Passive Voice

I liked this idea too. But from what I saw in social media, it's clear that people still use simple sentences or copy English grammar. Horngoth also had passive voice suffix but without any clarification how to use it.

As for your examples, it's hard to distinguish patient from agent of action. Either we should use -irzi (by) suffix as in English to mark  patient of passive action or make a strict rule for word order. Better both. I'm unsure if person suffix should depend on agent of patient of action. I prefer suffix based on subject=patient

Golug gimbaguzat olog-irzi - The elf was found by the troll
Golug gimbaguzat - The elf was found
Krith nazgu thrâkaguzut sharazu matûrzu**
** adjectives take plural ending too, but animated/person nouns do not have plural. Seems 99.9% people including myself always forget this. That's why this was changed in new dialect.

If you want get rid of English cliches, then I strongly recommend to familiarize yourself with Zhâburi dialect. It's based upon grammar of ancient Hurrian language.

I don't think that using stand-alone pronouns is strictly forbidden. Was it ever stated by Tolkien himself? And adjectives after nouns should also be quite a loose rule (but still a preferable variant).

Attaching pronouns to verbs is very tricky. First, I think attached pronouns of subject serve as suffix of verb's person (but this doesn't suit well with Shadowlandian 3rd person; should we add both -at and -ta then?).

Kuluz-izg dîsuga olog-irzi
Kul is not needed in easy active present tense sentences. But in such complex constructions it carries information about tense, and serves as anchor for subject pronoun.

Dîsaguz-izg olog-irzi
yes, -irzi suffix is not very important

I think Scatha's translations of English prepositions were quite literal. She tried to distinct space and time prepositions, but there was not any system. So "-irzi" can have both meanings.

This is very close to what I liked in Zhaburi. The word order there is Subject-Object-Verb, but as we can place object pronouns after verb, your construction suits it. However, Zhaburi has Ergative case marker for agent, but patient is not marked with case endings. However object pronouns have special form for joining the verb. In that case we don't need passive ending too. So, your sentence would be like "Olog-(ergative case marker) dîsuz-izish". And "Troll was attacked by me"/"I attacked the troll" both will be like "olog dîsuz-izg" or "izg-(erg. case) olog dîsuz". Please note that actual Zhâburi has different suffixes and vocabulary, and there's no examples when a pronoun is an agent of transitive verb.

In situation where both could show up, yes, we would use both. It would be:

She attacked him
ta dîsuzatta

ta dîs -uz -at -ta
she attack -past tense -he/she/it person -him/it (object)

I attacked him
dîsuztaizg

dîs -uz -0 -at -ta -izg
attack -past tense -I/we/us person -him/it (object) -I (subject)

I also don't think we need to add z between suffixes if they are separate syllables. Each suffix should be clearly understandable (at least in Classical Black Speech, Debased Black Speech can do whatever it wants, and since Orcs slur stuff together this makes good sense):

I attacked him
dîsuztaizg

dîs -uz -0 -ta -izg
attack -past tense -him/it (object) -I (subject)

I get the way kul-/-irzi is used now with pronouns, but I'm still not a fan of it. It could be a feature of Debased Black Speech, but English doesn't exist in Middle Earth (the conceit of the books having been translated from Westron into English by Tolkien) so it seems unlikely that it's grammar would show up anywhere on Middle Earth.

Is the lesson on Passive Voice something you wrote by yourself? I don't remember it being a part of Scatha's lessons.

The -ag suffix is what would turn it from "He attacked me" into "I was attacked by him" without moving the subject/object or using the kul-/-irzi helpers. This is my proposed way of doing it to replace the kul-/-irzi way.

I noticed I actually made a mistake, and it should be:

ta dîsaguzatizish
I was attacked by him

ta dîs -ag -uz -at -izish
He (subject) attack -passive marker -past tense -him/her/it person -me (object)

I've finished the verb code to parse all the suffixes, so now I have a way to double check that I'm doing things correctly. Hopefully this will lessen my mistakes. The forum I'm coding for the Red Hand site will have a built in way to link Black Speech to the parser, so it'll be easy to check words (at least when using LoS CBS). I tested it with your examples in lesson XIII: ghâshaguzatûk ufubulâzh azubulûk-izg

I'm still working on nouns, but I have a lot done. It currently parses the noun examples from lesson XIII too: dûshatârnizkûfzagus ronktablataghâraz

It still won't parse narsrinkhshârzusharzrad though. I'm not even sure what it means, lol! Can you break it down into it's parts so I can see what it's supposed to mean?

Don't delete them yet. I'm still learning stuff from them

I like the way you have them defined. It makes everything much more organized. I've always liked the subject pronoun going at the very end.

That might be a good idea. I will probably add both for now.

Okay. I see what the issue is. The adverbs have been turned into suffixes, while the rules on the adverb lesson say that adverbs come after the word they modify or at the end of the sentence, the word order in Appendix B also shows them as stand alone words, and the suffix guide in Lesson XIII doesn't list adverbs as being suffixes. It should be, according to the rules in the lessons (and my parser can understand it like this):

narsrinkhshaarz sharz rad
not-understand-adverb very now
not very understandably now

"understandable" would be the adjective form, so to get that English translation it would have to be narsrinkhshaûrz  sharz rad Not very understandable now.

Since neither -arz nor -ûrz are listed in the verb suffixes in lesson XIII, I decide to put them in with the infinitive suffix, as that suffix also changes the verb into something else, and they seem to be only added to the root of the word much like they are for nouns. None of the other suffixes will make much sense once they have been converted. At some point I'll be adding code to parse adjectives and adverbs (right now it only breaks down verbs and nouns) to better handle that anyway.