I've finally begun reading David Weinberger's Everything is Miscellaneous (amazon.com). It's been on my to-read list for awhile but I've only recently gained access to a copy.
I've also been reading about Discourse Analysis and Discourse Grammar (the latter has more to say about what's going on at the sentence-clause level). One of the primary principles of Discourse Grammar is described by Stephen H. Levinsohn in his article The Relevance of Greek Discourse Studies to Exegesis. Here it is.
Jan Firbas, a Prague School linguist, was a pioneer in recognizing that, in the majority of sentences in a natural text, the order of non-verbal constituents tended to follow what has come to be called the "Principle of Natural Information Flow" (Comrie 1989; see Firbas 1964). According to this principle, non-verbal constituents that convey established information are placed before those that convey new or non-established information. (Levinsohn 14).
So, according to the principle of natural information flow, established information occurs first, and new information comes after.
Really, this is related to Weinberger's book (amazon.com). Trust me.
Here's Weinberger, in the chapter "The New Order of Order", subsection "Everything has its places".
The two processes by which new things are introduced into our homes are typical of how we handle information: we go through new arrivals and then we put them away. We go through the mail and file it in the special places we have for bills (the desk), cards from relatives (the refrigerator door), and junk mail (the trash). We go through bags of groceries and put the food away within minutes of bringing it into our house. We address these elements of disorder—unsorted mail in the mailbox, groceries sorted by relative weight into bags by a clerk in the store—with remarkable alacrity. (Weinberger 11)
We know how to sort our mail because it is ours. It is familiar. We go through the jumble of the mailbox, and certain things stand out: the electricity bill that needs to be paid; the envelope with the hand-written return address that looks like a card or letter (we open that one first because it is 'good mail'). The advertising circular that always comes on Tuesdays is likely plopped in the recycle bin on your way in the house because you know you already get the best price because you use your club card when you shop at that store.
Anyway, we process the information as we encounter it and filter it. We deal with the known (mail we recognize by some feature—size, color of envelope, type of postage, return sender, etc.) and move to the unknown. In so doing, some pieces of mail become prominent. We open those first because they're likely worthy of opening (except that clever junk mail in the manilla business-letter-sized envelope that looked like a check from the bank) and because we either have an inkling of what is inside or because it looks juicy but we don't really know what it is.
My flash of inspiration upon reading Weinberger after having read some papers on Discourse Analysis and Discourse Grammar — Reading involves the same process.
We do it innately with our native language because, like our mail, it is ours. We know how we've processed it in the past and we have lots of experience to filter through the new batches and determine what is promenent (the mail we open first) and what isn't (the flyers we throw in the recycle bin and the junk mail we rip up without even opening it). In our native language we naturally supply the known information and naturally note the new information and assimilate it into our further reading of the sentence/paragraph.
It is, however, much more difficult with a non-native language like Hellenistic Greek. We simply don't have enough experience "filtering the mail" to know which envelopes to open first, and which to throw out in the recycle bin. We can read the sentence/paragraph and get the gist of what's going on by assembling the words (more of a code/decode process) but we have problems picking out the salient bits because we haven't really internalized rules to tell us what is salient as we process the bag-o-words.
And this is what can be helpful in approaching the text of the NT from a Discourse Analysis and Discourse Grammar approach: we get some ideas on how to filter the mail. It gives us clues as to what to "open first" as we process a sentence or paragraph or an even larger discourse. It helps the prominent/salient bits become more evident, and this in turn helps our exegesis focus on what is necessary in order to properly handle the text.
I'm still thinking through this stuff; I'm interested to know what anyone thinks about this. Please use the comments if you have more to add to the discussion.