Journal of Linguistics and Language Teaching
Volume 14 (2023) Issue 1
Coda Adaptation in Urhobo-English Loanwords:
A Constraint-Based Account
Philip Oghenesuowho Ekiugbo (Aba, Nigeria) & Cecilia Amaoge Eme (Awka, Nigeria)
Abstract
This study1 examines how codas of Urhobo-English loanwords are adapted and shows that the strategies adopted in repairing loanword coda in Urhobo are driven by syllabification constraints and universal conventions. Syllabification conditions in languages that forbid filled coda will require that all the consonant sounds in a phonological word that are to be found in the phonetic string are parsed as onsets. Assuming this is true, it has implications for loanword adaptation. Urhobo exclusively permits the open syllable type. Implicitly, all the coda elements of loanwords are likely to be licensed as onsets, which may result in a possible ‘illicit’ onset cluster given the onset condition requirement of the language. Accordingly, this study examines the attested patterns of adaptation of English coda in Urhobo loanwords and their motivations as well as implications. The discussion is built around the theory of constraints and repair strategies.
Keywords: Loanwords, adaptation, coda, repair strategies, syllabification
1 Introduction
Words display phonotactic patterns that are a consequence of constraints on syllabification (Katamba 2006). These constraints determine the distribution and patterning of sounds in a phonological word. Thus, phonotactic constraints shape the overall structure and phonological patterns of words in a language as well as the distribution of sounds and syllables, and ultimately contribute to the distinct sound patterns and characteristics of that language. The syllable plays a central role in phonotactics and has been argued to be the core of phonological representation (Ugorji 2013, Duanmu 2008, Pigott 1999, Blevins 1995). It is an element of phonological structure which consists of one or several segments organised in permissible intrinsic sonority sequences that might constitute the basis for a prosodic statement (Ugorji 2002: 89). The syllable is a crucial unit in phonological theory, with evidence from various languages supporting its formal definition (Hooper 1972). This is further reinforced by experimental studies, which have provided empirical evidence for the nature of the syllable, its structure, and its role in phonology (Zec 2007, Bruck Treiman & Caravolas 1995, Treiman 1983, Fallows 1981).
As has long been established in the literature, syllable structure requirement is the main conditioning factor underlying most phonological phenomena, including processes, repairs and alternations (Ekiugbo 2022, Ugorji 2013, Childs 2003, Hombert 1973). Such phenomena result from the need to meet the syllable structure requirement(s) of the language. Syllable structure requirement refers to the constraints or rules that govern the permissible combinations of sounds within a syllable. It specifies the possible arrangements of consonants and vowels in a syllable and defines the patterns that are allowed or disallowed in a given particular language. According to Nespor & Vogel (2007), the principles of syllabification group segments into well-formed syllables throughout a string. For Itô (1988), syllabification can be viewed as a case of template matching. Segments that are not matched with a slot in the syllable template are unlicensed and, hence, fail to appear in the surface representation. For instance, all the consonant sounds of a phonological word in languages having only open syllables will always be parsed as onsets. This is because of the effect of a syllabification constraint which forbids coda in such languages. This constraint can be represented as in (1) below:
(1) NoCoda: the occurrence in the output of syllables ending on a consonant is prohibited
The syllable in Urhobo consists of a vocalic nucleus with or without an onset element of one or two consonants, but no coda. This has also been reported in many other languages (Blevins 1995, Piggott 1999). Accordingly, the NoCoda constraint in (1) is active in these languages. The implication of this is that these languages will parse all the consonants in a lexical entry as onsets, however, to the extent consistent with other syllable requirements of the language. Example (2) below provides an illustration of this in Urhobo:
(2) /òʒàríkrɛ́/ ‘longsuffering’
(a) [ò. ʒà. rí. krɛ́]
(b) *[ò. ʒà. rík. rɛ́]
The above may also have implications for the structural adaptation of loanwords, especially with respect to the adaptation of codas. The present (?) study therefore examines how the codas of loanwords are adapted in the Urhobo language. It provides a new insight into the motivation and process of coda adaptation in the language drawing on insights from the Theory of Constraints and Repair Strategies (TCRS). Urhobo is a Niger-Congo language spoken in parts of Southern Nigeria. It belongs to the Edoid group of languages, a branch of Benue-Congo. The sound inventory of the language consists of seven vowels, each of which has oral and nasal counterparts (Rolle 2013, Aziza, 2002) as shown in Table (1), and thirty consonants (Kelly 1969) as shown in Table (2).
