530 
Marcello Lamberti 
Anthropos 82.1987 
Kapitano for Ko, Andreas Ijjigu and Kussiya 
Toloongiye for D’i, and Hayle Mikael-Kisamu 
for Mo. 
The following abbreviations are used in this 
paper: 
Af 
= Afar 
Al 
= Alaba 
Amh 
= Amharic 
Arb 
= Arbore 
Aw 
= Awqi 
Ba 
= Bayso 
Br 
= Burji 
CC 
= Central Cushitic 
Das 
= Dasenech 
Du 
= Dullay 
D’i 
= D’iraassh 3 
El 
= Elmolo 
Ga 
= Gawada 
Gi 
= Gide’o 
Go 
= Gollango 
Ha 
= Hadiya 
Kam 
= Kambatta 
Ko 
= Konso 
Ma 
= Mashile 
Mo 
= Mossiya 
Or 
= Oromo 
Rend 
= Rendille 
Sa 
= Saho 
Shi 
= Shinassha 
Sid 
= Sidamo 
So 
= Somali 
Ti 
= Tigre 
Ti.nya 
= Tigrinya 
Ts’a 
= Ts’amay 
Wo 
= Wolayta 
Ya 
= Yaaku 
Ye 
= Yemsa 
Za 
= Zaysse 
As far as the internal classification of the 
Konsoid group is concerned, Ko seems to be 
much closer to Ma than to D’i and Mo. The 
latter two are very close to one another and are 
to be regarded as two dialects of the same 
language. Ma seems to take an intermediate 
position between Ko and D’i/Mo, but it is 
probably closer to the former language than to 
the latter two. 
The transcription of the Konsoid lexemes 
reported below is a phonological one, implying 
that some phonological rules have to be taken 
into consideration and they are therefore sum 
marized below. The phoneme inventory of Ko 
is quite limited and consists only of the follow 
ing phonemes: 
lab. lab.- dent. pal. vel. uvul. glot. 
dent. 
plosives 
P 
t 
(c) k q 
’ 
ejectives 
w 
d’ 
r 
fricatives 
f 
s 
sh kh 
h 
nasals 
m 
n 
P 
liquids 
r/I 
glides 
w 
y 
b’ occurs only in a few records, for instance: 
hib’ta (lip) and it is possibly a variant of p. The 
sound c reported by Sim (1977: 10) is according 
to my material only a dialectal realization of 
the sequence ky, cf. alko-sin d’ise ca/kiya (that 
bed, lit. “this bed, which is there”), but it does 
not seem to be an autonomous phoneme. 
Through a couple of loanwords from Or a 
sound c’ (voiceless palatal ejective) has also 
been introduced into Ko, cf. mic’- (to wash), 
but it was not regarded by my informants as a 
phoneme of their language. The phoneme q can 
be freely realized either as uvular or as velar 
ejective, in both cases it is voiced, p, t, and c 
are realized as b, d, g, and j (combinatory 
variants of the phonemes p, t, k, and c) in 
voiced environment (for instance, intervocally, 
after nasals, etc.) 2 ; the voiceless pronunciation 
of the implosives implies their aspiration. The 
ejectives (including q) are on the contrary 
always realized as voiced. The fricatives are all 
voiceless; s is also voiceless, even in intervocalic 
position, but it has a free variant z (voiced 
sibilant fricative) after n, cf. -si (this) —» kipin-zi 
(this stick), kipin-si is also possible. All conso 
nants (also glides) can occur geminated. Conso 
nant gemination is phonologically relevant. All 
initial vowels are introduced by a glottal stop. 
For further details on the Ko phonology see 
Black (1973) and Sim (1977). 
The phoneme inventory of D’i largely cor 
responds to that of Ko, whereby the following 
should be taken into account: the Ko phonemes 
q, j’, kh, and p are missed and replaced by k’ 
(voiceless velar ejective), c’ (voiceless palatal 
ejective), h, and n respectively; besides this, 
D’i disposes of a phoneme t’ (voiceless dental 
2 The -t- of the suffixes -ta/-ayta/-ootaJ-teeta is 
curiously always voiceless, but not geminated.