How to describe morphosyntax?

[Ares Land]

This post only incidently has to do with Describing Morphosyntax.

My question is more about presentation of heavily inflecting languages. Basically, do you prefer when grammars:

1. dump complete inflectional paradigms on you, as huge reference tables and then discuss usage - in other words, pretty much what Mark does.
2. or for each form/prefix/suffix, discusses formation, allomorphs et al. and immediately goes into usage for that particular form - which I've seen plenty of scholarly reference grammars do?

I see advantages in both solutions; the first one seems more logical and is pretty much what I've been doing so far; on the other hand, I'm not sure it's very reader friendly (I don't mind scrolling through a few page of paradigms to find out what exactly form so-and-so does exactly but maybe others aren't that patient.)

[zompist]

The main reason I prefer giving the paradigms first is for my own reference. It's annoying to scroll through pages and pages of text just to remember how to form the 3s past form of conjugation III.

I also find that if I start to get into usage, it's hard to know when to stop. E.g., cases usually have multiple uses and drag in pretty complex constructions: passives, nominalizations, ditransitives, etc. It seems easier to give a hint on how the forms are used under Morphology, and discuss everything else in turn.

As for ease of reading, when redoing the Verdurian grammar I added a Basics section so people can get a feeling for the language before even reading about the phonology. There is certainly an argument for not starting out with morphological tables...

You can do things however you want, of course!

[Vardelm]

FWIW since I haven't produced a "complete" grammar or even conlang as of yet....

I used to like the idea of organizing a grammar into phonology, morphology, and syntax sections. "Organization" was what I prioritized. However, with my "primordial conlangs" project, I've started to mix morphology in with syntax. Like the "Basics" section of the new Verdurian grammar (BTW, I think it's a good improvement!), I now like to start with the syntax of the absolutely most simple sentences possible. From there, more & more morphology and syntax can be added in whatever fashion makes the most sense & is hopefully still very organized. These days I think that allows readers to get a handle on how your language functions, and therefor get a sense of its overall flavor, much quicker than dumping paradigm-arrhea on them (to use a recent ZZB term). That said, having a separate appendix or something with just the tables is helpful for reference.

[Kuchigakatai]

This post only incidently has to do with Describing Morphosyntax.

My question is more about presentation of heavily inflecting languages. Basically, do you prefer when grammars:

1. dump complete inflectional paradigms on you, as huge reference tables and then discuss usage - in other words, pretty much what Mark does.

• or for each form/prefix/suffix, discusses formation, allomorphs et al. and immediately goes into usage for that particular form - which I've seen plenty of scholarly reference grammars do?

I see advantages in both solutions; the first one seems more logical and is pretty much what I've been doing so far; on the other hand, I'm not sure it's very reader friendly (I don't mind scrolling through a few page of paradigms to find out what exactly form so-and-so does exactly but maybe others aren't that patient.)

I've seen some reference grammars do the first option.

• Allen and Greenough's 1903 Latin grammar starts with 150 pages of inflectional and derivational morphology, followed by 250 pages of syntax.
• Cowell's 1964 Syrian Colloquial Arabic grammar starts with 20 pages quickly introducing the basics of the morphology, proceeding with 180 pages of inflectional morphology, 85 pages of derivational morphology, and 260 pages of syntax.
• Badawi et al.'s 2002 Standard Arabic grammar has a 110-page-long chapter simply called "Forms" for inflectional morphology (and verbal derivational morphology, as it's sort of inflectional in Arabic), followed by 700 pages of syntax, finishing with a second morphology chapter on nominal/adjectival derivational morphology just 30 pages long.
• The RAE's 2010 Spanish grammar, in its shortened manual ("handbook") edition, starts with 76 pages of inflectional morphology, followed by 110 pages of derivational morphology, followed by 730 pages of syntax. (The unabridged edition of the grammar is twice as long, mostly because it has more examples.)

By the way, this is how the [ list ] tag is used for numbered items:

Code: Select all

[list=1][*]Option one.
[*]Option two.[/list]

1. Option one.
2. Option two.

[Ares Land]

By the way, this is how the [ list ] tag is used for numbered items:

Code: Select all

[list=1][*]Option one.
[*]Option two.[/list]

1. Option one.
2. Option two.

Ah, thanks. That's fixed. I knew I had messed it up, but for some reason I forgot to check the preview.

[nebula wind phone]

The way I look at it, there are two kinds of reference grammar. There are reference grammars designed for language learners, and there are reference grammars that are written by linguists for other linguists as the end result of a bunch of years of fieldwork. I associate long lists of declension tables with the first kind of grammar, and morpheme-by-morpheme discussion with the second kind -- though it's not absolute.

So I guess you could think of the choice between them as a genre decision. What kind of flavor of scholarship do you want to evoke for your reader? Classics, philology, and foreign language study? Or modern linguistic fieldwork?

[Torco]

I'm not as good at linguistics as a great many here, but for my two cents I'd think there's good reason to describe some languages with paradigmata laid out in complete tables, and others with just the rules: if there's 95 different rules going into it and all are super regular, then no, please don't give me nine pages of multidimensional tables: but if it's real irregular and super fusional, tables are indispensable.

[bbbosborne]

or for each form/prefix/suffix, discusses formation, allomorphs et al. and immediately goes into usage for that particular form - which I've seen plenty of scholarly reference grammars do?

this could just be an enormous chapter titled "morphosyntax." I'd say it's a bit more reader friendly, though I don't prefer it. it's logical since if you restrict a morphology section to only paradigms and their usages to a syntax section, then that wouldn't really be a "syntax" section. however this results in a confusing organization of the entire reference grammar.

I now like to start with the syntax of the absolutely most simple sentences possible. From there, more & more morphology and syntax can be added in whatever fashion makes the most sense & is hopefully still very organized. These days I think that allows readers to get a handle on how your language functions, and therefor get a sense of its overall flavor, much quicker than dumping paradigm-arrhea on them (to use a recent ZZB term). That said, having a separate appendix or something with just the tables is helpful for reference.

I think this is the most versatile format that can please the widest range of readers, regardless of their background or desires.

[gach]

this could just be an enormous chapter titled "morphosyntax." I'd say it's a bit more reader friendly, though I don't prefer it. it's logical since if you restrict a morphology section to only paradigms and their usages to a syntax section, then that wouldn't really be a "syntax" section. however this results in a confusing organization of the entire reference grammar.

You can, of course, always include both morphology and syntax centered descriptions in different parts of the reference grammar. That's to say that when describing case forms, for example, you'll first list all their uses in the morphology section with adequate examples and later revisit their usage under syntax whenever it's relevant for the construction under discussion. You may end up effectively writing a lot of things twice, but the end result is a much more user friendly grammar.

[Curlyjimsam]

I think it's often unhelpful to have all the morphology together followed by all the syntax. So you have paradigms for verbs, nouns, pronouns, adjectives etc. before you get around to explaining how any of them are actually used. I'd generally rather have the verbal paradigms followed immediately by discussion of what the different tenses are for (or whatever), followed by the same for nouns followed by discussion of the different cases, etc.

[Ares Land]

Thanks everyone for the advice!

So what did I choose?

I've picked option 2 - discussing usage immediately after (and sometimes before) describing the morphology. It makes for a more interesting read.
Option 1 doesn't work that well with unfamiliar morphosyntax; it works well with Indo-European languages, or Semitic languages when the reader has a general idea of how the language works beforehand; but dumping tables of patient noun-class agreement prefixes on the reader is pretty cruel