Table 1: The Vowel Inventory of Urhobo
Table 2: The Consonant Inventory of Urhobo
According to Ugorji (2013), the syllable formalisation for Urhobo includes onset and rhyme constituents. The onset may branch minimally, however with some restrictions, but a branching rhyme is not attested, as shown in Figure 1 below:
Figure 1: An Illustration of the Syllable Template of Urhobo
Lexically, the Urhobo language has borrowed much from various languages. The majority of its loanwords are from English, but it has also borrowed words from Portuguese, Bini, Yoruba, and Igbo (Darah & Ekiugbo 2014, Ojaide & Aziza 2014). However, vocabulary development in recent times has mostly been owing to loanwords from the English language because of the persistent contact between Nigerian indigenous languages and English, the official language in Nigeria. These loanwords cover a wide range of semantic domains and linguistic categories, especially nouns. According to Darah & Ekiugbo (2014: 13),
the Urhobo language borrowed words from languages with which it has contact, in order to have a word in the language, to describe new concepts or elements which do not exist at some point in the history of Urhobo.
They also note that the language has borrowed morphologically simple words to replace some compound or compound-complex words in the language for the purpose of economy.
2 Literature Review
Studies on the phonology of loanwords have shown that borrowed words usually undergo 'adaptation' processes to conform to the constraints of the borrowing language phonology (Kang 2011). The arguments over the nature of structural adaptation of loanwords continue to indicate that cross-linguistically, how languages adapt foreign terms aims primarily at satisfying constraints on phonotactic and syllable structure requirements of the borrowing language. In his study of English loanwords in Modern Standard Arabic (MSA), Al-Athwary (2017) notes that most of the regular adaptations at the syllable level are motivated by linguistic constraints inherited in the phonological system of MSA rather than by extra-linguistic motivations. Contrary instances may then be understood as cases of structural borrowing, given that borrowed items are not restricted to only lexical items but extend to other linguistic elements such as phonological constraints, morphological patterns, and syntactic constructions (Harb 2016, Curnow 2001).
Basically, there are three levels at which modifications can occur in loan word adaptation. These are segmental, phonotactic and prosodic. At each of these levels, borrowed words are adjusted mainly to accommodate the phonology of the recipient language, not that of the donor language (Harb 2016). Such accommodation involves applying different repair strategies to deviant structures or contents. The choice of repairs alluded to in the literature depends on the constraint, language, and theoretical purview (Guba 2018, Al-Athwary 2017, Harb 2016). Repairs in the TCRS model only include two universal phonological operations: insertion and elision. Insertion entails the addition of phonological material (a link or phonological content) to fill a gap that causes a constraint violation, while elision consists of erasing (or unlinking) material to repair the violated constraint (Rose 1995). Alquarni states:
When the constraint violation is due to the lack of phonological content or structure, insertion takes place to compensate for the missing information. For instance, if a language requires a final vowel in its native lexicon, foreign loanwords in such a language will be supported with a final vowel in conformity to its phonological requirements. In contrast, when the violation is triggered by the superfluous excess of segmental malformations, deletion applies and erases the problematic content. For example, in a language that disallows consonant clusters, a deletion strategy may apply and obliterate one consonant. (Alquarni 221: 427)
In the existent literature on Urhobo (Bolaji 2016, Ugorji 2013, Aziza & Utulu 2006), with respect to the repair of English coda, it is argued that the syllable requirement in Urhobo forces coda elements of loanwords to be repaired using either elision or epenthesis. According to Aziza & Utulu (2006), deviant structures of English loanwords in Urhobo are modified using one of two strategies: the deletion of a consonant or the insertion of a vowel, or a combination of both. The repairs are motivated by the need to obligatorily fulfil the two constraints which all syllables entering into the Urhobo language (and Yoruba) must observe: (i) no (illicit) complex onset, and (ii) no coda. They observe that a complex onset is simplified either by deleting one of the consonants or, more commonly, by inserting the vowel /i/ between the two consonants, while codas are repaired by either elision or epenthesis.
