Zompist Bboard Again

xxx wrote: ↑Sat Aug 31, 2019 1:45 am I see the forum as a club of explorers where anecdotes of travel in unknown countries are discussed rather than a specialized library where complete works are presented in large books that are difficult to manipulate...

I find this to be a beautiful statement of the purpose of the ZBB.

Moose-tache wrote: ↑Sat Aug 31, 2019 2:27 am I will definitely implement some of these ideas, especially the notion that I shouldn't be kicking down the door with the reference grammar.

So will I! I don’t have any conlangs anywhere near a publishable state at the moment, but I will definitely do this when I do publish one.

I've seen that grammar-last approach work well for other people (Did Mecislau create the Novogradian website before or after the reference grammar?).

I’m still unconvinced that the grammar-last approach is best. Personally, I think that the grammar should be published at the same time as other articles about the language, so that if you’ve finished reading about the interesting parts, you can go to the grammar to learn more about the language.

So the question then might not be how do you make a conlang reference grammar interesting to strangers, but how do you make it engaging to people who already have a passing curiosity about the language? How do you take that spark and keep it burning? There have been plenty of times when I was already exposed to examples of a language (say on Game of Thrones or Star Trek), only to be bored almost instantly when I tried to dive deeper into the available material.

This is a very interesting question; I don’t have an answer, but I wonder if anyone else has an idea?

A quick prompt to get examples flowing: after reading the Almean historical atlas, or Lord of the Rings, or Native Tongue or your favorite Star Wars fanfic, was there a language you sought out only to skim the grammar and give up early? What went wrong?

I haven’t actually tried to do this yet. However, I do find this to be an interesting question, so I’ll try to think of an interesting conlang to research, try and read its reference grammar, and report back on how it went.

I read the Láadan grammar by itself, not having read her novels or anything else. I was only interested in the language itself.

Pabappa wrote: ↑Sat Aug 31, 2019 8:54 am I read the Láadan grammar by itself, not having read her novels or anything else. I was only interested in the language itself.

for instance in Làadan, I always wonder what is the motivation why unnaturally lack some voiceless consonants...
impersonal grammars don't give answers for such questions...
(don't know for the novels...)

xxx wrote: ↑Sat Aug 31, 2019 9:47 am

Pabappa wrote: ↑Sat Aug 31, 2019 8:54 am I read the Láadan grammar by itself, not having read her novels or anything else. I was only interested in the language itself.

for instance in Làadan, I always wonder what is the motivation why unnaturally lack some voiceless consonants...
impersonal grammars don't give answers for such questions...
(don't know for the novels...)

The idea was to make it more easily pronounceable across speakers of disparate native languages, and one of the ways that was done was to not have the v/vl distinction. Why she went with the voiced sounds by preference, I don't know.

zompist wrote: ↑Fri Aug 30, 2019 10:48 pm I should try to create an issue of an Almean linguistics journal...

I'd be very curious to see that! Working through competing grammatical traditions could be a lot of fun.

For me, though, I couldn't really do these without having the reference grammars first, though perhaps not in complete form.

Yeah, I was definitely thinking more about what would interest me as an introduction to someone else's language, not what would be useful for actually working with the language.

Also FWIW, I like having a high-level introduction to the grammar-- I did this in rewriting the Verdurian grammar.

I liked this a lot.

A bit more generally, there's a real art in conveying a good sense of a language without needing a lot of words to do it. I hope I can figure this out someday!

akam chinjir wrote: ↑Sun Sep 01, 2019 12:04 am

zompist wrote: ↑Fri Aug 30, 2019 10:48 pm I should try to create an issue of an Almean linguistics journal...

I'd be very curious to see that! Working through competing grammatical traditions could be a lot of fun.

Indeed! On a similar vein, I've always been curious to see the in-world reconstruction of Proto-Eastern...

I consider a congrammar to be “engaging” when it answers “yes” to a proposal of marriage.

(Sorry, that’s all I could think of!)

Boooo, Tom! BOOOOOO!!!!!!

TomHChappell wrote: ↑Sun Sep 01, 2019 10:09 pm I consider a congrammar to be “engaging” when it answers “yes” to a proposal of marriage.

(Sorry, that’s all I could think of!)

(Grin) And just how many times have you seen that happen, Mr. Chappell?

Pedant wrote: ↑Mon Sep 02, 2019 7:00 pm (Grin) And just how many times have you seen that happen, Mr. Chappell?

The next time will be the first!
———
Serious answer; whenever a congrammar includes some feature I still haven’t figured out, I want to see how that conlanger handles it.

So my answer is probably in constant flux.

I do have some long-standing questions/problems/mysteries-to-me.
For instance, what if a language has tritransitive tetravalent verbs, and reciprocalization, and reflexivization?
It seems that, in such a language, there should be six different kinds of reciprocalization and six different kinds of reflexivization.


Assume one of those processes can be applied even after another one has already been applied.
What are the various things that could happen?

For instance, what are the various things that can happen to a tetravalent verb if it’s reflexivized then reciprocalized?
Or reciprocalized then reflexivized?
Or reflexivized then reciprocalized then reflexivized?
Or reciprocalized then reflexivized then reciprocalized?

—————

There are others.

—————

Some of them inspire puns.
Whenever someone has serial-verb constructions, I always want them to tell me about Jack the Ripper.
Whenever someone has a switch-reference system, I always want them to explain how the lights are controlled in the long corridor where my office is, that has three switches at each end, and three sets of lights.
And so on,

But my entire brain is pun-based; I think it’s hereditary.
So even my serious ideas start out as puns. I actually have more serious pun-based or chiasm-based or word-play-based ideas posted on these bboards, than attempts at humor. Normally, though, I work on them enough that the roots are hidden by the time I post them.

