New copperplate grant of Śrīcandra (no. 8) from Bangladesh1

Abstract This article is the first publication of the Bogra copperplate, the eighth known land-grant inscription issued by Śrīcandra (r. c. 925–975 ce), one of the kings of the Candra dynasty of Bengal. A diplomatic transcription is included, together with an annotated English translation and a critically edited text in Devanāgarī. The inscription describes a gift of land to a Brahmin named Śrīkaradatta Śarman, who probably hails from North Bengal (“Hastipada [in the region of] Śrāvasti”). While the praśasti (praise) portions largely parallel the king's other known inscriptions, the inscription contributes new information about place names and regions associated with the Candra dynasty, as well as attesting the movements of Brahmins associated with the Parāśara Gotra and Chandoga Caraṇa. The article also provides an overview and assessment of research on the inscriptions and history of the Candra dynasty, particularly in light of the discovery and identification of this new inscription.

city in North Bangladesh. It now forms part of a private collection in the USA, to which it travelled after having passed through many hands, including dealers in Denmark and Thailand. The majority of the other Śrīcandra plates have been found in more southern portions of the region (e.g. near Dacca, in Rampal), although the provenance of one plate issued by this ruler, the Paschimbhag plate, was in the north-east (i.e. Sylhet). The text of the inscription, however, corresponds most closely with the inscriptions found in the southern portions of Bengal (especially the Rampal plate).
This copperplate was brought to my attention by the current owner, Dave Smith, for identification and decipherment, and he also granted permission for the plate to be photographed and published. I refer to it below as the "Bogra plate" after its tentative find-site. The plate is 21 cm long and 17 cm wide. It bears an inscription, on both sides, for a total of 40 lines (25 on the obverse and 15 on the reverse). The inscription is in the proto-Bengali script current in the tenth and eleventh centuries CE (also called Northern Nāgarī; Majumdar 1929). 4 There are presently about twelve known plates of the Candra dynasty, 5 and all are written in this same script. The letters on the Bogra plate are small and neat, although they have some inconsistencies, probably due to their small size, which obfuscates some of the subtle differences between letters. The height of the letters is about 1 cm on average.
Overall the plate is clean and readable, with most of the wear being near the edges; the reverse is more legible than the obverse. There is a nice patina richly colouring the plate. In terms of the state of preservation, this is as fine an example of a Śrīcandra plate as is found anywhere. At certain points, the Bogra plate thus allows us to confirm or question some of the conjectured readings that past scholars have proposed for other plates issued by this king.
There is one small dent on the obverse of the plate, on the lower left, at line 23. This damage has resulted in a thin line through a small portion of the text; it is possible that the plate was dropped at some point or struck (by whatever cause) during its long history. This dent, in any case, does not appear to be originally been grouped with a number of Pāla dynasty sculptures and misidentified as an inscription of the Pāla dynasty. 4 The term "proto-Bengali" has been employed for a wide range of scripts, sometimes incorrectly, but it is most appropriate for those from the general time-frame of this inscription; see discussion in Dimitrov (2002, 30-33). Further palaeographic study of early forms of the Bengali script is needed to assess its development more accurately. 5 There are a total of four known copperplates issued by Śrīcandra's descendants. One is by Kalyāṅacandra (c. 975-1000), namely, the Dhaka plate (Dani 1961). Two are by Laḋahacandra (c. 1000-1020), both found at Maināmatī (Dani 1961(Dani , 1966Mills 1993, 85); there is also an inscription of this king on an image (Huntington 1984, 61-4, 242-4, plates 64-5). One known plate was issued by Govindacandra (c. 1020-1045) and was similarly found at Maināmatī (Dani 1961(Dani , 1966Mills 1993, 86); in addition, there is an inscription from the twenty-third year of his reign (i.e. the Betkā-Pāikpādā inscription) for the creation of an image of Vāsudeva by an inhabitant of Rālaja (Huntington 1984, 61-4, 244-5, plates 66-7). Govindacandra's name is also mentioned in the Tirumali inscription (1025 CE) of the Chola king Rājendracola, who calls him the king of Vangāladeśa (Sircar 1971, 132-3). Govindacandra seems to have been the last king of the Candra dynasty of Bengal. new. Nevertheless the damage does not affect the reading of the text, which is still legible in the area in question (see n. 46 below). The language of the inscription is Sanskrit. As is typical of these plates, a lengthy prose section comprises the main portion of the inscription. There are stanzas in four types of meter coming before this prose section, namely, Vasantatilakā (stanzas 1 and 3), Śārdūlavikrīḋita (stanzas 2 and 5), Upajāti (stanzas 4, 7 and 8), and Indravajrā (stanza 6). The three stanzas following the prose section (9, 10, and 11) are all in Anuṡt˙ubh meter.
