shaped rims appear to be fragments of large basins, as seen in their relatively larger external diameters as well as their almost vertical vessel walls which extends immediately from the rims. Moore described these basins as ‘heavily potted vessels with a wide, concave base, a slightly flaring side, often inturned at the top and a thick, rolled or everted rim’ and they apparently occur only in Phase I and II Sarawak sites. 161 Similar basins were also found in the Tioman 11th-12th centuries Guangdong stoneware ceramic assemblage as well as in the 14th 155 F. E. Treloar, “Stoneware Bottles in the Sarawak Museum: Vessels for Mercury Trade?” SMJ 20,40-1 (January-December 1972), p. 382-3. 156 Miksic, Archaeological Research, p. 69. 157 Kwan & Martin, “Introduction to the finds from Pulau Tioman,” p. 74; Treloar, “Stoneware Bottles in the Sarawak Museum,” p. 378, 83. 158 Heng, Sino-Malay Trade and Diplomacy, p. 189. 159 Treloar, “Stoneware Bottles in the Sarawak Museum,” p. 377-8; McKinnon, “Oriental Ceramics Excavated in North Sumatra,” Transactions of the Oriental Ceramic Society (TOCS) 41 (1975-1977), p. 76. 160 Kwan & Martin, “Introduction to the finds from Pulau Tioman,” p. 74. 161 Moore, “A Suggested Classification of Stonewares of Martabani Type,” p. 41-2. 191 century Turiang wreck. 162 Several 10th to 12th centuries Guangdong stoneware basins, including one with a ‘foliate pie-crust-like mouth rim,’ were also found in pre-colonial Philippines as well. 163 On the other hand, the provenance of the sampled STA porcelain sherds appears to be much more diverse than the simpler Guangdong-Fujian provenance dichotomy of stoneware sherds. The majority of porcelain sherds appear to be fragments of vessels produced and imported from Guangdong, Fujian, Jiangxi and Zhejiang provinces in China. From the 11th century onwards, Chinese trade ceramics were being shipped overseas en masse from the international ports of Guangzhou, Quanzhou, Fuzhou ( 福 州 Fú Zhōu, Fujian province), Mingzhou/Ningbo (明州/宁波 Míng Zhōu/Níng Bō, Zhejiang province) and Lin’an/Hangzhou (临安/杭州 Lín Ān/Háng Zhōu, Zhejiang province). 164 The prevalence of carved and incised decorations, characteristic of the ‘variable and individualized’ ceramics of Song potters, on the porcelain sherds, over moulding, luting, stamping and appliqué decorations widely adopted during the Yuan dynasty (1271-1368) of which only one example is present, suggests a general production period of the late-13th to early-14th centuries. 165 At the same time, the prevalence of green-ware sherds over other porcelain varieties reflect a distribution and consumption pattern that coincides with that of excavated archaeological sites of Southeast Asia as well as the records of various historical settlements of the same period. 166 Sherds with a fine pale grey to white fabric, thick and opaque green to olive or even golden glaze and commonly adorned with ‘lotus-leaf’ (foliated) reliefs on the exterior wall or fluted walls on the interior cavetto closely resemble green-ware ceramics made in the Longquan kiln complex of Zhejiang province. This renowned centre of green-ware production was known 162 Kwan & Lam, “Guangdong Wares,” p. 101; Brown and Sjostrand, Turiang, p. 33-4. Bautista et. al., “Guangdong Wares. 10th to 12th century,” in Guangdong Ceramics from Butuan and other Philippine sites, p. 108-10. 164 Heng, Sino-Malay Trade and Diplomacy, p. 35, 19-36; see also Lin Shimin, “Zhejiang Export Green Glazed Wares: Ningbo Data,” in New Light on Chinese Yue and Longquan Wares, p. 141-2, 151-4; Wanda Garnsey & Rewi Alley, China. Ancient Kilns and modern ceramics. A guide to the potteries (Canberra: Australian National University Press, 1983), p. 92, 118, 129. 165 Guy, Oriental Trade Ceramics in South-East Asia, p. 24; Catherine Teo, “Qingbai Ware for Export,” in Qingbai Ware: Chinese Porcelain of the Song and Yuan Dynasties, ed. Stacey Pierson (London: Percival David Foundation of Chinese Art, 2002), p. 246. 166 Ibid. 163 192 to have produced countless bowls, large dishes, jarlets, vases and many other vessel forms between the 11th to 16th centuries. 167 Longquan green-ware ceramics recovered from the Southeast Asian archaeological context however are typically dated by scholars to the 13th and 14th centuries. 168 The lotus-leaf decorations correspond specifically to those of ‘phase three, four and five’ (early-13th century to first half of the 14th century) Longquan green-ware ceramics recovered from the ‘eastern Longquan kiln district’ in Zhejiang. 169 One base sherd with a swirl relief on the interior medallion surface in particular is exactly the same as another with a recorded Longquan provenance at the Muzium Seni Asia. 170 Longquan green-ware ceramics also have a darker green glaze colour than those from Guangdong and Fujian. 171 This has been attributed to the use of coal as kiln fuel in the North (Zhejiang) and wood in the South (Guangdong and Fujian). 