Abstract

Fourteenth-century Singapore was an important node in the networks of commerce and communication which connected polities within Southeast Asia to their contemporaries in China and India. This chapter focuses on archaeological studies of two sites in Singapore: St. Andrew’s Cathedral (STA; excavated 2003–2004) and Fort Canning Spice Gardens (FTCSG; excavated 2010). Analysis of these sites can contribute to the scholarship on ancient Singapore (1300–1600) by providing a more comprehensive understanding of the ancient Singapore urban centre and how it functioned.^[1] These two sites were chosen in order to represent the two major components of 14th-century Singapore: a palace and temple complex and a commercial/industrial/residential zone. A comparison of these two sites can provide an indication of the differentiation between subdivisions of an early Southeast Asian seaport, and thus the complexity of life in the region at that time.

Introduction

Archaeological research in Singapore began in 1984 when John Miksic directed the first systematic archaeological excavation in the country’s history on Fort Canning Hill(Miksic 1985; Miksic 2013). As of 2024, Miksic and coworkers have recorded 17 sites in Singapore, of which 11 have been excavated, while the remainder were discovered by members of the public who collected surface finds and brought them to Miksic’s attention (such as in the case of Pulau Saigon).

Artefacts dating from the 14th century have been found spread over an area of Singapore covering approximately 85 hectares (Miksic 2006, 334; Miksic 2013). Early occupation on Fort Canning Hill was divided into four zones arranged vertically. The top of the hill (called the “Forbidden Hill”, Bukit Larangan in Malay in the early 19th century) was the location of the palace of ancient Temasek (the 14th-century name for Singapore). Artefacts found there, such as ornate gold jewellery, are indicative of this function (Miksic 2013, 222; Winstedt 1928/1969). On a slightly lower terrace on the northeast slope of the hill, fragments of ancient bricks discovered in Miksic’s excavations are remnants of extensive ruins which still existed when the British arrived in 1819 (Crawford 1828 [1967]). Crawfurd’s inference that these were the remains of ancient Buddhist shrines is probably correct. On a third terrace, excavations have revealed evidence of craftsmen’s workshops where gold and glass were among the raw materials fashioned into adornment for the palace elite (Miksic 2013, 283–5). On a fourth terrace, the author and Miksic uncovered remnants of an ancient footpath associated with more 14th-century artefacts. This chapter will discuss these discoveries.

Other sites (St. Andrew’s Cathedral [STA], Empress Place [EMP], Parliament House Complex [PHC], and Old Parliament House [OPH] etc.) along the Singapore River and near the former seashore have yielded evidence of a variety of activities ranging from warehouses/dockyards to marketplaces and industrial work. For instance, at PHC, metal slag has been found together with ancient Chinese coins and other metal implements such as fishhooks, wire, a Chinese padlock, and other bronze and copper objects (Miksic 2006, 339). Iron fragments, probably from a cauldron, were also found (Miksic 2006, 340). While the majority of the sites were excavated between the 1980s and early 2000s, analysis of the artefacts has continued over the past four decades, albeit at a slow rate due to the sheer volume of artefacts excavated from all 11 sites. STA alone yielded approximately 400,000-500,000 objects, which weigh approximately one ton.

This chapter focuses on a comparative analysis of the artefacts excavated from STA and Fort Canning Spice Gardens. The data can be used to test Miksic’s argument that 14th-century Singapore was not a simple fishing village; rather it was a complex urban centre with a ceremonial and palatial zone with accompanying crafts clusters located on Fort Canning Hill which was connected to the market, industrial, habitation, and port facilities in the rectangular area bounded by the Singapore River on the southwest, the ocean on the southeast, an earthen rampart at northeast, and the hill known as Fort Canning on the northwest (Miksic 2006, 335). Recent studies (Kwa et al 2019; Lim 2012; Heng 2009, Kwa et al 2009) have not challenged Miksic’s argument regarding the urban character of 14th-century Singapore since archaeological research began in 1984 (Miksic 1985; Miksic 2004 when he first discussed Singapore as a port-of-call in the 14th century; Miksic 2006 where he discusses the relationship between the seven sites then excavated within the context of the Southeast Asian region; Miksic 2013 where more sites and more intensive analysis was applied to the sites, their finds, and Singapore’s place in the region and globally in the context of the Southeast Asian Maritime Interaction Sphere [SAMIS]). Discussion of the excavation and preliminary analysis of the artefacts from these two sites pave the way for the construction of hypotheses regarding the functions of these two sites and the relationship between the ceremonial centre on one hand and the urban area on the other. This analysis does not challenge the overall view of Singapore’s early history, but rather seeks to provide more precise comparisons of the activities conducted at two sites within the site complex occupied by the 14th-century port.

