Article

Pottery

From Munshipedia, the MBRAS digital historical encyclopedia

Pottery

Pottery

Pottery constitutes one of the most persistent and archaeologically significant material cultures of the Malay Peninsula and Borneo, spanning from Neolithic cave deposits dating to approximately 4000 B.C. through to modern local production. The ceramic record in this region encompasses a remarkable diversity of forms and functions: hollow stands and perforated cones from northern Malayan caves, double-spouted earthenware vessels from the Niah Caves in Sarawak, Chinese export stonewares recovered from Sumatran trade sites, and locally made spouted water-vessels still in use in the early twentieth century. Together, these assemblages illuminate questions of indigenous technological development, long-distance maritime trade, ritual practice, and the complex interplay between local innovation and external influence that characterises the prehistory and early history of the region.

Neolithic Pottery of the Malay Peninsula

The earliest well-documented pottery from the Malay Peninsula comes from Neolithic cave sites in the north and north-west, where hollow clay stands and perforated cones have been recovered in association with cord-impressed wares and polished stone tools. Prince John Loewenstein’s 1959 comparative study established that the hollow stands excavated from Gua Cha, Gua Musang, and Tengku Lembu belong to a wide Old World distribution originating in Mesopotamia and transmitted through India to Southeast Asia [1, pp. 168–196]. The earliest known example of this type comes from Al Ubaid in Mesopotamia at approximately 4000 B.C., with subsequent appearances at Uruk (3500 B.C.), Farah, Ur, and Kish [1, p. 177]. In the Indus Valley, ring-stands, cylindrical stands, and hour-glass-shaped stands were recovered in large numbers at all levels of Mohenjo-daro, Harappa, and Chanhu-daro (c. 2500–1500 B.C.), and at Harappa, jar-stands were found embedded in the paved floors of kitchens and larders [1, pp. 166]. Loewenstein positioned the Malayan specimens within this broader chronological and geographical framework, noting that all known examples come from the northern or north-western part of the peninsula—precisely the regions where early Indian maritime contacts would have established footholds [1, p. 187]. The largest Malayan specimen, a cylindrical stand of red ware from Gua Cha, measures 204 mm in height with a flaring rim diameter of 277 mm and an opposite opening of 133 mm; a vessel resting on a similar cylindrical stand was found in situ at that site [1, p. 168].

A related class of objects, the perforated pottery cones, has generated sustained scholarly debate regarding their function. P. D. R. Williams-Hunt discovered a series of such cones at Bukit Kaplu, Kedah, in 1951, and initially suggested they had ritual significance, possibly Buddhist, or were associated with incense burning [2, p. 189]. G. de G. Sieveking challenged this interpretation in 1956, arguing on the basis of use-wear evidence that the cones served as practical working gear in Neolithic pottery production [2, pp. 189–194]. The cones, recovered from an undisturbed cave at Kodiang alongside three polished stone axes and fragments of at least twenty-five cones, are thick, hand-made, fired in an oxidising atmosphere to a reddish-yellow colour, and show clear evidence of being rolled up from a flat sheet of clay laid on a cord mat [2, p. 189]. Two opposing perforations, made with a flat instrument approximately 10 mm broad while the clay was still wet, penetrate the body of each cone [2, p. 189]. Sieveking’s central argument rested on two observations: the cones are markedly cruder in finish and consistently under-fired compared to the contemporary earthenware vessels found with them, and none shows any impregnation with burnt incense particles, whereas known-used earthenware censors in the Perak Museum (Kelantan and Negri Sembilan specimens) display a distinct blackish internal crust of incense residue [2, pp. 192–193]. He proposed instead that the cones functioned as supports or moulds in the potter’s workshop: the heavily striated rim would grip wet clay, the pointed end could be stuck into sand, and the perforations would serve as slots for wooden pegs allowing rotation during decoration [2, p. 193].