172 The glaze is also thicker and much more opaque due to the application of two or more glaze layers on ceramic body. 173 The relatively coarse quality of these ‘Longquan’-type sherds – gritty surface, imperfect glazing and partially or unglazed foot – points to the manufacturing period of the 13th to 15th centuries as well. 174 Decorations on Longquan ceramics are also known to be generally restrained until the middle of the 13th century, where they ‘became more elaborate in technique and frequently more florid in style.’ 175 Seen in this context, the STA Longquan-type green-ware sherds date to this transitional period as the decorations are neither too simple like those of the Southern Song dynasty nor too elaborate as found on 14th century examples from the Yuan dynasty up till the demise of the kiln complex in the late-16th century. 176 This dating – on the basis of differentiation in decoration – is 167 Kwan & Martin, “Introduction to the finds from Pulau Tioman,” p. 77. Ho, “Problems in the Study of Zhejiang Green Glazed Wares,” p. 191-2. 169 Kamei Meitoku, “Chronology of Longquan Wares of the Song and Yuan Periods,” in New Light on Chinese Yue and Longquan Wares, p. 56-66, 73-84. 170 Kwan & Lam, “Zhejiang Wares. Song and Yuan Dynasties 11th-14th century,” in A Ceramic Legacy of Asia’s Maritime Trade, p. 124 (Plates 276-9). 171 Lu Yaw, “Introduction,” in Chinese Celadons and other Related Wares in Southeast Asia, compiled by Southeast Asian Ceramic Society, Singapore (Singapore: Arts Orientalis, 1979), p. 26. 172 Ibid., p. 15. 173 Medley & Stacey Pierson, Illustrated Catalogue of Celadon Wares in the Percival David Foundation of Chinese Art (London: University of London, School of Oriental and African Studies (SOAS), 1997), p. 9; Rose Kerr, Song Dynasty Ceramics (London: V&A Publications, 2004), p. 92. 174 Kwan & Lam, “Zhejiang Wares,” p. 123-4; see also Kwan & Lam, “Ceramics found at Kampong Juara in the Muzium Sultan Abu Bakar, Pekan 12th-19th century,” in A Ceramic Legacy of Asia’s Maritime Trade, p. 142. 175 Medley & Pierson, Illustrated Catalogue of Celadon Wares, p. 9-10; see also Ye Wencheng, “A Preliminary Discussion of Fujian Wares Made In Imitation of Zhejiang Green Glazed Wares,” in New Light on Chinese Yue and Longquan Wares, p. 124-5. 176 Medley & Pierson, “Catalogue,” in Illustrated Catalogue of Celadon Wares, p. 19-60; Malcom V. Quie et. al., “Illustrated Catalogue of Exhibits. Chinese Celadons,” in Chinese Celadons and other Related Wares in Southeast Asia, p. 126-225. 168 193 reaffirmed with the recovery of ‘Southern Sung-Yuan Lung-ch’uan celadon wares’ from the Sinan shipwreck off the south-west coast of Korea. 177 Recovered Longquan green-ware sherds which are ‘symmetrically refined usually with no decoration or a simple decoration of carved lotus petals on the exterior of the vessel’ were attributed to the ‘Southern Sung period,’ whereas ceramics with more ‘ornate’ decorations were thought to ‘bear the accepted characteristics of Yuan wares.’ 178 The popularity of Longquan green-ware ceramics in overseas markets is archaeologically proven; more than 60 percent of total excavated artefacts between 1213 and 1274 at the port of Ningbo consisted of ‘Longquan ware’ which became the main export ceramic of Zhejiang province up to the mid-Yuan period (circa 1323). 179 Conversely, sherds with a thinly applied light green to green glaze on the body but none on the base or foot share the same characteristics typically found on Guangdong bowls and dishes manufactured between the 11th and 14th centuries. 180 A green-ware base sherd with a steep vessel wall gradient in particular resembles that of a ‘Bullet jar’ recorded to have been made at a number of Guangdong kilns. 181 Some of these green-ware sherds may have been from the Minqing (闽清 Mǐn Qīng), Anxi (安溪 Ān Xī) and Tong’an (同安 Tóng Ān) kilns of Fujian province as well, as they are also known to have produced finely made green-ware bowls with combed, incised and impressed decorations in the same period. 182 However, many other Fujian ‘interior’ and ‘coastal’ kilns as well as kilns in the Quanzhou vicinity also made green-ware ceramics with ‘close imitations of Longquan-type glazes, shapes and decoration’ specifically for export to overseas markets, with Quanzhou as the main port exporting these imitations. 183 Nonetheless, Fujian imitations can be differentiated from Longquan green-ware ceramics through a number of characteristics. Fujian imitations have unglazed stacking rings on the 177 John Ayers, “A Note on Celadon Wares from Sinan,” in Chinese Celadons and other Related Wares in Southeast Asia, p. 4955. 178 Quie, “A Chronological Study of Ceramics Excavated off the Sinan Sea Shore in 1976 and 1977,” in Chinese Celadons and other Related Wares in Southeast Asia, p. 43-4. 179 Lin, “Zhejiang Export Green Glazed Wares: Ningbo Data,” p. 151-4. 180 Lam, “Northern Song Guangdong Wares,” in A Ceramic Legacy of Asia’s Maritime Trade, p. 1, 15, 18; Kwan & Lam, “Guangdong Wares,” p. 86; Bautista et. al., “Guangdong Wares,” p. 90, 105. 181 Kwan & Lam, “Jiangsu Wares,” p. 120. 182 Kwan & Martin, “Introduction to the finds from Pulau Tioman,” p. 74-5; Kwan & Lam, “Fujian Wares,” p. 114-6; Bautista et. al., “Fujian Wares. 