Navigation – Plan du site

Diplomacy, Trade and Networks: Champa in the Asian Commercial Context (7th-10th Centuries)

Do Truong Giang
p. 59-82


Les études antérieures sur l’histoire du Champa sont principalement basées sur l’hypothèse que le Champa était un royaume « indianisé », profondément influencé par la civilisation indienne. Des études récentes, cependant, ont démontré qu’en dehors de l’influence indienne, le Champa a également fortement bénéficié de l’influence d’autres civilisations, en particulier la Chine et le Moyen-Orient. Cet article, qui se fonde sur un document historique chinois, sur les anciennes inscriptions de l’ancien Champa, ainsi que sur les récentes découvertes archéologiques dans le centre du Vietnam, vise à apporter un nouvel éclairage sur les relations diplomatiques et des liens commerciaux entre le Champa et la Chine et le Moyen-Orient au cours de la période allant du viie au xe siècle. Dans un premier temps seront examinés le commerce et la diplomatie du Champa tributaires des tribunaux chinois ; dans un second temps, l’engagement du Champa dans l’expansion du réseau commercial musulman dans l’océan Indien sera examinée.

Haut de page

Texte intégral

I am deeply indebted to the National University of Singapore (NUS) and to the Southeast Asian Studies Regional Exchange Program (SEASREP) for their generous grants which had enable me to conduct fieldwork in central Vietnam and other parts of Southeast Asia in 2012, 2013 and 2014. I would also like to extend my gratitude to Prof. Bruce Lockhart, Prof. John Miksic, Prof. Michael Feener, Dr. Geoff Wade (Singapore), and Mr. Tran Ky Phuong, Mr. Lam Du Xenh, and Ms Duyen Nguyen (Vietnam) of whom I have benefited greatly from their insightful comments and suggestions. The author would also like to thank the anonymous reviewers for their constructive comments, which helped me to improve the manuscript.

1Previous studies of Champa history mainly based on an assumption that Champa was an “Indianized” kingdom who profoundly influenced by Indian civilization. Recent studies, however, have convinced that apart from Indian influence, Champa did benefit greatly from other civilization centers, especially China and the Middle East. This article, relies on Chinese historical document and the old inscriptions of ancient Champa as well as recent archaeological findings in central Vietnam, aims to shed new light on the diplomatic relations and commercial links between Champa with China and Middle East during the period from the 7th to the 10th: Firstly, the tributary trade and diplomacy from Champa to Chinese courts will be taken into account; and secondly, the engagement of Champa in the expansion of Muslim trade network in the Indian Ocean will be investigated.

Maritime Trade Networks in Asia through the Tang Era

2The formation of the Great Tang Empire in 618 CE marked an important milestone not only in the history of China, but also of the Eurasian continent at large. The Tang Empire is recognized as one of the most flourishing dynasties in the thousands of years of Chinese history. The political stability, the cultural bourgeoning and the economic prosperity during the Tang had all made China a massive and desired market with which all the countries in Eurasia wished to establish the diplo-political relations, cultural exchanges and trade, using both maritime and overland routes.

3From the 8th century onward, however, the overland Silk Road was affected by increasing acts of pillage, and massacre of the merchants in the Central Asia. By this time, the rise of Arab merchants who worked to take over the trading networks also contributed to the break in trade between Tang China with the Mediterranean world via the overland Silk Road. This break resulted in both negative and positive consequences. On the one hand, the decline of trade via the overland Silk Road weakened the relationship between China and its traditional markets in Central and West Asia as well as in Mediterranean, depleting the fiscal revenues of the Empire, and consequently led to political and economic disorder in China. On the other hand, the stagnation in overland trade activities opened up new prospects for strengthening and increasing trade with potential markets in the south through maritime routes. From a reluctant substitute for the overland Silk Road, the maritime Silk Road gradually became the principal trading route linking the enormous Chinese market with other major markets of Eurasia.

4The rise of Arab world from the 7th century played a significant role in shaping, operating and developing both the overland and maritime trade networks of Eurasia in the following centuries. Before the sixth century, while Indian merchants were regarded as the master of the trade activities between South Asia and Southeast Asia, Arab traders played only a limited role in the operation of regional trade networks in West Indian Ocean. From the sixth century onwards, Arab traders, thanks to innovations in science and technology, especially in marine engineering, gradually broke down the traditional Indian networks, and finally established themselves in the maritime trade activities in the whole regions of vast Indian Ocean and South China Sea. From the 7th century, the Arab vessels started to sail directly further north, cross the South China Sea, to conduct direct exchanges at the coastal ports of Champa and South China, carrying along with them many valuable Western Asia-origin items such as glass, ceramics, and jewelry.

5From the 8th century CE onwards, the role of Arab traders, the stability of the Tang Empire as well as the innovational advancements in terms of navigation all led to major changes in the commercial system of Eurasia in which the maritime trade became increasingly significant to most countries in Asia. Seas and Oceans served as avenues for civilizational diffusion, cultural exchanges and trade activities. The connections between major civilization centers, in spite of the geographical distance, become easier and more frequent. With the increase in the maritime trade activities throughout the Indian Ocean and western Pacific Ocean, the coastal polities of Southeast Asia enjoyed new favorable conditions for economic take off, political stability and territorial consolidation.

6Champa, with its long coastline and because of its location in the middle of the safest maritime route connecting China with Southern Sea countries, rapidly engaged with the regional maritime trade networks and took full advantage of the new circumstance to expand its influences over other polities in the region. During the Tang period, Champa was not only the most active mandala in the Southern Sea area sending tributary and diplomatic missions to Tang China, but also emerged as one of the most important centers for Arab traders who, by the time, controlled and operated the maritime trade networks in Asia.

Tribute and Trade from Champa to Tang Court

  • 1 Jiu Tang Shu 197, cited from Maspero (1928: 85-87).
  • 2 Maspero held the idea that this mission was sent by king Sambhuvarman, the son of Rudravarman and a (...)

