Introduction
The term ‘structure in linguistics’ is mostly used to refer to a sequence of units that are in a certain linguistic relationship to one another. Thus no matter how minimal a sequence is, if it can be analysed in terms of a relationship within it, then that is a structural unit. For example, one of the structures of a noun phrase may be ‘article + adjective + noun’ as in the vicious circle (Richards & Schmidts, Reference Richards and Schmidt2002). Structures entail relationships which may be syntagmatic or paradigmatic. According to Finnegan (Reference Finegan2014) when we say language structure we are essentially talking about syntax, semantics and phonology of a language. When it comes to idioms the same principles apply. The structures of idioms are essentially their syntactic behaviour. This behaviour cannot be predicted solely on the basis of their form or figurative meaning alone, but it must be due to some relation between the form and meaning (Gibbs & Nayak, Reference Gibbs and Nayak.1989). An idiom is an institutionalised and conventionalized sequence of at least two words or free morphemes that is semantically restricted so that it functions as a single lexical unit, whose meaning cannot or can only to a certain extent be deduced from the meanings of its constituents. (Skandera, Reference Skandera, Allerton, Skandera and Tschichold2003: 60). To Nurnberg, Sag and Wasow (Reference Nunberg, Sag and Wasow1994), idioms are characterised as having conventional meaning, figuration, inflexibility of form, and proverbiality. Idioms have been called ‘multiword units’ (Grant & Bauer, Reference Grant and Bauer2004), metaphors (Gibbs, Reference Gibbs1993; Toris, Reference Toris2011), phrasemes (Howarth, Reference Howarth1998), fixed expressions (Moon, Reference Moon, Schmitt and McMarthy1997; Carter, Reference Carter1998) and formulaic expressions (Wray, Reference Wray2002). Structurally, idioms do not form a unique class of linguistic items such that all idioms belong to it, but that they share many of the same properties normally associated with more literal expressions.
Structurally there are two types of idioms, viz., lexical idioms which are syntactically dominated by one of the lowest syntactic (grammatical) categories (nouns, adjectives, verbs, etc.) and phrase idioms which are not headed by any grammatical category but form a unit that is at best phrasal in its entirety (Hockett, Reference Hockett1958). Idioms share all the syntactic features of other nonfigurative units (Vegge, Reference Vegge2012). Thus they can undergo modification, quantification, topicalisation, and transformation. Grant and Bauer (Reference Grant and Bauer2004) have adequately addressed this problem in their attempt to give criteria for identifying idioms. Idioms have both internal and external structure. The former is their internal organization while the latter deals with how materials contained within the idiom interact with materials that are not part of them. It is on account of their internal structure that idioms are noun phrase, verb phrase, and adjective phrase idioms. On account of the external structure idioms are predicative i.e. they serve the function of a predicate within the clause.
There have been some studies on the nature and specific properties of idioms – notably Nicholas (Reference Nicholas, Everaert, van der Linden, Schenk and Schreuder1995) and Mostafa (Reference Mostafa2010). There is, however, a dearth of research on the specific structure of idioms. This is perhaps because idioms have been considered as part of the language and any discussion of any language structure should also subsume that of idioms. This is especially true in the context of Nigerian English.
The term ‘Nigerian English’ (NigE) can be broadly defined as ‘the variety of English spoken and used by Nigerians’ (Adeniyi, Reference Adeniyi2006: 25). It is characterised by a lot of borrowings, nativisation and coinages, especially from the indigenous languages. Akere (Reference Akere2006: 9) describes this domestication as ‘the transformation of English as an alien medium, to make it respond to local imagery, figures of speech, sound patterns and the general cultural milieu of the region.’ This variety has helped Nigerians to express their worldview in an international medium of communication like the English language. That an internal norm in English language in Nigeria has developed is evident from the fact that numerous structures which would be totally unintelligible to native speakers of English are used officially at the highest level of government in Nigeria (Bamiro, Reference Bamiro1991:7).
