Return to the ALL LANGUAGES ArchiveForward to the Current ALL LANGUAGES discuss

senator16Thursday 26th of May 2005 04:16:28 PM
LANGUAGE OR DIALECT: - Language or dialect?:

Short definition: a language is a dialect promoted by élites.


Sometimes one can hear people speaking about "those tribes in Africa with all their dialects" while the same people speak about "European nations with their languages". Without necessarily intending to do so, one can in this way hierarchise people and what they speak. In Debi Prasanna Pattanayak's view (1991: 27-28), "the developed countries treat their respective dominant languages as resources, call them world languages, and use them to further their national interest', while those of the 'third world élites' who follow the West 'deride the mother tongues' in their own countries 'as dialect, slang, patois, vernacular, and condemn them to marginal use, or completely ignore them" (ibid., 28). But we can also hear a genuine question: is what XX speak a "language", or is it a "dialect"? Can the question be answered? What is the difference between a language and a dialect?

There are no linguistic criteria for differentiating between a language and a dialect (or vernacular or patois). Structural similarity or dissimilarity can only tell apart very dissimilar languages. It is easy to confirm that, for instance, Chinese and English, or Kurdish and Turkish are clearly different languages because their linguistic structures are so dissimilar. But despite being structurally very close to each other, Swedish, Danish and Norwegian are called different languages. Serbian and Croatian may be even closer to each other but they are now (again) called two different languages. Hindi, Urdu and Punjabi are both structurally and lexically very similar, Kannada and Marathi are structurally almost the same but lexically dissimilar -- all are called different languages. Structural similarity can thus mainly be used to differentiate between two languages in cases which are so clear that no linguists would be needed anyway to solve the problem. In other cases, linguistic criteria are not of much help.

"Mutual intelligibility" has also been used as a criterion: if you understand a "language", A, without being taught that "language", it is a dialect (or another variety) of your own "language", B. Or your own "language" B is a dialect of the one you can understand, A. Or what both of you speak (A & B), are dialects of some third entity, C, which is then called "a language". But if you don't understand A, it is a different language. But the criterion of mutual intelligibility is also far from unambiguous. Let us say that speaker A understands B, and speaker B understands C, who in her turn understands D. On the other hand, speaker A does not understand C, and speaker B does not understand D. Where is the boundary then between language and dialect. Or if A understands B but B does not understand A (non-reciprocal intelligibility), are A and B dialects of the same language for speaker A who understands both, but two different languages for speaker B who does not understand both? In situations where languages are oral (spoken) languages and have not been reduced to writing, people in neighbouring villages often understand each other, either well, or at least to some extent, despite the differences, but they may not understand people from villages much further away. These in turn understand their close neighbours, etc.

How well do the speakers need to understand each other? Is "semi-communication" enough (Haugen 1966: 102) or must the understanding be "complete" (and is it ever complete even between speakers of the same language)? Should the speakers who test the criteria be monolingual? It is, for instance, easy for me (Tove Skutnabb-Kangas), knowing other Indo-European languages like Danish, English, German, Latin, Norwegian and Swedish, to understand some Dutch, without having ever been taught Dutch. Would Dutch then be a separate language for a monolingual Swedish-speaker who does not understand Dutch, but a dialect of Swedish, or German or English, for me?

Is oral understanding enough, or should we rather use understanding of writing as a criterion? Or the opposite: is understanding writing enough, or should one also understand the oral mode? A Finn who has studied Swedish at school, understands some written Danish, but does not understand spoken Danish at all. Is oral Danish then a separate language from Swedish, while written Danish is a dialect of Swedish? And what about the deaf population?

Should the criterion be used only with language spoken by a native speaker, with normal speed, or can a second language speaker who speaks slowly also be used? Age, amount of formal schooling, degree of metalinguistic awareness, amount of exposure to the language or to other languages in general, learning styles, courage, motivation, fatigue, etc, obviously also affect intelligibility, in many situations much more than the "same language/different languages" question. Mutual intelligibility as a criterion thus discriminates well only in situations with structurally unrelated languages, as was the case with the structural linguistic criterion too.

Neither similarity or dissimilarity of structure, nor mutual intelligibility or lack of it can therefore differentiate between languages.

