Some syntactic features of Swazi English - Wiley Online Library

c:/3d/weng/arua.3d ± 26/5/98 ± 18:42 ± Sharon

World Englishes, Vol 17, No. 2, pp. 139±151, 1998.

0883±2919

Some syntactic features of Swazi English ARUA E. ARUA* ABSTRACT: The paper describes some stable syntactic features of Swazi English. It is thus an extension and elaboration of the discussion in Kamwangamalu and Chisanga's characterization of Swazi English. They point out that the wh-word is placed last in question formation, time expressions such as Ì met him last of last week' occur and idiomatic expressions such as `To see once is to see twice' are used. This paper discusses, among others, the use of the modal auxiliary must, the use of as to, the conflation of the emphatic do with the simple past tense and dangling modifiers as other stable syntactic features of Swazi English. It then elaborates on the use of idiomatic expressions, since in Kamwangamalu and Chisanga's discussion only a few were mentioned. Finally, as there are few descriptions of Swazi English, the paper advocates more studies on the subject.

INTRODUCTION

In a study published over a decade ago Y. Kachru (1985: 230) puts the research on what have been termed `New Englishes' in perspective when she identifies some of the major issues in such research: (a) Basic research: descriptive and empirical studies in different regions of the world at international, national, regional and local levels, in relation to roles, functions, attitudes, and expectations; indication of factors that ensure intelligibility and interpretability at the international level; development of models and research techniques appropriate to such investigations. (b) Applied research: studies of the implications of international vs regional distributions in language learning/teaching; development of frameworks and methodologies for teaching with new goals in mind; and development of curricula, materials, tests, etc.

In many countries in the African continent where the New Englishes exist, much basic research has been done (see, for example, Bamgbose, 1982; Chishimba, 1983; Magura, 1984; Chisanga, 1987). In some others such as Swaziland ± the focus of this paper ± there is a paucity of descriptive research. Kamwangamalu (1994, 1996) and Kamwangamalu and Chisanga (forthcoming) are the only studies in the field. The main reason for the paucity of descriptions of English in Swaziland must be the assumption that the descriptions and discussions of English in South Africa encompass English in Swaziland, a landlocked kingdom almost entirely encircled by South Africa. If a pan-southern African view is adopted, as Magura (1984, 1985), among others, has done, then English in Swaziland, South Africa and all the anglophone countries in the southern African region must be related. However, it is also true, at least initially, that the Englishes found in different politically and socioculturally independent entities like Swaziland or South Africa are, to some extent, unique, and therefore, must necessarily be different in certain respects from other Englishes found in other anglophone states. This view implies, rightly, that English in Swaziland is different, in some respects, from the Englishes found in other southern African states, notwithstanding the historical and linguistic ties of these states. * Department of English, University of Botswana, Private Bag 0022, Gaborone, Botswana. E-mail: [email protected] # Blackwell Publishers Ltd. 1998, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA.


140

Arua E. Arua

Swaziland is a small, stable, bilingual society. English and siSwati, rightly described as òfficial languages of unequal status' (Kunene, 1997), are the only languages available, sometimes simultaneously, for all official/formal and non-official/non-formal functions in the kingdom. However, English occupies a more elevated position, as Kunene has rightly observed. SiSwati is the main language of culture, as the indignation which greeted King Mswati III's use of English for the first time during the Incwala (a ritual in which the king is strengthened and rejuvenated) in 1997 shows. English, on the other hand, is the main language of education at all levels, except in the first two years of primary education (Undated Ministry of Education of Swaziland Document, p. 2). As Swaziland is basically monocultural, has one indigenous language, uses English as the main medium of instruction and as the language of many of its official functions, the English used in the kingdom has tended to be similar across social classes, and across formal and non-formal educational levels. Unlike complex linguistic and cultural societies such as Nigeria with over 200 languages and ethnic groups (Akindele and Adegbite (1992: 57) among others), and South Africa with nine official languages and an equal number of ethnic groups, it is fairly easy to identify the phonological, lexical and syntactic characteristics of the use of English in Swaziland. Kamwangamalu and Chisanga (forthcoming) have discussed the form and functions of English in Swaziland. With respect to syntax, generally regarded as the defining feature of different Englishes, they identify three features (pp. 12±14). The first is question formation in which the wh-word is placed last: (1) You are leaving when? (When are you leaving?) (2) People are how? (How's everybody?)