St. Andrew’s Cathedral and Fort Canning Spice Gardens Sites

The two sites covered in this study were excavated in the first decade of the 2000s. Analysis of the assemblages of tens of thousands of artefacts recovered from the two sites has now reached a level of fine-grained analysis which allows us to compare the sites and their artefacts using quantitative as well as qualitative means, and to come closer to a definitive understanding of the use of space in Singapore in the 14th century.

The St. Andrew’s Cathedral (STA) site was excavated in 2003-2004, and the Fort Canning Spice Gardens (FTCSG) in 2010. These two sites were selected for comparison on the basis of their locations:

• The FTCSG site is located in an area of Fort Canning Hill, which formed part of the periphery of the 14th-century palace zone. It was close to the site of the ancient palace garden (Miksic 2013, 12), where the archaeological dig site (Miksic’s FTC site, now called Artisan’s Garden) is located.
• The STA site lies at the southeastern corner of the main zone of habitation, commerce, and industry of Singapore between 1300 and 1600.

Sites on Fort Canning Hill fell out of use by the end of the 14th century (see Miksic 1985, 2004, 2006, 2013). Sites in the area between the Singapore River and the earthen rampart, which ran along what is now Stamford Road (see map), indicate that activities continued there from the 14th century to approximately 1600.

Comparison of the artefacts from the FTCSG and STA supports the assessment that the sites on the hill fell in disuse following the departure of the ruler’s family and retinue to Melaka at the end of the 14th century, but the plains of Singapore continued to function as activity zones where artefacts support the dating of these sites to the period between 1300 and 1600. The STA site was discussed briefly in Miksic 2013, but much more analysis of the finds has now been conducted. This publication is the first to present the results of the excavation at FTCSG.

Function of these Sites: Preliminary Postulations

The excavation at St. Andrew’s Cathedral (hereafter STA) took place over a period of seven months from September 2003 through April 2004; preliminary survey was conducted in March 2003 (Miksic et al, unpublished progress report). The soil at the site was mainly sand, which did not display clear strata. The ground at the site was probably swampy and undulating in the 14th century (Miksic 2013, 341-343).

STA might have been located on the fringe of the settlement whose inhabitants occasionally disposed of rubbish there, but some extraordinary objects such as carnelian beads, intact jars, and “a carved stone with a human head at one tip, a point at the other” were also found there (Miksic 2013, 343). The carved stone peg resembles items recovered from Kota Cina, Sumatra, and in the area of Kedah and the Isthmus of Kra on the Malay Peninsula, but is the only one of its type found in Singapore. Too small to be a hairpin, another hypothesis is that the stone artefact might have served as a touchstone for determining the purity of gold. This hypothesis is consistent with the recovery of two other valuable items at STA: a piece of gold and a simple gold ring. The anthropomorphic features of the stone suggest that the object was not an item of mass production which required skill to produce, and had symbolic value which is still mysterious (see section below discussing unique artefacts).

Other high-value items comprise artefacts made of glass, porcelain, and metal, such as Chinese coins. Glass items include fragments of an emerald green bangle, an indigo blue bottle with inlaid white decoration, a bangle with alternate white and black design, possibly of Indian origin, and beads of glass carnelian stone. Sherds of high-quality porcelain include rare pieces of a special bottle called yu hu chun in Chinese decorated with red copper paint, another with a white qingbai jarlet with decoration made with iron spots, covered boxes possibly for cosmetics made of special white porcelain made in Dehua, delicate bluish-white qingbai bowls and jarlets, and a wide range of white Chinese porcelain decorated with cobalt blue paint, ranging from Yuan dynasty bowls and cups to wares from the early through middle and late Ming periods. These fragments of high-quality ceramics were found jumbled up with ordinary Chinese stoneware, pottery from other parts of Southeast Asia, and everyday earthenware made in Singapore.

