Off and on I have been working on revising (and adding to) my grammar of Jamna Kopiai. The original is here. I'm well aware the original suffers from serious readability problems: large walls of text; a dearth of examples; a syntactic transformation notation that was intended to be simple and illustrative but quickly ballooned out of control.
Beyond these generalities, I am asking for specific suggestions on which parts of the grammar need to be re-written or re-structured. What parts are too big a wall of text? Where have I been too jargony for easy comprehension? Where have I droned on to a degree that becomes uninstructive? What things are confusing or poorly explained? Where are the worst problems in how the information is organized? Et cetera.
1. It seems really, really weird to me to break up morphology and syntax so deeply. I realise that there are other grammars that do this, and that it may be marginally more convenient for an expert who only occasionally needs to look up a few inflections or syntax rules but not both, but it's definitely anti-didactic, and for a conlang's grammar I think it's a lot more useful to assume the reader knows little or nothing about how the language works.
2. The "categorization" paragraph at the start of "issues in Jamna morphology" piques my interest with the irregularities it suggests and its talk of syntax and derivation—but the stuff it talks about is nowhere to be seen. I think you should try to make an effort to break up the absolute dead obvious noun and verb inflections (e.g. TAM, noun case, noun number, etc.) before you introduce grey areas (e.g. the this/that/which series) and finally clear-cut derivational morphemes, such as the first set of haaioari.
3. You introduce the stem endings for accusative, dative, and nominative without first saying definitively what cases the language has and what they're used for. This, I think, is the worst consequence of the grammar's overall division into syntax and morphology.
4. The verb section is a little better because it starts with a chart describing all the parts of the verb, which helps the reader get a bit of a sense of everything that's going on—I was happy with it up until the second sentence of the TAM complex section, which told me I had to go read the syntax section to figure out how this part of the language worked. Again, breaking up the grammar this way rebukes the student.
5. If you're referencing another section, can you throw in an internal link to an anchor tag? Kinda adding insult to agony here.
6. Both the syntactical and morphological sections on formants are disasters—the morphological one is basically redundant, and the syntactical one needs a summary table. The way this section is written suggests that formants don't have clear-cut uses, requiring the patient memorization of the entire thing in order to be sure one does not conflate them. (I'm particularly annoyed that it takes to the section paragraph of "-so" to figure out what it does.)
7. "multiple preference patterns" – empty heading before "other adjustments" and a number of empty headings within "other adjustments"... but I won't report on those any further.
8. I'm hesitant about the "other adjustments" section. On one hand, it's not immediately clear to me how to write it differently, but at the same time it feels like it sharply perpetuates the tradition of being overly analytical at the expense of didactic value. At a minimum this section needs a lot of examples and, ideally, headings that also describe what each adjustment is used for; this may just be me being an inexperienced linguist, but I have my doubts that someone asking "how do I say x in Jamna Kopiai?" is going to immediately jump to the section entitled "IDO extraction," particularly as I can't even figure out what it means. (The "postposing (de-extraction)" part is backward and needs to be introduced more clearly for the roles it provides, while the remaining adjustments seem somewhat easier to grasp at a glance.)
...hopefully that'll be enough to get you started; I realise it leaves at least half of the grammar uncovered.
Thank you! That is extremely helpful feedback of exactly the type I am looking for, and I really appreciate your taking the time to look through the grammar well enough to spot these things.
A few comments on some points:
1) The logic on splitting morphology from syntax is that the latter deals with all the grammatical machinery, while the former is meant to be something like a set of morphology tables gathered together in one place for easy reference so you don't have them spread willy-nilly across a whole book - but including description of how word shapes work when using them, since this is a matter somewhat accidental and disjoint from the big ball of rules that make up the grammar grammar. Some languages are better suited to this style of description than others - it is quite decent for Latin or German, but abysmal for the likes of Mandarin or Haida. I think I will be retaining the split, but I will do some re-thinking about how information is organized and located between the two sections, and ensure that relevant morphological tables are repeated within the syntax section for easier reference.
6) I knew they were disasters even as I was writing them originally - at the time, I couldn't think of any better way to go about it. The TC sections are not far behind. In both cases the sections need to be reunited and also rewritten, probably from scratch. Even how I treat the formants is probably going to change - my descriptive approach to them is awfully clunky and confusing. Not that I intend to change their nature; but might it be less confusing to first lay them out as grammatical voices, and then describe how they also affect semantic role interpretation, ending with how each verb is married to one of them whether the voice applies or not? Or something along those lines?
8) Well, I am not really envisioning many people trying to use the conlang, let alone learn it; so instructive value is not meant to be ignored, but is meant to come in (a close) second behind providing as comprehensive an analysis as I might. More examples I can and indeed should provide, as well as better description of what the various adjustments are for. Beyond that, I'm not sure what more I could do to improve instructive value.
1) One last plea: why not have both? Merge the syntax and morphology sections, then have an appendix of tables. After all, it works for teaching grammars.
6) That sounds promising. Above all else, include lots of examples.
8) My real issue is that I feel you set out a lot of vague descriptive categories without providing a concrete justification—e.g. a grammar of a Bantu language can justly get away with vaguely summarizing what each noun class tends to include without being exact, but it would be indefensible for an English grammar to go over the basic verb forms and merely mention the existence of irregular verbs. I feel there are some parts in the grammar where such gaps exist; the description of what verbs prefer the "-so" formant feels particularly obtuse and unhelpful. In the "other adjustments" section, this mostly manifests when you describe qualitative criteria (e.g. in object demotion and IDO extraction) without examples.
The result of these omissions is that the grammar errs on the side of being a featural overview, not a technical breakdown or teaching resource; it feels sometimes like someone was conducting a census for a overly-detailed, Akana-only version of WALS, and didn't really care about how the language works.