The sealing (mudra), affixed to the top of the plate, is just under 12.5 cm long and 9 cm wide. It bears the dharmacakra symbol, two deer facing a wheelan image common in Buddhist art since at least the Gupta era and a common royal emblem on the land-grant sealings of the Deva, Pāla and Candra dynasties. 6 It should be noted that the use of the dharmacakra symbol is not limited to Buddhist kings but is also found on sealings of some kings in the region who patronized Hindu gods. 7 Like all of these types, this sealing has raised print, having been pressed from a seal carved in intaglio. The name of the king is here written as Śrīśrīcandradevaḣ ; below the name is an image of the kīrtimukha ("mask of glory"), a common iconographic motif on architecture, temples, and sealings throughout India. Unlike the sealings of the Pāla dynasty, which are typically rounded on the top, this and other Candra sealings are pointed. 8 The inscription describes the gift of land from Śrīcandra to a Brahmin named Śrīkaradatta Śarman, who probably hails from North Bengal, from the areas where the Kāmarūpa Pālas were dominant (see below). As with all known Śrīcandra plates, the land-grant is issued from Vikramapura. 9 Moreover, as with most of the Śrīcandra grants, the gifted land is located in the province (bhukti) of Pauṅḋra. 10 The plate further identifies the land as located in the Samatat˙a region (maṅ ḋ ala) 11 in the Śrīnagara district 6 For discussion of sealings and their function in copperplates from Bengal, see Morrison (1970, 66-8). 7 See, for example, the Pāla dynasty sealing in N. G. Majumdar (1933, 150-9, plate I).
From their copperplate inscriptions it seems that Pāla kings patronized Buddhists and Brahmins as well as supporting worship of Śiva and Viṡṅu (Tripathi 1987, 192-201;G. S. Majumdar 1983, 24-8). Later kings in the Candra lineage appear to have patronized Vaiṡṅavite and Śaivite deities but maintained the dharmacakra as a royal emblem; see Sircar 1970, 199-200, 210, Plate I, III;1971, 253-7. 8 This feature resonates with architectural forms on temples in North East India and South East Asia, and it may speak to the stylistic traditions common in the geographic areas with which the Candra dynasty was most strongly associated. 9 On other inscriptions issued from Vikramapura and on its status as an administrative centre of the region, see Morrison 1970, 56-7. 10 So too the Rampal, Madanpur, Bangladesh Museum, and Dhulla plates. Sircar (1959-60, 137-8) notes that the Pauṅḋra bhukti was "the territorial division, the headquarters of which were originally at the city of Puṅḋravardhana identified with modern Mahāsthān in the Bogra District of North Bengal. During the age of the Pālas, this bhukti or province included considerable parts of South-Eastern Bengal. Although the Chandra empire does not appear to have included any portion of North Bengal, the name of this old administrative unit was continued in use during the rule of the Chandras". See also R. C. Majumbar 1971, 203. 11 This is also the location of the land granted in the Bangladesh Museum plate. On Samatat˙a, see the detailed discussion in Sircar 1971, 149-58; from the inscriptional (viṡ aya). 12 The town name (grāma) appears to be Vyāghravoraka, 13 and the plot is a single unit ( pāṫ aka) in size. There is no date of issue inscribed. 14 The inscription begins with a standard genealogy of Candra kings, which occurs, with minor variations, on all known Śrīcandra plates. This genealogy of Śrīcandra lists three prior generations: In a manner largely consistent with the other plates of this king, the Candra dynasty is here described as having once ruled in Rohitāgiri. The precise evidence, Sircar identifies it as "the present Tippera-Noakhali region of South-East Bengal". This maṅ ḋ ala is well known from outside sources as a famous centre of Buddhism, having been mentioned by the seventh-century Chinese pilgrims Xianzang and Sengzhe (see further Tripathi 1987, 188-91, 211). For other inscriptions that mention this region, see Morrison 1970, 115-19, 125. 12 With regard to the Śrīnagara district or viṡ aya, a similar place name, Śrīnagara-bhukti, is found in a plate of Devapāla; this is a medieval name of modern Patna in Bihar; see line 26 of the Nālandā copperplate inscription of Devapāla (c. 812-50 CE; Sircar 1983, 75). It is perhaps unlikely that the two districts are one and the same since the Śrīcandra plate occurs more than a century later. The region designation of bhukti changed to that of a viṡ aya, but it is unclear which size of land plot the different terms designate. It is likely one is a subdivision of the other. Śrīnagara is not, to my knowledge, associated with any other medieval city in north-east India. Here, it seems to be a specific district and not a generic term. 13 See line 18 of text. The term voraka recurs in a number of village names in inscriptions of Śrīcandra's successors (Sircar 1970, 200-01); see Mainamati plate no. 1, lines 38-9 and 42-3 (Sūpakāravoraka, Baleśvaravarddhakivoraka, Bappasiṁ havoraka, and Karavattīvoraka), as well as plate no. 2, lines 10-11 (Brāhmaṅadevavoraka and Suravoraka). Thus, we may reasonably speculate that Vyāghravoraka also represents the name of a town. Sircar suggests that voraka is related to the Bengali boro, "which essentially means a sort of rice sown in low swampy ground or near the banks of a river". He further proposes that in the context of the Mainamati plates, the term might have the sense of being "fit for boro cultivation" (1970,201). This could fit well with the general sense of prosperity and bounty associated with the gift of land, as evoked by other elements in the plate (see below n. 69). There is a well-known maṅ ḋ ala in the general Bengal region from this period with the name of Vyāghratat˙ī; while most likely distinct from the Samatat˙a-maṅḋala mentioned here (line 17), there is some debate as to whether the Vyāghratat˙ī-maṅḋala is in the north or south (A. Banerji 1960). The appearance of "Vyāghra-" here might suggest some correlation with the Samatat˙amaṅḋala at some point. As rightly stressed by Sircar (1965, 381), however, territorial designations are extremely fluid and difficult to determine with certainty. 14 It should be noted that only two of the eight known Śrīcandra copperplates include mention of dates of any kind (Paschimbhag, regnal year 5; Madanpur, regnal year 44). It is on the basis of the dated Candra plates, along with analysis of palaeography, that the dates of the reign of Śrīcandra have been suggested by scholars.