The second is idiomatic expression: (3) To see once is to see twice (Once bitten twice shy) (4) He has the liver of a crocodile (He is brave)

And the third is time expression: (5) I met him last of last week (I met him the week before (last)) (6) You can expect me between today and tomorrow (You can expect me today or tomorrow)

Their discussion of the syntactic features listed above is illuminating, but limited, not surprisingly, because the features are discussed alongside the functions of English in Swaziland and Swazi English lexis. The present study, therefore, discusses syntactic features not mentioned in Kamwangamalu and Chisanga, and extends the discussion of idiomatic expressions in their paper. The data for the study have been collected from various sources. The most important of these have been the Times of Swaziland (TOS) and its sister paper, Swazi News (SN), and The Swazi Observer (TSO). Another source is the responses to questionnaires (RQN) administered by the University of Swaziland sociolinguistics class of 1996. The questionnaire sought examples of the use of English at the levels of phonology, lexis and grammar, although only the syntactic examples are used in this study. All the questionnaires administered were returned. In addition, data were collected from the University of Swaziland third- and fourth-year English language students' continuous assessment essays and examination scripts (SSC) of courses in discourse analysis, psycholinguistics A Blackwell Publishers Ltd. 1998


Some syntactic features of Swazi English

141

and sociolinguistics. Finally, the present writer's observations (WO) and other texts specified in the paper also formed part of the data. The examples collected from students' essays occurred at least once in 75 percent of the scripts. The examples from the newspapers were not as frequent. The examples which were obtained mainly from the National, Home, or Community News pages of the newspapers occurred at least once in half of the items or stories examined for the study. The examples attributed to the writer's observation and to other texts in the paper were the most frequent. The examples occurred several times in the spoken English of Swazis of all classes. In discussing the examples, Standard English is used as the norm. The term Standard English has been used in two senses (Quirk et al., 1972). The first is in the sense of national standards of English. American and British Standard Englishes are much more institutionalized than other national standards such as Scottish, Irish, Canadian, Australian and New Zealand which are in various stages of institutionalization. The second, the sense adopted for use in this study, is that of World Standard English (simply Standard English in this study), which, as Quirk et al. (1972: 16) put it, is à single standard of English throughout the world, across a multiplicity of political and social systems.' In this sense, Standard English is seen as `supra-national, embracing what is common to all' the standards (Quirk et al., 1972: 17). The use of Standard English as the norm in this study is motivated, since Swazi English has come under various influences of which the most widespread in order of magnitude are British and American Englishes and cultures. Errors (in the non-Standard English sense) are not included in the discussion. Other examples, such as the omission of articles, the omission of prepositions or their replacement, which have been well documented in other Englishes are also not included in the discussion. DISCUSSION

The discussion is in two parts. Syntactic features are discussed first, and idiomatic expressions last. Syntactic features The modal auxiliary `must.' The use of the modal auxiliary must dominates that of other modal auxiliaries in Swazi English. It has two meanings. The first, as ascribed to it in Standard English, is that of compulsion or necessity. Therefore, a sentence such as: (7) You must do it.,

implies that no choice has been given to the addressee. This is the conventional meaning. Consider, however, the following sentences: (8) Must I do it? (WO). (Should I do it?). (9) Sir, the gardener says you must come (WO). (Sir, the gardener says you should come).

The Standard English phrasing for (9) should be either `Sir, the gardener would like to see you' or `Sir, the gardener wants to see you.' The point though in examples (8) and (9) is that must is used as a synonym of should. A Blackwell Publishers Ltd. 1998


