How do language borders work?

February 20, 2025 10:43 AM

Curious about how language borders work, especially inside a country. What makes a lanaguge just come to a stop at a certain point?

I understand language borders that are also country borders. Each country has its official language, borders are not so easy to cross, etc.

But I was recently looking at maps of countries like Belgium, Switzerland, and you have a village (A). One mile to the east is another village (B) that speaks the same language, one mile to the west is another village (C) that speaks a different language.

In the case I was looking at there didn’t seem to be any geographical barrier, and the two villages were in the same region (same canton of Switzerland).

Languages spread through contact so I wonder why, historically, the language spread from Village B to Village A but not to Village C.

Would neighbouring villages speaking different languages just not have mixed at all back in the day? Would people have viewed it is a foreign village and just not have interacted?

What makes a language just *end* at a certain point?

I suppose there’s maybe no clear answer to my question, is there a keyword I could look up to read more about this?

posted by sefsl to Society & Culture (19 answers total)

What makes a language just *end* at a certain point?

Mountains. Since one group found it hard to get over or around a mountain, the languages developed separately. The reverse is true of rivers: since it's easy to traverse a river, the same languages spread out over a great distance.

"Guns, Germs, and Steel" explains a lot of this.
posted by Melismata at 10:49 AM on February 20

One of the big things is that borders have shifted continuously throughout history, and what you may think of as Belgium and Switzerland once didn't exist or were split up very differently.
posted by Tell Me No Lies at 10:49 AM on February 20

The simplest answer is different people who spoke different langauges settled these villages and without a reason to change languages, they didn't, because changing languages is hard, and why would you not want to speak what you grew up speaking, even if you became multilingual? Groups of people may also be separated by religion, occupation, or a million other preassure causing but non-geographic formations.

A book that is much less problematic and just-so than Guns, Germs, and Steel is Danubia, by Simon Winder. He covers the ethnic mixes that existed in the Hapsburg lands up through the end of the First World War and notes that the varying ethnicities are often varying language groups and people could switch their group through marriage or moving.

Belgium was formed when the Catholics in the former Spanish/Hapsburg lands separated from the Protestants. So they shared religion but not language. Nowadays, it is administratively divided into language regions and services are provided in the given language, with the exception of Brussels, which is located within the Flemish portion of the country but provides services in both languages. Interestingly, this is starting to cause issues as gentrification pushes Francophones into the Flemish suburbs.
posted by dame at 10:56 AM on February 20

Not to threadsit but I am going to add a specific example. In Switzerland there is a town called Sierre. The next town over, which is along the river and basically just across a couple of fields, is German speaking. From what I can tell they have always been part of the same country. What could have happened there?
posted by sefsl at 10:56 AM on February 20

A group of Germans went together and founded a village. A group of French folks went and founded a village nearby because the valley was fecund or whatever. They were probably large clans. I think you need to motivate why people would change without being forced to at the end of a wallet or a gun.
posted by dame at 10:59 AM on February 20

This question is discussed at length in The Horse, the Wheel, and Language and the answer is not always intuitive. In particular, the author claims that language boundaries historically often don’t follow topological boundaries. On the other hand, cultural boundaries can create language boundaries that persist for centuries, even with no physical boundary. TLDR it’s complicated.
posted by Winnie the Proust at 11:00 AM on February 20

After this I am going to shut up, but it is also worth remembering that large areas where everyone speaks the same language are less natural than one might think. The UK and Ireland; France; and Italy, for instance, were all home to multiple language groups for a long time and that only changed under cultural and military force.
posted by dame at 11:05 AM on February 20

What could have happened there?

What could have happened is a perceived or actual cultural difference (beyond just the language) that prevented mixing, intermarriage, social interactions, even commerce. Each group had their own schools, stores and churches without the need to even learn the other's language. But nothing is static. Over time, these barriers break down to some extent, usually by the imposition of a lingua franca in which necessary government business, trade and education are carried out. Eventually maybe one language wins out over the other. Look at how native speakers of Manx disappeared. Or, people care enough about their own language, keep speaking it in their local settings, and preserve the difference. BTW all of this applies also to historic local accents, such as you can find all over England, and in regional and ethnic differences in the US.
posted by beagle at 11:07 AM on February 20

Countries in europe were really an urban phenomenon up until about 150 years ago. The book Peasants into Frenchmen is really about the transformation of rural France during the 1800s---most rural people in France in the 19th century didn't speak 'french' per se, but spoke some dialect. You could go on village over and have a hard time understanding those people. This may be more stark when its not a dialect but German and French, but the logic is the same. I forget how old written Russian is but in War and Peace everyone wrote in French.
posted by MisantropicPainforest at 11:19 AM on February 20

You'll also find that most of people in those mixed-language regions do know the other language - they just identify with the language spoken at home, and "the village" speaks publicly the majority language. People do move, but if you haven't lived long-term in a country with another official language you have no idea what a tiring headfuck it is. You literally forget your language of origin if you don't use it every day, and you do have to communicate with locals to live your life.

