Re: Ways to get good grammar coverage

On Sun, Aug 10, 2008 at 5:19 PM, Eric Christopherson <rakko@...>wrote:

> What strategies do you folks have for getting good coverage of grammar --
> i.e. thinking up grammatical distinctions, traits, and quirks? I
> occasionally notice some interesting grammatical point in a natlang, or
> think of one myself, that I then decide to put into a conlang, but I don't
> feel like I have a really *systematic, deliberate* way of doing this.
> Related to the recent discussion of canonicity in conlangs, I would like to
> have most of my bases covered fairly early on, rather than having to add
> things piece by piece for years after solidifying a conlang (although I know
> some of that is inevitable).
>
> The best I can think of is to read a lot of grammars, other books on
> linguistics, and other things (not necessarily language-related), and make
> sure to take notes on them, but my tendency to add books and web pages to my
> to-read list faster than I actually cross things off that list makes this
> difficult. And sometimes I don't realize a feature is really interesting
> until after I'm way past it, and then I don't remember all the details (this
> recently happened to me with a grammar of Ainu I read; I emailed the author
> afterward hoping that my vague description of the thing I half-remembered
> would allow her to identify it for me, but I didn't hear back.)


Thomas Payne's _Describing Morphosyntax_ serves fairly well as a template
for things to cover (at least WRT morphosyntax; he says next to nothing
about phonology). One of my favorite grammars, Beverly Crum and Jon Dayley's
grammar of Western Shoshoni provided the template for my online description
of Shemspreg (sort of; Crum and Dayley put off phonological discussion to
the end--I start with it). It is very clearly written with a non-specialist
audience in mind, and as a result, is very easy to read.

Of course, the choice of presentation will depend on the subject matter.
Using a grammar of a polysynthetic language as a model won't help much if
your own language is more analytic and isolating.

Dirk
--
Miapimoquitch: Tcf Pt*p+++12,4(c)v(v/c) W* Mf+++h+++t*a2c*g*n4 Sf++++argh
La----c++d++600

Reply