12th to 14th century,” in Guangdong Ceramics from Butuan and other Philippine sites, p. 119-21. 183 Ye, “A Preliminary Discussion of Fujian Wares Made In Imitation of Zhejiang Green Glazed Wares,” p. 122, 125-6. 194 interior bases, a feature not commonly found on Longquan ceramics as most vessels were placed in individual compartments or ‘saggars’ instead of being stacked on top of each other for firing by the 13th century. 184 Hence, sampled STA green-ware base sherds with unglazed stacking rings were more likely products of Fujian kilns. Fujian vessel forms are also primarily utilitarian – bowls, plates etc – and unlike Longquan ceramics, none were ever made for purely art’s sake. Just like Guangdong green-ware ceramics, the glaze on Fujian imitations is quite different from that of Longquan as well. The glaze is generally of a ‘greyish tone’ ranging from yellowish-green to greyish-green in colour and not ‘as neatly and evenly applied as the glaze on Zhejiang Longquan wares.’ 185 Green-ware ceramics were also made in Jiangxi, Shaanxi (陕 西 Shǎn Xī) Henan (河南 Hé Nán), Hubei (湖北 Hú Běi) and Hunan (湖南 Hú Nán) as well. 186 Nonetheless, they would probably still be transported to the ports designated by the Chinese imperial government for foreign trade and export in Guangdong, Fujian and Zhejiang. 187 The sampled white-ware sherds were most likely produced in southern Chinese kilns as well, with the two principal white-ware ceramics kiln areas at Dehua and Jingdezhen. 188 As mentioned earlier, sherds with ‘creamy offwhite tending to yellow or beige’ glaze are known to be ceramics produced in the Dehua kilns of Fujian province. 189 These sherds bear a close resemblance, if not similar, to Dehua ceramics – also known as blanc de chine – recovered from Tioman and Butuan dated to the period from the 12th to 14th centuries. 190 The provenance of the remaining white-ware sherds unfortunately cannot be precisely determined. From the late-8th or 9th centuries onwards, kilns across China were producing and refining white-ware ceramics in ‘the persistent search for a pure white body.’ 191 However, it was in the south that the first porcelain ceramics were produced due to the supplies of pure (low iron content) kaolinic clay 184 Li Dejin, “Technology of Longquan Ware Manufacturing,” in New Light on Chinese Yue and Longquan Wares, p. 89, 95. Ye, “A Preliminary Discussion of Fujian Wares Made In Imitation of Zhejiang Green Glazed Wares,” p. 123-4; see also Li, “Technology of Longquan Ware Manufacturing,” p. 90-2. 186 Ho, “Yue-Type and Longquan-Type Green Glazed Wares made Outside Zhejiang Province,” in New Light on Chinese Yue and Longquan Wares, p. 103, 108-10; Lu, “Introduction,” p. 27. 187 Heng, Sino-Malay Trade and Diplomacy, p. 39, 42-3, 48, 55, 65, 188 Guy, Oriental Trade Ceramics in South-East Asia, p. 71 189 Ibid.; see also Kwan & Martin, “Introduction to the finds from Pulau Tioman,” p. 75-6. 190 Kwan & Lam, “Fujian Wares,” p. 116-7; Kwan & Lam, “Ceramics found at Kampong Juara in the Muzium Sultan Abu Bakar, Pekan,” p. 140; Bautista et. al., “Fujian Wares, p. 118, 122-3; see also Guy, “Catalogue: Chinese Ceramics,” in Guy, Oriental Trade Ceramics in South-East Asia, p. 83-4 (Plates 18-21). 191 Medley, The Chinese Potter, p. 97-102. 185 195 and felspathic China stone (白墩子 Bái Dūn Zǐ) which were especially plentiful in the vicinity of Jingdezhen. 192 These were essential in the production of the vitrified translucent body that defines porcelain ceramics. A few white-ware sherds can be described as of the Qingbai (青白 Qīng Bái or, also known as 影青 Yǐng Qīng or Yingqing) sub-variety due to the bluish-white colour of their glaze. 193 This is significant in terms of dating as Qingbai ceramics are specifically products of the Song period but ceased in the Ming dynasty. 194 However, kilns in ‘forty-four counties of nine Chinese provinces’ were making Qingbai ceramics, with those at Jingdezhen produced Qingbai ceramics of the finest quality and in the greatest number. 195 Kilns in the Fujian and Guangdong provinces are known to produce both white-wares and Qingbai as well. 196 Another distinguishing feature of some sampled STA white-ware sherds is the application of dark olive brown or black underglaze spots as decoration. Composed of iron oxide, these ferruginous spots are found on a variety of vessel forms and porcelain ceramics as early as the 4th century. 197 Underglaze iron-spotting decoration was revived on green-ware (also known as ‘tobi senji’), white-ware and Qingbai ceramics of various vessel forms during the 13th to 14th centuries. 198 It became a popular treatment during the Yuan dynasty and was applied to a variety of vessel forms but especially on ewers, the likes of which are ‘commonly excavated in Southeast Asia.’ 199 The sampled spotted sherds resemble iron-spotted ceramics of the ‘greyish, puddy-like Guangdong type’ than their ‘whiter’ counterparts from Jingdezhen and Jiangxi. 200 A few sherds in particular appear to be fragments of a lobed Guangdong jarlet similar to one held at the Chinese University of Hong Kong. 201 Their distinct shape is also 192 Ibid., p. 100-1, 164-5; Rosemary Scott, “Introduction: Qingbai Porcelain and its Place in Chinese Ceramic History,” in Qingbai Ware, p. 7-8; Garnsey & Alley, China, p. 107. 