Nipah or Gebang?

A Philological and Codicological Study Based on Sources from West Java

In: Bijdragen tot de taal-, land- en volkenkunde / Journal of the Humanities and Social Sciences of Southeast Asia

This article takes up the question of writing supports, the physical media on which texts were recorded on the island of Java before paper and printing were introduced, with special focus on the western region and the Old Sundanese tradition. In the past, two types of indigenous writing materials prepared from the leaves of palm trees were identified, one known among scholars in the Dutch tradition as ‘nipah’, the other as ‘lontar’. While lontar is a common, widely used designation for the type of palm-leaf writing material used in the vast majority of surviving manuscripts, the nipah is rare and not commonly thought of as a writing material outside of scholarly circles. In an effort to understand the place of nipah in the tradition, the author turns to descriptions of writing materials in old written as well as oral sources and concludes that the terminology used there is at odds with the accepted idea that the second, rare type of leaf used in the palm leaf manuscript tradition came from the nipah palm. Instead, it was prepared from another palm species called gebang. At the same time, the author provides new insight into indigenous conceptualizations that differentiate the types of texts recorded on lontar and gebang materials.


This article takes up the question of writing supports, the physical media on which texts were recorded on the island of Java before paper and printing were introduced, with special focus on the western region and the Old Sundanese tradition. In the past, two types of indigenous writing materials prepared from the leaves of palm trees were identified, one known among scholars in the Dutch tradition as ‘nipah’, the other as ‘lontar’. While lontar is a common, widely used designation for the type of palm-leaf writing material used in the vast majority of surviving manuscripts, the nipah is rare and not commonly thought of as a writing material outside of scholarly circles. In an effort to understand the place of nipah in the tradition, the author turns to descriptions of writing materials in old written as well as oral sources and concludes that the terminology used there is at odds with the accepted idea that the second, rare type of leaf used in the palm leaf manuscript tradition came from the nipah palm. Instead, it was prepared from another palm species called gebang. At the same time, the author provides new insight into indigenous conceptualizations that differentiate the types of texts recorded on lontar and gebang materials.

It is widely accepted and regularly repeated among scholars that there are two types of palm leaves that were historically used as writing supports in manuscripts from the Indonesian archipelago: leaves of the sugar, or toddy, palm (Borassus flabellifer) and those of the nipa palm (Nypa fruticans). These palms and their leaves are known as lontar and nipah respectively. These two types of leaves can easily be distinguished by the naked eye: the latter are thinner and stiffer, and they are of a lighter colour than the former. The manner of inscribing text on these two sorts of leaves is also different: lontar are inscribed through a process of scratching or incising, while nipah are written upon directly using black ink.

In contrast to lontar-leaf manuscripts, which make up the vast majority of all known palm leaf manuscripts from Lombok to Sumatra, use of the nipah leaf as writing support has been the subject of only limited codicological research, receiving almost no attention in the literature (Van der Molen 1983:88). The only testimony to the use of nipah leaves for writing material is De Clercq (1927, as quoted in Van der Molen 1983:89), who states that he heard that formerly, in the hinterlands of South Sumatra, and perhaps up to the time of his report, nipah were used for writing ephemeral love letters. Beyond this, the literature is silent on the characteristics of this rare writing material. There are, to be sure, a small number of articles or catalogue notes that discuss nipah, but they are limited to the investigation of available nipah manuscripts; the writing support is in such sources categorized as nipah without further question. The processing and preparation of nipah leaves for writing also remains an uninvestigated question, as does their use.

The first notice mentioning the existence of manuscripts written on palm leaves from the Priangan area was published by Netscher (1853:474), although he still designated the writing material of the Arjunawiwāha codex (later acquired by the Bataviaasch Genootschap [hereafter BG] with accession number L 641) as lontarblad (lontar leaf). Holle was the first to use the term nipah in 1862 (NBG 1, 1862–1863:14; Van Lennep 1969:16).

Five years later Holle (1867) described three nipah leaf manuscripts donated by Raden Saleh.1 Holle identified these three manuscripts as MSS A, B, and C, and it can be determined that MS A is now catalogued as PNRI number L 632 (Kabuyutan Galuṅguṅ); MS B is L 630 (Saṅ Hyaṅ Siksa Kandaṅ Karəsian), while MS C is number L 631 (Chandakaraṇa)2 (Holle 1867:452–64).3

But the discussion which touches most directly upon the use of nipah leaf as a writing medium is the general sketch given by Holle in the introduction to his Tabel van Oud- en Nieuw-Indische alphabetten (1882), a work on palaeography that makes use of epigraphic sources as well as later records on organic materials. There are three important things to note from Holle’s study: (1) the sites where nipah manuscripts were collected; (2) the writing implements used to inscribe them; and (3) the style of script and other aspects of palaeography. Holle states that most nipah manuscripts originated in West Java, with a small number also acquired from the hermit-scholar collections in the Merapi-Merbabu mountain region of Central Java. With regards to writing implements, he says that nipah manuscripts were inscribed using a type of ink fabricated from the nagasari plant (cobra’s saffron, Mesua ferrea) and damar sela resin (Sundanese, harupat). In relation to the writing itself, Holle states that the letters inscribed on the nipah were in a quadratic Kawi script. He also provides a table of characters drawn from several nipah manuscripts found in West Java, that is, in Talaga, Cirebon, and Ciburuy (1882:7–8, 17, 25–6).

A few other investigations of nipah manuscripts by scholars of subsequent generations also need to be mentioned here. Among them is the promising investigation by Van Lennep in her 1969 undergraduate thesis done at Sydney University. Her research is important because it notes details of the original acquisition of nipah manuscripts by the BG, and takes a different look at aspects of nipah. Van Lennep also hypothesizes that the BG’s trove of nipah manuscripts was part of a royal manuscript collection of the Pajajaran Kingdom in West Java which, when under threat due to the rise of Islam in the sixteenth century, was removed to (or hidden in) the surrounding mountains, such as Mt Cikuray near Garut (Van Lennep 1969:29–33).

Another important codicological contribution is Van der Molen’s careful physical examination (1983:90–3) of one particular nipah manuscript containing a Kuñjarakarṇa text (LOr 2266), edited in his Leiden doctoral thesis. Through meticulous attention to small details he was able to observe traces of a press or pressure device in the form of marginal lines—one on the right, one on the left, and two in the centre of the leaves—sometimes quite clear, other times faint. Observations on the distance between these lines, measured to the millimetre, indicated that the processing of leaves for use as writing supports was a precision craft. Following Grader (1941:25), who established a relationship between the variability of dimensions of manuscripts and the use of different tools by different artisans, Van der Molen (1983:91) arrived at the hypothesis that there are two possibilities related to such variability. If tools or equipment are the source of variation, then indications regarding the identity of workshops can be derived from manuscript features such as length, breadth, distance between holes, and distance from holes to leaf-edge. If the type of leaf is the most significant variable, then measurements can be the same over a wide area.

The most recent study of interest for our investigation is Acri’s (2011a) investigation of the Dharma Pātañjala, an Old Javanese Shaivite text. This wide-ranging study includes identification of nearly every known nipah manuscript. The specific Dharma Pātañjala text Acri focused on was found in a manuscript from the Merapi-Merbabu area, not West Java, where most nipah manuscripts originated. With respect to the place where the manuscript was found, Acri suggested that there may once have been a relationship between scriptoria in West Java and the manuscript repositories of the Central Javanese massif. It is possible that several nipah manuscripts from West Java could have made their way to Merapi-Merbabu some time before the middle of the eighteenth century, and from there ended up in the great Windu Sono collection of Merapi-Merbabu manuscripts that was later transferred into the possession of the BG. Cultural ties between these two centres of literary production might have led to the exchange of manuscripts in the past (Acri 2011a:44–7).4 This completes the survey of principal sources on manuscripts hitherto identified as being inscribed on so-called nipah leaves.

