This article highlights a parallel between a verse-quarter of the Sanskrit colophon of the Tanjung Tanah codex, the oldest Malay manuscript known so far, and the opening verse of the Bhuvanakośa, a Sanskrit-Old Javanese text from Bali. On the basis of the parallel, and of the Malay gloss, I suggest an alternative reconstruction of the verse-quarter to that proposed by Arlo Griffiths (2010). Having traced the parallel back to identical verse-quarters from South Asian texts, and building on the recent discussion by Hunter (in Kozok forthcoming), I advance the view that premodern textual materials from Sumatra and Bali, although originating from distant geographical areas, stemmed from a common, supralocal Indic tradition.
An Indonesian-language critical edition-cum-translation and study of the so-called ‘Tanjung Tanah code of law’ was published by a team of scholars led by Uli Kozok in 2006.1 The 30-page codex unicus presented in the work, originating from Dharmasraya but found in Tanjung Tanah, further south in the Kerinci highlands of Central Sumatra, contains a legal text written in a form of Indic script and pre-Classical Malay language.2 Being the oldest Malay manuscript currently known (ca. fourteenth or early fifteenth century), this codex is to be regarded as a tremendously important document of pre-Islamic Malay literary culture.
Griffiths (2010) is a detailed and critical review of Kozok (2006). In the concluding part of his useful review, Griffiths (2010:137) made some significant remarks on the codex’s colophon:
The colophon is another aspect of the codex that clearly stands in a scribal tradition inherited from India. This is not the place for a full interpretation of its text, but what seems clear is that the sequence 30.5–31.2 pranamya śrisa divam· tṛlūkyadipāti stutim·, ṇāṇasattrudṛtaṃ vakitnitrisatrasamukṣayam, which contains the title, may be restored to acceptable grammatical and metrical form, and then translated, as follows:
praṇamya śrīmahādevaṁ trailokyādhipatistutam nānāśāstroddhr̥taṁ vakti nītisārasamuccayam
Having bowed to Śrī Mahādeva, who is praised as the Lord of the Triple World, [the author] expounds the Compendium of the Essence of Policy, extracted from various authoritative sources.
Griffiths’s preliminary analysis, and restoration, of the relevant Sanskrit portion of the colophon is insightful, and makes amends for the unsatisfactory treatment it received from Kozok and his team. In particular, his identification of the verse as a eulogy to Śiva, as well as of what seems to have been the original title of the text, namely ‘Compendium of the essence of policy’ (Nītisārasamuccaya), takes the matter to a whole new level, as it pinpoints the fundamentally Indic nature of the scribal tradition from which it stems. While building on Griffiths’s discussion, as well as on the recent re-analysis of the Sanskrit verse and its Malay gloss (saluka dipati) by Thomas Hunter (in Kozok forthcoming), I suggest an alternative reconstruction of the first quarter of the verse, which is in harmony with the Malay gloss.3 In doing so, I will proceed on the basis of philological considerations, and especially of a textual parallel found in a Sanskrit-Old Javanese Śaiva source—the Bhuvanakośa—that has been transmitted to us through Balinese manuscripts. The discovery of this parallel verse-quarter in the Bhuvanakośa, as well as in other Balinese and South Asian Sanskrit sources, carries wider implications for the history of pre-Islamic—that is, Indic—literary cultures of Nusantara, suggesting as it does some fundamental analogies between the text-building and commentarial practices of the early modern scribal milieux of Sumatra and Bali.
