Qian Zhi Fu

The location of Qiān Zhī Fú and the identification of the names recorded for it remain uncertain. The notice stands in the Nan Man (南蠻), or "southern barbarians", chapter of the New Book of Tang, whose sequence runs through Huan Wang, Pan Pan, Funan, Zhenla and Dvaravati.^{[1]: 26 [2]} Hoshino reads the Southwest Sea of that chapter as denoting not a sea but the part of mainland Southeast Asia lying south of the Red River basin and Nanzhao, distinguished from the Nan Hai of the middle Mekong.^{[1]: 39, 67} The notice describes Qiān Zhī Fú as a settlement of people from southern India and gives it the alternative name Bàn Zhī Bá, which Hoshino renders "Five Hills".^{[1]: 30, 34 [3]} The same passage describes the northern neighbour Duō Mó Cháng, whose ruler is said to command an army of 20,000 men equipped with various weapons but no horses, a figure that concerns that state and not Qiān Zhī Fú.^[1]: 30

On the basis of the name "Five Hills", Paul Pelliot tentatively identified Qiān Zhī Fú with Kancipura in southern India, which has a mountain of five peaks.^[1]: 30 Hoshino rejects this identification and instead places the polity inland, adjoining the Isan region, the neighbouring Mekong valley and Vietnam. He argues that the reference to southern Indians describes the settlement's historical connections rather than its geographical location, and that those connections had already declined by the time the Tang reports were compiled.^{[1]: 32, 34} Hoshino also cautions against the older identifications of the chapter's place names, noting that Pelliot's proposals were provisional and were subsequently adopted by Cœdès without re-examination. He further notes that Pelliot, encountering a passage that did not fit his reconstruction, considered it misplaced and proposed that it belonged in the section on India instead.^{[1]: 25, 29 [b]}

The chain of neighbours in the annals provides one line of evidence for the proposed location. Duō Mó Cháng stands immediately north of Qiān Zhī Fú, with Hē Líng (訶陵) further north.^[1]: 30 Pelliot and Cœdès identified Hē Líng with a polity on the island of Java, partly because it is placed south of Zhenla and because the Chinese sources also use Du Po for Java. On that reading, however, Qiān Zhī Fú would lie still farther south, at the southern limit of the Chinese geographical world, potentially in open sea or as far as Australia. Hoshino considers this implausible: Australia was not known to the Chinese at the time, the proposed region was not Indianised, and the islands of Southeast Asia contain no river comparable to the Ganges, although the annals place the group including Qiān Zhī Fú south of one.^{[1]: 30–31}

Hoshino therefore distinguishes two places called Hē Líng. He identifies the northern Hē Líng with a polity around Sipsong Panna, drawing on Pelliot's citation of the biography of Wei Gao, where Hē Líng appears as the second of eight West Mountain chiefdoms that submitted in 793.^{[1]: 31–32}

The position of Duō Mó Cháng also provides a geographical constraint from the north. The Chinese record states that travel across its territory was more difficult from east to west than from north to south. Hoshino interprets this as describing two or three river valleys separated by north–south mountain ranges, a configuration he associates with Lan Na, including the Yom and Nan basins and tributaries of the Mekong.^[1]: 60 If Duō Mó Cháng occupied the Yom and Nan basins, Qiān Zhī Fú would have lain in the plain immediately to its south, consistent with Hoshino's proposed location.^[1]: 34

The name itself has also been considered. The suffix fú 弗 corresponds to the Sanskrit pūra, meaning "town" or "city", an element Hoshino also finds in Phetburi, Chanthaburi, Phetchabun and Si Khottabun.^{[1]: 28, 30} The prefix qiān 千 remains uninterpreted, and no identification with a local name has been established. The term xiān 暹, however, is attested in Chinese and Đại Việt sources for peoples and polities of the Chao Phraya River basin in the early second millennium.^[4]

Among the mainland candidates, Hoshino considers Lopburi and Si Thep and prefers Si Thep on three grounds: its sacred hills, the presence of the region's earliest Sanskrit inscriptions, and the quality of the sculpture found in its vicinity. He further identifies the polity with the primordial Canasapura, treating Mueang Sema [de] as a regional administrative centre, and cites Dhida Saraya's 1985 study in support of this equation.^{[1]: 32–34, 42, 55} He describes these identifications as tentative and consistent with the textual evidence rather than established by it, and states that nothing in the reconstruction of the period is certain.^{[1]: 37, 46} Si Thep has been claimed for another polity as well. Claude Jacques proposed in 2009 that the site was Dvāravatī itself, on the strength of its early cult of Viṣṇu and of Kṛṣṇa, whose legendary capital in Indian literature carries that name.^[5]: 408