Among the studies mentioned above, only those of Holle will continue to retain our attention in this article, for Holle was the first to identify the writing medium that is the focus of our attention here as nipah and to furnish codicological explanations about both the material and the utensils used to write on it.

It is unclear on what basis Holle arrived at this identification, as he mentions no source. There seem to be two possibilities: either Holle himself established this botanical identification (with or without the help of an anonymous botanist), or he obtained this information from his local contacts in West Java, where he lived. There is something to be said for the first possibility, because we find that Kern wrote, just a few years later: ‘according to the opinion of the botanists it is palm leaf, I take it to be very thin bark’ (NBG 25, 1887: 179; Van Lennep 1969:16). The second possibility also makes sense, as the word nipah had already been recorded in the Sundanese dictionaries that date from the same period (Rigg 1862:s.v.; Geerdink 1875:s.v.), and was almost certainly known to Holle’s informants as some kind of palm tree.5

But we must remember that Holle’s explanations date to a time when, as reported by Netscher (1853), the practice of writing on nipah leaves was no longer a living tradition. For this reason, we may ask ourselves whether the identification received from an informant was merely based on the physical appearance of the leaves, or whether the informant was actually familiar with the use of the same leaves as reported from Sumatra by De Clerq a few decades later (1927). We may indeed ask ourselves whether Holle’s identification can be acceped at all. In this article I will propose a new identification of this type of palm leaf used as writing material, using sources that were not used by Holle in determining his identification.

Nipah Manuscripts

The total number of nipah manuscripts is a tiny fraction of those written on lontar. When all nipah manuscripts in Indonesian and European collections are added up, they number only 29 out of the many thousands of Indonesian palm leaf manuscripts in existence. At least 20 of these 29 manuscripts are in the collection of the Perpustakaan Nasional Republik Indonesia (National Library of Indonesia; hereafter PNRI) in Jakarta;6 three are in the Kabuyutan (hermitage) of Ciburuy, Garut;7 and a single manuscript is in the Sri Baduga Museum at Bandung.8 In Europe there are at least five such manuscripts: two in the library of Leiden University,9 one at the Bibliothèque nationale de France (French National Library) in Paris,10 one at the Staatsbibliothek (State Library) in Berlin,11 and one at the Bodleian Library, Oxford.12 Others may exist, but they have not yet been detected or described.

Although there are a few exceptions, nearly all nipah manuscripts have been found in West Java, as indicated in the institutional records on their acquisition history. The origins of the nipah manuscripts at the PNRI are indicative. Manuscripts L 374 and L 630–632 were obtained from the Galuh area of East Priangan by the well-known Javanese artist and cultural figure Raden Saleh (Cohen Stuart 1872; Holle 1867); manuscripts L 633–642 were acquired by the bupati of Bandung from near Cilegon, Garut (Netscher 1853; Krom 1914:71); L 643 originated from Talaga, Cirebon (NBG 4, 1866:118; Krom 1914:92); manuscripts L 1095, L 1097, and L 1099 all came from Kabuyutan Koleang at Jasinga, Bogor (Krom 1914:32). There is some uncertainty about the provenance of three PNRI manuscripts, that is, L 455, L 627, and L 628. However, in copies of these manuscripts in the K.F. Holle Collection (PNRI Peti 89) the reader finds notes indicating that the last three came from Merbabu (cf. Acri 2011a:46 n. 9).

The situation regarding provenance is similar for nipah manuscripts in European collections. The two manuscripts kept in Leiden, Kuñjarakarṇa (LOr 2266) and Tiga Jñana (LOr 2267), are thought to come from West Java (Pigeaud 1968:94; 1970:21, 56). The Rasa Carita at Oxford (MS Jav.b.1) is one of a pair of manuscripts donated by Andrew James during the seventeenth century and very likely to have originated in West Java, the other manuscript being in Old Sundanese (Noorduyn 1985). The provenance of the manuscript in Paris is unclear, but noting that its text contains the Saṅ Hyaṅ Hayu, which is also found at Ciburuy and in three PNRI manuscripts known to originate near Garut, it must be closely associated with the textual tradition of West Java. One exception is the Berlin collection, whose Dharma Pātañjala came from Mt Merbabu in Central Java (Pigeaud 1975:111–2; cf. Acri 2011a:44).

d2426912e591Figure 1

‘Nipah’ Manuscripts. a. Arjunawiwāha (Old Javanese, 1334 ce, cod. PNRI L 641); b. Saṅ Hyaṅ Hayu (Old Javanese, 1523 ce, cod. PNRI L 634); c. Saṅ Hyaṅ Siksa Kandaṅ Karəsian (Old Sundanese, 1518 ce, cod. PNRI L 630). National Library of the Republic of Indonesia.

Citation: Bijdragen tot de taal-, land- en volkenkunde / Journal of the Humanities and Social Sciences of Southeast Asia 171, 2-3 (2015) ; 10.1163/22134379-17101004

Local Sources

As mentioned above, De Clercq stated in 1927 that in South Sumatra nipah was used as a writing material for a specific function—love letters. The function of so-called nipah manuscripts in Java, that is to say, the type of texts found copied on nipah leaf manuscripts, was quite different. Judging by the surviving corpus of such manuscripts, they never contain love letters. This difference in use or function is what prompted Van der Molen (1983:89) to ask, ‘Can De Clerq’s information be applied to literature and to Java?’. In other words, are there sources from Java that designate nipah as a writing material? We can say from the outset that the answer to this question is resoundingly negative. Although in an article on Old Javanese writing materials Hinzler (2001) states that nipah was a type of writing support mentioned in older Javanese texts, she does not indicate the textual source upon which this claim is based. The word she is referring to is found in the Tuhañaru, Wariṅin Pitu, and Balawi inscriptions, as quoted in Zoetmulder’s OJED under the entry nipah (1982:1183). But the form of the word quoted there, anipah,13 and the fact that it appears in a context containing references to payuṅ wəlu (round parasol?), mopih (wrapping or covering), and ruṅki (a type of woven basket?), does not indicate any relationship to writing. In any case these inscriptions all originate in East Java. Outside of the realm of epigraphy, I have not found a single textual passage involving nipah that could be quoted here.

The identification of the rarer of the two types of palm leaf manuscript as nipah until now has hinged on a single scholarly assertion. It has become part of ‘common knowledge’ about palm leaf manuscripts in Java, though it is not based on thorough philological or codicological research. This article will show that this is a case of mistaken identification which has escaped the attention of researchers working on so-called nipah manuscripts. What we have come to know as nipah manuscripts are probably not actually made from nipah leaves. In any case, there are no known references to Nypa fruticans as a writing support in the textual traditions of Java and Sunda. Because so-called nipah manuscripts generally originated from West Java, sources from this region should interest us before all others in an effort to clarify the identity of these writing materials and the terminology associated with them. Relevant passages have been identified in several texts and in passages from the oral tradition of Sundanese pantun, as outlined in the following paragraphs.

Saṅ Hyaṅ Śāsana Mahāguru

The first source that mentions palm leaves and their use as writing materials is the Saṅ Hyaṅ Śāsana Mahāguru (hereafter Śāsana Mahāguru), a prose work. As far as is known there are two manuscripts containing this text, both written in Old Sundanese script and language on lontar leaves; they are PNRI manuscript L 621 and another manuscript, identified as kropak 26, in the collection at Kabuyutan Ciburuy. The text is a tutur presenting the teachings of a guru (saṅ pandita) to his student, a religious devotee (saṅ sewaka dharma), presented in typical form as a question and answer narrative. The parts of the text that are of specific relevance are the maṅgala, or introduction, as well as a section containing an enumeration of the ‘ten improvements’ (dasawṛddhi), or ten types of material used as writing media. An edition of this text based on manuscript L 621 is provided in Gunawan (2009).


