A True Martaban Jar A Burmese Ceramic Jar in the Ethnological Museum in Heidelberg (Germany)

EDITORS

AMY McNAIR, University of Kansas

ANNE McGANNON, New York

EDITORIAL BOARD

MILO C. BEACH, Former Director Freer and Sackler Gallery, Washington D.C.

EBERHARD FISCHER, Museum Rietberg Zurich

JAN FONTEIN, Museum of Fine Arts, Boston

B.N. GOSWAMY, Panjab University, Chandigarh

JEAN-FRANÇOIS JARRIGE, Musée Guimet, Paris

GREGORY LEVINE, University of California, Berkeley

KYÔTARÔ NISHIKAWA, Kanagawa Prefectural Museum of Cultural History

JESSICA RAWSON, Merton College, Oxford

ADELE SCHLOMBS, Museum für Ostasiatische Kunst, Cologne

RODERICK WHITFIELD, University of London

JOANNA WILLIAMS, University of California, Berkeley

PUBLISHERS

EBERHARD FISCHER, Museum Rietberg Zurich

JORRIT BRITSCHGI, Museum Rietberg Zurich

PRODUCTION

Layout and ProductionELIZABETH HEFTI,Winterthur, Switzerland

Chinese and Japanese GlossariesDAVID W. GOODRICH, Birdtrack Press, New Haven, CT, U.S.A.

LithographsTHOMAS HUMM, Humm DTP, Matzingen, Switzerland

PrintingWERK ZWEI Print+Medien Konstanz GmbH, Konstanz, Germany

© ARTIBUS ASIAE and the authors, 2014

ISSN 0004-3648

VOL. LXXIV, No. 22014

PUBLISHED BY THE

MUSEUM RIETBERG ZURICH

Reprint from

384

The contents of this and previous issues of ARTIBUS ASIAE are fully indexed in the ART INDEX.ARTIBUS ASIAE, VOL. LXXIV, 2014

CONTENTS

Articles

241 MELIA BELLI BOSE

Ancestors of the Moon: Bhati Mytho-History and Memorial Art at Bara Bagh, Jaisalmer

257 BRIGITTE BORELL

A True Martaban Jar: A Burmese Ceramic Jar in the Ethnological Museum in Heidelberg, Germany

299 ROSELYNE HUREL

A Ragamala Painted at Patna for Zain Ud-Din Ahmad Khan, Nephew and Son-in-Law

of Alivardi Khan

311 GAUTAMA V. VAJRACHARYA

Kirtimukha, the Serpentine Motif, and Garuda: The Story of a Lion That Turned into a Big Bird

337 FAN JEREMY ZHANG

Jin-Dynasty Pingyang and the Rise of Theatrical Pictures

1 Völkerkundemuseum der J. & E. von Portheim-Stiftung, Heidelberg, inv. no. 33296. The jar was donated to the

museum in 1980. My thanks go to Margarete Pavaloi, director of the museum, for giving me permission to study and

publish the jar and to procure a chemical analysis of the glaze. The museum also provided new photographs of the jar.

257

BRIGITTE BORELL

A TRUE MARTABAN JAR:

A BURMESE CERAMIC JAR IN THE ETHNOLOGICAL MUSEUM

IN HEIDELBERG, GERMANY

INTRODUCTION

The starting point of this paper is a large ceramic jar in the collection of the Ethnological Museum

in Heidelberg. Not much is known about its provenance except that it was acquired in southern

India, probably in Kerala.1 Nor do we have detailed information about its situation or use within India.

My research into large-sized storage jars in Southeast Asia and into this class of Burmese jars in

particular soon broadened into an inquiry into the term martaban, derived from the port town Marta-

ban (Mottama) in Lower Burma, and commonly used in sixteenth- and seventeenth-century sources.

In the past, it has often been explained as a popular name and generic term for large storage jars in

Southeast Asia. Although this view has been corrected at least in part, it continues to be used in an

indiscriminate and confusing manner. In the course of a necessary fresh review of the sources, I became

aware how abundant the historical sources on the jars called martaban are, on the one hand, and, on the

other hand, how often they were quoted only in short excerpts without discerning who the writers

were, what their situation was, and from what kind of experience they were writing, that is, whether

they were eyewitnesses or compilers of others’ accounts, perhaps with varying degrees of reliability.

The ensuing investigation focused first and foremost on the meaning of the term martaban in the

historical sources. My research shows that, far from being a generic term, in the sixteenth- and seven-

teenth-century sources, martaban is a determinate name designating exclusively a class of ceramic jars

produced in Lower Burma that were remarkable for their size and special qualities. Because of these

qualities, they were in great demand, not only domestically but particularly also in a wide export mar-

ket. The sources provide an abundance of other important information about the martaban jars not yet

fully recognized in ceramics research. They also give much insight into the various uses of these jars,

information that is usually lost and can only in exceptional cases be deduced from the archaeological

context.

Another point of importance that will be raised here is the distribution of these Burmese jars to

India and Sri Lanka. This is of particular interest with regard to the Heidelberg jar, which is said to

come from southern India. As the sources indicate, this was the primary destination of their maritime

export. However, with regard to the archaeological record, this line of investigation can at present only

be a beginning, since published finds of such jars are still scarce. In general, this class of Burmese jars

had an astoundingly wide, almost global distribution on board Asian and European ships, in an east-

ern direction to insular Southeast Asia, Japan, and Mexico, and in a western direction beyond South

Asia to the Middle East, the east coast of Africa, and farther into the Atlantic and its shores, even as

far as the eastern shore of Canada.

DESCRIPTION

The Heidelberg jar is already impressive for its measurements, though it is by far not one of the largest

of its kind (figs. 1–6, 9b). It is 73 centimeters high, with a maximum diameter of 77 centimeters and

a circumference of about 240 centimeters. It is capable of holding 210 liters. The clay is a dark brown-

ish-red.2 The lustrous glaze is transparent brown with an olive tinge, so that the glazed area of the jar

looks shiny dark brown. The lower part of the body and the base are unglazed, as is the interior, except

for a few splatters of glaze on the inner surface.

The bulbous body is slightly broader than high, its shape characterized by broad, high shoulders.

The neck is very short. Remarkable is the tapering towards the bottom: the diameter of the flat base

measures 33.5 centimeters, almost as small as the rim diameter. A precise figure for the rim diameter

is difficult to establish, as the mouth of the jar is not a perfect circle, but slightly distorted to an oval

form with diameters between 33.0 and 35.5 centimeters. The rim is everted and thickens toward the

rounded edge. On its upturned inner side, the rim has a broad, shallow groove. Four broad horizontal

strap-handles were placed on the shoulder; three of them are still extant, while the missing fourth han-

dle has been restored. Otherwise the jar is intact. Traces of wear are found in the rim area, in particu-

lar towards the inside of the neck. Here, in spots, the glaze is completely worn off, indicating regular

use over some time.

As was common in the manufacture of large vessels, the body of the jar was formed in stages in a

process that made use of sequential coiling and throwing for each stage of the jar. The lower part was

probably made by starting with a flat base and coiling up from it, then smoothing the wall on the fast-

spinning wheel. After this part had dried and stiffened enough to support weight, the potter added

more coils to form the upper part of the jar wall.3 The horizontal joint, just below the applied strip

around the body, is recognized on the inside by the thickening, which is only roughly smoothed. The

wall was nicely smoothed on the exterior whereas on the interior of the upper wall clear signs of the

coiling technique remain visible. The upper shoulder and rim were probably formed in a like manner

in a third stage.

Before any of the decoration in light-colored clay was applied, two fine horizontal lines were

incised: one at the transition between neck and shoulder, the other just below the handles. These lines

may have been intended to serve as a guide when applying the clay strips and placing the handles. For

decoration, strips of light-colored clay were applied in light relief. The strips appear in a pale yellow-

ish color under the glaze. A small area underneath one of the handles remained unglazed; here, the

original whitish color of the clay used for the strips is still discernible (fig. 4). Two horizontal stripes

258

2 Munsell Soil Color Charts (Newburgh, N.Y.: Macbeth, 1992), 2.5 YR 5/2, weak red.

3 For the technique, see Bernard Leach, A Potter’s Book (London: Faber & Faber, 1977), 108, figs. A, B, C. For additional

information on these technical aspects I am indebted to Louise Cort, personal communication, March 2013.

encircle the vase, one on the shoulder underneath the handles, the other on the lower part of the body.

The large zone between them is divided into eight panels by groups of vertical strips: alternately three

strips where the handles were later placed, and two strips in the intervals.4 Then the handles were placed

on the shoulder; the strap is attached on each end with an additional lump of clay, which is flattened

out and smoothed onto the surface. The traces of this process are still visible as an oval in very faint

relief, partly on the handle and partly on the shoulder. Finally, the exterior of the pot was glazed. A

chemical analysis of the glaze revealed a high amount of iron oxide, which is responsible for the dark-

brown color.5

The forming and firing of a jar of this size required experienced potters. Though primarily made

as utilitarian ware, the jar is striking not only for its craftsmanship but also for its aesthetic quality,

both of which deserve our esteem. In addition, as the sources relate, the qualities of jars of this kind

were equally astonishing.

DATE

The Heidelberg jar belongs to a class of large storage jars found widely dispersed in Southeast Asia.

Accordingly, they have received appropriate attention in Southeast Asian ceramics research,6 and quite

a number of them from several regions of Southeast Asia have been published. Characteristic of this

class are the dark reddish clay, the blackish-brown glaze, the shape with broad and high shoulders, the

tapering towards the base, which often has a smaller diameter than the rim, and the shape of the everted

rim with the shallow groove on its upturned inner side. Another diagnostic detail is the flattened and

smoothed oval shape of the additional lump of clay on each end of the strap-handle. The most obvi-

ous characteristic is the decoration made of light-colored clay in low relief, either stripes, as on our jar,

or rows of buttons resembling rivet heads. Among the shapes of these jars two types prevail, both bul-

259

4 In one of the intervals, the two vertical stripes appear to be absent, as if the potter had forgotten to apply them. How-

ever, on the shoulder at the lower edge of the horizontal stripe, their starting points seem to be discernible; perhaps

the stripes disappeared under the dark glaze, which is very thick here.

5 X-ray fluorescence microanalysis of the glaze on the Heideberg jar, inv. no. 33296, with values in weight percent:

n.d. = not detected

The non-destructive analysis was carried out in the laboratory of the Römisch-Germanisches Zentralmuseum, Mainz,

by Sonngard Hartmann. For their kind cooperation my thanks are due to Susanne Greiff and Sonngard Hartmann.

6 Roxanna Brown, The Ceramics of South-East Asia: Their Dating and Identification, 2nd ed. (Singapore: Oxford Univer-

sity Press, 1988), 104; Roxanna Brown, Southeast Asian Ceramic Museum, Bangkok University (Bangkok: Bangkok Uni-

versity Press, 2009), 73; Sakai Takashi, “Indoyô no tôji bôeki: Toruko to Higashi Ajia no kôryû wo megutte” (The

Ceramic Trade of the Indian Ocean: Concerning Exchange between Turkey and the Eastern Part of Asia), Jôchi Ajia-gaku 23 (2005): 261–62, 269–71; Nan Kyi Kyi Khai, “The Study on Myanmar Trade Ware: Martaban Jar and White

Dish,” Kanazawa Daigaku Kôkogaku kiyô 30 (2009): 101–13.

Na2O MgO Al2O3 SiO2 P2O5 SO3 K2O CaO TiO2 MnO FeO CuO ZnO SrO ZrO2 SnO2 PbO

3.05 2.82 8.99 44.75 0.45 0.29 1.30 18.80 0.80 0.05 9.31 0.10 0.09 4.16 5.10 n.d. n.d.

bous with high shoulders: one has a slender body; the other is more globular and very voluminous,

like the Heidelberg jar. Although the handles are normally horizontal, vertical handles also occur.7 In

addition, there are some variants in shape, for instance, a type with a squat, globular body. As an alter-

native to the everted rim and short incurving neck, an upright rim and a straight neck can also be

found.8

Closely comparable in shape and decoration to the Heidelberg jar are a jar from Borneo and two

others found in the Philippines.9 All three are without a dated context, however, as is often the case

with ceramic jars in Southeast Asia; most have been handed down for generations as highly valued

heirlooms.

Important evidence for the dating comes from shipwrecks that suggest a period for the manufac-

ture of these jars from the late fifteenth to the late seventeenth century. For instance, a jar of this class,

but of more slender shape, was found in the so-called Lena Shoal junk, wrecked off Busuanga Island in

the Philippines. Its cargo is dated to the late fifteenth or early sixteenth century.10 In addition, at least

two jars of this kind were recovered off Mombasa on the East African coast from a shipwreck, presum-

ably the Portuguese frigate Santo António de Tanná, which sank there in front of Fort Jesus in 1697.11

A jar very similar in shape to the Heidelberg example, although plain and without the applied

decoration, has been retrieved from the Portuguese wreck of the Espadarte, which sank off Mozam-

bique in 1558.12 Two further jars similar in shape and dimension to the Heidelberg jar were found along

with smaller vessels of this class in the cargo of a wreck off the Philippine island of Luzon, generally

assumed to be the Spanish galleon San Diego, which sank there after a battle on 14 December 1600.13

260

7 For examples, see Sumarah Adhyatman, Burmese Ceramics (Jakarta: Himpunan Keramik Indonesia, 1985), 21, fig. 21;

and Nan, “Myanmar Trade Ware,” 125, fig. 3 (jar from Oita, Japan).

8 See, for instance, Adhyatman, Burmese Ceramics, 21, fig. 24; and Christine L. van der Pijl-Ketel, ed., The Ceramic Loadof the “Witte Leeuw” (1613) (Amsterdam: Rijksmuseum Amsterdam, 1982), 225, inv. no. 12116 (from the Witte Leeuw).

For the upright rim and straight neck, see also Nan, “Myanmar Trade Ware,” 127, fig. 5; 130, figs. 12 and 13.

9 Barbara Harrisson, Pusaka: Heirloom Jars of Borneo (Singapore: Oxford University Press, 1986), col. pl. 60; Cynthia O.

Valdes, Kerry Ngyuen Long, and Artemio C. Barbosa, eds., A Thousand Years of Stoneware Jars in the Philippines(Makati: Metro Manila, 1992), 174, nos. 155–56; see also 196–97.

10 Franck Goddio and Gabriel Casal, eds., Lost at Sea: The Strange Route of the Lena Shoal Junk (London: Periplus, 2002),

228, no. 11.4 (H. 75 cm, maximum diameter 53 cm).

11 These are the jar MH 11 (H. 91.5 cm) and fragments of a similar one; Hamo Sassoon, “Ceramics from the Wreck of a

Portuguese Ship at Mombasa,” Azania 16 (1981): 109–10, fig. 9; Farid Paul Willoughby, “The Martaban Jars,” Insti-tute of Nautical Archaeology Newsletter 18, 2 (Summer 1991): 26, fig. 4. A third smaller jar, MH 1, came from the SantoAntónio de Tanná; Sassoon, “Ceramics,” 108, fig. 8; Willoughby, “Martaban Jars,” 27, fig. 5; Caesar Bita and Atthasit

Sukkham, “Martaban Jars Found in Kenya,” Southeast Asian Ceramics Museum Newsletter 7, 1 (2013): 3, fig. 2. See also

Paul J. Lane, “Maritime and Shipwreck Archaeology in the Western Indian Ocean and Southern Red Sea: An Overview

of Past and Current Research,” Journal of Maritime Archaeology 7 (2012): 9–41, esp. 10–16. Sassoon, “Ceramics,” 110,

mentions three more jars in Kenya, one privately owned in Mombasa (H. 104 cm) and two in Lamu, farther north on

the coast (H. 88 and 100 cm). Sakai, “Ceramic Trade,” 270, also includes in his lists an example in Zanzibar.

12 Southeast Asian Ceramics Museum Newsletter 5, 3 (2008): 3. This shipwreck is also known as the Fort San Sebastian wreck.

Lane, “Maritime and Shipwreck Archaeology,” 16–19.

13 Marie-France Dupoizat, “Asiatic Stoneware Jars,” in Treasures from the San Diego, ed. Jean-Paul Desroches et al. (Paris:

AFAA, 1996), 232–35, inv. nos. 3841 and 3696.

For these two jars, capacities of 280 liters by a height of 90 centimeters, and 350 liters by a height of

92 centimeters, have been established.

