92 holger gzella
(> /ayyā/) by /ɛ̄/, supposedly an assyrian form used for euphonic rea
sons (i.e., in order to avoid the cumbersome ending /āyayyā).59 the latter
became the dominant form in later eastern aramaic,60 but its frequency in
old aramaic cannot be confidently assessed. feminines ending in /āt/,
/īt/, and /ūt/61 generally lost the /t/ in the absolute singular and plural
but preserved the long vowel of the stem (e.g., rʿy /reʿī/ ‘pasture’ and mšqy
/mašqī/ ‘watering places’ in Kai 309:2.3).62 in the plural, however, these
long vowels would be expected to triphthongize before vocalic endings,
hence, absolute /awān/, construct /awāt/, emphatic /awātāʾ/ > /awātā/
for /āt/; /iyān/, /iyāt/, /iyātāʾ/ > /iyātā/ for /īt/; and /uwān/, /uwāt/,
/uwātāʾ/ > /uwātā/ for /ūt/. Yet the situation in old aramaic largely has
to be reconstructed in light of later evidence. Similar principles would then
apply to nouns ending in /ī/ > /ɛ̄/ (e.g., ʾrbh /ʾarbɛ̄/ ‘locust’ in Kai 222 a: 27
and participles as well as adjectives of verbal roots in /ī/ like, supposedly,
ʿnh /ʿanɛ̄/ ‘humble’ in Kai 202 a: 2): absolute and construct masc. singu
lar /ɛ̄/, emphatic /iyāʾ/ > /iyā/; absolute plural /ayn/, construct /ay/,
emphatic /ayyāʾ/ > /ayyā/; absolute fem. singular /iyā/, construct /iyat/,
emphatic /ītāʾ/ > /ītā/; absolute plural /iyān/, construct /iyāt/, emphatic
/iyātāʾ/ > /iyātā/.63 these should be distinguished from the small group
of nouns in /y/ (like ṣby /θ̣abī/ < */θ̣aby/ ‘gazelle’ in Kai 222 a: 33).64
this difference seems to be related to the dual nature of the glide /y/,
which combines properties of vowels and consonants.
pronominal suffixes can be attached to prepositions and the construct
state of nouns in order to render a genitive relationship with a pronomi
nal possessor. depending on whether the construct form of a noun ends
59 Kaufman 1974: 127f and Beyer 2004: 50.
60 gzella 2008: 100.
61 Beyer 1984: 454–456.
62 cf. Kaufman 1982: 164. Beyer 1984: 27, by contrast, seems to suppose that {y} here
atypically renders /ɛ̄/, but the former explanation is easier.
63 Beyer 1984: 456–458.
64 Sometimes nouns ending in /y/ can merge with those in /ī/, hence ṣdh ‘owl’
(Kai 222 a: 33), which is presumably to be read /ṣadɛ̄/ although it belongs to the same
class as /θ̣aby/. the same then applies to gdh /gadɛ̄/ ‘goat’ (Kai 223 a: 2) from /gady/.
comparative evidence (e.g., the hebrew plural kēlīm ‘vessels’ from /kily/ or /kaly/ as
opposed to the expected gḏāyīm ‘goats’) shows that such crossovers are less surprising
than degen 1969: 27 believes. an influence of the large group in /ī/ on the few similar
nouns in /y/ is especially likely. there is no need to assume (with, e.g., fitzmyer 21995:
90 and 126) an ending /ay/ for nouns written with {y} and a change of /ay/ to /ē/ for
those written with {h}. Monophthongization of */ay/ occurs by closing the mouth and
thus invariably leads to a “closed” ē, whereas {h} as a vowel letter is only attested for an
“open” ɛ̄.