clitic and orthotonic personal pronouns

[Bjoern Wiemer]

I have a question to those who are familiar with the diachronic development
of personal and reflexive pronouns in Slavonic languages:
        For the historically tesfifiable development of pronouns in the IE
languages it is normally assumed that the clitic forms arose out of "long"
ones, i.e. those that are put under stress, after prepositions,
sentence-initially etc.. E.g. Latin 'se' became a clitic, and only in some
cases it was reinforced, e.g. to 'sesez' in Surselvan (Raetoromanic). In
other cases a form that could be used both clitically or orthotonically, as
e.g. the ancestor of Scandinavian 'sig' (or 'sik'?), developed into a
postposed clitic from the unstressed variant. This, in turn, got wholly
affixized to '-sk' (and further '-st > -s') in older stages of the
Scandinavian languages. On the one hand, this can be regarded as semantic
bleaching (the marker of reflexivity proper turns into a rather universal
marker of "middle voice", or, better, all kinds of lexicalized recessive
diatheses). On the other hand, this evolution seems to correspond to the
usual picture of deteriorization of linguistic forms, as it is discussed,
e.g., in the literature on grammaticalization.
        But if we now turn to Slavonic material, we are confronted with
another picture, which in some respects is paradoxical compared with what we
know about other branches of IE languages (see above). As widely known, the
Russian reflexive postfix '-sja/-s'' derives from a properly reflexive
pronoun. But as far as I know it doesn't derive directly or indirectly from
the "full form" 'sebe/seb'a', but from 'sa~', which, in turn, was NOT a
simple clitic derived from 'sebe/seb'a', but existed as a clitic as well as
an orthotonic pronoun. According to M.A.Gadolina (1963): Istorija form
lichnyx i vozvratnogo mestoimenij v slavjanskix jazykax (Moskva: Izd-vo AN
SSSR), the ACC-form of the reflexive pronoun, as well as the personal
pronouns of the first and second person singular, were supplanted by 'sebja'
('menja', 'tebja') only on the onset of historically documented times, this
last form "invading" from the genetive. Thus, we would have a situation as
in early Germanic, which Kemmer  (1993): The Middle, Voice (Benjamins),
characterizes as a "one-form-language" (i.e. both the reflexive and the
middle coincide); compare Germanic 'sik' and ancient Slavonic 'sa~'.
        The difference, however, is that the Germanic morpheme either
disappeared altogether (Old Saxon, Old English) or was from the ACC
transferred to the dative (German after 1500). Nothing similar happened in
the historical stages of modern Slavonic languages. Instead, due to Gadolina
(ibd.), the dative pronouns from the beginning ("iskoni") showed a short and
a long form ('si' vs. 'sebe/sobe'), which were neatly distributed as clitic
vs. orthotonic.
        From a more general typological viewpoint such a case seems quite
improbable: if, after Kemmer (ibd.), we call "direct reflexive contexts"
those which refer to the second participant (normally encoded as direct
object in the ACC) and "indirect reflexive contexts" those in which
coreference is established between the first and the third (or even a
peripheral) participant, we see that direct reflexive contexts are more
central in that there are usually more forms to be exploited differently in
discourse. (For this reason Middle High German 'sich' was extended from the
ACC to the DAT and not the other way round.)
        I hope my somewhat lenghty reasoning demonstrates the problem. In
connection with this I should like to put forward the following questions:

1. Gadolina's above cited book is of a rather old date. Can anyone advise me
newer investigations relevant to the historical development of clitic /
orthotonic / affixal forms of pronouns in Slavonic languages? I would
greatly appreciate hints at work in a typological frame and/or with some
concern for the IE background.

2. I have the impression that the language stages Gadolina discusses are,
maybe, not ancient enough to render wholly reliable data suited for
comparison with Romance and/or Germanic.

3. Does anyone know (literature on) what was the development in Baltic
languages? Can we reconstruct the way from pronouns to the nowadays affix
'-si-' in Lithuanian and the history of the regular reflexive pronouns
'save~s (GEN)/ sau (DAT)/ save (ACC)' and their functional distribution?
And: did the affix '-si-' (or its ancestor) at any time distinguish case
(ACC vs. DAT)? Was it at any time related to the ancient Slavonic clitic
'si' (see above)?

4. Finally, can anybody advise literature on the (presumable) reasons for
which the clitic forms in Eastern Slavonic disappeared altogether, whereas
in Western Slavonic they are wholly "alive" until today? There must be some
more systematic reasons exceeding just the system of pronouns.

        I will summarize replies if there will be sufficient material to be
summarized. Anyway, for hints and discussion of the question touched on
above I should be grateful.
Bjoern Wiemer.

#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#+#

Bjoern Wiemer
Universitaet Konstanz
Philosophische Fakultaet / FG Sprachwissenschaft - Slavistik
Postfach 55 60 - D 179
D- 78457 Konstanz

e-mail: Bjoern.Wiemer at uni-konstanz.de
tel.: 07531 / 88- 2582
fax:  07531 / 88- 4007
                - 2741

*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*^*