INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
Optimality Theory: Application of Dahl’s law in Kindia  
Grace W. Muriithi  
University of Botswana  
DOI: https://doi.org/10.47772/IJRISS.2025.91200003  
Received: 09 December 2025; Accepted: 18 December 2025; Published: 30 December 2025  
ABSTRACT  
This paper looks at Dahl’s law, a voice dissimilation process prevalent in Bantu languages, and how it is  
manifested in Kindia, a Kikuyu dialect. Dahl’s law is realized in varied ways in Zone E languages at dialectal  
level hence making it a significant divider amongst dialects of the same language (Kanana, 2011). Dialects tend  
to vary at phonological and morphological levels. While these variations can be identified, discussions of their  
viability can be discussed through Universal Grammar. Optimality Theory postulated by Prince and Smolensky  
(Prince & Smolensky 2004) can be used to establish the constraints that may be at play in determining the  
realization of Dahl’s law in Kindia. Some Zone E Bantu languages allow for the dissimilation of the voiceless  
stops such as /t/ and /k/ to voiceless fricatives, /ð/ and /ɣ/ respectively depending on the verb stem (Elwell, 2010).  
However, Kindia shows marked difference in that the voiceless velar stop /k/ in the verb initial is retained  
regardless of the verb stem. This markedness, allowing the retention of the voiceless velar stop /k/, is one of the  
dominant features that distinguish Kindia from other Kikuyu dialects. Using Optimality Theory, this study seeks  
to demonstrate the diverse and allowable manifestations of the Dahl’s law in Kindia different from other Kikuyu  
dialects. In the other Kikuyu dialects (Nyeri, Murang’a and Kiambu), the voiceless velar stop /k/ is realized as  
voiceless velar fricative /ɣ/ if the verb stem has a vowel initial or has the voiced velar fricative, /ð/. Kindia ranks  
constraints differently than other Kikuyu dialects.  
Keywords: Dahl’s law, Optimality Theory, voice dissimilation, Bantu, Kindia  
INTRODUCTION  
Kindia is a Kikuyu [ISO: kik] dialect spoken in Kirinyaga County (ISO 639-2). Located in Central Kenya,  
Kirinyaga County has a population of over 600000 people (Kenya National Bureau of Statistics, 2019). Kindia  
is also known as ‘Gi-Kirinyaga’ as it is spoken by people from Kirinyaga County. The prefix Ki-Gi- is a Class  
7 noun class marker as categorized in Bantu languages (Kutik, 1983). Kindia is classified as a dialect of Kikuyu  
spoken in Southern Mount Kenya region (Iribemwangi, 2012). Kikuyu has other dialects spoken in Nyeri,  
Kiambu, and Murang’a counties in Central Kenya as well as other areas in central Rift Valley. Maho (2008) lists  
six dialects of Kikuyu, namely, North Kikuyu, South Kikuyu, Kindia, Gigichugu, Murang’a and Mathira.  
Iribemwangi (2012) affirms that these Kikuyu dialects can be linguistically discerned. Thus Kindia is also  
distinguishable from other Kikuyu dialects in pronunciation, intonation, phonemic system and in select  
vocabulary as is common with dialects.  
Key phonemic differences between Kindia and other Kikuyu dialects  
Two key phonemic aspects that distinguish Kindia dialect from other Kikuyu dialects is the non-existence of the  
glottal fricative /h/. In its place, the glottal fricative /h/is either omitted or realized as the bilabial fricative /β/.  