—————

Alright, I guess I can probably go out on a limb, and say if a conlang’s lexicon and morphology allows for some good puns, I’ll probably find it engaging.
Or some other good garden-path or punctuation-is-important type jokes, such as mistaken focus, in the syntax and pragmatics and so on.

I've never successfully published any of my dozen or so grammars because I always feel like there's there's a section or twelve that need to be completed and I just happen to find them a total slog to get through. Meanwhile, I'll often rewrite the same section on case assignment, valency adjustment or relational nouns fifteen times

Recently, I decided to simply open up my working Google Doc of my latest language as readable on the internet and it's been pretty sweet. Makes me think I should probably just port these things over to the open web ASAP and let it be ok that I don't find them 100% complete. Who's gonna fully read anyway? And the incomplete sections might spur someone to write me a message to complain which in turn might provide the necessary motivation to get it done. Who knows.

For my jobby job, I write and design brand guidelines, and I employ essentially the same basic rules for those as I do for grammar documents. Adapted from a work document, the rules are below:

- The overall structure is based on the principle of progressive disclosure. This means organizing the high-level sections in such a way that you never have to define a feature of the language far away from describing the feature in detail. If something demands a reference way down in the document to be understood, it may not be in the right place. However, sometimes it's inescapable. Typically, this principle leads to a structure of: Phonology, Nouns, Adjectives, Verbs, Word Order, Phrase Structure, Clause Structure, Complex Sentences etc. But often a particular language demands a different order. For Duriac, for example, the polysynthetic structure demanded me moving up Verbs ahead of Nouns, something I've literally never done before, but the little details only made sense that way. However, I then did have to give a high-level explanation of noun classes ahead of the verb chapter. So things like that will move around, based on what I discover is necessary building block. (I constnatly struggle with the location of Derivation, often placing it at the end in an appendix, but sometimes, integrate it into the morphosyntactic description of parts of speech).
- Never more than five levels of headlines, preferably three or less; subsections can be called out with simply bolding the keyword if the description is short enough. If a chapter is breaking down into more than five levels, it needs to be broken up into more sections.
- Every description of a grammatical feature is followed by an example, leading to a regular rhythm of: Feature description, example proof; feature description, example proof etc.
- The most generally applicable description comes first, followed by ever increasingly obscure or rare exceptions. By just glancing at the top of each section, you can understand the general rule.
- A defined term is italicized on first mention
- Ample use of bullets/numbered lists when breaking down multiple components of a feature
- The simplest, clearest language possible is used to describe everything and jargon is used sparingly. Watch out for grammar description weasel words like "simply" "just" "only" – it's very tempting to constantly try to make it sound like the language is easy. Just don't!

I will be very engaged by any congrammar which adequately handles, or valiantly attempts to handle, all or even most of the questions raised by this:
https://listserv.brown.edu/cgi-bin/wa?A ... 0yahoo.com
Or here: http://archives.conlang.info/zhu/felsho ... ghuen.html
(They’re the same thing, but one link might still work for you even if the other one doesn’t.)

A good grammar of course ought to be clearly written and well-structured. Use terminology which is standard in linguistics, but not too specialized. If you need uncommon terminology because your language has uncommon features (as in an alien language), introduce it well. Avoid writing within a specialized grammar-theory framework, as that will render the grammar unusable to people who aren't into that framework.

I also like it when a grammar tells something about the people speaking the language. It should have a good introduction which tells something about them. Not just something like "Skrillexx is the language of a species of mantid-like aliens in the Orion Arm". Write who they are, where they live, what other languages their language is related to, who are their neighbours, etc.

This goes on with examples. I like examples which tell something about the speakers' culture. My favourite example is Zompist's example of Xurno about the sculptor who hopes a govern a province one day. Many conlang grammars use examples which are too obviously constructed. (I have to admit that my own grammar of Old Albic, as it is currently presented on Frathwiki is badly wanting in this, but that 's just work in progress.) In natlang grammar writing, examples taken from actual texts and conversations are preferred; elicited examples are considered second-class evidence, and constructed ones no evidence at all.

And finally, your grammar ought to be descriptive rather than prescriptive. (Prescriptive grammars are OK if they are inworld texts. Inworld grammars can be fine if they are written well and actually useful - but then, many conworlds have no good grammar tradition, so all you get are useless, imprecise prescriptive grammars, in which case you really should write an outworld grammar!)

A lot of good answers have already been given here. Just another consideration on my part; when writing the Diachronic Phonology of the Sataw Language, I mostly struggled with:

a) writing coherent and grammatical English sentences.

but more importantly,

b) finding the way to include all the conworlding and cultural tidbits in a non-obtrusive way. Personally, the conlanging and Sataw folklore go hand in hand in about a 50/50 ratio. They inform eachother and one day I'm all morphosyntax, while the next I will obsess over creation myths.

Yet, the paper is devoted to the language, and devoted to a rather specific niche at that. To "respect" the space that a technical subject like diachronic phonology demands, I had to put a lot of the cultural tidbits in the subtext. So I considered which lexemes I cited. These were selected to stay in line with the general feel. For instance, you might find culinary and esoteric especially dominant. Other conworldy information I put in footnotes. The rest was put in the introduction. This way the hypotethical readership of the paper could "read around" conworlding stuff they might not be interested in.

In short, I find conworlding elements in a reference grammar extremely interesting, but it is advisable to make it so that hardcore conlangers can skip those parts and easily find the descriptions of the language.