The second source is a prose text in Old Javanese, the Bhīmaswarga, which recounts the adventures of Bhīma, second of the five Pāṇḍawa brothers, as he journeys to heaven. There are a number of versions of the Old Javanese Bhīmaswarga text. Hinzler (1981:194–203) notes one prose and two poetic renditions in the Balinese tradition. Among the Merapi-Merbabu manuscripts, Setyawati, Wiryamartana, and Van der Molen (2002) have recorded six copies of a prose Bhīmaswarga that differs from the Balinese version. There is also a prose Bhīmaswarga that originates in West Java that is different again from either of the aforementioned. This is the version employed in this study.

The West Javanese Bhīmaswarga, an edition of which is currently being prepared by this writer, is known from three manuscripts, two of which are currently held at the PNRI in Jakarta and one in Ciburuy. The first manuscript, L 455, is inscribed on nipah leaves in a script that resembles that found in the Kuñjarakarṇa manuscript, LOr 2266, at Leiden. The second is manuscript L 623, inscribed in Old Sundanese script on lontar leaves. The third manuscript from the Kabuyutan at Ciburuy is incomplete and has been separated into two separate fragments catalogued as lontar Ciburuy VII and kropak Ciburuy XII. It, too, is written in Old Sundanese script.

d2426912e828Figure 2

Source Manuscripts. a. Saṅ Hyaṅ Śāsana Mahāguru (PNRI L 621); b. Saṅ Hyaṅ Śāsana Mahāguru (Ciburuy Kropak 26); c. Bhīmaswarga (PNRI L 455); d. Bhīmaswarga (PNRI L 623); e. Bhīmaswarga (lontar Ciburuy VII); f. Saṅ Hyaṅ Swawarcinta (PNRI L 626). National Library of the Republic of Indonesia.

Citation: Bijdragen tot de taal-, land- en volkenkunde / Journal of the Humanities and Social Sciences of Southeast Asia 171, 2-3 (2015) ; 10.1163/22134379-17101004

Although the core text is entirely in Old Javanese, certain sections display affinities to the pre-Islamic manuscript tradition that is specific to West Java. First, both L 623 and the Ciburuy manuscript are written in local script. Second, the colophon found in L 623 is written in Old Sundanese and states that the text, here given the title Bhīmaləpas, was composed (or copied) at Mt Cikuray, situated in the Garut district of West Java.14 Finally, although it is recorded that L 455 was acquired from Mt Merbabu, the writing material indicates the possibility alluded to earlier, namely that the manuscript was originally produced in West Java before ending up in the Merbabu manuscript trove.

Saṅ Hyaṅ Swawarcinta

The third source is a poetic text in Old Sundanese, the Saṅ Hyaṅ Swawarcinta, available in PNRI manuscript L 626. This is a codex unicus. It is inscribed on lontar leaves in a form of Old Sundanese script. It contains a long narrative by an author who considers himself quite young (boñcah), and who asks his readers for permission to present them with ilmu, knowledge. Its contents include numerous depictions of the everyday lives of the Sundanese people at the time the text was written, such as the stories they recited, types of food, manners, and the like. An edition based on this manuscript was published in Wartini et al. (2011).


The last source is pantun literature. Pantun form part of the Sundanese oral tradition, consisting of tales about the initiation and exploits of cultural heroes. They are recited without reference to a written text by a juru pantun, usually accompanied by a kecapi lyre, in a performance that lasts most of the night. According to Noorduyn and Teeuw (2006:279) there is a historical relationship between Old Sundanese poetic literature and pantun, which is apparent not only in the shared feature of composition in octosyllabic lines, but also in the formulaic expressions found in both. My sources for these come predominantly from transcriptions of pantun tales made by Ajip Rosidi: Carita Kembang Panyarikan (1973), Tjarita Demung Kalagan (1970), Carita Gantangan Wangi (1973), and Tjarita Parenggong Jaya (1971).15

Nipah or Gebang?

In this section, passages containing information about writing materials from the sources listed above will be presented. In this way the tradition can itself testify about writing materials and their historical use.

In the maṅgala of Śāsana Mahāguru the writer dedicates his work to Bhaṭāra Gaṇa (Gaṇeśa), the creator of writing implements. Below is the relevant quotation:

article image

We see that in this maṅgala the writer dedicates the work he is composing to Bhaṭāra Gaṇa, whose trunk trumpeted and sprayed a gift to the realm of humankind. Through this gift he created (matəmahan) both gebang and lontar. After the first green shoots of gebang and lontar appeared (pupus), they were formed in precisely the same way (tinut pinada-pada) in their length and breadth, making them ideal to be used as writing materials that could subsequently ‘be sprinkled’ (tinitisan) by the four instruments of writing, that is, the hand (hasta), water (gaṅgā), pen or brush (wīra), and ink (tanu). These are all instruments used by a pandita to write a work that can then be read by someone lacking understanding so that their knowledge can grow; they are instruments by which the passing on of knowledge is determined or limited (winaləran) by the poet (saṅ kawi).

The quartet hasta-gaṅgā-wīra-tanu as writing instruments or elements can also be found in the text of the Bhīmaswarga below, with the difference that in this text we find a set of three elements, gaṅgā-wīra-tanu, while hasta is not mentioned:

article image

The words ‘black’ and ‘ink’ in this context are key. These terms are relevant only for manuscripts written on the palm leaf type identified by scholars as nipah. With lontar, as is well known, the writing itself is created by colourless incisions which are only subsequently made visible by the application of oily lampblack or a similar soot-based blackener.

Elsewhere in the Bhīmaswarga we find mention of gebang again in relation to books (pustaka), as can be seen in the following allegorical enumeration of Bhīma’s divine allies:

article image

In the Bhīmaswarga text, in addition to gods, the Pāṇḍawa are also represented in ways symbolically linked to a book or pustaka:

article image

This quote paints a picture in which pustaka is the term for book (manuscript) as a whole, which consists of several parts: gəbaṅ refers to the leaves; tali to the string passed through the perforation in the middle of the stacked leaves; papan refers to the box (kropak) or wooden cover boards of the manuscript.17

Based on these excerpts from West Javanese sources, we have at this point been introduced to two types of writing support: lontar and gebang. Furthermore, in the Śāsana Mahāguru it is stated that these two writing materials represent two of ten types of writing media:

article image

As enumerated here, the ten writing supports are (1) gold (omas), (2) silver (salaka), (3) copper (tambaga), (4) steel (cundiga), (5) iron (bəsi), (6) stone (batu), (7) wooden boards (paduṅ), (8) bamboo? (pəjwa), (9) lontar (taal), and (10) gəbang. For the purposes of this article, only the last two are of direct interest.

Further along the Śāsana Mahāguru gives a more detailed explanation of terms and functions related to writing on lontar and gebang:

article image

Thus, a technical demarcation is established between the two media in terms of use and function: writing on lontar is called carik, while writing on gebang is called cəməṅ. The word carik means ‘scratch, line’; lontar manuscripts are written by incising, cutting, and scratching shapes onto the leaf surface using a type of knife (Balinese: pengutik; Sundanese: péso pangot). As for cəməṅ, its meaning is simply ‘black’ (cf. Javanese cemeng). Perhaps ‘black’ is used to signify black ink, such as we find used on nipah, as stated by earlier scholars? Compare also the term hirəṅ as written in the opening of the Bhīmaswarga’s first passage on the pustaka. The material distinction between lontar and gebang as writing supports is paralleled by a distinction in the type of text that each is meant to bear: lontar are not meant to hold writings of an inherently sacred character (lain pikabuyutanən), while gebang leaf is specifically intended to be the medium for conveying sacred texts (pikabuyutanən).