Since analysis of the assemblage of the site is ongoing and will take years to complete, any characterisations of the possible function of the site are hypothetical and cannot be taken to be conclusive. In addition to the standard cataloguing of artefacts, some students have written theses on selections of artefacts in the STA assemblage. Lim Tse Siang (2012) argues in his thesis that the site should be considered a habitation area with different parts of the site corresponding to residences of communities of differentiated socio-economic classes. While Lim’s hypothesis that parts of the site may have been early Singaporean habitation areas is deserving of testing, Lim’s conclusion is flawed due to his faulty usage of the *chi-*square mode of analysis and problematic hypothesis that certain types of artefacts can be assumed to have been deposited there by people of specific social classes and ethnicities.

Lim’s sample was too small to be statistically valid. Lim argues that the proportion of earthenware compared to porcelain was indicative of the presence of people of a lower socio-economic class. This inference was based on incomplete and problematic sampling techniques. It is erroneous to assume that the presence of earthenware reflects sites where people of a lower socio-economic group lived. Other hypotheses should have been considered. For instance, earthenware vessels used as water containers and cooking vessels might have been used by a wider group of individuals and groups. Lim’s biased assumption is predicated on the assumption that people of different ethnicities would have lived in separate parts of the site. There is as yet no evidence that this was the practice in ancient Singapore. Miksic (2000) warns against ascribing ethnicity to ceramics. It is not admissible to assume that only ethnic Chinese used Chinese porcelain and stoneware, whereas Malay populations only used locally-made or regional pottery.

Spencer-Wood (1987, 4) discovered in her analysis of archaeological data from 18th- and 19th-century American households that “wealthy house residents bought and discarded not only expensive goods but also more inexpensive goods than less wealthy households”. While it would be difficult to discuss socio-economic differences among the population of 14th- to 17th-century Singapore due to the dearth of documents on groups and individuals and their possessions, it is important to note that inexpensive, and by extension, lower-quality products were probably also used by members of higher socio-economic classes. These artefacts may comprise items of daily use, including earthenware and stoneware vessels used in food preparation and storage. Excavated items would probably have belonged to multiple individuals and/or groups within complex social units in areas where rubbish was discarded.

The Fort Canning Spice Garden site was excavated in April and May 2010. The site comes under the purview of the statutory board, National Parks Board (NParks). NParks contacted Miksic and Goh to determine whether a location where a wheelchair access ramp was to be built should receive impact assessment work before the ramp was constructed. A watching brief of two test trenches dug at the location showed the presence of a 14th-century layer, whereupon Miksic and Goh proposed an archaeological salvage excavation to recover artefacts as well as to determine, as far as possible, the probable function of the site.

An online report on this excavation and an open-access database are available on the National University Press EPress website: Southeast Asian Archaeological Site Reports (see The Fort Canning Spice Garden Site Report, 2010: https://epress.nus.edu.sg/sitereports/spicegarden/).

A preliminary site report and database of the STA excavation and artefacts are also available on the same website (see The St. Andrew’s (STA) Preliminary Site Report, 2003-2004 https://epress.nus.edu.sg/sitereports/sta/)

Analysis of the distribution of artefacts at the FTCSG site shows that in ancient times this location was part of a path which would have been used by people climbing the hill to reach the area of the archaeological dig site (excavated since 1984), which is now called Artisan’s Garden in 2019. The centre of the path was kept clear of broken pottery, but a ditch parallel to the path was full of sherds.

Some Discussion of Unique Artefacts

The artefacts discussed here represent only a selection of the unique objects excavated from the two sites. It would be preferable to compare the quantities of the different types of artefacts, but that is not yet feasible. The first item of note is a small stone peg measuring about 2cm in length with an anthropomorphic face. It is an enigmatic object with its small stature and intricate designs. Similar items have been recovered from Kota Cina, Sumatra, Indonesia, Kedah, Malaysia, and south Thailand. The other pegs bear natural and other motifs, none of which bears a human image. Although shaped like a peg, the STA “peg-man” is too small to be a peg in dressing one’s hair, nor is it a game piece, in view of its rounded tip. This peg-like object is possibly a touchstone used by goldsmiths in determining the purity of gold. The rounded tip would have been used to rub against gold to determine its purity. Physico-chemical research is needed to test this hypothesis.

A gold wire recovered from STA confirms the presence of gold artefacts there. Gold items were also recovered from other Singapore sites, including Fort Canning.