The Sanskrit Verse and Its Reconstruction
Griffiths’ reconstruction of the corrupt (and unmetrical) verse-quarter pranamya śrisa divam into praṇamya śrīmahādevaṁ is acceptable on palaeographic grounds, and yields a perfectly metrical and grammatical clause. However, an alternative reconstruction, which has the advantage of being more faithful to the original text, would be praṇamya śirasā devaṁ, meaning ‘Having bowed down with the head before God’. The verse-quarter can be analysed as follows:
Griffiths assumes the substitution of an original ma with sa (of admittedly similar shape in the script) and the loss of a ha. The solution I envisage implies a somewhat simpler and altogether common process, namely: 1. the transformation (inversion) of śir into śri—śrī being a recurrent honorific prefix meaning ‘illustrious’; 2. the loss of the short a that followed the original sequence śir. This loss of a might have been triggered either by the inversion śir>śri or, conversely, the inversion might have taken place in an intermediate form: śīrṣa. The word sirsa does indeed appear in the Malay gloss—śīrṣa being a phonologically similar stem-form that is historically related to the Sanskrit śiras, ‘head, skull’. It must be noted that the word śrisa features also in the line immediately preceding the start of the colophon, where we read the (largely similar) sequence pranamya divaṁ śrisa maleśvaraṁ. Here maleśvaraṁ is likely to represent a corruption of maheśvaraṁ4 (the graphemes h and l being orthographically similar, and often confused in Nusantarian Indic scripts); thus, the sentence can be restored to praṇamya devaṁ śirasā maheśvaraṁ, meaning ‘Having bowed down with the head before God, the Great Lord (Maheśvara)’. Kozok and his team (2006:100, fn. 73) noted that both occurrences of śrisa—which to them must be read as sirǝsa, in the light of the Malay gloss—reflect a Sanskrit śīrṣa, ‘head’.
Griffiths’s emendation of the corrupt second half-verse ṇāṇasattrudr̥taṁ vakit nitrisatrasamukṣayam into nānāśāstroddhr̥taṁ vakti nītisārasamuccayam is satisfactory. However, the -sattr(a) occurring in two compounds in both the first and second quarters poses a philological riddle, multiple solutions to which are possible. While the emendation -śāstra constitutes the more natural outcome, one should also take into account, as done by Griffiths in the case of the second verse-quarter, -sāra. In fact, one notes the frequent occurrence of the cliché -sāroddhr̥ta (‘extracted from/as the essence’) on the one hand, and the relative rarity of -śāstroddhr̥ta (‘extracted from the treatises’) on the other, in Sanskrit texts of various genres from South and Southeast Asia, as well as in Old Javanese literature.5 Zoetmulder’s Old Javanese-English dictionary (1982:1697) glosses sāroddhr̥ta as: ‘(Skt) “the selected essentials of the books of learning”, name of a work containing rules for the conduct (policy) to be followed; cf Sārasamuccaya’, but does not give any instance of śāstroddhr̥ta.6 This fact suggests to me that nānāśāstroddhr̥taṁ in the third verse-quarter could be a shortened versified version of an original *nānāśāstrasāroddhr̥tam (‘extracted from/as the essence of various treatises’), where the element -sāra- was dropped to suit the metre. The compound śāstrasāroddhr̥ta is attested in the Tantri Dəmuṅ,7 as well as in the full title of the legal code Pūrvādhigama, that is, ‘Purvadhigama Sasana Sastra Saroddherta’.8 Conversely, -sattra in the third verse-quarter could be alternatively rendered as śāstra, namely Nītiśāstrasamuccayam. While the sequence (nīti)sāra(samuccaya) is well-attested in Javano-Balinese literature—witness the Sanskrit-Old Javanese legal code titled Sārasamuccaya, and the Old Javanese Nītisāra—the word nītiśāstra is also found in Old Javanese,9 as well as in Sanskrit.10 More importantly, Monier-Williams’s Sanskrit-English dictionary (s.v. Nītiśāstra) records a work named Nītiśāstrasamuccaya. Whatever the original form of the second half-verse might have been, the idea it conveys is clear: the work in question is a synthetic compendium containing material extracted from many legal codes.
The Malay Gloss
A set of Malay glosses on the saluka dipati is provided by Kuja Ali, the alleged author or re-compiler of the Tanjung Tanah codex. As argued by Hunter (in Kozok forthcoming), these glosses contain a number of lacunae and ‘imperfections’, which reflect the long process of ‘drift’ undergone by the local scribal tradition of enunciation, orthographic representation, and exegesis from the Indic tradition from which the codex, and especially its colophon, stem.