Research questions
This work is part of an expanded research on the structure and use of idioms in Nigerian English. The main objective is to discuss the structure of idioms in NigE and show how the indigenous languages contribute to this structure. The following research questions will be answered.
1) What is the structure of idioms in NigE?
2) How do the indigenous languages contribute to the structure of idioms in NigE?
Materials and methods
The procedure in this work includes: first determining the data source, then selecting sample sources and then proceeding to identify the samples from the sample source using designated idiom identification tools. This was done strictly according to Umar et al. (Reference Umar, Yap, Chan and Wong2016). The data source determined in this work are the newspapers, novels and academic articles written in English in Nigeria. The bulk of these sources come from two newspaper titles. Siepmann (2005: 25) has described newspaper language as a fountain from which living idioms are constantly renewed and which provides a ready source of language material. Specifically, two newspapers – Vanguard and Dailytrust – were used. Although they are not the only English medium newspapers, they were selected because they have more widespread coverage in the country. 30 issues of Vanguard dated from the 3rd November 2013 to 21st January 2014 and 30 issues of Dailytrust dated from the 24th February 2014 to 12th May 2014 were used, based on their availability. A total of 3,667 pages containing an estimated 3,428,625 words were checked. To give the task of the identification a sense of accuracy and a systematic outlook, an existing idiom identification method was used. The word sense disambiguation (WSD) model (Hashimota & Kawahara, Reference Hashimoto and Kawahara2009) was used to identify the idioms in context. The WSD was used to identify idioms in the Japanese language. Its major features include a consideration of the Part-of-Speech of Neighboring Words; a consideration of Single Words in the Surrounding Context; and a consideration of Local Collocations. Not all of its features are applicable to English language (cf. Hashimota & Kawahara, Reference Hashimoto and Kawahara2009). It is hypothesised that if an idiom can be found in any dictionary of English idioms, then it is not unique to Nigerian English (Umar et al., Reference Umar, Yap, Chan and Wong2016). Even if an idiom is found in any English idiom dictionary but that it is used with a different meaning, then that idiom is unique to Nigerian English. In order to check the identified samples, three idiom dictionaries published by British establishments, two published by American establishments, and two corpus-based sources were used: Longman Idioms Dictionary (2000), with 6000 entries; Cambridge Idiom Dictionary (2006), with up to 7000 entries; Oxford Idioms Dictionary for Learners of English (2006), with over 10,000 entries; Spears (Reference Spears2005), with 24,000 entries; Ammer (Reference Ammer2003) with 10,000 entries. The British National Corpus (BNC) and the WebCorpus were used to provide currency.
Results
An effort was exerted to minimise human error. A total of 2,121 samples were identified: 756 samples came from the Dailytrust newspaper, 1229 samples came from the Vanguard newspaper, while 136 samples came from other sources. The samples contained repetitions and conjugational/collocational variations. For instance there was in the data source, handle with kid gloves and treat with kid gloves, both having the same meaning; if only one of them appears in any dictionary then the other will be considered as its variant. The identified samples were then conflated into tokens to remove the repetitions, and their number became 1,415.
Data analysis
Idioms have both internal and external structure (this has been discussed above). The internal syntax of idioms from NigE is looked at following Kövecskes & Szabó (Reference Kövecskes and Szabó1996) and the external structure is considered following Moon (Reference Moon1998).
Internal structure of idioms
Based on the data available and a consideration of their internal structures, three types of idioms can be identified in NigE and they are discussed below.
General English idioms: These are idioms that can be found in most other varieties of English because they can be located in most English idiom dictionaries and corpora. Here they are referred to as general English idioms. Of the 1415 idiom tokens identified only 168 are not originally from the general English. The rest are either used the way they are in general English, or modified with a substitution, a deletion or an addition of an element. The idioms common to all varieties abound in as many instances as possible. They occur within the sentence as subjects or predicates – in form of verb phrases, prepositional phrases or even as adjuncts. Below are examples of such idioms.
