ARTICLES

 

African language-speaking plurilingual students' writing in English: Shifting the focus from 'error' to 'variation'

 

 

Msakha Mona^I; Ellen Hurst Harosh^II

^IUniversity of Cape Town
^IIUniversity of Johannesburg

Correspondence

 

 

---

ABSTRACT

The study aims to address epistemológica! erasure by challenging the way students' knowledge is assessed in written assessments. The problem lies in assessment practices that do not consider the social and linguistic context of students' writing, particularly for plurilingual speakers of English. The study analysed a set of essays by plurilingual students to make sense of ways in which the students' backgrounds found expression in their English writing. Using the decolonial concept of locus of enunciation, the study found that features considered "errors" in English essays are often influenced by students' plurilingual backgrounds. Thus, the study argues for a shift from framing plurilingual students' writing as "error", to "variation" which would be accommodated by a true translanguaging approach.

Keywords: decolonisation, epistemological erasure, hidden curriculum, plurilingualism, translanguaging, variation

---

 

 

Introduction and background

The implementation of translanguaging pedagogies in South African higher education can be seen as a decolonising move, in its recognition of African language-speaking students' linguistic and cultural resources and its challenge to the hegemony of English as the only language of academia (Hurst & Mona, 2017; Hurst, et al., 2017; Makalela 2015). However, assessment practices by lecturers and tutors remain a challenge in translanguaging pedagogies, and assessment of plurilingual students' essays and other submissions in English remains influenced by the "hidden curriculum" - 'all that is culturally arbitrary and contingent in the curriculum, for example, the degree to which their work meets the grammatical standards of native speakers of English' (Luckett, et al., 2019: 37). Aspects of English writing such as grammar and spelling, as well as the non-compliant use of English words by students who are plurilingual speakers of English, are commonly considered "errors" by markers. And yet, a true translanguaging approach would consider departures from Standard English as variation rather than error. The research project described here analyses a set of translanguaging essays, focusing on examples of written English that depart from Standard English in the writing of African language-speaking plurilingual students.

Our analysis of these suggests that many of the constructions and linguistic elements that would be framed as errors by markers, are in fact evidence of alternative epistemologies, and reflect the linguistic practices associated with these epistemologies. From a decolonial perspective, 'the geo-political and body-political location of the subject that speaks' (Grosfoguel 2011: 5), namely the students' locus of enunciation (Grosfoguel, 2011: 5), has significant influence on why students write the way they write. This study therefore has implications for the ways that we, as lecturers, mark our students' writing and train our tutors to mark essays. In the context of decolonising the university, framing students' knowledge as errors can be considered epistemological erasure.

This article begins with an overview of plurilingualism and translanguaging in South African higher education, as well as the framing of home language interference as error. We also consider the assessment practices surrounding this framing. The next section introduces the theory of locus of enunciation as the theoretical framework for the understanding of plurilingual variation in the paper. We describe the methods of data collection and analysis, and then present the results of the data analysis, by focusing on firstly the variation identified in the analysis, and secondly on how these examples of variation demonstrate the ways that locus of enunciation is reflected in the data.

 

Problem statement

Assessment practices by lecturers and tutors in English medium universities still marginalise students who do not speak English as a first language. A prominent way in which students are marginalised is through the treatment of language variation as errors which are often highlighted as incorrect grammar. Such assessment practices miss the opportunity to consider that students who are not first language English speakers also bring their own cultural and linguistic resources into their writing processes. Some of what may be considered errors by assessors could be better framed within a decolonial translanguaging approach as variation, as students navigate, or translanguage, between English and their home languages.

 

Plurilingualism and language 'interference'

South African African-language-speaking black students are commonly referred to as multilingual, often speaking English and Afrikaans in addition to home languages such as one or more of the South African official African languages: isiXhosa, isiZulu, Sepedi, Sesotho, Setswana, siSwati, Tshivenda, Xitsonga, and isiNdebele. However, we distinguish between multilingualism and plurilingualism on the lines of Canagarajah (2009: 27) [on plurilingualism] who explains that:

For plurilingual communities, the multiple languages in their locality (including those imposed from outside) eventually constitute the same system. They are also open to one language influencing the shape and sound of the other.

The interrelatedness of languages and linguistic practices in South Africa is better reflected by the term plurilingualism than multilingualism, which tends to imply separate or discrete monolingualisms (Canagarajah, 2009). This view concurs with Blommaert (2014) and others' shift from the idea of pure, bounded, distinct languages, the units of language, dialect, sociolect, and so on, to repertoire-meaning the linguistic resources gained and held by an individual 'and deployed in highly context-sensitive metalinguistically regimented social practices' (Blommaert, 2014: 3).

However, the plurilingual resources of students are often erased in university assessment practices and framed as errors. Ward-Cox (2012) for example, investigated errors made by distance education university students in South Africa, with the aim of improving their English writing skills. Based on her findings, she recommends urgent intervention 'especially in the case of Xhosa-speaking students from rural and township schools' (Ward-Cox, 2012: 132). She attributes errors to home-language influence, inadequate schooling, and a lack of exposure to written discourse alongside possible contributing factors such as poverty, urbanisation, and dysfunctional families (Ward-Cox, 2012). She claims that students demonstrated 'little knowledge of the language conventions of the target variety', and that 'errors of sentence structure and word choice frequently impeded meaning to such an extent as to render the writing almost incomprehensible' (Ward-Cox, 2012: 132). She argues that 'language accuracy remains a key factor in determining academic success', while acknowledging debates over the value of error correction, particularly in relation to 'World Englishes' (Ward-Cox, 2012: 132). In this regard, she argues it would be pedagogically and ethically unacceptable to implement an unproven theory advocating no error correction, as this may have far reaching impacts on students and their socio-economic status. She advocates an additive approach, in which

students should be aware that their own local variety is not inferior or "deviant", but that their linguistic range is being expanded to include the academic variety of writing in order to further their academic goals, including that of meeting the present testing criteria ... [and that] developing features of SAE (which includes BSAE [Black South African English]) have not yet stabilised sufficiently to the stage of the adoption of endonormative criteria in the assessment of academic writing. (Ward-Cox, 2012: 130- 131)

