Are there any examples of languages that are extremely similar but are nonetheless considered to still be separate languages? Or the converse, dialects of one language that are extremely different from each other?

The difference between "language" and "dialect" is of course not well defined, but I'm wondering if there are cases where the definitions of these words are particularly stretched, and why.

By "languages" and "dialects" I mean those generally considered to be so by linguists, non-linguist scholars, and/or the general public. So for example Serbian/Croatian or Hindi/Urdu are valid not examples of "similar languages" since most scholars consider them to be slightly different versions of the same language, and e.g. Mandarin/Cantonese are valid not examples of "dissimilar dialects" since they're often considered to be different languages.

  • 4
    I can assure you there are people who consider Serbian and Croatian to be separate languages. For purely political reasons that have nothing to do with language, of course, but even so, such people most certainly exist. I assume the same is probably true of Hindi and Urdu as well. Feb 1, 2021 at 19:53
  • @JanusBahsJacquet good point; edited question slightly
    – Aqualone
    Feb 1, 2021 at 21:01
  • I'm hoping for an answer in an extremely different dialect. Feb 2, 2021 at 5:18
  • Are you primarily asking this from a linguist point of view or from the point of view of native speakers? Because I can tell you that the Malay family of languages are complicated. In one extreme I personally don't understand the east coast dialect of my own country. In another extreme I can both read and understand standard Indonesian which Indonesians and increasingly Malaysians consider a different language. However I find it really hard to follow non-formal spoken Indonesian because our languages have diverged in the last 20 years or so
    – slebetman
    Feb 3, 2021 at 17:26

2 Answers 2


In Norway, Valdresmål and Bokmål are very different, perhaps comparable to the difference between Norwegian and Swedish. Most people, linguist and non-linguist alike, consider Norwegian and Swedish to be different language, although there is a minor (dismissive) sub-meme in linguistics that consider them to be dialects of Norwedish (a.k.a Swegian). Most linguists have no opinion about Valdresmål and Bokmål as language vs. dialect. I don't know if anyone has conducted a scientific survey, but I think most Norwegians consider them to be "dialects" of Norwegian.

The general rule is that most linguists are maximal splitters, so if it's reasonable to call A and B separate languages, they will do so. But we are largely led by local sentiments, to the extent that we know anything about such sentiments. This can lead to a historical progression of terminology, where categorization changes when we learn more. An example is Dogri, an Indic language of Jammu, which was considered to be a dialect of Punjabi. After a bit of real research on the language, it became apparent that it has to be treated as a separate language. Similarly, Hawrami has been treated as a dialect of Kurdish (this is an ethnic issue), but the linguistic evidence suggests (I'll remain neutral on the correctness of the conclusion) that it is not a dialect of Kurdish, it is on a separate branch of Northwestern Iranian. Similarly, Karaga, Kalanga, Ndau, Manyika, Korekore and Zezuru have been considered to be "dialects of Shona", but now the received opinion is that they are mostly if not all separate, related languages.

In other words, there are numerous cases where it is assumed that two speech forms are a single language and if one is identifiable (has a name), it is a dialect, but deeper digging reveals that the assumed language is really more than one language. SIL invented a term to cover this problem: macrolanguage. Thus what was once the "Luhya language" is not the Luhya macrolanguage, containing many specific languages (ones formerly termed "dialects").

  • Do most linguists consider Norwegian and Swedish to be separate languages? Most non-linguists certainly do, but in my experience, most linguists – in the Scandinavian countries themselves, at least – consider Danish, Swedish and Norwegian to be closer to dialect groups of the same (macro)language than really distinct languages. There are far greater dialectal differences within each ‘language’ than between the closest variants of neighbouring ‘languages’. Perhaps Scandinavian linguists are just more likely to be minimal splitters than elsewhere? Feb 1, 2021 at 21:08
  • I've seen more disagreement over than than any other set of well-known languages, also incidentally only considering the opinions of non-Scandinavian linguists. Unfortunately, there isn't a reliable polling system for getting opinions of "the linguists". Most linguists in Scandinavian countries know a Scandinavian language or two so they have a better basis for judging. But most Scandinavian linguists are outnumbered by most linguists, dessverre.
    – user6726
    Feb 2, 2021 at 1:18
  • I suspect a strong driving factor is that most Scandinavian linguists find all three languages, excepting perhaps some of the more extreme dialects (nobody understands this fisherman), to be completely mutually intelligible, and it instinctively just feels weird to be having a perfectly fluent conversation using all the same words, and then calling it different languages. Despite what The Bridge would have people think, many non-linguists do struggle more with the mutual intelligibility, which would make them more likely to see them as separate languages. Feb 2, 2021 at 1:35
  • 1
    Mutual intelligibility is not a binary thing. Standard Danish and Standard Swedish are often not mutually intelligible at all. I also don't agree with the premise that they "use the same words". Afrikaans and English aren't mutually intelligible because "my pen is in my hand" happens to mean the same thing in both languages when written down.
    – Agnes
    Feb 2, 2021 at 8:18
  • 1
    @Agnes Standard Swedish and Standard Danish are mutually intelligible to a great many speakers (though of course not all) with a little bit of practice. ‘Not mutually intelligible at all’ applies only to a very small minority of speakers. The coincidence in vocabulary is far greater than single sample phrases – at least 90% of the vocabulary is cognate, and the differences in grammar and syntax are no bigger than between, say, Italian dialects, or indeed between Scottish English and AAVE. Feb 3, 2021 at 0:28

There are several examples of languages that are very similar, but still considered different languages. This is mostly for political or religious reasons. Examples are Urdu and Hindi, which are mutually understandable on a basic level, but divert in their script (Arabic vs Devanagari), and in advanced vocabulary (derived from Arabic vs Sanskrit). Similar, Servo-Croatian was broken into two languages after the break-up of Yugoslavia, Serbian (Orthodox, Cyrillic script) and Croatian (Catholic, Latin Script).

On the other end of the spectrum, we have for example the various languages of the Philippines, which the Spanish, and after that the Americans and until this day many of the Filipinos themselves call dialects, although they are quite distinct and not mutually understandable.

Mutually understandability is not a symmetric thing: speakers of a more prestigious language or dialect will claim not to understand the less prestigious version, but the other way around you will hear no such issues. A language like Afrikaans, which is very closely related to Dutch, is fairly easy to understand for Dutch speakers, but the other way round, appears to be a bit harder.

An old linguists' joke is that a language is a dialect with a an army.

  • Welcome. Be aware that certain language pairs are explicitly mentioned in the question as not valid examples for this question. The question already mentions Urdu/Hindi and Serbian/Croatian, aka Hindustani and Serbocroatian. Feb 2, 2021 at 9:12
  • You're correct about that, but it would be a bit hard to generate examples, if the most widely known examples are excluded... The other part of the question was why these definitions are stretched, and that is actually the aspect I wanted to address. Languages are in a sense genetically related, borders are drawn pretty arbitrarily, based on politics, religious and cultural affinities, and socio-economic status.
    – Jeroen
    Feb 3, 2021 at 10:17

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.