In excavations conducted between 1996 and 2000, numerous fragments of such jars were retrieved

from the wreck of an early seventeenth-century Portuguese ship at the mouth of the Tagus River near

Lisbon, the so-called Pepper Wreck, identified with a high degree of certainty as the Indiaman NossaSenhora dos Mártires, which sank there in a storm in September 1606 after a nine-month voyage from

Cochin, including a three-month stop in the Azores. The wreck suffered further damage when large

rocks rolled over its remnants, probably during the tsunami following the earthquake of 1755. Owing

to this unfortunate history, only shards have been saved. However, their diversity suggests that they

come from a substantial number of individual jars of this class.14 A large part of the top of one jar, with

a rim diameter of 19 centimeters and a wall thickness of 1.2 centimeters, has been preserved (fig. 9 a).15

Though of smaller dimensions, its rim profile is similar to that of the Heidelberg jar. In addition, the

glaze seems quite similar in color and appearance. Surveys in 2002 and 2004 of a wreck at Faial Island

in the Azores, identified as the Portuguese carrack Nossa Senhora da Luz, lost there in November 1615

on her return voyage from Goa, have so far recovered at least three fragments of this class of jars.16

A large jar (figs. 7, 8, 9 c), about 75 to 79 centimeters in height and similar in shape to those from

the San Diego, was recovered from the wreck of the Witte Leeuw, a ship of the Dutch East India Com-

pany (VOC ). It was on its return journey from Bantam (Banten) in West Java to the Netherlands when,

in June 1613, it was sunk after a battle with two Portuguese carracks near the island of St. Helena in

the southern Atlantic.17 Fragments from a piece resembling the large jar from the San Diego were

retrieved from another Dutch ship, the Mauritius, also on the return journey from Bantam. The Mau-ritius was wrecked in the Atlantic in March 1609, off Cape Lopez in the Gulf of Guinea on the west-

ern coast of Africa.18

261

14 Filipe V. de Castro, “The Pepper Wreck, an Early 17th-Century Portuguese Indiaman at the Mouth of the Tagus River,

Portugal,” The International Journal of Nautical Archaeology 32, 1 (2003): 10 (Burmese stoneware); Filipe V. de Castro,

The Pepper Wreck (College Station, Tex.: Texas A&M University Press, 2005), 101, 203–40 artifacts list (Martaban

jars); Ines A. Duarte Pinto Coelho, A cer√mica oriental da carreira da Índia no contexto da carga de uma nau – A presumívelNossa Senhora dos Mártires (Master’s thesis, Universidade Nova de Lisboa, 2008), online at cham.fcsh.unl.pt/files/

varia/tese_inespintocoelho.pdf (accessed 26 February 2012), 75–77 and artifacts lists (Martaban). Drawings of three

fragments of Burmese jars from this wreck are included in Sara Teixeira Simões, “A Brief Study Concerning Marta-

ban Jars,” in Maritime Contacts of the Past: Deciphering Connections amongst Communities. Maritime Archaeology I, ed. Sila

Tripati (New Delhi: Kaveri, 2015), 252–74, figs. 4, 10, 11; the fragment shown in figure 8, probably also Burmese,

was excavated at Largo do Chafariz de Dentro, further upriver towards Lisbon.

15 Coelho, Cer√mica oriental, 77, fig. 28 and artifact list no. cnans.sjb.arq.94.0021.

16 José A. Bettencourt, “Os vestígios da nau Nossa Senhora da Luz resultandos dos trabalhos arqueológicos,” Arquipélago– História, 2nd series, 9 (2005): 258–59, fig. 22.

17 Pijl-Ketel, The Ceramic Load of the “Witte Leeuw,” 226–27, inv. no. 12126 (rim diameter 44 cm, maximum diameter

at the shoulder 81 cm, base diameter 33.5 cm). I am indebted to William Southworth of the Rijksmuseum Amster-

dam for checking the measurements.

18 Eight wall fragments, described as covered with a blackish-brown glaze, were retrieved. The two shards illustrated in

the publication show a decoration with stripes and dots in light-colored clay characteristic of this class of jars; Michel

L’Hour, Luc Long, and Eric Rieth, Le Mauritius: La mémoire engloutie (Grenoble: Casterman, 1989), 180, illustration

on the margin (bottom left), PG 2121. For the decoration, compare the jar from the San Diego in Dupoizat, “Asiatic

Stoneware Jars,” 233, inv. no. 3841.

Another wreck in the harbor of Galle, Sri Lanka, was identified as the Avondster, a VOC ship with

an unusual and long history of more than eighteen years.19 When it sank in 1659, it was considered to

be old and worn out. After excavation, the Avondster suffered a second disaster caused by the tsunami

of December 2004. Many artifacts and part of the documentation were lost. The 1993–2004 excava-

tions yielded fragments of at least two large jars that doubtless belong to this class.20 Although the

publication gives no precise measurements, the rim diameter of jar CE 171 appears to be of similar

dimension to that of the big jar from the Witte Leeuw. The rim profiles of both CE 171 and CE 177 are

also of the same type.

In addition to the finds from shipwrecks, archaeological excavations of sites in Indonesia and Japan

have yielded fragments of jars of this class. In Japan, fragments of fourteen jars were found on the islands

of Okinawa and Kyushu. Fragments of one such jar were found in Oita at the Otomo-funai-machi site

in a context of damaged ceramics, interpreted as a destruction layer of the year 1586 when the Shimazu

forces overwhelmed the Otomo clan; hence the ceramics would antedate the year 1586. The other finds

from Kyushu and Okinawa are not closely dated, but the archaeological contexts of the sites appear to

be consistent with dates in the sixteenth and seventeenth centuries. The rim profile of the Oita jar, as

well as those of the other examples from Japan, is comparable to that of the Heidelberg jar.21 The frag-

ments of two jars from Okinawa also suggest a similar globular shape, with maximum diameters of

60 and 70 centimeters – only slightly smaller than our jar.22 The two excavation sites in Indonesia

with jar fragments, Banten on Java and the fortress Wulio on Buton Island, Sulawesi, are also consis-

tent with a sixteenth- to seventeenth-century date.23

In his study on the ceramic trade of the Indian Ocean, Sakai Takashi suggests classifying jars of

this kind into three chronological and stylistic groups based on dated contexts. His first group is dated

from the late fifteenth to the late sixteenth century and includes the jar from the Lena Shoal wreck.

The second group ranges from the end of the sixteenth to the end of the seventeenth century, and com-

prises the finds from the San Diego and from Oita, Japan. The third group dates from the eighteenth

century to the present and includes the modern production in Mandalay province in Upper Burma.24

262

19 Robert Parthesius, Excavation Report of the VOC Ship Avondster (1659) (Amsterdam: Centre for International Heritage

Activities, 2007), 12–13.

20 Hans Bonke, Robert Parthesius, and Christine van der Pijl-Ketel, eds., Artefacts Catalogue Avondster Site, 1998–2004(Amsterdam: Centre for International Heritage Activities, 2007), 19, fig. 2.14; 47 (cat. no. CE 171, consisting of forty

fragments of rim, wall, and base) and 48 (cat. no. CE 177, rim fragment). Their rim profiles and the broad strap-han-

dles clearly indicate an attribution to our class; in the publication they are assigned to the group “Stoneware, A.1 type

1.b,” which also includes the fragments CE 172, CE 173, CE 175, and CE 176. Although the fragments were recovered

in the aftermath of the tsunami, no illlustrations of them are given. Fragments CE 178 and CE 174, lost during the 2004

tsunami but at least illustrated in some manner, may come from jars of this class on the grounds of their size (the base

diameter of CE 174 would be about 23 cm) and the color of the clay and glaze, judging from the drawing or photographs.

21 Nan, “Myanmar Trade Ware,” 125, fig. 3 (Oita); 124, figs. 1a, 1g (Okinawa); 125, fig. 2 (Hakata). The graphic recon-

struction of the jar MM-1 from Nagasaki (ibid., 126, fig. 4a), shows an unusually wide mouth. However, some cau-

tion is advisable, since the estimate for the rim diameter is based on only one rim fragment of a few centimeters; this

fragment may be from the long side of a distorted mouth, as on the Heidelberg jar.

22 Ibid., 124, figs. 1e, 1f (Okinawa).

23 Sakai, “Ceramic Trade,” 296, figs. 9, 10; Nan, “Myanmar Trade Ware,” 107–8, 118, 127, figs. 6f, 6g.

24 Sakai, “Ceramic Trade,” 262, 269–71.

The Heidelberg jar clearly belongs to his second group, into which most of the known examples from

Southeast Asia fall. In this group, our jar finds close comparisons not only in terms of overall shape,

but also for the particular detail of the rim shape, which is shared by the fragmented jar from Oita25and

jars from the San Diego,26 the Nossa Senhora dos Mártires,27 and the Witte Leeuw.28 These comparisons

suggest that the Heidelberg jar dates to the late sixteenth or early seventeenth century.

ORIGIN

Initially, when only few examples of this class were known, they were classified as Chinese.29 Later,

they were thought to originate in Vietnam or Thailand, and were even assigned to the Si Satchanalai

production, although nothing similar had been found at the kiln sites there, and the clay body of Si

Satchanalai stoneware is different.30 These attributions resulted in part from the erroneous assump-

tion that the large storage jars, commonly called martabans, were only exported from the port town of

Martaban in Lower Burma, but were principally made elsewhere. However, Pamela Gutman, using

epigraphic and literary evidence, argued already in 1978 for an active ceramic production in Burma

going back many centuries. Unfortunately, this important conference paper was known only to a few

researchers until its publication in 2001.31

When Gutman presented her paper in 1978, archaeological evidence for ancient ceramics from

Lower Burma was still scarce. Credit must be given to Sumarah Adhyatman, who, based on textual

evidence and studies in the field in 1984, traced the origin of the class of jars under discussion to Lower

Burma.32 Indeed, broadly similar large jars are still being made to this day in Myanmar and used for

water storage.33 They are now produced at Kyaukmyaung near Shwebo in Upper Burma. However,

263

25 Nan, “Myanmar Trade Ware,” 125, fig. 3 (Oita).

26 Dupoizat, “Asiatic Stoneware Jars,” 234, inv. no. 2251; 235, inv. nos. 2991, 3696, 2882, 2141.

27 Coelho, Cer√mica oriental, 77, fig. 28.

28 Pijl-Ketel, The Ceramic Load of the “Witte Leeuw,” 226–27, inv. no. 12126.

29 E. W. van Orsoy de Flines, Guide to the Ceramic Collection (Jakarta: Museum Pusat Djakarta, 1972), pls. 75–76; William

Sorsby, South-east Asian and Early Chinese Export Ceramics (London: W. Sorsby Ltd., 1974), 49, no. 74. On the prob-

lems of their classification, see the references in n. 6 above.

30 Roxanna M. Brown, Otto Karow, Peter W. Meister, and Hans W. Siegel, Legend and Reality: Early Ceramics from South-East Asia (Kuala Lumpur: Oxford University Press, 1977), 255, no. 182; Harrisson, Pusaka, 46 (addressed as

Sawankhalok and assigned to the sixteenth or seventeenth century). See the critical revision in Brown, Ceramics of South-East Asia, 104; Si Satchanalai stoneware usually fires grey and is often speckled with black particles; ibid., 71–72.

Roxanna Brown, The Ming Gap and Shipwreck Ceramics in Southeast Asia: Towards a Chronology of Thai Trade Wares(Ban gkok: River, 2009), 66; Lucas Chin, Ceramics in the Sarawak Museum (Kuching: Sarawak Museum, 1988), 150–51

(described as “Vietnamese, sixteenth–seventeenth centuries”).

31 Pamela Gutman, “The Martaban Trade: An Examination of the Literature from the Seventh Century until the Eigh-

teenth Century,” Asian Perspectives 40, 1 (2001): 108–18.

32 Adhyatman, Burmese Ceramics. See also Sumarah Adhyatman and Abu Ridho, Tempayan di Indonesia/Martavans inIndonesia, rev. 2nd ed. (Jakarta: The Ceramic Society of Indonesia, 1984), 64–67, in particular the post script.

33 Sumarah Adhyatman visited production centers in the south and north of Myanmar. In Twante in the south, glazed

and unglazed wares of moderate size were produced. The large jars of sizes up to one meter in height were made in the

Shwebo region in the north. Adhyatman, Burmese Ceramics, 10. See also Mick Shippen, The Traditional Ceramics of South-East Asia (London: A&C Black, 2005), 145–62.

the potters are Mon who were brought there from the south as captives of war in the mid-eighteenth

century, when the Burmese conquered Pegu. According to their tradition, the Mon potters from the

Shwebo area trace their ancestry back to the Gulf of Martaban region, a “place of big kiln sites (O-yon)

near Mudon in Mon State,” south of Moulmein (Mawlamyine).34

Over the last three decades, knowledge about ancient Burmese ceramic production has been greatly

augmented.35 The ceramics found at burial sites in the Tak and Omkoi areas along the border between

Thailand and Myanmar have led to an awareness and recognition of the diversity of Burmese ceramic

production, in particular Burmese celadons and lead-glazed wares dated to the fifteenth and sixteenth

centuries.36 The lead-glazed earthenware was apparently the predominant tradition in Burma, which

includes glazed architectural ceramics.37

Four different areas of ceramic production are now known in the Irrawaddy Delta region of Lower

Burma. Two of them, the Ngaputaw and Pathein areas, are in the western delta. The other two are far-

ther east: one is in the area between Yangon and Bago (ancient Pegu), with an estimated one hundred

kilns; the other is in the Twante area. Both lie on the western coast of the Gulf of Martaban. Appar-

ently, the most active area was Twante, where several hundred ancient kilns have been identified. Pre-

liminary investigations of a few kilns have revealed a celadon production dated to a period from the

fifteenth through the seventeenth century.38

The attribution of the group of jars under discussion to Lower Burma, and most likely to the region

of the Gulf of Martaban, is now generally accepted, although precisely which kilns made these dark-

brown glazed jars has not yet been determined.39 In light of the tradition on the origin of the present-

264

34 Nan, “Myanmar Trade Ware,” 111–12. See also Myo Thant Tyn, “Tradition of Myanmar Glazed Ceramics and Its His-

torical Status in Southeast Asia,” Kanazawa Daigaku Kôkogaku kiyô 26 (2002): 22; Tsuda Takenori, “Myanmaa seyû

tôji: Seisan gijitsu to hennen no tame no shiryô” (Glazed Ceramics in Myanmar: Their Manufacturing Technique and

Historical Documents for Dating), Jôchi Ajiagaku 23 (2005): 58; and Louise Cort (2 March 2008), http://seasiance-

ramics.asia.si.edu/search/object.asp?id=F1994.16 (accessed 19 February 2012). Similar information was collected in

Adhyatman, Burmese Ceramics, 10. See also Shippen, Traditional Ceramics, 145–46.

35 A recent recapitulation appears in John N. Miksic, “Kilns of Southeast Asia,” in Southeast Asian Ceramics: New Lighton Old Pottery, ed. John N. Miksic (Singapore: Southeast Asian Ceramic Society, 2010), 66–68.

36 Sumitr Pitiphat, Ceramics from the Thai-Burma Border (Bangkok: Sunta Press, 1992), 38–40.

37 John Guy, Ceramic Traditions of South-East Asia (Singapore: Oxford University Press, 1989), 5–14; Sylvia Fraser-Lu,

Burmese Craft: Past and Present (Kuala Lumpur: Oxford University Press, 1994), 187–205.

38 Myo Thant Tyn and Dawn F. Rooney, “Ancient Celadon in Myanmar: A New Ceramic Discovery,” Orientations 32, 4

(2001): 57–61. See also Myo Thant Tyn and U Thaw Kaung, “Myanmar Historic Earthenware,” in Earthenware inSoutheast Asia: Proceedings of the Singapore Symposium on Premodern Southeast Asian Earthenwares, ed. John N. Miksic

(Singapore: Singapore University Press and the Southeast Asian Ceramic Society, 2003), 291–99, esp. 298–99; South-east Asian Ceramics Museum Newsletter 1, 1 (2004): 1–2; and Southeast Asian Ceramics Museum Newsletter 5, 1 (2008): 4

(Burmese celadons from the Santa Cruz shipwreck, dated to the late fifteenth to early sixteenth century).

39 Southeast Asian Ceramics Museum Newsletter 4, 3 (2007): 2, fig. 1–2. The statement in Nan, “Myanmar Trade Ware,”

108–9 and 111, suggesting four areas for the production of big jars in Lower Burma, namely Twante, Bago, Myaungmya,

and Mottama, still has to be verified. The kiln sites at Kyat-chaung and Phayagyi north of Bago are said to have pro-

duced big jars, but details as to their type and date are lacking. Kiln no. 3 at Myaungmya in the western Irrawaddy

Delta yielded large tubular supports that are believed to be evidence that large jars were fired there. However, there

is as yet no decisive evidence for assuming that the few jars of our class now in local collections in the presumed pro-

duction areas were actually produced there; they could just as well have been brought from kiln sites elsewhere to

day potters in Kyaukmyaung, the search for the ancient kilns should probably be directed to the area

south of Martaban. Such were the conclusions when this article was submitted in July 2012. Shortly

thereafter, more than fifty kilns were discovered to the southeast of Martaban along the road between

Kyaikmayaw and Kawkareik. It was reported that they appear to have been used for the production

of the famous jars.40 In March 2014, a group of specialists in ancient ceramics and kiln technology

started an investigation of the area east of Martaban along the Gyaing River. They discovered eleven

ceramic production sites of martaban jars.41 It is not surprising to find the kilns close to a river since

waterways were needed for the transport of the heavy jars. Further investigation of these kiln sites is

necessary, but these discoveries seem to prove that production of the jars under discussion indeed took

place in the area to the south and east of Martaban, as has been proposed here on the basis of written

sources and reported local tradition.

