Conlangery SHORTS 30: Revising a Grammar

Podcast – Conlangery Podcast

George spends some time talking about his recent revisions of his Istatikii grammar, with a focus on organizing writing to serve the needs of the language and the readers.

You will find the two drafts of the Istatikii consonant processes below.

istatikii-phonproc-draft1Download

istatikii-phonproc-draft2Download

Script below the fold, see the history here.

Welcome to Conlangery, the podcast about constructed languages and the people who create them. I’m George Corley. I’m making a short for this month, due to some scheduling issues for our planned episode. Before we get on with the show, I have a couple of announcements:

First, we are supported entirely by our patrons over on Patreon. Part of my recent revisions mean that if you’ve pledged at $10 an above, you may already know just about what I’m going to say today, because you’ve had access to my script as I’ve been writing it for about a week prior to recording. I appreciate patrons who contribute at every level, as it helps me ensure I can keep the podcast going.

Second, the Eighth Language Creation Conference will be held on the 22nd and 23rd of June at Anglia Ruskin University, Cambridge, United Kingdom. If you’re going there, I’m trying to arrange for someone to run around with a recording device interviewing people for me, so look out for that. For those who won’t be there, the Language Creation Society will be streaming it live. I’ll link to some more information in the shownotes.

Now, onto our topic for today:

After finishing my dissertation, I took a little time to look at my grammar for Istatikii. My goal is to finish it up enough that I can present a proper “published” version before moving on to another conlang project. Life has been a bit hectic, so progress has been slow and halting, but I did manage to start some revisions, and I wanted to talk about something that came up in those revisions.

My first task is to go chapter by chapter and revise what I have written. I followed a pretty classic format for a reference grammar breaking up topics into phonology, morphosyntax, etc. So, the first chapter I have set about revising is the phonology chapter. I had broken that down into sections on consonants and vowels, with each of those having sections on the phonological processes that affect them. On my reread, I noticed that the organization of the phonological processes wasn’t really working for me.

The problem was mainly in the consonants, as they had a long list of changes, where vowels simply have the vowel harmony system, which is complex but contained. My consonant phonological processes were organized into three categories: assimilation, deletion, and “other”, with that “other” section containing only the final devoicing rule. This was a logical idea when I was first writing, but on a second reading it didn’t really work for me.

I noticed that organizing by type of process did not aid the natural connections between processes. For instance, stop gemination and nasal deletion interact in Istatikii, because the nasals participate in gemination, and thus stop gemination can bleed nasal deletion by turning a nasal coda into an oral stop. But in the old version, those two rules were far apart, and there was just a minor reference back to stop gemination. Someone reading through might miss the connection or have to jump back and reread. There was a lot of this going on in the rules — someone would have to jump around the list to refresh their memory or see a rule that hadn’t been mentioned yet. Non-linear reading is of course going to happen in a grammar, and you do want to be sure you cross-reference between chapters and sections, but small sections should be readable linearly as well.

Seeing this problem I looked at my rules and all their feeding and bleeding relationships and reorganized them in a more reader friendly manner. For Istatikii, that meant an organization based on natural classes — the stops and nasals interacted as above, and liquids had their own interaction between a deletion and gemination rule, so I made stops and nasals one section, and liquids another section. There is a third section entirely devoted to glottal deletion rules, as the glottal stop and /h/ have deletion rules that interact with both nasal deletion and liquid deletion. In that case, the idea is to present it last so that the reader is able to relate these rules to rules they already know.

Is this organization the best possible for Istatikii? I’m not sure. Will it be the go to for every language? I think on that the details are going to really depend on your language. Everyone will have rules interacting in different ways, and every conlang is unique. I think any conlanger who wants to present their conlang to others should do a readthrough of their documentation from the perspective of someone new to their language. Have someone else take a look if you can, as well.

I’m thinking about how I should present things in other sections of the grammar as well. For instance, Istatikii devotes a huge section of morphosyntax entirely to prepositions, giving a great deal of time to the two classes of inflecting prepositions.. In fact, the Class I prepositions section basically goes through each preposition and lists out its uses with examples. At the time, I felt that was the best way to present it, and it is similar to how I handled Aeruyo’s moods, which also have lots of noodly uses that need listing, but I have been reconsidering it recently.

The main issue I have is that the information I have on those prepositions is going to be replicated in a much condensed but still similar form in the dictionary. Redundancy is not necessarily an enemy — it can be useful to present the same information in different ways in your documentation, because readers won’t necessarily read linearly and they won’t necessarily get things the first time. However, I have been considering recently that I could pare down the prepositions just to their inflectional paradigms and the basic syntax of the prepositional phrase, and then use the examples I have in a section that covers syntactic constructions more generally. This would help me to build the syntax section, which I’m a bit stuck on how to start, and to think more about constructions aside from how prepositions are used in them.

That, of course, is a different problem I’ll be thinking about in the weeks to come. Circling back to the phonology example, your rules might not interact in the same ways as my rules, and the details of my organization may not work for you, but the basic principle here is: If you write a grammar for public consumption, take a step back and reread it from the perspective of someone who knows nothing about the language. Because that’s who it’s for — we are conlangers, there is no greater expert in our language than ourselves. Go back in and restructure things in a way that teaches about your language and builds upon knowledge already presented. In truth, this is just a general tip for writing, but it certainly applies for us conlangers as much as for everyone else.

I’m going to attach two different versions of the consonant processes in my Istatikii reference grammar — one in the old format with my editing notes included, and a second with the new reorganization so people can compare. It’s by no means a finished product, I still need to go through and get the text right on the individual rules. Take a look at it. If you have your own suggestions, let me know.

I’d also love to hear from conlangers on how they reorganized or revised their own grammars. I put out a question about this topic on social media and the answers I got were more about the larger scale organization of the grammar, such as what chapters you would include, or technological issues such as preference for PDF vs HTML grammars. Those are things I might cover in the future, but I really do want to ask — have you ever looked at one section, or a subsection, even — and reorganized things to make it an easier read? What did you change? How did it turn out for you?

And as always, Happy Conlanging!