Lezghian
lez has a single ISO 639-2 form and no ISO 639-1 equivalent, so BCP 47 tags also use lez; ISO 639-3 keeps it as an individual language.
Search logs show the string lez in many contexts that have nothing to do with linguistics, from usernames to abbreviations in other languages. In ISO 639-2 the three letters have one meaning only: Lezghian, a language of the eastern Caucasus. The Library of Congress record at http://id.loc.gov/vocabulary/iso639-2/lez gives the English label Lezghian and the French lezghien; other English spellings in common use are Lezgian and Lezgi.
Definition and setting
Lezghian belongs to the Lezgic branch of the Northeast Caucasian family. Its speakers live on both sides of the Russia–Azerbaijan border in the Samur valley region: in southern Dagestan, where it is one of the republic’s literary languages, and in northern Azerbaijan. Like most languages of the area it has a large consonant inventory and a rich case system, which made it a frequent subject of descriptive linguistics; a widely cited reference grammar by Martin Haspelmath appeared in 1993.
Within Dagestan, Lezghian sits next to Avaric and several dozen smaller languages. Only a handful of them have ISO 639-2 codes, which affects how collections from the region are cataloged.
Coding decisions for Lezgic material
The rules are short:
- Lezghian itself takes lez in every system, because there is no two-letter alternative.
- Other Lezgic languages (Tabasaran, Agul, Rutul, Tsakhur, Budukh, Kryts, Udi, Archi) have individual codes in ISO 639-3 only. Where a schema accepts ISO 639-3 or BCP 47, those codes are the right values.
- Where a schema is restricted to the ISO 639-2 list, the fallback for them is cau (Caucasian languages), plus a note naming the actual language.
Using lez for a Tabasaran text because the two are related, or because the cataloger recognized the region, misleads every later user. The collective code at least says honestly that the language is not individually coded.
Records from the region
Twentieth-century publications from Dagestan and Azerbaijan are often bilingual or trilingual: Lezghian with Russian, or with Azerbaijani (aze). MARC 21 records list all of them in field 041, with the dominant one in 008. A 546 note recording the script helps whenever a collection also holds pre-Soviet Arabic-script or 1930s Latin-script material, because the language code alone hides the difference.
Romanization of Cyrillic Lezghian titles in English-language catalogs uses the ALA-LC table for non-Slavic languages in Cyrillic script, which handles the palochka and the additional digraphs. Systems that strip combining marks or treat the palochka as a Latin capital I produce inconsistent titles; a check for U+04C0 against U+0049 in titles is a useful cleanup query.
In Dublin Core and on the web
The example above shows a newspaper issue with two language values. In simple Dublin Core the strings lez and rus are enough; linked-data records use the id.loc.gov URIs. On web pages the tag lez (or lez-Cyrl where a page also reproduces older scripts) is valid and gives browsers and screen readers what they need. Few speech engines support the language, but correct tagging keeps them from reading the text with Russian rules.
Example in RDF
<rdf:Description rdf:about="https://example.org/serials/lezgi-gazet/1995-01.pdf"
xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
xmlns:dcterms="http://purl.org/dc/terms/">
<dcterms:title>Newspaper issue, January 1995</dcterms:title>
<dcterms:language rdf:resource="http://id.loc.gov/vocabulary/iso639-2/lez"/>
<dcterms:language rdf:resource="http://id.loc.gov/vocabulary/iso639-2/rus"/>
</rdf:Description>
Questions about lez
What language does the code lez stand for?
lez is the ISO 639-2 code for Lezghian, also spelled Lezgian or Lezgi, a Northeast Caucasian language. The French label in the code list is "lezghien". The code is published at http://id.loc.gov/vocabulary/iso639-2/lez.
Where is Lezghian spoken?
In the southern part of the Republic of Dagestan in the Russian Federation and across the border in northern Azerbaijan, along the Samur River and in the neighboring districts. Communities also live in other parts of Russia and Azerbaijan.
Does Lezghian have a two-letter ISO code?
No. There is no ISO 639-1 code for Lezghian, so lez is used everywhere: in MARC records, in Dublin Core metadata and as the BCP 47 primary subtag in web and XML documents.
Which script does Lezghian use?
Lezghian is written in a Cyrillic-based alphabet, which uses the palochka (Ӏ) for ejective consonants. Arabic script was used before the Soviet period and a Latin alphabet briefly in the late 1920s and 1930s. In BCP 47 the script can be stated as lez-Cyrl.
Are Tabasaran and Agul covered by lez?
No. They are related languages of the same Lezgic branch with their own ISO 639-3 codes, but ISO 639-2 has no individual entries for them. A schema limited to ISO 639-2 uses the collective code cau for them, not lez.