Another useful source is the Old Sundanese text Saṅ Hyaṅ Swawarcinta, ll. 447–52, which provides the following overview:

article image

The meaning of tanah (l. 450) raises several questions.19 Zoetmulder (1974:129–35) has identified this writing implement in the Old Javanese tradition as a sort of slate pencil made from a soft stone; it could be sharpened with the fingernail and was simply thrown away when it was broken or worn down to a nub. Interestingly, the quote links tanah directly to lontar and gebang, while in kakawin sources it is regularly paired with karas (a writing slate) and never with palm leaves. In the excerpt above it is stated that the tanah is black. Can the tanah intended here be linked to the spines (lidi) of Enau sugar palm fronds (Arenga pinnata), called harupat in Sundanese, as stated by Holle (1882:17)? Can the characteristics of tanah as identified by Zoetmulder be linked to those of the lidi of the Enau palm, which in fact is black, easily bent, and capable of being sharpened with the fingernail?

Let us turn our attention now to the final set of sources: the pantun of the oral-tradition. The relevant portions for the purposes of this article are the rajah in the pantun stories. Rajah, which are usually performed by the juru pantun at the beginning of a performance, are sung prayers asking the gods to prevent the occurrence of any disturbances during the performance of the tale:

article image

The word pus is a short form of pupus, meaning ‘the young (just unfolding) leaf of the banana and the lirang-palm’ (OJED, s.v. pupus). In the next line we find as synonym of pupus the word pondok, which appears to be a mispronunciation by the juru pantun of, or a mistranscription/misprint for, the more archaic word pondoh, which is still noted in dictionaries with the meaning ‘young palm leaf’ (Danadibrata 2006:s.v. pondoh). This may be compared with the case of the synonyms pucuk, which also means ‘young leaf’, and pondok (again mispronounced/mistranscribed for pondoh) in the pantun Kembang Panyarikan cited below.

Lontang is also a mispronunciation/mistranscription for lontar. Sabeulit, ‘one twist or turn’, indicates a length of string wrapped once through the perforation at the middle of the manuscript leaves (Balinese: song) and then wound once around the cover boards. In the context of the quoted sentence, the word sabeulit indicates that the juru pantun intends to finish the entire narrative in a ‘single wrap’ of the manuscript (though in fact the performance does not involve actually reading from a physical manuscript). The entire story must be finished as a tamba pamali, or protection from forbidden things.

The rajah (introduction, introit) reproduced above is parallel with the rajah at the opening of the pantun called Kembang Panyarikan, as below:

article image

In this passage, gebar is a mispronunciation/mistranscription of gebang, the same phenomenon of /r/ and /ng/ being switched (as in lontar < lontang) seen in the Demung Kalagan excerpt. The quote above also shows the word lawe being used as a synonym for apus, meaning string or cord. The word lulumbang is assumed to have the meaning of lulumban in modern Sundanese (ng < n), meaning ‘to be joyful’ (Satjadibrata 2005:235). Lulumbang siang may perhaps have a meaning similar to baranang siang, ‘a clear day’, which, in the above context, indicates a bright and joyful mood. It appears the writer intended to offer good tidings through the gebang and lontar, whose leaves have been opened:

article image

Abis gobang is a mispronunciation by the juru pantun (or mistranscription) of apus gəbang (gebang cord), as becomes clear from the next phrase, lawe lontar (lontar string), a near synonym. Bandepung may be taken as a variant of mandepun, a piece of cloth used to cover up objects set on a tray. A mandepun is also used to wrap up ‘manuscripts’ and other sacred heirlooms (Panitia Kamus Sunda 1976:301). Once again we see the equivalence of /ng/ in pantun usage to /n/ in modern Sundanese.

The excerpts from the rajah of pantun stories offered above strengthen Noorduyn and Teeuw’s assertion (2006) that pantun are closely related to the Old Sundanese written tradition. In practice, juru pantun performed pantun tales without relying on a written text, but when the reciter began his performance by uttering the opening rajah, it is clear that the story about to be related had originated in a text recorded in a manuscript (kandana baheula) consistently described as gebang and lontar.


As has been shown above, the analysis of West Javanese sources, both in the form of ancient manuscripts preserved in West Java and in oral tradition, provides reasonably strong evidence that the term gebang is used to refer to a type of writing material in the manuscript tradition of West Java. By contrast, the word nipah never figures anywhere in ancient sources in the contexts that concern us here. Although there are physical similarities between nipah and gebang, and both belong to the same botanical family (namely, the Arecaceae palm family), they are distinct species belonging to different genera. There appears to be no indication in Sundanese, Javanese, or Balinese sources that nipah has ever been commonly used as a writing support.20

It is worth noting in this connection how palm leaves were historically used as a writing material in other, related cultural settings in South and Southeast Asia. Throughout this extensive zone, Borassus flabellifer L. and Corypha umbraculifera L. (both members of the subfamily Coryphoideae) are the two types of palm most widely used (Jahn 2006:923). A survey of studies of palm leaves as writing support from Tibet to the Philippines confirms that the genera Corypha and Borassus are the raw materials of first resort in societies that manufacture writing materials from palm leaves. Writing about India, Hoernle (1900:93) explains that two types of palm leaves were traditionally used as writing supports there—just as we have seen in Java. Those two species of palm were the tāḍatāla (Borassus flabellifer) and tāḍitāli (Corypha umbraculifera). Tāḍatāla is the same species as lontar. Tāḍitāli is the leaf of a different type of palm, called talipot; its leaves are thinner, wider, and have have a smoother surface than the Borassus. Thus we see that in India, too, two types of palm leaves are used as writing materials, one identical to lontar, the other a close relative of gebang (Corypha utan or Corypha gebanga).21 Indeed in a very recent article, Perumal (2012:159) notes that Corypha utan is a third type of palm leaf used as a writing support in Tamilnadu, South India, in addition to Corypha umbracullifera and Borassus flabellifer, while Nypa is notably absent.22

When we regard these plants in their natural environments in Indonesia, gebang is a common and well distributed variety of palm tree (Rigg 1862s.v.). It is naturally known by a variety of terms in the languages of the archipelago. The Dayak people know it as gabang, the people of Timor call it gawang, in Madura it is pocok, to the Betawi it is pucuk, among the Batak and Sasak it is ibus, while in Minahasa it is silar. Physically, a mature specimen ranges between 15 to 20 metres in height. Gebang leaves form a fan shape, like the fingers of an open hand, with a diameter of 2–3.5 metres, joined at the tip of a stalk. This tree is most commonly found in coastal areas near rivers and swamps; it is also encountered in hilly countryside, though more sparsely. Gebang have a slow rate of growth and are not found above elevations of 300 metres (Heyne 1922:301). The geographical origins of this species are not clear, but its distribution today includes tropical Africa, India, Burma, Thailand, the Philippines, Malaysia, and Indonesia.

By contrast, the nipah palm is more rarely found in Java (Gallop 1998:16). Local toponymy gives a clear reflection of the relative abundance, importance, and familiarity of natural landscape features, flora, and fauna. An examination of gazetteers reveals that nipah is almost completely unknown in the toponymy of West Java. Gebang, by contrast, is widely used in place names, including Bantar Gebang in Bekasi, Gebang sub-district in Cirebon, and Ciawi Gebang in Kuningan. Several of these place names are located in lowland areas (such as Bantar Gebang and the Cirebon sub-district); others are in areas of low hills (Ciawi Gebang). One can deduce that areas named after this type of palm tree are places where the gebang once grew in a way that made it stand out. Nipah stands out only by its absence as an element in place names.