Similar propositions are presented in Ugorji (2013) in which the procedures for dealing with ‘illicit’ segments, structures and suprasegment of English loanwords in Urhobo are also discussed. The study drew on insight from the tenets of the generalized syllable syntactic framework, a CV-phonology approach. It noted that adaptation of English loanwords in Urhobo is motivated by the need for (re-)syllabification, and is achieved through prothesis, epenthesis, syncope, apocope and substitution. These accounts, however, do not explain why some source coda are deleted while others are not, or why source nasal coda is realised as a nasal feature on preceding vowel sounds.
3 Theoretical Framework
The present study adopts the TCRS model because of its derivational approach to phonological constraints. It is a theory of phonology which posits that the phonological well-formedness in a language is governed by both universal and language-particular constraints which may be positive or negative and are said to be inviolable. The phonological grammar in the TCRS model consists of positive constraints that define acceptable configuration as well as a “negative constraint that states that a certain configuration is illicit” (Calabrese 2022: 530), as well as an operation known as repair which is used when a constraint is violated (Rose 1995). Thus, a repair operation only applies when a constraint is violated to preserve the constraint. According to Murphy (2019: 1), “a repair can be characterized as a particular (often generally unavailable) structural change licensed to avoid an illicit output configuration.”
Following Paradis & Lacharite (1997), a repair strategy is a universal, non-contextual rule that is always driven by a given constraint, i.e. it only applies to preserve a violated constraint. There are two types of constraint violations: intrinsic violations, i.e. violations internal to the language, and extrinsic violations, i.e. violations external to the language. Following Paradis & El Fenne (1995), violations can result intrinsically from underlying ill-formed phonological material, morphological and syntactic operations, or constraint conflicts, while extrinsic sources of violations include borrowings and paraphasias. Also, repairs are not made arbitrarily. They are governed by three principles, which are minimality, preservation and threshold principles.
4 Coda Adaptation in Urhobo Loanwords
Certain configuration constraints govern the arrangement of consonants and vowels within a syllable. Such constraints specify the nature of the nucleus and whether marginal elements (onsets or codas) are permissible in the language, as well as which and how many consonants can appear in such positions, if it is permissible. The basic composition of the Urhobo syllable consists of an obligatory nucleus, an optional onset and no coda. Only vowels ([-cons] element) may occur as the nucleus and any consonant as a simple onset. Where a consonant cluster involving two consonants, the first consonant is a labial or velar sound, and the second a liquid. However, our concern in this study is with the constraints on marginal elements, that is, the constraint barring coda and the constraint on onset condition.
Implicitly, the open syllable is the only syllable type attested in Urhobo. Thus, the syllable constraints in the language will require that all syllables be unchecked. This is a standard phonological principle for languages that disallow codas. Given that the syllable templates in such languages have a dispreference for filled coda, it is therefore expected that the languages will not license coda elements. Implicitly, coda elements of loanwords may be parsed differently from what they are in the source language. Following Paradis & LaCharité (1997: 380), loanwords frequently enter the borrowing language (L1) with structures (segments and sequences) that are ill-formed from the perspective of the borrowing language. Such structures are often repaired minimally by universal phonological operations that are triggered by the constraints which we observed on the surface. One of the areas in which loanwords of English origin in Urhobo are likely to violate the phonotactics of the language is the borrowing of words with a coda. In what follows, data are shown and discussed in relation to how the NoCoda constraint as in example (1) impacts the adaptation of the coda of loanwords in Urhobo:
(3) a. bles [blɛ̀.sí] ‘bless (v)’
b. brɛd2 [i.brɛ.di]3 ‘bread’
c. pɒt [ipɔtu] ‘pot’
d. ʃaɪn [ʃai.ni] ‘shine’
e. fɪlm [i.fi.mu] ‘film’
f. bʌlb [ì.bɔ́.bù] ‘bulb’
g. belt [i.bɛ.ti] ‘belt’
h. gɒlf [i.gɔ.fu] ‘golf’
i. bɒks [i.bɔ.su] ‘box’
j. dʒɪst [ʒi.s.ti]4 ‘gist’
(4) a. kɒn.tækt [kɔ̃̀tátì] ‘contact’
b. ˈbʌ.tn [ìbɔ́tìnĩ̀] ‘button’
c. ˈsʌb.dʒekt [ìsɔ́jɛ̀tì] ‘subject’
d. ki.bɔd [ìkíbɔ̀dù] ‘cupboard’
e. əˈlɑːm [alamũ] ‘alarm’
f. ˈænθəm [a.tɛ.mũ] ‘anthem’
g. ˈbækgraʊnd [i.ba.gra.nĩ] ‘background’
h. bandɪt [i.bã.di.ti] ‘bandit’
i. ˈblæŋkɪt [i.blã.kɛ.ti] ‘blanket’
j. ˈkʌmfət [kɔ̃fɔtu] ‘comfort (v)’
The phonetic strings of some Urhobo loanwords of English origin, both in the source language and the borrowing language, are shown in examples (3) and (4) above, all of which have source codas, which is the focus of this study. These examples are instances of source monosyllabic and disyllabic words, respectively. The following may be noted from the examples about the adaptation of coda in Urhobo loanwords:
In a simple coda of the monosyllabic words, the consonant is resyllabified as onset, with an epithetic vowel inserted (3a-d).
In a complex coda of the monosyllabic words, the last consonant sound(s) is resyllabified as onset to the extent consistent with the syllable requirement of Urhobo, while the other consonant(s) does (do) not appear in the phonetic string (3e-j). Furthermore, where the first consonant is a nasal, the preceding vowel is realized as /Ṽ/ (3g-j).
In di-/polysyllabic words, simple codas of final syllables are treated as (i), and complex ones are treated as (ii); the coda of non-final syllable is deleted, while ambisyllabic consonants are realized as onsets (4).
Where the non-final syllable is a nasal, the preceding vowel is realized as /Ṽ/ (See 4a).
Where a nasal coda is deleted, a nasal feature is copied on the preceding vowel, except where the vowel is in the initial onsetless syllable, as in (4f).
From the foregoing, there are three bipartite areas that seem to trigger different repair strategies in the repair of the loanword coda in Urhobo, at least superficially. These are:
the nativization of final coda versus non-final coda;
the nativization of simple versus complex coda; and
the copying of nasality on preceding vowels with respect to non-onsetless syllable versus onsetless syllable.
For instance, while final codas are retained, non-final codas are not. Also, while final simple codas, in addition to having the epenthetic vowel attached after the coda, are re-syllabified to become onset elements, only the rightmost consonants of complex codas are retained while all the other consonants are deleted. Implicitly, this seems to suggest that the different natures and environments of codas are possibly governed by two different constraints, at least, superficially. This is because different repair strategies are invoked.
In this study, it will, however, be argued that in Urhobo, only a single coda constraint governs all the natures (simple vs. complex) and environments (final vs. non-final) of no coda constraint violations, and that, in fact, the same repair strategies apply. To provide a systematic account of these observations, this study makes the following assumptions:
Firstly, the syllabification constraint, which requires all consonants to be parsed as onsets in Urhobo, also applies to loanwords. Thus, all loanword codas are (re-) syllabified as onsets, but only to the extent that they are consistent with the language's permissible syllables. As already noted, the Urhobo language has three syllable types, which are V, CV, and CCV, with the C’s being optional elements. There are also restrictions on onset clusters in the language both at the phonological and phonetic levels, such that only a velar or labial (/p, b, k, g, m, f, v, x, ɣ/) can occur as C1, while C2 must obligatorily be a liquid. Implicitly, it is expected that any structures that do not conform to the above are not linked to any timing slot and are therefore omitted from the resulting phonetic string.