Demana (2022) investigated written English language errors made by level one students in a South African university. Their findings cited errors of omission, addition, and misinformation. These were broken down further into errors relating to copula "be" and other auxiliaries, third person singular, pronoun use, prepositions, plural markers, articles, coordinating conjunctions, apostrophes, possessives and past tense markers. Demana (2022: ii-iii) states that the possible causes of errors 'were ascribed to a variety of factors including cross-linguistic differences between English and the students' L1, over-generalisation, carelessness on the part of the student, insufficient mastery of the English language system and hypercorrection resulting from the students' strict observance and over-caution regarding the English language structure'.

Similarly, Ekanjume-Ilongo and Morato-Maleke (2020) analysed errors in the essays of second year students at a University in Lesotho (an enclave within South Africa). They found that most errors were 'committed as a result of negative transfer and interference of the students' L1 (Sesotho) (71)', as well as ineffective feedback from teachers. They propose that further appropriate 'corrective techniques' should be implemented to correct student errors, namely that systematic errors should be explained and the correct forms should be provided. We view this quest for error correction as a bias towards Standard English, against which plurilingual students are assessed.

 

Variation and translanguaging

In South Africa, this view of plurilingual students' writing as error is reflective of the dominance that English enjoys in the university environment. Bangeni and Kapp (2007) highlighted this dominance of English as contributing to the amount of linguistic negotiation that African-language speaking students faced. These challenges often entailed that students struggled to embrace their "original" identities when they spoke in English. In their study, students 'codemixed between English, Afrikaans and Xhosa, interspersed with Kwaito-derived slang' (2007: 265) in a context where they strived to keep in touch with their sociocultural backgrounds while adapting to a new English-dominant university context.

Bangeni and Kapp's (2007) work falls within a well-established field problematising the deficit positioning of plurilingual African language speaking students' linguistic resources in South African universities (Gough & Bock, 2001; Liebowitz, 2005; Paxton, 2004). Part of this work has emphasised the recognition of different discourses and literacies that students draw on in their university work. For example, Paxton (2004) found that students intertwined diverse discourses such as African oral, literary and narrative traditions, biblical rhetoric, television and the internet, with more academic discourses. Similarly, Gough and Bock (2001: 12) found features of isiXhosa oral traditions in written essays: 'Perhaps the most striking feature of these essays is their similarity to spoken discourse, with long chains of clauses often making up whole paragraphs'.

These, along with other linguistic and discursive features, are part of the makeup of BSAE. Makalela's (2013) work on BSAE foregrounds demographic factors as key to ways in which South Africans have adapted English for their localised contexts. Thus, he explains that 'there are more non-native speakers than occidental native speakers of English' (Makalela, 2013: 96). Due to these demographics, the English that is being transferred among non-native speakers is influenced by forms and functions of the local languages (Makalela, 2013: 96). This phenomenon of transference of non-native forms and functions has culminated in a variety of English, hence the BSAE described by Makalela, and what Gough and Bock describe in the written form as 'the kinds of patterns [which] have become a very 'comfortable' way of speaking and writing English' (2001: 12). Thus, the variations that emerge in students' written work are also reflective of the fact that the society from which these students come has normalized using the English language in ways that are influenced by local languages.

Participants in Makalela's (2013) research also spoke English without following the rules of Standard English so as to insert meanings that are culturally located and conceptualized in the speakers' mother tongue. Although Makalela's (2013) study of BSAE looked at speech patterns as opposed to written language, the principle of local languages influencing the use of English are relevant for the issues we deal with in this study. Moreover, it is worth noting that student's writing practices are also linked to speech experiences - as we will show in some of the examples.

An alternative approach to variation, other than framing it as error, can be informed by recent sociolinguistic theory as discussed above, which views language as a process or practice which draws on the full range of linguistic resources of a particular speaker (Blommaert, 2014). In such scholarship, languages are not seen as complete, clearly demarcated entities, but rather as ideological constructions. Language in actual practice meanwhile is messy, and involves inherent variation, including practices such as choice of register, style, codeswitching, and what is described often within pedagogical accounts as translanguaging, as well as non-normative varieties of standardised languages. Translanguaging theory suggests that

rather than possessing two or more autonomous language systems, as has been traditionally thought, bilinguals, multilinguals, and indeed, all users of language, select and deploy particular features from a unitary linguistic repertoire to make meaning and to negotiate particular communicative contexts. Translanguaging also represents an approach to language pedagogy that affirms and leverages students' diverse and dynamic language practices in teaching and learning. (Vogel & Garcia, 2017: 1)

The adoption of translanguaging pedagogical approaches within South African higher education can be understood as a decolonial, social justice approach (Hurst & Mona, 2017; Hurst, et al, 2017). Makalela (2015: 15) makes a case that 'fuzziness and blurring of boundaries between languages in the translanguaging classes are (i) necessary and relevant features of the 21st century to enhance epistemic access for speakers in complex multilingual spaces, and that they are (ii) indexical to the pre-colonial African value system of ubuntu', referring to the 'ubuntu worldview of interconnectedness where one language or culture is incomplete without the other' (Makalela, 2015: 28). The ubuntu framework entails that 'speakers of languages with the "ntu" or "tho" root have a complex identity matrix of unboundedness, confluence and overlap, which is embedded in the typology of the languages they speak' (Makalela, 2015: 191). Makalela's (2015: 191) conceptualisation of the relationship between identity and language concurs with the concept of locus of enunciation, which we use in this paper to explain that language is more sophisticated beyond being a means of communicating. In other words, 'languages are a representation of the human cultural logic of being and they are therefore inseparable from the soul of their speakers'. Building on Makalela's framework of ubuntu translanguaging, we also argue that to confine the assessment of plurilingual students to one standard language, is to displace their loci of enunciation.

