This is a framework commonly used by German sociolinguistics to talk about different ways that languages is defined and distinguished from dialects (see also: languages vs dialects). The main terms are:
- ausbau language: this is a standard variety of a language; it is defined as a language because it’s the language of a government, taught in schools, has style guides for publications written in the language, etc.
- abstand language: this is a language which is defined as a language because it is sufficiently different from other languages that it cannot be defined as a dialect of one of those (i.e. it fails the mutual intelligibility test).
- dachsprache: this is a language variety (generally an ausbau language) that provides a “roof” over other language varieties that may or may not be mutually intelligible with it. For example: Modern Standard Arabic as a dachsprache for Arabic, Southern Quechua as a dachsprache for Quechua, standardised Romansh as a dachsprache for the natively spoken Romansh dialects, or the fact that identical dialects spoken just across the Netherlands-Germany border are considered dialects of Dutch or German, respectively, depending on which side of the border they’re on (they’re “roofed” by different ausbau languages).
The first two terms were coined by Heinz Kloss in 1952. Žarko Muljačić coined “dachsprache”.
Basically, the framework helps to explain the different ways that people do define and distinguish between languages, rather than how people “should” define or distinguish between languages. So if someone were to ask, say, “Are Norwegian and Swedish dialects, or independent languages?” you could answer, “They’re independent ausbau languages, but not independent abstand languages.”
There is a difference between a pluricentric language and mutually intelligible language varieties that are standardised as separate ausbau languages. In Kloss’s original framework, he recognised three degrees of separation between language varieties. Where the standard languages are based on an “identical or near-identical” language variety, like with Serbo-Croatian, English or Portuguese, that makes for a pluricentric language. Where the standards are more divergent, but mutual intelligibility still exists, that’s where you get the situation of languages that are different ausbausprachen but not different abstandsprachen. Finally, you have languages that are different on both measures, like Dutch vs German.
To me, that still raises the question of where you draw the line between “near-identical” and “not identical but still mutually intelligible”. Like, I would personally consider Spanish a pluricentric language, and Spanish’s own language academies consider it a pluricentric language, but its dialects are more different than I would call “near-identical”. Perhaps it doesn’t matter because those language academies are not claiming to regulate different languages? But then you have the Catalan context, where the Valencian right-wing (who are no longer able to do their old trick of calling Valencian a dialect of Spanish) now insist that Valencian is its own independent language (so not the same language as Catalan). And Valencian is not called “Valencian Catalan”, it’s called Valencian. The two standards are not very different but I also wouldn’t call them “near-identical”. I would consider this a pluricentric language too, and Valencian’s own language academy agrees, but apparently the majority of Valencians (albeit mostly older and uneducated people) see it as a different language. I guess it can still only be a separate ausbau language to the extent that it is recognised as a separate language at all, though, and that’s by pretty much no one with linguistic training.