A close reading of the word-by-word Malay gloss supports the reconstruction of the first quarter of the Sanskrit verse proposed above. The text runs as follows:11
Pranammya nāma, tunduk mañambaḥ, sirsa na[ma] kapāla, diva nama divata.
Pranammya (<praṇamya) means: to bow the head and pay homage. Sirsa (<śirasā) means:12 head. Diva (<devaṁ) means: a divinity.
Notwithstanding the fact that all the Sanskrit words appear in a ‘corrupted’ or, rather, ‘localized’ form, their Malay glosses are correct, and restitute what must have been the original import of the verse-quarter, namely ‘to pay a respectful homage by bowing down before the deity with the head’—or, more idiomatically, ‘having bowed the head before God’.
The remaining part of the gloss (as transcribed, translated, and paraphrased by Hunter in Kozok forthcoming) runs as follows:
Many of the Malay glosses are consistent with the meaning of the Sanskrit original, but in some cases a ‘shift’ has occurred. As noted by Hunter (in Kozok forthcoming), in dipati there is an ‘expansion’ of the meaning of the original Sanskrit adhipati (‘ruler, king’, here referred to as God, ‘the paramount in the Triple World’) that underlines the superior status of the bearers of this title of political rank. The gloss of dhr̥tam does not capture the meaning of uddhr̥tam (‘extracted’), but ‘refers rather to the aspect of speaking that can be understood as recoverable from the overall sense of the phrase nānāśāstroddhr̥taṁ vakti, “he speaks that which has been extracted (raised up) from many authoritative texts”’. While to Hunter the gloss of satra seems to have been incorrectly understood by Kuja Ali in terms of a local ‘tadbhava form’ of kṣatriya, ‘member of the warrior caste’, it is also possible that here satra—also attested as such in the Sanskrit verse—could be a ‘tadbhava form’ of śāstra, which Kuja Ali could have understood in terms of an (authoritative/normative) ‘text’ or ‘manual’ (satra nama yang satra = ‘that which is a śāstra’). I also add that the gloss of samuksayam (< Skt samuccayam) as ‘all and everything’ cannot be said to be, strictly speaking, incorrect, for the Sanskrit samuccayam, besides ‘compendium, collection’, can also mean ‘totality, aggregate’.
If we take into account, with Hunter, the fact that Kuja Ali ‘was working with a Sanskrit original that had already gone through processes of drift in terms of both phonological shape and meaning’, and, at the same time, that ‘his understanding of the local Sanskrit form of the Sanskrit original was imperfect, so that he only partially succeeded in giving an expert gloss’, the result is something remarkably similar to
the typical text-building strategies of a long tradition of didactic texts that may have been pioneered in the Buddhist institutions of Śrīvijaya, and to modern practices of extemporaneous glossing like those of the ‘reading clubs’ (Pesantian, Sĕkaha Mabasan) of Bali (Hunter in Kozok forthcoming).
What Hunter means by ‘extemporaneous glossing’ are hermeneutical strategies that do not (primarily) involve analytical means, drawn from a systematic knowledge of morphology and grammar, or try to establish historical derivations and etymologies; they are mainly based on contemporary/popular lexical know-how, ‘folk etymology’, and associative thinking effected through homology, metaphor, and assonance. It may be argued that these hermeneutical practices are not uniquely the expression of a ‘local’ or ‘embedded’ milieu, but represent a continuation, or ‘demoticization’, of the codified norms of semantic analysis and argumentation that characterized the Sanskrit śāstric tradition.