Although our focus here has been on Java and in particular its western stretches, it should be noted that there are references outside of West Java to gebang as a writing support. Ketut Ginarsa (1975:92) once wrote that ‘in Bali, in addition to lontar, gebang (Corypha utan), sometimes called “forest lontar”, is also said to be related to writing implements (for example in the prayer accompanying wayang performances uttered by dalang in North Bali)’. Hinzler (1993:438) quotes Cox (1931) in the opening to her article on palm leaf manuscripts in Bali, adding that the gebang palm was, at least in the 1930s, very rare in Bali, whereas lontar palms grew abundantly in dry areas. No further information offering insights into their use along the lines found in West Javanese sources is mentioned.


Holle’s misidentification of gebang as nipah more than 130 years ago was taken up uncritically by subsequent researchers. It is contradicted by sources from West Java, both manuscripts and the oral pantun, which describe use of two types of palm leaves as complementary writing materials: gebang and lontar. Nipah does not enter into these passages, though other types of material that can be used for writing on are enumerated. As long as no other sources are found that contain references to nipah, one should rely on the term gebang (Corypha gebanga, Corypha utan Lam.) to designate the palm-leaf writing support that is written on with ink, not incised with a knife or stylus. Indeed, the character of manuscripts copied through use of ‘black’ (cəməṅ) rather than ‘incising’ (carik), is differentiated in important ways. Writing on lontar is ‘not intended to be used to preserve a sacred text’ (lain pikabuyutanən), that is, a text imbued with sacral power and dealing with matters of transcendent knowledge. This theoretical notion is substantiated by the fact that, in general, Old Sundanese texts found on lontar tend more to have the characteristics of ‘literature’ as understood by modern readers. Writing on gebang, by contrast, is ‘intended for use in preserving sacred texts’ (pikabuyutanən). This statement, too, is substantiated by the fact that nearly all texts written on gebang are of the tutur or tattwa variety, that is to say, didactic religious works in prose containing teachings of sacred knowledge structured as an exchange between a guru and a student, and often accompanied by several śloka.23 The special, religious character of these texts is reflected in the use of Old Javanese, the cosmopolitan language of Java and Bali in that period, which is relatively dominant in texts written on gebang.24 It is worth noting that in Sanskrit and descendant Indo-Aryan languages Corypha is also called śrītāla (Jahn 2006:929). The addition of the element śrī (sacred) to the word tāla in this compound may indicate the special character of Corypha compared to Borrasus.

Gebang manuscripts available to us at this time provide evidence that, although the gebang leaves are physically less robust than lontar, as a writing support they are very long-lasting and able to survive in the humid tropical climate of Indonesia. The oldest of the gebang manuscripts, giving a text of the Arjunawiwāha (PNRI L 641, dated 1334 ce), is nearly seven hundred years old. Although it has been late in coming, further codicological research on the production process by which leaves become writing supports, such as that carried out by Hoernle for palm leaf manuscripts in India more than a century ago, remains an important task to be pursued today. The conclusions reached here on philological grounds should also be strengthened by botanical research on nipah and gebang through laboratory experiments. Although this could not yet be offered here, such experimental evidence will be important to determine with greater precision which type of leaf was actually used in the past to produce the type of writing material that I propose from now on to designate as gebang.

Pun. Leuwih luangan kurang wuwuhan.

Finished. Whatever is excessive, please reduce; whatever is deficient, please supply.


Acri, Andrea (2006). ‘The Sanskrit-Old Javanese tutur literature from Bali: The textual basis of Śaivism in ancient Indonesia’, Rivista di Studi Sudasiatici 1:107–37.

Acri, Andrea (2011a). Dharma Pātañjala: A Śaiva scripture from ancient Java studied in the light of related Old Javanese and Sanskrit texts. Groningen: Forsten Publishing.

Acri, Andrea (2011b). ‘Javanese manuscripts of the Tattvajñāna’, in: ManjuShree (ed.), From beyond the eastern horizon: Essays in honour of Professor Lokesh Chandra, pp. 119–29. New Delhi: Aditya Prakashan.

Acri, Andrea and Undang Ahmad Darsa [2009]. Situs Kabuyutan Ciburuy: Collection description. Endangered Archives Programme (EAP) 280—‘Retrieving Heritage: Rare Old Javanese and Old Sundanese Manuscripts from West Java’. (accessed March 2014).

Cabaton, A. (1912). Catalogue sommaire des manuscrits Indiens: Indo-Chinois & Malayo-Polynésiens. Paris: Leroux.

Carita Gantangan Wangi. 1973. Carita Gantangan Wangi. Dipantunkan oleh Ki Asom (Pringkasap, Subang), diusahakan oleh Ajip Rosidi. Bandung: Proyek penelitian pantun dan folklor Sunda. [Mimeographed.]

Carita Kembang Panyarikan (1973). Carita Kembang Panyarikan. Dipantunkan oleh Ki Kamal (Lebak Wangi, Kuningan), diusahakan oleh Ajip Rosidi. Proyek Penelitian Pantun dan Folklor Sunda. [Mimeographed.]

Cohen Stuart, Abraham Benjamin. [1872]. Eerste vervolg catalogus der bibliotheek en catalogus der Maleische, Javaansche en Kawi handschriften van het Bataviaasch Genootschap van Kunsten en Wetenschappen. Batavia: Bruining & Wijt; ’s Gravenhage: Nijhoff.

Clerq, F.S.A. de. (1927). Nieuw plantkundig woordenboek voor Nederlandsh Indië. Met korte aanwijzingen van het nuttig gebruik der planten en hare beteekenis in het volksleven, en met registers der inlandsche en wetenschappelijke benamingen. Amsterdam: De Bussy. Tweede, herziene en vermeerderde druk bewerkt door A. Pulle, met medewerking voor het taalkundig gedeelte van den heer A.H.J.G. Walbeehm te Leiden.

Cox, J. (1931). ‘Monografie Bali en Lombok’. [Unpublished typewritten report dated 1931.]

Danadibrata, R.A. (2006). Kamus basa Sunda. Bandung: Panitia Penerbitan Kamus Basa Sunda in collaboration with Kiblat Buku Utama & Universitas Padjadjaran.

Darsa, Undang A. (1998). Sang Hyang Hayu: Kajian filologi naskah bahasa Jawa Kuno di Sunda pada abad XVI. [Magister thesis, Universitas Padjadjaran, Bandung.]

Darsa, Undang A. (2012). Séwaka Darma dalam naskah tradisi Sunda Kuno abad XV–XVII Masehi. [Phd thesis, Universitas Padjadjaran, Bandung.]

Darsa, Undang A. and Edi S. Ekadjati (2004). ‘Sanghyang Raga Dewata’, in: Fatimah in West Java: Moral admonitions to Sundanese gentlewomen dan kajian lainnya mengenai budaya Sunda, pp. 134–79. Bandung: Pusat Studi Sunda & Kiblat Buku Utama. [Seri Sundalana 2.]

Ensiklopedi Sunda (2000). Ensiklopedi Sunda: Alam, manusia, dan budaya (termasuk budaya Betawi dan Cirebon). Jakarta: Pustaka Jaya.

Gallop, Annabel T. (1998). ‘Indigenous writing materials: Seals’, in: John H. McGlynn (ed.), Language and literature, pp. 16–7, 44–5. Singapore: Archipelago. [Indonesian heritage 10.]

Geerdink, A. (1875). Soendaneesch-Hollandsch woordenboek. Batavia: H.M. van Dorp & Co.