Luckett and Hurst Harosh (2021: 43) highlight some affordances of translanguaging as a pedagogic strategy, suggesting that it allows students to 'utilise their full linguistic repertoires to make meaning, express their feelings and judgments and draw on "cultural knowledge" through the use of their first or home languages, leading to high levels of student engagement'. Similarly, Hurst, et al. (2017) highlight how translanguaging pedagogies firstly enabled students to surface their language resources, and secondly enabled lecturers to acknowledge and value them, and in addition, students were able to utilise translingual practices to develop their understanding of key concepts.

Translanguaging pedagogies encourage learners to use resources from their full linguistic repertoires in learning contexts, with resources referring not just to standard languages (separate monolingualisms which may imply a code-switching approach) but to all "varieties" and linguistically blended practices which speakers employ. In a translanguaging framework, the influence of home languages on language practices including writing could be framed as variation rather than interference or error, and the English used by plurilingual students as varieties of English. In this way, students' translingual practices might be valued and even rewarded, rather than penalised.

 

Assessment and the hidden curriculum

Progressive approaches, such as those conceptualised under translingual pedagogies, are often not reflected when it comes to assessment practices. Assessment of essays is in general a fraught issue, with questions raised such as how to achieve a reliable score based on specific criteria such as content and grammatical accuracy, particularly between different markers. In South Africa, Gamaroff (2000) found that there was substantial variability between markers given the same criteria; some markers penalised students for spelling and/or grammatical errors while others considered mainly content. The author suggests the need for an objective standard of 'what is, for example, (good) grammar and (good) spelling, and agreement on what importance to attach to particular criteria' (Gamaroff, 2000: 31).

Lephalala and Pienaar (2008) looked at feedback on ESL student writing in South Africa and suggest that while students value feedback, markers pay little attention to assessment criteria, and instead focus on addressing errors at word level only or give non-text-specific feedback. They show that only 10% of markers in their study focused on content and organisation, which was considered the most effective feedback for the students. In a study by Jansen, et al. (2021), spelling errors negatively influenced all assessment judgements. They further suggest that an awareness by markers of the negative judgements caused by spelling errors can reduce the effect. This suggests that training might assist with teachers/lecturers' assessment of plurilingual English writing.

On the other hand, variation in written English can reflect the use of home language resources by plurilingual students to engage in meaning making, such as in the use of metaphors highlighted by Postma (2015). By focusing on the creative use of metaphors in ESL students' essays, Postma moved away from the use of only 'Standard British English' in the classroom, and instead enabled metaphors to be 'mixed or reinvented by L2 speakers, rather than viewed as inappropriate because they do not meet the conventional form of the language' (Postma, 2015: 61). Postma (2015: 60) argues that 'South African learners who speak an African language at home are familiar with a metaphor-rich cultural heritage' and that the use of these forms in their writing could provide 'encouragement and acceptance of the cultural heritage of African learners to make use of what is familiar to them when telling their stories and reporting on their life worlds, [and] could decrease the effect of unfamiliarity and alienation experienced by learners in the L2 or EFAL class' (Postma, 2015: 61). She furthermore argues that these strategies should be recognised in assessment practices. Unfortunately, she notes that '[t]eachers assessing these essays in the final exam did not indicate any acknowledgement of learners' metaphor use' and that instead markers focused on errors, structural shortcomings and logical issues (Postma, 2015: 61).

Assessments are often influenced by the perception of language errors, even when rubrics or marking criteria do not make explicit mention of particular requirements for academic English. This relates to what is described as the hidden curriculum. For analytical purposes, the three dimensions of curriculum outlined by Le Grange (2016) are useful. These are "the explicit, hidden and null curriculum" which are explained as follows:

The explicit curriculum is what students are provided with such as module frameworks, prescribed readings, assessments guidelines, etc. The hidden curriculum is what students learn about the dominant culture of a university and what values it reproduces. The null curriculum is what universities leave out. (Le Grange, 2016: 7)

We therefore locate the framing of plurilingual students' variations as errors, within the hidden curriculum, as this is the dimension of the curriculum where implicit cultural dominance affects how students are assessed. The hidden curriculum hides cultural value systems such as 'the continued dominance of colonial languages and cultures in the modern university and the delegitimisation of those of previously colonised peoples' (Luckett, et al., 2019: 37-38). This framework of the hidden curriculum is significant in that it can be used by markers to reflect on the kinds of cultural value systems that may dominate their assessment of plurilingual students.

 

Locus of enunciation as a theoretical framework

The concept of locus of enunciation, borrowed from decolonial scholarship, can be defined as 'the geo-political and body-political location of the subject that speaks' (Grosfoguel, 2011: 5). Applying this conceptual frame to students' writing challenges us to appreciate that students, too, express themselves from particular places, physically and otherwise. This is evident in the use of metaphors and discourses by plurilingual students as highlighted above in the work of Gough and Bock (2001), Paxton (2004), and Postma (2015). The case presented by Postma demonstrates that students approach their writing from the positions of their heritage, thus the metaphors that are part of these heritages form part of the student's written text. In other words, a student writes their essay from a particular geography, culture, and body. These elements of the locus are intertwined in that the geography has implications for the culture, the culture includes and affects the language, while the body affects how one expresses themselves among other things. For example, Jabulani who is a male from a Zulu family that resides in Soweto is likely to adopt the "kasi" culture alongside his Zulu culture. His Zulu language is likely to be blended with "Tsotsitaal", and his expressions are likely to include body gestures that are part of communication in the township. Moreover, his gender is also likely to influence his gestures and speech. Due to his geographical location, Jabulani is likely to speak English only as a minor part of his linguistic repertoire. Even when Jabulani speaks English, he will likely not be thinking from English, at least not in its "standard" form, nor will he have the same access to cultural references that someone from an English-speaking locus will have. Due to his geographical location that leads to a culture of translingualism, Jabulani operates between multiple cultures and languages. However, in academia Jabulani does not get rewarded for his cultural and linguistic resources, while someone from an English-speaking locus who might be proficient only in English gets an advantage due to English being the preferred language in education.

