What non-Semitic languages have templatic non-concatenative morphology?

Question

Which languages, if any, outside of Semitic have something like Semitic-style root-and-template morphology, with roots that can be analyzed as consisting of consonants only, and the vowels coming from a separate morphological tier? I've heard that some Native American languages have similar systems, but I don't know which.

I realize that Germanic ablaut (for example) is a similar kind of phenomenon, but it's not basic to morphology in Germanic the way that it is in Semitic; I'm asking about languages where this is the basic strategy of verbal and/or nominal inflection.

Answer 1

A pure separation of consonants=root and vowels=affix would be way out at the extreme end of templatic morphology — so extreme that it doesn't describe what's going on with the nouns in Semitic languages or even a substantial proportion of the verbs. Almost always in "root-and-pattern" systems, you'll find roots that have their own inherent vowels and CV-patterns (or syllable structures), but the templates often ride roughshod over those vowels and CV-patterns, and sometimes the consonants too.

The most notable/studied native American language of this type is Yowlumni (usually called Yawelmani until recently). You'll find some similar features in Muskogean languages, and probably several others.

You'll also find templatic morphology at the next level up from Semitic, in other Afro-Asiatic languages like Berber and Hausa.

Answer 2

The oldest level of Proto-indo-european had such morphology, but it was somewhat simpler with two consonants or consonant groups and one vowel between. The strong verbs in Germanic languages is a rest of this state. There are evidence indicating that the Semitic type evolved from this type, and this again is used as a support for the Nostratic hyotesis.