142

Arua E. Arua

The following excerpt from a document (p. 3) on the implementation of an Academic Communication Skills Unit/Centre at the University of Swaziland, prepared by a local academic, not only highlights the frequency of the use of must, but also shows that the grammatical feature, generally found in the spoken form, has now crept into the written form: (10) 2. IMPLEMENTATION OF ACADEMIC COMMUNICATION SKILLS (ACS) UNIT (LATER: CENTRE ± see Bint 1993: 12) 2.1 STAGE 1: 2.1.1 BUDGET (April: 1995) The budget for Academic Communication Skills must be separated from that of the Department of English as from April 1995. The ACS Co-ordinator should receive his/her monthly statements through the Dean's office like all Heads of Department in the Faculty of Humanities. 2.1.2 BUDGET FOR LIBRARY BOOKS ETC. (APRIL 1995) The library should provide money for Academic Communication Skills so that books, journals etc. that are relevant to Academic Communication Skills may be ordered. The ordering of Library books etc. should start immediately if this has not been happening. The ACS staff must do it. This budget must also be separated from the vote of the Department of English. 2.1.3 AVAILABLE EQUIPMENT (MARCH 1995) The Academic Communication Skills Section must submit to the Bursar the list of equipment that was donated under ACS section. These should include the serial number, the make and where it is housed or kept.

Must is used in this manner throughout the document. If Standard English is used as the norm, then all occurrences of must in the excerpt would be incorrect. However, this is not so in Swazi English, because must is a synonym of most of the other modal auxiliaries. Indeed, the use of must, especially in the spoken form, appears to be one of the major grammatical elements that define Swazi English as a variety in its own right. Non-Swazis initially feel disoriented when it is used in its all-encompassing synonymous/hyponymous sense. However, like everybody else in the kingdom, they learn to use context where and when necessary to distinguish its various meanings. As must has largely suppressed other modal auxiliaries that serve various functions including politeness marking, and obscured the power relations that such auxiliaries are used to show, the question is: what language structures are used to achieve them? It should be clear then that the concepts of politeness and power as found in Standard English would have to be redefined in Swazi English. Indeed, several scholars (for example, Chisimba, 1982; B. Kachru, 1982; Keenan, 1976) have rightly raised doubts about the universal applicability of the cooperative principles as well as other discourse and sociolinguistic theories. They suggest the need for a revision of these theories to accommodate the discourse functions performed by must and other grammatical/language structures yet to be discussed in this paper, and similar structures that many other studies have shown to be important. Preference for `to have' `to get' in place of `to have.' In its spoken form, Swazi English prefers the perfective to have to get form in the place of the simple present tense of the verb to have: A Blackwell Publishers Ltd. 1998



143

(11) Error analysis and contrastive analysis have got (have) a negative attitude towards errors (SSC). (12) If you have got (have) a product to sell, contact Swazi TV marketing department (Swazi TV Olympics advertisement). (13) The other one said if language A has got (has) a label for certain things, . . . (SSC). (14) The Republic of South Africa has got (has) a lingua franca (SSC). (15) For example, South Africa has got (has) 9 official languages, which include English, Afrikaans, Zulu, Xhosa, etc (SSC).

Each of the has/have got examples above can be substituted with has or have. Although both forms are correct and are attested in other varieties of English, including native speaker varieties, the frequency of occurrence of the has/have got form indicates an overwhelming preference for it. Reported questions. Swazi English inverts the auxiliary verb and subject in reported questions as opposed to Standard English. Consider the following examples (the Standard English versions are in brackets): (16) When I asked one of them what was he looking for (what he was looking for) in the bush, he told me he lost a plastic bag containing soap and E200 (TOS 17/6/96 p. 32). (17) In subjects like literature, geography and others we gain nothing and one wonders why doesn't the education ministry include (why the education ministry doesn't include) new subjects which will help us . . . (SN 15/6/96 p. 10). (18) This was disclosed at a press conference by Education Minister, Solomon Dlamini, when responding to a question on how will government cover up for lost time (how government will cover up for lost time) (TOS 10/7/96, p. 32). (19) The thematic structure enables us to establish what is the passage all about (what the passage is all about) (SSC). (20) The secretary general Saladin Magagula wondered where will the government get (where the government will get) the terminal benefits for all the teachers since they are claiming that they do not have money (TOS 10/7/96, p. 32). (21) Ask your mummy can you go with us (WO).

