ISO 639-2 Three-Letter Codes — Extended Language Coverage

What Is ISO 639-2?

ISO 639-2 extends the ISO 639 standard with three-letter codes that cover approximately 500 languages — far more than the 184 covered by ISO 639-1. It was first published in 1998.

Two Variants: /T and /B

ISO 639-2 has a unique feature: about 20 languages have two different codes:

• /T (Terminological): Derived from the language's native name
• /B (Bibliographic): Derived from the English name

Language	/T Code	/B Code
German	deu	ger
French	fra	fre
Chinese	zho	chi
Czech	ces	cze
Dutch	nld	dut
Greek	ell	gre
Persian	fas	per
Romanian	ron	rum
Welsh	cym	wel
Basque	eus	baq

Which Form to Use?

• Modern web standards (BCP 47, IANA subtag registry) prefer the /T form
• Library cataloging (MARC records) traditionally uses the /B form
• When in doubt, use the /T form — it is more widely adopted in software

When ISO 639-2 Is Needed

ISO 639-2 codes are used when:

• A language does not have a two-letter ISO 639-1 code
• BCP 47 requires a primary language subtag for a language not in ISO 639-1
• Library and archival systems need to catalog materials in less common languages

Special Codes

• mul — Multiple languages
• und — Undetermined language
• zxx — No linguistic content
• mis — Uncoded language