The social functions of languages, measured, for instance, by the speakers' own views on what are different languages, are based partly on the two linguistic criteria (structural similarity, mutual intelligibility), but mainly on extra-linguistic criteria. One possible criterion which has been suggested is standardization. Only dialects which have been reduced to writing (a prerequisite for standardization) and been standardized are languages, everything else is something else (dialect, vernacular, patois). Peter Trudgill's old definition (1983: 16) reflects this; for him "languages" were "independent, standardized varieties ... with, as it were, a life of their own". This would drastically reduce the number of "languages" in the world. Very few indigenous languages and only a handful of sign languages would qualify as languages according to this definition. But it can be understood in the sense that it only becomes natural to speak about a language as a specific, discrete unit, distinct from other similar units, when there is a written form of that language, claims Tore Janson, earlier Professor of Latin, now Professor of African Languages, at the University of Gothenburg, Sweden (1997: 125). The written forms of today's languages in Europe displaced and replaced other ways of writing. In most cases, a written form came first and a name for the language only afterwards. One or some of the dialects were chosen as the basis for the written form, and the choice was obviously made by those or to benefit those who "needed" the written form in the first place: the élites, the state builders, the church representatives. These choices were also decisive for inclusion and exclusion: the rulers decided where the borders would be placed in the dialect continua between what was called one language and what another language.

Thus, the main criterion for whether something is a dialect of another language or a separate language (and what is being standardized, what not) is the relative political power of the speakers of that language/dialect. The decisions about what are "languages" and what are not, are thus political decisions. Those with enough power can claim that what they speak is a language and what less powerful groups speak are dialects. Political definitions of a language would be: "a language is a dialect with an army (and a navy)" or "a language is a dialect with state borders" or "a language is a dialect promoted by elites". (TSK)

keaThursday 26th of May 2005 06:01:37 PM
- I can bring an example about languages and dialects around here. In Estonia we speak Estonian, which is called a language but in South part of Estonian country they still speak vőru keel, which is clearly related closely to standard Estonian but still different enough that standard Estonian speakers don't understand the speakers of vőru keel really. But officially it is a dialect of Estonian.
But at the same time there are two very small languages (around 10 speakers left of each language) that are very closely related to Estonian - Livish and Votjan. And the distance between Estonian and those two languages is no bigger than the difference between Estonian and the South Estonian dialect Vőru. But Livish and Votjan are called languages. I think the reason is geographical, because when Vőru dialect is spoken inside Estonian borders, where the official language is closely related to this dialect then Livish speakers (the number of them is many, many times smaller than the speakers of Vőru dialect) live in Latvia and the Votjan speakers live in Russia. And Latvian and Russian are not related to Livish and Votjan at all.
UlvenSaturday 28th of May 2005 02:45:10 PM
- If you haven't already Senator, see this thread--->
[url]http://phrasebase.c../../discuss/read.php?TID=6201[/url], called The Mis-use of the term 'Arabic', by Ashlee.

Myself and Ashlee have run into major dilemnas in our pursuit of the Lebanese language (aka Arabic). There is a website link which shows the extreme shortcoming of the mis-labelling of all Arabic script languages as 'Arabic'. There's no doubt in my mind that the Arabic tongues are languages, not dialects.
I also pursue the Scandinavian languages, and they're no voids compared to Arabic tongues. But, you do need seperate learning material for each of Danish, Swedish and Norwegian. But whether they're dialects or languages, I could go either way. I'll wait till I'm better at them before I decide.
But as for Arabic tongues, that's a no-brainer, really. I'd say there are at least three seperate languages under the title 'Arabic'.
But when it cuts close, dialect vs language comes down to the difficulty level for the individual. I don't think the label can ever be gospel,(as you yourself seem to have pointed out).
TeupSaturday 28th of May 2005 04:30:02 PM
- For those interested, there's been a topic on it in the Dutch discuss [url=http://www.phrasebase.c../../discuss/read.php?TID=3121]here[/url]. Labeling is indeed rather arbitrary, in reality there is no "language" or "dialect", everone speaks a littlebit different, it's all on a continuum. There are no clear cut boundaries and it depends on your point of view. Many variants of Dutch could be considered different languages for example, because they're as far apart as for example Swedish and Norwegian or Serbian and Croatian (therefore generally Serbo-Croat is preferred). Afrikaans is considered a different language, although it's about just as far away from Dutch as some dialect variants. It depends on politics, or rather social factors in general - here the goal is preserving a unity (and Afrikaans is spoken in a different country anyway), in the Balkans it's about creating a seperate group identity.
UlvenSunday 29th of May 2005 01:01:34 AM
- You're right Teup. Naming something a language can represent a goal for the future of a community. Even if the difference between two dialects now are small, if they intend to go their seperate ways, like Serbia and Croatia, and form seperate identities, it makes sense for them to get the ball rolling and label their intent from the outset. Politics certainly can polute the passion behind any drive for seperateness, but that doesn't mean there isn't a legitimate passion in forming one's own identity. It's exciting forming uniqueness, and it has to start somewhere.
I suppose the Arabic world hasn't strived to name their languages seperately because they have more concern seperating themselves from western society. If things calmed down there, then we'd see a pusuit for Arabic's so-called dialects to assert their seperateness. They'd have greater freedom to sit with their language and contemplate. But at the moment, they have more pressing needs than thinking about what to name their speech patterns. There's quite some complexities involved. Even I admit to missing alot of them.