location of Rohitāgiri is disputed by scholars, with Sircar (1971, 152) identifying it with modern Rohtāsgaḋh in the Rohta district of Bihar (37 km south-west of Sasaram), while A. M. Chowdhury (1967, 160-1) places it in eastern Bengal, near Comilla. 15 It is said that Pūrṅacandra, the head of the lineage, was proclaimed through various media: idols (arcā) in stone, copper inscriptions, and pillars (lit. "victory columns"jayastambhas). It has been argued (R. C. Majumdar 1971, 201) that Pūrṅacandra's son, Suvarṅacandra, was the first of the lineage to convert to Buddhism; but the text merely states that he was a Buddhist (bauddha) and was associated with the Buddha's "hare incarnation" (śaśakajātaka). 16 As to whether or not the lineage was previously Buddhist, the inscriptions are silent.
Suvarṅacandra's son, Trailokyacandra, is distinguished, according to the royal genealogy on the plate, for having exercised influence at Harikela and for establishing himself as ruler in Candradvīpa which, according to Sircar (1971, 133), may be taken as identical with "Van . gāla-deśa".
Trailokyacandra's precise relationship with Harikela is not entirely clear from these genealogical notices, although much has been made of the passage (stanza 5 in the present text) with regard to the history and growth of the Candra dynasty as a whole (e.g. Sircar 1971, 133;1959-60, 135;Basak 1949-50, 53-4; R. C. Majumdar 1943, 194-5). It is possible to speculate, for example, that Trailokyacandra took refuge in Harikela's court on account of conflict or unrest within his own kingdom, although some have suggested that he undertook more vigorous expansion. 17 With regard to the reconstruction of the history of the three ancestors of Śrīcandra mentioned on the plate, however, some scepticism is warranted. There are no corroborating external sources referring to these kings, and all that we know about them has been filtered through royal genealogical notices that may well be idealized or inspired, serving primarily as a means to enhance the authority of other, later, Candra kings. This material is thus interesting in its own right as a medieval representation of royal power and ideology (Ali 2000: 165-229), but its use for historical reconstruction must be pursued with caution. It is hoped that new discoveries may eventually help to clarify the dynasty's history prior to Śrīcandra. 15 Sircar (1971, 152) further suggests that the Candras migrated into east Bengal, where they rose to power due to the decline of Pāla power in the region at the time. R. C. Majumdar (1971, 200-1), however, tentatively follows Chowdhury (1967, 160-1). Sanderson (2009, 83) has assigned dates for both Pūrṅacandra and Trailokyacandra as "r. c. 850-875" and "r. c. 900-925" respectively. It is unclear how Sanderson determined these dates as there is no inscriptional record for these figures and their earliest mention is on the Śrīcandra plates themselves. 16 For a discussion and critique of Majumdar's argument see Sanderson (2009, 83 n. 141). 17 R. C. Majumdar (1943, 194-5) employs this passage to suggest that Trailokyacandra "laid the foundation of the greatness of the family"; Basak (1949-50, 53-4) proposes that the king ruled over both Candradvīpa and Harikela entirely. The most extensive record of Trailokyacandra's political enterprises, such as they are described, remains the Paschimbhag plate of Śrīcandra (Sircar 1968, 294, 298-300).
The genealogy is followed by an account of the birth of Śrīcandra and a celebration of his mother, Śrīkāñcanā. 18 The inscription suggests that it was under Śrīcandra that the Candra dynasty finally crystallized its authority in the Bengal region. It is in his name and by his authority that the text thus goes on to record the king's donation of land to a Brahminas described in the next portion of the text.