This bilabial fricative /β/ is conspicuously absent from the other Kikuyu dialects as illustrated below:  
Other Kikuyu dialects  
Kindia  
aβa  
Gloss  
haha  
harea  
ehuta  
‘here’  
βarea  
eβuta  
‘there’  
‘move away’  
Page 25  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
huura  
kohoha  
hurutera  
hiɣo  
βuura  
‘wipe the surface’  
‘to wilt away’  
‘fan me’  
koβoβa  
mburutera  
mbiɣo  
‘a kidney’  
kehuruta  
hio  
kemburuta  
mbio  
‘butterfly’  
‘hot item’  
ihoha  
iβoβa  
‘a boil’  
From the above examples, the glottal fricative /h/ found in other Kikuyu dialects is firstly conspicuously absent  
in Kindia. Secondly, it is either realized as the voiced bilabial fricative /β/ or as a pre-nasalized bilabial stop  
/mb/. This prevalence of the glottal fricative /h/ makes the Kindia speakers refer to the other Kikuyu dialects as  
‘ke-haha’ speakers, that is, those who use the glottal fricative /h/ sound as illustrated in the words above.  
Optimality Theory  
Optimality Theory, hereafter referred to as OT, is a development of Generative Grammar focusing on formal  
description of and quest for universal principles (Kager, 2007). Having noted that Universal Grammar contains  
a set of violable constraints, it therefore means that surface forms of a language reflect resolutions of conflicts  
between competing demands (Archangeli, 1999). The surface form is optimal if it incurs the least serious  
violations of a set of constraints, taking into account their hierarchical ranking in each language. OT identifies  
the constraints in a language and assigns them roles in a language (Archangeli, 1999). These markedness and  
faithfulness constraints are at play as they govern the output. On the one hand, faithfulness constraints determine  
the preservations of the input in the output by asking for the exact preservation of the input in the output along  
various dimensions. On the other hand, markedness constraints govern the well-formedness of the output. Thus,  
OT grammar employs this input-output mechanism and that each input has only one output Kager (2007). This  
input-output relation thereby generates a set of candidates which are then evaluated to determine the one that  
best satisfies the relevant constraints, and ranked in order of importance. Therefore,  
Input →GEN→ Candidate set →EVAL → Optimal candidate  
Kager (2007) provides the following mapping of input to output in OT grammar:  
Table 1: Mapping of input to output in OT grammar  
C1  
»
C2  
»
Cn  
Candidate a →  
Candidate b →  
Candidate c →  
Candidate d →  
Candidate ... →  
→
→
Input →  
→
→
→
→
→
Output  
Source: Kager, 2007 pg. 8  
Page 26  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
Each language prioritizes some constraints over others based on domination; the higher-ranked constraint has  
priority over the lower-ranked one regardless of the violations thus resolving the conflicts by domination (Kager,  
2007).The ranking of constraints in languages gives rise to systematic variation between languages and even  
between dialects as shall be indicated in this paper.  
Dahl’s Law in Zone E languages  
Dahl’s law is a voicing dissimilation process where Bantu voiceless stops become voiced in certain  
circumstances as observed in Zone E languages1. This dissimilation process applies in CVCV sequences where  
both consonants (C) are voiceless and thus turns the first consonant into its voiced counterpart (Werner, 2011).  
Voiceless stops, such as /p t k/, become voiced (/b d ɣ/) when immediately followed by a syllable with another  
voiceless stop.  
Elwell (2010) illustrates that in Ekegusii, a Zone E language, the velar stops dissimilate with the voicing of the  
following onset.  
Dahl’s law in Ekegusii  
Word  
Gloss  
oɣotama  
okoraaɣera  
‘to flee’  
‘to eat’  
Using OT, voicing is the markedness constraints in Ekegusii in the CVCV sequence. Thus, the initial voiceless  
stop is realized as a voiced fricative if followed by voiceless stop. In cases where the initial stop is followed by  
a lateral, the stop after the lateral becomes a voiced fricative /ɣ/. The faithfulness constraint on the other hand  
ensures that the CVCV syllable structure is preserved. No segments are deleted in both examples, making MAX-  
C a dominant constraint. Therefore, a tableau for these realizations can be mapped as follows:  
Table 2: okotama/ → [oɣotama]  
Candidate *VOICELESSSTOP+STOP IDENT(CVCV) IDENT (voice) Optimal  
0
0
0
0
1
0
1
√
oɣotama  
okotama  
ogotama  
1!  