Secondly, syllabification in Urhobo is described in terms of onset prominence and is assigned from right to left. In onset prominence, the syllabification phenomenon applies as the demarcation of syllable margins. The supposition here is that peak assignment is really not the driving force behind syllabification, as proposed in the traditional theory of syllabification such as the maximal onset principle. According to Schwartz,
since phonology is concerned with discreteness, and discreteness is defined in terms of boundaries, it might be reasonable to question whether the “nucleus”, the only syllabic position that does not by definition define a boundary, should play any role in the delineation of phonological constituents. (Schwartz 2015: 1)
Lastly, the preservation principle will require that “…segmental information is maximally preserved” (Paradis & Lacharité 1997: 384). The preservation principle protects the input of a formative, preferring repair processes involving insertion rather than deletion. This means that repair processes, at least, with TCRS are often achieved through the insertion or deletion of content (such as features or timing units) or structure (links between features or various levels of structure). However, preference is generally given to insertion, because it helps to maximize the underlying form. The functionality of the preservation principle is guided by the threshold principle.
The Threshold Principle5 posits that
all languages have a tolerance threshold to the amount of repair needed to enforce segment preservation and that
this threshold is the same for all languages: two steps (or two repairs) within a given constraint.
Given these assumptions, it is argued that all the consonants in loanwords are first licenced as the onset, such that all adjacent consonants are earmarked as likely onset clusters. However, where the possible resulting onset clusters are illicit, i.e., do not conform to a language particular requirement or a universal tendency for onset clusters, the nonconforming elements are not linked to any slot during parsing and, by extension, are deleted, given universal conventions for syllabification.
(5) Universal conventions (Paradis, 1988:76):
Segment licensing: a timing unit (a slot) must be attached to a segment or deleted.
A segment must be attached to a syllabified slot or be delinked.
In what follows, we illustrate how the structural adaptation of loanword coda in Urhobo operates drawing on insights from the assumptions discussed above:
(6) i. source form ii. licencing iii. Processes iv. Surface form
a. /səb.dʒekt/ (s)ɔ(bʒ)e(kt) sɔ.Øʒe.Øt V [sɔ̀ʒètí]
b. /sʌb.let/ (s)ɔ(bl)e(t) sɔ.ble.tV [sɔbleti]
c. /tʃeə.mən/ (ʃ)ɛ(m)a(n) V.ʃɛ.ma.nV [iʃɛmani]
d. /tʃeə.mən.ʃɪp/ (ʃ)e(m)a(nʃ)i(p) V.ʃe.ma. Øʃi.pV [i.ʃe.mã.ʃɪ.pi]
A minimally specified strategy for structural adaptation of loanwords coda in Urhobo is shown in example (6) above. The first column shows the source forms for the words subject, sublet, chairman and chairmanship. In the second column, all the consonants in the source form are shown to be licensed as a possible onset. Thus, during parsing, they get linked to onset slots to the extent consistent with the syllable requirement of the Urhobo language. To illustrate with (6a), given that /bʒ/ is neither a permissible onset cluster in Urhobo nor does it follow universal tendencies, only /ʒ/, being the closest element to the peak slot, is attached to the onset slot, while /b/ is deleted. That this is the case is evident in the repair of /sʌb.let/ ‘sublet’ in (5b), which is repaired as [sɔ.blɛ.ti]. The coda of the first syllable in the source form (/b/) is not elided as is the case in (7a). This is because the /bl/ cluster is permissible in the language. Further illustration is shown in (6c) in which the coda of the second syllable (/n/) is retained because it is a final simple coda, but this coda is deleted in (6d), given that it is not a final coda in the morphologically derived word chairmanship.