The struggle faced by Jabulani, and many other students with rich plurilingual backgrounds, is against Anglonormativity - the hegemony of English - which entails the expectation that students 'should be proficient in English, and are viewed as deviant or deficient if they are not' (McKinney, 2016: xvi). According to McKinney (2017: 42) the monolingual orientation, particularly Anglonormativity, is borne out in educational practice and policy which treats named languages as autonomous objects that must be kept separate from each other. This practice in South African education creates a hierarchy in which English becomes the most valued language.

Thus, an Anglonormative orientation to assessing plurilingual students' writing disregards all the linguistic resources that students employ in their writing, to rather point out that which students are perceived to be lacking in terms of the rules of Standard English. The loci of enunciation of students such as Jabulani, therefore, are undermined by Anglonormativity.

Anglonormativity works alongside the universalisation of the modern episteme through the modern university system, as pointed out by Grosfoguel (2013: 74). He explains that the modern disciplines generally operate through the five hegemonic (ex-colonial) European languages meaning that the onus is on speakers of non-European languages to "catch up" and adapt to, in the case of South Africa, "Anglonormativity" (McKinney, 2017), including native speaker levels of proficiency in English (Luckett & Hurst-Harosh, 2021).

In other words, the loci of enunciation that plurilingual students bring to the modern South African university are not recognized as they are geo-political and body-political locations outside the Anglo-European locus.

In order to recognize African students' loci of enunciation in their writing, we advocate for a move from assessing plurilingual students solely from the perspective of Standard English to account for the fact that plurilingual students do not 'work' English from an English locus. As Canagarajah (2009: 17) explains, 'monolingualist orientations impute deficiency to the unique strategies that multilinguals employ to work English'. He further suggests that an 'emic perspective will help explain why multilinguals adopt the forms and conventions they do in their writing' (Canagarajah, 2009: 17). Canagarajah (2009: 18) defines error 'as those items one or both members of the interaction refuse to negotiate'. The refusal to negotiate meaning lies on the side of assessors who assume Anglonormativity in assessment. However, in a context where both the student and assessor assume multi/plurilingualism, Canagarajah (2009: 18) argues that error cannot exist: 'Since multilinguals are open to negotiating difference, there are bound to be no errors in their communication'. The emic perspective to language which Canagarajah advocates for provides opportunities for student's loci of enunciation to be considered when markers decode meaning from students' writing. A translanguaging pedagogy allows for this emic perspective by allowing for and recognising students' plurilingualism.

 

Methods of data collection and analysis

In 2018 members of a research team in an Education Development Unit at a South African university established a database of translanguaging essays. The university in question is a historically white institution; the unit seeks to address/ redress historical inequalities/ coloniality in the faculty teaching and learning practices.

The essay which provides the data for this database is called 'My Language History' and elicits information on students' language backgrounds in order to surface and value students' linguistic resources. These essays commonly feature English alongside other languages in the students' repertoires. The extract below is from the essay instructions:

The language we use - both spoken and written - reflects who we are. Where we were born and grew up plays an important part in determining the language we speak. The language that our parents and family spoke is probably the most important influence when we are small, and then our friends, schooling and jobs all add to the pool of language resources from which we choose. Reflecting on your personal language background can help you to understand how your use of language has developed.

Think for a moment about the questions below and then answer them. Write it as a piece of prose, not as a list. You can write this essay in any language you like. You can switch between languages as you write - you can write one word, or sentence or paragraph in Tswana, another in English, another in Zulu, or write the whole essay in Sotho, or write the whole essay in English, whatever you prefer.

The instructions further ask the students to reflect on their language experiences and factors that affected their language across 3 stages, namely: from birth to school age, through school, work and study. Since this is the very first assignment in the beginning of the course, it enabled us to formulate 'an emic perspective' (Canagarajah, 2009: 17) into student's loci of enunciation (Grosfoguel, 2011), in so far as their loci have a bearing on the languaging processes they bring with them to the course. We label these essays as "translanguaging essays" because of the manner in which principles of translanguaging were applied by students in their writing. Although the assignment instructions did not explicitly use the phrase "translanguaging", theoretically that is what students were invited to do - as reflected in the last paragraph of the instruction extract above.

In the initial coding of the database, the authors of this article noticed several features which might be considered errors in the English of these essays, but which appeared to be influenced by aspects such as home language constructions, morphosyntax, and metaphors/idioms. To explore this proposition, the authors conducted an analysis of these variations against the home language of Nguni language-speaking students, to unpack the linguistic and contextual causes and meanings behind them. For the analysis in this paper a subset of 10 essays were used from a total of 77 essays submitted. The home language of the students in the subset was isiXhosa, and in some cases, students also identified other African languages they spoke growing up such as Setswana, Sesotho and isiZulu. We focused on Nguni home language speakers because Nguni languages formed the largest African language subgroup in the data and are the expertise of the first author. The inclusion of additional languages would have entailed further contrastive analyses, for a very small sample.

We draw on a tradition of contrastive analysis to highlight ways in which English and Nguni languages differ. The contrastive analysis was based partly on linguistic descriptions of English and Nguni languages, but also partly on our own expertise - the first author is a speaker of Nguni languages and a critical education specialist, and the second author is English and a linguist. The results are partly subjective interpretations of concrete linguistic features.

In the analysis phase, we first went through and identified variations in the English and categorized them according to linguistic feature. The features we identified were prepositions, phonology/ spelling/ tone, articles, sentence structure, idioms, and discourse markers.