Baker and Pasuk rank Si Thep as probably the second largest town of the Chao Phraya plain, a moated site of the Pa Sak valley occupied from the 6th century, enlarged many times and given several large monuments, among them Khao Klang Nok, 64 metres square and about 20 metres high, built of large laterite blocks to a design without parallel. When many sites founded in the 6th century declined or were abandoned from the late 9th, Si Thep and Nakhon Pathom were among the few that were enlarged instead, and Si Thep stood with Mueang Sema [fr] and Si Mahosot [th] on what they call the cultural border between the plain and the plateau, where Angkorian-style temples were built.^{[6]: 16, 21}

The Chinese records describe a network of overland and river routes crossing the middle Mekong region and linking several of the interior polities. Hoshino places Lopburi at its western end and proposes that the route continued east beyond the mountains to the South China Sea. He points to the number of archaeological sites and finds along its course in Isan as evidence of movement across the region from prehistoric times through the Tang period.^[1]: 50 Traffic across the Mekong was not new in the seventh century, but he judges that its volume rose sharply during the Tang and that a place standing both on this route and at the crossing of others would have been enriched by it; when the traffic first became heavy enough to produce such a hub he leaves open.^[1]: 53 The inland states along it are distinguished in his account from the coastal territories dominated by Dvaravati, whose wealth came from maritime trade.^[1]: 30

Several of the states recorded in Chinese sources near the eastern arm of the route are dated to the seventh and eighth centuries. Hoshino interprets them as Tai-dominated, although Hiram Woodward considers this part of the argument tendentious and argues that it should neither support nor invalidate the geographical reconstruction.^[c] They include Gān Bì (甘毕), which could mobilise 5,000 soldiers and which he places in the vicinity of present-day Savannakhet;^[1]: 46 Jūn Nà Lú (郡那盧; Cantonese gwan1 naa5 lou4), rendered in Thai and Lao as "แคว้นนาลาว" and "ແຄວ້ນນາລາວ", "country of Lao rice fields", which he places in the area of modern Sakhon Nakhon and counts among three states lying up-river from the Tang prefectural centre of Changzhou [zh] rather than within it;^{[1]: 60–61} and an unidentified Gǔ Láng Dòng (古郎洞), which Hoshino takes for the name of a Tai state, on the road between that prefecture's two districts, the second of which he places near Thakhek.^{[1]: 49–50} Further north the same chapter gives an overland route from Duō Mó Cháng to Jiaozhou by way of Lin Yi in central Vietnam.^[1]: 30

The tribute mission of 657 provides evidence of contact among several of these polities. Qiān Zhī Fú travelled as one of five, under the leadership of Zhān Bó; the others were Pó Àn (婆岸), whose name Hoshino suggests survives in that of Mueang Phon in Khon Kaen province, Shě Bá Ruò (舍跋若), which he places at Suphan Buri, and Mó Là (摩臘), which he identifies in one passage as a Cham port on the South China Sea, comparing the name with Mloi, and in another as a Malay or Mon town.^{[1]: 30, 42, 47, 68} The party reached the Tang court with elephants and rhinoceroses. Hoshino interprets their joint reception as an acknowledgement of Zhān Bó's leadership; if Mó Là was indeed a Cham port, he suggests that the envoys may have completed the journey by sea from there.^{[1]: 43, 47} He compares the group with the chain of allied princes and chiefs named in the Ram Khamhaeng inscription five centuries later, while noting that the 657 mission comprised only five small states.^[1]: 42

Tatsuo Hoshino has suggested that Qiān Zhī Fú was one of the early Siamese kingdoms of the trans-Mekong network, later Taicized.^[1]: 54 Other polities proposed as contemporary Siamese-related states included Cān Bàn,^{[1]: 39, 68} Xiū Luó Fēn,^[1]: 54 and Wen Dan.^{[1]: 39, 53, 68} His argument for grouping them rests on the New Book of Tang itself, which places Qiān Zhī Fú, Xiū Luó Fēn and Gān Bì in a single category and reports that the customs of each resembled those of the others; all three were fortified, and Hoshino reads their military organisation as characteristic of Tai-dominated states.^{[1]: 53–54}