There are two observations to be made regarding the examples above. The first is that there is subject-verb inversion in all the examples. The inversion appears to be a strategy for keeping the original questions in view. In other words, the inversion makes the original question easily recoverable. The second observation is that, given certain examples, it might be thought that the construction here is awkward, if Standard English is taken as the norm. This is true of example (21), where there seems to be an omission of if or whether and the auxiliary can is not marked for the past tense. Notice, however, that the insertion of if or whether in (21) would make the sentence ungrammatical. The omission, therefore, appears to be deliberate, as a grammatically awkward sentence is better than one which is completely ungrammatical. A Blackwell Publishers Ltd. 1998


144

Arua E. Arua

The use of às to.' The following examples show the ways in which às to' is used in Swazi English: (22) This essay will attempt . . . to consider as to which of these adequately explains errors in second language acquisition (SSC). (23) Dlamini also wanted to know from Makhubu as to what was the Times of Swaziland doing to develop and educate its reporters (TOS 14/6/96 p. 2). (24) As a result we do not expect any confusion as to who is the chief of the area as he is well known to all the residents (SN 15/6/96 p. 5). (25) The Minister also said he was puzzled as to why the students are camped at Bishop Ncamiso Ndlovu's house as this residence has no connection with the Ministry of Education (TSO 10/7/96, p. 7). (26) It (The Committee) said that as to how they will be compensated is an administrative decision that falls squarely on the shoulders of the Ministry of Agriculture in consultation with the Cotton Board (TOS 10/7/96, p. 32).

If the meanings and phrasing of the sentences in (22)±(26) are considered, only examples (24) and (25) would be grammatical according to the norms of Standard English, as they express two of the four meanings of às to,' namely: concerning, on the subject of, according to and by, listed in the Longman Dictionary of Contemporary English. For sentences (22), (23) and (26), the use of às to' would be considered superfluous in Standard English. This is because its deletion in the three sentences would not have any adverse effect on meaning. Now, consider the consistency of occurrence of as to in the environment of wh-words in the examples. It becomes obvious that its use is not arbitrary. On the contrary, there appears to be a grammatical rule being applied that can be stated thus: prefix as to to wh-clauses occurring as the object of the main verb or as qualifier in the object noun phrase of a sentence.

It should be clear by now that these supposedly aberrant grammatical occurrences do not arise as a result of there being no `clear semantic basis for a grammatical convention' (see Y. Kachru, 1985: 226); they are rule governed and therefore can be systematically described. Conflation of the emphatic `do' and the simple past tense. The conflation of the emphatic `do' and the simple past tense is found in the spoken form of Swazi English. The following are some examples: (27) My teacher did beat me (WO) (My teacher beat me). (28) He did do it (WO) (He did it). (29) I did go to the butchery (WO) (I went to the butchery). (30) She did take the book (WO) (She took the book). (31) If you did receive the copy directly from Germany, please ignore this (Memo from Dean's office at Uniswa).

In Standard English, the co-occurrence of the verb do with another verb shows A Blackwell Publishers Ltd. 1998



145

emphasis. It functions to remove doubt as to whether an action was performed or not. Consider example (33) which is a response to the question in (32): (32) Did he do it? (33) Yes, he did (do it).

In Swazi English, do is used not only in the sense of (32), but also generally in place of the simple past tense. In this sense, it is similar to the grammatical preference shown by the variety for has/have got rather than the simple present has/have. Dangling modifiers. One of the prescriptive rules of English is that modifiers should be placed near the objects they modify. When they are not so placed, the modifiers are said to be dangling. Therefore, the following italicised examples should be dangling modifiers, if Standard English is regarded as the norm: (34) The money is in respect of prices of goods sold and delivered to him during last year, which he has failed or refused to pay (SN 29/6/96, p. 3). (35) The government must respect the King's National Council and not take issues to court which are already before the King (SN 29/6/96, p. 3). (36) This was revealed by a cattle herder at the Manzini Pound who was giving evidence before the Commission of Enquiry (SN 29/6/96, p. 2). (37) Government yesterday said she is still uncertain about how much to pay teachers and civil servants in salaries who were engaged in the recent strike (TOS 17/7/96). (38) Some reference sources can be accessed through electronic media such as `The Encyclopedia Britanica' (Uniswa IDE ACS draft document, Unit 2). (39) It is very common that certain sources are in high demand but in short supply, such as past exam papers, rare books etc. (Uniswa IDE ACS draft document, Unit 2)).