Ginarsa, I Ketut (1975). ‘The lontar (palmyra) palm’, Review of Indonesian and Malaysian Affairs 9:90–103.

Gonda, J. (1933). ‘Agastyaparwa: Een Oud-Javaansch proza-geschrift’, Bijdragen tot de Taal-, Land- en Volkenkunde 90:329–419.

Gonda, J. (1936). ‘Agastyaparwa: Vertaling’, Bijdragen tot de Taal-, Land- en Volkenkunde 94:223–85.

Grader, C.J. (1941). ‘De vervaardiging’, in: C.J. Grader and C. Hooykaas, ‘Lontar als schrijf-materiaal’, Mededeelingen van de Kirtya Liefrinck-Van der Tuuk 13:23–6.

Gunawan, Aditia (2009). Sang Hyang Sasana Maha Guru dan Kala Purbaka (suntingan dan terjemahan). Jakarta: Perpustakaan Nasional RI.

Heyne, K. (1922). De nuttige planten van Nederlandsch-Indië: Tevens synthetische catalogus der verzamelingen van het museum voor economische Botanie te Buitenzorg. Batavia: Ruygrok & Co.

Hinzler, H. (1981). Bima Swarga in Balinese wayang. The Hague: Nijhoff. [Verhandelingen van het Koninklijk Instituut voor Taal-, Land- en Volkenkunde 90.]

Hinzler, H. (1993). ‘Balinese palm-leaf manuscripts’, Bijdragen tot de Taal-, Land- en Volkenkunde 149:438–73.

Hinzler, H. (2001). ‘The book in ancient Java: Books, writing, writing material, and reading in written and visual evidence’, in: Marijke J. Klokke and Karel R. van Kooij (eds), Fruits of inspiration: Studies in honour of Prof. J.G. de Casparis. Groningen: Forsten.

Hoernle, A.F. Rudolf (1900). ‘An epigraphical note on palm-leaf, paper and birch-bark’, Journal of the Asiatic Society of Bengal 69:93–134.

Holil, Munawar and Aditia Gunawan (2010). ‘Membuka peti naskah Sunda kuna di Perpustakaan Nasional RI: Upaya rekatalogisasi’, in: Perubahan pandangan aristokrat Sunda dan esei-esei lainnya mengenai kebudayaan Sunda, pp. 104–46. Bandung: Pusat Studi Sunda dan Kiblat Buku Utama. [Seri Sundalana 9.]

Holle, K.F. (1867). ‘Vlugtig berigt omtrent eenige lontar-handschriften afkomstig uit de Soenda-landen’, Tijdschrift voor Indische Taal-, Land- en Volkenkunde 16:450–70.

Holle, K.F. (1869). ‘De Batoe-Toelis te Buitenzorg’, Tijdschrift voor Indische Taal-, Land- en Volkenkunde 17:483–8.

Holle, K.F. (1882). Tabel van Oud-en Nieuw-Indische alphabetten. Bijdrage tot de paleographie van Nederlandsch-Indië. Batavia: Bruining; ’s Gravenhage: Nijhoff.

Jahn, Samia Al Azharia (2006). ‘Comparative studies on different concepts about the origin of writing on palm leaf: Botany—traditional technologies—divine teachers’, Asiatische Studien / Études Asiatiques 60–4:921–61.

Krom, N.J. (1914). Rapporten van den Oudheidkundigen Dienst in Nederlandsch Indië. Inventaris der Hindoe-oudheden op den grondslag van Dr. R.O.M. Verbeek’s Oudheden van Java, Eerste Deel. Weltevreden: Albrecht & Co.

Lennep, D. van (1969). Some observations on the nipah leaf kropaks from West Java, with an analysis of content and historical relevance of the manggala to the Old-Javanese Amaramala. [BA honours thesis, University of Sydney.]

Lokesh Chandra (1997). ‘Chanda-Karaṇa: The art of writing poetry’, in: Lokesh Chandra, Cultural horizons of India. Vol. 6, Studies in Tantra and Buddhism, art and archaeology, language and literature, pp. 140–242. New Delhi: International Academy of Indian Culture and Aditya Prakashan.

Molen, Willem van der (1983). Javaanse tekstkritiek: Een overzicht en een nieuwe benadering geïllustreerd aan de Kuñjarakarna. Dordrecht: Foris. [PhD thesis, University of Leiden.]

Moriyama, Mikihiro (2005). Sundanese print culture and modernity in 19th-century West Java. Singapore: Singapore University Press.

NBG (various years). Notulen van de algemeene en directievergaderingen van het Bataviaasch Genootschap van Kunsten en Wetenschappen. Batavia: Lange & Co., later Bruining, Kolff & Co; ’s Gravenhage: Nijhoff.

Netscher, E. (1853). ‘Iets over eenige in de Preanger-regentschappen gevonden Kawi- handschriften’, Tijdschrift voor Indische Taal-, Land- en Volkenkunde 1:469–79.

Noorduyn, J. (1982). ‘Bujangga Manik’s journeys through Java: topographical data from an old Sundanese source’, Bijdragen tot de taal-, land- en volkenkunde 138: 413–42.

Noorduyn, J. (1985). ‘The three palm-leaf mss from Java in the Bodleian Library and their donors’, Journal of the Royal Asiatic Society of Great Britain and Ireland 117:58–64.

Noorduyn, J. and A. Teeuw (2006). Three Old Sundanese poems. Leiden: KITLV Press. [Bibliotheca Indonesica 29.]

Panitia Kamus Sunda (1976). Kamus umum basa Sunda. Compiled by Panitia Kamus Lembaga Basa & Sastra Sunda. Bandung: Tarate.

Perumal, P. (2012). ‘The Sanskrit manuscript in Tamilnadu’, in: Saraju Rath (ed.), Aspects of manuscript culture in South India, pp. 157–72. Leiden: Brill [Brill’s Indological library 40.]

Pigeaud, Th. (1968). Literature of Java: Catalogue raisonné of Javanese manuscripts in the library of the University of Leiden and other public collections in the Netherlands. Vol. II, Descriptive list of Javanese manuscripts. The Hague: Nijhoff.

Pigeaud, Th. (1970). Literature of Java: Catalogue raisonné of Javanese manuscripts in the library of the University of Leiden and other public collections in the Netherlands. Vol. III, Illustrations and facsimiles of manuscripts, maps, addenda and a general index of names and subjects. The Hague: Nijhoff.

Pigeaud, Th. (1975). Javanese and Balinese manuscripts and some codices written in related idioms spoken in Java and Bali. Descriptive catalogue. Wiesbaden: Steiner. [Orientalische Handschriften in Deutschland 31.]

Poerbatjaraka, R.M.Ng. (1926). ‘Arjuna-wiwāha. Tekst en vertaling’. Bijdragen tot de Taal-, Land- en Volkenkunde van Nederlandsch-Indië 82:181–305.

Ricklefs, M.C. and P. Voorhoeve. (1977). Indonesian manuscripts in Great Britain; A catalogue of manuscripts in Indonesian languages in British public collections.

Rigg, Jonathan (1862). A dictionary of the Sunda language of Java. Batavia: Lange & Co.

Robson, S.O. (1976). ‘More about tanah and karas’, Bijdragen tot de Taal-, Land- en Volkenkunde 132:353–5.

Rubinstein, Raechelle (2000). Beyond the realm of the senses: The Balinese ritual of kakawin composition. Leiden: KITLV Press. [Verhandelingen van het Koninklijk Instituut voor Taal-, Land- en Volkenkunde 181.]

Sarkar, Himansu Bhusan (1935). ‘The Old-Javanese copper-plates from Sidotĕka of the śaka year 1245’, Journal of the Greater India Society 2:133–41.