193 Kerr, Song Dynasty Ceramics, p. 96. 194 Pierson, “Qingbai Porcelain: Technology, Forms and Decoration,” in Qingbai Ware, p. 17, 20. 195 Scott, “Introduction,” p. 6; see also Amy Barnes, “Catalogue entries,” in Qingbai Ware, p. 27-233. 196 Kwan & Martin, “Introduction to the finds from Pulau Tioman,” p. 73-5; Kwan & Lam, “Guangdong Wares,” p. 86-94, 96-8; Kwan & Lam, “Fujian Wares,” p. 114-6; Kwan & Lam, “Ceramics found at Kampong Juara in the Muzium Sultan Abu Bakar, Pekan,” p. 140; Bautista et. al., “Guangdong Wares,” p. 90, 92-3, 96-7, 99-104; Bautista et. al., “Fujian Wares, p. 118, 121-2, 124. 197 Medley, The Chinese Potter, p. 69. 198 Ibid., p. 152; see also Guy, “Catalogue: Chinese Ceramics,” p. 88-91 (Plates 42-9); S. R. Parker, “Chinese Celadons and Other Wares Excavated in Sarawak,” in Chinese Celadons and other Related Wares in Southeast Asia, p. 61 (Plate F); Quie et. al., “Illustrated Catalogue of Exhibits. Chinese Celadons,” p. 146 -7 (Plate 56), 192-5 (Plates 116, 118-121) 199 Barnes, “Catalogue entries,” in Qingbai Ware, p. 130 (Plate 68), 178 (Plate 96). 200 Lam, “Decorative Techniques and Motifs in Guangdong Trade Wares of the Song Dynasty,” in Guangdong Ceramics from Butuan and other Philippine sites, p. 51. 201 Lam, “Northern Song Guangdong Wares,” p. 74. 196 similar to a 14th century ‘balimbing’ jarlet found in the Philippines which resembles the shape of a starfruit. 202 Most exported Qingbai ceramics are also observed to be relatively ‘coarser’ than those produced at the Jingdezhen kilns which reflect ‘a fairly consistent standard of production with relatively few variations in quality.’ 203 Unlike previous porcelain varieties, most of the sampled blue-and-white sherds have a firmer provenance: the Jingdezhen kilns of Jiangxi province during the Yuan dynasty. The earliest Chinese ceramics decorated with a blue pigment derived from cobalt ore are dated as early as the Tang dynasty as testified by their presence on the Belitung wreck. 204 However, the design of the underglaze cobalt blue motifs on the sherds matches that of blue-and-white ceramics during the Yuan period, which is fortuitously in sharp contrast to that of the later Ming and Qing blue-and-white ceramics. Yuan blue-and-white ceramics typically have ‘rich’ but ‘neat’ motifs (裝飾繁縟工整 Zhuāng Shì Fán Rù Gōng Zhěng) of ‘thick’ cobalt blue (青料濃 厚 Qīng Liào Nóng Hòu) which is sometimes ‘uneven’ in shade (青花色調濃淡不均 Qīng Huā Sè Diào Nóng Dàn Bù Jūn), as opposed to the ‘unrestrained’ and ‘lively’ brushstrokes (飾紋的 用筆奔放、活潑 Shì Wén De Yòng Bǐ Bēn Fàng、Huó Pō) of the Ming period. 205 The motifs themselves, consisting mainly of ‘formal’ and floral patterns, are also much simpler during the Yuan period, as opposed to the more elaborate depictions of humans, creatures, buildings and landscapes which dominate Ming blue-and-white ceramics. 206 Yuan blue-and-white vessels recovered in the Philippines bears close resemblance to the sample sherds in the decorations and forms as well. 207 Specifically, a sampled base sherd bears the exact same encircled foliage motif as found on a 14th century Yuan ‘Shufu-type bowl,’ whereas the two angled body sherds 202 Larry Gotuaco, “Coloured Plates. Chinese Blue and White Ceramics. Yuan Period. 14th century,” in Chinese and Vietnamese Blue and White Wares Found in the Philippines, ed. Larry Gotuaco et. al. (Makati City: Bookmark, Inc, 1997), p. 74 (Plate Y37b). 203 Teo, “Qingbai Ware for Export,” in Qingbai Ware, p. 246. 204 Guy, Oriental Trade Ceramics in South-East Asia, p. 73; see also Jessica Hallett, “Pearl Cups Like the Moon. The Abbasid Reception of Chinese Ceramics,” in Shipwrecked, p. 80. 205 楊勝德 (Yáng Shèng Dé) & 馬紀恩 (Mǎ Jì Ēn), “中國的青花瓷 (Zhōng Guó De Qīng Huā Cí or Blue-and-white Ceramics of China),” in Chinese Blue & White Ceramics, p. 80. 206 S. T. Yeo & Jean Martin, “Catalogue of the Exhibits. Yuan Dynasty 1280-1368,” “Early Ming 1400-1487,” “Middle Ming 1488-1566,” “Late Ming 1566-1643” & Transitional Ming/Ching 1630-1660,” in Chinese Blue & White Ceramics, p. 95-101, 10321, 123-61, 163-215 & 217-23; see also Kerr, The World in Blue and White. An exhibition of Blue and White ceramics, dating between 1320 and 1820, from members of the Oriental Ceramic Society (London: The Oriental Ceramic Society, 2003), p. 5-51; Rita C. Tan, “Coloured Plates. Early Ming Period. Late 14th-Mid 15th Centuries,” in Chinese and Vietnamese Blue and White Wares Found in the Philippines, p. 109-79. 207 Gotuaco, “Coloured Plates. Chinese Blue and White Ceramics. Yuan Period. 14th century,” p. 32-74. 197 appear to be either the body fragments of an octagonal ‘mei-ping’ (梅瓶 Méi Píng or ‘Plum vase’) or the neck fragments of an octagonal ‘yuhuchun’ vase of the same period, both of which are noted to be ‘extremely rare’ due to their distinctive shape. 208 It should be noted that Chinese archaeologists believe the production of iron-spotted Qingbai and blue-and-white ceramics directly replaced that of plain Qingbai and white-ware ceramics during the 14th century. 