#4 is *not* empty; but your classifications are too simplistic for the actual data. In particular, I consider the assumption that if there is a genetic relationship and one is in 639-2 that therefore it should be a macrolanguage to be invalid.
Potential cases of languages
- with the same
- genetically related
- one is in 639-2 but is not a macrolanguage encompassing the other(s)
include the following (item in 639-2 listed first):
bas vs. bzw: These are two Niger-Congo languages separated at a genetic level that encompasses 961 languages.
bem vs. bey: These are two Bantoid languages separated at a genetic level that encompasses 513 languages.
fan vs. fak: These are two Bantoid languages separated at a genetic level that encompasses 659 languages
luo vs. luw: luw is not yet classified, so this pair may or may not fit the criteria. They are not regionally contiguous. Also, luw is all but extinct (known speaker population = 1), so there's likely no body of documents, and hence no useful reason to consider it encompassed by luo.
Now, in the first three cases, geographic proximity might provide a basis to say that a macrolanguage is appropriate in spite of the genetic distance. But because of the distance it's really a stretch to suppose that anyone would find it useful to tag documents from the two varieties the same.
Peter
_______________________________________________ Ltru mailing list Ltru at ietf.org https://www1.ietf.org/mailman/listinfo/ltru
Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.