<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">
<title></title>
</head>
<body text="#000000" bgcolor="#ffffff">
<p class="MsoNormal">Dear Florian,</p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">I am surprised that so far nobody has mentioned
the case of
Ethiopian Semitic and Cushitic languages. </p>
<p class="MsoNormal">In Tigrinya, only the old personal pronouns of the
first
persons (singular and plural) have been retained. 2nd and 3rd person
pronouns
were replaced by an element ness(e)- in the singular and nessat- in the
plural,
to which the possessive suffixes are attached, thus<br>
2sm nesse-xa, 2sf nesse-xi 'thou', 3sm ness-u 'he', 3sf
ness-a 'she', 2pm nessat-ku 'you', etc. Etymologically, ness(e)- is
*nafs-
'soul', nessat- its plural *nafsaat- 'souls'. The original meaning of
the new
pronouns is thus 'your soul', 'his soul', 'your souls', etc.</p>
<p class="MsoNormal">The Northern Cushitic language Beja follows a
similar
pattern: old Afroasiatic pronouns of the 1st person singular and plural
are
retained ('ane 'I', hinin 'we'), while the rest is replaced by a
noun-like
element with possessive suffixes, here bar- for masculine, bart- for
feminine,
and baraa- for plural. Strangely enough, plural forms take singular
suffixes,
e.g. barúu(-s) 'his "bar"' = 'he', baráa(-s) 'his "bars"' =
'they'. I have not been able to figure out what the original meaning of
*bar-
was. The relevant body part terms are all different. But in view of
grammaticalized politeness forms such as Spanish Usted or Egyptian
Arabic
HaDritak ("your presence") it is perhaps worth while looking at other
semantic areas as well. There is also a demonstrative element ba-, but
I don't
know if this is related.</p>
<p class="MsoNormal">Incidentally, it occurs to me that the Ancient
Egyptian
paradigm displays a very similar pattern: there is 'nk 'I', 'nn 'we',
both
looking like old Afroasiatic forms. However, the remainder of the
independent
pronouns are all based on an element nt- to which the possessive
suffixes are
attached: nt-k 'thou masculine', nt-f 'he, nt-s 'she', exactly as in
Beja
("your nt", "his nt", etc.) Again, I have no idea what nt-
might have meant.<o:p> </o:p></p>
<p class="MsoNormal">In Amharic, the old Semitic pronoun of the 3rd
person was
replaced with an expression using the noun 'head', again with the
possessive
suffixes attached: originally *re'es-u 'his head', *re'es-wa 'her
head', in
Modern Amharic phonologically eroded to ersu ~ essu 'he', erswa ~ esswa
'she'.<o:p> </o:p></p>
<p class="MsoNormal">Finally, let me mention the case of the South
Ethiopic Soddo
language, a member of the Gurage group, which has replaced the
independent
pronouns of the 1st and 2nd persons singular and 2nd person plural by
an
element (ä)d- + possessive suffixes. It is possible that (ä)d- comes
from the
word for 'hand' (Proto-Semitic *yad-), though this remains a matter of
dispute
(cf. Hetzron 1977, The Gunnän-Gurage Languages, p.58).<o:p> </o:p></p>
<p class="MsoNormal">The existence of a "pronoun replacement area" in
this region is all the more remarkable as otherwise Afroasiatic
languages are
comparably conservative with regard to personal pronouns; changes
usually come
from analogical processes triggered by patterns from "inside" the
system (such as, for example, adjustment to pronominal suffixes on the
verb or
the addition of a possessive suffix to an already existing independent
pronoun
to increase distinctiveness). I am not aware of any case of borrowing
of
pronominal forms from a non-Afroasiatic contact language.</p>
<p class="MsoNormal">Best wishes,</p>
<p class="MsoNormal"><o:p></o:p></p>
<p class="MsoNormal">Hans-Jürgen Sasse</p>
</body>
</html>