The clausal and phrasal modifiers italicized in (34)±(39) respectively modify money, issues, cattle herder, teachers and civil servants, sources and sources. However, the modifiers are not placed near the structures they modify. There are a number of observations to be made regarding the structures. First, they are widespread in Swazi English. Second, the sentence in (36) is ambiguous. It is not at all certain whether the Commission of Enquiry took placed at the Manzini Pound or whether the cattle herder resides there. This, however, is not remarkable, since ambiguity is an inherent part of language. Third, apart from (36), the semantic content of all the other sentences is clear, regardless of where the clausal and phrasal modifiers have been placed. And fourth, the structures are attested even in native speaker English as this example from a British scholar living in Swaziland shows: (40) Ask each other questions, and try to explain things to them which you understand (Uniswa IDE-ACS draft document, Unit 14).

Example (40) raises the question of whether the native speaker has been influenced by English in Swaziland or whether he already had the structure as part of his native speaker repertoire. The extent to which native speakers living in stable second language situations are influenced by the Englishes spoken in such places, or the extent to which they are A Blackwell Publishers Ltd. 1998


146

Arua E. Arua

willing to learn the Englishes, should be investigated. In conclusion, the so-called dangling modifiers are a stable characteristic of Swazi English. Tag questions. There seems to be only one dominant tag question isn't it in Swazi English. In this connection, it behaves like must, which has already been discussed. Now, consider the following examples: (41) The brakes are on, isn't it ? (WO) (42) The announcement provided a lot of excitement, isn't it? (WO) (43) He came to the house, isn't it? (WO) (44) The University will reopen very soon, isn't it? (WO)

The tag question isn't it manifests a couple of characteristics which make it unacceptable to some people. The first is that there is no number distinction between the tag question and main clause verbs, as are and is show in the sentence in (41). The second is that the proper tag forming main or auxiliary verb (did or would) is not selected in the sentences in (42)±(44). This tag question selected, that is, isn't it, is not unique to Swaziland, however. Todd (1982), for example, considers it a general characteristic of non-native varieties of English. It occurs in Chinese, Indian, Nigerian, Singaporean English, and in many other non-native varieties of English. In addition, Bamiro (1995: 202) shows a related tendency in the use of not so? in East African English, as his example, from wa Thiong'o's I will Marry When I Want shows: (45) You were the one who said we should cook food for the visitors, not so?

The tag question isn't it appears to be an imprecise translation of the siSwati word angitsi. Angitsi contains the negative particle an, the pronoun gi and the root of the verb of saying tsi. It can be fully translated as: Isn't what I have said correct? As siSwati has only one tag question angitsi (apart from its synonyms), it must have been convenient not only to substitute isn't it? for the full translation above, but to use it as the dominant tag question in Swazi English. The use of am . . . ing. A spoken characteristic of Swazi English is the use of the am . . . ing form: (46) I am having a new car (I have a new car) (RQN). (47) I am asking the way to Dr Dlamini's place (Could you show me the way to Dr Dlamini's place?) (RQN). (48) I am asking Tom (Is Tom in? or Can I see Tom?) (RQN). (49) I am failing to send the message (I couldn't send the message) (RQN). (50) I am asking for a chair (Can I have a chair?) (RQN).

These constructions are translations from siSwati into English. For example, am failing is translated as ngiyehluleka and am having as ngine. The constructions therefore involve, mainly, the use of am and the progressive aspect of lexical verbs. They are used not only to A Blackwell Publishers Ltd. 1998



147

show the progressive, which is the way such constructions function in Standard English, but also for various other functions, as the sentences in brackets in (46)±(50) show. àm asking' is used to make requests. The requests could be for information as the sentence in (48) shows, or direction (47) or even to obtain an object as (50) shows. àm having' in (46) is used in the sense of ownership. This kind of possession is shown by have in Standard English (I have a new car). It is also used in the sense of `holding,' for example, Ì am holding a party'. In the context in which it is used, àm failing' surprisingly means that an action has been completed. Standard English would use `have failed' in its place, for example, Ì have failed to send the message.' From the foregoing, it is clear that the constructions have peculiar usage in Swazi English. What is not clear is whether all lexical verbs entering into the àm . . . ing construction are also used in peculiar ways. The data are not sufficient for a generalization to be made. The use of `borrow'. In the spoken form, Swazi English uses borrow in the place of lend. Examples include: (51) Will you borrow me your pencil? (RQN) (52) Please borrow me your pen (RQN). (53) Please borrow me E5 (RQN). (54) Borrow me your book (RQN).