THE TERM MARTABAN

The term martaban or martavan has often been used for large ceramic jars in Southeast Asian ceramics

research. The use of this term has a long history. Unfortunately, in scholarly research it has caused con-

fusing discussions and interpretations, sometimes with misleading results.

The term is derived from the port town of Martaban in Lower Burma (Myanmar), situated on the

eastern coast of the gulf named after it. In the late thirteenth and first half of the fourteenth century,

Martaban was the capital of the Mon kingdom founded by Waweru. Although the capital of this

Ramanya kingdom was later moved to Pegu (Bago), Martaban remained an active port for trade until

the middle of the sixteenth century,42 and was later succeeded by other ports in the Martaban Gulf –

Pegu, Dagon (Yangon), and finally Syriam. In addition, Lower Burma had another port, Bassein

(Pathein), in the western Irrawaddy Delta. In terms of the history of Lower Burma, the period of man-

ufacture of the martaban jars spans the later part of the Ramanya kingdom (1287–1539), the First Toun-

265

their present locations. For instance, the jar in the Bago Archaeological Museum, illustrated in Nan, “Myanmar Trade

Ware,” 109, fig. 8d, is certainly not Burmese, but of twentieth-century southern Chinese manufacture, probably from

Guangdong; cf. Stephen Markbreiter, Arts of Asia 8, 1 (1978): 90, fig. 1–2.

40 They were discovered by accident during road construction in October 2012; see Cherry Thein, “More Clues to Marta-

ban Jar Mystery,” The Myanmar Times, 26 November 2012, online at http://www.mmtimes.com/index.php/national-

news/3359-more-clues-to-martaban-jar-mystery.html (accessed November 2012).

41 The group was led by U San Win and included Daw Nan Kyi Kyi Khaing, as well as Don and Toni Hein; Louise Cort

and Don Hein, personal communication, January 2015.

42 Donald F. Lach, Asia in the Making of Europe, vol. 1, The Century of Discovery (Chicago: The University of Chicago Press,

1965), 539–41; Michael Aung-Thwin, “Lower Burma and Bago in the History of Burma,” in The Maritime Frontier ofBurma, ed. Jos Gommans and Jacques Leider (Amsterdam and Leiden: KITLV Press, 2002), 47–49; Gutman, “The

Martaban Trade,” 111–12. For sources on Muttama (Martaban) pre-dating this period, see Tilman Frasch, “Coastal

Peripheries during the Pagan Period,” in Gommans and Leider, Maritime Frontier, 61–65. Discussions of the short

periods of claims of Sukhothai’s or Ayudhya’s domination of Martaban are here left aside; see David K. Wyatt, Thai-land: A Short History (New Haven, Conn., and London: Yale Univerity Press, 1982), 50, 55 (late thirteenth to early

fourteenth century), 90–91 (mid-sixteenth century), 111 (1662). Jan M. Pluvier, Historical Atlas of South-East Asia(Leiden: E. J. Brill, 1995), 16 § 24, 17 § 26, maps 12, 13.

goo empire (with the capital in Pegu from 1539 to 1599) – the historical sources often speak simply of

the Kingdom of Pegu – and the Restored Toungoo empire (1599–1752) with the capital in Ava.

The heyday of the term martaban in European and Ottoman sources is the period of the sixteenth

and seventeenth centuries. In this period, the term occurs as a popular name for the large storage jars

from Martaban. Although both the European and Ottoman sources of the sixteenth and seventeenth

centuries explicitly mention the production of large jars at Martaban, late nineteenth- and twentieth-

century researchers disregarded this, and assumed that the sources stated erroneously that the jars

acquired at Martaban were also made there. Instead, they posited that the jars were only shipped from

Martaban, but made elsewhere.43 The jars were thought to have arrived there either in transshipment

or to have been brought overland from other production centers. Hence, the popular use of martabanentered modern scholarly writing as a generic term for large ceramic jars, whether made in China, Thai-

land, or other regions of Southeast Asia, thereby generating considerable confusion.44

On the basis of the present state of knowledge about Burmese ceramic production, it is most

rewarding to take a fresh look at the historical sources from the sixteenth to eighteenth century that

use the term martaban. The review will provide clear and reliable information not only about the place

of manufacture of martaban jars, but also about the precise and determinate use of the name. In addi-

tion, the sources offer a vast corpus of information about the use of martaban jars and the demand for

them in inter-Asian trade.

In the early sixteenth century, around 1518, the Portuguese Duarte Barbosa wrote about the large

black-glazed jars made at Martaban that were highly valued as trade goods.45 Writing towards the end

266

43 Brief overviews are found in Brown, Ceramics of South-East Asia, 99, 100; Adhyatman and Abu Ridho, Tempayan,

48–49; and Regina Krahl, Chinese Ceramics in the Topkapi Saray Museum Istanbul: A Complete Catalogue (London:

Sotheby’s, 1986), 884. A nineteenth-century forerunner was the (now obsolete) controversial discussion between the

ethnographer Adolf Bernhard Meyer (“Ueber die Herkunft gewisser Seladon-Porzellane und ueber die Martabani’s,”

Österreichische Monatsschrift für den Orient 11, 1 [1885]: 11–13) and the orientalist Josef von Karabacek (“Die Mart√ban-

Seladon-Frage,” Österreichische Monatsschrift für den Orient 11, 1 [1885]: 29–35). See also n. 77 below; and Julian Raby,

“Fa≈fur, Merteban, and Other Terms for Porcelain and Celadon,” in Krahl, Chinese Ceramics in the Topkapi Saray, 83

n. 15. For compilations of the sources, usually quoted without giving the context, see for instance: Henry Yule and A.

C. Burnell, A Glossary of Colloquial Anglo-Indian Words and Phrases (1886; reprint, Richmond, U.K.: Curzon Press,

1995), 559–60; R. C. Temple, “Notes on Antiquities in Ramannadesa,” The Indian Antiquary 22 (1893): 364–65; Gut-

man, “The Martaban Trade,” 113–14; and Myo, “Tradition of Myanmar Glazed Ceramics,” 21–23.

44 For instance, Nanne Ottema, Chineesche Ceramiek: Handboek (Amsterdam: J. H. de Bussy, 1946), 125–37, 126, fig. 145

includes under the term “martavanen” even jars made in Guangdong dating from about the ninth century; Shelagh J.

Vainker, Chinese Pottery and Porcelain (London: British Museum Press, 1991), 146–47; and Jessica Harrison-Hall, Cat-alogue of Late Yuan and Ming Ceramics in the British Museum (London: British Museum Press, 2001), 445–48.

45 Duarte Barbosa, Livro em que dá relação do que viu e ouviu no Oriente Duarte Barbosa, ed. A. Reis Machado (Lisbon: Agên-

cia Geral das Colónias, 1946), 196: “Martabão . . . Tambémse fazem neste lugar muitas e grandes jarras de porcelana,

mui grossas, rijas e formosas; há aí delas que levam uma pipa de água; sao vidradas de preto e muito estimadas entre

os mouros, e entre eles valem muito, as quais eles daqui levam com muito beijoim em paes.” Duarte Barbosa, A Descrip-tion of the Coasts of East Africa and Malabar in the Beginning of the Sixteenth Century, trans. Henry E. J. Stanley (1866;

reprint, London: Hakluyt Society, 1970), 185; Duarte Barbosa, The Book of Duarte Barbosa, trans. Mansel Longworth

Dames (1918; reprint, London: Hakluyt Society, 1967), 158–59. The earlier entry for the year 1508 in Yule and Bur-

nell, Glossary, 560, an excerpt of a letter from the Viceroy Dom Francisco Almeida to the Portuguese king, has to be

deleted from the list of references to martaban jars: the commodity is lac, and the vessels from Pegu and Martaban

of the sixteenth century, the Dutchman Jan Huygen van Linschoten gives a detailed account in his

Itinerario (1596) of the large pots made in Martaban and brought in great numbers to India, where –

as he says – they were generally called martabans after the name of that town.46

Both Duarte Barbosa and Linschoten were in Portuguese service in India – Barbosa for about six-

teen years, from around 1500 to around 1516 or 1517, mainly in Kannur, and Linschoten for about five

years, from 1583 to 1588 in Goa – and hence familiar with the jars from Martaban arriving in India. As

far as we know, neither Barbosa nor Linschoten ever went to Martaban, and therefore neither of them

actually could have seen the jars being made there. But we may justly assume that they had reliable

information. The period of Portuguese presence at the ports of Lower Burma begins directly after the

conquest of Malacca in 1511, when Afonso de Albuquerque sent an ambassador to the court at Pegu. In

1512, a Portuguese factor sailed from Malacca to Martaban and concluded some business deals. And

from 1514 to 1516, the Portuguese had a permanent office in Martaban. António Dinis, secretary to the

Portuguese factor, stayed permanently in Martaban for two years, from 1514 until the expulsion in 1516.

After an interruption of a few years, a peace treaty was signed with Pegu at the end of 1519.47 Linschoten

himself discloses his sources of information, although in a different context: on the one hand, the Por-

tuguese who regularly went to Pegu and therefore had day-to-day experience, and on the other, the peo-

ple from Pegu themselves, as a number of them were living in India, which here surely refers to Goa.48

Similarly, Sidi Ali Reis, a Turkish admiral and commander of the Ottoman Indian Ocean Fleet,

explains the name of the jars in his nautical handbook, Muhit, of 1554, written during his stay in

Gujarat: “Because the porcelain mertebans which come to the Land of Rum [Turkey] with preserves in

them are produced in Merteban, a port in the province of Siam, people call them mertebans.”49 Apart

from Ibn Battuta’s Rihla, which I shall discuss later, this is to my knowledge the earliest written evi-

267

mentioned in this text are not pots but ships (naos in the Portuguese original). See Gaspar Correa, Lendas da India,vol. 1, part 2 (Lisbon: Academia Real das Sciencias, 1859), 900–1: “porque estas naos partem cedo, e as naos que o

trazem de Pegú e Martabao, vem tarde.” This has likewise been misunderstood by J. George Scott and J. P. Hardi-

man, Gazetteer of Upper Burma and the Shan States, part 1 (Rangoon: Government Printing, 1900), 2:399.

46 Jan Huygen van Linschoten, Itinerario: Voyage ofte schipvaert van Jan Huygen van Linschoten naer Oost ofte Portugaels Indien,1579–1592, ed. H. Kern (The Hague: Martinus Nijhoff, 1955–57), 1:76; Jan Huygen van Linschoten, The Voyage ofJohn Huyghen van Linschoten to the East Indies, ed. Arthur Coke Burnell (1885; reprint, London: Hakluyt Society, 1970),

1:101.

47 Geneviève Bouchon, “Les premiers voyages portuguais à Pasai et à Pegou (1512–1520),” Archipel 18 (1979): 127–52;

Sanjay Subrahmanyam, The Political Economy of Commerce: Southern India, 1500–1650 (Cambridge: Cambridge Univer-

sity Press, 1990), 99–100; Geneviève Bouchon, “Nouvelles de Martaban 1516: Lettre d’un agent commercial por-

tuguais,” in Études birmanes en hommage à Denise Bernot, ed. Pierre Pichard and François Robinne (Paris: École française

d’Extrême-Orient, 1998), 218–19; Om Prakash, “Coastal Burma and the Trading World of the Bay of Bengal,

1500–1680,” in Gommans and Leider, Maritime Frontier, 96–97.

48 Linschoten, Itinerario, 1:75; Linschoten, Voyage, 1:100.

49 English translation after Raby, “Fa≈fur, Merteban, and Other Terms,” 83. See also D. Bittner and W. Tomaschek,

trans. and comm., Die topographischen Capitel des indischen Seespiegels Mohit (Vienna: K. K. Geographische Gesellschaft,

1897), 60–61. Sidi Ali Reis completed the Mohit in Ahmedabad, Gujarat, in 1554; the two manuscripts used for the

translation are copies dated to 1558 and 1570. He relied on earlier texts, in particular those of Sulaiman al-Mahri from

the early sixteenth century, but he also added information from his own experience, in particular remarks on the prod-

ucts of the different countries; see Bittner and Tomaschek, Mohit, 5, 6; and G. R. Tibbetts, A Study of the Arabic TextsContaining Material on South-East Asia (Leiden and London: E. J. Brill, 1979), 14, 181, 189.

dence for the popular name martaban used for such jars and, moreover, accompanied by the explana-

tion that this term is derived from the port of the same name. Likewise, the Portuguese physician Gar-

cia da Orta, who lived in Goa for thirty years until his death in 1568, colloquially uses the name “jar-ras martabaas” in his famous book on medicinal drugs, published in Goa in 1563, which long served as

a standard European work on tropical diseases.50

Duarte Barbosa explicitly mentions their dark (“preto”) glaze, which is characteristic for this class

of Burmese jars. He was also receptive to the aesthetic qualities of the jars in that he described them

as beautiful (“formosas”).

Later, in the early seventeenth century, François Pyrard de Laval wrote of the large storage jars man-

ufactured in and exported from Martaban. He, too, was attracted by their appearance and described

them “as very beautiful jars, better glazed and made than any I have seen elsewhere.”51 Pyrard de Laval

originally set out from Saint-Malo in northwestern France as a merchant to the East Indies in 1601. He

encountered a series of misfortunes there and for almost five years was a captive in the Maldives. After

his return in 1611 he published an account of his voyage. His discussion of the martabans is included in

the description of a Southeast Asian ship “from Sunda,” wrecked on Guradu Island in the Maldives.

He points out the advantages of different methods of storing water on board ship (see below). The

method used throughout the whole of India – this wording would include Southeast Asia – was to

store the water in special jars of enormous capacities made in Martaban. Pyrard de Laval had traveled

in the Gulf of Bengal region as far as Chittagong (Chottogram), although he says nothing about ever

having been in the town of Martaban. He probably knew these jars from Goa or other places in India,

and possibly also from a later voyage as a soldier in Portuguese service to the islands of the Malay archi-

pelago and to the Moluccas.

The martaban jars continued to be in great demand in inter-Asian trade throughout the seventeenth

century, now mainly exported from Syriam, then the chief port of Lower Burma. A recent study by

Wil O. Dijk has compiled an overwhelming corpus of evidence for their role in the trade of the Dutch

VOC with Burma. From her lists of Dutch exports from Burma in the years from 1634 to 1680, it

becomes apparent that thousands of martaban jars left Burma on Dutch ships alone. Year by year, they

were exported in large numbers: for instance, five hundred jars are listed for the year 1650. When in

1677 the Dutch decided to close down the VOC factories in Burma, they took pains to secure the

remaining goods, among them 1,100 martaban jars. A total of 3,243 martaban jars appears in Dijk’s lists

compiled from the VOC archives for the whole period.52 However, this figure is incomplete, because

268

50 Garcia da Orta, Colóquios dos Simples, e Drogas he Cousas Mediçinais da India (Goa, 1563), 184r.

51 “Somme qu’en toute l’Inde, ils n’ont point notre invention des pipes, mais usent seulement de jarres les plus belles,

mieux vernis et les mieux façonnées que j’aye vues ailleurs. Il y en a qui tiennent autant qu’une pipe et plus. Elles se

font au royaume de Martabane, d’où on les apporte et d’où elles prennent leur nom par toute l’Inde. L’eau ne se g√te

et corrompt jamais là-dedans, et elles se ferment avec la clef.” François Pyrard de Laval, Voyage de Pyrard de Laval auxIndes orientales (1601–1611), ed. Geneviève Bouchon and Xavier de Castro after the third edition of 1619 (Paris: Chan-

deigne, 1998), 244–45. François Pyrard de Laval, The Voyage of François Pyrard of Laval to the East Indies, the Maldives,the Moluccas, and Brazil, trans. Albert Gray and H. C. P. Bell (1887–89; reprint, London: Hakluyt Society, 1964),

1:258–59.

52 Wil O. Dijk, Seventeenth-century Burma and the Dutch East India Company, 1634–1680 (Singapore: Singapore Univer-

sity Press, 2006): 116–17, 120–22, 194, Appendix 1, Table B, 7–17. Tijs Volker, Porcelain and the Dutch East India

some years are missing in the available documentation, and in some cases only the value of a consign-

ment of martaban jars is mentioned, not the number of jars. Therefore, an even higher number of marta-bans exported by the Dutch has to be estimated. In addition, we have to bear in mind that the goods

exported by the Dutch represent only a fraction of the total, and that a host of other traders – Muslim

and Hindu from India, in particular the famous Gujarati merchants, but also Turks, Persians, Arme-

nians, and Europeans other than the Dutch, not to forget the Peguans themselves – were active in the

export of trade goods from Burma before and after the period of Dutch presence, although compara-

bly detailed written documentation for the activities of those other traders and for their shipments is

lacking. In the early sixteenth century, Lower Burma sent about fifteen to sixteen junks and twenty

to thirty cargo pangajauas to Malacca and Sumatra every year.53

The Dutch also contributed to the distribution of martaban jars in the Southeast Asian archipel-

ago. Ambon in the Moluccas, where the Dutch had one of their earliest and most important factories

for the spice trade, is the island where Georg Eberhard Rumpf (Rumphius) lived for almost half a cen-

tury until his death in 1702. Originally from Hanau (Hesse), Rumphius had enlisted as a young man

with the Dutch VOC, arriving in Batavia in 1653, and in the following year he came to Ambon, where

he began his studies of the local flora and fauna, culminating in his magnum opus, the HerbariumAmboinense, posthumously published in 1741. His chapters about the coconut palm also comprise an

account of its use and powers, and he relates how the coconut shells are used to ladle water from the

huge vases, the martavans and Siamese pots; he adds the explanation that these were very large and

sturdy jars made in the regions of Martaban and Siam.54 In one of his other works, the Amboinsche

269

Company (Leiden: E. J. Brill, 1971), 221, using the Daghregister, arrives at a figure of about 1,300 martaban jars imported

at Batavia.