When listing the genealogy of the donee, Śrīkaradatta Śarman, the inscription seems to convey three prior generations: The text states that the donee comes from the Parāśara Gotra, the Tryārṡa Pravara, and the Chandoga Caraṅa. 19 The Parāśara Gotra possibly denotes a subdivision of the Kārṡṅāyana Gotra mentioned in the Chāndogya Upaniṡ ad (Bhandarkar 1913, 16). Past speculation about the connection of these gotras may find some confirmation in Śrīkaradatta's affiliation with the Chandoga Caraṅa. This caraṅ a (i.e. school) is known in relation to the Sāmaveda as well as to the Chāndogya Upaniṡ ad, the latter being one of the oldest of all the Upaniṡads. There are some other references to this Brahmanical school in pre-modern copperplates from Bengal and Assam, both before and after Śrīcandra. References to Brahmins of the Chandoga Caraṅa are found, for instance, in seventh-century inscriptions issued by Kumar Bhaskara Varman, who ruled in Kāmarūpa (part of modern Assam; Niyogi 1967, 67-72). 20 The details of the lineage-names and scholastic affiliations are, of course, of interest to scholars who wish to trace the movements of Brahmanical schools 18 On the attention to mothers of monarchs in copperplate inscriptions from ancient Bengal, see Tripathi 1987, 35-8. 19 On the basis of the Candra plates available in the 1960s, Niyogi (1967, 8) had gone so far as to suggest that the Candras may have been less interested in recounting the scholastic background of Brahmanical donees than other kings of the time. The evidence of the Bogra plate certainly changes the picture. Other Śrīcandra plates also include information about the gotra and pravara of the donee(s); the former differs, but the latter is most often tryārṡ a (so too Rampal and Dhulla plates). The Bogra plate is the only known plate issued by this king that includes information about the caraṅ a of the donee. The Bangladesh Museum and the Paschimbhag plates, however, do mention the gotras of the families receiving grants. 20 There are only a few other known references to the Brahmins of the Chandoga Caraṅa; these occur in plates from the eleventh century, decades after Śrīcandra's reign. One is the Irdā plate of king Nayapāladeva (a member of the Khaboj Pāla dynasty); see N. G. Majumdar 1933, 156, esp. line 28 of the transcription. Another is the Nidhanpur plate of Bhāskaravarman; see Niyogi 1967, 9, 30, 48-9, 67-72. and families in medieval times (e.g. Niyogi 1967;Witzel 1993). Also of particular interest is the reference to the town and region from which the donee came, since such details are not included in other known plates issued by Śrīcandra; a compound in line 4 on the reverse side of the Bogra plate, śrāvastihastipadagrāmavinirgata-, suggests that Śrīkaradatta "hails from the town of Hastipada (in the region of) Śrāvasti". The hometown of the donee, Hastipada, is sometimes placed by scholars in North Bengal, although there is some debate on its precise location. Datta (1989, 44) suggests that Hastipada may be identical with "Hastigrāma in Magadha" (near Vaiśāli in North Bihar). 21 What is notable, for our purposes, is that Hastipada is often mentioned in other copperplates as the hometown of Brahmins who were granted settlement land by rulers in Bengal and Orissa (Niyogi 1967, 44-5). With only one exception, the Brahmins from this town belong to the Parāśara gotrathe same gotra as Śrīkaradatta Śarman, the donee of the Bogra plate. 22 The Bogra plate's location of Hastipada in the Śrāvasti region would thus seem to support the general view that "Hastipada" was the name not just of an ancient city in Madhyadeśa, but also of a newer Brahmanical settlement in north Bengal. Sircar (1971, 297), for instance, suggests that "a large number of Brahmanas of Sravasti in the ancient Kosala country in Madhyadeśa (i.e. the Set-Mahet region in the Gonda and Bahraich Districts of U.P.) . . . were settled in the Hili-Balurghat area in North Bengal and that these Brahmanas renamed their new settlements after their old habitations in U.P.". In addition, it may be argued that parts of the Kāmarūpa district overlapped with that of the Northern Bengali Śrāvasti. This possibility is suggestive in light of the possible association of the donee of the Bogra plate with the region of Kāmarūpa (Sircar 1971, 159-66, 298;Barua 1952, 52). 23 In any case, from this preliminary exploration of inscriptional data, it can be suggested that the Bogra plate represents the earliest reference to the Chandoga Caraṅa in this particular part of Bengal. The plate is shown in Figures 1 and 2. 21 See also Niyogi 1967, 44-5;R. C. Majumdar 1971, 493 n. 46. 22 Niyogi (1967 has, in the back of his book, laid out a large series of medieval Bengali land grants in a chart ("A tabular statement of classified details of land grants in Ancient Bengal"); those donees having "Hastipada" as the place of origin are consistently said to belong to the Pāraśara Gotra. Datta (1989, 136-7, 226) reiterates this assessment. That the Bogra plate also follows this pattern helps to confirm these earlier evaluations. 23 It may be noted that the Bāṅagaḋa copperplate inscription of king Mahipāla I (c. 988-1023 CE) of Bihar, although it probably post-dates the Bogra plate by a few decades, also mentions a donee named "Varadatta" who "hails from Hastipada". Like the Varadatta of the Bogra plate, he too belongs to the Parāśara Gotra (Mukherjee and Maity 1967, 197-208). The migration of Brahmins from the northern Śrāvasti region appears to have continued.