We then jointly cross-analysed them from a perspective of a) Nguni linguistic structure and b) discursive meaning. In this way we further identified several variations that appeared to arise as a result of students' loci of enunciation. Finally, we also reflected on what the students themselves were saying in their essays about language and power. The analyses are presented in the following section.

 

Data analysis

In the analysis we present data to evidence variation and show some linguistic effects of locus of enunciation on student writing. In the examples below, the variations which we analysed are highlighted in bold and underlined. We first present an example of a specific category, and then offer our interpretation of it.

Example 1: prepositions

However, on my middle and high school life, all those home language lessons got to an end, as we started to be introduced in English as a first additional language and all the subjects were now taught in English.

To analyse this example, it is worth noting that the Nguni languages in South Africa differ significantly from English in the formation of prepositions. Mathonsi (2001: 172) demonstrates that prepositions are formed morpho-syntactically in isiZulu, which is part of the Nguni languages, because 'there are no prepositional phrases in Zulu, but phonological phrases, although there could be prepositions as closed class syntactic categories in the deep structure'. English, on the other hand, has more than 100 prepositions which are usually a single word that functions as the head of a prepositional phrase. It is generally acknowledged that English prepositions are difficult to learn by speakers of other languages. Hendricks (2010: 24) for example highlights a number of difficulties including 'that prepositions are often conceptually different from one language to the next, and direct translation cannot be relied on'.

In example 1, instead of on, the correct preposition should be 'in' or 'during'. However, in Nguni, which is part of the locus of this student, there would not be a direct equivalent of 'in' or 'during'. The preposition e- would be used in Nguni to denote location, for example e middle "in the middle"; e high school"in high school"; e taf/leni'On the table" e manzini'"inside water". Whereas in English, a speaker or writer has to choose between "in", "on", "at", or "during" among other prepositions, in Nguni e- works in the place of any of the English prepositions. This possibly creates an extra burden for a Nguni student who enunciates from a world where one preposition works in multiple settings, to now having to figure the correct preposition amongst many - a burden that those who enunciate from a more English background might not have.

Regarding the use of m, the correct preposition would be "to". The Nguni equivalent of "to" presents an epistemic problem in this context. Hypothetically, "introduced to" would be ukwaziswaku. However, the epistemic problem is that from the point of view of Nguni languages, one cannot be introduced to something that is not alive, so if one thinks strictly from this position, then a person cannot be introduced to a language since a language cannot know a person. Thus, beyond the "in" being a grammatical variation in English, from the locus of a Nguni speaker such an enunciation, even if correct in English, is not normal. In this example, in order for the student to write the correct English, they would have to make an epistemic jump into a scenario where people can be "known" by non-living things. This example potentially reveals that to fulfil the correct grammatical requirements, a Nguni student might need to perform some epistemic shift exercises - a process which a first language English speaker does not have to undergo.

Example 2: prepositions

By using English, especially here at the university I have a feeling that I am slowly losing touch to my Tsotsi-taal and my home language.

In the case of to in example 2, the correct standard English preposition would be "with", which would be selected from a large pool of prepositions. As previously mentioned, Mathonsi (2019) claims that prepositions in Nguni are formed in the deep structure. Simango (2019: 319) on the other hand argues that prepositions 'represent a category for which isiXhosa has a very impoverished inventory'. What both authors concur on, is that the Nguni languages do not have the same construction of prepositions that English has, and that English has a wider range of prepositional phrases whose prepositional meanings are more stable. In this case, the isiXhosa ne might be used, which is a combination of prefix na with noun prefix i - becoming ne. This combination can carry different meanings such as "and", "too", and "with" depending on the sentence in which it is found. Yet in English, the student has to choose one specific preposition that carries a more fixed meaning. This can therefore be seen as a variation in English usage influenced by the student's plurilingualism.

This raises a question of whether it is fair to expect a student who enunciates from a locus that does not have prepositional phrases to readily know the correct English preposition, particularly as the collocation of prepositions with particular words is often arbitrary rather than rule-driven. Yet, equal tacit knowledge is expected of students from differing loci. When a plurilingual student does find the correct preposition, they will not be rewarded for their effort any more than an English-language speaker will be rewarded for enunciating from tacit knowledge. Social justice theory would classify such a process as one of disparity of participation, especially against the understanding that 'the most general meaning of justice is parity of participation' (Fraser & Honneth, 2003: 16). The disparity of participation entails that participants who come from a diversity of linguistic backgrounds are expected to fulfil a fixed set of grammatical requirements based on one language that some of the participants had less access to than others.

Example 3: vocabulary/equivalency

As I know that the first word I was able to call was Mama.

The correct English word in this sentence rather than call would be "say", which in isiXhosa would be ukutsho. However, ukutsho can also be used interchangeably with ukubiza which means "to call". For example, an isiXhosa speaker can say ulibize igama or ulitsho igama, both of which can mean "enunciate the word/name" (note that "word" and "name" can also be used interchangeably in Nguni - for example, HIV is euphemistically referred to as amagama amathathu"three names" (because it is a three letter acronym), as opposed to amazwi amathathu which would literally mean "three words"). What this example points out is that the student comes from a locus where there is more flexibility and equivalence between these words, and they now have to fit within an English frame that has less equivalence between the words.

Example 4: comparative adjectives/equivalency

And because I was constantly speaking and having people around me speak English I would say I speak English the best than I would with IsiXhosa.

While the correct English in example 4 would be "better than", in Nguni we often say ngconoto refer to both "better", and "best", and the final meaning would depend on the specific sentence in question. From a Nguni perspective, ngcono carries a greater range of meaning, whereas in English the equivalent word would have to be changed if the meaning shifts (between "better" and "best"). It is also worth noting that in Nguni the concept of ngcono takes a more qualitative approach to reach meaning, while the English version takes a more quantitative approach. For example, in English you use the word "best" usually when comparing more than two things (there is 1(good), 2(better), 3(best)), whereas in Nguni, the number of the things or people being compared will not matter, what matters is only the state of quality being compared. Therefore in Nguni, one person can be ngcono than another, and one person can also be the ngcono one in a group of three. This is another example of cultural properties that are often taken for granted when students' writing is assessed - the qualitative vs quantitative ways of deriving meaning used in the different languages.

