Learning language in Kovol is like trying to drink from a fire hose. There’s always so much and you feel completely overwhelmed with it all. It’s kind of nice to have time to sit and pick over language data in a way you never have time for in the bush. It’s not a productive way of learning language. The fire hose of immersion is a much better way to learn to speak, but it certainly satisfies my sense of completion to be able to spend weeks poring over the data I elicited previously without a stack more arriving in my inbox.

I’ve been poring over the verbs that change based on tense and speaker. There are endings for I, you, he, we, you (plural) and they in the remote past, recent past and future tenses too. There are a few other endings too, but right now I’m concentrating on the main ones.

aminim, “to talk”
Remote pastRecent pastFuture
2samɔŋaŋgɔŋaminiŋ ig
2pamomwaaŋgɔmaamwa ig

There are patterns. I feel like we can predict the endings 70% of the time. Here’s another one:

piginim, “to put”
Remote pastRecent pastFuture
2spigɔŋpigɔŋpiginiŋ ig
2ppigomwapigimapigwa ig

I’ve written a Python script that takes the first singular remote past and recent past and uses those two conjugations to predict the rest. It works pretty well. When it doesn’t work out I have to ask myself a few things:

  • Is the prediction wrong?
  • Did I write the verb down wrongly (the answer is wrong)?
  • Did the speaker say it wrongly?

Learning an unwritten language is messy work! I remember the good old days of linguistic training where we had data sheets containing the 100% correct incontrovertible data to work with. Now I’m very aware of the fact that fallible me gathered the data!

That counts as correct right?

Then there are the times my predictions are just totally wrong. I wonder if we have a group of irregular verbs, or if we have different kinds of patterns or both. Some verbs seem to be unique, but there does seem to be a group of verbs that change as the tense changes.

For lack of a better name I’m calling them tense change verbs because the tense causes the root to change. Normally only the endings change, but here we have tɛβ changing to tɔβ. Here’s a list of some ‘tense change’ verbs we’ve found:

lɛβinim, “to come down”
mɛŋgɛrɛminim, “to search for something”
nɛnim, “to work”
tɛnim, “to throw”
tɛβinim, “to come from the side”
wɛndinim, “to hit or dig”
wɛnim, “to shoot / to sing”

Notice a pattern? They all have ɛ as their first vowel. Unfortunately the pattern isn’t always the case and some verbs that have ɛ as their first vowel follow the normal pattern! Argh!

Will further study reveal a crystal clear unambiguous answer? Or will it be a messy mess that we just muddle through? My vote is it will be the messy mess of the real world of learning an undocumented language 😀

Wow, that was nerdy. Who wants to see a picture of some babies? They now sleep from 11:30pm till 6:30am. Good job babies 😀


Colette Harding · 29/06/2021 at 9:34 pm

Thanks for the insights on the challenges of learning and writing an unwritten language. We pray for all of you involved in this difficult but needful work.

Lois Snyder · 30/06/2021 at 8:22 am

Very cute babies! Glad they are sleeping better. Of the verbs you listed, I like “piginim” best because the stem stays completely the same. And it also has many of the “recent past” and “remote past” verbs that are the same. Good job deciphering verb patterns! It looks like you have a good start.

Leave a Reply