The STA site also yielded coins, of which a rare example is a Sri Lankan coin dating to the 13th century. Chinese coins were also recovered from the site, including semi-melted coins. The presence of these fused coins attests to the melting of coins to yield molten metal, which could have been used for making metal objects, such as fishhooks.

Other unique objects include glass items such as an Indian bangle with inlaid glass of multiple colours and a Chinese green bangle, as well as a broken fragment of an indigo blue bulbous vial with inlaid white decoration.

Another type of artefact comprises beads of different materials. The first is of probable Indian origin and consists of carnelian stone beads. The second type comprises glass beads of different colours, such as yellow, blue, red, and white, from China (see Borell 2010). The glass beads bear evidence of a Chinese method of production which consists of winding molten glass along a wire to form spirals which were later cut into individual beads.
The glass beads and shards of inlaid glass formerly of jarlets or vases from FTCSG are similar to examples recovered from the Fort Canning (FTC) and STA sites. While these FTCSG artefacts may be similar to items at FTC and STA, two examples of unique artefacts from the FTCSG site comprise a dressed or worked stone, possibly from a statue and a large slab of 14th-century brick.

The Two Sites and Singapura between 1300 and 1600

How do the two sites fit together? FTCSG and STA represent two locations with the Singapore site complex connecting two subzones: the former as part of the palatial area and the latter as part of the area where the main population resided. FTCSG represents a link between the main settlement on the plain and the elite subzone on Fort Canning Hill. As the footpath leading to a workshop, possibly one of several on the lower terraces of Fort Canning Hill (then the forbidden hill or royal hill of 14th-century Singapore), it would have provided a route via which the artisans would have made their way up to the workshop.

This study suggests that the patterns of distribution and presence/absence of archaeological materials at each site are indicative of the “patterns of participation in consumer behaviours of the cultural subgroups” (Spencer-Wood 1987, 1) and serve as an indicator of the differences spatially and temporally. The Fort Canning Spice Gardens site was used for a shorter span of time and formed a marginal part of the palace complex, whereas the St. Andrew’s Cathedral site was located on the interstitial bounded areas of what would have been the residential and market areas of 14th-early 17th-century Singapore. It is important to note here that the “cultural subgroups” here do not refer to ethnic groups in the Singapore context. As Spencer-Wood argues, one cannot assume that patterns of consumption necessarily suggest social stratification and ethnicity; other variables include access to products via marketplaces, functions of the sites in question, life cycles of the sites, and the completeness of the data available. The last point here is important as unlike more recent periods when local data can be compared with data from other sites within the region and globally, the completeness of our data is important as it can be directly correlated with level of confidence.

Cultural groups and sub-cultural groups hence refer to groups and sub-groups within an archaeological culture. What is an archaeological culture? As a key concept, various notable archaeologists have defined the term. Willey and Phillips (1958: 16-17) in response to the criticisms they received from Spaulding revised their definition of archaeological culture to that of “an archaeological unit as a provisionally defined segment of the total continuum, whose ultimate validation will depend on the degree to which its internal spatial and temporal dimensions can be shown to coincide with significant variations in the nature and rate of cultural change in that continuum”. Willey and Phillips (1958, 15) had initially characterised “archaeological culture as an arbitrarily defined segment of the total continuum”. Lewis Binford and David Clarke adapted the framework of systems theory in their development of a new approach to doing archaeology, commonly referred to as New Archaeology (processual archaeology in the US and analytical archaeology in the UK). Neither scholar explicitly referred to archaeological culture. Binford’s definition appropriates Leslie White’s (1959, 8) characterisation of culture as an “extrasomatic adaptive system,” which cannot be “reduced to normative ideas” (Binford 1965, 209) in which “people, things, and places are components in a field that consists of environmental and sociocultural subsystems, and the locus of cultural process is in the dynamic articulations of these subsystems” (Binford 1965, 205). Clarke (1978, 86) wrote that “Culture is an information system, wherein the messages are accumulated survival information plus miscellaneous and random noise peculiar to each system and its past trajectory”. The culture system comprises “relationships explicitly or tacitly inferred by their repeated patterns of activities, artefacts, and beliefs” (Clarke 1978, 86).