In the example (6d), the /n-C/ cluster is licensed as a possible onset. Considering that this cluster is not permitted in the language given the syllable requirement for onsets which permits only a Consonant-Liquid cluster, and syllabification applies from right to left, only the rightmost consonant is linked to the onset slot. This explains why some non-final codas are deleted, while word-final simple codas are retained. It also explains the nature of the simplification of final complex codas. As already noted, elided nasal codas leave their nasal feature behind. This nasal feature can be perceived auditorily on the preceding vowels as shown in example (7) below:
(7) a. /bæn.dɪt/ [i.bã.di.tʰi] ‘bandit’
b. /tent/ [i.tʰɛ̃.tʰi] ‘tent’
c. /lɔnʃ/ [lɔ̃.ʃi] ‘launch’
d. /bæŋk/ [i.bã.kʰi] ‘bank’
e. /brɒnz/ [i.brɔ̃.zu] ‘bronze’
f. /pænt/ [i.pã.ti] ‘pant’
g. /fræŋk/ [ì.frã́.kì] ‘frank’
h. /prɒmpt/ [prɔ̃́.tù] ‘prompt’
i. /plæŋk/ [i.plã.ki] ‘plank’
j. /leŋθ/ [i.lɛ̃.ti] ‘length’
From these examples, it can be observed that where the elided coda element is a nasal sound, the surface form of the lexical item in the target language is realised with a nasal vowel preceding the environment from which the source nasal coda was deleted. However, this apparent phenomenon is restricted to syllables having onsets, as (initial) onsetless syllables do not behave in this manner as shown in example (8) below:
(8) a. ˈænθəm [a.tɛ.mũ] ‘anthem’
b. /ɪn.spek.tə/ [ispɛtɔ] ‘inspector’
c. /'ɪŋ.glɪʃ/ [i.gli.ʃi] ‘English’
d. /ɪŋ.glənd/ [i.glã.di] ‘England’
To explain the first of the two latter observations, a distinction is made between repair strategies which are processes resulting directly from constraint violation and phonological processes. The nasal spread phenomenon here is not a process directly triggered as a result of constraint violation, but a ‘product’ of regressive nasal spread which precedes the application of repair processes (Eme & Ekiugbo, forthcoming) as illustrated in example (9) below. This corroborates arguments which claim that contrastive nasal vowels have their nasal feature from diachronically elided nasal coda (Salffner 2013, Osiruemu 2011):
(9) a. Source Phonetic Form: ˈkɒn.tækt
b. Licensing: kɔ.(n)ta.(k)tV
c. Phonological Processes: kɔ̃.(n)ta.(k)ti
d. Repair Processes: kɔ̃. Øta. Øti
e. Surface form: [kɔ̃̀tátì] ‘contact (V)’
As already stated above, non-final codas are often deleted given that if they are retained as part of the onsets of the following consonants, the resulting sequence may not be permissible. Thus, the non-final coda in example (9) will be deleted as expected. However, the resulting surface form shows the presence of a nasal feature on the vowel preceding the nasal coda in the source form. This is due to a process of nasal assimilation, which is a general process in the language applied to the structure prior to the application of the repair process (see 9c-d).
5 Conclusion
The arrangement of consonants and vowels within a syllable is governed by configuration constraints, which specify the nature of the nucleus and whether marginal elements (onsets or codas) are permissible in the language as well as their nature. In Urhobo, the basic composition of the syllable consists of an obligatory nucleus, an optional onset, and no coda. Thus, syllabification constraints will, among other things, prohibit the occurrence of coda. Implicitly, the constraints will require that all consonants be parsed as onsets. Thus, given that loanwords of English origin in Urhobo are likely to violate such requirements, it is to be expected that the language will adapt all coda entering the language.
This study therefore examined how the coda elements of loanwords are handled in Urhobo. It investigates the different strategies involved in the adaptation process in Urhobo. It showed three bipartite areas that trigger different repair strategies in the repair of the loanword coda in Urhobo: the nativization of final coda versus non-final coda; the nativization of simple versus complex coda; and the copying of nasality on preceding vowels with respect to non-onsetless syllable versus onsetless syllable.
The three superficially different coda environments are governed by the same constraint, NoCoda. Accordingly, the same repair processes are applied. The strategies employed in the repair process are insertion and deletion. The apparent difference results from factors such as syllabification, onset composition requirement and general process. Syllabification rules in the language parse all consonants as onsets, and repairs are only applied when the resulting sequence is inconsistent with other syllable requirements in the language, especially requirements on onset composition. Accordingly, repair is only applied to ensure that the resulting surface form follows the syllable requirement of the language. This explains why some source codas do not appear in the phonetic string of the borrowing language, yet others do, but as onsets.
References
Al-Athwary, Anwar A. H. (2017). The Phonotactic Adaptation of English Loanwords in Arabic. In: Arab World English Journal 8 (3), 392-406.