Example 5: vocabulary/equivalency

Varsity reflects the outside world and English becomes the medium of instruction, varsity is made up of a very diverse culture so to get noted you need to put your home language aside unless English.

While the correct English in example 5 would be "noticed", both "noticed" and "noted" denote recognition. Someone with tacit knowledge of English would know the distinctions between the adjective "noted" meaning "celebrated", the verb "to note" meaning "acknowledged" and the verb "to notice". This is another example that challenges markers to look for an intended meaning in students' writing. In this case, if "recognition" is the intended meaning, the student should not be penalized for what is quite a complex grammatical distinction. Yet linguistic training on the part of teachers and markers is not a standardized requirement, and unpacking the cause of this type of variant is not common practice amongst markers.

For the second highlighted term unless, in Nguni, this could be equal to ngaphand/ekwe, which can also be used in the place of "besides". From an equivalency perspective, therefore, the Nguni ngaphand/e kwe should carry the same value as either "besides" or "unless"; in which case, "put your home language aside, besides English" could be seen as grammatically correct, while "unless" leaves the sentence incomplete.

Example 6: vocabulary/equivalency

Less did I knew that my high school was mixed race which suggests that I had no alternative but to learn to communicate in English as fast as I could.

While the correct English in example 6 would be "little" rather than "less", in Nguni the word ncane can be used to mean "less'", "little", and "small". All the three English words have equivalency to the single Nguni word ncane, which creates the possibility for a Nguni writer to use these words interchangeably because they carry the same value in the Nguni locus.

Example 7: phonology/spelling

I remember one of my teachers in Grade 12 who use to teach English he would say to us as learners, "kids listen! Listen! You must not laugh as if you are uneducated, not walk like you are ill discipline and don't talk English like blacks."

This example brings in the body aspect of locus of enunciation which in this case could have a bearing on writing. While the right English spelling would be "used to", when speaking it is normal to blend the consonants d and t such that the pronunciation can sound like "use to". It is therefore possible that this student's writing is influenced by their bodily (the mouth) treatment of English, and thus written based on the sound. If the text is spoken verbally, "use to" is not likely to sound wrong especially in the South African context. This example reminds us to bear in mind that students, although they might submit written text, also enunciate from nonwritten loci, such as the body. Language proficiency, too, is not limited to written text. Thus markers need to be able to appreciate the link between writing and other loci of language proficiency. In this example, it would be helpful for the tutor to imagine what the student would sound like if they were speaking out the sentence. Although it is beyond the scope of this paper to dwell on the problematics of centring written text in assessments, it is worth bearing in mind that communication is broader than writing. Thus, markers of written text, such as in this example, should ask "what could the students be trying to communicate" as opposed to relying overly on "how is the student saying what they are supposed to be saying". The latter question runs the risk of reducing the answers to either "correctly" or "incorrectly" while the former challenges the marker to try and recognize the content that the written text seeks to represent. Markers of written work must therefore think outside writing and outside English, which is what students also do, as these examples show us.

Example 8: Idioms

I was born a Xhosa girl. Ho ya ka setso, it is "cultural" for a child to inherit the paternal name and culture,, culture is tied to language, you cannot have one without the other. So in my case my father is Xhosa and of cause speaks the language, IsiXhosa,, however he married a Sotho speaking woman, no ha se Mosotho.

In this example, the presence of an English idiom leads to variation. The correct English idiom would be "of course", which is metaphorically derived from the course or direction of a river and means "in the normal order". Course and cause can sound the same when spoken, so the student could be writing this from their spoken repository. The metaphor "of course" is tacit or cultural knowledge, while the student's variation "of cause" presumably relates to "because/ causes" and is actually a more logical spelling in terms of the meaning of "of course" as "by consequence".

Example 9: discourse markers

Even though I would say I am able to speak English, but writing it for me has always been the most challenging thing, as I have to first think in Xhosa then try to translate my thought into information that is written in English.

From a correct English perspective, but is not needed in example 9 because the sentence starts with "even though". However, there is background to why this would make sense in a South African plurilingual context: speakers of African languages, particularly in urban contexts where there is heightened language contact, often borrow discourse markers or logical connectors from English or Afrikaans and use these in the same sentence as a marker/connector that is from the African language. For example, to say "it is hot", one could say kodwa kushushu mara. The word kodwa is isiXhosa for "but", while mara, borrowed from Afrikaans (in Afrikaans it is written as "maar") also means "but". Therefore, the direct English translation would be "but it's hot but"- in essence the second conjunction becomes a discourse marker. The double use of conjunctions in example 9 is normal from this perspective. This student is therefore possibly influenced from urban isiXhosa forms.

 

Language and power

Students reflect explicitly on the role and status of their various linguistic resources in their essays, which we will briefly discuss in this section. Although students were not asked to reflect on power dynamics in the essay, the examples they shared revealed a hierarchy amongst the languages that characterised students' language histories. This was tied to the perceived benefits of being fluent in Standard English, such the "need to put your home language aside" to be recognized in the university, which is noted in example 5.

In all the essays that were analysed, English enjoyed dominance both explicitly and implicitly. English is seen to have had a hegemonic status in the students' upbringings. For example, the phenomena of talking "English like blacks" shown in example 7 above, shows English as a marker of status, which calls for alienation from blackness. The essays in general reflected attitudes to English as global, international and universal. They emphasised its utility, convenience, and usefulness as a common medium of communication, but also highlighted its dominance as a struggle or challenge which undermined student's feelings of intelligence and self-worth. English was presented as the most powerful language even in contexts that are culturally diverse, as seen in example 6 where the student attended a multiracial school. Although the school is in South Africa, diversity was not navigated through African languages, hence the student states that 'I had no alternative but to learn to communicate in English as fast as I could'.