7Soon after the establishment of Tang Empire in 618, the Tang economy bloomed and then reached its peak during the period from 623 to 684 due to the increase of population, the expansion of cultivated land and new volume of trade (Wang 1998: 72). In 623, Lin-yi and “Chenla” were among the first polities from the south sea to send tributary missions to the Tang court in order to establish the diplomatic relationship with the newly established giant in the north.1 This mission is said to have been sent by king Fan Fanzhi of Linyi who had established a new capital after the invasion of Liu Fang in 605 CE.2 In the following years, Linyi kept sending tribute missions to China - 10 missions were reported between 625 and 670 (Wang 1998: 82, n. 11). In 672, Yijing sailed from China to India on ships belonging to the ruler of Srivijaya.

8In the late 680s, trade in the Nanhai re-emerged after a decade of decay due to the growth of corruption in Guangzhou, the central trading port of Tang China as well as Nanhai trade system. Linyi was again the first polity sending a diplomatic mission to the Tang court in 686. In following years, the tribute missions from Linyi, along with those from Srivijaya, continued to visit regularly the Tang Court even during the decay period after the An Lu Shan revolt in 755 (Wang 1998: 75). Champa sent tribute missions in 691, 695, 699, 702, 703, 706, 707, 709, 711, 712, 713 and 731 (Maspero 1928: 93, n. 1) and another mission in 749 (Wang 1998: 84, n. 38, Maspero 1928: 93 n. 3). In these tribute missions, elephants were the most popular tribute objects from Linyi to the Tang court (Maspero 1928).

9The rebellion of An Lushan in 755 caused the fall of the great market-cities of Changan and Luoyang, and as a result, the decline of Nanhai trade as a system. From the middle of the 8th century until the end of the 9th century, Arab and Persian merchants became important middlemen in the trade of maritime Asia. The Muslim merchant in Canton could even revolt and led to the short-term decline of the trade in the region. Consequently, during the period from 758 to 878, there were only few tribute missions from the South sent to the Tang court (Wang 1998: 79, 86 n. 66).

10From 758 onwards, the Chinese historical records stopped using the term “Linyi” to indicate the Cham polity in central Vietnam and started to call it “Huanwang” instead (Maspero 1928: 95). The precise reason behind this change of name of Champa in Chinese records is unknown. Previously, Champa scholars followed Maspero’s study, thus believing that by this time the political center of “Champa” had moved to the south, from Trà Kiệu in Amaravati to Kauthara and Panduranga (ibid.). This view, however, has been questioned by recent scholars who assume that there was in fact no “shift” of capital of Champa during this period since Champa was not a unified state but a confederation of small kingdoms. William Southworth has made an attempt to explain this change from the economic perspective and suggests that:

[…] the collapse of the early Tang trade and the demise of Guangzhou during the mid 8th century A.D. also destroyed the commercial dominance of the Thu Bon valley system, and when the South China Sea trade was transferred to the ports of the Red river delta during the late 8th century A.D., an entirely new pattern of trade emerged. The financial importance of Hanoi, and the emergence of a burgeoning maritime commerce with Java, encouraged a major shift in the displacement of trade wealth to south central Vietnam, in particular to the ports of Nha Trang and Phanrang, on the sea-route from Java to northern Vietnam. The independent states of Kauthara and Panduranga, which flourished here during the late 8th and early 9th centuries AD, became of central importance to the continuing history of Campa during this period. (Southworth 2001: 318)

11Only one tribute mission of rhinoceros and buffalo from Huanwang was recorded in 793 (Maspero 1928: 104-105). In the meantime, Champa suffered from naval attacks by Malay and Javanese in 774 and 787 respectively. Soon after that, the Chinese record that Huanwang attacked the Chinese Protectorate of Annam twice in 803 and 809. These events provide strong evidence showing Champa’s ambition in competing with its northern rivalry ports in order to gain control over the maritime trade in the region (Wang 1998: 90). After the attack of Huanwang in 809, Chinese records cease to mention “Huanwang”.

12As observed from tribute missions sent by Linyi and Huanwang to China during the Tang era, it was apparently that Champa was a major source of important commodities for China market and in return the maritime trade provided the financial and commercial basis for the great epigraphic and sculptural traditions of the 7th and early 8th centuries CE in Champa (Southworth 2001: 318). Champa’s trade with Tang Empire was mainly in its local products including ivory, rhinoceros horns, gharu-woods, tortoise-shell, amber and manufactured articles of gold and silver. In return, Champa received great amount of goods from China, especially the trading ceramics as presented below. The active engagement of Champa into the tributary system of Tang Empire was among the major factors leading to the prosperity of the trade in South China Sea during the Tang era (Wang 1998: 89).

The Expansion of Muslim networks in Maritime Asia and Champa’s engagement

13The rise of Arab world from the 7th century CE had a significant implication in shaping, operating and developing both overland and maritime trade networks of Eurasia in the following centuries. Starting from the initial base in Konkan and Gujarat, Persians and Arabs gradually expanded their territory to the east and dominated the maritime trade routes along the coast of Indian Ocean by the 9th century, turning the Indian Ocean into an “Arab Mediterranean” (Wink 1990: 65, 1997: 1). Arab traders then expanded their network to Southeast Asia and southern China by the 10th century (Wink 1997 : 1, Salmon 2004). By the late 12th century, the maritime trade in South China Sea was basically in the hands of Muslim and Tamil traders (Wade 2009: 234).

14As the controllers of maritime trade in the South China Sea, Arab traders and travellers appeared to have rich knowledge of the mercantile ports along the maritime trade routes of the region. In the 9th century, Ibn Khordadbeh recorded the trade ports at the furthest extension from the Arab world as Luqin (likely Loukin, situated in what is today Vietnam), Khanfu (Guangzhou/Canton), Khanju (Quanzhou), with Qansu (Yangzhou) marking the end of the maritime route (Wade 2010: 367-368). With the emergence of Arab networks in the region, the Cham soon recognized the opportunity and actively participated into the regional maritime trade networks. Champa ports, especially the ports in Amaravati region became a rival of the ports in northern Vietnam, and were seen as regularly destinations for international trading vessels. Sanf – the term used by Arab traders designating both the country of Champa as well as its mercantile port, was repeatedly mentioned in various travelling records by Ya’Qūbī (9th century), Mas‘udi (10th century), Ibn al-Nadīm (988 CE), Aja’ib al-Hind (c.1000 CE), the Mukhtasar al-Aja’ib (c.1000 CE) and Idrisi (mid-12th century) (Tibbetts 1979, Ferrand 1913-1914).

