Learn Mando'a

Wikipedia: "In linguistics, an agent noun (in Latin, nomen agentis) is a word that is derived from another word denoting an action, and that identifies an entity that does that action. For example, driver is an agent noun formed from the verb drive."


So what are some examples of this in Mando'a? Well, we have:

Lead- Alorir
Leader- Alor

Protect- Cabuor
Protector- Cabur

Win- Parjir
Winner- Parjii

Not exactly a pattern, especially with words that might look entirely different from their verb forms because they weren't created with the verb in mind. For example, Commander is the Leader of Soldiers instead of the One Who Commands (which begs the question, how much can you shorten a word that it can still be considered that word? Isn't "al" the word for "but"?).

In any case, I haven't looked at every single example, so I might be complelty glossing over something. Using my very limited grammar knowledge and Mando'a skills, what I came up with is just taking the three examples listed above and following what it is. If the verb without the verb ending (-ar, -er, -ir, -or, -ur) ends with an "r", we simply remove the ending. If the word without the verb ending ends with a vowel, we replace the verb ending with an "r". If the word without the verb ending ends in a consonant, we replace the "r" with the prior vowel.

Thus:

Buy- Verborir
Buyer- Verbor

Speak- Jorhaa'ir
Speaker- Jorhaar

Fight- Akaanir
Fighter- Akaanii


That's about it. If I missed something, and I'm thinking I did, well, mirdi'ni utreekov.

Interesting look into this pattern- I hadn’t considered it myself. I figured I’d add that the way I usually create the part of speech you’re describing is to just add ‘ad to the end of whatever I was trying to agent-ify. To relate it to your example, verborir might become verbor’ad. Or, akaan would be akaan’ad- I find this one to be especcially backed up by the fact that akaan’ade, the plural, is listed in the dictionary as ‘army’. (The alternative I’ve been using for inanimate objects is adding ‘keb (from kebi/kebise), but this hasn’t come up often- the one time I can remember trying to use this is to come up with a word for something along the lines of a locker. I concluded with ‘jur’keb’, still not sure how I feel about that…) On the other hand, I see an extra dimension in one of your examples- adding ‘ii’ also seems to indicate a person’s origin, as in jetii, aruetii, or kaminii.
Just some thoughts and experiences that seemed related. I’m glad you posted this, I hadn’t realised how distinct a grammatical concept it was to think about! I’m always looking for new ways to get a point across, and as someone who is very intetested in languages, a grammar pattern newly spotted in the wild is too tempting not to explore.

How unfortunate for me to have discovered Naastar-Naast for Destroy-Destroyer just now then. We will have to look at this further. Your idea of using 'ad' and 'kebi' is interesting.

Unless we cheat and simply say that stems ending with "t" act the same as the ones with "r".

I suppose a better way to go about it is to just call Naast an exception to the rule instead of coming up with a new rule.

I'm kinda in the middle on this, I like -ad and -keb for their clearness, but the existing corpus makes it clear that it can't be a hard rule. We might have to accept there's multiple ways to indicate agent status, and all of these methods share a measure of validity.

Well, Mando'a is a very personal language. Canonically, people just say things based on a very large template but with their own quirks, I think. Am I confusing this concept with something else?

But with -ad or -keb, might it not seem that the action is being done to a person or a thing instead of the -ad or -keb modifying the word?

I dunno, it's not the first assumption I'd make. Plus, I feel in most cases, the context would make it clear enough

Alright then. -ad and -keb it is, I suppose.