Bamgbose (1982) considers borrow, as used above, an error which should be separated from the more authentic characteristics of the New Englishes. The use of borrow has, however, persisted. Perhaps, it should also be considered as a stable feature, since second language speakers of English in Africa now select it deliberately, regardless of whatever negative connotations it has. The use of ìn as far as.' Swazi English sometimes adds in to Standard English's as far as. The following are examples: (55) He disclosed that in as far as his organisation is concerned, the DPP did nothing wrong because his suspension was invalidated by the High Court (TOS 6/8/96, p. 2) (56) He claims that in as far as this feature is concerned, there is a sender and a receiver (SSC). (57) The Kingdom of Swaziland and the Republic of South Africa have different cases in as far as languages are concerned (SSC).

A similar occurrence is in as much as: (58) However, we must all know that our country does not have enough resources to have the squad at training in as much as I would have liked (TOS 9/8/96 p. 34).

The in in in as far as and in as much as appears to have been added to the two phrases for emphasis. It appears to be an instance in which a preposition is inserted where it should not otherwise have been. Although its origin has been difficult to establish, it is generally becoming a stable characteristic of Swazi English. Idiomatic expressions A discussion of the idiomatic expression below now follows. A Blackwell Publishers Ltd. 1998


148

Arua E. Arua

(59) This leaves no one in a forest whenever someone speaks (SSC). (60) You must hold the sun not to set (Problems will bring you back) (SSC). (61) Who am I with a stomach like that of a bird's gizzard (I will not require or need much) (SSC). (62) He died of a clay pot (SSC). (63) I have no mouth (SSC). (64) He became a chicken with a mouth cut (He could not utter even a single word) (SSC). (65) He makes himself a person by me (He has a negative attitude towards me) (SSC). (66) They caught me a bull (I was robbed) (SSC). (67) He has the liver of a crocodile (He is brave) (SSC). (68) The horse kicked him on the chest (He can't keep a secret) (SSC).

The meanings of the idiomatic expressions in (59), (63), (64) and (67) are transparent. The transparency, however, is not in terms of deducing their meanings from their constituent parts, but in terms of their being more easily understood in the contexts in which they have been used. (59), `This leaves no one in a forest whenever someone speaks,' is derived from translating the siSwati word kubasehlatsini (to be in a forest). `To leave one in a forest' in a conversation is to confuse the person. The expression is equivalent to Standard English's lost (for example, `You lost me') when it is used in a similar context. (63), Ì have no mouth,' and (64), `He became a chicken with a mouth cut,' are derived from similar translations: Angisena (I don't have) mlomo (mouth) (I don't have (a) mouth). Inkhukhu (chicken) ijutjwe (cut) umlomo (beak) (The chicken's beak is cut).

Both expressions imply inability to speak, squeak, squawk, squeal and so on, as a result of the absence of a mouth or beak. The Standard English version of these two local idioms is to be at à loss for words.' (67), `He has the liver of a crocodile,' is translated from Kuba (have) nesibindzi (bravery, liver) sengwenya (crocodile)

The word nesibindzi is polysemous; it means bravery or liver. Se in the expression sengwenya is the siSwati equivalent of the similarity marker like. In view of the foregoing, two translations of the siSwati expression: Kuba nesibindzi sengwenya are possible. The first is an idiomatic translation: `To be brave like a crocodile' or `to have bravery like a crocodile.' The second translation, the one elicited from the data (see 67), is literal. This translation, however, has to be interpreted to get the meaning: Às brave as a crocodile.' This idiomatic expression is used extensively in Africa, although heart rather than liver is used in many places. It has been difficult to establish why liver instead of heart is used in Swaziland. It should be noted that `crocodile' is not an essential part of the expression. The liver of any fearsome animal can be used. This is in consonance with how heart is used; it is simply said that someone has or has no heart. It is remarkable that Swazi English idiomatic expressions replicate not only the A Blackwell Publishers Ltd. 1998



