Garre language
[3] The purported copyright violation copies text from Map of the Somali Dialects in the Somali Democratic Republic and Schizzo grammaticale del dialetto Karre di Qoryooley (Copyvios report); as such, this page has been listed on the copyright problems page.
Also, the typical Digil plural morpheme —tə has been replaced in some Garre dialects (especially in those around Baydhaba) by the common southern Somali morpheme—yaal.
(OCCL': occlusive is not exploited) This happens in particular if the two words are morphologically united, as in the periphrastic verbal forms:[10] (3) usu wa ara[g]šrə 'he was selling' (habitually).
On the other hand, spirantization of voiced stops in intervocalic positions, such as is present for example in Dabarte, does not seem to occur in Garre.
- at least of the tuuf dialect variety of Qoryooley -[11] The spirantization and the passage of b to w are instead quite frequent in the list of words Garres, e.g.
As in the other dialects of the region,[13] in final q and realized ʔ: E4: q → ʔ/__≠ (4) duq → [duʔ] __≠ "elderly In the initial position of speech, in front of vowel and sometimes realized a glottal occlusion, especially at the beginning of a sentence or in the case that the word in question is emphasized: but in most cases this exclusion glottal is not ubidable[check spelling].
a- - and encounters with other consonants are excluded (which instead are allowed with suffixes): 5) maʔag="name" naʔas = "'breast" towaʔdii = "the jackal" (towaʔ + -tii anaphoric determinant: ) ʔ also appears at the end of a word: (6) karaʔ = "Rope used to wear/put the camel's chest" suusaʔ "Milk curds of camel" kubaʔ = "Thunder" The phonemic of ʔ at the end also confirmed by the treatment of the masculine morpheme k of the article, which falls after ʔas after the other consonants.
While the phonemic stats of bilabial nasal 'm' and of a dental 'n' and also, to a limited extent, of a palatal 'p', do not create problems, different is the case of the velar 'ɳ'.
(with the same meaning ɲ can be pre-glottidalized: ʔɲaa, we also need j-a); it is natural to connect this form to the singular imperative of "eating" in Oromo: nyaaddhu.
In many cases the presence of "ɲ" in the median position is determined by the application of E7 in the encounter between a theme in the nasal and a suffix; for example, wiin + yahay→ wiiɲahay "he is great", lan + yaw ~ laɲaw "oh name!"
However, "ɲ" appears in the lexicon, albeit to a limited extent; for example in: (11) maaɲo “tare” gaɲɲuuf “saliva” mapaʔ "Milk at the first stage of acidity" Given the impossibility of deriving these ʔ cases from the application of a more general phonemic process, I consider ɲphonemic in K. The status of 'ŋ' in Garre is rather complex: it is excluded in the initial word position and appears only in very few words in the middle position, for example in haŋuur "'food" and foŋor "cracking (of vase)" .
- the latter form derived by dissimilation from + foggorr: Ali (1985: 345: n, 59} reports faqqar (glossing pot missing pieces), It's the same shape as faŋar.
In K. - at least in T. dialect - a process seems to operate which resolves a nasal + velar connection by assimilating the latter to the nasal, which becomes velarized in ŋ; this process is lexically determined, given that with other determinants (the anaphoric and the indeterminate / interrogative) the link ŋk is preserved (naturally realized [ŋ k], E7); but it could also depend on the rhythm of the sentence (more frequent with a fast pace and an informal style).
On the basis of its possibility of occurrence in the intervocalic position - where the other nasals are naturally also allowed - we recognize the phonemic status of ŋ.
In fact, phonological representations such as lanka and islaanta would be not only too distant from the phonetic forms laŋa and islaana, but the latter.