0
Table 3: /okoraaɣera/ → [okoraaɣera]  
Candidate *VOICLESSSTOP+STOP IDENT(CVCV) IDENT (voice) Optimal  
okoraaɣera 0  
0
0
1
0
√
okoraakera 1!  
In both tableaux, candidates (b) and (c) violate the markedness constraints (*VOICELESSSTOP+STOP) where  
obstruents are in sequence, and thus are deemed least optimal outputs. Penalization occurs where voiced stops  
occur in onset positions and where voiceless stops occur in sequence. Since Candidate (a) has the least violations,  
it makes it the optimal output. IDENT (voice) and IDENT(CVCV) preserve the input voicing and the syllable  
structure. In Tableau 2, markedness constraint IDENT (voice) because the second stop after lateral becomes /ɣ/,  
satisfying markedness.  
1
Zone E are Bantu languages spoken in Kenya.  
Page 27  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
Dahl’s Law in Tiania  
Dahl’s law does not just include the dissimilation process; at times it may involve deletion of sound  
correspondences. Hyman and Mbuui (2022) illustrate /g-/ deletion in Tiania, a Meru dialect as shown below:  
Underlying form  
ko-goro  
Tiania  
ko-oro  
ko-rua  
oti-a  
Meru  
Gloss  
ko-goro  
ko-ruga  
otig-a  
‘leg’  
ko-ruga  
‘to cook’  
‘to leave behind’  
tig-a  
These examples demonstrate that g-deletion occurs due to the voicing dissimilation triggered by Dahl’s law  
making it a one of the distinguishing factor of Tiania as a dialect against other Meru dialects. This means that in  
Tiania, Dahl’s law activates the markedness constraint through the g-deletion thus ensuring that the output is  
well-formed. Faithfulness constraints ensure that much of the input is preserved in the output. The g-deletion is  
ranked high in Tiania making it a distinguishing factor from the other Meru dialects. Therefore, Table 4  
illustrates the intervocalic /g/ deletion prevalent in Tiania:  
Table 4: /ko-goro/ → [ko-oro]  
/ko-ruga/ → [ko-rua]  
/tig-a/ → [oti-a]  
Candidate  
*G/V_V  
MAX  
Candidate  
*G/V_V  
MAX  
Candidate  
ONSET  
DEP-V  
ko-goro  
*!  
ko-ruga  
*!  
tig-a  
*!  
*
*
*
☞ ko-oro  
☞ ko-rua  
☞ oti-a  
While Meru prioritizes markedness constraints over faithfulness, Tiania prioritizes faithfulness constraints over  
markedness. The intervocalic /g/ is deleted in Tiania while it is retained in Meru.  
Dahl’s Law in Kindia  
Miti (2006) demonstrates that in Kikuyu the voiceless stop /k-/ is realized as velar fricative /ɣ-/ when it occurs  
before /t, ð, ʃ, k/. Using the same verbs that Miti (2006) uses, the realization of those verbs in Kindia is illustrated  
below:  
Underlying form  
ko-teɣa  
Other Kikuyu dialects  
ɣoteɣa  
Kindia  
ɣoteɣa  
Gloss  
‘to trap’  
‘to read’  
‘to return’  
‘to milk’  
ko-ðoma  
ɣoðoma  
koðoma  
ɣosoka  
ɣokama  
ɣo-ʃoka  
ɣoʃoka  
ko-kama  
ɣokama  
Dahl’s law in Kikuyu involves the alternation between the voiceless velar stop /k-/ and velar fricative /ɣ-/. If a  
prefix which is to be attached to a stem has the velar stop /k-/, and the following syllable initial segment in the  
stem is /t, ʃ, k/, then the voiceless velar stop /k-/ of the prefix is realized as a velar fricative /ɣ-/ (Bennett, et al.,  
1985).  
Therefore, Dahl’s law in Kindia can be illustrated below as:  
/k/ → /ɣ/_ /t, s, k/  
Page 28  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
Kindia realizations have the /k-/, a voiceless consonant, occurring before voiced segments of the stem while /ɣ-  
/, a voiced consonant occurs before stems starting with a voiceless segment. Furthermore, the prefix for class 7  
nouns is /ke-/ɣe/ depending on the noun stem due to Dahl’s law.  