Alqarni, Muteb (2021). Arabic Loanwords in Seven Ethiopia Languages. In: Brill’s Journal of AfroAsiatic Languages and Linguistics 13, 423-473.
Aziza, Rose O. (2002). Nasality in Urhobo: An Autosegmental Perspective. In: Journal of West African Languages 29 (2), 11-21.
Aziza, Rose O. & Don C. Utulu (2006). Loanword Phonology: English in Urhobo and Yoruba. In: Journal of West African Languages 33 (2), 3-22.
Blevins, Juliette (1995). The Syllable in Phonological Theory. In: Goldsmith, John (Ed.): Handbook of Phonological Theory. Oxford: Blackwell, 206-244.
Bolaji, Ezekiel T. (2016). Constraints Interaction in the Phonology of Urhobo-English Loanwords”. In: Odebunmi, Akin, Ayo Osisanwo, Helen Bodunde & Stella Ekpe (Eds.): Grammar, Applied Linguistics and Society: A Festschrift for Wale Osisanwo. Ile-Ife: Obafemi Awolowo University Press, 207-221.
Bruck, Maggie, Rebecca Treiman & Marketa Caravolas (1995). Role of the syllable in the processing of spoken English: Evidence from a nonword comparison task. In: Journal of Experimental Psychology: Human Perception and Performance 21(3), 469-479.
Calabrese, Andrea (2022). Historical Notes on Constraint-and-Repair Approaches”. In: Dresher, B. Elan & Harry van der Hulst (Eds): The Oxford History of Phonology. Oxford: Oxford University Press, 530-550.
Childs, G. Tucker (2003). An Introduction to African languages. Amsterdam: John Benjamins.
Curnow, Timothy Jowan (2001). What Language Features Can Be Borrowed? In: Aikhenvald, Alexandra Y. & Robert M. W. Dixo (Eds.): Areal Diffusion and Genetic Inheritance: Problems in Comparative Linguistics. Oxford: Oxford University Press, 412-436.
Darah, Moses E. & Philip O. Ekiugbo (2014): Dissecting the Phenomenon of Borrowing in Urhobo. In: Journal of Linguistics, Language and Culture 1 (1), 17-27.
Duanmu, San (2008). Syllable Structure: The Limits of Variation. Oxford: Oxford University Press.
Ekiugbo, Philip O. (2022a). Rule Ordering in Uvwie. In: Macrolinguistics 10 (1), 101-110.
Ekiugbo, Philip O. (2022). Another Chapter in the Story of /s/: The Voiceless Alveolar Fricative in Urhobo Loanword Adaptation. Paper Presented at the 1st International Conference of the Department Of Linguistics. Nnamdi Azikiwe University, Awka.
Eme, Cecilia A. & Philip O. Ekiugbo (forthcoming): Hiatus Resolution in Urhobo: A Constraint-Based Account. In: An-Najah University Journal for Research – B (Humanities) 38 (2) (https://journals.najah.edu/journal/anujr-b/issue/anujr-b-v38-i2/article/2155/in-press/; 15-08-2023).
Guba, Mohammed N. A. (2016). Phonological Adaptation of English Loanwords in Ammani Arabic. PhD dissertation, University of Salford.
Harb, Mustafa A. (2016). An Optimality-Theoretic Account of English Loanwords in Hawaiian. In: Open Journal of Modern Linguistics 6, 230-242.
Hombert, Jean-Marie (1973). Speaking Backwards in Bakwiri. In: Studies in African Linguistics 4, 227-236.
Hooper, Joan B. (1972). The Syllable in Phonological Theory. In: Language 48 (3), 525-540.
Itô, Junko (1988). Syllable Theory in Prosodic Phonology. New York: Garland Publishing.
Kang, Yoonjung (2011). Loanword Phonology. In: M. van Oostendorp, C. Ewen, E. Hume & K. Rice (Eds.): Companion to Phonology. Malden, MA.: Wiley Blackwell, 2258-2282.
Katamber, Francis (2006): Phonetically Motivated Word Formation. In: Brown. K. (Ed): Encyclopedia of Languages and Linguistics. Amsterdam: Elsevier, 411-414.