A separate line of evidence comes from the terms Siam and Syam, attested in several inscriptions and interpreted by earlier scholars as references to a specific location or population within the Chao Phraya Basin.^[8]: 69 These include K557 (611 CE),^[9]: 21 K79 (639 CE),^[9]: 69 K127 (683 CE),^[9]: 89 K154 (685 CE),^[9]: 123 and K904 (713 CE).^[10]: 54 Syam later occurs in slave lists on inscriptions of the Champa and Khmer kingdoms, dated to the 11th and 12th centuries.^{[11]: 140 [12]: 124 [13]: 62} From approximately the same period, a well-known bas-relief panel at Angkor Wat depicts mercenaries of the Angkorian army identified as syam-kuk, which has been interpreted as referring to Siamese people.^[8]: 70

A Monic element in the population is attested separately. The Nyah Kur, who speak a conservative Mon dialect, still inhabit the highlands around Si Thep, and their presence has been taken to indicate that the early inhabitants of the region were Monic and were absorbed into Tai culture over time, a process associated with the arrival of northern Chiang Saen and eastern Lao-Phuthai groups and with trade contacts toward southern China and Đại Việt.^{[citation needed]} A comparable shift is reported for the Kaleun people, thought to descend from the Austroasiatic Bru people and to have adopted a Southwestern Tai language.^[14][15]

An inscription discovered at Ban Wang Pai, Phetchabun province (K.978), records the enthronement of a new monarch, identified as a son of Prathivindravarman. George Cœdès's original reading dated the inscription to around 550 CE and identified this son with Bhavavarman I of Chenla, suggesting the existence of dynastic connections between Qiān Zhī Fú and Chenla, although the king's own personal name remains unrecorded in the text itself.^[45][46] Thai scholar Kangwon Katchima argues that the Bhavavarman mentioned in the inscription may not correspond to Bhavavarman I of Chenla, given that the inscription was likely transcribed after 627, several decades subsequent to Bhavavarman I's reign. The reference could therefore pertain to Bhavavarman II—whose origin remains unclear—or to another monarch of the same name.^{[47]: 17–19} Anurak Depimai supported Katchima's redating and rejected the identification of the king as Bhavavarman II, noting that the Wat Kud Tae inscription identifies Bhavavarman II as a son of Isanavarman I, not of Prathivindravarman. Depimai instead identified the Wang Phai king as a local ruler of Si Thep who adopted Chenla-style royal titulature.^{[48]: 39–42}

The use of the South Indian regnal title varman among Qiān Zhī Fú rulers indicates substantial influence from south Indian culture,^[46] consistent with Chinese sources reporting that the polity was once a South Indian colony before attaining independence at the time of its diplomatic mission to the Tang Dynasty under Emperor Gaozong of Tang circa 656–661.^{[1]: 30, 32, 34}

If Qiān Zhī Fú at Si Thep was the primordial Canasapura, then it is plausible that the Canasapura monarchs listed in the Śri Canāśa Inscription K.949, discovered on the island of Ayutthaya,^[f][49] may have reigned at Si Thep. Notably, an Angkorian inscription dated 946 records the victory of king Rajendravarman II over Rāmaññadesa (lit. 'country of the Mon'),^[50] coinciding with a period during which the regnal titles of Si Thep Canasapura rulers transitioned from the Buddhism-associated "datta" or "krama" to the Hindu-oriented "varman".^[51] Archaeological evidence from Si Thep dating to this era further corroborates Angkorian influence.^{[17][35]: 303, 308–309}

The Pali Jinakalamali recounts the enthronement of Camadevi at Haripuñjaya in 662 CE and mentions her spouse, Ramaraja, who has been interpreted by multiple Thai scholars as the monarch of Ramburi (รามบุรี; lit. 'city of Rama').^[52] The identification of Ramburi remains contested; it has been alternatively proposed that Ramaraja was a prince of Lavo,^[52] a prince of Srivijaya,^[52] a ruler of Mawlamyine,^[53]: 46 or a sovereign of Ayutthaya's antecedent polity, Ayojjhapura at Si Thep.^[21]

Additionally, the 9th-century Sanskrit Phu Khiao Kao inscription K.404, located approximately 100 kilometers northeast of Si Thep in present-day Kaset Sombun district of Chaiyaphum province, mentions a king named Jayasimhavarmman. His precise identity remains uncertain; prior scholarship has variably suggested that he may have ruled Wen Dan, Bhavapura,^{[54]: 90 [55]} or Champa.^[56] Definitive identification, however, is currently lacking.

The following is the list of kings of Qiān Zhī Fú's Si Thep; some of them have currently been the subject of debate.