209 Since the majority of sampled porcelain ceramics originated from Chinese kilns, it is reasonable to assume that most, if not all of the misfired or degraded sherds share the same provenances as well. While it is difficult to differentiate between the two conditions, it is clear that this group of sherds are of a much lower quality and hence, value, than the abovementioned porcelain varieties. In his study of similar misfired sherds with underglaze blue decorations recovered from the FTC site, Miksic has suggested that these sherds may ‘represent an attempt by less skillful technicians to produce porcelain’ at a different kiln area outside Jingdezhen. 210 As the sherds were deposited on the former coastline of Singapore, degradation would definitely occur due to weathering processes from the sand and seawater over the centuries; it is also entirely possible that part of the maritime ceramic trade involved less-than-perfect porcelain articles which bear some defects, particularly in glaze quality, but not too much to be discarded as kiln wasters. The slightest accidents or smallest mistakes at any stage of the ceramic production sequence will result in defects or deformities, as illustrated by the variety of terms used to describe them in A Dictionary of Chinese Ceramics. 211 A few porcelain sherds however appear to have non-Chinese provenances. Multiple carved lines on the interior surface of a green-ware sherd form a ‘stylized onion-skin medallion’ motif which is commonly and uniquely found on exported Thai Sawankhalok (also known as Si Satchanalai) green-ware ceramics. 212 A few examples can be found within various 208 Ibid., p. 60 (Plate Y25), 33 (Plate Y1), 50 (Plate Y15). Ibid., p. 74. 210 Miksic, Yap, Sam Fong Yau Li & Kebao Wan, “Analyses of Some Yuan Dynasty Artifacts Excavated in Singapore.” Conference Paper. Fifth Chinese Symposium on Archaeometry. University of Science and Technology of China, Hefei. October 14-9, 1998, p. 1, 3. 211 Wang, A Dictionary of Chinese Ceramics, p. 143-6. 212 Brown, The Ceramics of South-East Asia, p. 73. 209 198 catalogues as well as the artefact assemblages of the Nanyang, Longquan, Ko Khram (circa 1450 to 1475) and Royal Nanhai wrecks. 213 A popular regional green-ware ceramic alternative, Brown observed that Sawankhalok green-glazed vessels was exported as early as the late-14th century, with Sawankhalok celadon plates comprising the primary cargo for all mid-15th century shipwrecks found in the region. 214 Four blue-and-white sherds may also be Vietnamese as they bear underglaze motifs stylistically different from the rest of the Chinese blue-and-white samples. Underglaze cobalt blue decoration was first applied on Vietnamese ceramics during the second half of the 14th century, reaching its peak as an export good in Southeast Asia during the 15th century. 215 Unfortunately, the sherds were too fragmented and motifs too incomplete for precise identification. Given the looseness of the sand in which these artefacts were deposited, it is quite possible that these sherds were intrusive artefacts from the previous stratigraphic layer which sank into the 14th century. Porcelain vessel forms and functions are the most varied of the three ceramic categories. Dishes, basins, plates, cups, bowls, jarlets, covered boxes and vases can be reasonably discerned from the shape and characteristics of the sampled sherds. Like many riverine and coastal sites such as Kota Cina in Sumatra, these porcelain vessels were probably used alongside stoneware and earthenware ceramics for domestic purposes in pre-colonial Singapore. 216 While cups and bowls are common across porcelain varieties, some vessel forms appear to be more associated with specific varieties. Dishes and plates are only found within green-ware samples, whereas covered boxes are mainly white-ware ceramics. Despite their existence and distribution across Southeast Asia in the same period, noticeably absent from the porcelain samples are blue-and-white plates, dishes and jarlets and iron-spotted green-ware ceramics (‘tobi senji’), as well as other contemporaneous vessel forms common to all three 213 Brown, The Ming Gap and Shipwreck Ceramics of Southeast Asia, p. 113 (Plate 30), 116 (Plate 33), 120 (Plate 37); see also Brown & Sjostrand, Turiang, p. 40-58; Quie et. al., “Illustrated Catalogue of Exhibits. Other Related Wares,” in Chinese Celadons and other Related Wares in Southeast Asia, p. 278-9 (Plate 240), p. 284-5 (Plate 248); Hiromu Honda & Noriki Shimazu, “The Collection,” in Hiromu Honda & Noriki Shimazu, The Beauty of Fired Clay. Ceramics from Burma, Cambodia, Laos and Thailand (Kuala Lumpur: Oxford University Press, 1997), p. 92 (Plate 72a-c), 98 (Plate 79), 104 (Plate 87 a, b), 164 (Plate 203a, b). 214 Brown, The Ming Gap and Shipwreck Ceramics of Southeast Asia, p. 52, 54; see also Miksic, “Kilns of Southeast Asia,” p. 625. 215 Guy, Oriental Trade Ceramics in South-East Asia, p. 73. 216 McKinnon, “Yue and Longquan Wares in Sumatra,” in New Light on Chinese Yue and Longquan Wares, p. 291. 