Loewenstein’s 1959 study resolved the question in the opposite direction. Constructing a full-scale plaster model and conducting a controlled burning experiment, he demonstrated that five grams of incense placed on a hot plate beneath a reconstructed cone produced a steady flow of smoke through the upper perforation for forty minutes, consuming the entire quantity of resin [1, p. 183]. He corroborated this with a letter from B. A. V. Peacock, who visited the site in December 1956 and obtained confirmation from a Buddhist chief priest at Padang Sira that the cones were indeed used as incense burners [1, p. 185]. Loewenstein further connected these Malayan censers to the perforated “brazier” vessels of the Indus cities and to compact hearth-stands from Ashur, which share the characteristic of a hollow foot with lateral perforations designed to increase air circulation [1, pp. 179–182]. The debate between Sieveking and Loewenstein thus represents one of the earliest methodological confrontations in Malayan prehistoric studies, pitting comparative use-wear analysis against experimental reconstruction.

Loewenstein extended his diffusion argument to the Far East, citing hollow stands in Yayoi ware from southern Japan (2nd–3rd century A.D.), aeneolithic sites in southern Manchuria, and a hollow bronze stand from South Korea (c. A.D. 700) [1, p. 189]. He argued that these Far Eastern examples are linked to Austronesian migrations, which passed through the Malay Peninsula between 2500 and 1500 B.C., and that the pottery type was adopted during that sojourn [1, p. 195]. He also corrected an earlier assumption by H. D. Collings that modern Malayan rotan stands evolved from Neolithic clay prototypes, arguing instead that the plaited fibre stand is the older form and that the clay stand is a later imitation of it [1, p. 191]. A living link to the ancient type persists at Parit Buntar, Perak, where ceramic ring-stands are still being produced in limited numbers exclusively by Indian potters for ritual and ceremonial occasions [1, pp. 192–193].

Prehistoric Pottery of Borneo: The Niah Caves

The prehistoric pottery of Borneo is best documented from the Niah Caves in north-west Sarawak, where Tom Harrisson’s excavations over more than a decade produced a rich and varied assemblage of earthenware. Among the most distinctive and hitherto neglected classes are the double-spouted vessels, of which fifty examples were recovered across nine caves within the Niah limestone massif [3, p. 39]. These vessels are unknown in historic times and have not been found outside the Niah massif despite extensive search in burial caves along the Bakong and Baram rivers within fifty miles [3, p. 38].

Harrisson established a typology comprising four categories: Type I, with separated upswept angular spouts rising from a common head; Type IIA, with rounded spouts each placed separately on a domed top, unbridged; Type IIB, the same but bridged between the spouts; and Type III, with two rounded bridged spouts attached to a flat top [3, pp. 44–45]. The richest assemblages come from Jeragan (11 vessels), Kain Hitam upper (16), and the West Mouth of the Great Cave (12) [3, p. 39]. A general developmental trend runs from Type I in the purely Neolithic caves towards Types IIB and III in the later, metal-age contexts [3, pp. 56–57].

Radiocarbon dates anchor the sequence: a Jeragan coffin burial yields c. 2,250 B.C. (4,300 ± 160 years), a Jeragan jar burial c. 1,120 B.C. (3,070 ± 410 years), and a Magala burial c. 1,180 B.C. (3,130 ± 240 years); West Mouth urn contents date to 1,330–1,120 B.C. [3, pp. 69–70]. Four wooden “death-ship” dates from Kain Hitam and Gua Samti range from c. 350 B.C. to c. 915 A.D., confirming the survival of Type III vessels well into the Chinese ceramic import trade of the T’ang and Sung dynasties [3, p. 70]. Double-spouted vessels thus span approximately two millennia, representing nearly one hundred generations of potters [3, p. 71].

A remarkable negative feature of all double-spouted vessels is the complete absence of paddle-beater marking on the body, despite paddle decoration accounting for 29 per cent of all Great Cave earthenware [3, pp. 41]. Decoration, where present, is confined to the spouted top section and comprises incised or impressed circles and straight lines arranged in geometric patterns [3, pp. 59–64]. Harrisson noted that the full design effect is intended to be seen only when looking down onto the top of the pot from above, and that the arrangement of lines and circles gives the impression of a coded system or “alphabet” of arrangement, though this remains unproven [3, pp. 63–64].