15Although the relationship between Champa and the Muslim network since the 9th century is quite clear, the issue of when exactly Islam penetrated into Champa still remains debatable and has not yet been studied comprehensively. Rie Nakamura assumes that there were apparently two waves of contacts between Champa with the Muslim world. The first wave, according to her, probably took place no later than the 9th century and featured the key role of the Persians, Arab, Indian and Chinese Muslims. The second wave of contacts occurred during the sixteenth and seventeenth centuries when the Malay Muslim trade was at its height in the regional waters (Nakamura 2000: 58). During the first period of contacts, individuals of Champa converted to Islam, however, Rie suggests, “these conversions would have been rather isolated instances, limited to commercial circles and the court which had relationships with foreign communities.” (Ibid.: 59.)

  • 3 In the middle of the 10th century, the Arab Mas΄ūdī recorded Arab trade with Southeast Asia and Ch (...)

16In History of the Five Dynasties, Chams are said to have the same customs as “those of the Arab [TA-SHIH].” The Envoy sent by Cham ruler to the Zhou court in 958, for example, presented rose water, flasks of ‘Greek fire’, and precious stones. This envoy was led by P’u Ho-san (probably Abu Hasan in Arab). P’u Ho-san again led another envoy of Champa to the newly established Song court where he offered ivory, camphor, peacocks, and twenty “Ta-shih [Arab] vases” (Shafer 1967: 75). In 987, Zhao Rugua noted that the Cham polity of Panduranga brought tribute to Song court with the company of Ta-shi people/merchants/officials (Hirth & Rockhill 1967: 51). In 986, a member of the Pu clan – Pu Luo-e (Abu Nur), led his family and people to seek refugee in Hainan to escape the pressure from Jiaozhou (Wade 2011). Two years later, Hu-xuan (Husayn), another Muslim from Champa led 300 members to Guanzhou (375-376).3 The migration of these people from Champa might be associated with the historical event occurring in Champa in the late tenth century and was recorded in both Vietnamese and Arabic materials. Ibn Al-Nadim recorded in 988 that:

[…] at the time when I was in India the king who ruled in Sanf was called Lajin. The monk of Najran told me that at that time the king was a king called the King of Luqin who had invaded Sanf, ravaged it and enslaved all the inhabitants. (Tibbetts 1979: 41-42)

17Vietnamese historical records also mention a noble, known as Lưu Kỳ Tông, who fled to Champa and then established himself as king of Champa by end of the 10th century.

18The earliest Arab accounts of Champa are included in the Kitab al-masalik wa’l-mamalik of Ibn Khurdadhbih and in the Akhbar al-Sin wa’l-Hind, which dated to around 850 CE (Tibbetts 1979: 25-29). The first of these texts includes a section describing the maritime route from Persian coast to the East. A journey of three days from the Khmer realm (Qmār) to Champa (Sanf) before sailing to Annam (Luqin) is described as follow:

  • 4 Pulau Tioman.
  • 5 The Khmer polity, likely around the Mekong delta.
  • 6 William Southworth suggests that the Cham port used on this journey was probably Nhatrang (Southwo (...)

[…] On leaving Mā’it, one find to the left, the island of Tiyūma4 […] From there one goes in five days to Qmār5 […] From Qmār to Sanf is three days following the coast. The aloeswood of Sanf, known under the name Sanfi, is superior to that of Qmar…From Sanf6 to Lūqīn which is the first step of China, is a hundred parasangs either by land or sea. (Tibbetts 1979: 28-29)

  • 7 Kadrang = either Pulao Condore or Panduranga/Pandran in Panran.

19Akhbar al-Sin wa’l-Hind describes an itinerary of ten days from Kadrang7 to Sanf, where they collected fresh water before sailing further north via Sundur Fulat (the island of Hainan) to Guangzhou (Canton):

  • 8 Pulau Tioman.
  • 9 Tibbetts suggests various locations and equivalents for Kadrang (Tibbetts 1979: 157-59). Ferrand s (...)
  • 10 Southworth suggests this journey probably stopped in Quảng Nam Province (2004).
  • 11 The island of Hai-nan.

Then the ships travel to a place called Tiyūma,8 which has fresh water for anyone who desires it, and the distance to it is ten days. Then the ships set sail for a place called Kadrang and arrive there after ten days. There one also finds fresh water. Such are the islands of India, wherever you find springs you will find fresh water in them. At Kadrang9 is a high mountain where slaves and robbers sometimes hide. The ships next reach a place called Sanf10 situated at a distance of ten days; there they find also fresh water, and from this place is exported a kind of aloeswood called Sanfī. This place is a kingdom. The inhabitants are brown and each one of them wears two fūtas. When the ships have taken fresh water they set sail for a place named Sundur Fūlāt,11 which is an island in the sea, ten days distance away [from Sanf]. Here also fresh water is found. Then they set sail as far as sea called Sankhai, as far as the ‘Gates of China’. (Tibbetts 1979: 26-27)

Fig.1. Place names mentioned in Ibn Khurdadhbih (mid-9th cent CE)

Fig.1. Place names mentioned in Ibn Khurdadhbih (mid-9th cent CE)

Source: Tibbetts (1979).

20In the Chinese travelling accounts, Champa ports also depicted as important spots for long-distance ships to anchor while sailing from South China ports to the Southern Seas. In about 800 CE, Jia Dan (730-805) wrote a book with a range of itineraries describing major maritime trade route from China to the Persian Gulf. Although Jia Dan’s book is now lost, but this maritime route was recorded in the Xin Tang Shi. As could be seen below, the importance of several ports along Champa coast were featured:

  • 12 Xin Tang shu, juan 43C, 18b, cited from Wang (1998: 98). For further details, see Pelliot (1904).