The Balinese and South Asian Parallels
An occasion to link a fourteenth-century Sanskrit-Malay document to the hermeneutical and text-building practices found in Sanskrit-Old Javanese textual sources that have come down to us from the Balinese scribal tradition is provided by a parallel in the Śaiva tutur text Bhuvanakośa (The storehouse of worlds). This voluminous work, consisting of some 200 Sanskrit verses provided with Old Javanese glosses, is of uncertain date. While my preliminary reading would date this text to the fifteenth or even early sixteenth century, extensive portions of it are likely to stem from several centuries earlier.13
The opening Sanskrit verse of Chapter One presents a quarter that is identical to the first quarter of the verse of the Tanjung Tanah codex. The verse and the ensuing Old Javanese commentary run as follows:
praṇamya śirasā devaṁ14| vākya(ṁ) munir amanmathaḥ15 | devadeva mahādeva | parameśvara śaṅkara || 1.1
Having bowed down with the head before God, the sage who is free from (erotic) desire spoke: ‘O God of Gods, O Great God, O Supreme Lord, O Śaṅkara!’
śrī muni bhārgava | sira mahyun tumakvanakǝn ikaṅ pada nirbāṇa ri bhaṭāra | maṅkana pvābhiprāyanira | manambah ta sira ri bhaṭāra | śirasā | makakāraṇa hulunira sira | ri tlasnira manambah | mojar ta sira || he devadeva | kita devaniṅ devata kabeh | he mahādeva | kita bhaṭāra mahādeva ṅaranta | he maheśvara | kita bhaṭāra maheśvara ṅaranta | he śaṅkara | kita bhaṭāra śaṅkara ṅaranta ||
The sage Bhārgava desired to ask the Lord about the stage of extinction. Such was his intention. He worshipped the Lord with the head—he used his head. Having finished worshipping, he spoke: O God of Gods—you are the God of all the Gods! O Mahādeva—You are Lord Mahādeva by name! O Maheśvara—you are Lord Maheśvara by name! O Śaṅkara—you are Lord Śaṅkara by name!
The Old Javanese active verbal form manambah glosses the Sanskrit absolutive praṇamya, ‘having worshipped’; the Sanskrit śirasā is quoted in the commentary and glossed as makakāraṇa hulunira sira, ‘he used his head’, literally ‘he used as the instrument his head’. The gloss presents an exemplary semantical and grammatical analysis of the Sanskrit in the first verse-quarter.16
Insofar as it obviously represents an introductory eulogy to (various manifestations of) Śiva, the verse mirrors the colophon of the Tanjung Tanah codex with respect to both context and function. The two passages contain a verbal praise (stuti) of, and represent a physical act of worship towards, a paramount deity—that is, Śiva17—by either the text’s author or, in the case of the Bhuvanakośa, the main interlocutor of Śiva and ‘revealer’ of the text. As we have seen, both the Malay and Old Javanese glosses of the identical first quarter of their Sanskrit verses can be considered as faithful renditions of its meaning.
Additional evidence comes from the corpus of Sanskrit hymns (stuti, stava) handed down to us through Balinese palm-leaf manuscripts. The opening verses of several hymns attest the sequence praṇamya śirasā + name of the deity (for instance, rudram, sūryam, viṣṇum) to which the hymn is devoted.18 The verse-quarter praṇamya śirasā liṅgaṁ, found in the first stanza of the Liṅgastava (Stuti 670), is accompanied by an Old Javanese gloss:19
praṇamya śirasā liṅgaṁ divyaliṅgaṁ maheśvaram | sarvadevātidevanaṁ tasmai liṅgāya vai namaḥ || 1
ai kita bhaṭṭāra, (t)aṅa-taṅa tiki səmbah iṅ ulun riṅ bhaṭṭāra, jə̄ṅnira umuṅgv iṅ śirah. Paran ta don iṅ ulun magave namaskr̥ta riṅ bhaṭṭāra śivaliṅga? Kumva kadivyan bhaṭṭāra, an kita maśarīra liṅga viśeṣa, sājñā bhaṭṭāra. Sarva devatī devatā paśarīra bhaṭṭāra. Saṅ muṅgv iṅ liṅga bhaṭṭāra sinəmbah.
I bow my head down to the earth before the Liṅga, the divine Liṅga which is the Great Lord, the superior Lord of all Lords; to that Liṅga homage be rendered.