The vessels are consistently associated with burial remains, particularly secondary burials, and no example shows clear signs of domestic use [3, pp. 38–39]. Harrisson proposed that the double spout is connected to the idea of the journey of the dead from this world into another, with passage through the body of the pot or through the bridge between the two spouts representing a symbolic passage of the spirit [3, p. 71]. He linked this to the hornbill’s beak and casque, perhaps the most persistent head symbol in Borneo, noting that a piece of hornbill casque was actually excavated in the West Mouth in 1965 [3, p. 71]. The Samti vessel, with its angular, nearly pyramidal bridge and two virtually identical rounded spouts, may represent a final evolution of Type III, possibly placed as a separate spouted disc divorced from any body [3, pp. 56].

Harrisson argued firmly for a fundamentally local, indigenous origin rather than external diffusion, countering the tendency of Western scholars to attribute everything fine or good to outside influence [3, p. 72]. He acknowledged parallels in adjacent lands, including a Sumatran double-spouted vessel examined in the Leiden Museum and a single bronze double spout from central Vietnam discussed by Louis Malleret, but maintained that the Niah sequence, with its clear developmental trajectory over two millennia, points to internal creation [3, pp. 72–74].

Trade Ceramics and Later Production

The ceramic record also documents the arrival of Chinese export wares in the region. E. Edwards McKinnon’s 1979 field note reports the recovery of surface finds of Chinese export ceramics at Bukit Seguntang and Talang Kikim near Palembang, Sumatra, during a June 1978 survey [4, pp. 41–47]. The Yueh-type stoneware sherds from Bukit Seguntang, displaying two distinct types of kiln-support spur-marks on their interior surfaces, are dated to the late T’ang to Five Dynasties period (c. 906–960 or slightly earlier) [4, pp. 40–41]. The small quantity of material—probably representing no more than ten or twelve vessels—led McKinnon to propose that the bowls were likely offering vessels placed at a Buddhist sanctuary on the hill, rather than evidence of dense domestic habitation [4, p. 44]. At Talang Kikim, a mixed assemblage ranging from late Sung and Yuan green-glazed porcellaneous stonewares to Ming-period blue and white wares was found alongside fragments of wafer-like bricks [4, p. 45]. These finds help to close the chronological gap between the seventh-century Srivijayan inscriptions and the well-documented fourteenth- and fifteenth-century remains in the Palembang area [4, p. 41].

In the Malay Peninsula, I. H. N. Evans documented in 1923 the persistence of a spouted clay water-vessel type, the kendi (a spouted vessel for pouring liquid), in Negri Sembilan and Perak, with modern locally-made examples held in the Perak Museum at Taiping [5, p. 248]. He compared these with specimens from the West Coast of Sumatra and Aceh, noting that the Sumatran pots most closely resemble the Negri Sembilan type while the Aceh examples are taller with a small air-hole rather than an open top [5, p. 248]. Ancient comparanda include Chinese blue-and-white and green-glazed vessels from the Salayer Islands, a Javanese Hindu-period carving of Bhrkuti holding a spouted vessel, and a stone model from the Kedoe Residency catalogued as a gendi [5, p. 249]. Evans concluded that the type most likely originated in India and reached Java, with the Chinese examples possibly resulting from Buddhist transmission or from Chinese potters producing non-Chinese forms for export trade [5, p. 249].

A separate question of local versus imported production was addressed by Sieveking in a 1956 note correcting the earlier identification of stamped earthenware from Johore Lama as Han dynasty Chinese pottery [6, p. 194]. He reassigns the material to local Malay Peninsula production of the nineteenth or twentieth century, arguing that the paste, firing, and design quality differ markedly from typical Han pottery and instead closely resemble yellowish-white paddle-marked vessels from Singgora in the Patani states [6, p. 194]. The dating was confirmed when a large rim and body fragment was recovered in 1953 from between the inner and outer skins of a Chinese twakoo (Boat B) excavated at the site, a vessel probably engaged in trade with Singapore and the Riau Archipelago some fifty to a hundred years before the excavation [6, p. 194].

