Consonants

The consonant chart below lists the Lusoga sounds which have been found to provide phonological contrast. The sounds in parentheses have been attested in the language, but they are very rare. They have not been included in any of the numerical counts in this paper.

While the Upper Lunyole consonant system consists of 62 consonants (Namulemu Reference Namulemu2006), Lusoga has 70. The size of this consonant inventory is to be considered as large given that the mean consonant inventory size across the world's languages is 22.7 (Maddieson Reference Maddieson, Dryer and Haspelmath2011).

Lusoga has plosives at five places of articulation with a clear phonemic distinction between a dental and an alveolar place of articulation. This is evident from near-minimal pairs like [ὲbì ὲpὲɺὲ] ‘fried cookies’ vs. [ὲbítὲɺὲkὲ] ‘parcels’ and [ músàː à] man’ vs. [ kùsàːdà] ‘to shake a liquid in a container’.

Ladefoged & Maddieson (Reference Ladefoged and Maddieson1996: 20–23) suggest that dental plosives tend to be laminal with tongue contact on both the teeth and the anterior part of the alveolar ridge, while alveolar plosives tend to be apical with tongue tip contact in the middle of the alveolar ridge. The palatograms in Figures 1–6 show that this also appears to be the case in Lusoga.

Figure 1 Palatogram of the dental plosive in [a a]. The bulk of the occlusion is against the rear of the upper teeth.

Figure 2 Palatogram of the alveolar plosive in [a t a] with well-defined contact on the alveolar ridge only.

Figure 3 Palatogram of the voiced dental plosive in [a a] showing tongue contact with the back of the upper teeth and the anterior portion of the alveolar ridge. Dental contact is asymmetrical in the midsagittal plane.

Figure 4 Palatogram of the voiced alveolar plosive in [a d a] with contact on the alveolar ridge only.

Figure 5 Palatogram of the dental nasal in [a a] showing tongue contact with the back of the upper teeth and the anterior portion of the alveolar ridge.

Figure 6 Palatogram of the alveolar nasal in [a n a] with contact on the alveolar ridge only.

At all places of articulation, the plosive pairs are distinguished in terms of voicing. This is witnessed by (near-)minimal pairs like [ kùpìkà] ‘to put pressure into something’ vs. [ kúbìkà] ‘to relay bad news’; [ὲbì ὲpὲɺὲ] ‘fried cookies’ vs. [ὲbì ὲɡὲɺὲ] ‘chains’; [ kútàɺà] ‘to get ready to fight’ vs. [ kúdàɺà] ‘to be jolly’; [ kùcá] ‘to stop; to become day’ vs. [ kùɟá] ‘to go’; [ kùkàːwà] ‘to be sour’ vs. [ kùɡàːwà] ‘to go bad (of food)’.

Voiceless plosives have a short Voice Onset Time (< 30 msec in the recordings provided with this illustration), while the voiced plosives have considerable prevoicing. The palatal plosives [c] and [ɟ] are typically realized as affricates [tʃ] and [dʒ]: [càːβàzíː^ŋɡà] (title of the Busoga king) and [ kùɟá] ‘to go’. The glottal stop only occurs as the strong onset of word-initial vowels: [ʔí àífὲ] ‘queen of Busoga’. The occurrence of a glottal stop often gives rise to a significant creaky voice quality on the preceding and following vowels especially in the context of low tones.

The labial, dental, alveolar and velar plosives also occur with labialization and these contrast with the plain plosives: there are clear minimal pairs for [b], e.g. [ὲː^m b ^wá] ‘dog’ vs. [ὲː^m bá] ‘jaws’, [t], e.g. [ kùt ^wâːɺà] ‘to take’ vs. [ kútàɺà] ‘to get ready to fight’, [d], e.g. [ ːⁿ d ^wáíɺέ] ‘diseases’ vs. [ὲːⁿ dà] ‘stomach/pregnancy’, [k], e.g. [ kùk ^wàːjà] ‘to make noise (of paper, leaves, plastic)’ vs. [ kùkàːjà] ‘to make bitter’ and [ɡ], e.g. [ mùɡ^wâːɡ^wá] ‘stupid person’ vs. [ múɡáːɡà] ‘type of tree’. There is no minimal pair for [p], [ ] and [ ] but there is evidence that they have labialized counterparts in similar phonetic environments, e.g. [p ^wìːp ^wìːp ^wì] ‘very early morning’ vs. [ kùpìkà] ‘to put pressure into something’ and [ βù ^wá] ‘intelligence’ vs. [ βú àmà] ‘dirtiness’. Furthermore, the voiceless and voiced alveolar plosives occur in phonemic contrast with the palatalized alveolar plosive, e.g. [ kùt ^já] ‘to fear’ vs. [ kùtá] ‘to put’ and [ kúɡùd ^jà] ‘to bite severely’ vs. [ kùɡùdà] ‘to gulp’.