Text of the inscription 24
L. 1. Siddhaṁ 25 svasti | vandyo jinas sa bhagavān karuṅaikapāt[ram] dharmmo py asau vijaya- 24 My transcription from the original text of the Bogra plate follows the recommendations for dealing with scribal errors, etc., in Salomon 1998, 163. In cases of obvious scribal errors in orthography, etc., I have transcribed the akṡ ara(s) occurring on the inscription and provided suggested readings in parentheses directly following; a question mark has been added in those cases where emendations are less certain. Readings of worn or otherwise unclear akṡ ara(s) have been placed in square brackets. In cases where akṡ ara(s) or elements thereof appear to be missing due to scribal errors of omission, these are provided in square brackets followed by an asterisk. In cases where it is necessary to separate out consonants of a single akṡ ara for the sake of clarity, an "equal" (=) sign is employed. As with many ancient Bengali inscriptions, that on the Bogra plate features frequent exchange of ś and s. In such cases, I have chosen simply to transcribe the text; in this, I am following Salomon's assessment (1998, 163) that "a degree of caution is advisable, as the editor runs the risk of engaging in hypercorrection by applying to old inscriptions strict orthographical standards which have actually developed only in relatively recent times in connection with the printing of Sanskrit texts" and that "hypercorrection when carried to extremes can actually distort linguistic and paleographic data". 25 Siddhaṁ is represented by a symbol, which is also transcribed as oṁ (Basak 1913-14, 138; N. G. Majumdar 1929, 11); however, as Mills notes, "in some inscriptions the symbol preceded oṁ . . . which would never have been the case if the two were identical"   Majumdar 1929, 46); and the Barrackpur copperplate of Vijayasena (N. G. Majumdar 1929, 61). The use of the siddhaṁ symbol and oṁ together appears common, especially in the Bengal region. 26 Metre: Vasantatilakā. 27 Śriyām; the "i" is also rendered as "ī" on some plates. N.G. Majumdar (1929, 2) notes the retention of "m" before the letter "v", which here functions redundantly following the anusvāra. 28 Here the akṡ ara "mpa" is rendered as "pya", but this is either a scribal error or variant orthographic representation, as the phrase must be arccānām padapīṫ hikāsu here; see the Dhulla (line 4), Madanpur (line 5), and Rampal (line 4-5) plates. 32 Rendered variously as atithibhis and atitibhis, the akṡ ara "thi" is here plain and clear. The Bogra plate thus confirms the conjectural reading by Sircar (1959-60, 138) for line 9 of the Dhulla plate. 33 The akṡ ara "to" has been amended to "ro" here, as found in all other plates containing this verse. 34 Metre: Śārdūlavikrīḋita. 35 This correction also follows the Bangladesh Museum plate line 13 (Mills 1993, 79) and the Madanpur plate line 13 (Basak 1949-50, 57 With regard to the compound śrīnagaraviṡ ayasaṁ ("belonging to the district of Śrīnagara"), the akṡ ara "-saṁ " probably designates "belonging to" and functions much like a genitive or locative ending. Sircar has noted that this is a contraction of saṁ baddha, meaning "attached" or "belonging to" (1966,288); this abbreviation also appears in the Mainamati plates (1970,206  The term dāṅ ḋ ika is here added in front of the usual daṅ ḋ apāśika daṅ ḋ anāyaka. Elsewhere among the Śrīcandra plates, dāṅ ḋ ika is only found in the Paschimbhag plate (line 32). See further n. 67. 46 The underlined portion represents the only physically damaged part of the plate (aside from the general wear of ageing). It looks as if it has been struck here with some sharp object. The letters, however, remain legible. 47 Mills (1993, 83 n. 5) makes note of a discrepancy between pūti/yūti here, but the corresponding line of the Bogra plate, in my opinion, clearly has pūti. That said, the orthography of "pa" and "ya", as with some letters of this region and time period, is such that context is often needed to determine which is intended; this is especially so given the very small size of the letters. 48 Here "sā" and "śā" may be interchanged, due to metathesis, that is, a scribal error involving the transposition of adjacent letters. As discussed above (n. 24), however, the letters "śa" and "sa" are also interchanged elsewhere in the plate, as in other inscriptions; see Mills 1993, 83 n. 9;Salomon 1998, 96-7. The phonetic interchange of sibilants is not restricted to Sanskrit or even to South Asian languages but is common across other language families as well; see e.g. Young 1993, 187-92. 49 Parāśara-sagotrāya: here parāśara is the family name of the donee. See above n. 19. L. 29. ne śrāvastihastipadagrāmavinirgatabrahmapāl[ā]ya bhat˙t˙aputra 50 voḋhada 50 Datta (1989, 172) posits that Bhaṫt˙aputra is a prefixed name used exclusively for Brahmins, in a manner similar to Śarman, although the latter only ever functions as a suffix. Datta  . It is interesting to note that Bhaṫt˙aputra Nārāyaṅa (c. eleventh century CE) is said to come from "Hastipada", like the donee of the Bogra plate, suggesting a possible connection between the name Bhat˙t˙aputra and that town. 51 As in the Rampal plate (lines 27-8; Basak 1913-14, 139), the lineage is in descending order: prapautra, pautra, putra (i.e. great-grandson, grandson, son). 