The Ancient Languages of Indonesia

In the last post I described what we mean when we talk about ‘medieval Indonesia’ or ‘ancient Indonesia’. Plenty of historical work is set more or less in that region, a region that actually amounts to the main landmasses, and generally the coastal parts of those, in western Indonesia and Malaysia: Sumatra, Java, Bali, Borneo, and the Malay Peninsula. Eastern Indonesia is treated entirely separately by historians. The reason for that is that writing and Indian influence were not really present east of Borneo and Bali, at least not to the same extent as they were in that core region. And this is true; it’s not how I think history should be done these days (‘prehistory’ is just part of human history) and eastern Indonesia played an enormous role in making the archipelago what it was when Muslims and Europeans started arriving in large numbers. But it is certainly true that ‘Indianization’ and inscriptions were restricted to the western islands.

So if you want to use primary sources on Indonesia/Indo-Malaysia before the sixteenth century then the languages you need are mostly from western Indonesia and India. Chinese and Arabic are also helpful, the former especially so. There are also European sources from the fourteenth and fifteenth centuries (Polo, Odoric, Conti). The reports in Portuguese by Tomé Pires, Duarte Barbosa, and Afonso de Albuquerque the younger (among many others) from the early-to-mid sixteenth century are extremely useful in making sense of the history of poorly documented parts like Ternate.

Ethnohistory is important in Indonesia because of the relative lack of local sources, and I’ll write a post on European, Middle Eastern, and Chinese texts on the archipelago at some point. In this post, though, I’ll give a brief survey of the local languages that a historian of Indo-Malaysia up to 1500 CE might need. In sort-of-chronological order, they are: Sanskrit, Malay, Javanese, and Sundanese. Aside from Sanskrit, they are all related languages — all are members of the Malayo-Polynesian branch of the Austronesian language family, and therefore related to Māori, Malagasy, Fijian, and countless others in the Indo-Pacific.


a) Sanskrit.

The first inscriptions in the archipelago come from eastern Borneo, way away from the main trade routes. Nobody really understands why this is, actually, but either way they were inscribed in the fourth century CE in Sanskrit. Sanskrit is an Indo-European language of North India, but the script used in these early inscriptions (along with a few other clues) suggests that it was brought to Southeast Asia by South Indians — Tamil speakers, perhaps — and first by Hindus rather than Buddhists. Sanskrit seems to have been the only language of Indonesian inscriptions for at least a few centuries; other Sanskrit inscriptions include the Purnawarman stones from West Java, dated to the fifth century, and the Canggal inscription from Central Java, dated 732 CE. Sanskrit was gradually replaced by local languages, however, and for the period I’m interested in (fourteenth and fifteenth centuries) Sanskrit isn’t a particularly useful language to know, except to the extent that it helps you to figure out the meanings of loanwords in the local languages.

One of the inscribed stones from Kutai, eastern Borneo, c. 350 CE? Currently in the Museum Nasional, Jakarta.

Sanskrit is difficult and time-consuming. It has a tricky inflectional morphology that takes time to fully understand. I do not know Sanskrit, although I’m familiar with a good chunk of the common vocabulary. If you’re really into this early period, however, it seems essential.


b) Malay

Malay is a big language — really a group of closely related dialects spread over a wide geographical area with lots of loanwords mixed in, somewhat akin to modern English. It has something like 300 million speakers, although most of these grow up bilingual with something else (Javanese, maybe, or Ternate, or Minangkabau); either way it’s one of the biggest languages on Earth by number of speakers. One of its dialects, Indonesian (bahasa Indonesia), is the national language of Indonesia; another, Malaysian (bahasa Malaysia), is the national language of Malaysia. It likely served as a lingua franca throughout the archipelago from quite an early point, and there are other (generally less prestigious dialects) scattered among the islands from Terengganu Malay to Kupang Malay to Ternate Malay to Betawi, the modern dialect of the city of Jakarta.

Malay was also the first language indigenous to the archipelago to have been written down, beginning in 683 CE with the Kedukan Bukit inscription. The inscription is one of several from the kingdom of Śrīvijaya in southern Sumatra, whose capital appears to have been at Palembang (now the second-largest city on the island). Śrīvijaya became wealthy and powerful by monopolizing trade in the Strait of Malacca; it was this trading advantage that probably gave Malay a significant boost in the direction of becoming the archipelago’s lingua franca. The stage of the language represented by the inscriptions from Śrīvijaya is known as Old Malay, characterized by a lack of distinction between the voiced labial consonants /b/ and /w/ and by a large proportion of Sanskrit loans. Inscriptions in Old Malay can be found in Sumatra, the Malay Peninsula, Bangka (off the east coast of Sumatra by Palembang), and even Java.