And even if a family moves permanently, the kids will learn the local language at school and get to the point - common with second-generation immigrants to America, for example - where they understand the family's original language but don't speak it. By generation three or four it's just "oh his mum is of X origin". Especially if the kids keep marrying locals, the assimilation is quick. Where subcommunities do preserve their language, it's usually part and parcel of being culturally distinct enough to make intermarriage more difficult, including by making raising kids in X culture a condition of creating the mixed family.

Fun lesson from forced mass population movements in Eastern Europe after World War II: the only way to fully change the language spoken in an area is to move a new community there so that it becomes a majority. Even banning Silesian and Kashebian didn't make people not speak it at home (and banning education in Polish under the Russian Empire was an even more fun failure). But eighty years after the displacements, the mayor of Lviv speaks with the same Polish accent as people from Wrocław - because most Poles from Lviv were relocated to Wrocław, while some stayed in Lviv and the mayor learned Polish from their children.
posted by I claim sanctuary at 11:21 AM on February 20

From what I can tell they have always been part of the same country.

Without very detailed study, how would you know this? There were very frequently moving borders in this part of the world, over many centuries. It would not be surprising at all for there to have been a border at some crucial point.
posted by ssg at 11:22 AM on February 20

I forget how old written Russian is but in War and Peace everyone wrote in French.

That would be fashion though - since Peter the Great and forced Westernisation of the upper classes (so language-by-edict and to some extent fashion because you don't want to speak like the peasants you exploit). Cyrillic is from the ninth century and the oldest written Russian text is around a thousand years old.
posted by I claim sanctuary at 11:24 AM on February 20

One thing I'd ask is just whether your map shows areas of extensive bilingualism / multilingualism.

My recollection is there was sort of a famous study by the sociolinguist John Gumperz where he looked at multilingualism in a village in India and found really remarkable convergence between languages from different language families. Googling a bit, I suspect it was this study: Gumperz, J.J. and R. Wilson. 1971. "Convergence and creolization: A case from the Indo-Aryan/Dravidian border."

Maybe disregarding the keyword creolization--a different problem with its own issues, rejected in this follow-up--what Gumperz and Wilson actually found is summarized in this more recent study / update: Kulkarni-Joshi, S. 2016. "Forty years of language contact and change in Kupwar: A critical assessment of the intertranslatability model":

G&W made one observation and one prediction in their study:

i. Marathi and Hindi-Urdu ... and ... Kannada in the village of Kupwar had achieved morpheme-to-morpheme intertranslatability.

ii. ... multilingualism across languages belonging to two different language families would be maintained as long as the ethnic (i.e. religious) separateness of home life remained important for the communities in the contact situation.

In other words, the map may be misleading, bilingualism / multilingualism may be common, interesting things can happen in that contact zone, and yet (unsurprisingly) social issues probably determine a lot to do with language maintenance.

On the other hand, that re-study of Kupwar reached a conclusion that changes to the map itself may have had implications for more recent language change:

In 1960 the bilingual region was reorganised into two distinct states each with a separate, officially sponsored language, each belonging to a different language family ... [L]inguistic reorganisation of the region in 1960 was proposed as the prima facie cause of recent language change and divergence in the region.

I haven't really read this study too closely though--it just had the best summary I could find of the older study, and the updates seemed like a bonus.
posted by Wobbuffet at 11:35 AM on February 20

Canada has a bit of this going on in its official language communities as well, which can be seen in Western Quebec and Eastern Ontario. People just kind of settled around their own language group and did somewhat mix with the neighbouring ones, so you can almost see that the other language knowledge is kind of like a gradient, getting stronger where the language communities border each other (in western quebec and eastern ontario, for example) and with very little second language knowledge in the regions far from each other, like eastern Quebec north of the St Laurent and, say, most places in British Columbia where there are six languages (including English) spoken by more people in BC than French.

And even with that high knowledge of the second language, in places like Ottawa you still see the francophones living mostly in the east of the city, in the historically more francophone areas like Orléans and Vanier, and the anglos in the western parts of the city. Montreal is also very similar. It's definitely about access to services in one's mother tongue and proximity to other people who speak it too.
posted by urbanlenny at 11:38 AM on February 20

If you think about it, there might be a version of this phenomenon, or one related to it, going on where you currently live - especially if you live in a city with a lot of diversity. As an example, do I understand all of the slang and linguistic patterns people younger than me use? Probably much of it, if they're from the same overall socioeconomic and ethnic group as me; but the farther I move away from my age and my background, the more likely it is that there'll be things I won't understand, and that I'll just speak differently, overall, than they do. This would happen because my life doesn't intersect with theirs much; the people I spend most of my time with talk more like I do, and maybe the people they spend most of their time with speak more like they do. Even though we live in the same city, in the same time period, with a decent overlap in the media and cultural content we're exposed to.