O Lord, pay attention to this salutation of mine to You! Your feet are before my head. What is my purpose in performing a reverent salutation to the Lord Śivaliṅga? With your permission, thus is Your divine quality, as You embody the supreme Liṅga: All the goddesses and gods form Your body. The [goddesses and gods] who dwell in Your Liṅga are worshipped.
The Old Javanese gloss depicts the image of a worshipper performing a sǝmbah by literally bowing down with the head to touch the feet of the paramount Lord Maheśvara (that is, Śiva) embodied in the Liṅga, which is obviously a rendering of the Sanskrit verse-quarter praṇamya śirasā liṅgaṁ. The passage features clear similarities to the glosses found in both the Tanjung Tanah codex and the Bhuvanakośa.
The verse-quarter praṇamya śirasā devaṁ is found in the similarly eulogistic first stanza of the Gaṇapatistava (Stuti 667), a hymn dedicated to the elephant-headed god Gaṇapati, son of Śiva. As pointed out by Goudriaan and Hooykaas (1971:390–91), this hymn finds a close parallel in the Saṁkaṣṭanāśanagaṇeśastotram, a Sanskrit hymn of South Asian provenance published in the Br̥hatstotraratnākara collection. Besides in the Sanskrit hymn, an identical verse-quarter is also attested in a number of Sanskrit sources, spanning from the Mahābhārata and Rāmāyaṇa epics to the Atharvavedapariśiṣṭa, various Purāṇa and Śaiva and Vaiṣṇava Tantric texts.20 This fact shows that the form praṇamya śirasā devam was not a Nusantarian invention, but a cliché originating from South Asian Sanskrit sources.
My analysis has confirmed the following: 1. the first verse-quarter of the Sanskrit colophon of the Tanjung Tanah codex can be restored on the basis of a parallel found in the opening verse of the Sanskrit-Old Javanese Bhuvanakośa, as well as in several other Sanskrit texts from Bali and South Asia; 2. the restored reading of this quarter of the Sanskrit verse is in harmony with its Malay gloss. More importantly, the existence of these formal as well as contextual parallels suggests that Sanskrit materials from the highlands of central Sumatra and the island of Bali, although separated by a significant geographical (and perhaps, although not necessarily, chronological) distance, are witness to the existence of a supralocal textual tradition of, ultimately, a South Asian origin and a theistic (Śaiva) character. Although the distinct vernacular glosses discussed here were produced in different local contexts, characterized by different histories of reception, socio-cultural factors, and religious contingencies, they appear to have developed—in parallel, as it were—from what must once have been a common tradition of text-building and exegetical practices. This tradition might have had its heyday in the premodern Indicised cosmopolitan courts of Sumatra and Java, and, as suggested by Hunter (in Kozok forthcoming), might have continued—in a transmogrified form—to the present day through contemporary Balinese ‘reading clubs’.
Acri, A. (2012). ‘Bhuvanakośa. Critical edition and English translation’. [Unpublished manuscript.]
Acri, A. and A. Griffiths (2014). ‘The romanization of Indic script used in ancient Indonesia’, Bijdragen tot de Taal-, Land- en Volkenkunde 170:365–78.
Budha Gautama, Wayan (2009). Tutur Bhuwana Koṣa. Surabaya: Paramita.
Creese, H. (2009). ‘Judicial processes and legal authority in pre-colonial Bali’, Bijdragen tot de Taal-, Land- en Volkenkunde 165:515–50.
Goris, R. (1926). Bijdragen tot de kennis der Oud-Javaansche en Balineesche theologie. Leiden: A. Vros. [PhD dissertation, Leiden University.]
Goudriaan, T. and C. Hooykaas (1971). Stuti and Stava (Bauddha, Śaiva and Vaiṣṇava) of Balinese Bauddha Brahmans. Amsterdam/London: North Holland Publishing Company.
Griffiths, A. (2010). ‘Review of Uli Kozok (2006), Kitab undang-undang Tanjung Tanah: Naskah Melayu yang tertua’, Bijdragen tot de Taal-, Land- en Volkenkunde 166–1:133–8.