199 porcelain varieties such as jars, ewers and kendis. 217 However, not all porcelain ceramics were used according to their intended functions. Two gacuks, one hewed from a Longquan-type green-ware ceramic and another from a from a white-ware base sherd, are clearly identified among the porcelain samples. Similar discs cut in different sizes and thickness from terracotta or glazed foreign ceramics were found across Southeast Asia and were hypothesized to be used in either disc-throwing games or for ceremonial purposes. 218 Although absent from the present assemblage, Ming blue and white sherds were also known to have been ‘smoothened into handy disc shapes for use as gambling counters’ in Malayan villages as well. 219 Porcelain ceramics were not just imported commodities, but likely functioned as media of exchange between Chinese and local merchants in the conduct of trade in pre-colonial Singapore. Southeast Asian societies were known to adopt monetary standards based on ‘a directly measurable quantity’ of highly valued substances or commodities. 220 When the Song court banned the export of precious metals such as gold and silver as well as copper cash in 1127, 1133, 1175 and 1219, porcelain ceramics were one of the commodity-substitutes used in overseas maritime trade and transactions. Ceramics were specifically named as one of the products to be used in place of gold and silver in the 1219 ban. 221 In exchange, Chinese traders received various aromatic products, spices and precious items such as pearls, ivory, rhinoceros horns and other tropical produce to satisfy the ‘basic needs of China’ from Southeast Asian merchants. 222 III. Metals and Stones: Craft and Currency The material culture of pre-colonial Singapore extends to non-ceramic artefacts as well. A portion of recovered metal artefacts were likely crafted by local blacksmiths who probably 217 See Quie et. al., “Illustrated Catalogue of Exhibits. Chinese Celadons,” p. 126-225; Yeo & Martin, “Catalogue of the Exhibits. Yuan Dynasty 1280-1368,” p. 95-101; Guy, Oriental Trade Ceramics in South-East Asia, p. 78-103; Miksic & Ong, “Catalogue,” p. 144-152. 218 Hasan Muarif Ambary. Laporan Penelitian Arkeologi Banten 1977. Jakarta: Pusat Penelitian Arkeologi Nasional, 1977, cited in Sri Soejatmi Satari, “A Preliminary Study on the Local Ceramics of Trowulan,” in Studies on Ceramics, p. 43. 219 Michael Sullivan, “Notes on Chinese Export Wares in Southeast Asia,” TOCS 33 (1961-62), p. 61. 220 Wicks, Money, Markets, and Trade in Early Southeast Asia, p. 301. 221 Heng, Sino-Malay Trade and Diplomacy, p. 165, 173-4. 222 Wong, “Chinese Blue-and-White Porcelain and Its Place in the Maritime Trade of China,” p. 51. 200 worked alongside the earthenware potters as well. This is supported by the recovery of metal slag and objects together in the same stratigraphic layer, which clearly indicates that both smelting and smithing processes had occurred either within or in the vicinity of the STA site at the same time. The presence of prestige goods – such as the bronze bangle – together with utilitarian tools represented by fishhooks, nails and coiled wires also gives an indication of the wide variety of metal goods produced and consumed in pre-colonial Singapore. This comes as no surprise, as the general level of metallurgical knowledge of Southeast Asian smiths between 500 and 1500 CE ‘should have been quite high.’ 223 Just like stoneware and porcelain ceramics, some of the metal artefacts could possibly be fragments of imported cast-iron objects or wrought-iron ingots from China as well. Too corroded and fragmented to be clearly identified, these metal artefacts may have well been fragments of an assorted variety of objects including cauldrons (鐵鼎 Tiě Dǐng) and even needles (鐵針 Tiě Zhēn). 224 According to Heng, China was the major exporter of these iron products as well as a limited amount of copper- and bronze-based objects to Southeast Asia in the pre-colonial period. 225 Importing from China makes sense, as Singapore Island is completely devoid of metal ores. Although structurally part of the Malay Peninsula, the island does not share the deposits of natural resources such as tin, iron and gold found on the peninsula. 226 On the other hand, the level of exploitation and export of these metal resources remained at a relatively low level until after 1500. Even so, most of the metals consumed by Southeast Asians then were imported from extra-regional trade with Europe, India, China and Japan. 227 The Santubong river delta site at Sarawak, a major contemporaneous iron-mining settlement site, was also proposed by Shah Alam as another possible source of iron ore for precolonial Singapore. 228 223 Bronson, “Patterns in the Early Southeast Asian Metals Trade,” p. 97. Wheatley, “Geographical Notes on some Commodities involved in Sung Maritime trade,” p. 117. 225 Heng, Sino-Malay Trade and Diplomacy, p. 151-67. 226 See Lee Kim Woon & Zhou Yingxin. Geology of Singapore. Singapore: Defence Science and Technology Agency in collaboration with Nanyang Technological University, Building and Construction Authority, 2009; Charles S. Hutchison and Denis N. K. Tan. Geology of Peninsular Malaysia. Kuala Lumpur: University of Malaya and Geological Society of Malaysia, 2009. 