Contrary to the post-processual criticism of processual archaeology as paying little heed to the agency of groups and individuals and not incorporating a multitude of perspectives, both Binford and Clarke highlighted its dynamic and interactive nature in their definitions of archaeological culture, wherein the participation of human actors was emphasised. In post-processual archaeologists’ mission to illuminate the “structures of power and structures of meaning, the role of agency, the central importance of history, and epistemological questionings concerning the relationship between subjects and objects” (Hodder 2014, 6083), they reject New Archaeology’s aim to demonstrate archaeology’s potential to produce generalisable statements. In likening New Archaeology’s approach to a science, Binford, Clarke, and their peers were committed to undertaking a systematic method for conducting archaeological studies. Post-processualism adopts the cultural or literary turn inspired by “Bourdieu and Giddens (2014, 6084) who were the first to develop an “interpretive archaeology” (Hodder et al 1995) in the 1990s which gave way to variety of influences from phenomenology, postcolonial studies, studies of materiality of objects, indigenous studies, gender, feminism, and various approaches in the 2000s. Post-processual archaeology does not provide any particular position on its interpretation of archaeological culture, or culture. Shanks (2009: 134), for instance, suggests that “society was seen not as an extrasomatic means of adaptation (the premise taken up in processual archaeology), but a communicative medium”, making a case for a more “reflexive” and interpretative form of archaeology.

Appropriating the processual archaeological view of culture, the archaeological culture of Singapore comprises the complex networks of multiple sites, which interacted with one another between the 1300s and late 1600s and then again from the 1800s, and is represented by the distribution and composition of artefacts across the sites and compared with other external sites. What this chapter provides is two snapshots of two sites within this larger internal network within an expanded international network. This discussion gives a perspective from within a single site complex (in this case, ancient Singapore) comprising several local sub-nodes, and by discussing the relationships between these sub-nodes, one could extrapolate and expand these facets of relationships to produce a hypothetical articulation of the 14th to 17th-century Singapore node which could then be examined within a regional and international scope. It demonstrates the importance of examining the singular nodes in conjunction with other nodes as well as at the grander scale of connected networks. This emphasises the need to determine the character and relationships between the sub-nodes (FTCSG and STA) in providing a more accurate description of the nature, functions, and capabilities of Singapore as a node. The linkages are essential in providing an image of the patterns and flows of the archaeological culture of 14th-–16th-century Singapore.

Singapore: the hyper-node in the Southeast Asian Maritime Interaction Sphere (SAMIS)

The reference to Singapore as a node of a regional and global network of trade points to the importance of evaluating the usefulness of the network analysis model. Network analysis has a long history dating to the 1970s (Brughmans 2013, 624); Collar et al (2015) suggest that network analysis became more prevalent in the 2010s. Network analysis is predicated on the work which sits in studies of “biological and cultural worlds that people inhabit” that “are organized into networks of nodes…and the relational ties or edges that connect them (Newman 2010)” (Collar et al 2015, 2). All the variant approaches “focus on relationships between entities and on the patterns that emerge from them” (Brughmans 2013, 625).

A “network model represents the conceptual process researchers go through, explicitly or implicitly, in deciding whether the phenomena under investigation can be usefully abstracted using network concepts and represented as network data” (Collar et al 2015, 4), and trade and its patterns are perceived to be able to benefit from such analogy. Collar et al (2015, 4) refers to “social entities” as nodes and the “connections that allow for, or arise out of, the flow of commercial information and goods between them can be represented as edges, linking the nodes together”. Two types of networks are presented by Brandes et al (2013, 10; cited in Collar et al 2015. 4): binary and valued, both emphasising how one edge can affect any other edges in the same network. The relationships within each network are important in determining its “structure” (Collar et al 2015, 5) and represent a key concept of network analysis (Collar et al 2015, 6).

As Collar et al (2015, 10) suggest, “sites or assemblages of material culture form natural nodes” of which archaeological studies can focus on “the dynamics by which they come to be characterised the way they are”. The different studies presented in Collar et al’s edited volume present ways in which network analysis has been used at the site, intra-site and inter-site and intra-regional levels. For instance, Moli et al use “ego networks” to examine intra-site assemblages to “produce a ‘thick’ network picture of multiple different interactions within and between groups in the Caribbean” (Collar et al 2015, 12). These provide different ways in which network analysis can be used to inform the study of 14th-century Singapore and its intra-site connections and its role in the trade and communications networks within and beyond the region.