In the essays, isiXhosa was referred to in terms of love, pride and identity, but also in terms of loss, in that students felt they were losing their home language(s) to English, or that African languages were being diluted by English. Students also described how speaking isiXhosa felt like a disadvantage (see also Hurst (2017) for an analysis of attitudes and ideologies to language in the essays).

Having established that students' language practices are a reflection of their loci of enunciation, the constant need to divorce their home languages in favour of English suggests a separation of the lifeworlds that inform their languaging. We see this as a process of disempowerment in a multifaceted way: the most direct is the struggle of shuttling between their home languages and Standard English, against the possibility of being marked down for their variations that are perceived as errors. Another dimension of the disempowerment is the demotion of their home languages as students acquire formal education.

 

Discussion and recommendations

The examples above problematise the dominance of English as the preferred language of creating meaning for students who do not come from English speaking backgrounds. This dominance of English can also be described as Anglonormativity, which 'refers to the expectation that people will be and should be proficient in English, and are deficient, even deviant, if they are not' (McKinney, 2017: 80). In other words, Anglonormativity entails ignorance of all other meaning-making processes and linguistic resources that African language-speaking black students employ that do not conform to the expected standard of English. The deficiency that is attributed to these students by an Anglonormative perspective, is complicit in the unfair assessment practices that students often face - wherein students are marked down on the basis of language rather than their ability to make meaning from linguistic resources that are available to them. As the examples above demonstrate, the students are capable of making meaning from their loci of enunciation. There are in fact layered meaning-making processes beneath the apparent errors. However, insistence on assessing such meaning solely against Standard English risks collapse of the intended meaning and subsequently hinders the students' success.

There are various ways of responding to this. On the one hand, academic development in universities could focus on sensitising and training markers (lecturers and tutors) to see examples such as those above as variation, rather than errors, and engaging with it rather than penalising such variation. Resources could be developed for markers to understand the different ways that students might write English in a university context in South Africa. Examples of types of variation such as those above would be useful in such a training guide. Assessment rubrics could also be developed that explicitly state that grammatical variants related to home language are not penalised. There can then be intentional effort on the part of markers to engage with students' submissions beyond the actual written text. Teachers and markers could create assignments that are conducive to recognizing the kinds of resources that students bring with them. Just as the examples of the essays used in this research show, when students' lifeworlds as well as language resources are invited and recognized in assignment tasks, the marker can be empowered to take a more emic approach to making sense of student's writing.

On the other hand, a more radical approach might involve a theoretical and pedagogic shift to understand variation as translanguaging in action, and rather than simply not penalising it, assessments could in fact reward students for drawing on their full range of linguistic and cultural resources. This is in fact the aim of translanguaging pedagogies, as they are 'an approach to language pedagogy that affirms and leverages students' diverse and dynamic language practices in teaching and learning' (Vogel & Garcia, 2017: 1). This would also constitute a response to calls for social and linguistic justice in a higher education system that is still rooted in its colonial heritage and refuses to acknowledge the loci from which the majority of its students speak.

 

Conclusion

Locus of enunciation as a frame through which we assess students' texts entails that we must appreciate that students come from different backgrounds - and different loci than the Anglonormative frame that modern South African universities are still bound by. We have to expect that students will bring properties from these backgrounds as they articulate themselves, and we need to find ways to accommodate and reward it. When we ask students who were not brought up in an English-speaking world to write essays in English, we are asking them to detach from their backgrounds, and they are assessed on their ability to enunciate from a place they have never been, while expecting them to perform at the same level as those who are from an English-speaking locus.

The domination of English in academia leads to plurilingual students not only getting penalised for not grasping a language that is foreign to them, but also leads to a lack of recognition of the rich linguistic resources these students bring from the languages they use daily. Those who speak English as a first language, on the other hand, get rewarded even when they are not competent in the indigenous languages of the country where their education is offered. Such a process of assessment perpetuates the dominance of English at the expense of indigenous languages.

Much of these processes of penalty and reward take place through assessments implicitly via the hidden curriculum, in judgements about students' language skills or the competency of their native-speaker production of English. By focusing on these judgements and reframing errors as variation, we have suggested a way to take into account students' loci of enunciation, the geo-political and body-political locations from which they come. The examples used in this paper illustrated that plurilingual students are certainly influenced in their English writing by their other languages, and by other factors that shape the way they make sense of the world; by geography, culture, epistemologies, and their bodies. A decolonial translanguaging framework invites students to use their full range of resources in their responses to disciplinary questions; the challenge for higher education is how to take on that emic perspective, how to recognise these resources, and how to reward them.

 

Author Biographies

Msakha Mona is a Lecturer in the Language Development Group, University of Cape Town. His work includes decoloniality and social justice through education and arts. He works through over 8 South African languages, while learning other languages. Msakha is a member of the South African National Commission for UNESCO, Culture Sector.

Ellen Hurst Harosh is a Research Associate with the Department of Languages, Cultural Studies and Applied Linguistics, University of Johannesburg. Her research focuses on African youth language practices with an emphasis on sociopragmatic approaches to peer group and online interactions, language and globalisation/migration, and multilingualism in higher education in South Africa.

 

Acknowledgements

The authors would like to thank the UNISA Curriculum Transformation Unit who supported the development of this research by organising and funding a writing retreat and webinar, and assisting with the editing of the article. We are grateful for database funding from the UAE-based research cluster Languaging and Higher Education in Bilingual Contexts, led by David Palfreyman at Zayed University.