149

transparency of idiomatic expressions in Standard English, as the foregoing discussion shows, but also its opacity as the following discussion shows. The examples in (60), (61), (62), (65), (66) and (68) are Swazi English expressions whose meanings are not easily deducible from their constituent parts and the contexts in which they are used. (60), `You must hold the sun not to set,' is used mainly to indicate the inevitability of an offender, especially a child, getting his/her deserts. The only way the offender can escape punishment is to cause the sun not to set, for when the sun sets, the offender is forced to return home where he/she is punished for his/her offence(s). `You must hold the sun not to set' means `problems (for example, the setting of the sun) will bring you back.' Translated from Ubolibamba lingashoni (to hold the sun not to set)

the expression may also be used in contexts in which a person is forced to return to an option which he/she rejected earlier. It is doubtful that Standard English has an equivalent idiomatic expression. (61), `Who am I with a stomach like that of a bird's gizzard,' is derived from the siSwati sentence: Sisu (stomach) semhambi (traveller) singang (equals) engingila (gizzard) yenyoni (bird).

This literally translates into À traveller or stranger's stomach is equal to a bird's gizzard.' From this is derived the idiomatic expression `Who am I with a stomach like that of a bird's gizzard?' The meaning of the expression may be given as Ì will not require or need much, since I am a visitor' or À traveller is ``satisfied'' by a small quantity of food.' It is clear from the discussion of (61) that the first two translations or expressions operate at different levels of meaning ± the literal for the first, and the metaphorical for the second. It is this fact of the second expression being different from the first (both of which are derived from the same siSwati words), and of the second expression needing interpretation that shows unequivocally that it is idiomatic. (62), `He died of a claypot,' is translated as follows: Kufa (to die) lufa (crack) lwembita (claypot) (to die the death of a claypot).

The translation is idiomatic, not literal, since `crack,' the second verb in the siSwati expression, is not accounted for in the translation, and `the death' which appears in the initial translation in English has been elided. The expression itself centres on the irreversibility of some misfortunes. If a misfortune is irreversible, then it is comparable to death ± a broken claypot from the traditional Swazi viewpoint cannot be mended or repaired. `To die the death of a claypot,' therefore, is to be overtaken by a terrible and irreversible misfortune. (65), `He makes himself a person by me,' is derived from the siSwati sentence: Utenta (make) umuntfu (person) ngami (me) (He is making himself a person out of me).

The expression is used when one person ridicules another. The person being ridiculed diminishes in stature while the person ridiculing gains. The context in which the idiomatic expression is used is normally public, especially when somebody who wants fame or public attention thinks that the best way to get it is to attack and ridicule someone who is greater than he/she is in stature. A similar explanation is contained in Kamwangamalu and A Blackwell Publishers Ltd. 1998


150

Arua E. Arua

Chisanga. The expression is the equivalent of Standard English's `riding on somebody else's back.' The example in (66), `They caught me a bull,' is derived from one of the practices associated with the Swazi Incwala. After delivering lusekwane (plant used in the Incwala ritual) at the Royal Cattle Byre, it is required of the boys to demonstrate their daring and bravery by catching and killing a bull (inkunzi) without weapons. This art and act of catching a bull is now used to describe the daring with which robbers (the boys) rob or steal (catch) their victims (bulls). The similarity extends to the fact that the robbers operate without weapons. The robbery must have taken place and the robbers gone before the victims realise they have been robbed. The last idiomatic expression to be considered is (68), `The horse kicked him on the chest.' It is derived from: Kubabete (not to have) sifuba (chest) (To have no chest).

The traditional Swazi view is that secrets are stored in people's chests. A horse kick, or more appropriately, a donkey kick to the chest, will open the chest and cause the person in whom the secrets were confided to spill them. In other words, such a person is incapable of keeping secrets. A related Standard English equivalent is `to spill the beans' ± an indiscreet or unintentional divulsion of information. The discussion has highlighted a number of things. First, it has shown that the meanings of some of the idiomatic expressions can be directly accessed, and that others have to be interpreted in order for their meanings to become clear. This is the transparency and opacity mentioned in the discussion. Second, the discussion has shown that all the idiomatic expressions are translated from siSwati into English. Therefore translation is a major method of creating them in English. This is not surprising, for as Kachru, Y. (1985: 228) rightly notes, it is now obvious that the first language exerts a lot of influence on the Englishes spoken in stable second language situations. Third, all the examples used in this section are legitimate Swazi English expressions, regardless of how they originated, and the awkwardness (basically arising from translation) of some of the expressions. The discussion has also shown that there are Standard English idiomatic expressions that are equivalent to the Swazi English expressions used in this study. Why are alternative Swazi English formulations used when existing Standard English idiomatic expressions could easily have been used? The answer is not far to seek. Language is used to express a people's culture ± way of living, of knowing, of appreciating, of speaking and of doing things generally. Therefore, like the many examples that have been cited in many other studies, Swazi English idiomatic expressions show that the English used in Swaziland rightly describes the Swazi sociocultural environment. CONCLUSION