Hooykaas, C. (1964). Āgama Tīrtha; Five studies in Hindu-Balinese religion. Amsterdam: Noord-Hollandsche Uitgevers Maatschappij.
Hunter, T.M. (forthcoming). ‘Sanskrit in a distant land: The Sanskritised sections’, in: U. Kozok (forthcoming). The Tanjung Tanah codes of law [with contributions by Thomas Hunter, Waruno Mahdi, and John Miksic], Chapter 7. Singapore: ISEAS Publishing.
Juynboll, H.H. (1911). Supplement op den catalogus van de Javaansche en Madoereesche handschrif-ten der Leidsche Universiteitsbibliotheek; Deel 2: Nieuwjavaansche gedichten en Oud-, Middel- en Nieuwjavaansche prozageschriften. Leiden: Brill.
Kozok, U. (2004). The Tanjung Tanah code of law: The oldest extant Malay manuscript. Cambridge: St Catherine’s College and the University Press.
Kozok, U. (2006). Kitab undang-undang Tanjung Tanah: Naskah Melayu yang tertua. Jakarta: Yayasan Naskah Nusantara/Yayasan OBOR Indonesia.
Kozok, U. (forthcoming). The Tanjung Tanah codes of law [with contributions by Thomas Hunter, Waruno Mahdi, and John Miksic]. Singapore: ISEAS Publishing.
Mirsha, I Gusti Ngurah Rai et al. (1994). Buana Kosa. Alih aksara dan alih bahasa (Brahma Rahasyam). Denpasar: Upada Sastra.
Monier-Williams, M. (1899). Sanskrit-English dictionary. Oxford: Clarendon Press.
Sanderson, A.G.J.S. (2003–2004). ‘Śaivism among the Khmers; Part I’, Bulletin de l’École française d’Extrême-Orient 90/91:349–462.
Sternbach, L. (1975). ‘General appeal of Subhāṣita literature in Sanskrit’, in: V. Raghavan (ed.), Proceedings of the First International Sanskrit Conference, pp. 370–97. Delhi: Rashtriya Sanskrit Sansthan.
Zoetmulder, P.J. (1982). Old Javanese-English dictionary [with the collaboration of S.O. Robson]. ’s-Gravenhage: Nijhoff. Two vols. [KITLV.]
I thank Arlo Griffiths for his useful comments on a draft version of this article. Unless otherwise specified, the romanization of Old Javanese (and pre-Classical Malay) conforms to the norms adhered to by the present author in previous publications, which are discussed in Acri and Griffiths 2014.
See Kozok 2006. An English-language monograph of the same title was published by Uli Kozok in 2004. A substantially revised and expanded English-language monograph by a team of authors led by Kozok is currently in press (see Kozok forthcoming). I thank Thomas Hunter for making accessible to me his chapter in the aforementioned publication, of which I have made grateful use here.
As the form of language preserved by the text is still free of the Arabic loanwords that characterize Classical Malay, it may be concluded that Tanjung Tanah Malay is an intermediate form of Malay standing somewhere between Old and Classical Malay.
With hindsight, one may partially revisit Griffiths’s (2010:138) claim that ‘[t]he interesting Sanskrit/Old Malay gloss, which terminates this principal part of the codex, is not a reliable guide for interpreting the colophon’. Admittedly, the preliminary treatment offered by Griffiths in his book review was not the place for a philological justification of the proposed restoration of the verse, or for an analysis of its Malay gloss. Hence, my purpose here is only to point at the existence of parallels that are helpful in restoring the first verse-quarter, and at the same time tell us something about the supralocal dynamics that shaped the literary cultures of premodern and early modern Nusantara.
Pace Kozok and his team (2006:100 fn. 74), who suggest to read amaleśvara, namely amala (‘spotless’) + īśvara (‘Lord’). The occurrence of the sequence śrisa maheśvara in this line is presumably the reason why Griffiths restored śrīmahādevaṁ in the verse.
