A written grammar is really just someone making assertions about a language and backing them up with examples. For example
Examples are important - but when you're reading a grammar you take them for granted because they're just there, in print, looking very legitimate and impressive. Some of them are quite famous. Any Tibeto-Burmanist would instantly recognise "He said that he went there" as being from Hale's seminal work on the conjunct/disjunct system (the rest of you don't need to worry about it at all).
Anyway, it's easy to forget that these examples come from somewhere. Hale actually sat down at some point and asked a Kathmandu Newari speaker to say "He said that he went there". Also, when you start thinking about it, a lot of example sentences are quite strange and you wonder where they come from. Were they prepared by the researcher? Recorded on the fly during a session? Part of a narrative? How do you get such wonderful sentences as "(It is said that) I used to love my little dog a lot" (taken from Aksu-Koc & Slobin 1986)?
I'm building up a fair corpus of wacky sentences - especially as I've been turning to Nepali textbooks for inspiration. I wonder at this early point what will take my fancy later as being clear and pertinent examples of what is important in this language. I hope it's not some of the sentences from recent sessions such as:
"why do you beat the dog?"
"the earth is round"
"you never spoke English while living in Australia"
"before eating I wash my hands"
"he was cured because of good medicines"