227 Bronson, “Patterns in the Early Southeast Asian Metals Trade,” p. 86-104. 228 Shah Alam, Metal Finds and Metal-working, p. 38.. 224 201 Base-metals like iron and copper were not just imported as commodities, but likely served as media of exchange in pre-colonial Singapore as well. Commercial trade between China and the Malay Peninsula was primarily conducted through barter during the 13th and 14th centuries. 229 Metal alloys were adopted as the primary means of exchange in the Malay Archipelago. 230 Iron ingots were specifically used by Chinese traders during the Song dynasty as a bartering article in Srivijaya and Fo-lo-an on the isthmian tract of the Malay Peninsula. 231 Yet the recovery of six metal coins suggests that Chinese ceramics and base-metal objects were not the only media of exchange in pre-colonial Singapore. These coins however were not native-stuck gold or silver sandalwood flower coins circulated at various parts of Sumatra and the Malay Peninsula prior to the 13th century. 232 Instead, five coins are undoubtedly Chinese copper cash coins, as characterized by their circular shape, a square hole in the middle of the coins and with average external diameters of 2-3cm. The presence of Chinese characters, typically depicting naming the reign title of a specific emperor and on one side of the coin, appears only on three coins, whereas fragmentation and corrosion have eroded these characters from the remaining two coins. However, only two of the three Chinese coins have a legible script as that of the third coin was too worn to be read. While the third coin had all four, albeit illegible, reign title characters, unfortunately only two of four reign title characters are present in each of the first two coins due to their fragmentary nature. The first coin has characters written in ‘regular’ script (‘楷书’ Kăi Shū or ‘真书’ Zhēn Shū), which are arranged clockwise as ‘明道 - -’ (Míng Dào - -), whereas the second has characters written in ‘clerk’ or ‘official’ script (‘隶书’ Lì Shū), arranged clockwise as ‘崇 - - 寶’ (Chóng - - Băo). No other characters were found on the reverse side of both coins. Since the use of reign titles on Chinese coins only began in the Tang Dynasty (618-907), it is therefore fairly certain that these characters represent reign titles spanning from the 7th century to the collapse of the last 229 Wicks, Money, Markets, and Trade in Early Southeast Asia, p. 238. Heng, Sino-Malay Trade and Diplomacy, p. 163. Wheatley, “Geographical Notes on some Commodities involved in Sung Maritime trade,” p. 117. 232 Wicks, Money, Markets, and Trade in Early Southeast Asia, p. 219-35. 230 231 202 imperial Chinese dynasty in the 20th century. Fortunately, the first character of both coins and their respective scripts match two respective reign titles within this period of 1300 years. As such, the complete set of characters of the first coin should read clockwise as ‘明道元宝’ (Míng Dào Yuán Băo), the currency of the fourth Northern Song Emperor ‘仁宗’ (Rén Zōng) minted during the ‘明道’ (Míng Dào) years between 1032 and 1033. On the other hand, the complete set of characters of the second coin could be read, in a top-bottom-right-left manner, as ‘崇宁 通宝’ (Chóng Níng Tōng Băo), ‘崇宁元宝’ (Chóng Níng Yuán Băo), or ‘崇宁重宝’ (Chóng Níng Zhòng Băo). Regardless, all three possibilities represent currency denominations minted from 1102-1106 during the ‘崇宁’ (Chóng Níng) years of the eighth Northern Song Emperor ‘徽宗’ (Huī Zōng). 233 It should be noted also that two Tang dynasty coins, 90 Northern Song dynasty (960-1127 CE) coins and six Southern Song dynasty (1127-1279 CE) coins were also found within the pre-colonial Singapore layer at the PHC site. 234 At FTC, 16 coins dating again to the period between the Tang and Southern Song Dynasties were also found. 235 It is therefore safe to assume that the remaining three unidentifiable Chinese coins from the STA assemblage date to this period as well. The last coin, found in the SW quadrant, weighs 1g and measures 1cm in external diameter and is therefore very small in size (1Pc/g). It bears a motif on one side similar to that of a Sri Lankan coin – minted during the reign of Bhuvanaika Bahu I (1273-1302) – which was recovered from the PHC site. Although the motif is still visible, it is unfortunately too worn to be clearly identified. However, the similarity in motifs as well as the context and matrix within which both coins were found justifies at least the provisional ascription of a ‘medieval’ (circa 13th to 14th centuries) South Asian provenance to the coin. Moreover, a survey of the numismatic history of Sri Lanka reveals that coins with similar motifs were only first stuck 孫仲匯 (Sūn Zhòng Huì), 《中國古錢》 (Zhōng Guó Gǔ Qián or Ancient Chinese Coins) (台北 : 藝術圖書公司, 1994), p. 623; 劉巨成 (Liú Jù Chéng), 《中國古錢譜》 (Zhōng Guó Gǔ Qián Pǔ or Illustrative Plates of Chinese Ancient Coins) (北京: 文 物出版社, 1989), p. 183, 212. 234 Shah Alam, Metal Finds and Metal-working, p. 34; Low, “Singapore from the 14th to 19th Century,” p. 31-4. 