ISO 639-3 has many language codes for Arabic, but only one for English.

I'm an Arab who is familiar with multiple Arabic dialects. We do not call it anything but "لهجات" which is translated to "dialects". We also call it "عامية" which is translated to "slang". I've never heard an Arabic man call its region's dialect a language. Yet the ISO classification for Arabic contain more languages than Arabic countries.

Arabic with over 30 language: http://www-01.sil.org/iso639-3/documentation.asp?id=ara
English with just one: http://www-01.sil.org/iso639-3/documentation.asp?id=eng

I know there are many difference between the American and British English. And there are differences between the South African, Canadian, North Ireland, Scottish and the Australian English. There is also the old Shakespeare period English language.

Still just one English language under the ISO 639-3.

Update: AFAIK No official reorganization for any dialect from any Arabic country. Arab countries teach only the main Standard Arabic language in schools and it's used in official documents.

It's the official language in the highest Arabic body, the [Arab League](language of the Arab https://en.wikipedia.org/wiki/Arab_League)

Also except for some variations in North Africa Arabs understand each other very clearly similar to different speaker of English (like English in India - English in Scotland).

  • 1
    We are very sensitive to the individuality and autonomy of others. I will actually see if we can get that changed.
    – user6726
    Commented Apr 3, 2016 at 17:55
  • Perhaps because it is so listed in ISO 639-2? The macrolanguage groups appear to line up between the two. Old and middle English have separate designators, by the way. Commented Apr 4, 2016 at 4:11
  • 2
    It's extremely common for native speakers of other varieties of English to trouble understanding some native speakers of both Indian and Scottish English, especially speakers with stronger accents. Commented Apr 17, 2016 at 16:21
  • Forsooth, methinks that this is an excellent question. An upvote I cast unto thee. Commented Feb 16, 2017 at 16:42
  • 3
    Arabic is hardly the only case here, for example bs sr hr and sr-ME overlap, Chinese has nuances, German dialects like Alemannic also have codes, as does Scots English. Hindi and Urdu both have a code. Commented Feb 19, 2017 at 5:53

5 Answers 5


ISO639-3 is derived from The Ethnologue. The author of this work have a certain point of view on the definition of "language" (vs. "dialect") with a strong tendency to split languages into smaller units. Note that the authors of the Ethnologue have a open agenda that is different from doing science. Divide and Conquer.

  • 6
    SIL are controversial sure, and their criteria for language vs dialect may be different from other people's, but as far as the Ethnologue is concerned, it's still a good scientific linguistic work.
    – curiousdannii
    Commented Apr 4, 2016 at 11:06
  • 1
    @curiousdannii: I have seen what The Ethnologue does do German and its dialects and their point of view looks almost ridiculous to me. Commented Apr 8, 2016 at 8:42
  • 3
    You're both right. SIL are the only people to have data at all for a huge chunk of the world's languages, particularly including Papua New Guinea; you simply can't assemble a list of the world's languages without SIL. They are also overenthusiastic splitters, and the quality of work of the missionary linguists is variable (as you'd expect from people who aren't professional linguists, and as I can confirm from having been RA on a project using their material). While the SIL has an agenda, though, I don't see how elevating 50 dialects of German to languages advances that agenda. (cont) Commented Jul 11, 2018 at 2:17
  • 2
    And if anything, being a splitter rather than a lumper makes it easier to get along with the tribes they work with, who themselves are splitters about how distinctive their languages are. It's not just the SIL who respect native speakers' notions of how distinctive their languages are: the linguist count of 200 Australian indigenous languages is heard less now, because there are 600 tribes historically. Lumping is what an external linguist does, with taxonomical motivation, splitting is what a native speaker does, with identity-driven motivation. Commented Jul 11, 2018 at 2:20
  • 1
    As someone who has never heard of this source, I'm curious why you say they have an agenda to divide and conquer. Is there a reason for this? Commented Feb 4, 2021 at 19:49

Are the languages spoken in various Arabian countries actually mutually intelligible? If no then it makes more sense to regard them as separate languages.

In China the government likes to officially categorize various Chinese languages as "dialects", but the reality is that the difference is really huge between some of them, e.g. comparing Mandarin and Cantonese is more like comparing Italian with Spanish than American English with British English. Most Mandarin speakers don't understand nor speak Cantonese at all. Therefore there are many language codes for various Chinese languages as well and it makes sense to me.

I wonder whether the situation is similar in Arabic: If you can't even understand some of the Northern African Arabic then how can you claim they're the "same language"? Some people might try to do so politically but linguistically it would be far-fetched. The situation is just fundamentally different from American vs. British English I suppose.

  • SIL set 30 Arabic languages in 639-3. I said Arabs that are not from Morocco may have problems to understand the Moroccan slang language. So for now it's 1 of 30 that is 20% or less different from Arabic maybe. What about the other 30 like the one in the Gulf and in Egypt?
    – uak
    Commented Apr 9, 2016 at 10:10

Unifying and subdividing speech forms under an ISO code is not a rigorous ontological claim: it is the standardization statement "this linguistic thing is to be abbreviated that way". "Quechua", "Luyia", and "Arabic" (also "Latvian") are examples of macrolanguages, under ISO 639-3. The individual languages are different enough that they are not all mutually intelligible. Cypriot, Juba Arabic (pga), and Chadian are quite different, and it is well-known that Moroccan Arabic is "challenging". I haven't experiences Tajiki or Uzbeki Arabic but I don't find the claims to be too incredible: we can leave it at that (I am skeptical about N and S Levantine by the mutual intelligibility criterion). The point is that there are significant linguistic differences between these varieties of Arabic, and linguists have documented even more differences than are presently recognized in the ISO system. (See for example the numerous varieties documented in T. Prochazka's book Saudi Arabian dialects). All of the forms are still unified with a single identifier ara which refers to the Arabic macrolanguage; individual varieties also have their identifiers. The Glottolog system makes even more distinctions (40) – subsuming Maltese under their arab1395 (not included under ara in the ISO system), subdividing Libyan Arabic into 4 varieties.