Class 7  
Gloss  
Class 8  
i-ɣwa  
Gloss  
ke-ɣwa  
'sugarcane'  
'head'  
‘sugarcanes’  
‘heads’  
ke-oɲgo  
ɣe-tongoro  
ɣe-sisio  
ke-ondʒe  
ke-ðayo  
ke-ɣota  
si-oɲgo  
i-tongoro  
i-sisio  
‘onion’  
‘mirror’  
'cripple'  
'fool'  
'onions'  
‘mirrors’  
‘cripples’  
‘fools’  
si-ondʒe  
i-ðayo  
'lazy person'  
i-ɣota  
‘lazy people’  
These examples illustrate the alternates of initial /k-/ and /g-/ depends on the initial sound of the stem. If it is  
voiced, then the voiceless velar stop /k/ is applied, if voiceless, then the velar fricative /ɣ-/ is realized. Whereas  
the above realizations of initial /k-/ and /ɣ-/ cut across all Kikuyu dialects, there are differences in regard to  
dental fricatives /ð/ and /ʃ/. The velar stop /k-/, in the prefix dissimilates in voicing and is realized as a velar  
fricative, /ɣ-/. Notably, unlike Kikuyu, Kindia allows for /k-/ and /ð-/ sequence as illustrated in further in the  
examples below:  
/k/ and /ð/ sequences  
Underlying form  
/ko-ðeka/  
Kindia  
koðeka  
koðoma  
ɣokeða  
keðandoko  
keðori  
Other Kikuyu dialects  
ɣoðeka  
Gloss  
‘to laugh’  
‘to read’  
‘to harvest’  
‘a box’  
/ko-ðoma/  
ɣoðoma  
/ko-keða/  
koɣeða  
/ke-ðandoko/  
/ke-ðori/  
ɣeðandoko  
ɣeðori  
‘chest’  
These sequences are characteristic of Kindia, where the occurrence of the voiced dental fricative –ð- stem  
initially does not trigger Dahl’s law, an indication that Kindia has lost this effect compared to Kikuyu, its mother  
language. This prevalent phonological process triggered by Dahl’s law determines the morpheme alternations of  
simultaneous morpheme markers, -ke- / -ɣe-.  
Markedness and Faithfulness Constraints in Kindia  
OT categorizes constraints as either markedness constraints or faithfulness constraints. The realization of Dahl’s  
law in Kindia identifies these constraints at play and their ranking. Evidently, Kindia alternates between  
maintaining voicelessness and introducing voicing depending on the phonological environment. Agreement in  
voice, is the marked constraint as it governs the well-formedness of the output, therefore the adjacent consonants  
must agree in voicing. Consequently, Kindia penalizes sequences of two adjacent voiceless obstruents, thus one  
has to become voiced. Moreover, to preserve faithfulness in the output, the segments must preserve their voicing  
specification in one segment. Consonants are neither deleted nor epenthesized. This implies that in Kindia, the  
IDENT-IO (voice) constraint is ranked higher than the AGREE(Voice) as illustrated below.  
Page 29  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
Constraint set and ranking in Kindia  
Using OT, it can be deduced that the following constraints are predominant in Kindia:  
MAX-C: no deletion of consonants  
IDENT-IO (voice): preserve voice in input  
AGREE (voice): Adjacent consonants agree in voicing  
*VV less Seq: Penalize sequences of adjacent voiceless obstruents  
These constraints can be ranked as follows:  
Ranking: MAX-C, IDENT-IO (voice) ≫ AGREE (voice), *VVlessSeq  
Tables 5, 6 and 7 demonstrate these constraints, their ranking and the rationale for the determination of the  
optimal output.  
Table 5: Input /ko-ðoma/  
Candidate MAX-C IDENT-IO (voice) AGREE (voice) *VVless Seq Optimal  
koðoma  
0
0
0
1
1
0
0
0
√
ɣoðoma  
Rationale: Faithfulness to voicing is high-ranked; no pressure to change /k-/ before a voiced onset.  