53 Tomé Pires, The Suma Oriental of Tomé Pires, trans. Armando Cortesão (London: Hakluyt Society, 1944), 98 (bring-

ing mainly rice and other foodstuffs, lac, benzoin, musk, precious stones, etc.). Pires gives a good picture of the extent

of mercantile activities in the Indian Ocean and Southeast Asian waters in the early sixteenth century, especially for

merchants to Pegu, in ibid., 17 (from Aden), 42, 45–46, 101 (merchants from Gujarat and Cambay, who dominated

the trade, often accompanied in their ships by other traders from regions ranging from the eastern African coast to

Turkey and Persia); for Pegu trade, 98 (to Malacca and Sumatra), 144 (to Pase), 195 (to Java; the Pegu junks used for

the transport were also part of the merchandise and likewise sold, not only in Java but also in Sumatra and the Malay

Peninsula, i.e. in Pedir, Pase, Pahang, and Palembang), 283 (to Malacca).

Joost Schouten of the VOC visited Malacca immediately after its conquest by the Dutch in 1641; in his report to

the Governor-General Antonio van Diemen, he says that the Peguans formerly imported into Malacca gold, rubies,

musk, martavan jars, rice, and other foodstuffs, although for some years this trade was disrupted due to the internal

wars of that kingdom; see P. A. Leupe, ed., “Stukken betrekkelijk het beleg en de verovering van Malakka op de Por-

tuguezen in 1640–1641,” Berigten van het Historisch Genootschap 7 (1859): 258–375, esp. 315–16; see also Marie A. P.

Meilink-Roelofsz, Asian Trade and European Influence (The Hague: Martinus Nijhoff, 1962), 70, 138, 364 n. 13.

For the seventeenth century, see also Dijk, Seventeenth-century Burma, 146–74. The VOC documents mention two

ships owned by Indians from Bengal and Surat, and one ship owned by two Armenian brothers leaving Syriam in 1680,

fully laden, including martaban jars among other goods, ibid., 149; two ships under the account of an English trader

sailed from Syriam to Masulipatnam in 1663 and among the goods on them were a number of martaban jars, ibid. 163.

54 Georg Eberhard Rumpf (Rumphius), Herbarium Amboinense (Amsterdam, 1741), book 1, ch. 3, p. 18: “en scheppers,

waar mede men het water uit de groote Martavanen, en de Siamsche potten schept, (zynde zeer groote, en dikke pot-

ten, in de landen van Martavaan, en Siam gemaakt, die men in geheel Indien vervoert, om alderhande vogtigheit daar

in te bewaren) . . ..”; translation: “and dippers [are made from the shells of the coconut] to ladle the water from the

Rariteitkamer, he mentions two large martavans flanking the famous “Moon of Pejeng” in Bali, a huge

bronze drum with a tympan diameter of 160 centimeters and a height of 186.5 centimeters.55 Presum-

ably the two martavan jars on either side of the “Moon,” the latter at that time lying on the ground,

were also specimens of imposing size. Rumphius was a man with a wide range of interests, and was

certainly kept well informed through his contacts, not least among them his friend François Valen-

tijn, a minister based in Ambon from 1685 to 1694. Obviously Rumphius was well aware of the differ-

ent origins of the ceramic jars, and the discriminating eye of the botanist might well be trusted to dis-

cern between a Burmese jar and one from Siam.56

The most important account of the jars made in Martaban is given by Joris van Coulster in his final

report on Pegu for the VOC dated 1675. He was in Burma for about thirty-four years until 1671, and

for nearly twelve years was one of the chief factors of the Dutch VOC, in the years from 1666 to 1671

in the port of Syriam. His list of the country’s products includes an entry on martaban jars (fig. 10):

Martavans take their name from the town of Martavan, to the east of Pegu, from whence they are

transported by boat along inland waterways to Syriam, generally in the rainy season when their

purchase can be done at the cheapest. The material with which the aforementioned pots were

glazed, and the way in which this is done, the people [the Martavan potters] would not disclose

to us, fearing a loss of their craft and thus of their livelihood. But their fear is probably groundless,

since jars as large as these could not be made in any other country or from any other clay. Whether

they might have changed their minds [and told their secret of glazing], might be possible.

In a note on the margin, under the heading “Martavanen,” he adds, “As reported to us, the clay the

potters use for them is found nowhere else in the country.”57 James Low, too, attests to the renowned

quality of that clay in his geological survey (1833): “Potter’s earth is obtained in abundance near Marta-

270

large Martavans and the Siamese pots (these are large and thick pots, made in the countries of Martaban and Siam,

which are brought into all of India to store various liquids) . . ..”

55 Georg Eberhard Rumpf, D’Amboinsche Rariteitkamer (Amsterdam: F. Halma, 1705); Georg Eberhard Rumpf, TheAmbonese Curiosity Cabinet, trans. E. M. Beekman (New Haven, Conn., and London: Yale University Press, 1999),

239–40 (book 3, ch. 7). The “Moon of Pejeng” at that time, i.e. the later seventeenth century, was lying as “fallen out

of the sky.”

56 For instance, the Amboinsche Rariteitkamer, book 3 has two chapters with elaborate accounts of ceramics: chapter 23 on

porcelains, with details about the differences between Japanese and Chinese porcelains, and chapter 24 on celadons;

Rumpf, Ambonese Curiosity Cabinet, 269–75. On the complex structure of the VOC’s shipping networks, which allowed

a high degree of fast communication between the trading posts and the central control in Batavia and hence access to

a range of detailed information, see Robert Parthesius, Dutch Ships in Tropical Waters: The Development of the DutchEast India Company (VOC) Shipping Network in Asia, 1595–1660 (Amsterdam: Amsterdam University Press, 2010),

113–21, 171–72.

57 Joris van Coulster, “Verclaringe sommiger saecken des Coninckrijcks Peguw, te praesenterenaen den Ed. Heer Gou-

verneur-Generaal Joan Maetsuycker en d’Ed. Heeren Raden van Indie,” Batavia, 30 March 1675. VOC archives 1313,

ff. 168–81, esp. f. 175 v and f. 176 r: “Martavanen; hebben de naem nade stadt Martavan, oostwaerts vande stadt Pegu

gelegen, van waer deselve met vaertuijgen binnen door tot Sierjangh werden aengebragt, en meest in den regen tijd

als wanneer den incoop [f. 176] op het sevielste te geschieden dient, de spetie daer mede voorsz: potten werden ver-

glaest, en hoedanigh met deselve dient gehandelt sulx heeft dat volckjen d’onse niet willen aenleren, dugtende hun

handtwercq, ende neringe mits dien mogte comen te verminderen, maer die sorge is voor haer niet seer nodigh, om

dat in andere landen ofte van andere aerdt niet wel soo grote potten kunnen werden gemaect, ofse hun hier over nogh

ban. Of this, most of the utensils known by the name of Pegu jars, were formerly made.”58 Of partic-

ular interest is Coulster’s information that the pots were brought from Martavan to Syriam along

inland waterways. The low-lying land of Lower Burma had a system of interconnecting waterways nav-

igable at least in the rainy season with barges, which were the usual means of transport. Consulting a

map of the Syriam River and Pegu copied for the Atlas Blaeu-van der Hem around 1670 (fig. 11), we

find to the south of Pegu an eastern tributary to the Syriam or Pegu River with the addendum: “River

from Martaban that runs to the sea.”59 Such connecting waterways from the Salween to the Sittang and

finally to the Irrawaddy are found in two late eighteenth-century maps published by Francis Hamil-

ton, who describes them as navigable at least in the rainy season. One of these maps was drawn for

Hamilton by a native of Toungoo (fig. 12), the other by a slave of the crown prince of Burma.60 John

271

hebben bedagt, sulx can wel sijn. [On the margin below on f. 175 v] Nota: De aerd’ of soodanige als de pottebackers,

daer toe gebruijcken, is op geen andere plaetse in ’trijcq te vinden, soo ons is berigt.” The transcription of the hand-

written text has been generously provided by Wil O. Dijk. I am also greatly indebted to her for help with its transla-

tion and interpretation. See also Dijk, Seventeenth-century Burma, Appendix V, 81.

58 James Low, “Observations on the Geological Appearances and General Features of Portions of the Malayan Peninsula

and of the Countries Lying betwixt it and 18˚ North Latitude,” Asiatic Researches 18 (1833): 128–62, esp. 153; on the

soil in the region of Martaban, Low writes, “The sub-stratum in the lower parts is commonly a stiff clay . . ..” See also

James Low, “History of Tennasserim,” The Journal of the Royal Asiatic Society 3 (1836): 27, 50.

59 For the complete map that had been copied from the Secret Atlas of the East India Company, see Peter van der Krogt

and Erlend de Groot, eds., The Atlas Blaeu–van der Hem (t’Goy-Houten: HES & De Graaf, 2005), 5:386, map 41:16.

On the transport along inland waterways from Martaban to Syriam, see also Pieter van Dam, Beschrijving van de Oost-indische Compagnie, vol. 2,2, ed. F. W. Stapel (The Hague: Martinus Nijhoff, 1932), 87 (probably taken from Joris van

Coulster’s account); generally Dijk, Seventeenth-century Burma, 76–81, esp. maps 9, 10.

60 Francis Hamilton, “Account of a Map Constructed by a Native of Taunu, of the Country South from Ava,” EdinburghPhilosophical Journal 5 (1821): 75–84 with descriptions of these anastomosing or inosculating rivers, esp. 75–76, 82–83,

pl. 5. Francis Hamilton, “Account of a Map of the Kingdom of Pegu,” Edinburgh Journal of Science 1 (1824): 267–74,

esp. 273, pl. 10. During his stay in Ava from March to November 1795, Francis Hamilton had several maps made for

him by local inhabitants. On Francis Hamilton (formerly Francis Buchanan), see Michael Symes, An Account of anEmbassy to the Kingdom of Ava, sent by the Governor-General of India, in the Year 1795 (London: W. Blumer & Co., 1800),

reprinted in SOAS Bulletin of Burma Research 4, 1 (2006): 59–60, 161–62. Joseph E. Schwartzberg, “Introduction to

Southeast Asian Cartography,” in History of Cartography, vol. 2,2, Cartography in the Traditional East and SoutheastAsian Societies, ed. John B. Harley and David Woodward (Chicago: The University of Chicago Press, 1994), 690, 698;

Joseph E. Schwartzberg, “Southeast Asian Geographical Maps,” in Harley and Woodward, Cartography, 743–52, 803,

877–78, esp. 749, with a short discussion of the first of these two maps, raising doubts about the reliability of the pat-

tern of anastomosis. Schwartzberg is certainly correct in his criticism insofar as this should not be interpreted as a per-

manent pattern of drainage. Henry Yule, A Narrative of the Mission to the Court of AVA in 1855 (London: Smith, Elder

& Co., 1858; reprint, Kuala Lumpur: Oxford University Press, 1968), 145–46 footnote, 270–72, 284, already pointed

out that such communications in their upper valleys as represented in the maps are undoubtedly imaginary, as moun-

tains form a barrier between the Sittang and the Salween, as do hills between the Sittang and the Irrawaddy. How-

ever, Yule admits inosculating branches in the lower part of the course of the Sittang and the Irrawaddy; see ibid.,

285, where he speaks of a natural channel between the Sittang and the Pegu rivers.

Nevertheless, for the low-lying southern lands near the coast, such a particular connection of waterways, i.e.

creeks and canals, from Martaban might have worked in the rainy season – and at high tide – for flat-bottomed river

barges. Otherwise it is inexplicable why, independently of each other, the different reports about it exist, all based on

information received from the locals. The eighteenth- and nineteenth-century British accounts certainly had no knowl-

edge of Joris van Coulster’s unpublished handwritten report of 1675 for the VOC.

Crawfurd, in his journal of a visit to Martaban in late March and early April 1826, gives a similar

account of interconnecting branches and canals from the Salween to the Sittang and Pegu rivers.61

Further details of interest on the jars are found in the highly original book by Alexander Hamil-

ton published in 1727. Alexander Hamilton was a captain who navigated the seas between the Cape

of Good Hope and Japan for more than thirty-five years, from 1688 to 1723. He made at least two vis-

its to Pegu, one in 1709 and one in 1710/11, and also gives an eyewitness account of Martaban. He notes

the decline of this town, which had previously been one of the most flourishing ports of trade in the

East. He writes about the still-existing ceramic production in Martaban and the manufacture of very

large jars. Furthermore, he notes the use of a lead glaze.62 This tradition of applying a lead glaze con-

tinued into the present-day production of large jars and other ceramics at Kyaukmyaung in Upper

Burma.63 However, from chemical analysis, we know that the glaze of the Heidelberg jar contains no

lead.64 This raises the question whether, at a certain point in time, a change occurred in the practice

of the potteries, or whether different parallel traditions existed. Lead-glazed earthenwares have a long

history in Burma, best known from architectural plaques and green-and-white wares.65 More archaeo-

metric research on the glaze of Burmese jars from different periods is needed here.

272

61 John Crawfurd, Journal of an Embassy from the Governor of India to the Court of Ava (London: Colburn, 1834), 2:76,

reprinted in SOAS Bulletin of Burma Research 3, 2 (2005): 538: “When twelve miles above Martaban. . .. We passed

the Kadachaong creek, which leeds to Rangoon through the Setaang and Pegu rivers, and thence again through sev-

eral cross channels to Bassein, a direct distance of more than two hundred miles. The internal navigation of Pegu

appears to me to possess natural facilities far beyond any other Asiatic country . . ..” See also the map in John Craw-

furd, Journal of an Embassy to the Courts of Siam and Cochin China (London, 1828; reprint, Kuala Lumpur: Oxford Uni-

versity Press, 1967): it is called “Kyouksaree C.” [C. for Creek] in the map; Yule, Mission to the Court of AVA in 1855.

See also the maps in John James Snodgrass, Narrative of the Burmese War (London: Murray, 1827), and in Thomas A.

Trant, Two Years in Ava: from May 1824 to May 1826 (London: Murray, 1827) with connecting waterways at least

between the Rangoon River (= Pegu River) and the Sittang. See also Carl Ritter, Die Erdkunde im Verhältnis zur Naturund zur Geschichte des Menschen, part 5,2, vol. 4,1, Die Indische Welt (Berlin: G. Reimer, 1835), 135. Of course, consider-

ing that these low-lying lands were subject to flooding and silting processes, causing certain geomorphological

changes, the precise course of those interconnecting waterways of the early nineteenth century is not necessarily iden-

tical with that of the late seventeenth century. On the extent of the geomorphological changes, see the remarks by

Elizabeth H. Moore, Early Landscapes of Myanmar (Bangkok: River Books Press, 2007), 42–45.

62 Alexander Hamilton, A New Account of the East Indies (Edinburgh: John Mosman, 1727), 2:63; he says he was in Pegu

in 1709 (p. 40). For a chronology of Hamilton’s visits to Pegu, see the introduction in William Foster, ed., A NewAccount of the East Indies by Alexander Hamilton (London: Argonaut Press, 1930), 1:xxvii; and Justin Corfield and Ian

Morson, eds., British Sea-Captain Alexander Hamilton’s A New Account of the East Indies (17th–18th Century) (Lam-

peter: Edwin Mellen Press, 2001), xvi, 306–7, 352, 367–68.

63 Adhyatman, Burmese Ceramics, 9; Shippen, Traditional Ceramics, 157–58. Late nineteenth-century reports on the ceramic

production in the Twante region near the west coast of the Martaban Gulf explicitly mention the current manufac-

ture of large jars up to heights of about 120 cm and, in particular, note the use of a lead glaze made from a mixture of

galena and rice water. H. R. Spearman, British Burma Gazetteer (Rangoon: Government Press 1879–80), reprinted as

Gazetteer of Burma (Delhi: Gian, 1987), 1:418–19; 2:559, 849. See also Scott and Hardiman, Gazetteer, 401; J. G. Scott,

Burma: A Handbook of Practical Information, 3rd rev. ed. (London: Alexander Moring, 1921), 280; and Fraser-Lu, BurmeseCraft, 209.