Lusoga has 20 prenasalized plosives. Prenasalized consonants in this paper have been considered as unitary segments for at least the following six reasons: (i) the overall duration of these sounds falls well within the range of what can be expected for a single sound; (ii) the prenasalizations are always homorganic, so their phonetic realization is dependent on the place of articulation of the plosive; (iii) if syllables are taken to start with a cluster consisting of two segments, the sonority hierarchy predicts that the segment with the lowest sonority (i.e. the plosive) occurs first; (iv) Lusoga has minimal pairs contrasting prenasalized plosives with plain ones, e.g. [ kúɡûː^m bà] ‘to gather; to grow’ vs. [ kùɡùbà] ‘to become dirty’, [ mùpûːⁿ tà] ‘surveyor’ vs. [ mùtápùtà] ‘interpretor’, [ kúwàːⁿ dà] ‘to spit’ vs. [ kùwàdà] ‘to accuse falsely; try’, [ ː^ŋ kàtà] ‘head cushion’ vs. [kàtá] ‘almost’, and [ kúsíː^ŋɡà] ‘to win’ vs. [ kúsìɡà] ‘to sow’; (v) Lusoga has minimal pairs contrasting prenasalized consonants with full nasal phonemes, e.g. [ kǔː^m pà] ‘to give me’ vs. [ kùmà] ‘you light a fire’; (vi) the prenasalized plosives participate in the same processes of labialization and palatalization as the plain plosives.

A comparison with prenasalization of the plosives in the UPSID corpus (UCLA Phonological Segment Inventory Database; Maddieson Reference Maddieson1984) reveals that 53 of the 451 languages included in UPSID (11.75%) have prenasalized plosives. The total number of prenasalized plosives in Lusoga is exceptionally high in comparison to the UPSID mean (2.92).

Nasals occur at four places of articulation: labial, dental, alveolar and velar. The Lusoga variety does not have a palatal nasal which occurs in the other varieties. The labial, dental and alveolar nasals contrast with a labialized counterpart; there are minimal pairs for [m], e.g. [ m ^wíːzὲ] ‘return him/her’ vs. [ mìzέ] ‘you have swallowed’, [ ], e.g. [ὲc n ^wá] ‘bundle of firewood’ vs. [ὲcí à] ‘gecko lizard’ and [n], e.g. [ὲcìn ^wá] ‘ugly mouth’ vs. [ὲcíːná] ‘hole’. Furthermore, the labial nasal occurs in opposition with a palatalized labial nasal, e.g. [ kútὲm ^jà] ‘to blink’ vs. [ kútὲmà] ‘to cut’.

Geminate nasals also occur and these typically surface as the result of Meinhof's Law (or the Ganda Law): ‘a nasal + voiced consonant sequence becomes a geminate nasal when the next syllable also begins with a nasal’ (Hyman Reference Hyman, Nurse and Philippson2003: 52). Nouns in classes 9 and 10 (which take the prefix eN-) are especially affected: e.g. eN-[βaː^m b a] > emmamba [ mːàː^m bà] ‘meat’, eN-[j aː^ŋɡɛ]> ennhange [ ːàː^ŋɡὲ] ‘dove’, eN-[ɡɛːⁿ dɔ]> eŋŋendo [ὲŋː ːⁿ d ] ‘journeys’. Geminate nasals also surface with the first person singular morpheme (-N-), either as subject (e.g. N-[jɛːⁿ d a] > nnhenda [ ὲːn dà] ‘I want’) or as object (e.g. [a]-N-[βiː^ŋɡ] [a] > amminga [àmːìː^ŋɡà] ‘he chases me’). Nasals are the only sounds in Lusoga which occur as singletons and geminates.