52 See lines 55-6 of the Paschimbhag plate; the corresponding line in the Bogra plate has koṫ ihoman . gatavate where the Paschimbhag plate reads śrāvaṅ aravisan . krāntau. The Rampal plate (lines 28-9) also attests koṫ ihoman . gatavate. Sircar (1959-60, 137) mentions the koṫ ihoma in his article on the Dhulla plate: "Pītavāsa received a grant for performing a ceremony called Kot˙ihoma". 53 Metre: Anuṡt˙ubh. 54 Metre: Anuṡt˙ubh. 55 It should be noted that "ṙ i" is not uncommon in these plates and may indicate an alternate spelling for kṙ mi rather than a scribal error per se. 56 Metre: Anuṡt˙ubh. This stanza corresponds to stanza 11 of the Bangladesh Museum plate (lines 43-5) which supports my emendations. Cf. also Salomon 1998, 95. through the worship of both of which the whole great community of monks attains the end of the cycle of existence [1]. Now, in the lineage of the powerful and prosperous Candras, rulers of Rohitāgiri, 57 was one like the full moon who was celebrated in the world as the blessed Pūrṅacandra. He was mentioned before his descendants in the pedestals of images (as well as) on victory columns, which had freshly chisel-hewn benedictions, and on copperplates 58 [2]. His son, Suvarṅacandra, a Buddhist, 59 was renowned in the world as if because he was born into the respectable family of the Moon, that is, the Lord possessing beams that are a mine of nectar, who lovingly bears the Buddha's hare incarnation stationed in his spot 60 [3]. It is said that on the new moon day his mother, on account of her pregnancy while desiring to see the orb of the moon rising, was gratified because of (the sight of) the golden moon (suvarṅ acandra) (and) so they call (him) "Suvarṅacandra" [4]. His son Trailokyacandra, the sanctifier of both families, who was the abode feared by improper conduct, 61 was known in the three worlds (trailokya) through his qualities by travellers of all directions. He who resembled Dilīpa became king on the island that has the name "Candra" prefixed to it, 62 and he was a receptacle of the good fortune that had blossomed 57 Basak (1913-14, 140 n. 11) suggests that this may refer to Rohtāsgaḋh or Rohitāsgaḋh, a hill-fort in the Śāhābād District. Luce (1969, 120 n. 24) proposes that "Rohitāgiri, meaning the 'red hill', appears only to be the Sanskritised name of the Lālmāi (lit. 'red soil') range which occupies the centre of the Tippera district", although he also notes that Sen (1942, 370) doubts this theory. 58 The self-reference to the material of the copperplate (tāmra) is interesting to note. This tendency is not necessarily seen, for instance, in other Indian literary genres (e.g. Epics, Purāṅas), which typically mention the act of recitation and remembrance of a given work, rather than its written and material form. That inscriptional media may have been conducive to such self-references, however, is suggested by parallels in early Latin inscriptions (c. 2 nd c. BCE-2 nd c. CE). See, e.g., Corpus Inscriptionum Latinarum, vol. I.2, no. 1202; vol. VI, nos. 10069, 21096, 21715; I thank Daniel Harris-McCoy for these references. 59 It is from the term bauddha that earlier scholars have proposed that Suvarṅacandra was the first in his line to become a Buddhist (e.g. G. S. Majumdar 1983, 50;R. C. Majumdar 1971: 201). For a critique of this reading see Sanderson 2009, 83. 60 The compound an . kasaṁ stham has been translated in a number of different ways.
R. Basak (1913-14, 140) renders it as "fixed in his spot"; N. G. Majumdar (1929, 7) as "in his lap"; Mills (1993: 81) breaks up the compound as "in his curve . . . in the form of a mark". I here follow Basak's translation, in that he accounts, most literally, for the content. The intention of this image, in any case, is to evoke the well-known myth, expressed across Asia, of the spots on the moon representing a rabbit. 61 It is possible to translate the element kaulīna in the compound kaulīnabhītāśayas in a couple of ways. One, positive, is "belonging to a noble family"; a second, negative, is "an evil report" or "scandal". Various translations of this compound employing one or the other meaning are found in the scholarly renderings of the corresponding verse in other plates; see Mills 1993, 81 ("refuge of the well-born"); N. G. Majumdar 1929, 7 ("that were afraid of scandal"); Basak 1913-14, 141 ("that were afraid of [the] evil report"); Mukherji and Maity 1967, 227 ("afraid of scandal"). The different translations lose either the sense of bhīta ("fear", "afraid") or of āśaya ("abode", "resting-place", as well as "disposition of mind"). 62 The implied meaning here is "the island of Candra" or "Candradvīpa", which is likely a region name. Sircar (1971, 140), for example, suggests that Candradvīpa was the name of a region of southern Bengal.
