Javanese
jav has a single ISO 639-2 form; the ISO 639-1 equivalent is jv (an earlier two-letter code, jw, was withdrawn), and ISO 639-3 keeps jav for the language.
Javanese has one of the largest speaker populations of any language without national-language status, and a written tradition reaching back more than a thousand years. Its code in ISO 639-2 is jav, published by the Library of Congress at http://id.loc.gov/vocabulary/iso639-2/jav; the two-letter ISO 639-1 form is jv.
The language
Javanese is an Austronesian language native to the central and eastern parts of Java, with large communities elsewhere in Indonesia and in Suriname, New Caledonia and Malaysia, a legacy of labor migration in the colonial period. Its best-known feature is a system of speech levels: everyday vocabulary (ngoko) and polite registers (krama) differ for many common words, and the choice depends on the relation between speaker and listener.
The language is written in Latin letters today, but a large manuscript heritage uses Javanese script, a Brahmic script known as Aksara Jawa or Hanacaraka, and some religious texts use a form of Arabic script called Pegon.
Codes that sit near jav
| Code | Meaning | Notes |
|---|---|---|
| jav / jv | Javanese | modern language |
| kaw | Kawi | used for Old Javanese literary texts |
| sun / su | Sundanese | western Java, separate language |
| mad | Madurese | Madura and eastern Java |
| ind / id | Indonesian | national language |
| map | Austronesian languages | collective fallback |
The withdrawn two-letter code jw deserves its own line. It was replaced by jv decades ago, yet locale lists in some software kept it, and data exported from such systems still carries it. A cleanup rule mapping jw to jv (and to jav for three-letter fields) is a routine step when migrating older collections.
Metadata practice
Catalogs. MARC 21 records use jav in 008/35-37. Manuscripts in Javanese script commonly have romanized titles and a 546 note naming the script, since the language code does not say which script was used. For collections of palm-leaf (lontar) and paper manuscripts, that note is often what researchers search for first.
Dublin Core. The id.loc.gov URI in dc:language gives an unambiguous value in a repository profile built on the ISO 639-2 vocabulary. A description or a separate script field covers the writing system.
Web and digital text. BCP 47 tags Javanese as jv, with jv-Java for the traditional script. Javanese script has been in Unicode since version 5.2, but older digital editions sometimes used custom fonts mapped onto Latin code points, so the stored characters are Latin letters that only look like Javanese with the original font installed. Format and preservation records for such files should say so plainly; once the font is lost, the text is unreadable without the mapping. Converting such files to real Unicode Javanese is a migration in its own right, and the original file should be kept alongside the converted one.
Recurring confusion
The commonest error is coding Javanese as Indonesian because a publication comes from Indonesia. Bilingual publications are frequent, and they need both codes. Less often, Sundanese material from West Java is coded as Javanese because of the shared island. The two are distinct languages, mutually unintelligible in everyday use, and have separate codes in every part of ISO 639.
Example in RDF
<rdf:Description rdf:about="https://example.org/manuscripts/serat-047"
xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
xmlns:dcterms="http://purl.org/dc/terms/">
<dcterms:title>Serat, palm-leaf and paper manuscript 47</dcterms:title>
<dcterms:language rdf:resource="http://id.loc.gov/vocabulary/iso639-2/jav"/>
<dcterms:description>Javanese in Javanese script (Aksara Jawa).</dcterms:description>
</rdf:Description>
Questions about jav
What is the ISO code for Javanese?
The ISO 639-2 code is jav and the ISO 639-1 code is jv. The Library of Congress publishes the three-letter code at http://id.loc.gov/vocabulary/iso639-2/jav. ISO 639-3 also uses jav, so catalog records and language-technology resources share the three-letter form.
Why do some old files use jw for Javanese?
jw was an earlier two-letter code for Javanese. It was withdrawn and replaced by jv, but legacy software and older datasets still contain it. jw is not a valid ISO 639-1 code today and should be mapped to jv during migration.
Is Javanese the same as Indonesian?
No. Indonesian (ind, id) is the national language of Indonesia, a standardized form of Malay. Javanese is a separate Austronesian language, native to central and eastern Java, with its own literature and a system of speech levels.
How is Javanese script tagged?
The ISO 15924 code for Javanese script is Java, so text in the traditional script can be tagged jv-Java in BCP 47. Most modern Javanese is written in Latin letters (jv-Latn), which is usually left implicit.
Which code fits Old Javanese texts?
ISO 639-2 has a separate code, kaw (Kawi), which libraries use for Old Javanese literary texts. Modern Javanese material takes jav. Middle-period texts sometimes need a cataloger's judgment and a language note.