A two-mode and multi-scalar analysis allows the researcher to examine intrasite relationships between FTCSG and STA, and inter-regional relationships between Singapore and other nodes within the larger Asian trade network. This “two-mode” analysis can provide “visualizations incorporating sites, objects, and contexts as nodes” (Collar et al 2015, 13). Brughmans (2013, 627) describes a “two-mode” network as “an affiliation network”, and these networks are connected via relationships with attributes of “ties”, where a tie can refer to the “number of pottery sherds found on a site” (see Brughmans 2010). An approach focused on networks of consumption, examining materials utilised and disposed of within the node and its sites, is useful for studying early Singapore. For the 1300s through 1600s, the materials consisted predominantly of ceramics, and analysis of the usage and discard patterns of these can provide insight into the economic and socio-cultural structure of early Singapore. This specific examination of two sites representing two different periods in Singapore’s fluorescence represents an attempt to evaluate the usefulness of using network analysis for understanding Singapore between the 14th century and the end of the 17th century. As Brughmans (2013, 16) states, “network methods are part of an archaeological research process, not a replacement of it”, and the use of network models makes it possible to examine networks, such as trade and communication networks, in a dynamic manner.

Knappett (2011) advocates the usefulness of the two-mode network analysis for discussing the patterns of relationship and nature of nodes at the site complex (for example, Singapore between 1300 and 1699) and individual site levels (for example, FTCSG and STA in this chapter). The significance of the nodes is determined by their connectivity (number and limitations of these links) and accessibility (distance between the nodes and other access points). Actors and their actions are “viewed as interdependent rather than independent autonomous units” (Brughmans 2013, 632). The patterns of relationships between actors and the actions of these actors reveal aspects of the structure of the society or nodes.

Two-mode network of Singapore, 1300s–1600

The FTCSG and STA sites can be used to provide a discussion of the first-level modal study of a network within the site complex of Singapore between 1300 and 1600. The second-tier mode will require a separate study with data from other nodes in Singapore’s regional network, which cannot be achieved within the scope of this chapter, but a brief discussion using existing scholarship could show the potential for such a study.

Two-mode network, described as an affiliation network (Brughmans 2013, 627), comprises relationships established through ties, such as the ceramics excavated from the two sites and their attributes, presented in terms of their types, number of sherds, and weights of different ceramic types and their varieties.

Only two per cent of the ceramics of the 14th century through 1600 have been processed (sorted, labelled, and data entered). The data in the pie chart is skewed toward earthenware and porcelain due to these being the subjects of three final-year projects and two Master’s theses by students at Nanyang Technological University Singapore. For this reason, the percentages of earthenware and porcelain will be used as points of comparison with the more complete analysis of the Fort Canning Spice Garden (FTCSG) materials.

For this study, percentages will be used as reference points for “ties” rather than the number of sherds. A comparison of the proportions between the two sites shows that the ties (percentages of types of ceramics) and attributes (varieties of ceramic types, such as tempered versus fine paste earthenware, for instance) are consistent. These also fit the general descriptions of other sites referred to in Miksic (2013).

Based on a limited comparison of the earthenware and porcelain from FTCSG and STA sites, the proportions of tempered to fine-paste earthenware and green porcelain to white porcelain show a limited difference between the two sites. Within the category of earthenware, tempered earthenware comprises the larger number, while among porcelain, green porcelain was available in greater quantities than white and blue and white porcelain. The proportions of tempered earthenware to fine-paste earthenware are similar in both sites. The two sites show a close affiliation to each other synchronically within the 14th century and over the next two centuries until 1600. The similar distribution of ceramics across the sites projects an image of continuity.

A detailed analysis of the proportions within the porcelain category between the two sites demonstrates that the differences: 33 per cent to 29 per cent to 6 per cent (green to white to blue and white porcelain) for the STA site and 15 per cent to 7 per cent to 0.002 per cent (green to white to blue and white porcelain) for the FTCSG site could be explained hypothetically. Due to the longer period of occupation at the STA site as well as the greater number and diversity of population in the area near the Singapore River, there was greater access to and availability of the range of porcelain. That green porcelain represents the largest percentage of porcelain in Singapore sites is not unusual (see Miksic 2013; Miksic 2006). Green porcelain continued to be produced following the Yuan dynasty into the Ming (from the 14th century through 1600) period, possibly accounting for the large quantities of green porcelain available in ancient Singapore. A point of interest regarding the proportions of green to white porcelain at the STA site pertains to the relatively large quantities (in sherds) of white to green porcelain. Green porcelain sherds are four per cent more than white porcelain, where usually there would have been a greater differential between the two types of porcelain.