1. I met my better half. (VNN)
2. He would as usual sweep the matter under the carpet. (DTN)
3. Heavily armed police men sealed off the Baraje group's secretariat (VNN)
When one says general English here, one actually means all other varieties of English apart from NigE. This includes native speaker and non-native speaker varieties of English – essentially Kachru's (Reference Kachru1986, Reference Kachru1990) inner, outer and expanding circles of English (for a detailed discussion of the structure of idioms in this category see Makkai, Reference Makkai1972, Fernando, Reference Fernando1978, Moon, Reference Moon1998, among others).
Modified Idioms: The second category of idioms is that of idioms that have been modified with all or part of the meaning in the idioms retained. This modification is manifested in instances of the addition of lexical item(s) to the string (the item added can be an affix, a word or even a phrase); the substitution of some lexical items from outside the original source; the deletion of some lexical items from the string (meaning still the same); suppletion; and grammatical modification (where the position of a lexical item is modified, i.e. singular to plural). The examples below give instances of modified idioms
4. He was only describing those who were fighting him through newspaper publications as fifth coloumnists
5. All they do is to … shout foul at every given opportunity
6. The adverse economy in Nigeria has seen everybody struggling to eke a living;
7. Security of lives and property is at the lowest ebb
Idioms unique to Nigerian English: The third type of idiom category identified is that of idioms unique to NigE. These are idioms that can only be found in NigE. All the idioms in this category were subjected to a rigorous crosschecking in five idiom dictionaries, one mainstream dictionary and two web corpora. Where not a single instance of use can be detected except from Nigerian sources, then an idiom is considered as being unique to NigE. Since the idioms here are unique to NigE and cannot be found in any idiom dictionary, wherever any instance is cited, glosses are provided. 168 tokens have been identified as being unique to NigE; 29% of them are from the DTN, 51% are from VNN and 20% are from other sources.
The external structure of idioms in Nigerian English
The external structure of the idiom has to do with its relationship to other elements within the clause. Following Moon (Reference Moon1998), the position here is that the minimum structure of an idiom is that it is a phrase. Whatever is less than a phrase cannot be considered as an idiom based on this consideration.
Although idioms are minimally phrases, maximally they can manifest themselves as clauses. In all the data at hand all such clauses always appear as parts of sentences rather than alone. Grant and Bauer (Reference Grant and Bauer2004) claim that the compositionality, the institutionalisation and frozenness that are features of idioms are also factors to consider in discussing the external structure of idioms. The arguments for the three above are well documented in the literature. All that needs to be done is to state whether our tokens satisfy the conditions for compositionality, institutionalisation and frozenness. None of our tokens is completely compositional. The logic is that if meaning can be derivable from the surface then it will be a normal string and cannot have any idiomaticity. All of them enjoy some institutionalisation since they are used in formal contexts. Some of the idioms are completely frozen but a lot are fairly decomposable.
Additionally idiom strings act or function as subjects/objects, complements, adjuncts and predicates. Based on the data at hand the predicative idioms are more in number than any other type. These are idioms that serve as the predicate of the clause within which they occur. According to Nicholas (Reference Nicholas, Everaert, van der Linden, Schenk and Schreuder1995), there are seven types of predicative idioms.
The noun groups are the second-largest and they include adjectives. There are a lot of verb phrase groups but sometimes they occur as part of a predicate and are thus considered predicative. There are also prepositional structures that serve as complements of nouns or noun groups. Here it should be pointed out that the external structure of idioms unique to the NigE is not different from that of other idioms in general English. All the tokens identified as being unique to NigE can conveniently be explained along the same line as other idioms in general English as far as their external structure in concerned. It is with the internal structure that differences arise.