Satjadibrata (2005). Kamus basa Sunda. Bandung: Kiblat Buku Utama.

Setyawati, Kartika, Kuntara Wiryamartana, and Willem van der Molen (2002). Katalog naskah Merapi-Merbabu, Perpustakaan Nasional Republik Indonesia. Yogyakarta: Penerbitan Universitas Sanata Dharma; Leiden: Opleiding Talen en Culturen van Zuidoost-Azië en Oceanië, Universiteit Leiden.

Swellengrebel, J.L. (1936). Korawāçrama: Een Oudjavaans gedicht. Groningen: Mees.

Tjarita Demung Kalagan (1970). Tjarita Demung Kalagan, dipantunkeun ku Ki Kamal (Lebakwangi, Kuningan), diusahakeun ku Ajip Rosidi. Bandung: Proyek Penelitian Pantun dan Folklor Sunda.

Tjarita Parenggong Jaya (1971). Tjarita Parenggong Jaya, dipantunkan oleh Ki Samid Tjisolok—Sukabumi, diusahakan oleh Ajip Rosidi. Bandung: Proyek Penelitian Pantun dan Folklor Sunda.

Wartini, Tien, Ruhaliah, Mamat Ruhimat and Aditia Gunawan (2011). Sanghyang Swawar Cinta: Teks dan terjemahan. Jakarta: Kerjasama Perpustakaan Nasional RI dan Pusat Studi Sunda.

Wiryamartana, I. Kuntara (1993) ‘The scriptoria in the Merbabu-Merapi area’, Bijdragen tot de Taal-, Land- en Volkenkunde 149:503–9.

Zoetmulder, P.J. (1974). Kalangwan: A survey of Old Javanese literature. The Hague: Nijhoff. [KITLV Translation Series 16.]

Zoetmulder, P.J. (1982). Old Javanese–English dictionary. With the collaboration of S.O. Robson. Two vols. ’s-Gravenhage: Nijhoff.

Appendix 1. Gebang Manuscripts in the Collection of the National Library of Indonesia


  1. NBGNotulen van het Bataviaasch Genootschap van Kunsten en Wetenschappen.
  2. TBGTijdschrift van het Bataviaasch Genootschap van Kunsten en Wetenschappen.
article image

Appendix 2. Gebang Manuscripts from Kabuyutan Ciburuy

Information on manuscripts below is drawn from the descriptions in Acri and Darsa (2009). These descriptions were appended as an index for the Ciburuy manuscript digitalization project, sponsored by the British Library through its Endangered Archive Program (EAP) in 2009. Digitalized copies are kept at the National Library of Indonesia with an additional copy deposited at the British Library as the sponsoring institution (

article image

I would like to thank Henri Chambert-Loir, Willem van der Molen and Dina Isyanti for their valuable suggestions on the first draft of this article. Special thanks go to Arlo Griffiths for his generosity in providing relevant sources from the EFEO library in Jakarta and for important comments on and corrections to this article. This article appears in its present form thanks to Tim Behrend, who helped translate it into English with great care, at the same time providing valuable feedback and critical commentary. For this, I am sincerely grateful. Nevertheless, any mistakes that might remain are entirely my own responsibility.


Raden Saleh’s gift in 1866 totaled 38 manuscripts (NBG 5, 1867:155), but Cohen Stuart (1872) only recorded 35 of them. The three manuscripts not noted by Cohen Stuart were all made from nipah leaves. This is rather surprising recalling that Holle 1867 had already provided descriptions of them. See also Van Lennep 1969:11.


A complete edition of this manuscript is available in Lokesh Chandra 1995. See also Appendix 1 of this article.


A quick word on the fraught matter of spelling in this article. In order to avoid confusion due to variation in orthography in quotes from various printed and hand-written sources in three languages (Sundanese, Old Sundanese, and Old Javanese), all quotes from older literary sources have been standardized according to the system used in Zoetmulder’s Old Javanese–English dictionary (OJED) (1982), with slight changes as follows: the e-pepet is rendered as ə, not ĕ, while ŋ becomes . Furthermore, because the orthographic system used in Old Sundanese manuscripts does not distinguish between the vowels ə and eu, it is not necessary nor even, in my opinion, desirable to distinguish between the two in transcription. I have therefore used only the character ə where modern orthography distinguishes e and eu. All quotes from pantun have been standardized according to Modern Sundanese spelling as used in the Kamus umum basa Sunda (Panitia Kamus Sunda 1976).


On the Merapi-Merbabu scriptoria, see Wiryamartana 1993. The relationship of those scriptoria with West Java was indirectly referred to by Bujangga Manik, a sixteenth-century Hindu-Sundanese pilgrim who visited Damalung (an old name for Mt Merbabu) to study there (Noorduyn 1982). Damalung is also mentioned in the Sri Ajñana (ll. 45–53, Noorduyn and Teeuw 2006) as the name of a place where a heavenly protagonist was exiled to earth in punishment for his sins.


One of Holle’s most important informants was his friend Muhammad Musa, a religious leader (penghulu besar) from Garut and an important figure for Sundanese literature in his time (see Moriyama 2005:100–42). One example of the important role that Musa played as an informant for Holle is the fact that in his study of the Batu Tulis inscription at Bogor, Holle refers to the authority of his friend when he translates the word ñusuk found in this inscription with the verb ‘to establish’: ‘Volgens den Panghoeloe van Garoet wordt het Soendasche njoesoek nu nog wel eens in de beteekenis van een dorp of stad stichten gebruikt’ (Holle 1869:485).


Consult Holil and Gunawan’s study (2010) on pre-Islamic, West Javanese manuscripts (including so-called nipah manuscripts in Old Javanese) in the collection of the PNRI.


Acri and Darsa 2009 initially identified four nipah manuscripts from Kabuyutan Ciburuy in Garut, in addition to 23 lontar manuscripts kept in the same location. But based on more recent data, it appears that there are only three nipah and 24 lontar manuscripts in the Kabuyutan Ciburuy. I would like to thank Andrea Acri for this information (private email communication, 19 March 2013), and for his permission to include the list of manuscripts in appendix 2 to this article.


A manuscript titled Saṅ Hyaṅ Raga Dewata, MS 07.106 (Darsa and Ekadjati 2004).


The manuscript mentioned earlier (LOr 2266, containing a text of the Kuñjarakarṇa) and LOr 2267, containing the Tiga Jñana (Pigeaud 1968:94).


Manuscript no. Mal. Pol. 161 (Cabaton 1912:254). In Cabaton’s catalogue this manuscript is not given a title, but Acri (2011a:645) has identified it as containing a text of the Saṅ Hyaṅ Hayu. Several other manuscripts with the same title are found in the collection of the PNRI (L 634, L 637, L 638); an edition of Saṅ Hyaṅ Hayu based only on PNRI L 634 and L 637 has been provided by Darsa 1998.


The manuscript is titled Dharma Pātañjala, Cod. Schoemann I-21 (Pigeaud 1975:111–2). An edition of this manuscript has been offered by Acri 2011a.


Manuscript titled Rasa Carita, MS Jav.b.1; a portion of this manuscript was photocopied, transliterated, and annotated by Voorhoeve and Soegiarto, and is available as item LOr 8515 in the Leiden collection (Ricklefs and Voorhoeve 1977:177; see also Pigeaud 1968:479).


In his study of the Tuhañaru inscription from Sidoteko, Mojokerto, Sarkar (1935) cautiously suggested that anipah is based on sipah (from səpah?) with prefixed a-, and that its meaning relates to ‘sirih-plums?’.