under the royal umbrella insignia of the king of Harikela 63 [5]. As Jyotsnā to Candra, Śacī to Jiṡṅu, Gaurī to Hara, and Śrī to Hari, his belovedhaving the splendour of gold (kāñcana)was to him whose commands were respected (called) "Śrīkāñcanā" [6]. Possessing Indra's radiance (and) understanding prudent policy, he (i.e. Trailokyacandra) at a moment made auspicious through a lunar conjunction attained through her (i.e. Śrīkāñcanā) a son, Śrīcandra, who resembled the moon [and] in whom astrologers observed the marks of (a future) king [7]. Having made the earth embellished with a single umbrella, he (i.e. Śrīcandra) was not obedient to fools, put his enemies into prisons, and made the four directions fragrant with his fame [8]. You should know that, 64 from the illustrious triumphal-headquarters, which were established at the blessed city of Vikrama (i.e. Vikramapura), hean excellent follower of the Buddhais the rightful successor 65 of the blessed lord Trailokyacandra and a king among great kings. Lord Śrīcandra, the highest Lord, the noble Lord, a king among great kings, is prosperous. With regard to this (one) pāṫ aka of land, (in the town of) Vyāghravoraka, belonging to the district of Śrīnagara, in the region of Samatat˙a, in the middle of the great province of Pauṅḋra, he (i.e. Śrīcandra) appropriately honours, addresses, inscribes, and commands all who have gathered together: the queen, noblemen, princes, king's ministers, regional rulers, military planners, ministers of foreign affairs, generals, record keepers, administrators, head of the king's doors, minister of forts, those who catch dangerous robbers, those who remove thieves, officiators over the fleets, elephants, horses, cows, buffaloes, goats, and sheep, etc., 66 commanders of troops, tax inspectors, junior police, sergeants, and police chiefs, 67 governors of 63 It should be noted that the genitive compound harikelarājakakudacchatrasmitānāṁ relating to śriyāṁ has been translated variously by editors and translators of the other plates of this king. Compare Basak 1913-14, 141 ("of the royal majesty smiling in the royal umbrella of the king of Harikela"); Majumdar 1929, 7 ("of the Fortune goddesses [of other kings] smiling at [i.e. joyful on account of] the umbrella which was the royal insignia of the king of Harikela"); Mukherji and Maity 1967, 227 ("of royalty, smiling in the royal umbrella of the king of Harikela"); Mills 1993, 81 ("of the power that had bloomed under the regal canopy of the king of Harikela"). 64 For a discussion of the term khalu, see M. B. Emeneau (1969, 251-62;esp. pp. 258-9). 65 In inscriptions pādānudhyāta has the sense of "is the rightful successor of", although it has been translated more literally as "contemplates at the feet of" (Mills 1993, 82) and "who meditates on the feet of" (Basak 1913-14, 141). Cédric Ferrier and Judit Törzsök have argued convincingly that such "literal" translations are in fact inaccurate and that pādānudhyāta should be considered passively in the sense of "favoured/blessed by the respected" (2008,93), with pāda being understood only as an honorific title, at least until after the tenth century when the practice of foot-veneration came to be more widely practised (2008,99). 66 Sircar (1966, 213) has a discussion of the compound found here, naubala-hasty-aśva-go-mahiṡ -āj-āvik-ādi-vyāpṙ ta, suggesting "an officer in charge of the royal navy and cattle", etc. 67 See n. 45 above. This series of terms (i.e. "junior police, sergeants, and police chiefs" corresponding to dāṅ ḋ ika-daṅ ḋ apāsika-daṅ ḋ anāyaka) probably stands for categories of disciplinarians or those who extract fines. Dāṅ ḋ ika possibly means "one who inflicts punishment". Sircar (1966, 80) suggests that daṅ ḋ anāyaka is "probably a translation of Greek Strategos; a general; a leader of forces, an army officer". He further suggests that daṅ ḋ apāśika is "a police officer; an official who was probably the leader of a group of Dāṅ ḋ ikas; officer in charge of punishment (i.e. criminal justice), according to some" districts, and the like, as well as other servants of the king: those said to work as inspectors and those who were not announced here: members of the communities of Cāt˙as and Bhat˙as, 68 citizens, farmers, and leading Brahmans. It must be understood by all of you (present) that this land, as inscribed (in this plate) above, which is limited by its own boundaries, is flanked on all sides with pastures of pūti-grass, with its low-lying and elevated land, along with its mango, breadfruit, betel-nut and cocoa-nut trees, with salt (derived from sea water), with (fresh) water and dry land, with its salt-rich soils and caves, 69 with the (removal of) the ten faults, with the eradication of robbers, with all sorts of impositions removed, it is without interference from (the communities of) Cāt˙as and Bhat˙as, free from taxation which is associated with levies of gold on all royal subsidies (is granted by us), having first performed the water ceremony according to the rules, (I grant this land) to the one who officiates the Kot˙ihoma ritual, the blessed Śrīkaradatta Śarman, who is the son of Varadatta, grandson of Varadatta, great-grandson of Bhat˙t˙aputravoḋhadattaśarman, and a protector of Brahmans, 70 who hails from the village of Hastipada in the Śrāvasti region.