Outcome: koðoma is optimal because violating lower-ranked AGREE (voice) is tolerated when IDENT-IO  
(voice) is satisfied as stem begins with voiced /ð/.  
Table 6: Input /ko-teɣa/  
Candidate MAX-C IDENT-IO (voice) AGREE (voice) *VV less Seq Optimal  
ɣoteɣa  
0
0
1
0
0
1
0
1
√
koteɣa  
Rationale: Adjacent voiceless obstruents are disfavored; AGREE (voice) and *VVlessSeq push for [ɣo-].  
With IDENT-IO (voice) lower-ranked than these markedness pressures in this environment.  
Outcome: ɣoteɣa is the optimal output; the single violation of IDENT-IO(voice) is preferred over dual  
markedness violations (AGREE + *VVlessSeq) as the stem begins with voiceless /t/.  
Table 7: Input /kokama/  
Candidate MAX-C IDENT-IO (voice) AGREE (voice) *VVlessSeq Optimal  
ɣokama  
0
0
1
0
0
1
0
1
√
kokama  
Rationale: The sequence of voiceless obstruents is penalized, favoring [ɣo-] because stem begins with voiceless  
/k/.  
Page 30  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
Outcome: ɣokama is the optimal output because minimizing markedness violations outweigh faithfulness  
violations.  
This means that before voiced onsets such as /ð/ the voiceless obstruent /k/ is retained. However, before voiceless  
onsets such as /t, k/, the optimal output is / ɣ/. MAX-C is ranked high in Kindia as no consonants are deleted.  
CONCLUSION  
As illustrated in the discussions above, the diachronic process of Dahl’s law comes into play when determining  
the morpheme alternations of the phonological /k/ and /ɣ/ segments depending on the initial segment of the stem  
due to the dissimilation process (Davy, J. I. M & Nurse, D., (1982). However, Kindia deviates from other Kikuyu  
dialects in that Dahl’s law operates with any voiceless obstruent and not just with plosives. This feature  
distinguishes Kindia from other Kikuyu dialects specifically the realizations of the /k, ɣ/ before /ð/ and /s/  
fricatives as summarized in Table 8 below.  
Table 8: /ko-ðoma/  
Candidate  
a. [ɣo-ðoma]  
b. [ko-ðoma] *!  
c. [go-ðoma]  
d. [o-ðoma]  
e. [toma]  
AGREE IDENT (stem) IDENT (prefix) *[+voice] [+fric] MAX DEP  
*
*
*!  
*!  
*!  
*!  
*
*
*
Table 9: /ko-soka/  
Candidate  
a. [ɣo-soka]  
b. [ko-soka] *!  
c. [go-soka]  
d. [o-soka]  
AGREE IDENT (stem) IDENT (prefix) *[+voice] [+fric] MAX DEP  
*
*
*!  
*!  
*!  
*!  
e. [goka]  
*
*
*
This paper demonstrates that Dahl’s law not only influences Zone E languages at language level but also at  
dialectal level. The morpheme alternations found in Kindia are distinct enough to set it apart from other Kikuyu  
dialects. Kindia dialect applies Dahl’s law more broadly to voiceless obstruents unlike other Kikuyu dialects  
where voicing dissimilation primarily targets plosives. This results in unique realizations of /k, ɣ/ before  
fricatives /ð, s/ and a tolerance fro /k-ð/ which other Kikuyu dialects avoid.  
Using OT it is evident that faithfulness constraints of voice (IDENT-IO (voice) and consonant retention (MAX-  
) are ranked higher than markedness constraints allowing retention of voicing in specific environments while  
still penalizing adjacent voiceless sequences. Thus /k/ is preserved before voiced onsets but alternates to /ɣ/  
before voiceless ones.  
Page 31  
www.rsisinternational.org  
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)  
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025  
These findings underscore the role of constraint interaction in determining the variation of dialects and highlight  
Kindia’s phonological autonomy from other Kikuyu dialects. Future research could explore whether similar  
constraint hierarchies operate in other Bantu languages and their respective dialects contributing to a broader  
understanding of voicing dissimilation and dialect evolution.  
