86 holger gzella
(Kai 214: 1; Kuttamuwa l. 5), presumably a defective spelling for /ðenā/,
and a byform znn /ðenān/ (Kuttamuwa l. 9).37 defective spellings like
zn and ʾl, as well as the much later byform dn in aramaic texts from
Qumran, could entail that the stress was originally on the first syllable in
these forms.38 proper distal counterparts (‘that, those’) are only attested
for later periods,39 whereas old aramaic, like phoenician, hebrew, and
much of official aramaic, used the thirdperson independent pronouns
for this purpose.
the common aramaic relative marker zy /ðī/ (later > /dī/, eventually
spelled dy), by contrast, has evolved from a determinativerelative pro
noun and preserves a fossilized genitive that no longer inflects.40 it can
introduce relative (Kai 214: 1) and, especially in combination with prepo
sitions, other subordinate clauses (e.g., it marks a causal relationship in
Kai 201: 4) or act as an analytical genitive marker. the latter function
is but rarely attested in the oldest aramaic texts, uncontested examples
being dmwtʾ zy hdysʿy ‘the image of haddayisiʿ ʼ (Kai 309: 1) and mʾnyʾ
zy bt hdd ‘the vessels belonging to the house of hadad’ (Kai 309: 16–17).
Both occur in the tell fekheriye inscription and have perhaps been
reinforced by akkadian ša due to extensive bilingualism in that area.41
this particle can also resume another antecedent, as in mlk gzn wzy skn
wzy ʾzrn ‘the king of gozan and the one of Sikan and the one of azran’
(Kai 309: 13).42 in subsequent stages of aramaic, however, analytical geni
tive marking increasingly competes with the construct state, especially
for further qualifications such as materials. together with the preposition
/la/ and a suffix, zy can form an independent possessive pronoun (zy ly
‘what belongs to me’, Kai 224: 20).
despite the otherwise genderbased nominal and pronominal system,
the two interrogatives mn /man/ ‘who?’ and mh /mā/ ‘what?’ distinguish
37 the form znn, which is now first attested in the new Kuttamuwa stele, has a later
parallel in dnn, which occurs frequently in the aramaic legal papyri from naḥal Ḥever
(Beyer 2004: 380). however, there is a tendency in aramaic to expand final long vowels in
certain forms by /n/ (Beyer 1984: 149), so one does not necessarily have to posit a direct
relationship between both forms.
38 cf. Beyer 1984: 555f.
39 evidence from official aramaic is conveniently summarized by folmer 2011: 142.
40 the alleged fem. singular variant ʾzh ‘which’ in Samʾalian (Kai 215: 2) is debated and
may be a ghost form (see hoftijzer – Jongeling 1995: 61 for the reading ʾlh ‘conspiracy’[?]).
41 Kaufman 1982: 151. the title šbṭ zy nrgl ‘the plague of nergal’ in the same text
(Kai 309: 23) may be calqued directly from akkadian and bears less weight for determin
ing the function of zy in aramaic.
42 for examples from later periods, see Beyer 1984: 549.