Author Topic: Language isolate inside a language family?  (Read 4666 times)

Offline Raptor

  • New Linguist
  • *
  • Posts: 3
Language isolate inside a language family?
« on: January 19, 2020, 09:09:37 AM »
First of all, yes, I know what a language isolate is (eg. Basque, Elamite, Sumerian etc.)

I am wondering if there is such a language that is classed in one language family (eg. Romance, Semitic etc.) but inside that family is relatively isolated from the rest of the languages in its family.

For argument's sake "Language A is in the X language family with languages B, C, D, E and F, however Language A is not very mutually intelligible with them"

Could you point me at the languages where this is the case?

Offline vox

  • Jr. Linguist
  • **
  • Posts: 37
  • Country: fr
Re: Language isolate inside a language family?
« Reply #1 on: January 20, 2020, 08:42:55 AM »
I'm not an expert but I think that Armenian, Albanian and Greek are often considered to be the only members of their respective branch in the Indo-European language family, even though they have different dialects (there's maybe a Hellenic branch for Greek).

Those languages are part of the Indo-European family but you mentioned Romance and Semitic families. It means that there's maybe an issue about the level of classification to determine whether a language is isolated or not. Being isolated inside the Romance family would be a very relative isolation. "Isolation" could be used as a mere word to describe such a situation but I don't think that as a taxonomic concept it would be relevent at such a low level of classification.
« Last Edit: January 20, 2020, 09:28:50 AM by vox »

Offline Daniel

  • Administrator
  • Experienced Linguist
  • *****
  • Posts: 2036
  • Country: us
    • English
Re: Language isolate inside a language family?
« Reply #2 on: January 21, 2020, 01:53:30 AM »
Vox's answer is good. The idea of 'isolate' is really misleading, because it's only about our current state of knowledge, not the actual (pre)history of a language. Basque, for example, must have had ancient (prehistoric) relatives, even though we cannot identify them today.

The term "family-level isolate" (or similar) is used for what you describe, and Vox gave good familiar examples, but you'll find many, many more, just depending on how you define "family-level" (or similar), and based on how much information is available for that group.

Similarly, to address both points, typical "isolates" are isolates at the family level, and where those families happen to have only one member to our knowledge. Note that this really is just about our knowledge and not real-world facts, such as the example of the Yenesian family containing Ket, along with several recently-extinct relatives (thus Ket is not typically called an isolate). But this idea of isolated families is important because there is some clear, inherent truth to Japanese and Basque being "isolates" (at least in the traditional use of the term, but also meaningfully in terms of not being able to link them to other languages elsewhere, at least not without being speculative), but both of these are actually more complex. Japanese is often said to not be an "isolate" because it is actually part of the small Japonic family containing Japanese and Ryukyuan, which was traditionally considered a dialect of Japanese but now a separate language, so Japanese enthusiasts will insist that in some sense Japanese is not actually an isolate, but this really misses the point: in that sense, then, Japonic is the isolate, and so would be Indo-European, etc. The case of Basque is also worth consideration because Basque is really a dialect cluster/continuum, without complete mutual intelligibility at the edges, but a standard (Batua) used today allows communication for everyone. The point is that "isolates" don't necessarily need to be simple scenarios to be isolated, and that the simple idea of a "language isolate" may be a myth.

Family-level isolates are trickier to define (what's a language, what's a dialect, what's a family?), but intuitive examples are found in many families. Vox mentioned Greek and others in Indo-European. There's Egyptian/Coptic in Afro-Asiatic. And within smaller families (hence the difficulty in defining "family"), there's only Gothic within East Germanic (although originally there were multiple varieties of Gothic, even though only one survives at a reasonable level of attestation for comparative work). Or similarly we might think of Romanian as the sole survivor of Eastern Romance, but that's actually misleading because it's just the only major language (e.g. national, with many speakers in the group), alongside minority languages like Aromanian, in addition to other varieties attested historically.

In the end, either an isolate is simply a genetic grouping of one language without any known relatives, or it's up to somewhat subjective interpretation on defining limits.
Welcome to Linguist Forum! If you have any questions, please ask.