The colophon reads: itiḥ kahuwusaniṅ pu[s]taka bimaləpas haranika, samapta sampun sinurat [riṅ] wulan kasa, saṅ anurat panadaan saka pat pun, ə[ñ]cu nu ṅaheraṅ bukit cikuray samapun Ø ‘This is the end of the book called Bhīmaləpas, its writing was completed in the first month, written by pa na da an (?) Śaka year four (?), grandson of the one practising meditation (at) Mt Cikuray. Finished.’


Further information about pantun performance is found in Ensiklopedi Sunda 2000:493. On the relationship between pantun and Old Sundanese literature, see Noorduyn and Teeuw 2006:10–1; 279–81.


In the Korawāśrama (Swellengrebel 1936:112) Citragotra is also said to have the responsiblity for writing pustaka (bhagawān citragotra pwa masəḍahan pustaka). The character referred to here is none other than Citragupta, the assistant or clerk of Dharmarāja (Yama), who keeps the book of humanity’s karma.


In an Old Javanese text from Bali, the Tutur Aji Saraswatī (Cod. Gedong Kirtya 2289), there is also a symbolic identification of the Pāṇḍawa with the components of a pustaka. In place of gebang, this text has lontar. The twins, Nakula and Sahadewa, between them symbolize the two cover boards (cakəpan kalih), Arjuna the lontar leaf (əntal), Bhīma the string, and Dharmatanaya the textual contents (śāstra) (Rubinstein 2000:56–7).


I choose to translate kabuyutan as ‘sacred text’, not as ‘place of worship’. The latter meaning does not suit the context because kabuyutan (in the sense of a place) actually denotes a place for storage of manuscripts of various types. Besides the meaning ‘place of worship’, OJED (s.v. buyut) also records other senses of this word, namely ‘object of worship, status of elder’. Furthermore, in the actual contexts where we find the word kabuyutan in Old Sundanese sources, it clearly does not always denote a place, but also denotes things that are ‘holy, sacred’ in a more general way, for example, ‘words’, as in Sewaka Darma (lines 513–4): saur dipikabuyutan, sabda dipirahasea ‘words are held sacred, sounds are kept secret’ (Darsa 2012:383); or ‘cloth’ in The sons of Rama and Rawana (lines 179–81): diais ku sabuk wayaṅ, dibaur dəṅ kabuyutan, saṅ hyaṅ gula gumantuṅ ‘He carried the boy in his decorated sash, with his sacred cloth, the sacred Gula Gumantung’ (Noorduyn and Teeuw 2006:183).


On the meaning and use of tanah in Old Javanese, see Zoetmulder 1974:129–35 and Robson 1976.


As for sources in Old Javanese not transmitted in West Java, I first note that Zoetmulder (1982:505, s.v. gĕbaŋ) cites just one text, namely the Pārthayajña, which mentions this word among other trees. A so far unrecognized occurrence of the word gəbaṅ is to be found in a passage in the Agastyaparwa (Gonda 1933:381 lines 27–30, 1936:266): kunaṅ anak bhagawān pulastya i saṅ wirudhinyaṅ odwad, salwiriṅ odwad, sakwehiṅ rumambat mwaṅ dukut, pətuṅ, nyuh, hano, tal, gəbaṅ (em., gəṅaṅ ed.), salwirniṅ twaksāra. nahan tānak bhagawān pulastya i saṅ wirudhi ‘And the children of Bhagawān Pulastya with Wirodhi, they are the hanging plants, all types of hanging plants; all creepers and grasses, bamboos, coconut-palms, aren-palms, lontar-palms, gebang-palms, all types of trees. Those are the children of Bhagawān Pulastya with Wirudhi.’ The emendation to gəbaṅ is unproblematic in palaeographic terms ( and b are very similar in Balinese script), and evident in the light of the West Javanese textual evidence presented above. The Agastyaparwa passage and this emendation were pointed out to me by Arlo Griffiths.


For more details on local names for Borrassus flabellifer and Corypha umbraculifera in India and Indonesia, see the table provided by Jahn (2006:927–8). Interestingly, in this table, Corypha umbraculifera and Corypha utan are considered to be the same.


This article presents two misconceptions that need to be clarified. First, it states that Corypha umbraculifera grows in dry climates. In reality the opposite is true: this species does not thrive in dry zones. Second, Corypha utan is referred to as lontar and is said to grow in wet zones. Lontar is not Corypha utan but Borassus flabellifer, and only grows in dry zones.


A detailed acount of tutur or tattwa, including chronology, typology, and relationship between these texts and their Sanskrit antecedents, can be found in Acri 2006 and 2011a:8–10.


Such as Saṅ Hyaṅ Hayu (PNRI L 634, L 637, L 638, see Darsa 1998), Dharma Pātañjala (cod. Schoemann I-21, see Acri 2011a), Siksa Guru (PNRI L 627, L 628, L 643), Kuñjarakarṇa (LOr 2266, see Van der Molen 1983), Arjunawiwāha (PNRI L 641, see Poerbatjaraka 1926), and others.

  • 6

    Consult Holil and Gunawan’s study (2010) on pre-Islamic West Javanese manuscripts (including so-called nipah manuscripts in Old Javanese) in the collection of the PNRI.

  • 9

    The manuscript mentioned earlier (LOr 2266containing a text of the Kuñjarakarṇa) and LOr 2267 containing the Tiga Jñana (Pigeaud 1968:94).

  • 10

    Manuscript no. Mal. Pol. 161 (Cabaton 1912:254). In Cabaton’s catalogue this manuscript is not given a title but Acri (2011a:645) has identified it as containing a text of the Saṅ Hyaṅ Hayu. Several other manuscripts with the same title are found in the collection of the PNRI (L 634 L 637 L 638); an edition of Saṅ Hyaṅ Hayu based only on PNRI L 634 and L 637 has been provided by Darsa 1998.

If the inline PDF is not rendering correctly, you can download the PDF file here.

  • 6

    Consult Holil and Gunawan’s study (2010) on pre-Islamic West Javanese manuscripts (including so-called nipah manuscripts in Old Javanese) in the collection of the PNRI.

  • 9

    The manuscript mentioned earlier (LOr 2266containing a text of the Kuñjarakarṇa) and LOr 2267 containing the Tiga Jñana (Pigeaud 1968:94).

  • 10

    Manuscript no. Mal. Pol. 161 (Cabaton 1912:254). In Cabaton’s catalogue this manuscript is not given a title but Acri (2011a:645) has identified it as containing a text of the Saṅ Hyaṅ Hayu. Several other manuscripts with the same title are found in the collection of the PNRI (L 634 L 637 L 638); an edition of Saṅ Hyaṅ Hayu based only on PNRI L 634 and L 637 has been provided by Darsa 1998.

  • View in gallery
    Figure 1

    ‘Nipah’ Manuscripts. a. Arjunawiwāha (Old Javanese, 1334 ce, cod. PNRI L 641); b. Saṅ Hyaṅ Hayu (Old Javanese, 1523 ce, cod. PNRI L 634); c. Saṅ Hyaṅ Siksa Kandaṅ Karəsian (Old Sundanese, 1518 ce, cod. PNRI L 630). National Library of the Republic of Indonesia.

  • View in gallery
    Figure 2

    Source Manuscripts. a. Saṅ Hyaṅ Śāsana Mahāguru (PNRI L 621); b. Saṅ Hyaṅ Śāsana Mahāguru (Ciburuy Kropak 26); c. Bhīmaswarga (PNRI L 455); d. Bhīmaswarga (PNRI L 623); e. Bhīmaswarga (lontar Ciburuy VII); f. Saṅ Hyaṅ Swawarcinta (PNRI L 626). National Library of the Republic of Indonesia.

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 616 199 25
PDF Downloads 94 66 8