Sailing from Canton, it is 200 li to the south-east, to T’un-men Shan [an anchorage south of Pao-an and opposite the north coast of Lantao Island], and two days towards the west to Chiuchou Shih [Ch’I islands or Taya Islands, north-east point of Hainan], and two days towards the south to Hsiang Shih [Tu-chu-shen or Tinhosa Island, or even further south], and three days towards the south-west to Chan-pu-lau Shan [Culao Cham], which is in the sea 200 li east of Huanwang, and two days towards the south to Ling Shan and one day [towards the south] to Men-tu Kuo and one more day [towards the south] to Ku-ta Kuo and half a day [towards the south] to Pen-t’o-land Chou and two days [towards the south-east] to Chun-t’u-nung Shanand five days [further to the south-east] to a Strait, 100 li from north to south, which the barbarians call che [Straits of Malacca], and three days sailing westwards out of the Straits to Ko-Ko-seng-ti Kuo, many of whose people were devoted to robbery and plunder, and were much feared by those who travel in ships and four or five days [towards the north-west?] to Sheng-teng Chou [east coast of Sumatra, near Deli?], and five days towards the west to P’o-lu Kuo [north-west Sumatra, Perlak or Perkir or Baros, the P’o-lu-shih of I-ching] and six days [towards the west or north-west] to Ch’ieh-lan Chou [one of the Niclobar Islands].12

21Following the suggestion of Chia Tan’s maritime route from South China to the Straits of Malacca and compare with the availability of both inscriptions and archaeological evidences along the coastal central Vietnam, we could to some extent identify important mercantile ports of Champa during the 9th century CE. The maritime route originated from Canton – the most important port of China for centuries, especially during the Tang era. From Canton, it took several days to reach the first position on the coast of Champa namely Chan-pu-lau Shan. Chia Tan suggested that Chan-pu-lau Shan is “in the sea” and its exact location is “200 li east of Huanwang.”The Chan-pu-lao Shan, thus, could be easily identified as the Cù Lao Chàm Island, off the coast of modern Hội An port city. Huanwang, as Chinese accounts and archaeological evidence show, is undoubtedly the Trà Kiệu Citadel where served as the political center of nagara Amaravati in the 9th century. The Newer Tang History provides another account of Champa in the name of Huanwang:

Huanwang, originally Linyi, is called Zhanbulao and also Zhanpo. It is situated 3,000 li by sea south of Fuzhou. Its land area east to west is more than 300 li and 100 li north to south. It is bounded to the west by Mount Wuwen of Zhenla and is bordered on the south by Benlangtou zhou. (Goble 2014: 9)

22Leaving Cù Lao Chàm, it took two days going to the south to reach Ling Shan. While Zhan-pu-lao Shan is generally accepted by scholars as Cù Lao Chàm Island, the term “Ling Shan”, however, has called for discussion among scholars. The Ling Shan or “Ling Mountain” has conventionally been associated with Cap Varella, near the Cả pass in Phú Yên province but the course of this itinerary suggests that Ling Shan in Chia Tan’s record must be located at between Cù Lao Chàm Island and Kauthara/Nha Trang. Southworth suggests  that “Ling Shan” could be identified with the Cù Lao Re island but given the fact that the distance between these two islands is not too far and it take less than one day by sail from this island to one another (2000: 242). Wang Gungwu speculates that Ling Shan was possibly Cape Sa-hoi [Sa Huynh], in the south of Quảng Ngãi province (Wang 1998: 98). The new discoveries in Bình Định province, however, suggest that Ling Shan should be associated with a position around the modern Quy Nhon city, Bình Định province. It is interesting that there is still a pagoda in the name of Ling Shan located at the top of the mountain in Phuong Mai peninsula, which lies between the Thị Nại bay and the eastern Sea/South China Sea. A Champa statue with few Sanskrit words carved on its back is still visible to researchers and thus suggest that this position once was a temple during Champa period. It should be noted that Thị Nại gulf, just next to the Ling Shan Pagoda, is a very good water area for huge trading vessels. Nagara Vijaya, where the Thị Nại port and the Ling Shan temple located, was emerging as a rival of nagara Amaravati in the north by 9th century, and thus it is possible that this area was on the maritime itinerary of the Chinese. (Note that Jia Dan’s account refers to several different “guo”, or countries, indicating separate Champa polities.) The two-day journey from Cù Lao Chàm to Ling Shan in Bình Định province is also acceptable.

  • 13 Previous scholars, like those as Pelliot and Wang Gungwu placed Men-tu in Bình Định province, see P (...)
  • 14 Momoki Shiro also believes that Mendu was in Phú Yên province, not in Bình Định. See Momoki (1998).

23According to Chia Tan, from Ling Shan, it took one day to go to Men-tu Kuo. Previous studies tend to place Men-tu in Bình Định province.13 But I believe that Men-tu was in fact located at the estuary of the Ba/Darang River in present-day Phú Yên province. The presence of a complex of citadel, inscriptions and sculptures in this area during the 8th and 9th centuries CE would confirm this interpretation.14 And then, “one more day [from Men-tu Kuo] to Kuda Guo.” Kuda Guo is largely identified with nagara Kauthara of Champa where there is a famous complex of Po Nagar temples locates. There is one, inscription there date to 8th century (C38A, B, C, and D1), and two another inscriptions found there in 9th century (C37 and C31 C2). From Kuda Guo it took “half a day [towards the south] to Pen-to-land Chou” [Panduranga] where is generally associated with the port in Panran, in present-day Ninh Thuận province. In the 8th century, there are two inscriptions from this region, i.e. C217 and C216. In ninth century, there are much more inscriptions were found in this region, including those of C24, C25, C23, C19 and C14.

Fig. 2. Maritime itinerary in Southeast Asia during the Tang era

Fig. 2. Maritime itinerary in Southeast Asia during the Tang era

Source: Wang (1998: 97).

Nagara Amaravati and the Cross-Cultural Exchange through the Tang Era

24Throughout the history, Champa maintained the situation of simultaneous existence of several different nagaras/polities based on their ability to control the trade networks along the major rivers. The history of Champa noted the existence of major powerful centers/nagaras including the region north of Hai Van Pass (from present-day Hue Province to Quảng Bình province), nagara Amaravati (from Da Nang to Quảng Ngãi province), nagara Vijaya (including Bình Định and Phú Yên provinces), nagara Kauthara in Khánh Hòa province and nagara Panduranga (in Ninh Thuận and Binh Thuận provinces). These centers of power had existed simultaneously, competed and allied with each other. The topographical features, the lack of large agricultural communities, the shortage of manpower, as well as the absence of a centralized administrative tradition all contributed to the failure of Cham in building a centralized state like the Vietnamese and the Khmer had done successfully. The history of Champa, however, witnessed the emergence of certain nagaras in some period of time who could control administratively and economically other nagaras and put other territories under their direct control. Among them, nagara Amaravati had served as a dominant center of Champa civilization where the Cham built two large sanctuaries at My Son and Đồng Dương, the capital city in Trà Kiệu and a system of commercial ports in Đà Nẵng, Hội An and Cù Lao Chàm island.

