I plead not guilty. I think the augment is lurking in the shadow of a
number of other language families also. In that case the "isogloss"
becomes almost pan-IE, and what is wrong with that?
I accuse the augment of having supplied the long vowel of OCS be^ 'was,
used to be', which may be *e-H1es-t with added b- (like German bin). I
like the idea expressed by Karl Praust that the same protoform underlies
the consuetudinal past of Old Irish -bi: 'used to be' with the same
addition of b-.
I also explain the Latin imperfect subjunctive in /-se:-/ as based on the
2sg forms: The s-aorist 2sg sbj. *weg^h-se-si rhymed with *H1esi 'thou
art', so the corresponding preterite would be made to rhyme with *e-H1es >
*e:s 'thou wast'. That made a type *weg^h-se:-s and, voilà, the se:-type
was born.
I see no good way of accounting for the Tocharian imperfects with
*e:-vocalism than by analogy with augmented *e-H1es-t, i.e. /e:s-/ from
/es-/.
If it is strange to the point of being unacceptable that the preterite
just was unmarked from the very start, then the prehistory with the
augment + injunctive will have to do for Anatolian also. I see no problem
with that.
I know of no other good candidates for Greek-Armenian-IndIr innovations to
support the diagnosis of this as a tight-knit special group. Where are
they?
Based on the paradigm of 'earth' and the forms of words showing "thorn"
elsewhere, I still think Anatolian was the first group to split off (or
stay behind), but I do not find anything in the augment of relevance to
this question.
The generalized sentence openers of Hittite and Luvian, nu- and a-
respectively, the latter of which was identified by Watkins with the
augment, do indeed make very good sense as two antonyms 'now' and 'then',
being then apparently once restricted to present and preterite use,
respectively. I have the impression this idea is not too widely accepted
these days, but I like it and think it has a very good chance of being
correct. But is that the full story of what the augment *used to be*? Is
it not rather *one of several uses* a word meaning 'then' could have? In
the latter event it offers little basis for further inferences.
Jens
On Mon, 21 Apr 2003, P&G wrote:
> > None that I would really accept myself, but it is somehow inherent in the
> > story, is it not?
>
> That's what I thought you might say. To suggest that Anatolian had the
> augment raises a number of other elements of the story, not least a
> Greek-Armenian-IndIr-Anatolian isogloss. Can such an isogloss or grouping
> be supported by any other evidence, in the way the Greek-Armenian-IndIr
> group without Anatolian can? Such an isogloss would also question the
> "Anatolian splits first" idea.
>
> Or alternatively, we have an augment in generalised PIE before Anatolian
> splits off, and so no isoglossic grouping is needed. But then we have to
> say that all the other branches lost it, rather than allowing
> Greek-Armenian-IndIr to innovate - yet we know that this grouping was highly
> innovative.
>
> So on the whole, I'm not yet inclined to be persuaded that the augment is
> "inherent in the story" of Anatolian.
>
> Peter
>
>
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>