The paper discussed, among others, the use of the model auxiliary must, the use of as to, dangling modifiers, the conflation of the emphatic do and the simple past tense, the preference for has got form, the inversion of the subject and verb in reported questions and certain socioculturally determined idiomatic expressions, as stable characteristics of Swazi English. As the paper shows in places, some of the features occur mainly in the spoken form, some other occur mainly in the written form and yet others are used in both media. Some A Blackwell Publishers Ltd. 1998



151

of the features cited here are not unique to Swaziland, as they have been attested in other parts of southern Africa, in East Africa and in West Africa. This attestation of similar features in a multiplicity of second language situations argues for the validity of the features and therefore of the varieties of English to which they belong. This point, which has been made in numerous other papers, bears re-emphasising. It remains to be said that much more work needs to be done in the context of Swaziland, in order for the Swazi variety of English to be better described. REFERENCES Academic Communication Skills Department, University of Swaziland (1995) Proposal for the implementation of Academic Communication Skills Centre/Unit. Akindele, F. and Adegbite, W. (1992) The Sociology and Politics of English in Nigeria. Ile Ife: Debiyi-Iwa. Bamgbos.e, Ayo.. (1982) Standard Nigerian English: issues of identification. In The Other Tongue: English Across Cultures. Edited by Braj Kachru. Urbana, IL: University of Illinois Press. pp. 99±111. Bamiro, Edmund O. (1995) Syntactic variation in West African English. World Englishes, 14(2), 189±204. Chisanga, Teresa (1987) An investigation into the form and function of educated English in Zambia as a possible indigenized non-native variety. Ph.D theses, University of York. Chisimba, Maurice (1982) African varieties of English: text in context. Ph.D dissertation, University of Illinois at Urbana-Champaign. Kachru, Braj B. (ed.) (1982) The Other Tongue: English Across Cultures, Urbana, IL: University of Illinois Press. Kachru, Yamuna (1985) Discourse analysis, non-native Englishes and second language acquisition. World Englishes, 4(2), 223±232. Kamwangamalu, Nkonko M. (1994) siSwati-English codeswitching: the matrix language principle and linguistic constraints. The South African Journal of African Linguistics, 14(2), 70±77. Kamwangamalu, Nkonko M. (1996) Sociolinguistic aspects of siSwati-English bilingualism. World Englishes, 15(3), 295±305. Kamwangamalu, Nkonko M. and Chisanga, Teresa (forthcoming) English in Swaziland: form and function. In English Around the World: Focus on Southern Africa. Edited by V. D. Clerk. Rhodes University, South Africa. Keenan, E. O. (1976) The universality of conversational postulates. Language in Society, 5, 67±80. Kunene, E. C. L. (1997) Official languages of unequal status: the case of SiSwati and English. Paper presented at the 2nd World Congress of African Linguistics, University of Leipzig, Leipzig, 27 July±3 August 1997. Longman (1987) Longman Dictionary of Contemporary English. London: Longman. Magura, Benjamin J. (1984) Style and meaning in African English. Ph.D dissertation, University of Illinois at Urbana-Champaign. Magura, Benjamin J. (1985) Southern African black English. World Englishes, 4(2), 251±256. Ministry of Education of Swaziland (n.d.) The Nine Year Programme of Instruction for English. Government Printer. Quirk, Randolph, Greenbaum, S., Leech, G. and Svartvik, J. (1972) A Grammar of Contemporary English. London: Longman. Todd, Loreto (1982) The English language in West Africa. In English as a World Language. Edited by Richard W. Bailey and Manfred GoÈrlach. Cambridge: Cambridge University Press. pp. 281±305. (Received 15 February 1997.)

A Blackwell Publishers Ltd. 1998