64 See the table of chemical analysis in n. 5 above.

65 Guy, Ceramic Traditions, 5–14; Fraser-Lu, Burmese Craft, 195–205; Tsuda, “Myanmaa seyû tôji,” 55–80.

Fig. 1 The Heidelberg jar. H. 73 cm. Ethnological Museum, Heidelberg (33296).

Photo: Alexandra Landerer.

Fig. 2 The Heidelberg jar. Drawing by the author.

0 6 12 cm

Fig. 3 The Heidelberg jar. Photo: Alexandra Landerer.

Fig. 4 The Heidelberg jar. Photo: Alexandra Landerer.

Fig. 5 The Heidelberg jar. Photo by the author.

Fig. 6 The Heidelberg jar. Photo: Alexandra Landerer.

Fig. 7 Jar from the Witte Leeuw (1613). H. 75.5–79 cm. Rijksmuseum, Amsterdam

(NG-1978–127–12126-W).

Photo courtesy of the Rijksmusem,

Amsterdam.

Fig. 8 Jar from the Witte Leeuw (1613).Rijksmuseum, Amsterdam (NG-

1978–127–12126-W).

Photo courtesy of the Rijksmusem,

Amsterdam.

Fig. 9 Rim profiles of three different jars.

Drawings by the author.

a.

Jar from the Nossa Senhora dos Mártires (1606).After Coelho, Cer√mica oriental, 77.

b.

The Heidelberg Jar.

c.

Jar from the Witte Leeuw (1613), Rijks-museum, Amsterdam (NG-1978–127–12126-

W). After van der Pijl-Ketel, Witte Leeuw,227.

0 5 10 cm

Fig. 10 Joris van Coulster’s account of Pegu of 30 March 1675, VOC Archives 1313, the entry on martaban jars. Courtesy of the National Archives, Den Haag.

a. Detail of folio 175 verso (bottom).

b. Detail of folio 176 recto (top).

Fig. 11 Map from the Atlas Blaeu-van der Hem, copied around 1670. The lower image shows a detail of the Syriam River

and Pegu (“Caert van de rivier Siriangh, en de Stadt Pegu”). To the east of Pegu are depicted the hills of Martaban (“’t

Hoge Landt van Martavan”). Below, it notes an eastern tributary to the Syriam River: river from Martavan which runs to

the sea (“Rivier van Martavan die jn Zee loopt”). Courtesy of the Austrian National Library, Vienna.

Fig. 12 Detail of a map of the country south of Ava constructed by a native of Taunu. After Francis Hamilton, “Account of a Map,” The EdinburghPhilosophical Journal 5 (1821): pl. 5. Photo courtesy of the University Library, Heidelberg.

Fig. 14 The market at Goa. Detail from Linschoten, Itinerario (1596). Fig. 15 Pottery in Nwe Nyein, Shwebo district, Upper Myanmar. Two

men with a dry pot lashed into a bamboo cradle carry it to the kiln. Photo:

Mick Shippen.

Fig. 13 Jan Huygen van Linschoten, Itinerario (1596), plate depicting the market at Goa. Photo courtesy of the Herzog August Library Wolfenbüttel:A:22.3Geom.2˚.

Fig. 17 Jar with locking iron lid, 67� 56 cm. Museum of International

Folk Art, Santa Fe, New Mexico. Gift of the Fred Harvey Collection,

IFAF Collection (FA.1979.64.58). Photo: Paul Smutko.

Fig. 18 Jar found at Ferryland, Newfoundland, Canada. Photo courtesy

of the Colony of Avalon Foundation.

Fig. 16 Jars in the Travancore Palace at Padmanabhapuram, Tamil Nadu.

Photo: Ramesh Chandran Sreevalsam.

Taken together, the historical sources from the early sixteenth century onwards leave no doubt that,

in Lower Burma in the area of Martaban, a large-scale production of huge storage jars existed. At least

by the mid-sixteenth century, the sources attest that these jars were popularly called martabans after

their place of manufacture. Without doubt, the martaban jars of the sources are to be identified as the

class of Burmese ceramic jars discussed here. Archaeological finds from dated contexts, mainly ship-

wrecks, attest a period for these Burmese jars similar to that covered by the sources, that is, from the

late fifteenth century through the sixteenth and seventeenth centuries.

So far, this review of the written sources has left Ibn Battuta’s Rihla out of consideration. Dating

from the mid-fourteenth century, it is usually quoted as the earliest evidence for the term martaban.

However, on closer inspection, several points in the relevant text seem odd. The term martaban is used

in an episode set in the kingdom of Tawalisi in Southeast Asia, where Ibn Battuta arrived in the course

of his voyage from Sumatra to China. Before he departed, the Princess of Tawalisi presented him with

a range of goods, mainly provisions for the voyage, and among them “four martabans, large jars, filled

with ginger, pepper, citrus fruit, and mangoes, all salted as for a sea voyage.”66 Several scholars

attempted to identify Tawalisi – the places suggested range from Vietnam (Champa), Sulawesi, and

Borneo, to the Moluccas and Mindanao in the Philippines – without arriving at a satisfactory solu-

tion. Henry Yule and, more recently, David Waines have pointed out the novelistic character of the

episode, which might explain why it is so difficult, and perhaps impossible, to arrive at a conclusive

identification for Tawalisi.67

The term martaban for ceramic jars is not known from other sources of this early period, but occurs

only in Ibn Battuta’s Rihla.68 This is strange if, as the text implies, it was already a commonly used

name for storage jars. The later sources discussed above usually emphasize the origin of these jars and

explain their popular name. In contrast, Ibn Battuta’s text omits any reference to the port town of

Martaban in Lower Burma. His voyage from Bengal to Sumatra, described earlier in his travelogue,

was rather direct; only one stop is mentioned, probably in the Andaman or Nicobar Islands. He cer-

tainly says nothing about a visit to Martaban.69

281

66 Charles Défrémery and Beniamino R. Sanguinetti, trans. and eds., Voyages d’Ibn-Batoutah (Paris: Impr. Impériale,

1853–58), 4:248–54, esp. 252–53; H. A. R. Gibb and C. F. Beckingham, trans. and eds., The Travels of Ibn Battuta,A.D. 1325–1354 (London: The Hakluyt Society, 1994), 4:884–87. I thank Raif Georges Khoury and Werner Arnold,

both in Heidelberg, for translating this passage for me from the Arabic.

67 Recent studies on Ibn Battuta’s Rihla generally warn about accepting everything in his travelogue at face value. It has

even been disputed whether he reached China at all. For recent summaries of the problems concerning the Tawalisi

episode, see Ross E. Dunn, The Adventures of Ibn Battuta (Berkeley: University of California Press, 1986), 258, 264

n. 34; Gibb and Beckingham, Ibn Battuta, 884–85 nn. 25, 27–28; David Waines, The Odyssey of Ibn Battuta (London:

I. B. Tauris, 2010), 9–10, 191–92; and Ralf Elger, trans. and ed., Ibn Battuta: Die Wunder des Morgenlandes (Munich:

C. H. Beck, 2010), 209.

68 Acccording to Tibbetts, Arabic Texts, 230–35, the port of Martaban seems to appear not earlier than the fifteenth cen-

tury as a destination in the texts of the Arab navigators. It appears in the texts of Ibn Majjid of the late fifteenth cen-

tury and of Sulaiman al-Mahri, writing in the early sixteenth century. The fifteenth century is the period that brings

prosperity to the port; Anthony Reid, Southeast Asia in the Age of Commerce, 1450–1680 (New Haven, Conn.: Yale Uni-

versity Press, 1993), 2:54; Victor Lieberman, Strange Parallels: Southeast Asia in Global Context, c. 800–1830, vol. 1, Inte-gration on the Mainland (New York: Cambridge University Press, 2003; reprint, 2010), 46, 122, 129.

The question I wish to raise here with all due caution is whether the passage in Ibn Battuta’s Rihlais to be accepted as reliable evidence for the production of jars known under the name martaban already

in the mid-fourteenth century, as has been assumed, or whether the term martaban in Ibn Battuta’s

text might possibly be a later interpolation.70 This is not to deny that Burma already produced ceram-

ics in the fourteenth century, nor that storage jars of various origins were used and widely traded in

Southeast Asia in this period. At issue is the use of the term martaban for a specific class of ceramic jars.

On the evidence of all the other sources, this term seems to belong to a later period. It is hoped that

future philological research will clarify this matter one way or the other.

In 1349/50, Wang Dayuan, a Chinese contemporary to Ibn Battuta, published an account with

descriptions of countries and their inhabitants in insular Southeast Asia and in the Indian Ocean, based

on personal experience from his travels abroad. In his twenty-fifth chapter, he gives a description of

Baduma, which is thought to refer to Martaban. However, Wang Dayuan makes no mention of spe-

cial jars among the local products of Baduma listed in his text.71

When and where the term martaban came to be used in a more general manner is still an open ques-

tion.

282

69 According to his account, Ibn Battuta went from Bengal rather directly to Sumatra. The text mentions only one stop

in the country of the dog-faced Barahnakar people, known as a literary topos also from other sources, and usually located

on the Andaman or Nicobar Islands. Other commentators suggest a place on the west coast of Upper Burma or near

Cape Negrais at the western entrance to the Gulf of Martaban. For further details of the discussion, see Tibbetts, Ara-bic Texts, 97, 155; Gibb and Beckingham, Ibn Battuta, 874 n. 1; and Elger, Wunder des Morgenlandes, 207. Dunn, Adven-tures of Ibn Battuta, 255, remarks that “this part of the Rihla is possibly murky and disarranged.” He reconstructs two

stops on the way to northern Sumatra: the first is the Barahnakar stop, supposedly on the coast of Upper Burma; the

second is the stop at Qaqula on the western coast of the Malay Peninsula, which is described later in the text.

70 Even in the Arab world, documented knowledge of Ibn Battuta’s Rihla appears rather late. So far, the earliest seems

to be al Maqqari (1577–1632); see Stephan Conermann, Die Beschreibung Indiens in der “Rih˙la” des Ibn Battuta (Berlin:

K. Schwarz, 1993), 25–27. The famous version compiled by Muhammad al-Bailuni (d. 1674) was the first known to

Europeans; Elger, Wunder des Morgenlandes, 7–9; here, in al-Bailuni’s version of the Tawalisi episode, no explicit men-

tion is made of the martaban jars. See also Édouard Dulaurier, “Description de l’archipel d’Asie, par Ibn-Bathouta,

traduite de l’arabe,” 2e partie, Journal Asiatique 9 (1847): 226, 239, 252 n. 31.

For a fruitful exchange of views, I am indebted to Ralf Elger, Halle, personal communication, November 2011.

He informed me about his observations in Ibn Battuta’s text that some of the geographical names in the Maledives

and in Ceylon are known from later Portuguese sources, but are not found in other Arabic texts. This suggests that

we might have to reckon with some later additions to Ibn Battuta’s text. With regard to the term martaban, the case

would be decided if it appears in the version of a manuscript which itself dates from the fourteenth or early fifteenth

century.

71 William W. Rockhill, “Notes on the Relations and Trade of China with the Eastern Archipelago and the Coast of the

Indian Ocean during the Fourteenth Century,” T’oung Pao 16 (1915): 255–56; Wang Dayuan, Daoyi zhilüe jiaoshi, ed.

Su Jiqing (Beijing: Zhongsha shuju, 1981), 130–32. Nor is a mention of jars found in the ninety-ninth chapter on the

country of Wudie, which has been variously identified as Orissa or Pegu; Rockhill, “Notes,” 444–45; Wang Dayuan,

Daoyi zhilüe, 375–78; Sun Laichen, “Chinese Historical Sources on Burma: A Bibliography of Primary and Second-

ary Works,” Journal of Burma Studies 2 (1997): 19–20. On the structure of Wang Dayuan’s book, see Roderich Ptak,

“Glosses on Wang Dayuan’s Daoyi zhilüe (1349/50),” in Récits de voyages asiatiques: Genres, mentalités, conception de l’e-space. Actes du colloque EFEO-EHESS de décembre 1994, ed. Claudine Salmon (Paris: École française d’Extrême-Orient,

1996), 132.

In the course of the eighteenth century, the name “Pegu jars” appears as a new term for these large

storage jars.72 Nineteenth- and early twentieth-century reports confirm the continuing manufacture

of such jars in Lower Burma, now in the vicinity of Pegu, and also in Upper Burma,73 although export

of these jars no longer had to be taken into account. In the catalogue of the exhibition of Indian art at

Delhi in 1902–1903, the text on the Burmese jars comments that the “Pegu jars” are no longer exported,

not even to India, and includes information about “jars made in Delhi which bear the vernacular name

of martabán.” Their local production is attested to be in existence already by 1869, although it is

assumed that up to about 1800, “the martabáns sold in India . . . were entirely imported from Burma.”74

From an early nineteenth-century travel account, it appears that in Oman as well, the use of the term

martaban could be extended to local products. James B. Fraser lists among Oman’s locally made prod-

ucts “earthen jars called murtuban.”75

The discussion on the name martaban for jars is related to that on another meaning of the same

term adopted by the Ottomans for celadon dishes. A detailed analysis of this discussion would breach

the scope of this paper, and the question can only be briefly touched upon here. The terms mertebanand nerdüban appear in documents from the late sixteenth century in the Topkapi Palace in Istanbul.

Although the earliest example, dating from AH 986/1578 CE, lacks clear specification as to which kind

of ware it refers to, seventeenth-century documents and other sources make clear that it is a term for

celadons. In the Palace documents, the new term merteban as a generic name for celadons replaced the

older names zeytuni, derived from the name of the Chinese port Zeytun (Quanzhou), and babaguri.76

It has been thought that – in the same way as assumed for the martaban jars – the name was derived

from Martaban, which functioned as a transshipment port for celadons made elsewhere. However, the

matter might be more complex. Katib Çelebi, a scholar of the Ottoman empire writing in the mid-

seventeenth century, relates some interesting information in his world geography, Cihannüma. He says

that Martaban produces “valuable dishes and vessels; these are not as pure and as figural as the porce-

283

72 See the list in Yule and Burnell, Glossary, 559–60; and Temple, “Antiquities in Ramannadesa,” 364–65.

73 See n. 63 above.

74 George Watt, Indian Art at Delhi 1903. Being the Official Catalogue of the Delhi Exhibition, 1902–1903 (Calcutta: Gov-

ernment Printing, 1903), 94–95: “There would seem little doubt that the martabáns sold in India, a century ago, were

entirely imported from Burma as regular articles of trade even to such remote inland towns as Delhi.” See also Scott,

Burma: A Handbook, 279.

75 James B. Fraser, Narrative of a Journey into Khorasan in the Years 1821 and 1822 (London: Longman, 1825; reprint, Delhi:

Oxford University Press, 1984), 18. In modern Palestinian Arabic, the word martaban denotes the present-day small

glass jars used for preserves. For this information I thank Werner Arnold, Heidelberg, who started a thorough search

in the dictionaries for me; personal communication, June 2014. Such a use of martabans for food preserves certainly is

in accordance with Ibn Battuta’s text; one may speculate whether this modern usage of the term was inspired by the

revival of Ibn Battuta’s Rihla from the mid-nineteenth century onwards after the edition by Defrémery and San-

guinetti was published; see Elger, Wunder des Morgenlandes, 8.

76 Raby, “Fa≈fur, Merteban, and Other Terms,” 83; see also p. 27 for a quotation from the journal of Antoine Galland,

who stayed in Istanbul in the years 1672–73. Likewise, in his translation of the Thousand and One Days published in

1710–12, Pétis de la Croix describes in a footnote the “grand bassin de Martabani” as “porcelain verte”; François Pétis

de la Croix, Les mille et un jours (Paris, 1778), 3:55 and note b.

lain of Northern and Southern China, but on the other hand they are solid and strong.”77 No explicit

mention is made as to the color of the glaze, but it seems reasonable to understand these as celadons,

as has been done in the past. Again, others assumed that Katib Çelebi erroneously concluded that wares

exported from Martaban were also made there. However, judging from the present state of knowledge

about a lively production of celadons in the Irrawaddy Delta, the explanation might be simpler and

more direct. It might in fact refer to celadons made in the Martaban Gulf region that were prized, even

if they could not compare to the Chinese celadons.

In other sources, green-glazed tablewares also appear under the name “ghori dishes.”78 They were

valued as detectors of poisoned food, as it was a general belief that celadons would react on contact

with poison. Understandably, they were in great demand, also in India, Persia, and farther west. In

seventeenth-century VOC documents, they figure as one of the typical export commodities taken from

Burma to the Coromandel coast, to Pulicat, and to Masulipatnam, where they were sold at great

profit.79 One thousand ghori dishes were taken to Golconda alone every year. Ghori dishes were also

among the high-prestige items offered by the Burmese in official gift exchanges.80 Although these may

have occasionally included antique and imported celadons,81 the bulk of the celadons exported at that

time were probably produced in Burma. At least one production area for celadons has been identified

in Twante in Lower Burma (see above). The evidence from shipwrecks indicates that, from the late

fifteenth century on, exported Burmese celadons began to replace the celadon plates of the Si

Satchanalai kilns in central Thailand.82 The VOC documents on the export of ghori dishes from Burma

point to a continuous large-scale production of Burmese celadons and other glazed tableware through

the seventeenth century.