REFERENCES  
1. Archangeli, D. B. (1999). Introducing Optimality Theory. Annual Review of Anthropology, 28, 531–  
552. https://doi.org/10.1146/annurev.anthro.28.1.531 Retrieved 4 July 2025  
2. Bennett, P. R., Bierster, A., Gikonyo, W., Hershberg, S., Kamande, J., Perez, C., et al. (1985). Gikuyu ni  
kioigire: A first course in Kikuyu (Vol. 1). Madison, Wisconsin, United States of America: The African  
Studies Program of the University of Wisconsin-Madison.  
3. Davy, J. I. M and Nurse, D. (1982) Synchronic versions of Dahl’s law: The multiple applications of a  
phonological dissimilation rule. Journal of African Languages and Linguistics 4: 157-195.  
https://doi.org/10.1515/jall.1982.4.2.157 Retrieved 3 September 2025  
4. Elwell, R. (2010). Finite State Methods for Bantu Verb Morphology. Retrieved October 31, 2025  
5. https://localizely.com/iso-639-2-list/ Retrieved 5 December 2025  
6. Hyman, L. M. and Mbuui. M. G. (2022) Dahl’s Law and G-Deletion in Tiania: A Dialect of Kimeru  
(Bantu, Kenya). In Language in Africa 3 (2) pp. 212-238. Doi: 10.37892/2686-8946-2022-3-2-212-238  
7. Iribemwangi, P. (2012). A Case for the Harmonization of Kikuyu, Kiembu and Kimbeere Phonology and  
Orthography. In N. O. Ogechi, J. A. Oduor, & P. Iribemwangi (Ed.), The Harmonization and  
Standardization of Kenyan Languages (Vol. 87, pp. 21-38). The Centre for Advanced Studies of African  
Society.  
8. Kager, R. (2007). Optimality Theory (7th ed.). Cambridge, United Kingdom: Cambridge University  
Press.  
9. Kanana, F. E. (2011) Dialect Convergence and Divergence: A case for Chuka and Imenti. Selected  
proceedings of the 40th Annual Conference on African linguistics (ACAL). Eds. Bokamba et al., 99.190-  
205. Somerville, MA: Cascadilla Proceedings Project. Retrieved October 31, 2025  
10. Kanana, F. E. (2011) Meru Dialects: The Linguistic evidence. Nordic Journal of African Studies 20(4):  
300–327 (2011) Retrieved October 31, 2025  
11. Kenya Census. (2019). Retrieved from https://www.knbs.or.ke/reports/kenya-census-2019/ Retrieved 4  
July 2025  
12. Kutik, E. J. (1983). Noun Class Assignment of English Loanwords in Kikuyu. In J. Kaye, H. Koopman,  
D. Sportiche, & A. Dugas (Eds.), Current Approaches to African Linguistics (Vol. 2, pp. 345-359).  
Dordrecht, The Netherlands: Foris Publications.  
13. Maho, J. (2003). A Classification of Bantu Languages: An Update of Guthrie's Referential System. In D.  
Nurse, & G. Philippson (Eds.), The Bantu Languages. London: Routledge.  
14. Miti, L. (2006). Comparative Bantu Phonology and Morphology (Vol. 40). Cape Town: The Centre for  
Advanced Studies of African Society (CASAS).  
15. Prince, A., & Smolensky, P. (2004). Optimality Theory: Constraint Interaction in Generative Grammar.  
Oxford, United Kingdom: Blackwell Publishers.  
16. The International Organization for Standardization https://www.iso.org/obp/ui/#iso:std:iso:639:-2:ed-  
1:v1:en Retrieved October 31, 2025  
17. Werner, A. (1919). Introductory Sketch of the Bantu Languages (1st ed.). Routledge.  
https://doi.org/10.4324/9780429462160 Retrieved 31 October 2025  
Page 32  
www.rsisinternational.org