A further category of coastal finds is represented by H. D. Noone’s 1941 report on pottery sherds recovered from a sea beach near Sepang, South Selangor [7, pp. 217–218]. The sherds, ranging from light buff earthenware with an impressed elephant motif to grey earthenware with flame patterns, incised sprigs, and traces of brown glaze, some bearing signs of lettering, were assessed by Noone as part of the cargo of a wrecked vessel washed in from the bay rather than evidence of a settlement [7, p. 217]. He concluded that the finds indicate associations with an ancient coastal trade from the north of the Indo-Chinese peninsula [7, p. 218].

Research and Documentation

The Society’s literature on pottery reveals a clear evolution in methodological approach and research priority over the course of the twentieth century. The earliest contributions, such as Evans’s 1923 note on the kendi, are brief comparative observations drawing on museum collections and a limited range of comparanda [5]. The 1940s and 1950s saw the emergence of more rigorous archaeological methodology: Noone’s 1941 beach find report, while modest in scope, applies contextual reasoning to distinguish settlement debris from shipwreck cargo [7]; Sieveking’s 1956 papers introduce use-wear analysis and experimental reasoning to the study of Neolithic material [2], [6].

The 1959 Loewenstein study marks a significant shift towards large-scale comparative archaeology, assembling evidence from Mesopotamia, the Indian subcontinent, Japan, Korea, and Manchuria to place Malayan Neolithic pottery within a global diffusion framework [1]. This approach directly challenged the earlier claim by Collings that ancient examples of perforated pottery “seem to be confined to Malaya” [1, p. 168], and extended the comparative framework established by Tweedie and Williams-Hunt, positioning the Malayan Neolithic within a broader Old World network rather than as an isolated phenomenon.

Harrisson’s 1971 monograph on the Niah double-spouted vessels represents the most substantial single contribution to the pottery literature in the Society’s journal [3]. It corrects the 1959 Solheim, Harrisson and Wall survey, which had identified only “about half a dozen” double spouts and suggested a “general deterioration” of pottery in caves other than the Great Cave; Harrisson’s quadrupled sample and wider cave distribution overturn both assumptions [3, pp. 40–41]. The paper also reflects a broader historiographical shift away from diffusionist explanations towards recognition of indigenous creative capacity in island Southeast Asia [3, p. 72].

The 1979 McKinnon note, while brief, illustrates the growing importance of Chinese export ceramics as a tool for establishing chronological frameworks in areas where indigenous material evidence is sparse [4]. It builds on the work of Van Orsoy de Flines, whose planned ceramic survey of Sumatra was forestalled by the Japanese invasion, and situates the Bukit Seguntang finds within the broader pattern of early Chinese export ceramics reaching the Malay world [4, p. 47].

Across these contributions, several recurring themes emerge: the tension between local and external origins for distinctive pottery types; the methodological challenge of assigning function to fragmentary or contextually ambiguous objects; and the use of pottery as a chronological marker in regions where other dating evidence is limited. The debate between Sieveking and Loewenstein over the function of the perforated cones, and between Harrisson and Malleret over the origin of double-spouted vessels, exemplify the kind of productive scholarly disagreement that has characterised the field.

MBRAS Sources

References

  1. J. Loewenstein (1959). Hollow clay stands and censers from Neolithic Malaya, and their western prototypes JMBRAS 32(1): 168–196. Read on JSTOR
  2. G. de G. Sieveking (1956). Pottery cones from Kodiang, Kedah JMBRAS 29(1): 189–194. Read on JSTOR
  3. Tom Harrisson (1971). Prehistoric double-spouted vessels excavated from Niah Caves, Borneo JMBRAS 44(2): 35–78. Read on JSTOR
  4. E.E. McKinnon (1979). A note on the discovery of spur-marked yueh-type sherds at Bukit Seguntang Palembang JMBRAS 52(2): 41–47.
  5. I.H.N. Evans (1923). On the persistence of an old type of water-vessel JMBRAS 1(1): 248–250.
  6. G. de G. Sieveking (1956). The stamped wares from Johore Lama JMBRAS 29(1): 194–195. Read on JSTOR
  7. H.D. Noone (1941). A find of pottery sherds on a beach near Sepang, Selangor JMBRAS 19(2): 217–218.