In his Amboinsche Rariteitkamer, Rumphius writes about the “Poison dishes or Gori”:

284

77 Quoted from Raby, “Fa≈fur, Merteban, and Other Terms,” 83 with note 16 (from the second version begun in 1654).

Josef von Karabacek, “Zur muslimischen Keramik,” Österreichische Monatsschrift für den Orient 10, 12 (1884): 281–92,

esp. 286–87 n. 37. On the controversy between the orientalist Josef von Karabacek and the ethnographer Adolf Bern-

hard Meyer, see n. 43 above; for a balanced view on the controversy, see Friedrich Hirth, Ancient Porcelain: A Study inChinese Mediaeval Industry and Trade (Leipzig: F. Hirth, 1888), 22–23.

78 Dijk, Seventeenth-century Burma, 251 n. 48, Appendix V, 74–75. The origin of the name ghori is still unclear; for expla-

nations, see Volker, Porcelain, 112; and Raby, “Fa≈fur, Merteban, and Other Terms,” 84.

79 Dijk, Seventeenth-century Burma, 125.

80 Ibid., 106, Appendix IV, list 5, 62–63.

81 The extremely expensive “nine gory dishes from Awa” brought from Pegu to Coromandel in 1652 may have been

antiques, as Volker, Porcelain, 106, suggests. They must have been exceptional pieces, and different from the common

ghori dishes; presumably they were imports either from China or, more likely, from Thailand, and already antiques.

Another entry for 1652, (pp. 106–7, ) says that in Persia, “veritable antique gouris” fetched comparably high prices;

cf. p. 182. One document of 1665 features the interesting remark that a set of twelve ghori dishes presented by the

Burmese king to the Dutch governor was “made in the Siamese way”; Dijk, Seventeenth-century Burma, Appendix V,

75. The large Si Satachanalai celadons were already in decline in the early sixteenth century, although Si Satachanalai

ceramic production continued at least until the end of the sixteenth century or even into the seventeenth; see Nancy

Tingley, “A Brief Note on the Terminal Date of the Si Satchanalai Kilns,” in Living a Life in Accord with the Dhamma:Papers in Honor of Professor Jean Boisselier on his Eightieth Birthday, ed. Natasha Eilenberg, M. C. Subhadradis Diskul,

and Robert L. Brown (Bangkok: Silpakorn University, 1997), 483–87.

82 Brown, Ming Gap, 62–65, 67.

They are bought in Pegu, where they come from Ava, and are transported to the Old Indies, where

they are held in great esteem, so that the Guzatts [Gujaratis] once paid 150 to 200 Rixdollars for

these Dishes, because, they said, they would not tolerate any poison, but since the greed of the peo-

ple from Pegu made them imitate and fake them, they came to be despised, and, as I am led to

believe, they remain in the Warehouse of the Company in Golconda, because the Chinese will not

buy them, since their conceitedness makes them discount anything that did not come from Sina ....

And so there appears to be a difference between the green Poison dishes and Gorys from Pegu which,

according to the Chinese, were once made in Sina.83

This seems to point to two different supply sources for ghori dishes: one up in the north – it produced

the dishes brought downriver from Ava on the Irrawaddy – and one in Lower Burma, presumably the

Twante area near Pegu. A northern source of supply for celadons can also be deduced from Joris van

Coulster’s account of 1675, in which he notes that the Peguans brought the ghori dishes on their return

from the border with China.84 Although these could be celadons brought from China, previous spec-

ulations about a possible celadon production in the Shan States now seem to be confirmed, since a num-

ber of ancient kilns have been discovered in the area north of Mongkung.85 Research on Burmese

celadons is still incipient. The preliminary reports speak of hundreds of kilns in the Irrawaddy Delta,

in the south, and in the Shan States in the north, still awaiting further research. This points to a large-

scale ancient production of celadons in Burma. Future resurveys of collections in India and farther west

may result in more identifications of exported Burmese celadon ware.

Important archaeological evidence for the broad export of celadons from Twante to western des-

tinations comes from Tatsuo and Hanae Sasaki’s excavations at Julfar, a port site of the fourteenth to

seventeenth century on the west coast of the Persian Gulf. Archaeometric analyses of excavated celadons

confirm their origin from kilns in the Twante region. These Burmese ceramics, the dominant ware

among the celadons found at the site, come from layers dated to the fifteenth to sixteenth century,

whereas the less abundant Chinese Longquan celadons found at the same excavation location come

mainly from the earlier strata. The excavators also identified Burmese celadons among finds at other

sites along the coast of the Arabian Peninsula, from Iran and from India.86

285

83 Rumpf, Ambonese Curiosity Cabinet, 273 (book 3, ch. 24).

84 Dijk, Seventeenth-century Burma, 129. A somewhat different version is given by Pieter van Dam, compiled between 1693

and 1701, although thought in principle to be borrowed from the same report by Joris van Coulster; van Dam, Beschri-jving, 2:2, 81–82. See also Volker, Porcelain, 5 n. 7: “and the Chinese annually come with pack animals to the frontier

town of Bomo [Bahmo] carrying along gory [celadon] dishes.”

85 Brown, Ceramics of South-East Asia, 104–6. See also Southeast Asian Ceramics Museum Newsletter 3, 5 (2006): 1. Louise

Cort informed me that some kilns making green-glazed stoneware continue to operate in the Shan States north of

Mongkung today; personal communication, March 2013.

86 Sasaki Tatsuo and Sasaki Hanae, “Southeast Asian Ceramic Trade to the Arabian Gulf in the Islamic Period,” in

Archaeology of the United Arab Emirates, ed. Daniel T. Potts (London: Trident Press, 2003), 254–62; Sasaki Tatsuo,

Kira Fumio, and Sasaki Hanae, “Myanmaa tôji no hakken” (Discovery of Myanmar Green Ware), Bôeki Tôji Kenkyû(Trade Ceramics Studies) 23 (2003): 111–13, 162. Interestingly, a Burmese origin has also been surmised for some celadons

in the Topkapi Saray Museum; Brown, Ceramics of South-East Asia, 102; see also Krahl, Chinese Ceramics in the TopkapiSaray Museum, 232, col. pls. 375–76, under the heading “Southeast Asian Celadons,” labeled as “15th or 16th century.

Perhaps Vietnamese.”

In conclusion, the evidence now available from the kiln sites, from shipwrecks, and from archae-

ological excavations demonstrates that Burmese celadons were not only produced but also exported

in great numbers, predominantly in a western direction, to Indian Ocean countries and beyond. It

therefore seems plausible to assume that the dishes from Martaban described by Katib Çelebi as infe-

rior to the Chinese in fact simply refer to celadons of Burmese origin.

When we return to the sixteenth- and seventeenth-century sources referring to the martaban jars

discussed above, the picture takes form that the writers had in fact a rather precise knowledge about

the origin of the martaban jars in Lower Burma. From the seventeenth-century VOC documents, it

becomes apparent that the term martaban was reserved for the jars of this specific class manufactured

in the Martaban area, where they were produced on a large scale. They were a commodity used by the

Dutch in their inter-Asian trade, and they appear predominantly in the shipment lists from Pegu to

Coromandel, where they were sold or transshipped to other destinations. On special demand, they were

also shipped from Coromandel to Batavia and beyond, mainly for the Company’s own use, because

these jars were highly valued for their excellent properties in the preservation of perishable goods, and

particularly for their special quality in preserving water, gunpowder, and even butter, among other

things (see below). From the sources, it is obvious that they were recognized as a class of their own. A

cross-check is revealing. Significantly, in the seventeenth-century VOC sources, large jars coming

from China, sometimes packed with porcelain, are not called martabans but figure just as large pots.87

To what extent and at what date the name martaban came into popular use for storage jars of ori-

gins other than Burmese is difficult to assess. However, from the evidence available – with the excep-

tion of Ibn Battuta’s testimony – it appears to be a later development. By the early 1820s, Dutch offi-

cials in the East Indies colloquially called all sorts of jars martevanen, in the Dutch version of the term.

The word could be used to denote small earthenware pots of contemporary manufacture available at

the market in Palembang (Sumatra) in 1822,88 or pots listed among the regular exports from Japan.89

The confusion began when scholarly research adopted the name martaban as a generic term for any of

286

87 Volker, Porcelain, 221, notes that the “martavan pots . . . during our period [1602–1682] were nearly all shipped from

Pegu.” However, since he assumes they were only transshipped from that port, he fails to recognize the true impact

of the seventeenth-century documentation he compiled. The sources contain no evidence that the name martaban was

also given to storage jars from other regions; quite the contrary: shipment lists of ceramics loads from China do not

mention martabans but just “large pots,” sometimes packed with finer porcelain; see pp. 197 (a shipment from Macao

in 1637), 198 (a shipment from Fuzhou in 1637), 199 (shipments from Amoy and Fuzhou in 1638), 202 (a shipment

from the China coast in 1646), 215 (a shipment from Macao in 1678).

88 See “Appendix A: Gewone marktprijzen der levensmiddelen te Palembang in 1822,” in Verhandelingen van het Konink -lijk Bataviaasch Genotschap van Kunsten en Wetenschappen 9 (1823): 122: “Aarden kleine martevanen het stuk f. 0. 7 ?.”

89 Hendrik Doeff, Herinneringen uit Japan (Haarlem: François Bohn, 1833), 1:73; Hendrik Doeff, Recollections of Japan,

trans. and annot. Annick M. Doeff (Victoria, B.C.: Trafford, 2003), 43, mentioning martevanen among the regular

exports from Japan. (Doeff was Dutch commissioner in Dejima from 1799 to 1819). See also the interesting remark by

Johannes Fredrik van Overmeer Fisscher, who was first clerk, then package house master in Dejima, in his Bijdragetot de Kennis van het Japansche Rijk (Amsterdam: J. Müller & Co., 1823), 52; he speaks of “martevanen” as a product of

Hizen. Similar is Ernst Christoph Barchewitz, Allerneueste und wahrhafte Ost-Indianische Reise-Beschreibung (Chemnitz:

Stößel, 1730), 539–40; he says of some old “Martafanen” accidentally found on Banda Island that they were made in

Japan.

the older jars, in combination with the erroneous deduction that no such jars were produced in Lower

Burma.90 In the future, the term should be avoided or, alternatively, reserved only for this special class

of Burmese jars.

THE EXPORT OF MARTABAN JARS TO INDIA

From the historical sources of the sixteenth and seventeenth centuries, it becomes clear that the main

direction for the export of martaban jars from the Gulf of Martaban was to India. The backbone of this

commerce was the trade in Indian textiles in return for rice, gold, rubies, and other precious stones

from Burma. The ports of the Coromandel Coast such as Pulicat, Nagapattinam, and, farther north,

Masulipatnam, the port of the Golkonda Sultanate, were among the direct destinations for this trade.91

Crossings of the Bay of Bengal are dictated by the monsoon seasons; the southwest monsoon blows

from mid-May to September, and the northeast monsoon from October to March. The accessibility of

the Burmese ports was thus very limited; during the heights of the monsoons, from June to August

and in October and November, virtually no shipping took place. According to the VOC records, the

preferred months for crossings from Coromandel to Lower Burma were September (the most favored

month), followed by December, February, and May (presumably the earlier part of the month), and

the return voyages took place in December through January and March through April.92 In contrast,

shipping from Coromandel to Batavia appears to have been possible almost year round except for in

October and November, when the northeast monsoon was at its height.93 Valentijn (1726) mentions

the martaban jars among the goods of a typical return load from Pegu to Coromandel. From there, they

could be further distributed in the VOC’s intra-Asian trade and for the VOC’s own use, as well as to

Southeast Asia.94 Similarly, in August 1650, the factors of the English East India Company in Bantam

in western Java wrote to Fort St. George (Madras) and urgently asked, “when it shall please God to

arrive the shipp from Pegu,” to send them as many large martabans “as possible you can and to send

them by the first conveyance fild with rice and wheate or in case you have any at present by you, pray

send us them.” Likewise, in another case in September 1689, the frigate Diamond was sent from Fort

St. George to Syriam to “buy rice and salt pork in Martavan jars.”95

287

90 For instance, Adolf Bernhard Meyer, Alterthümer aus dem Ostindischen Archipel und angrenzenden Gebieten, Königliches

Ethnographisches Museum zu Dresden (Leipzig: A. Naumann & Schroeder, 1884), 4:13–14. See also nn. 43 and 77

above.

91 Ashin Das Gupta, “Indian Merchants and the Trade in the Indian Ocean, c. 1500–1750,” in The Cambridge EconomicHistory of India, vol. 1, c. 1200–c. 1750, ed. T. Raychaudhuri and I. Habib (Cambridge: Cambridge University Press,

1982), 407–33, esp. 413, 431–32; Subrahmanyam, Political Economy, esp. 94–101, 138–39; S. Subrahmanyam, “‘Per-

sianization’ and ‘Mercantilism’: Two Themes in Bay of Bengal History, 1400–1700,” in Commerce and Culture in theBay of Bengal, 1500–1800, ed. Om Prakash and Denys Lombard (Delhi: Manohar, 1999), 47–85; Om Prakash, “Coastal

Burma,” 96–97; S. Subrahmanyam, “And a River Runs Through It: The Mrauk-U Kingdom and Its Bay of Bengal

Context,” in Gommans and Leider, Maritime Frontier, 108–9.

92 Dijk, Seventeenth-century Burma, 71–75.

93 Parthesius, Dutch Ships, 119; Dijk, Seventeenth-century Burma, 72.

94 François Valentijn, Oud an Nieuw Oost-Indien (Dordrecht: van Bram, and Amsterdam: Gerard, 1726), 4:260 (book 5,

ch. 3).

95 D. G. E. Hall, Early English Intercourse with Burma (1587–1743) (London: Longmans, Green, and Co., 1928), 89, 145–48.

Needless to say, the trade goods exported by ship from Burma, among them the martaban jars, also

proceeded to Sri Lanka, the western coast of India, and even farther west. As Roxanna Brown has

already suggested,96 the search for exported Burmese jars should not only consider Southeast Asia, but

should rather be turned to India, as this was the initial direction of their export. This is of particular

interest for the context of the Heidelberg jar because, according to the information about its prove-

nance, it is one of the jars that had been exported to India.

In the illustrations to his Itinerario, Linschoten includes a plate depicting the market in Goa (figs.

13, 14), held there in the mornings from seven to nine o’clock. Goa, where he stayed for more than five

years in the service of the Portuguese archbishop, was certainly the place he knew best, and his claim

that he made the drawings for the market scene “from nature” (polo natural in the Portuguese caption)

is probably reliable.97 In the foreground on the left, two porters are depicted carrying a huge globular

jar. From the bottom up to the shoulder, the jar is enclosed in basketry, a common practice to facili-

tate the transport of large-sized ceramics in Asia. Ropes through the basket handles serve to fasten it

to a thick pole, which the porters shoulder. In all likelihood, this jar is one of those imported from

Martaban that Linschoten describes.98 In fact, its shape compares well to the globular type of the class

of jars discussed here. The jar the porters carry is certainly empty and not filled with water, as has been

suggested.99 A seventeenth-century writer emphasizes that even an empty martaban jar needs two peo-

ple to carry it.100 Even in present-day Burmese potteries, it is still common for two men to carry a dry

pot to the kiln in such a manner (fig. 15).101 A jar of such size, even empty, has a considerable weight.

In a VOC memorandum of 1687, the weights of five different martaban jars are given; they range

between 65.7 and 111.2 kilograms.102 Given the reported and measured volumes of the martabans (see

288

96 Brown, Ceramics of South-East Asia, 104; Gutman, “The Martaban Trade,” 112.

97 Ernst van den Boogaart, Civil and Corrupt Asia (Chicago: The University of Chicago Press, 2003), 62, pl. 5.

98 Linschoten’s text in the Itinerario, 131–32, as well as the Latin text accompanying the plate in the Icones (Boogart, Civiland Corrupt Asia, 62, 64), explains that, at the Goa market, an immense quantity of goods was displayed, imported

from many countries, among which Pegu is mentioned: “An immense quantity of goods are displayed in a vast array

as if on stage, and the unparalleled wealth of Arabia, Armenia, Persia, Cambaia, Bengal, Pegu, Siam, Malacca, Java,

China, etc. is sold there according to a fixed and universal agreement.” The globular shape of the jar is indeed quite

comparable to the group of jars under discussion, even if, with respect to typological details, it might not stand the

critique of archaeological scrutiny. A drawing of archaeological precision was certainly not Linschoten’s objective; in

addition, we have to make allowances for some alterations by the engraver of the plate, Joannes van Doetecum.

99 Boogaart, Civil and Corrupt Asia, 22. See also Friedemann Berger, ed., De Bry, India orientalis, vol. 1 (Leipzig and

Weimar: Kiepenheuer, 1979), explanations to pl. 51, “Der Markt zu Goa,” no. 4: “Diese [referring to the porters]

lassen sich zu allerhand Arbeiten gebrauchen, tragen auch Krüge voll süßen Wassers durch die Stadt feil.”