See, for example, the Sanskrit portion of the Sdok Kak Thom inscription, which speaks of the Brahman Hiraṇyadāman who ‘carefully extracted the essence of [various Tantric] texts’ (dvijas samuddhr̥tya sa śāstrasāraṁ)—‘evidently a manual for practical application, a Paddhati or Pañjikā’ (Sanderson 2003–2004:356). The Śaiva Siddhāntatantra Sarvajñānottara (Jñānapāda, 1.4–5) characterizes itself as ‘a compendium of all scriptures’ (sarvaśāstrasamuccayam), the essence extracted from all scriptures (sarvaśāstreṣu sāram uddhr̥tya), just like honey is the essence extracted from all flowers. See also: vedasāroddhr̥taṁ, Jñānāmr̥tasārasaṁhitā 1.1.83; kātyāyanasmr̥tisāroddhāraḥ, Kātyāyanasmr̥ti 1; sarvaśāstrārthasāroddhāra, Ṣaṭsaṁdarbha 4.73; viśvasāroddhāre, Padārthādarśaṭīkā ad Śāradātilaka ch. 20, passim. For śāstroddhr̥ta/śāstroddhāra, see: śāstroddhāram, Mokṣopāyaṭīkā 1.1.3, and subhāṣitāni … śāstroddhr̥tāni, Mahāsubhāṣitasaṅgraha 7546.
On the basis of Sorāndaka 1.64, 2.61; Tantri Dəmuṅ 4.20b, 24a, 29b, 1.3b (saroddhr̥ti).
See 1.3b: Aji Uttara-manava mvaṅ sastrasarodrəti; 4.24a, 29b: tan tut sastra sarodrəta manuh don iṅ atəmah kevuh (Mānavaśāstra) (as cited in Zoetmulder 1982, and standardized according to the romanization system used by Acri and Griffiths ).
See codex LOr 5023 = HKS 5268, Purwadhigama Siwasasanasastrasaroddhreta (mentioned in Creese 2009:543), LOr 5098 (ff. 1–18a), Purwadhigama Sasana Sastra Saroddherta (Juynboll 1911:191), LOr 3723 (ff. 1–18. Purwadhigama, in the colophon Siwasasana Saroddherta, Juynboll 1911:191), and LOr 6203 (transcript of mss. KBG Brandes No. 467 and No. 478; see also LOr 3852), Purwadhigama, Siwa Sasana Saroddhrta, also containing a chapter captioned Sara Samuccaya. One Adhigamasasanasastroddhreta is documented in various textual and inscriptional accounts mentioned by Creese (2009:535), including the aforementioned Old Javanese legal code Pūrvādhigama, which begins with the phrase ‘nihan Purwadhigama sasana sarodreta sastra purwaramba sang telas wredhacarya sang purohita’, and also attests the sequence ‘sang hyang adigamasastrasarodreta’ in its preamble (romanization as in Creese 2009).
See Zoetmulder (1982:1200): ‘(Skt) the science of or a work on political ethics Sut[asoma] 4.5: varahən riṅ nītiśāstra; TK 34.23: mahyun iṅ vivāhakrama saṅke śāstrāgama mvaṅ nītiśāstra; 16.27’. In fact the Nītisāra and Nītiśāstra are probably the same work (see Sternbach 1975:394).
This may be either a general term meaning ‘the science of political ethics’ or the title of a specific work (or a class thereof). Several works bearing this sequence as part of their title are known to exist in the Sanskrit and Tibetan tradition (see Sternbach 1975:370–3).
My transliteration differs slightly from the ‘transliterasi diplomatik’ by Kozok and his team (2006:101), which reads pranemmiya nama, tunduk mañambah, sirsa na[ma] kapala, diwa nama diwata.
The use of nama in precisely the function and position of ṅaran (‘name’, ‘with the name’/‘called’ > ‘means’) in Old Javanese glosses is noteworthy. Kozok and his team insert °ma so as to fill a gap between °na and ka° in the codex. However, as Griffiths suggests (email dated 13-01-2014), the use of na as an abbreviation of nama may be identical to the very common use of ṅa as an abbreviation of ṅaran(ya) in Old Javanese texts.
