How did Miskito ligatures develop?

Question

I'm interested in the use of "ligatures", but I'm most interested in the way the ligatures and the possessives can use a combined infix-suffix. I don't understand how such a mixed system could develop. I understand the progression

u tla 'house' > utla (u or tla is class marker?)

u a tla 'house (-a- ligature) > watla

u a m tla 'your house (m 2nd person) > wamtla

but I don't understand how you could get mixed classes.

duri 'boat' > dwarka 'boat (-a- ligature)

duri > dwairka 'my boat'

duri > dwarkam 'your boat'

http://en.wikipedia.org/wiki/Miskito_grammar

Answer 1

I don't know the history of the Miskito case, but doubling up on morphology is a known development in the history of language. Here are two examples from English/Germanic:

1. The English word child once had plural childer (regularized from the original invariant plural child). Another pluralizer -en was then added, giving modern children

2. In Proto-Germanic, there is a class of verbs called preterite presents. These were perfective in Proto Indo-European, but reanalyzed as presents in Germanic. The past tense was then formed by addition of a suffix (cognate with Mod. Engl. -ed). So, the past tense forms of verbs carried two "past" markers -- the original PIE ablaut and the Germanic suffix.

If I had to guess what happened in Miskito, I would say that these words were originally in an infixing class (duri ~ *dwari), and were subsequently reanalyzed as being a suppletive stem in the -ka class. (Suppletion is the operation that occurs when a stem changes form after the addition of a suffix. An example from Mod. Engl. verb inflection is the alternation between bleed and bled (< bleed + ed).) But that analysis is only a guess.