Murphy, Andrew (2019). Resolving Conflicts with Violable Constraints: On the Cross-Modular Parallelism of Repairs. In: Glossa: A Journal of General Linguistics 4 (1), 1-39.
Ojaide, Tanure and Rose O. Aziza (2014). Urhobo Etymology. In: Aridon 1, 103-112.
Osiruemu, Evarista O. (2011). A Re-Appraisal of Nasal Vowels in Esan. In. Journal of Linguistics Association of Nigeria 14 (2), 319-327.
Paradis, Carole (1988). On constraints and Repair Strategies. In: The Linguistic Review 6, 71-97.
Paradis, Carole & Caroline Lebel (1994). Contrasts from Segmental Parameter Settings in Loanwords: Core and Periphery in Quebec French. In: Toronto Working Papers in Linguistics 13, 75-94.
Paradis, Carole & Darlene LaCharité (1997). Preservation and Minimality in Loanword Adaptation. In: Journal of Linguistics 33, 379-430.
Paradis, Carole & Fatimazohra El Fenne. (1995). French Verbal Inflection Revisited: Constraints, Repairs and Floating Consonants. In: Lingua 95 (1-3), 169-204.
Paradis, Carole & Jean-François Prunet (2000): Nasal Vowels as Two Segments: Evidence from Borrowings. In: Language 76 (2), 324-357.
Pigott, Glyne L. (1999). At the Right Edge of Words. In: The Linguistic Review 16 (2), 143-186.
Rolle, Nicholas (2013). Phonetics and phonology of Urhobo. In: UC Berkeley Phonology Lab Annual Report 9, 281-326.
Rose, Yvan (1994). L’Adaptation des Voyelles Nasales en Kinyarwanda: L’effet du Principe de Preservation. In: Journees de Linguistique VIII, 162-165.
Rose, Yvan (1995). Minimalité, Préservation et Tolérance dans les Emprunts Français en Kinyarwanda. MA Thesis, Université Laval.
Salffner, Sophie (2013). Final nasal consonants and nasalized vowels in Ikaan. In: SOAS Working Paper in Linguistics 16, 311-332.
Treiman, Rebecca. (1983): The Structure of Spoken Syllables: Evidence from Novel Word Games. In: Cognition 15 (1-3), 49-74.
Ugorji, Christian U. C. (2002). The Structure of Syllables in Mbieri: A Multilinear Approach. In: Kiabara 8 (2), 89-99.
Ugorji, Christian U. C. (2013.: Phonology of English Loanwords in Urhobo Corpus. In: Akande, A. T. & R. Taiwo (Eds.): Contact Linguistics in Africa and Beyond, 181-196. New York: Nova Publishers.
Zec, D. (2007). The syllable. In: Lacy, P. (Ed.): The Cambridge Handbook of Phonology. Cambridge: Cambridge University Press, 161-194.
Authors:
Philip Oghenesuowho Ekiugbo
Faculty Member
National Institute for Nigerian Languages
PMB 7078, Aba
Abia State
Nigeria
Email: oghenesuowho@gmail.com
Cecilia Amaoge Eme
Professor
Faculty of Arts
Linguistics Department
Nnamdi Azikiwe University
PMB 5025, Awka
Anambra State
Nigeria
Email: ca.eme@unizik.edu.ng
_______________________________
1 This study is a part of a research project on the phonological grammar of Urhobo. The data for the project was collected through part-funding from the Chief Mukoro Mowoe Research Scholars Fund (2021).
2 Throughout this article, IPA symbols are used to allow for consistency and to eliminate confusion. Thus, all non-IPA symbols in previous studies have been replaced with their corresponding IPA symbols.
3 Nouns in Urhobo begin vowels (initial onsetless syllable). This is also applicable to loanwords.
4 Following Ekiugbo (2022), /s/ is extraprosodic, hence it is not affected during repair.
5 Paradis & Label (1994) and Rose (1994), respectively, show that this limit to preservation holds for Fula and Kinyarwanda, evidence for which Paradis & Lacharité (1997: 385) posit that two repairs represent the universal limit for adapting ill-formed structures. However, they also state that “… should the threshold be set differently in other languages, the second part of the principle would have to be parametrised.” (Paradis & Lacharité 1997: 385)