The next two diagrams show the proportions of the different types and varieties of ceramics according to weight. When compared to the above diagrams based on the number of sherds, one would note that stoneware weighs more than porcelain and earthenware. This demonstrates the point that in utilising network analysis, the variables used are important and could have important implications for the results.

The above comparison of the two sites represents a limited use of the two-mode network analysis, which would benefit from a bigger sample of sites and their artefacts. Another point of comparison which would be advantageous consists of complete sets of data from all the Singapore sites.

Networks and Ceramics within an Interaction Sphere

In 2017, Miksic and Goh discussed the use of the “interaction sphere” concept for examining networks of exchange in Southeast Asia between the 9th and 16th centuries. They appropriate the term from archaeological studies which focused on the central United States, where “the term can be applied to the analysis of the distribution of a special type (or types) of artifact(s) made of similar material and displaying a common style over a large area” (Miksic and Goh 2017, 810). For this study, they identified six interaction spheres of connectivity within a hierarchy which cuts across space and time. These comprise the different types of ceramics from the most widespread, earthenware, through fine paste earthenware, Khmer stoneware, glazed ceramics of the 15th century of Myanmar, Thailand and Vietnam; Martaban stoneware jars, and Chinese porcelain.

The interaction sphere represents a useful analogy for discussing the roles of the nodes and the patterns of interaction between the nodes within the networks of exchange and trade. One way of imagining the relationship between the “interaction sphere(s)” and networks is to see the interaction sphere as consisting of “ties” with similar attributes, which can be analysed. The networks discussed in this chapter operate within spheres, with ceramics as artefacts or signals which ping to reveal flows, patterns and scope of distribution. The reference to “ties” allows for the measuring of significance in terms of quantity between different sites/sub-nodes and at the regional level between nodes.

Connectivity, Interaction Spheres and Networks

Another trending concept, especially in the circuits of ASEAN studies and community, is connectivity. It is an important feature to examine in studying trade, cultural interaction, and communication as it reflects the need to look at the abstract and intangible pulses of flows which can be emitted at different levels of intensity and in different directions: uni-, bi-, and multi-directions. As Miksic and Goh (2017, 827) discuss in their chapter, the analysis of “the distribution of Chinese ceramics, and its relationship with other, smaller spheres of ceramic exchange” presents “a dynamic system of early connectivity on various levels over a great span of time and space”.

Future directions

A study conducted by Mills et al (2013) amalgamated “social network analysis with geographic information systems approaches to reconstruct network dynamics” across a period of 250 years in order to determine the role played by migration and long-distance networks. This study examines the pre-Hispanic US Southwest between 1200 and 1450 CE and utilises a “comprehensive artefact database”. The site reports and databases produced for Singapore by Miksic and the current author represent an example of a similar approach to use of comprehensive data from the Singapore sites to test and analyse our hypotheses regarding the networks of trade and connections within Singapore and between Singapore and other contemporary polities and urban centres in the 1300 to 1600 period. The longer-term aim is to determine how the data can be used to assess the nature of the networks of which 14th-century Singapore represented an important node and how the networks of interaction changed over time. Mills et al’s project database comprises “4.3 million ceramic artifacts from more than 700 archaeological sites” and obsidian artefacts from the western Southwest (Mills et al 2013, 5785). Network analysis provides an interesting approach to examining sites such as Singapore as it is “the relationships among nodes (e.g., individuals, households, settlements), rather than the nodal attributes traditionally studied by archaeologists such as status, function, or size” (Mills et al 2013, 5785).

Bibliography

Footnotes

1. This study is made possible by funding generously provided to the author by a Tier 1 grant, RG68/17 When Pots Whisper: Accounts of the Places, the Purchasers, and the Producers of Ceramics in Medieval Myanmar and Singapore provided by the Singapore Ministry of Education and a Heritage Research Grant from the National Heritage Board of Singapore, Research and Database on Fort Canning Spice Gardens (FTCSG) and St. Andrew’s Cathedral (STA) Sites. ↩︎