  • 15 K.53, Kdei Ang inscription.
  • 16 Inscription C81 in My Son (between 709-717), see Golzio (2004: 25).
  • 17 In inscription C95 in Mỹ Sơn (d.1056/57); inscription C94 in Mỹ Sơn (d.11th century); in inscriptio (...)

25The inscription C96 in My Son (d.658 CE) provides information for the first time about the name of Campapura that was closely associated with king Prakasadharma. A few years later, the Khmer inscriptions also mentions its eastern neighbor under the name Campesvara (“Lord of Champa”).15 The appearance of name Campapura along with the term visaya referring to an administrative unit (probably means “district”) suggest that, by middle of seventh century, an administrative system had been initially established in Champa though we do not know exactly about its geographical scale and effectiveness. It is worth noting that during the reign of king Vikrantavarman the term rajadhiraja had appeared for the first time referring to “king of kings”.16 The evidence from archaeology, inscriptions and texts all tend to demonstrate that Trà Kiệu played a key role as political center of nagara Campapura from the seventh century. Trà Kiệu citadel probably served as the capital of Campapura where the Champa royal family and other famous family lines residing as well as a site to undertake the royal ritual ceremonies. Trà Kiệu citadel or Siṅhapura in Sanskrit/Cham inscriptions continued to appear in inscriptions in later period, especially during the eleventh and twelfth centuries.17

26From this time onwards, the name of Campapura, Campa or Campapuri began to appear regularly in Champa inscriptions. The kings of Champa called Campadesa while the place where he resides called Campapura/city or citadel of Champa that is Siṅhapura/Trà Kiệu since the 7th century. The inscriptions by this period refer repeatedly to Valkimi and Indian epic, and this suggests that the rulers of Champa had begun to apply the traditional Indian political and administrative models in Champa. The king of Siṅhapura (or of nagara Amaravati) was the king of kings [of entire nagara Champa], the gods in Mỹ Sơn sanctuary were the gods of all territories of Champa and Siṅhapura citadel was the capital of mandala Champa as a whole.

27The period from 9th to 12th centuries could be seen as the most prosperous period in the entire history of the nagara Amaravati both economically and politically. Firstly, it was during this period that the Đồng Dương sanctuary was formed and, along with the near-by Mỹ Sơn sanctuary, served as the primary religious center of entire mandala Champa. The prosperity of the Đồng Dương dynasty had also given way for the birth of a brilliant artistic style in the art history of Champa namely the Đồng Dương art style. Secondly, the numbers of inscriptions in the nagara Amaravati increased significantly. It’s worth noting that, Amaravati and Kauthara were the two centers of inscription production of Champa during this period. The content of inscriptions provides invaluable information for our awareness on various issues of Champa history including the political organization, religion, society and economic activities. Finally, the foreign trade and mercantile ports of Amaravati flourished during these centuries as will be presented below. Economic prosperity laid the foundation for the political stability and development of Đồng Dương dynasty, the economic basis for the construction of Đồng Dương and Mỹ Sơn sanctuaries as well as the proliferation of Champa inscriptions during this period.

  • 18 Suchandra Ghost. “Ministers and Nobles in the kingdom of Campa: looking through the epigraphic lens (...)

28In the 10th century, the inscriptions show that mandala Champa had deeply integrated with the regional maritime trade and had broadened its diplomatic relationship with various countries in the region. The Nhan Bieu inscription in present-day Quảng Trị province (d.911/912), for example, furnishes a detailed account of Pov Klun Rajadvarah, the son of the queen’s cousin, who had become a favorite of king Jayasimhavarman and was sent on a diplomatic mission to Java. Pilih Rajadvarah, the son of the cousin of Jaya Simhavarman’s queen, continued to occupy a high position under Bhadravarman and was again sent on a diplomatic mission to Java. The Bo Mung inscription in Quảng Nam province (d.890) also provides information about a diplomatic mission sent to Java (Majumadar 1927: 62, Golzio 2004: 73-76). In the meantime, the Bang An inscription refers to the multitude of royal ambassadors coming from different countries. It’s thus no doubt that by the tenth century, mandala Champa in general and nagara Amaravati in particular were internationally recognized as an important and powerful polity and actively took part in the international politics.18

  • 19 Ibid.

29The political dynamics and economic development of Champa in the tenth century were much associated with the emergence of new elite families who maintained and benefited from close relationship with the royal court and consequently took control of the economic activities of the country. These new nobles appeared in Champa inscriptions bearing the title “Sarthavaha.”19 The Hóa Quê inscription, for example, refers to an important collateral branch of the royal family which supplied high officials to the kingdom. The founder of this family, the original Sarthavaha, was a nephew of Rudravarman II, the first king of the dynasty, and a brother of the principal queen of Indravarman II. His three sons Ajna Mahasamanta, Ajna Narendra nrpavitra and Ajna Jayendrapati occupied high position in the kingdom as ministers of king Bhadravarman. These three brothers, together with a younger sister, called Ugradevi, erected an image of Maha Rudradeva, “out of devotion to, and in imitation of the features of their father”. Other religious endowments of the members of the family are also referred to in the same inscription (Majumadar 1927: 64). The noble who associated with the sarthayaha title is often attributed to those were merchants or traders who had sufficient knowledge of the different foreign countries and those were in charge of loading and unloading of commodities (Prakash Charan Prasad 2011: 118).