 

References

Bangeni, B. & Kapp, R. 2007. Shifting language attitudes in linguistically diverse learning environment in South Africa'. Journal of Multilingual and Multicultural Development, 28(4): 253-269.         [ Links ]

Blommaert, J. 2014. From mobility to complexity in sociolinguistic theory and method. Tilburg Papers in Culture Studies, paper 103.

Canagarajah, S. 2009. Multilingual strategies of negotiating English: From conversation to writing. JAC, 29(1/2): 17-48.         [ Links ]

Demana, V. 2022. Aspects of written English language errors made by level-one students in a South African University. Unpublished PhD diss., University of Venda        [ Links ]

Ekanjume-Ilongo, B. & Morato-Maleke, M. 2020 Error analysis in the essays of second year students of the National University of Lesotho. Journal of African Languages and Literary Studies (JoALLS), 1(1): 71-91.         [ Links ]

Fraser, N. & Honneth, A. 2003. Redistribution or Recognition? A Political-Philosophical Exchange. New York. Verso        [ Links ]

Gamaroff, R. 2000. Rater reliability in language assessment: The bug of all bears. System, 28: 31-53.         [ Links ]

Gough, D.H. & Bock, Z. 2001. Alternative perspectives on orality, literacy and education: A view from South Africa. Journal of Multilingual and Multicultural Development, 22(2): 95-111.         [ Links ]

Grosfoguel, R. 2011. Decolonizing post-colonial studies and paradigms of political economy. Transmodernity Journal of Peripheral Cultural Production of the Luso-Hispanic Word, 1(1).         [ Links ]

Grosfoguel, R. 2013. The structure of knowledge in westernized universities: Epistemic racism/sexism and the four genocides/epistemicides of the long 16th century. Human Architecture: Journal of the Socoogy of SeA-Knowledge, 11(1): 73-90.         [ Links ]

Hurst, E. & Mona, M. 2017. Translanguaging as a socially just pedagogy. Education as Change, 21(2): 126-148.         [ Links ]

Hurst, E., Madiba, M. & Morreira, S. 2017. Surfacing and valuing students' linguistic resources in an English-dominant university. In Palfreyman, D. & Van Der Walt, C. (eds.). Academic bbiliteracies.Bristol: Multilingual Matters, 76-95.

Jansen, T., Vögelin, C., Machts, N., Keller, S.D. & Möller, J. 2021. Don't just judge the spelling! The influence of spelling on assessing second-language student essays. Frontine Learning Research, 9(1): 44-65.         [ Links ]

Krashen, S.D. 1981. Second Language Acquisition and Second Language Learning US: Oxford: Pergamon Press Inc.         [ Links ]

Le Grange, L. 2016. Decolonising the university curriculum. South African Journal of Higher Education, 30(2): 1-12        [ Links ]

Leibowitz, B. 2005. Learning in an additional language in a multilingual society: A South African case study on university-level writing. TESOL Quarteriy, 39(4): 661-681.         [ Links ]

Lephalala, M. & Pienaar, C. 2008. An evaluation of markers' commentary on ESL students' argumentative essays in an ODL context. Language Matters, 39(1): 66-87        [ Links ]

Luckett, K., Morreira, S. & Baijnath, M. 2019. Recontextualization, Identity and Self-critique in a Post-apartheid University. In Quinn, L. (ed.). Demonizing the Curriculum. Stellenbosch: African Sun Media, 23-43.

Luckett, K. & Hurst-Harosh, E. 2021. Translanguaging pedagogies in the Humanities and Social Sciences in South Africa: Affordances and constraints. In Paulsrud, B., Tian, Z. & Toth, J. (eds.). English Medium Instruction and Translanguaging. Bristol: Multilingual Matters, 43-61

Makalela, L. 2013. Black South African English on the radio. World Englishes, 32(1): 93-107.         [ Links ]

Makalela, L. 2015. Translanguaging as a vehicle for epistemic access: cases for reading comprehension and multilingual interactions. Per Linguam, 31(1): 93-107        [ Links ]

Mathonsi, N.N. 2001. Prepositional and adverb phrases in Zulu: A linguistic and lexicographic problem. South African Journal of African Languages, 21(2): 116-175.         [ Links ]

McKinney, C. 2017. Language and Power in Post-Colonial Schooling: Ideologies in Practice. New York: Routledge        [ Links ]

Na Phuket, P.R. & Othman, N.B. 2015. Understanding EFL students' errors in writing. Journal of Education and Practice, 6(32): 99-106        [ Links ]

Paxton, M. 2004. Intertextuality in student writing: The intersection of the academic curriculum and student voices in first year economics assignments. Unpublished PhD diss., University of Cape Town        [ Links ]

Postma, M. 2015. Metaphor and meaning: Analysing metaphor use in the creative work of second language learners of English. Per Linguam, 31(3): 49-64.         [ Links ]

Prah, K.K. 1999. African Renaissance or Warlordism? In Makgoba, M. (ed.). African Renaissance: The New Struggle. Sandton: Mafube, 37-61.

Simango, S.R. 2019. English prepositions in isiXhosa spaces: Evidence from code-wwitching. In Hickey, R. (ed.). English in Multilingual South Africa: The Linguistics of Contact and Change. Cambridge: Cambridge University Press (Studies in English Language), 310-328.

Vogel, S. & Garcia, O. 2017. Translanguaging. In Noblit, G. & Moll, L. (eds.). Oxford Research Encyclopedia of Education. Available at: https://oxfordre.com/education/view/10.1093/acrefore/9780190264093.001.0001/acrefore-9780190264093-e-181 (Accessed: 4 December 2024).

Ward-Cox, M.W. 2012. A critical review of language errors in the writing of distance education students. Unpublished MA Diss. University of South Africa.         [ Links ]

 

 

Correspondence:
msakha.mona@uct.ac.za and ellenhurstharosh@gmail.com
@Msakha
@msakhamona

 

 

Submitted: 08 November2023
Accepted: 22 November 2024