Lusoga has no trills, but it has an alveolar lateral flap in words like: [ŋː^wáɺí] ‘crested crane’, [ múwùːɺù] ‘umarried man’, [ὲː^m p ^wíɡùɺú] ‘owl’ and [ ɪβàːɺὲ] ‘stone’. The speaker who has read the words for this Illustration displayed significant variability in the pronunciation of the lateral flap. Sometimes it is realized as an alveolar tap as in [ὲcìβìɾìːtì] ‘matchbox’, [ ɪβὲːɾὲ] ‘breast’ and [ ːⁿ zìɾ ] ‘soot’. In other instances, it appears as an alveolar lateral approximant: [ ɪd ^wâːlìl ] ‘hospital’, [ kùl βà] ‘to refuse’. Alveolar lateral flaps are rare in languages of the world: UPSID lists nine languages (2%) with this sound. One of the better-known examples is Japanese (Okada Reference Okada1991). In Lusoga, the alveolar lateral flap occurs contrastively with a labialized flap: [ kùɺ^wà] ‘to be late’ vs. [ kùɺà] ‘you grow up’, [ὲcíɡ ːⁿ dὲɺὲɺ^wà] ‘aim, goal’ vs. [ὲcíɡ ːⁿ dὲɺὲɺà] ‘is intended’. In addition, it contrasts with a palatalized flap, e.g. [ kùɺ^já] ‘to eat’ vs. [ kùɺà] ‘you grow up’, [ ɪɺ^jà] ‘marriage’ vs. [ ɪɺà] ‘later’.

Lusoga has fricatives at six places of articulation: the labio-dental and alveolar fricatives are represented by a voiceless and voiced member each, while the labial and velar places have a voiced fricative only. There is substantial variability in the phonetic realization of the velar fricative [ɣ], which may range from palatal/prevelar to outright uvular. Nevertheless, palatal and uvular realizations are not phonemic. The glottal fricative is very rare. The plain alveolar fricatives contrast phonemically with their labialized counterparts: [ kùs ^wáːɺá] ‘to be ashamed’ vs. [ kùsàːɺà] ‘to make a hissing sound’ and [àːⁿ z ^wíːɺὲ] ‘he/she has found me’ vs. [ ːⁿ zìɺ ] ‘soot’. Very exceptionally, labialized labio-dental fricatives are heard, but they are not contrastive: [ kùf ^wàːw ] and [ kùfàːw ] ‘to become extinct’, [ kùv ^wàːw ] and [ kùvàːw ] ‘to leave’. The labial fricative contrasts with a palatalized labial fricative, e.g. [ὲβ^jâːⁿ dà] ‘long span of time’ vs. [ὲβàːⁿ dà] ‘it hits’.

Lusoga also has five prenasalized fricatives which – just like the plosives – have been treated as unitary segments for reasons stated earlier. Two (near-)minimal pairs are: [ m ^w ː^ɱ vù] ‘ripe’ vs. [ m ^wέːvù] ‘educated’ and [ὲː^ɱ vú] ‘grey hair’ vs. [ ɪvù] ‘ash’. Only seven UPSID languages (1.55%) have prenasalized fricatives.

Lusoga has two approximants: [w] and [j].

A comparison of the Lusoga secondary articulations with languages in UPSID reveals that labialization occurs in 84 out of the 451 UPSID languages (18.63%). In this database, the number of labialized sounds varies between 0 and 29 with a mean of 4. Lusoga has 20 labialized sounds with a complete series of labialized stops and nasals, and an incomplete set in the fricatives. As far as palatalization is concerned, 35 of the UPSID languages (7.76%) have palatalized sounds with a mean of 5.2 and a range between 0 and 17. Lusoga has 13 palatalized sounds, none of which constitute a complete series.

Vowels

Lusoga has five qualitatively different vowels with a phonemic length distinction. With this system, it has the most frequent vowel system across the languages of the world. In addition, Lusoga has three rising diphthongs which are not the result of morphophonology. Although there are some examples of dipthongs in Lower Lunyole, their occurrence is rare when compared to the other Lusoga varieties.

Prosody

Preliminary research indicates that Lusoga has a reversive tone system. Bantu languages whose tone system falls in this category have inverted the tones of Proto-Bantu reconstructed forms, i.e. H is realised as L and vice versa (Marlo Reference Marlo2013). Lusoga has four tones: H, L, HL and LH. Some of the tones create lexical contrast, as illustrated in the examples below.

Transcriptions