Cross-Cultural Exchange in Hội An – Cù Lao Chàm Island

30Recent archaeological excavations in Đà Nẵng and Quảng Nam provinces have unearthed a huge amount of trading ceramics originated from various regions of Asia and thus intensely affirmed the economic prosperity of this region in the past. Archaeological finds reveal that the economic activities of nagara Amaravati had undertaken mostly at the northern bank of Thu Bon River. The Estuary of Thu Bon river was recorded in Chinese accounts as the economic center of Huanwang kingdom—the Linyi Town (V: Lâm Ấp Phố). The Thu Bon river, as Tran Quoc Vuong and Tran Ky Phuong have both suggested, served as the nexus of economic activities of nagara Amaravati that connecting several economic zones stretching from the western mountains to the eastern river estuary (Trần Kỳ Phương 2008, Trần Quốc Vượng 1995). At the Hau Xa 1 site in the north side of Thu Bon estuary, archaeologists have excavated many Chinese ceramics sherds of the Tang and Song periods, as well as the Islamic ceramics sherds dating back to the 9th-11th centuries (Kikuchi 2010: 72). Also locate at the north side of Thu Bon estuary, the sites of Hậu Xá II, Trang Soi and Bàu Đà have unearthed many Chinese ceramic sherds from Yuezhou kilns in South China in 10th century (ibid.: 73). Based on these archaeological findings, Seiichi Kikuchi suggests that the northern bank of Thu Bồn river had become a favorable location for commodity exchanges of merchant ships in the East-West maritime trade route and thus the Lâm Ấp Phố in Chinese records could be identified with this area (Kikuchi 2010:81, Trần Quốc Vượng 1998: 443).

Fig. 3. The 9th-10th centuries Changsha and Yuezhou ceramics on display at Hội An Trading Ceramic Museum

Fig. 3. The 9th-10th centuries Changsha and Yuezhou ceramics on display at Hội An Trading Ceramic Museum

©Do Truong Giang

31In the south of Thu Bồn estuary, archaeologists have found both ceramic artifacts and ancient coins. Tran Quoc Vuong and his colleagues had discovered many Chinese and Vietnamese ceramics and coins of the 9th-12th centuries in the Trung Phuong area (Trần Quốc Vượng 1995: 444). In 2012 and 2013 respectively, the author of this current study carried out surveys in southern bank of Thu Bồn estuary in searching for Chinese trading ceramics but the results were not fruitful. Perhaps, we have to wait until more thorough archaeological excavations being undertaken to have a firm conclusion about the economic activities of the Cham in this area of south Thu Bồn River.

32While the Thu Bồn estuary served as the primary economic zone of Amaravati, the evidences from archaeology suggest that there were several other nearby places had also played important part in exchanging commodities with foreign merchants. The maritime voyages of both Chinese and Arab merchants noting that Cù Lao Chàm, an island off the Thu Bồn Estuary, was an important location for the seafaring sailors because they could obtain the fresh water, firewood, foods for their long-day journeys. The archaeological study of Cù Lao Chàm was kicked off in 1992 with a large-scale survey conducted by a joint group of Vietnamese and Japanese scientists. During the period from 1997 to 2000, the local archaeologists in conjunction with those from Vietnam National University had conducted detailed surveys and archaeological excavations in which the 1999 excavation was the largest mission. These archaeological studies in Cù Lao Chàm have important implications for the study of the commercial history of Champa in general, and of the nagara Amaravati in particular, as they provide physical evidence of the presence of a large volume of trading artifacts those originated from China, Southeast Asia, India and Middle East.

33In the 1999 excavation, most of the Chinese ceramics were found dating from the 7th to the 10th centuries CE and were literarily classified as “Tang Pottery” by Vietnamese archaeologists. Regarding morphology, Chinese ceramics were relatively diverse, including pitchers, bowls, kendy, plates, jars, pots and lids. The recovered artifacts show that they are products of Yuezhou, Guangdong and Changsha kilns in Southern China. These ceramics are said to be similar to the ceramic artifacts discovered in numbers of archaeological sites in Thailand, Malaysia and other place throughout Asia.According to the excavation report, there were 55 pieces of Islamic pottery along with other Middle-East origin items founded during this excavation. The Islamic objects excavated in Cù Lao Chàm have contributed significantly to the general map of Islamic Items found at the archaeological sites in central Vietnam. In addition to Cù Lao Chàm, the Islamic pottery was also found in other sites including Trà Kiệu (Quảng Nam), Trảng Sỏi (Hội An), Ngu Hanh Son (Đà Nẵng), Ly Son (Quảng Ngãi) and Phú Yên with total of 100 specimens. Apart from ceramics, number of pieces of Islamic glass had been also found in excavation, including jars, vases, plates and vessels.20 Arab coins in nagara Amaravati has also been found. According to Anton Nguyen Truong Thang, who has created a special collection of Champa artifacts at Trà Kiệu Cathedral, Quảng Nam province. Among various valuable artifacts of Champa civilization, there was an Arab coin recovered at Chiem Son—a neighboring village of Trà Kiệu Cathedral. The coin was inspected by experts of the British Museum and was affirmed to be an Abbasid dinar of the caliph al-Muktafi billar r.289-95 AH/902-8 CE.21

Fig. 4. Dinar coin of Trà Kiệu collection

Fig. 4. Dinar coin of Trà Kiệu collection

©Nguyen Truong Thang

Champa’s Cross-Cultural Trade and Long-Distance Exchange: A View from Maritime Archaeology

34Recent archaeological discoveries in central Vietnam have made it possible to reconstruct a maritime history of Champa during the Tang period in which it features the important role of Champa as a hub for cross-cultural exchange in the maritime silk route. The excavations in Cù Lao Chàm have provided solid evidence to demonstrate the significance of this island in the maritime silk route as mentioned in Chinese records and presented earlier. The Châu Tân shipwreck and its commodities, as will be introduced below, indicate the presence of international merchants on the coast of Champa by the ninth century.

Fig. 5. Location of Châu Tân Shipwreck

Fig. 5. Location of Châu Tân Shipwreck

Source: Google map, modified by Do Truong Giang.

  • 22 In 2013, I reported the preliminary research of this finding at The first SEAMEO-SPAFA Conference, (...)

