Dialect, language and who gets to decide

Linguists are asked constantly where the line is between a language and a dialect, and the honest answer is that there is not one that works.

Mutual intelligibility fails

The intuitive criterion is whether speakers understand each other. It falls apart quickly.

Intelligibility is often asymmetric — speakers of variety A understand B better than the reverse, usually because A's speakers have more exposure to B through media and education.

It is also gradient rather than binary, and it is influenced by attitude. Speakers who consider themselves to belong to the same group report understanding each other better than speakers who do not, given the same linguistic material.

And the criterion produces results nobody accepts. Several varieties officially designated as separate languages are more mutually intelligible than varieties officially designated as dialects of a single language. Speakers across some national borders understand each other easily and call their speech different languages; speakers within some countries cannot understand each other and call their speech the same language.

The dialect continuum

The underlying reality in many regions is a continuum. Each village's speech is slightly different from the next, differences accumulate with distance, and at some point along the chain speakers can no longer understand each other — with no sharp break anywhere.

Such continua historically ran across large parts of Europe, and comparable situations exist elsewhere. Drawing a line anywhere along one is arbitrary from a linguistic point of view.

Where the lines do get drawn is at political borders, which do not correspond to anything in the speech itself. Once drawn, they become real, because education and media on each side pull the varieties toward different standards.

How a standard gets made

Standard varieties are not the oldest or purest form of a language. They are, in most cases, one regional variety — usually the one spoken near a centre of power or by an economically dominant group — that was selected, codified and promoted.

The process typically involves: selecting a variety, codifying it in grammars and dictionaries, giving it functions in administration, education and publishing, and elaborating its vocabulary for new domains.

Once a standard exists, everything else becomes non-standard by definition, and speakers of other varieties are treated as speaking incorrectly rather than differently.

The consequences of that framing

Sociolinguists have documented the effects extensively.

Children who arrive at school speaking a non-standard variety are frequently treated as having a language deficit rather than a different variety, with measurable effects on educational outcomes and on their own assessment of their ability.

Non-standard varieties are associated with lower socioeconomic status, and speakers face documented discrimination in employment and in legal settings. There is a body of research on how listeners' judgements of competence and credibility vary with accent, independent of content.

None of which reflects anything about the varieties themselves. Every natural language variety has a full grammar, is rule-governed, and can express anything its speakers need to express. That is not a diplomatic statement — it is what the evidence shows.

The famous line

The linguist Max Weinreich popularised the observation that a language is a dialect with an army and a navy. It is a joke and it is close to accurate as a description of how the distinction is actually made.

Cases where a variety's status changed following political change, without any change in the speech, are numerous. So are cases where varieties that were considered separate languages were reclassified as dialects of a national language as part of a unification project.

What speakers do

Most people in most places command more than one variety and shift between them according to situation — a phenomenon documented everywhere it has been looked for.

This is not confusion or corruption. It is a skill, and it is the normal condition of language use. Speakers who use a regional variety at home and a standard at work are doing something linguistically sophisticated, and the framing that treats one as correct and the other as sloppy misdescribes it.

What has changed recently is that written informal language has become far more visible. Text and social media are frequently written in something close to spoken varieties, which has made non-standard writing publicly visible in a way it was not when writing was mostly published or professional. The reaction to that from people invested in standards has been predictable and it is a very old argument in new clothes.

The practical point

Standard varieties are useful. A shared written standard allows communication across a large area, permits education at scale, and is genuinely worth teaching.

What is not useful, and is not supported by anything, is the idea that the standard is inherently better, or that the varieties people actually grew up speaking are degraded versions of it. Those varieties are the language. The standard is a selection from it that was given a job.