[Lingtyp] Fwd: syntactic construction formula

[Christian Lehmann]

Am 20.12.20 um 15:11 schrieb Elena Skribnik:
> "hier wohnt Maria/Maria wohnt hier": there is a pronounced IS 
> difference. Do you have a separate layer like "IS" in any terms, topic 
> - focus, old - new, whatever? For languages like Russian or German 
> with IS-governed constituent order it is necessary; I'm not sure about 
> /John suddenly screamed / John screamed suddenly./////What about Cabecar?

Two considerations come to mind:

  * One thing is the sheer possibility of having two or more orderings
    for the components (or just one component) of a given syntagma
    (construction, if you will). It just means that in texts, such
    orderings may be found. Another thing is the function of such
    orderings. Barring the extreme of free variation - which everybody
    agrees is just the result of the linguist throwing in the towel -,
    we assume that the variation makes some semantic or functional
    difference. If it is information structure, I would hope that the
    same operations of information structure apply not only to /hier
    wohnt Maria/ ~ /Maria wohnt hier/, or even only to declarative
    clauses based on a two-place predicate, but more widely to
    constructions of different sorts. Consequently, my construction
    formulas contain no layer of information structure. But may grammar
    will treat it in a separate chapter.
  * If we view grammar writing as a process, then understanding of the
    workings of the grammar steadily grows over time (so we hope). At an
    earlier stage, we come across some variation, which we note and
    represent. At a later stage, we find out what the variation is all
    about. It is well possible - and relevant cases are documented in
    the history of our discipline - that the later stage comes after
    publication of the grammar. In this sense, I will, at any rate, draw
    that diagram representing the variation, hoping to reach stage 2
    before publication of the grammar.

Cheers,

Christian

-- 

Prof. em. Dr. Christian Lehmann
Rudolfstr. 4
99092 Erfurt
Deutschland

Tel.: 	+49/361/2113417
E-Post: 	christianw_lehmann at arcor.de
Web: 	https://www.christianlehmann.eu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20201220/8f72a3f4/attachment.htm>