35The Châu Tân shipwreck was discovered and salvaged by local fishermen since 2012 on the coast of Quảng Ngãi province, about hundred km south of Hội An and Cù Lao Chàm island. A large amount of artifacts which include a hundred thousand ceramic shards, sections of ship hulls and other objects recovered from this shipwreck were then collected and stored at a local private museum. Mr Lâm Dũ Xênh, the owner of that local private museum then invited a limited number of archaeologists to visit and do research on his new acquisitions from Châu Tân shipwreck. Dr Nishimura Masanari, who suddenly and sadly passed away in 2013 after a tragic accident in Vietnam, took the pioneering and leading role in an attempt to examine thoroughly and scientifically this new collection. After his death, a joint team of Vietnamese and Japanese archaeologists took over his project and continue to do research on Châu Tân shipwreck (Nishino et al. 2015: 225-234). Thanks to the evidence from ceramics, inscriptions inscribed on ceramics and the shipbuilding technique, it is quite clear now that the Châu Tân shipwreck is up-to-date the earliest shipwreck found on the coast of central Vietnam, former territory of Champa, and could be dated back to the ninth century. While the detailed study on building techniques of the Châu Tân shipwreck is being examined by a group of experts, this paper focuses merely on introducing the trading ceramics recovered from this shipwreck and put it in the context of nearby and related archaeological sites. The author of this current study had undertaken several study trips to this special collection as well as related sites in central Vietnam respectively in 2012, 2013 and 2014.22

  • 23 I am in debted to Prof. Pierre Manguin, Dr. Michael Flecker for their participation in the online g (...)
  • 24 I would like to thank Prof. Louis Cort, Prof. John Miksic, Prof. Pierre Manguin, Dr. Li Baoping, Dr (...)

36Châu Tân ship is a typical lashed-lug and stitched plank vessel. This ship building technique is said to be of Southeast Asian tradition and was widely practiced in the region over thousand years from around second century to at least the early fourteenth century (McGrail 2004). The fact that the Châu Tân ship is still stitched suggests that it was made no later than the 10th century.23 Basing on the original keel timber, experts suggest that the Châu Tân ship could have been nearly 25m long and approximately 9m width (Nishino et al. 2015: 232). The salvaged ceramics from Châu Tân shipwreck consists of 400 sand bags of which the ceramics from Changsha, Yue and Ding kilns constitute the largest assemblages. Changsha ceramics include bowls, ewers, plates, dishes, white glaze, brown and green glaze, variety in shape and rim decoration. Yuezhou ceramics include bowls, plates, dishes, ewers, boxes, cups. Ding and Xing-type white porcelains include large dishes, dishes, bowls, ewers, cups and boxes. A large number of medium and large sized yellow-glazed storage jars from Guangdong are also recorded. Brown glaze basin-shape with impressed flower probably products of the Xicun kiln. Islamic green-glazed shards also recovered though not great in quantity.24

Fig. 6. Ding-Xing white wares

Fig. 6. Ding-Xing white wares

©Do Truong Giang

Fig. 7. Changsha ceramics

Fig. 7. Changsha ceramics

©Do Truong Giang

  • 25 This translation provided by Prof. Michael Feener of National University of Singapore.
  • 26 Michael Feener, personal communication, 2013.

37In this collection, the inscriptions on the bases of jars and basins with two types of incised and ink inscriptions are very stimulating. According to Prof.Aogama and Nishimura project team, 147 shards with incised inscriptions which are mainly found on bowls and jars have been identified up-to-date (May 2015) (Nishino et al. 2015: 225-234). Out of 438 shards with ink inscriptions, there are 19 Chinese, 27 Arabic and 201 Indic scripts, while 191 shards remain unidentifiable. Interestingly, several Indic inscriptions bear the name Ambārak which appears to be a location in present-day Iran. Ambārak is known to be a place in the north-west of ancient port of Siraf on the Persian Gulf where was a well-known destination of the maritime trade route during the Tang period. In the meantime, the incised Arab inscriptions provide evidence of the presence of Arab merchants/travellers on board of the ship. It appears that all the pieces with bits of Arabic on them carry the same text as “amdada dadu muhammad wa dafa…” which could be translated as “To aid Dadu[?] Muhammad and to resist/ward off…”25 The Arab inscriptions of Châu Tân collection are mainly end with a six-pointed or five-pointed star. Dr Michael Feener, an expert of Islamic history, identifies this symbol as the “seal of Solomon”, an emblem that was ubiquitous across the medieval and early modern Muslim world, and carried with it magical and talismanic qualities in many contexts.26

Fig. 8. Arabic scripts inscribed on the bases of Guangdong basins

Fig. 8. Arabic scripts inscribed on the bases of Guangdong basins

©Do Truong Giang

Fig. 9. Chinese inscriptions on Ceramics

Fig. 9. Chinese inscriptions on Ceramics

©Do Truong Giang

Fig. 10. Unidentifiable scripts

Fig. 10. Unidentifiable scripts

©Do Truong Giang

38The discovery of Châu Tân shipwreck off the coast of Quảng Ngãi province has a significant implication for those studying the maritime contacts between ancient Champa and the outside world. Most importantly, it provides the physical evidence for what had been written by contemporary Arab and Chinese writers during the ninth century. This finding affirm that the coast of nagara Amaravati was a favourable destination for long-distance travellers and merchants during the Tang era. The recovered assemblages of this shipwreck as briefly introduced above could be seen as one of the fullest representation of cross-cultural exchanges by the ninth century which see the involvement of Arab traders, Chinese commodities, Southeast Asian ship technology. Champa’s coast, as mentioned in both Arab and Chinese maritime itineraries and strongly re-affirmed by Châu Tân shipwreck, appears to be the favorable destination of the cross-ocean vessels.

39Combining evidence, we might sketch out the tentative/possible itinerary of the unfortunate Châu Tân ship as follow: Canton port in south China, where Arab and Persian traders had firmly established a trading colony by end of Tang period, appeared to be the departure place. This ship might call at Hainan coast for fresh water and goods before arriving at the coast of Champa. Although the Châu Tân ship was unfortunately sank off the coast of Quảng Ngãi, roughly 100km south of Thu Bồn estuary and Cù Lao Chàm island, we, however, believe that this vessel had called at either Cù Lao Chàm island or Thu Bồn river estuary before setting sail to the south. Information from Champa inscription and the discovery of Chinese and Islamic artifacts along the Thu Bồn river banks suggest that a short stay of Arab and Chinese traders for the purpose of exchanging goods and visiting the local rulers is arguably imaginable. Leaving the port of Amaravati and sailing along the coast, the ship was shortly sank due to either an ocean-storm or hitting the reef. The later reason appears to be more reasonable when we note that many shipwrecks have been discovered off the coast of Quảng Ngãi in recent years. If the ship had not sank, its next destination would probably be the mercantile port of nagara Panduranga in south of Champa before leaving for Pulao Condore. After leaving the sea of Champa, this ship might follow the same itinerary of the contemporary Belitung Ship and thus might anchor at a mercantile port of Java sea. The abundance of Arabic and Indic inscriptions written on the recovered ceramics suggest that the final destination of this ship may have been the Persian Gulf, or more precisely the Ambarak in north of ancient Siraf port if the information from the Indic inscriptions are accurate.