There is no point at which the Old Malay phase can be definitively considered to have ended because few texts from this period have survived. Insects, heat, humidity, and the fragility of the materials used all meant that early (pre-paper) manuscripts had a limited shelf-life. Periodisation is thus tricky. Only one manuscript in Malay has survived from before the sixteenth century, and it’s a legal text known as the Nītisārasamuccaya, written in late-fourteenth century Dharmasraya, Sumatra, on bark paper (daluang). Commonly known as the Tanjung Tanah manuscript (TK 214), it was preserved in an attic in Kerinci, Sumatra, where the smoke from the hearth kept insects and damp away. Uli Kozok is the chief (but not sole) investigator of the text at the moment, and you can find his website here.

After that, though, Malay entered a stage now called ‘Classical Malay’. Nearly all of the texts in Classical Malay are written in Jawi script, a writing system derived from Perso-Arabic script. As you might expect, Classical Malay is therefore also associated with the rise and spread of Islam in Indo-Malaysia. The oldest extant example of Jawi is the Terengganu Inscription Stone, dated to 1302 CE and found in the state of Terengganu on the east coast of the Malay Peninsula. The oldest Malay historical text, Hikayat Raja Pasai, is dated to later in the fourteenth century on internal references, but most Classical Malay texts are later even than that. Most of the manuscripts themselves are considerably more recent; the oldest MS of Hikayat Raja Pasai was penned in 1797 (and can be found here, completely digitized by the British Library).

What Classical Malay looks like in the Jawi script. BL MS Or. 14350, f.3v

Malay is fairly easy to learn. Since the 1950s it has been written primarily in the Roman script. It has very little if any inflectional morphology, its derivational morphology is a bit trickier but no harder than that of any European language, and the phonology is simple and easy to pronounce for speakers of, well, most languages. Tense marking is not obligatory and is performed by independent morphemes (e.g. sudah ‘already’ = past tense, placed before the verb). The sole difficulty is that there are lots of dialects; I know Standard Indonesian and can understand everything they say on the news, and I can get by reading Classical Malay, but the other dialects take me a bit of time. Nevertheless, if you have the opportunity to take a class in Malay/Indonesian, you will be amply rewarded. It’s a great language to know.


c) Old Javanese

Javanese is now spoken by about 80 million people, most of whom live in Central and East Java. There is a significant Javanese diaspora in Suriname, Singapore, Malaysia, and the Netherlands (and elsewhere), but the vast bulk of Javanese people are Muslims who live in Java, representing perhaps the largest single ethnic group in Southeast Asia and one of the largest Muslim ethnolinguistic groups in the world. The only thing that stops Javanese from consideration as a major world language is that it is not a national language.

The form of Javanese found in pre-sixteenth-century texts is called Old Javanese. Some scholars like to distinguish another phase known as ‘Middle Javanese’ from the fourteenth century on, but it’s probably not distinct enough to justify such a division and work in ‘classical’ Old Javanese was still being produced late into the fifteenth century (including the work that inspired my twitter handle, the Śivarātrikalpa, a beautiful kakawin written around 1470). P. J. Zoetmulder’s Old Javanese-English Dictionary, first published in 1982 and available for free online here, does not mark the distinction.

Old Javanese philology is where the real meat of ancient and medieval Indonesia is considered to be, at least in the Dutch tradition — which, by the way, is probably the greatest unknown contribution to the humanities of which I am aware, with so many fantastic texts transcribed and studied. Transcribed from what, though? What has survived?

Well, most texts in Old Javanese are actually from Bali, the island east of Java. Bali has remained predominantly Hindu-Buddhist-indigenous in religious orientation, deliberately and sometimes violently resisting Muslim domination for more than five centuries. Bali has its own language (see below) and its religious traditions today are doubtless different from those of fourteenth-century Java, but the island has in any case served as the repository of a lot of non-Islamic Javanese texts. Some Old Javanese texts survived in Java and there are non-Muslim Javanese people still, but most manuscripts are from Bali or the Balinese-speaking bits of Lombok, the island just to the east of Bali. Most of these are palm-leaf manuscripts, with several species of palm used to make them. The most popular is probably lontar (Borassus flabellifer). Some of these texts are superb pieces of world literature, and represent the only surviving literature of any size in any ancient or medieval Southeast Asian language.