100 Erasmus Petter, in his journal of Johan van Leenen’s journey from Batavia to Isfahan in 1689/90; here he describes the

water basins for the king’s exotic animals on the Meidan in Isfahan and compares their thickness and weight to the

martabans: “the Martavans, which had kept the drinking water and which, when empty, require two persons to carry

them”; Valentijn, Oud an Nieuw Oost-Indien, 5:254 (book 4, ch. 6).

101 I thank Mick Shippen for the generous permission to publish his photograph, which also appears in Shippen, Tradi-tional Ceramics, 148. See also Adhyatman, Burmese Ceramics, 36, fig. 73, a photograph showing how a jar is carried to

the kiln: fastened with straps from below to a bamboo pole and shouldered by two people.

102 Memorandum of Joachim Nieuwstadt, Batavia, October 1687; van Dam, Beschrijving, vol. 1,1, 531–50, esp. 531. Given

the capacities of these jars, such a jar filled with water would have an additional weight of several hundred kilograms.

below), such a jar filled with water could only serve as a stationary water reservoir. The jar depicted in

the market scene should more likely be interpreted as one of the commodities sold at the market, and

the porters now carry it to its new owner.

Published evidence from India for such Burmese jars is still scarce, although recently, fragments

of Burmese jars are said to have been excavated at sites along the coast of Orissa and Andhra Pradesh.103

Reports of excavations at Manikapatna, a port on the Orissa coast that flourished from the early his-

toric period to the nineteenth century, mention, among an array of other ceramics, fragments of brown-

glazed Burmese ceramics.104 They may very well belong to jars of the class under discussion; unfortu-

nately, no fragments with the characteristic decoration or other special features seem to have been pub-

lished thus far. Among the finds of the Sunchi Reef excavation off Goa, assumed to derive mostly from

an early seventeenth-century Portuguese shipwreck, numerous storage-jar fragments were retrieved.

From the descriptions in the preliminary reports, it seems likely that several of them are from Burmese

jars. So far, only one fragment of these jars has been illustrated. Judging from the color of the clay and

glaze, the way the strap handle was made, and the decoration of applied buttons (though rather large

and made of the same clay as the pot), this fragment appears to stem from a Burmese jar, albeit of a

different type.105

289

Such a weight would definitely be too much for the two porters in Linschoten’s market scene, not to mention the bas-

ketry handles. According to the sources, the largest jars could hold up to two hogshead, an old measurement for vol-

ume that varied according to region, period, and contents. However, an equivalent of the Dutch hogshead (okshoofd)

of about 230 liters may be a reliable measure for estimates; J. R. Bruijn, F. S. Gaastra, and I. Schöffer, Dutch-AsiaticShipping in the 17th and 18th Centuries (The Hague: Martinus Nijhoff, 1987), 218; accordingly, two hogsheads would

amount to about 460 liters; see also n. 117 below.

103 Louise Cort, personal communication, October 2011.

104 Debaraj Pradhan, Pradeep Mohanty, and Jitu Misra, “Manikapatana: An Ancient and Medieval Port on the Coast of

Orissa,” in Archaeology of Orissa, ed. Kishor K. Basa and Pradeep Mohanty (Delhi: Pratibha Prakashan, 2000),

2:473–94, esp. 480. This brown-glazed ware has been identified as Burmese by Louise Cort, personal communication,

12 March 2013. K. S. Behera, “Ancient Ports of Bengal and Orissa,” in India and the Eastern Seas, ed. Alok Tripathi

(Delhi: Agam Kala Prakashan, 2007), 1–20, esp. 13, fig. 1.13.

105 Sila Tripati, A. S. Gaur, Sundaresh, and S. N. Bandodker, “Exploration for Shipwrecks off Sunchi Reef, Goa, West

Coast of India,” World Archaeology 32, 3 (2001): 355–67, esp. 365; Sila Tripati, A. S. Gaur, and Sundaresh, “Explo-

ration of a Portuguese Shipwreck in Goa Waters, Western Coast of India,” Bulletin of the Australasian Institute for Mar-itime Archaeology 30 (2006): 127–36, esp. 133–34; Sila Tripati, “An Overview of Shipwreck Exploration in Goa Waters,”

in Asia-Pacific Regional Conference on Underwater Cultural Heritage Proceedings, Manila, 2011, ed. M. Staniforth et al.,

published online by The Museum of Underwater Archaeology, University of Rhode Island, Kingston, R.I., at

http://www.themua.org/collections/items/show/1233 (accessed February 2015), 1–2, esp. 2 and 4, fig. 2 bottom left.

A fragment that seems similar to this one was found in Kodungallur on the Malabar Coast north of Cochin. Noburo

Karashima, “Dharmadam/Pandalayini-Kollam/Kudungallur/Kollam,” in In Search of Chinese Ceramic-sherds in SouthIndia and Sri Lanka, ed. Noburo Karashima (Tokyo: Taisho University Press, 2004), 44–46, esp. 44–45, pl. 35,1 left,

second from bottom. However, among the finds from the Sunchi Reef are also fragments of Chinese storage jars. See,

for instance, Tripati, “Overview,” 4, fig. 2 (row 3 left) with a Chinese character impressed on the shoulder. On this

fragment, see also Southeast Asian Ceramics Museum Newsletter 2, 7 (Sep.–Oct. 2005): 2, left bottom.

A complete jar of our class, with the characteristic applied stripes and buttons of light-colored clay,

but with an upright rim, is kept in the Historical Mansion Museum in Galle Fort, Sri Lanka.106 Quite

likely, it arrived on board a VOC ship. Shipments of martaban jars, forwarded from Coromandel to

Galle, are reported in VOC sources of the mid-seventeenth century.107 In addition, Ceylon received

regular shipments of rice from Burma.108 It was apparently common practice also to include martabanjars in such shipments. The jars were first lined up in the hold, then the hold and the jars were filled

with rice. In this way, the bulky jars barely took up any extra space and were protected against dam-

age.109

Fragments of at least two Burmese storage jars of this class have been recovered from the Avond-ster (1659) in the harbor of Galle. Apparently they were part of the ship’s equipment. According to the

reports, remains of one such large jar were found in the midship section, an area used for food prepa-

ration.110

Several examples of complete Burmese jars are collected in the Padmanabhapuram Palace in pres-

ent-day Tamil Nadu, close to the border with Kerala. It was once the seat of the rulers of Travancore.

In a room below the dining hall, twelve huge jars are lined up in two rows (fig. 16).111 At least seven of

them closely compare to the Heidelberg jar in shape. They are made of dark reddish-brown clay with

a dark-brown or blackish shiny glaze. They are of the same bulbous shape, with very broad and high

shoulders. Some of the jars also sport the typical decoration of stripes and rows of buttons made of light-

colored clay in low relief under the glaze. Without doubt, these jars belong to the class of Burmese

jars discussed here. They are impressive material evidence for the export of martabans to southern India

in the sixteenth and seventeenth centuries.

290

106 Ewa Jenke, personal communication, February 2012. So far, no further information about the jar could be obtained.

107 Volker, Porcelain, 106–7, for the years 1651 (25 martaban pots to keep gunpowder in) and 1652 (25 large martabans);

178, for the year 1661 (7 large martaban pots); 189, a note of transfer at Malacca in 1678 mentions fifty-five “martavanpots containing each 6.5 Dutch pounds [3.2 kg] of gunpowder.”

108 Dijk, Seventeenth-century Burma, 124, Appendix V, 82–83, in the years 1651, 1652, and 1653.

109 Van Dam, Beschrijving, vol. 1,1, 550; Volker, Porcelain, 201; Dijk, Seventeenth-century Burma, Appendix V, 81. In a sim-

ilar way, pepper was used as a filling between other goods in order to pack everything so tightly “that not even a cat

could creep in between,” making full use of the available space and, at the same time, preventing a shift of cargo;

Valentijn, Oud an Nieuw Oost-Indien, 4:260 (book 5, ch. 3).

110 Robert Parthesius, Karen Millar, and Bill Jeffrey, “Preliminary Report on the Excavation of the 17th-Century Anglo-

Dutch East-Indiaman Avondster in Bay of Galle, Sri Lanka,” The International Journal of Nautical Archaeology 34, 2

(2005): 228–29; Parthesius, Excavation Report, 36. Although, unfortunately, neither the excavation report nor the arti-

fact catalogue make it clear to precisely which jar this refers, it may be inferred that it was jar CE 171, recovered in

forty fragments; Bonke et al., Artefacts Catalogue, 19, fig. 2.14; 47. The matter is further obscured by the indiscrimi-

nate use of the name “Martavan” as a general term for storage jars in this publication.

111 I thank Ramesh Chandran Sreevalsam for permission to publish this photograph. Local lore explains the jars as Chi-

nese jars for pickles. The label in the Ootupura (Dining Hall) of the Padmanabhapuram Palace explains: “The kings

of Travancore were known for their generous hospitality. Over two thousand people were served free meals in this

grand dining hall on a daily basis. Each storey of this two-storeyed building is built to accommodate one thousand

people at a time. The huge Chinese jars which were used to store pickles for the feast are exhibited in [sic] the ground

floor.” On the palace, see Ramu Katakam, Glimpses of Architecture in Kerala: Temples and Palaces (New Delhi: Rupa &

Co., 2006), 191–207.

The export of martaban jars from Lower Burma to India apparently continued into the nineteenth

century, when they were also called “Pegu jars.” However, by at least the 1860s, similar large jars were

also made in Delhi. Interestingly, these jars made in Upper India in the nineteenth century and later

retained the name martaban in the vernacular. Accordingly, the export of such jars from Pegu ceased,

though they continued to be made in Pegu for the domestic market.112

USE OF THE JARS

Ceramic jars have been used for centuries in Asia for all sorts of storage purposes. They could serve as

containers for liquids and preserved foodstuffs, and in this function they were an essential requisite on

long sea voyages. Their perishable contents rarely left a trace in the archaeological record. From finds

in shipwrecks, we know that they were also used to transport small ceramic ware or even glass beads.113

Although in such cases the jars served primarily as containers for finer, high-priced wares, this was at

the same time an economical way to transport the bulky jars, which were also traded as a commodity

in their own right.

Likewise, the martaban jars could serve a multitude of purposes. They were made “in all sorts, both

small and large,” and were used for liquids such as oil, wine, and water; for storage of foodstuffs such

as rice and pickled pork; and for all sorts of goods exported in them from Burma: loaves of benzoin

resin (styrax), lac, wax, green ginger preserved with sugar from Bengal, nipa arrack from the Tenasserim

area, and many others are mentioned in the sources.114 In addition, there was a wide and varied range

of secondary uses for the jars in the different regions of their final destinations.115

291

112 Watt, Indian Art at Delhi 1903, 90, 94–95. See also Scott, Burma, 279. Earlier in the nineteenth century, James Low

noted that, before the war, these jars were exported in large number; Low, “History of Tennasserim,” 50 (the war men-

tioned is presumably the 1824 war).

113 Dupoizat, “Asiatic Stoneware Jars,” 224–25, 229, 248–49, Chinese jars of utilitarian type from the San Diego, which

still contained pork, beef, and chicken bones. Goddio and Casal, Lost at Sea, 34–35 (filled with glass beads).

114 Linschoten, Itinerario, 1:76, 78; Linschoten, Voyage, 1:101, 103; Duarte Barbosa, Livro, 196; the appendix (p. 231) also

lists sugared ginger preserves, but this appendix is known only in the Italian version collated by Ramusio and was

subsequently translated back into Portuguese; see Giovanni Baptista Ramusio, Delle navigationi et viaggi, vol. 1

(Venice: Giunti, 1554), 357v; see also Duarte Barbosa, The Book of Duarte Barbosa, 2:228–29 and commentaries, 217,

and 1:lxxvii-lxxxx; Dijk, Seventeenth-century Burma, Appendix V, 81; XII, 109–10; Hall, English Intercourse, 147; and

the references in n. 45 above.

115 A closer investigation into this topic is beyond the scope of this paper; it may suffice here to hint at the special mean-

ing and importance of large ceramic vessels among the native tribes in Borneo and the Philippines, where several

Burmese jars of our class have been reported. See Rolf B. Roth, Die “Heiligen Töpfe” der Ngaju-Dayak (Zentral-Kali-mantan/Indonesien) (Bonn: Holos, 1992); Brigitte Borell, “Töpfe mit und ohne Drachen,” in Flussaufwärts: Die Bor-neo Sammlung Hilde May, ed. Stefan Dietrich and Margareta Pavaloi (Heidelberg: Völkerkundemuseum vPST, 2013),

173–86. The Burmese jars were reproduced by Chinese potters in Singkawang, West Kalimantan (Borneo), in the

1980s; see Naniek Harkantiningsih, “The Singkawang Dragon Kiln in West Kalimantan, Indonesia,” Arts of Asia20, 1 (Jan.–Feb. 1990): 137–40, fig. 6; in place of a light-colored clay, the appliqué reliefs of dots and stripes are col-

ored yellow before glazing.

In his memorandum dated Batavia, October 1687, Joachim Nieuwstadt states that the pots came

in three sizes: full size, three-quarters size, and half size. The measurements – apparently for a full-

size jar – are given as 134 centimeters in height and 130 centimeters in width.116 If this height was the

standard, this would imply that the Heidelberg jar, being 73 centimeters high, was in the half-size

category. Its capacity, with a volume of 210 liters, would also qualify it as a half-sizer. According to

the sources, large martabans could hold as much as one pipe – the equivalent of two hogsheads, or about

460 liters.117 In comparison, the two large Burmese jars from the San Diego, with heights of 90 and 92

centimeters, have capacities of approximately 280 and 350 liters. Another source mentions a further,

one-eighth size. The sloop Siriangh, arriving at Masulipatnam from Pegu in January 1680, carried on

board 50 full-size martaban jars, as well as 134 one-eighth-size martaban jars; the latter contained a total

load of 12,235.5 pounds of namrack, a black lac. Another special product of outstanding quality from

the Martaban region, it was in great demand in Japan and could be sold there at a high profit.118

The predominant use of the large martaban jars, as emphasized in the sources, was for water stor-

age. The storage of water is essential in regions with a dry season, as on the Southeast Asian mainland

and in India. The practice of using ceramic jars for water storage can still be observed today in many

rural regions of mainland Southeast Asia. In present-day Burma, the modern production of such jars

292

116 Measurements given here have been converted from the Amsterdam pound (494.09 grams) and foot (38.3133 cm),

which were used in the source: Van Dam, Beschrijving, vol. 1,1, 531–50, esp. 531: height = 3.50 feet, width in the mid-

dle (i.e. maximum diameter) = 3.40 feet, width of top and bottom (rim and base diameters) = 1.45 feet. The weights

of the five pots are given as 182, 188, 133, 225, and 150 pounds, respectively. This was important information for load-

ing the ships. Dijk, Seventeenth-century Burma, Appendix V, 81 n. 218: VOC 1432, f.402v. Such a standardization of jar

sizes, which would have facilitated stowing them on board ships, is also known from Thai jars made from the late

fifteenth through the seventeenth century, first in the kilns at Si Satchanalai and later mainly in the Maenam Noi

kilns. They came in three sizes, with heights of about 60, 45, and 30 centimeters. Mukai Kou, “The Study on Brown

Glazed Storage Jars, Exported from Thailand,” Bôeki Tôji Kenkyû (Trade Ceramics Studies) 23 (2003): 90–105, 161 (Eng-

lish summary).

117 Different values are given in the sources, and most of these units of measurement varied. Pyrard de Laval, Voyage,244–45, speaks of a capacity of one pipe or more. Hamilton, New Account, 2:63, says two hogsheads. Bruin et al., Dutch-Asiatic Shipping, 215 and 218, give for the Dutch pipe the equivalent of 435 liters, for the oxhoofd (which corresponds to

the hogshead) the equivalent of 230.4 liters; see also n. 102 above. Taking this as a guideline, the volumes of large jars

would be somewhere between 435 and 460 liters. Valentijn, Oud an Nieuw Oost-Indien, 4:260 (book 5, ch. 3), says that

some of them were capable of holding one or two aam or less (about 155 liters or 310 liters). Linschoten, Itinerario, 76,

even speaks of a two-pipe capacity. Scott, writing in the early twentieth century, says about the Pegu jars that “some

of them are capable of holding 150 gallons” (about 680 liters). Scott, Burma, 279. Crawfurd, Embassy to Ava, 103,

reprinted in SOAS Bulletin of Burma Research 3, 2 (2005): 498, says that some of the “Pegue jars” were so large as to

contain 200 viss of oil – about 182 gallons or 827 liters.