(p. 81). He connects dāṅ ḋ ika also with cauroddharaṅ ika, meaning "an officer in charge of the recovery of stolen property; a police officer" (p. 72). There is probably a related, hierarchical, relationship between the three categories in any case. See further Sircar 1966, 82;Mills 1993, 84 n. 30. 68 Regarding the compound cāṫ abhaṫ ajātīyān, the word jātīya is treated by Sircar (1966, 134, 136) under "jñātīya"; he defines both terms in the sense of "belonging to a community", as typically used in compounds (e.g. Prāgvāṫ ajñātīya, "belonging to the Prāgvāt˙a community"). Accordingly, I here translate Cāt˙a and Bhat˙a as communities of people. Typically these names are left untranslated because of the ambiguity of the terms, which can denote thieves or undesirables as well as those who remove such elements. However, treating them as peoples may avoid some of this confusion. On various usages of cāṫ a and bhaṫ a, see e.g. Sircar 1966, 67-8. 69 Here I have translated ūṡ ara (from the compound sagartoṡ arā) as meaning "salt-rich soil"; contrast the more common rendering of the term as "barren tracts" or the like in translations of other Śrīcandra plates (e.g. Mukherji and Maity 1967, 228). While brackish salt (lavaṅ a) has already been listed, it should be noted that the Bengal regionand Samatat˙a in particularwas renowned for salt resources in a variety of forms. It has been argued, for example, that medieval salt production in Bengal was more diverse than it is today, with salt derived from ocean waters, from the ground, and possibly from rocks (Jahan 2006, 64-5). Salt seems to have been a major trading commodity in the region (see further, e.g., Ahmad 1994, 101;1990, 174, 184-7;R. C. Majumdar 1971, 342). In my view, it is thus plausible that more than one kind of salt might be mentioned in a description of a plot of land. In any case, given the context of this passage, which relates the bounty and benefits of the granted land (mango trees, etc.), it seems to me unlikely that "barren tracts" would be included in the description; accordingly, I propose that what is here intended is to suggest bounty that derives from the production of salt for trade and the like. See also n. 72 below regarding the related issue of the interpretation of the compound bhūmicchidranyāya. 70 The precise meaning of -brahmapāla here is uncertain. I have suggested that the donee, Śrīkaradatta, is perhaps some kind of leader among Brahmans and thus deserving of the epithet "protector of Brahmans"; however, Śrīkaradatta is the only Brahman named on the plate. The term brahma could also be a reference to speech or poetry and, thus, to the Vedas. I have found no parallel use of this phrase among contemporary plates other than in the context of kings, i.e. the name of a king from the Kāmarūpa district, He (i.e. Śrīkaradatta) is a student of the Vedas belonging to the Kuthumin school (kauthumaśākhā), part of the Chandoga school (caraṅ a), as well as the ancestry ( pravara) connected to three Ṙ ṡis, and belonging to the Parāśara lineage (gotra). 71 In the name of the venerable lord Buddha, for the increased merit and fame of our parents, until the time of the destruction of the sun and the moon, (this land) is given by us according to the laws of land-division (bhūmicchidranyāya), 72 recording it in a copperplate grant with the seal of the blessed Dharmacakra. Henceforth, by everyone, this gift of land is to be respected; by future kings, in particular, it is to be recognized and protected, in the case of (attempts at) taking it back, on account of the importance of the potential of the gift and for fear of falling into the Great Hell (Mahānaraka); and by farmers dwelling there, who are obliged to hear this command, (this gift of land) is to be made to bring about enough yield as is appropriate. Furthermore, here are some verses in regards to relating the (wisdom of) Dharma 73 [8]: Both the one who receives this land and the one who gives the land produce merit and are sure to go to heaven [9]. Land has been given by many kings from Sagara onwards. Whosoever owns land, that one gains its benefits [10]. He who gives back this gift or who takes (the land) of another man shall be tormented along with his ancestors, after turning into a worm in excrement [11]. 74

Appendix A
To supplement the transcription of the Bogra plate above, I here provide a critical rendering, in Devanāgarī, of the text, emended to correct probable scribal errors, etc. Emendations have been made in consultation with the six other published Śrīcandra plates, and relevant parallels have been compiled in the notes.
Brahmapāla (r. mid-10 th c. CE); Sircar 1978, 9-11;G. S. Majumdar 1983, 28;Singh 1994, 88. 71 For a discussion of various usages of the terms ārṡ a, pravara, and gotra, see Brough 1953, 1-7. 72 O. von Hinüber (2005 has recently surveyed the history of research on the term bhūmicchidranyāya, re-examining the widespread view that it is a maxim concerning the transfer of permanent ownership of land to the one who cultivates it for the first time. He demonstrates that there is little reason to interpret the term as referring only to barren or fallow land. In his view, the maxim does not imply that the land granted is fallow; rather, it is oriented only towards emphasizing the longevity of the donation itself (p. 495). Notably, von Hinüber's understanding of the term is consistent with its use in the Bogra plate, where the granted land is described in terms that emphasize the many resources already on it (see above). 73 What follows is a version on the common imprecatory verse-types found often at the end of copperplate inscriptions. In other inscriptions, similar verses are attributed to the poet Vyāsa and associated with the Mahābhārata. Interestingly, verses 10 and 11 exhibit some parallels with passages in the Southern recensions (i.e. Telugu and Garntham) of the Āśvamedhikaparvan (see, e.g., Appendix no. 4 of the critical apparatus lines 1109 and 1125-6, corresponding to 11a-b and 10a-d of the copperplate respectively). 74 The particular reference to "a worm in excrement" can be found in other medieval sources (e.g. Garuda Purāṅ a 1.229.3; 2.42.15).
Of the six plates, four correspond closely to the Bogra plate, namely the Bengal Museum (BM), Rampal (R), Dhulla (D) and Madanpur (M) plates; for these, parallels and variants will be listed in the notes. Parallels in the Kedarpur (K) and Paschimbhag (P) plates are more limited and will be treated more selectively. Only two passages in the Kedarpur plate overlap significantly with the Bogra plate (i.e. stanza 1 and the opening of the prose section, where the Kedarpur plate ends), whereas the Paschimbhag plate includes close parallels only to stanzas 1, 9 and 11, as well as to portions of the prose section. 75 When relevant, however, it will be noted where the Kedarpur and Paschimbhag plates share passages with the Bengal Museum, Rampal, Dhulla, and Madanpur plates.