Haut de page

Note de fin

1 Jiu Tang Shu 197, cited from Maspero (1928: 85-87).

2 Maspero held the idea that this mission was sent by king Sambhuvarman, the son of Rudravarman and author of the inscription C.73 A at My Son. See Maspero (1928: 82, n. 5).

3 In the middle of the 10th century, the Arab Mas΄ūdī recorded Arab trade with Southeast Asia and China: circa 900 C.E. when “the ships of Basra, Sīrāf, Oman, India, the islands of Zābaj and Sanf came to the mouth of the river of Khānfū in China with their merchandise and their cargo” (Tibbetts 1979: 37).

4 Pulau Tioman.

5 The Khmer polity, likely around the Mekong delta.

6 William Southworth suggests that the Cham port used on this journey was probably Nhatrang (Southworth 2004: .227-228).

7 Kadrang = either Pulao Condore or Panduranga/Pandran in Panran.

8 Pulau Tioman.

9 Tibbetts suggests various locations and equivalents for Kadrang (Tibbetts 1979: 157-59). Ferrand suggested that “Kandrang” is an equivalent of the Chinese name Jun-tu-long, a reference to Poulo Condore (Ferrand 1913-194: 14-17).

10 Southworth suggests this journey probably stopped in Quảng Nam Province (2004).

11 The island of Hai-nan.

12 Xin Tang shu, juan 43C, 18b, cited from Wang (1998: 98). For further details, see Pelliot (1904).

13 Previous scholars, like those as Pelliot and Wang Gungwu placed Men-tu in Bình Định province, see Pelliot (1904: 217) and Wang (1998: 98).

14 Momoki Shiro also believes that Mendu was in Phú Yên province, not in Bình Định. See Momoki (1998).

15 K.53, Kdei Ang inscription.

16 Inscription C81 in My Son (between 709-717), see Golzio (2004: 25).

17 In inscription C95 in Mỹ Sơn (d.1056/57); inscription C94 in Mỹ Sơn (d.11th century); in inscription C90A-C,D in Mỹ Sơn (d.1081), inscription C100 in Mỹ Sơn (d.1157/58); and inscription C4 in Chợ Dinh, Phan Rang (d.1227/28).

18 Suchandra Ghost. “Ministers and Nobles in the kingdom of Campa: looking through the epigraphic lens (c.909-919 CE)”

19 Ibid.

20 Lam My Dung, Report on field survey and excavation in Cù Lao Chàm during the period from 1998 to 2000, unpublished document. I would like to express my gratitude towards Dr Lam My Dung for sharing her valuable suggestions and important document for my current research.

21 retrieved in June 2015.

22 In 2013, I reported the preliminary research of this finding at The first SEAMEO-SPAFA Conference, Thailand.

23 I am in debted to Prof. Pierre Manguin, Dr. Michael Flecker for their participation in the online group discussion on these materials.

24 I would like to thank Prof. Louis Cort, Prof. John Miksic, Prof. Pierre Manguin, Dr. Li Baoping, Dr. Geoff Wade and Dr. Edward McKinnon for their participation in the online group discussion on these materials.

25 This translation provided by Prof. Michael Feener of National University of Singapore.

26 Michael Feener, personal communication, 2013.

Haut de page

Table des illustrations

Titre Fig.1. Place names mentioned in Ibn Khurdadhbih (mid-9th cent CE)
Crédits Source: Tibbetts (1979).
Fichier image/png, 94k
Titre Fig. 2. Maritime itinerary in Southeast Asia during the Tang era
Crédits Source: Wang (1998: 97).
Fichier image/png, 670k
Titre Fig. 3. The 9th-10th centuries Changsha and Yuezhou ceramics on display at Hội An Trading Ceramic Museum
Crédits ©Do Truong Giang
Fichier image/png, 748k
Titre Fig. 4. Dinar coin of Trà Kiệu collection
Crédits ©Nguyen Truong Thang
Fichier image/png, 321k
Titre Fig. 5. Location of Châu Tân Shipwreck
Crédits Source: Google map, modified by Do Truong Giang.
Fichier image/png, 710k
Titre Fig. 6. Ding-Xing white wares
Crédits ©Do Truong Giang
Fichier image/png, 350k
Titre Fig. 7. Changsha ceramics
Crédits ©Do Truong Giang
Fichier image/png, 660k
Titre Fig. 8. Arabic scripts inscribed on the bases of Guangdong basins
Crédits ©Do Truong Giang
Fichier image/png, 540k
Titre Fig. 9. Chinese inscriptions on Ceramics
Crédits ©Do Truong Giang
Fichier image/png, 659k
Titre Fig. 10. Unidentifiable scripts
Crédits ©Do Truong Giang
Fichier image/png, 240k
Haut de page

Pour citer cet article

Référence papier

Do Truong Giang, « Diplomacy, Trade and Networks: Champa in the Asian Commercial Context (7th-10th Centuries) », Moussons, 27 | 2016, 59-82.

Référence électronique

Do Truong Giang, « Diplomacy, Trade and Networks: Champa in the Asian Commercial Context (7th-10th Centuries) », Moussons [En ligne], 27 | 2016, mis en ligne le 17 mai 2016, consulté le 26 mars 2017. URL :

Haut de page


Do Truong Giang

Do Truong Giang is currently a history PhD Student at National University of Singapore (NUS) and research fellow at Vietnam Academy of Social Sciences. Giang was a graduate fellow at Asia Research Institute (ARI) – NUS in 2009 and Affiliate at Center for Southeast Asian Studies (CSEAS), University of Michigan in 2014. He is now writing his Doctoral Dissertation, “Re-constructingthe history of Champa (9th-15th centuries CE)” at NUS. Apart from his PhD, Giang is also participating in two other research projects: Campa-Khmer Angkor Links project and Vietnam Maritime Archaeology project.

Haut de page

Droits d’auteur

Licence Creative Commons
Les contenus de la revue Moussons sont mis à disposition selon les termes de la Licence Creative Commons Attribution - Pas d’Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page