118 Dijk, Seventeenth-century Burma, Appendix V, 81, 82 (for namrack). The source for the one-eighth size is a document in

the VOC archives (VOC 1360, “Missive van de opperhoofden Hendrick van Outhoorn en Pieter Moutmaecker en de

Raet tot Masulipatnam aen haer Eds. tot Batavia, 16 Maert 1680”), f. 1459v to 1460r; Wil O. Dijk, personal commu-

nication, May 2012. I am greatly indebted to Dr. Dijk for this information and for providing me with an excerpt of

the document. This one-eighth size brings to mind a certain smaller variety of Burmese jar of this class with a squat,

globular shape like the one from the Witte Leeuw; Pijl-Ketel, Witte Leeuw, 225, inv. no. 12116 (H. 24 cm); see also Adhy-

atman, Burmese Ceramics, fig. 24.

still serves this purpose.119 It was predominantly in this capacity that the large and bulky jars were

exported as a commodity from Lower Burma to India and beyond. But they were also used for water

storage in insular Southeast Asia. When in the mid-seventeenth century the Banda islands in the

Moluccas experienced a long drought, the Dutch sent martaban pots and handed them out among the

inhabitants for 16 and 20 rijksdaalders; the number of jars sent was far from sufficient, and many more

were demanded.120

In addition, the jars served as containers for freshwater on ships, and some of the huge jars known

from shipwrecks were probably used in this capacity on board.121 On Asian ships, jars were commonly

used in this way, and Europeans soon discovered the advantages of the glazed ceramic jars for this pur-

pose and adopted them. The martaban jars were particularly prized for their qualities of preserving and

even purifying water. Linschoten (1596) describes the water storage on board ships that sailed from

Goa to China, Japan, Bengal, Malacca, and Hormuz. The water for the crew was stored in a wooden

cistern, whereas the captain, pilot, merchants, and other passengers had their water stored in marta-ban pots.122 Obviously the martabans were the better and probably also more costly choice. Pyrard de

Laval emphasized that the properties of the martaban jars were so beneficial that the water stored in

them never went bad. James Low even claimed that the water was purified in the martaban jars.123

Pyrard de Laval adds the interesting observation that the jars could be locked with a key – a quite

understandable measure for securing their precious content, the water, during a long sea voyage.124 A

293

119 After the 2008 cyclone in the delta region of Myanmar, UNICEF distributed several thousand ceramic jars to fami-

lies in the affected region. Each family received two jars, each jar able to hold 227 liters; that is the equivalent of 50

gallons (hence the nickname “50-gallon jar”). The contents of the two jars are said to correspond to around 25 percent

of the water needed by a family in the dry season. Source: http://www.unicef.org/eapro/media_10552.html (accessed

6 December 2011).

120 Volker, Porcelain, 204. See also W. Ph. Coolhaas, ed., Generale Missiven (The Hague: Martinus Nijhoff, 1964), 2:594–95.

In comparison, the prices for large martabans bought in Burma are 11.36 to 11.76 viss/ganza; see Dijk, Seventeenth-cen-tury Burma, Appendix XII, 109–10. This would be less than 2 rijksdaalders, according to the conversion tables in Dijk,

Seventeenth-century Burma, Appendix XVII, 117, 119; that is less than one-eighth or one-tenth of the prices asked in

Banda. The prices for martabans in the list for the years 1638 to 1664 are similar (Appendix I, 7–12), which comes to

an average of 5.50 guilders per jar. That is again less than 2 rijksdaalders (= 6 guilders); however, only a general aver-

age has been calculated here, as we do not know whether the totals were for smaller or larger jars. In any case, the large

martabans were not cheap: in comparison, in 1646 the monthly pay of an “assistant” in the VOC in Burma would have

been 24 viss/ganza = 12 guilders. Dijk, Seventeenth-century Burma, Appendix XIV, 114.

121 Dupoizat, “Asiatic Stoneware Jars,” 225–26, summarizes from seventeenth- and eighteenth-century sources how

highly appreciated the large glazed jars were for this purpose, and how rainwater was collected during the voyage by

means of a cleverly devised system using mats and bamboo tubes to refill the jars.

122 Linschoten, Itinerario, 2:30; Linschoten, Voyage, 1:267–68.

123 Low, “History of Tennasserim,” 50. See also Nicolas Gervaise, Histoire naturelle et politique du royaume de Siam (Paris:

Claude Barbin, 1688), 9, on such a practice in Siam; Nicolas Gervaise, The Natural and Political History of the Kingdomof Siam, trans. and ed. John Villiers (Bangkok: White Lotus, 1989), 14: “The water of the river is extremely clear,

light, and excellent to drink; during the rains it becomes somewhat muddy and then it quite often causes dysentery,

unless care is taken to guard against this by allowing the water to stand in great jars made specially for the purpose,

in which it loses this bad quality.”

124 Pyrard de Laval, Voyage, 244–45; see also n. 51 above. Imported ceramic containers encased with metal bands and pro-

vided with a lid, a suspension hook, and a device to lock the lid are known from Kerala. Margareta Pavaloi, “Gärten

Burmese jar of our class from Mexico, now in the Museum of International Folk Art in Santa Fe, is fur-

nished with such a cover that could be locked (fig. 17); however, when and where this iron cover was

added is unknown.

In fact, the thick-walled martaban jars must have had remarkable properties for insulating against

tropical heat and humidity. The famous sixteenth-century physician Garcia da Orta advocated that a

martaban jar with a high neck and glazed interior be used when shipping the valuable “China root”

from Goa to Portugal to ensure that it arrived in good condition.125 The Dutch even regularly trans-

ported butter in martaban jars from Bengal and from the Coromandel coast to Batavia.126 The jars were

also favored as containers for gunpowder and were used by the Dutch gunpowder mill at Pulicat. Prop-

erly sealed, the gunpowder could be kept for at least ten years without even being stirred. This exam-

ple was followed by the Dutch at Batavia in 1634/35, who even “pried open the ceiling of one of their

storage cellars and installed eighteen martaban jars in order to store their powder in the Pulicat way.”127

By special order from Galle, twenty-five martabans for storing gunpowder were sent from Coroman-

del to Ceylon in 1651.128 William Dampier describes how, during his time as a gunner of the English

fort at Bencouli (Bengkulu, Sumatra) in 1690, he renewed spoiled gunpowder, “which was like

mud.”129 It had already been taken out of the cask and transferred into martaban jars with the inten-

tion to send it to Fort St. George (Madras) to be renewed there. Dampier tried his skills on it, and by

mixing, beating, and sieving it – presumably aided by the special properties of the jars – he succeeded.

William Hunter, who was in Pegu for the East India Company in 1782 and 1783, relates another,

rather exceptional use for the large jars. Since foreigners were not allowed to take their native wives

with them when they left, they would smuggle their wives and children out of the country by hiding

them in the jars.130

294

des Reichtums: Die Inseln unter dem Winde,” in Die Gärten des Islam, ed. Hermann Forkl, Johannes Kalter, Thomas

Leisten, and Margareta Pavaloi (Stuttgart: Hansjörg Mayer, 1993), 194, fig. 295, although the date of the metal addi-

tions remains unknown.

125 Da Orta, Colóquios, 184r: “e o pao que ouuerdes de leuar pera Purtugual seia metido em jarras martabaas decolo alto

porque sam vidradas per dentro essostem muy ho pao sem se danar.” The China root was in that period a much sought-

after medicinal substance, in particular for the curing of syphilis. I thank Sara Teixeira Simºes for help with the trans-

lation from the Portuguese.

126 Dijk, Seventeenth-century Burma, Appendix V, 81.

127 Ibid., 45, 225 n. 172; Appendix V, 81; Appendix XIX, 122–23.

128 Volker, Porcelain, 106.

129 William Dampier, Voyages and Descriptions (London: James Knapton, 1700), 2:298–99.

130 William Hunter, A Concise Account of the Kingdom of Pegu (Calcutta: John Hay, 1785), 65; reprinted in SOAS Bulletinof Burma Research 3, 1 (Spring 2005): 185. Crawfurd, Embassy to Ava, 103; reprinted in SOAS Bulletin of Burma Research3, 2 (Autumn 2005): 498. Their use on fire rafts is another exception. In the 1824 war, such rafts were sent against the

British ships; alongside firewood and other combustibles on the rafts were Pegu jars filled with petroleum, “which

rising in a flame, created a tremendous blaze.” Thomas A. Trant, Two Years in Ava: From May 1824 to May 1826 (Lon-

don: John Murray, 1827), 40. See also G. E. Harvey, History of Burma (London: Longmans, Green, and Co., 1925), 156.

DISTRIBUTION OF THE JARS

From the review of the sources, as well as of the locations of shipwrecks and surviving jars, it becomes

clear that the distribution of the martaban jars was almost global and followed the arteries of maritime

trade routes of the period. In that way, they were brought not only to India and Southeast Asia, but

also to western Asia and beyond. Three of these Burmese jars are exhibited in the collection of the Top-

kapi Saray Museum. As there are no records about the formation of the collection, we can only specu-

late as to how these pieces reached the Topkapi Saray.131

Among Europeans, these jars were in great demand for such essentials as freshwater storage, not

only on board ships, but for regular household use in tropical climates. Their special quality for preser-

vation made them invaluable for life in the tropics. We must bear in mind that, to a great extent, their

distribution was dictated by the needs of Europeans for these jars. On board Portuguese ships, they

arrived at the Portuguese forts on the eastern coast of Africa such as Fort Jesus, Mombasa, and Mozam-

bique. Linschoten says of the island of Mozambique: “They have no freshwater in this island to drink,

but they have to fetch it from the mainland, from a place called by the Portuguese Cabasera [“Caba-

ceira” on Linschoten’s map of Mozambique], and they use in their houses large jars brought over from

India to store the water in them.” 132 In all likelihood, these were the large martabans imported to India

from Burma as described by Linschoten in his later account of Pegu. According to him, many of the

martaban jars were also brought to Portugal.133 In fact, the so-called Pepper Wreck at the mouth of the

Tagus river near Lisbon had quite a number of such Burmese jars on board, as did other homebound

ships that sank in the Atlantic.134

Here, it shall suffice to point out just a few of the more exceptional places where these Burmese

jars have turned up. The jar in the museum in Santa Fe (fig. 17), mentioned earlier, was acquired in

Mexico. A photograph taken around 1910 shows it standing on a street in Mexico City.135 Presumably,

the jar reached Mexico from the Philippines on board one of the Manila galleons, possibly as a fresh-

water container. Its slender, more ovoid shape – although the rim is hidden under the lid – is closely

295

131 Krahl, Chinese Ceramics in the Topkapi Saray, 884, 899, with an illustration of the medium-sized jar (H. 64.5 cm). Illus-

trations of all three jars are in Sakai, “Indoyô no tôji bôeki,” 264, 293–94, figs. 3–5; the largest jar (H. 92 cm), ibid.,

fig. 4, has a bulbous shape like the Heidelberg jar.

132 Linschoten, Itinerario, 1:21; Linschoten, Voyage, 1:28. On the find of such a jar in the Espadarte wreck from 1558 off

Mozambique, see n. 12 above. The jars from Mombasa have already been mentioned; see n. 11 above.

133 Linschoten, Itinerario, 1:76; Linschoten, Voyage, 1:101. See also Pierre du Jarric, Histoire des choses plus memorables adv-enues tant ez Indes Orientales (Bordeaux: S. Millanges, 1614), 3:845 (book 6, ch. 31): “Ici se faisoient les vases ou grands

pots de terre, communément appellez de Martauan, fort estimez par toute l’Inde, à cause qu’ils sont tre-excellens, pour

y tenir & conseruer l’eau, le vin, l’huyle, & autre telles liqueurs. Et en y autoit de si grands, qu’ils contenoient deux

caques, qui sont un demy muy. Tous ces quartiers d’Orient usoient pour la pluspart de ces vases, & mesmes s’en enuoy-

oient en Portugal.” A French Jesuit living in Toulouse and Bordeaux, du Jarric may have taken this directly from Lin-

schoten; in general, he relied on Portuguese and Spanish reports.

134 This may also apply to the Portuguese Nossa Senhora da Luz and the Dutch Mauritius and Witte Leeuw. A complete jar

(H. 61.5 cm), now in the municipal collection at Cascais near Lisbon, survived in a house said to have belonged to a

pilot of the Portuguese East India Company; see Simões, “A Brief Study,” 265–66, fig. 9.

135 Christine Mather, Colonial Frontiers: Art and Life in Spanish New Mexico – The Fred Harvey Collection (Santa Fe, N.M.:

Ancient City Press, 1983), 96, fig. 111; see also 75, fig. 89. Dupoizat, “Asiatic Stoneware Jars,” 232.

comparable to the reconstructed shape of jars found in Japan, from Hakata, Nagasaki, and from Oita

(the latter presumably with a terminus ante quem of 1586), as well as to one of the jars from the WitteLeeuw (1613).136 All these comparisons fall into Sakai’s second group, ranging from the late sixteenth

to the end of the seventeenth century, and such a date clearly applies to the Santa Fe jar as well. Two

of the Topkapi Saray jars also display a similar shape and decoration.137

So far, the most “exotic” location where a Burmese martaban has been found is Ferryland, New-

foundland, on Canada’s east coast (fig. 18). The jar was reconstructed from fragments excavated at a

site believed to date to the second half of the seventeenth century.138 Its elegant shape, with broad shoul-

ders and a strong tapering toward the base, and the decoration, with stripes and dots in light-colored

clay, appear to be consistent with a seventeenth-century date. An unusual feature is the relatively high

and narrow neck with a slightly everted rim.139

296

136 Nan, “Myanmar Trade Ware,” figs. 2–3, 4b, 4d; Pijl-Ketel, Witte Leeuw, 226, inv. no. 12121.

137 Sakai, “Indoyô no tôji bôeki,” figs. 3, 5; Krahl, Chinese Ceramics in the Topkapi Saray, 899, no. 1952.

138 Southeast Asian Ceramics Museum Newsletter 4, 2 (Mar.–Apr. 2007): 3. On Ferryland, see Charles E. Orser Jr., ed., Ency-clopedia of Historical Archaeology (London and New York: Routledge, 2002), 209. For permission to publish the pho-

tograph I thank Barry Gaulton, Archaeology Unit, Memorial University of Newfoundland, Canada.

139 Although necks, when present, are usually very short, somewhat higher necks occasionally occur. See, for instance,

Pijl-Ketel, Witte Leeuw, 226, inv. no. 12121; Adhyatman and Abu Ridho, Tempayan, 177, no. 183; 180, no. 187; and

Nan, “Myanmar Trade Ware,” figs. 3, 6h.

CONCLUDING REMARKS

The jar in Heidelberg’s Ethnological Museum finds its place among examples of a class of Burmese

jars dated to the late sixteenth and early seventeenth centuries. Clearly, these jars, which were pro-

duced from the late fifteenth to the early eighteenth century, are the martaban jars described in the his-

torical sources of that period.

For a long time, these ceramics, manufactured in Lower Burma, were misclassified in ceramics

research alternatively as Chinese, Vietnamese, or Thai. This is linked to a certain extent with the mis-

interpretation of the term martaban, as it was thought that the large storage jars were named after the

port town of Martaban only because they were exported from there, and that they were actually man-

ufactured elsewhere. A fresh investigation of the sources makes it abundantly clear that the writers

were indeed well informed, and their statements that these special jars were made in Martaban are reli-

able. It is paradoxical that, on the one hand, scarcely any other type of Southeast Asian pottery has

received so much attention in contemporary sources as these outstanding jars from Martaban, and that,

on the other hand, ceramics research has for so long failed to realize their full impact.

A review of the sources leads to the conclusion that the sixteenth- and seventeenth-century writ-

ers, when speaking of the martaban jars, refer exclusively to this particular class of Burmese ceramics

made in or near Martaban, where a special clay was available for use in their manufacture. Although

not all writers had local experience, such as Linschoten in Goa or, later, Valentijn and Rumphius in

the Moluccas, they were very well informed through day-to-day exchange with people who did have

that kind of knowledge. Of particular weight, of course, is the account by Joris van Coulster of 1675,

written after more than thirty years of on-site experience in Burma.

Particularly informative is the letter sent from the EIC factory in Banten in Java to Fort St. George

(Madras), praying that the ship from Pegu would arrive safely with the desired jars, and asking that they

be sent from the Coromandel Coast to Java, where they were urgently needed. It appears that the marta-ban jars were outstanding products with very special properties that could not easily be substituted by

any other kind of ceramics. The modern indiscriminate use of the term martaban is therefore not only

unjustified, but also unjust to these remarkable and highly admired works of Burmese craftsmanship.

These jars had special properties that were highly praised in the sources. The main use of these jars

was for the storage of freshwater in normal household use – in those days, an essential requirement in

a tropical climate with a dry season – as well as on board ships. Europeans quickly recognized the advan-

tages of these ceramic jars and adopted them. In addition, the martabans served to store countless other

goods. Their export from Lower Burmese ports was predominantly directed to India, but they had a

wide distribution farther west, as well as to the east. To no small extent the Europeans, first the Por-

tuguese, then the Dutch and others, contributed to the wide and almost global distribution of these

jars because they needed them for their own use.

Although martaban jars were designed as utilitarian ware for regular household use, the potters

nevertheless gave them an attractive appearance in shape, glaze, and decoration. These storage con-

tainers for daily use have a high degree of aesthetic appeal. The great demand for these particular jars,

as indicated in the sources, attests that they were indeed a very special product of Burmese craft, and

not easily rivaled.

297