The same is true of the various Luyia languages, which are unified under luy, but also distinguished with particular identifiers (lts for Tachoni, lri for Marachi, and so on); idem the Quechua languages. Somali is presently not indicated as a "macrolanguage", it is an individual language, but the main reason for the failure to subdivide is the lack of information on the extent of difference.

It is thus not surprising that these languages enjoy subdivisions in ISO 639-3 classification. What is somewhat surprising is that there is no subdivision in the ISO analysis of English, where there are dozens of varieties that are mutually unintelligible in the same say that Moroccan, Syrian and Juba Arabic are. Glottolog does, however, subdivide ("Mercian") English into 35 varieties, plus 4 Scots varieties (which ISO registers as a distinct language, sco, from English).


"Also except for some variations in North Africa Arabs" - this isn't just "some variations" - Moroccan Arabic and Tunisian Arabic are significantly different from Lebanese or Saudi (and from each other as well - with Algerian being a sort of middle between the two), far more so than British and American, more like the difference between Portuguese and Spanish or Spanish and Italian.

As to people being able to understand each other, especially across the Middle Eastern variants of Arabic, that has a lot more to do with TV and Pop Music than with linguistic unity. Everybody understands Egyptian for example, because of exposure to Egyptian movies, but if you examine Egyptian Arabic, it is actually very different from other forms of Arabic.

As a native speaker of Arabic myself (Maghrebi Arabic), I would argue that Arabic as a single language with the different regional variations being mere dialects or slang has a lot more to do with Arab nationalism than it does with proper linguistic classification.

  • 2
    Yes yes yes, the OP's post mostly reflected whatever he was taught to think. Statements like "I've never heard an Arabic man call its region's dialect a language" are non-falsifiable, because we can't really disprove that he hasn't heard the prominent eg Lebanese who call for exactly that, and tautological, because the Maltese for example simply aren't considered Arabs. Commented Sep 3, 2019 at 15:08
  • Could you please, provide evidence for your claims?
    – uak
    Commented May 1, 2020 at 2:12
  • @alex Well, that's your own opinion. I've stated in my comment here my thoughts about the "Maghrebi" variation. Does this variation have any official recognition or is it taught in school or recorded by official language bodies? linguistics.stackexchange.com/questions/17232/…
    – uak
    Commented May 1, 2020 at 2:16
  • 1
    @Akkad so you are denying my, and millions more Tunisians, Algerians, and Moroccan's heritage because it is not recognized by any official body? Or is it because it offends your Nationalism? Many times I've spoken pure Tunisian Arabic to an Egyptian or to a Palestinian and was asked to "Can you please stop speaking French to me" - even though all of the words I said were words with Arabic roots. That's how different it is. If you don't want to believe it, that's your problem, not a scientific fact. Commented May 1, 2020 at 22:44
  • @AlexKinman Your language have been effected by the French invasion so much that when you speak Arabic it sounds like it's French. like when an Arabs starts learning French they will sound speaking an Arabic version of French. Do you accept the French effect in your heritage?
    – uak
    Commented May 3, 2020 at 3:26

I'm not able to give you an authoritative answer, but here is my attempt at one. See "Macrolanguage" in Wikipedia and in SIL for more information.

The distinction between a language and a dialect is mostly a political or social question, not a linguistic one. Thus Serbian, Croatian, and Bosnian were universally regarded as one language (Serbo-Croat) until the breakup of Yugoslavia.

Having said that, mutual intelligibility is often taken as an important indicator. Although there have been British (especially Scottish) films that have been subtitled for US audiences, nearly all English dialects are mostly comprehensible to speakers of other dialects. This does not appear to be the case for Arabic: according to a Wikipedia article, "Some varieties of Arabic in North Africa, for example, are incomprehensible to an Arabic speaker from the Levant or the Persian Gulf."

As far as I know, there has never been a strong move anywhere in the Anglosphere to treat different varieties of English as different languages: I don't know whether or not there has been in the Arabic world; but the people who assigned ISO639-3 codes (members of SIL) seem to have regarded the different varieties of Arabic as different enough to require them to be listed separately within a macrolanguage.

  • 1
    Thank you for your explanation, yet there is no official reorganization to any dialects in any Arabic country as far as a I know. You are right we do not understand some varieties of Arabic language in North Africa but what about the remaining 30 other language?
    – uak
    Commented Apr 4, 2016 at 3:30
  • 5
    I don't know what you mean by "official reorganisation". What I do know is that the fact that most parts of the Arabic speaking world teach MSA in schools is irrelevant to the fact that they speak different varieties (whether you call them dialects or languages). Pretending that there is only one Arabic may suit certain political or nationalistic aims, but it misrepresents reality.
    – Colin Fine
    Commented Apr 4, 2016 at 22:57
  • Sorry I meant to write "official recognition". If we can understand each other like English speakers from all around the world understand each other, How come it became languages?
    – uak
    Commented Apr 5, 2016 at 21:07
  • I don't know. You'll probably have to ask SIL, or find an account of their deliberations, to get a definitive answer.
    – Colin Fine
    Commented Apr 6, 2016 at 10:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.