AAVE, a dialect of English with entire grammatical structures of its own, similarly often exists right alongside other dialects of English. Two people who have lived in the same city their whole lives, maybe in neighborhoods right next to each other, might speak different dialects of the same language, and not even be 100% intelligible to each other.

The AAVE example brings up another factor that can reinforce this diversity, which is that the language you speak can be part of your identity. You might code-switch and speak or write in a different way in some contexts, and some of that might just be what comes naturally to you, but some of it might also be a conscious choice to preserve and reinforce your identity, to treat language as a cultural and group marker, and so on. Talking like people in a different group do can feel like losing a part of yourself. A Flemish-speaking area in Belgium is not going to just up and switch to French for a lot of reasons, but one of them is that it would spark a huge cultural identity conflict/crisis. In Barcelona, insisting on keeping Catalan alive and strong instead of assimilating to Spanish has political weight.

Someone made an important point about how the idea of national identity and national language in Europe is fairly new. Identity-wise you used to belong to some city or some lord's domain. That (often, and especially for cities) dictated which people existed in your social circles, and also which people you felt kinship with and wanted to identify with (or felt enmity with or were restricted from interacting with. Combine that with the relative difficulty of travel, and other cultural dividers like religious affiliation and historical descent, and it's not weird to get a situation where village A has a different linguistic identity than village B and they both feel strongly about maintaining those identities.
posted by trig at 12:07 PM on February 20

Just some conceptual food for thought... It sounds like your mental model for how linguistic geography changes is akin to a virus, or maybe a wildfire-- where contact itself has a high likelihood for change. However, it's not like either of those. People need a reason to learn a new language (or even moreso, to raise their kids speaking some language other than the one they grew up with). You're coming at the thought process from the standpoint of "Why wouldn't these people start speaking Language X?" but the more deciding question is really "What would impel people to raise their kids to use Language X foremost?"
posted by dusty potato at 12:07 PM on February 20

This Wikipedia entry on French Switzerland seems to suggest it's more mixed at the edges than you are suggesting. And while Sierre is majority French, there's about 8% of the population that speaks German.

Seems like the valley with Sierre was French speaking and then some Germans moved in partway:
"Historically, the linguistic boundary in the Swiss Plateau would have more or less followed the Aare during the early medieval period, separating Burgundy (where the Burgundians did not impose their Germanic language on the Gallo-Roman population) from Alemannia; in the High Middle Ages, the boundary gradually shifted westward and now more or less corresponds to the western boundary of the Zähringer possessions, which fell under Bernese rule in the late medieval period, and does not follow any obvious topographical features. The Valais has a separate linguistic history; here, the entire valley, as far as it was settled, would have been Gallo-Roman speaking until its upper parts were settled by Highest Alemannic speakers entering from the Bernese Oberland in the high medieval period (see Walser)."
posted by bluedaisy at 1:02 PM on February 20

You'll also find that most of people in those mixed-language regions do know the other language - they just identify with the language spoken at home, and "the village" speaks publicly the majority language.

This, many people in Switzerland speak multiple languages and are used to dealing with languages they don’t speak. People absolutely interact and always have interacted outside their dominant local language. But there is also a really strong local identity and strong local dialects even within the same language. This article on Alemannic German is fascinating and includes a neat map. I’ve lived in Basel and later Zurich for more than ten years now but I still don’t understand all Swiss German variants with ease…for context, I am a native speaker of German.

There is the concept of place of origin for those born Swiss. To become a naturalised Swiss citizen the process involves three levels of approval. The first is being granted municipal citizenship in your local municipality, often a small town or village. There are formal criteria around language and length of residency but also around integration. If they don’t think you’re integrated in the local community and don’t want you, you don’t get municipal citizenship. If they approve, the canton follows and the federal government rubber stamps these approvals and gives you a Swiss passport. So it is all about local identity to this day.
posted by koahiatamadl at 2:46 PM on February 20

The reasons for language borders often vary. The reasons in Switzerland are most likely to be strong affiliation with a home village. In central Europe it's often to do with the history of war and settlement, and the rise and fall of different powers. In eastern Europe, whether someone was considered to speak language A or related language B might depend more on their religious affiliation than anything else. As centralised states developed, related dialects typically came to be seen as versions of the named language for that state - explaining why the langue d'oc and the langues d'oïl spoken in France came to be seen as dialects of French but the much more linguistically similar plattdeutsch and leegsaksies spoken in north Germany and east Netherlands were seen as dialects of German and Dutch respectively. (A more linguistically sensible thing to do might have been to say that they are neither German nor Dutch but dialects of a different language, Low Saxon.)

In Wales, the language border between English speaking and Welsh speaking areas is fundamentally a function of immigration or lack of immigration from England during the industrial revolution with the impact of forced English making the Welsh area smaller than it might otherwise have been. Except for very tiny children, everyone in a Welsh speaking village will speak English fluently so they could all choose to use English exclusively if they wanted to. But I think Welsh would only stop being the primary language of the village if either too many monolingual English speakers moved in or too many Welsh speakers moved away.
posted by plonkee at 6:10 AM on February 21