An Old (Middle?) Javanese inscription from Candi Sukuh in the east of Central Java. The date (the final text box) says 1361, equivalent to a Common Era date of 1439. These chunky letters are unusual and restricted entirely to two fifteenth-century sites in Central Java, Candi Sukuh and Candi Ceto. Image found here on Leiden University Digital Collection’s wonderful searchable website: https://digitalcollections.universiteitleiden.nl/view/item/8307

There are also plenty of Old Javanese inscriptions written in a range of often very different scripts that changed a lot over time. Learning the language isn’t too hard and the tools are there if you want to do it (including this free textbook by Willem van der Molen, as well as the aforementioned dictionary), but the palaeographic skill takes time to develop. I can make sense of texts in Old Javanese with the help of a dictionary and I can sight-read simple prose texts in transcription, but inscriptions require some serious up-close investigation — and even then scholars don’t always agree. There’s a thriving Indonesian palaeography blog scene, though, and you can often find a solution to a palaeographic difficulty by googling the right term in Indonesian.

Old Javanese morphology is more complex than that of Malay, but it’s nowhere near as complex as e.g. Latin or Sanskrit, or even French. It uses a few affixes to mark passive moods, and there’s an arealis form, but that’s really about it. A lot of the vocabulary, though, is Sanskrit.

I should also point out that Bali has its own language (Balinese) and that there are early texts in an early form of it known (unsurprisingly) as Old Balinese. I don’t know any Old Balinese and the language was replaced by Old Javanese for formal occasions at some point after the twelfth century. The reason I’m not discussing it in detail here is because, frankly, I don’t know enough about it; most texts are, as far as I’m aware, copperplate and stone inscriptions.


d) Old Sundanese

Before 1500 Javanese was spoken principally in Central and East Java. West Java was (and still is!) home to Sundanese, a language with a large number of speakers these days — around 40 million — and an interesting ancient literature. Sundanese is fairly close to Malay and some core vocabulary is actually the same, although it is marginally more morphologically complex (it has a plural form, for example — Malay just reduplicates the noun, e.g. tas ‘bag’, tas-tas ‘bags’). Sundanese has also been profoundly influenced by Javanese at various points, including in the medieval period.

The oldest inscriptions in West Java are in Sanskrit and Malay and then Old Javanese. Old Sundanese language and script developed rather later than Old Javanese and most Old Sundanese texts use a lot of Old Javanese loanwords, including both inscriptions (like the Kawali inscriptions, probably fourteenth century) and manuscripts (like the wonderful Sanghyang Siksakandang Karesian, dated 1518, the full transcribed text of which can be found here). Old Sundanese is a fourteenth-to-sixteenth-century language, and while its manuscripts often can’t be easily dated and tend to languish unread in Western collections, it appears to have fallen out of use in the eighteenth century, when (modern) Javanese took over as the literary language of West Java.

There is no textbook for Old Sundanese, nor is there a dictionary. This is a pity. Texts have to be interpreted etymologically and contextually through comparison with Old Javanese/Malay/modern Sundanese vocabulary. About half the texts are prose and half are poetic, and as far as I know all the poetic texts are organised in metrical/syntactic lines of eight syllables. Some of these texts have incredible value both as historical documents and literary works, and their neglect by all but modern Sundanese scholars and a select few Western academics is a real shame, not just for West Java but for the world. My particular favourite is Bujangga Manik, a poetic work of 1757 octosyllabic lines about the journey of an ascetic around Java in the late fifteenth century.


Naturally I haven’t said everything there is to say about ancient languages in Indonesia. There are also a few inscriptions in Tamil (as far as I’m aware exclusively from Sumatra, but I could easily be wrong), and Arabic texts are found on tombstones from quite an early period. Chinese people have been in the region for centuries; it’s hard to believe they didn’t write down texts in Chinese while there. It is an incredibly diverse place — at least as diverse as Europe when you include the Outer Islands. Poor preservation means that there are fewer texts to read than for comparably sized regions with literate traditions of similar antiquity, but that’s not an excuse for ignoring Indonesia entirely, and if you want to be able to read some of these languages then you should know that they really aren’t as difficult as they’re exotic names make them sound to European and American ears.

Like what you read? Give Medieval Indonesia a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.