Although the Bhuvanakośa was studied in the first half of the twentieth century by the philologists Goris and Zieseniss, neither a critical edition nor an English translation have been published. The versions that have been published in Bali in recent years, which come with an Indonesian rendering of the Old Javanese portions (see, for example, Mirsha et al. 1994 and Budha Gautama 2009), appear to all be based on a single (as a rule, unspecified) source, namely a (rather faulty) romanized, typed transcription of a palm-leaf manuscript belonging to the collection of the Pusat Dokumentasi Budaya Bali in Denpasar. Although these publications are of great importance for the ‘socialization’ of the Hindu religion in the contemporary Balinese community, they are not very useful for scholarly purposes. Here I have used my own provisional critical edition of the text (Acri 2012); an edition of the first six Sanskrit verses plus Old Javanese glosses of the text, on the basis of the single codex LOr 5022, may be found in Goris 1926:78–81.
Deva is my emendation of deva, found in all the manuscripts that I have consulted.
Amanmathaḥ is my emendation of amanmatha, found in all the manuscripts. The Sanskrit manmathaḥ means ‘agitating’, ‘love or the god of love’, ‘amorous passion or desire’, therefore the adjectival compound amanmathaḥ could mean ‘he who is free from desire’. The form vākya(ṁ), from √vac, ‘to speak, declare, proclaim’, et cetera, is problematic. The verse-quarter could be the result of a contamination between a nominal sentence with nominative and genitive (vākyaṁ muner amanmathasya, which is, however, hypermetrical), where vākya > vākyaṁ is the substantive ‘speech’, and a verbal sentence (vakti munir amanmathaḥ); the present form vakti, it should be noted, also occurs in the Sanskrit colophon of the Tanjung Tanah codex. In Bhuvanakośa 1.3c we find vākyan te, which is glossed as nihan vuvusan iṅ hulun i kita (‘my speech to you is as follows’).
Yet, as Arlo Griffiths suggests (email dated 13-01-2014), the Old Javanese gloss mahyun (‘desired’) might represent the Sanskrit amanmathaḥ (‘free from [sexual] desire’) in the second verse-quarter, where the commentator will have taken a- as an Old Javanese prefix rather than as the Sanskrit alpha privans that it is.
Note the mention of Mahādeva in the line preceding the colophon of the Tanjung Tanah codex, and the epithets Mahādeva, Parameśvara, and Śaṅkara in the Bhuvanakośa.
See Rudrakavaca (Stuti 673 and 676), Dvādaśādityastava (Stuti 679), and Viṣṇustava (Stuti 682) in Goudriaan and Hooykaas 1971.
Ms. K 1843, ed. Hooykaas 1964:151; the translation of the Sanskrit verse is by Hooykaas, while that of the Old Javanese gloss is my own.
See Atharvavedapariśiṣṭa 70b.1.1, Bhagavadgītā 11.14, Mahābhārata 5.129.13, Rāmāyaṇa 7.10.15, Brahmapurāṇa 45.17, 176.28, Nāṭyaśāstra 1.1 (variant devau, eulogy to Maheśvara), Harivaṁśa 106.37, Niśvāsatattvasaṁhitā, Mukhasūtra 1.24, Svāyaṁbhuvasūtrasaṅgraha, Śivapūjāpaṭalaḥ 39, Kriyākālaguṇottara 1.1, and Ahirbudhnyasaṁhitāpariśiṣṭa (Sudarśanasahasranāmastotra) 1.1, to mention only a few. Several instances with the variant śambhum or devīm exist (see, for example, Uḍḍāmareśvaratantra 1.1, Kubjikāmatatantra 22.29, Rasārṇava 1.3, and Ānandakanda 1.6.1, 1.13.1).
See Kozok 2006. An English-language monograph of the same title was published by Uli Kozok in 2004. A substantially revised and expanded English-language monograph by a team of authors led by Kozok is currently in press (see Kozok forthcoming). I thank Thomas Hunter for making accessible to me his chapter in the aforementioned publication of which I have made grateful use here.