Discussion
The distribution of the tokens appears lopsided in favour of the Vanguard newspaper as opposed to the DailyTrust newspaper. The simple reason is that the English language serves a bigger role of communication in southern Nigeria as there is no other indigenous language that competes with it and thus its ascendancy is complete. In northern Nigeria, however, Hausa language serves as the regional lingua franca, taking away much need for English language. The sources of NigE can also be sources for idioms unique to NigE (cf. Adegbija, Reference Adegbija2004; Peter & Wolf, Reference Peter and Wolf2007; Mgbemena, Reference Mgbemena2014; Kperogi, Reference Kperogi2015). The sources can be summarised as below. Kperogi (Reference Kperogi and Afọláyan2010) gave four sources of NigE. There is not much argument with his position, because he cited ample examples; however only one source can be attributed to idioms unique to NigE, i.e. that of linguistic improvisation. Adegbija (2003) gives six sources of idioms in NigE (analogical creation; native source equivalents; metaphorical provenance; personality-related; the media; and slangs). The sources cannot account for all of the data at hand. Four more sources can be added which have not been captured by Adegbija (2003). One of the sources, i.e. religion, has been corroborated by Kperogi (Reference Kperogi2015). The four additional sources identified through an analysis of the data at hand are discussed below.
a) Nigerian Pidgin English (NPE) is a source of idioms unique to NigE. The majority of users of English, especially from southern Nigeria, first learn NPE informally before starting formal schooling. They thus bring along usage habits into the mainstream NigE (Peter & Wolf, Reference Peter and Wolf2007). Bad belle (DTN11) is an expression that has long been used in NPE (Mowarin, 2010). A search in some old newspapers and records conducted by the researcher showed that only recently has its presence been noticed in formal newspapers. ‘Belle’ is a corrupt form of belly. To have bad belle for someone is to harbour ill will against him. ‘Belle’ here stands for the heart, which is the repository of feelings.
b) Religions, especially Christianity and traditional religions, have provided verifiable sources for idioms unique to NigE. This is not surprising. In a study conducted by Mgbemena (Reference Mgbemena2014), 15% of her respondents stated specifically that they acquired English from the church. She concludes that in Nigeria, Christianity has been central to the evolution of NigE. There are some idioms that obviously have their sources from religions (Kperogi, Reference Kperogi2015). Their religiosity is lost as they are integrated in popular use, since not all users of English follow Christianity or traditional religions. Shout hosanna as a Christian religious expression is not unique to Nigerians; its mundane use is, however, unique to Nigerians. In this data shout hosanna (VNN) has been used as an expression in a non-religious circumstance.
c) Idioms are also formed from world history and the classics. As a result of a fact from history, or even based on the exigencies of the time, idioms can be formed. Some other idioms are from the classical histories of previous civilisations. Some are modified versions of existing idioms. Nigerian factor (VNN) is an idiom formed to reflect the prevalence of self-imposed problems that bedevil Nigerians, such that things work imperfectly. When something fails to function properly in Nigeria, people say it is the Nigerian factor.
Conclusion
It can be said that based on their structural contents, there are three major categories of idioms in NigE. Structurally modified idioms and idioms unique to NigE present a new phenomenon and thus increase the linguistic stock of world Englishes. This is the point of reference. Even in the Standard British English there are variations in the forms of idioms. There are even structural variations among L1 varieties – like the one between American English and British English. Not all features of NigE can be identified in our idiom tokens. This is perhaps because of the limitation of the span of our data coverage.
ALIYU MUHAMMAD UMAR is a lecturer at the Abdu Gusau Polytechnic Talata Mafara Zamfara State in Nigeria. He graduated with BA in English language in 1991 from the Usmanu Danfodiyo University Sokoto, Nigeria. He also had his MA linguistics from the same university in 2005. He had his PhD in English language from the Universiti Putra Malaysia in 2016. During his over 20 years of service with Abdu Gusau Polytechnic, he has served in various capacities as head of department, director, dean, and deputy rector. His areas of interest include language acquisition, ESL, ESP, language documentation, literacy studies, theoretical linguistics, CALL, CMC, corpus linguistics and idiom studies. Email: aliyuumar2011@gmail.com