235 Miksic, “14th-Century Singapore,” p. 49-50. 233 203 during the 7th and 8th centuries as gold coinage, later in silver, copper and silvered copper varieties, known as ‘Kahavanu.’ Kahavanu-based coinage continued to be minted by the foreign Chola monarchs from 1010-1070 and successive Sinhalese rulers until the 15th century, where new forms of coinage called ‘larins’ and ‘fanams’ replaced the Kahavanu. 236 These dates generally coincide with those of the Chinese coins found across pre-colonial Singapore. The small size of the STA coin also suggests that it was probably the smallest denomination of Kahavanu-based coinage called the ‘Massa.’ 237 Other than at the PHC and STA sites, no other Sri Lankan or South Asian coin has ever been found in any other pre-colonial Singaporean site. Besides the crafts of pottery-making and metal-working, the presence of stone artefacts gives evidence to stonemasonry in pre-colonial Singapore. One should not forget also the neat and ordered inscription engraved on the Singapore Stone supposedly split by artificial means, the old wall as well as the holed sand-stone pedestal blocks observed by Crawfurd in 1822. Hindu-Buddhist architecture – primarily stone monuments and temples built according to that stylistic and cultural tradition – were already being built in island Southeast Asia from the first millennium up to the 13th century. 238 It is therefore possible to imagine the existence of stonemasonry as an established and well-refined craft in pre-colonial Singapore as well. Evidence of stonemasonry is no better illustrated by the presence of the ornately carved stone fragments found in the NW quadrant. While the precise function of these stone artefacts may never be known, it is possible that some of these stone fragments were used as structural or building materials and ornamentation which adorned them. Given that the STA site sits on a beach, as attested by the sandy matrix of the pre-colonial stratigraphic layer in which they were found, it is unlikely that the remaining stone pieces had occurred naturally in the environment. Although lacking in ornate carvings, the smoothened surfaces and angular edges of these stones suggest that they are more likely shaped from artificial than natural processes. 236 G. P. S. H. de Silva, History of Coins and Currency in Sri Lanka (Colombo: Central Bank of Sri Lanka), p. 35-8, 53-69. Ibid., p. 36. Daigoro Chihara, Hindu-Buddhist Architecture in Southeast Asia (Leiden: E.J. Brill, 1996), p. 83-8, 210-21; Louis Frédéric, The Temples and Sculpture of Southeast Asia (London: Thames and Hudson, 1965), p. 11-6, 143-5. 237 238 204 Some of the worked stones may also have been used as tools in the metal-working industry. One stone artefact with a crucible-like rectangular depression may have functioned as a crucible or mould for metal-working, given that the pre-colonial settlement has already been shown by Shah Alam to have had a thriving metal-working sector located at the PHC site. It should be noted that the majority of metal slag among the STA artefact samples were found concentrated in the SE quadrant where the crucible-like stone was found as well, although the two categories of artefacts were found in separate excavation units a distance away from each other. It is also noteworthy that no other crucibles of any make has been found in any other archaeological site despite Shah Alam’s metal-working sector at the PHC site and Miksic’s ‘craftsmen’s quarter’ specializing in glass-recycling at FTC. 239 239 Shah Alam, Metal Finds and Metal-working, p. 36; Miksic, “Beyond the Grave,” p. 51, 55-6; Miksic, “14th-Century Singapore,” p. 51-2. 205 APPENDIX B. STA SOIL STRATIGRAPHIC PROFILES: SAMPLE EXCAVATION UNITS This section contains the soil stratigraphic profiles for all 20 excavation units at the STA site sampled for this thesis. All illustrations and information presented here are faithfully transcribed from original drawings and records containted within the site’s ‘Pit Excavation Record: Layer/Level’ forms. 240 These forms were compiled by the excavators of each STA unit and document their observations by stratigraphic layer as they excavate their respective units. It is the intention of the present author to reproduce this data without any editing in order to preserve its integrity and accuracy for future research. 240 All illustrations contained within this section are designed by Carrick Ang, Singapore. Copyright 2011. 206 NE QUADRANT 207 208 209 210 211 212 213 214 NW QUADRANT 215 216 217 218 219 220 221 222 223 224 225 226 SE QUADRANT 227 228 229 230 231 232 233 234 235 SW QUADRANT 236 237 238 239 240 241 242 243 244 245 246 247 APPENDIX C. STA RAW ARTEFACT DATA: PRE-COLONIAL SAMPLES Enclosed in this section is a DVD which contains the raw data of all sampled artefacts, including their observed characteristics, measured dimensions and plate numbers of their respective photos within Microsoft Office Excel Spreadsheets. They can be referenced to the photographic archive of the sampled artefacts contained in the DVD as well. 248