‘Stop Measuring Black Kids with a White Stick’: Translanguaging for Classroom Assessment

Introduction

Speakers have a wide-ranging linguistic and semiotic repertoire that they can access for various language practices. For those deemed traditionally to be bi/multilingual, including many Australian Aboriginal people, this may include the use of more than one language. However, when such speakers are students in schools, those parts of their language that do not align to ‘classroom/school talk’ are often overlooked, particularly with respect to assessment, despite the conceptual, cultural and linguistic richness that might be encompassed. In this paper, we propose that translanguaging – a fluid approach to language use – should be embraced within assessment because it addresses the inherent requirements of valid and fair assessment and provides opportunities to address current injustices. To do so, however, will require changes in teaching practices. We also document some of the social, cultural and linguistic biases that underpin aspects of assessment, and make suggestions for improvement. In particular, we examine how teachers could approach assessment so that students with Aboriginal backgrounds are able to draw on their full linguistic repertoire and, in this way, address the issues surrounding discriminatory assessment practices that are founded on monolingual mindsets (Gramling, 2016). We take up the gauntlet to ‘stop measuring black kids with a white stick’ ¹ and seek positive and embracing ways for all students with Aboriginal backgrounds to engage with assessment.

The Conceptual Understanding of Translanguaging

Recent research in applied linguistics has begun problematising traditional concepts such as mono/bi/multilingualism for failing to address the linguistic resources and cultural and semiotic repertoires that have been constructed out of a transnational diversity in late modernity (Dovchin and Lee, 2019; Gramling, 2016). The ideology underpinning these traditional conceptual frameworks seems to create the sense of utopian bi/multilingualism, which is understood through a pluralised monolingualism, rebuffing not only other linguistic possibilities, but also other identity expressions that are closely attached to other language possibilities (Pennycook and Otsuji, 2015). For some scholars, such a view of separate languages is an abstract ideological construction, which is highly questionable when it comes to describing and analysing everyday language use (Lee, 2017). These highly ideologised views of bi/multilingualism, in which the co-existence of two or more linguistic systems is central, and ‘the enumerative strategy of counting languages’ (Makoni and Pennycook, 2007: 251) seem to presuppose that there is a clear border between languages, and that these languages can be counted and classified (Dovchin, 2021; Dovchin and Dryden, 2021). The importance of opting for a more critical perspective is acknowledged by Heller (2007: 1), who provides an alternative way to understand language practices, as ‘socially and politically embedded’. For example, she suggests that the notion of bilingualism needs to shift away ‘from a focus on the whole bounded units of code and community, and towards a more processual and materialist approach which privileges language as social practice, speakers as social actors and boundaries as products of social action’ (Heller, 2007: 1). Not only is it problematic, and at times needless, to differentiate linguistic features according to specific languages or linguistic systems, but it is also evident that the smooth and fluid movement and (re)merging between languages that is often observed requires different ways of thinking and new terminologies to describe them. As Møller (2008: 218) asks:

What if the participants do not orient to the juxtaposition of languages in terms of switching? What if they instead orient to a linguistic norm where all available linguistic resources can be used to reach the goals of the speaker? Then it is not adequate to categorise this conversation as bilingual or multilingual, or even as language mixing, because all these terms depend on the separability of linguistic categories.

Nevertheless, it is also possible that the participants may orient to distinct modes and codes, or orient to distinct resources even when they use their whole linguistic repertoire to make meaning (Canagarajah, 2018). While they may see their use of codes and modalities as a mix, which we refer to as translanguaging, they also see them as distinct in situated interaction (e.g. Oliver and Nguyen, 2017). As shown by Oliver and Exell (2020) Aboriginal speakers demonstrate high degrees of language awareness as they orient to particular codes depending on audience and context. In line with this thinking, recent studies in applied linguistics have been critical of the romanticisation of linguistic plurality based on putative language counts, while refocusing on linguistic features rather than languages (Otheguy et al., 2015). Such studies suggest that it may be more useful to talk, for example, in terms of ‘translanguaging’ – an approach to the use of language, bi/multilingualism and the education of bi/multilinguals that considers the language practices of bi/multilinguals not as two or three autonomous language systems as has been traditionally the case, but as one linguistic repertoire with features that have been societally constructed as belonging to two or three separate languages (Li, 2018). Thus, translanguaging has been introduced to capture the critical complexity of language practices that are receiving greater attention in the context of late modernity, advocating for the fluid movement between linguistic codes that absorb characters and features from a wide range of semiotic resources (Canagarajah, 2018). Addressing language beyond its systematic and formal features, translanguaging tries to close the gap in the old and seemingly discrete dichotomy between the visibly ‘systematic features of the language, such as syntax, grammar, or the relatively fixed meanings of words, and their unsystematizable contexts, which interact with such stable features in any actual conversation’ (Sultana et al., 2015: 2). The focus is on both language users’ and learners’ available ‘fluid and creative adaptation of a wide array of semiotic resources,’ (Hawkins and Mori, 2018: 2–3) to make relevant meanings and to achieve one's communicative aims. Translanguaging is viewed as multilayered complex processes of entangled and intertwined linguistic and semiotic resources – the (dis)assemblages of fluid, mixed, kaleidoscopic, and semiotic resources, styles, modes, registers, acts, genres and repertoires. When Makoni and Pennycook (2007) called for disinventing and reconstituting languages, they indicated that what counts as a particular language is often negotiated by people as they still have the concept of particular codes, dialects, languages, even if they are socially constructed. The so-called languages (English, Russian, etc.) become continuously ‘dis-invented’ (Makoni and Pennycook, 2007) and ‘resemiotised … [a] process by means of which every “repetition” of a sign involves an entirely new set of contextualisation conditions and thus results in an entirely “new” semiotic process, allowing new semiotic modes and resources to be involved in the repetition process’ (Varis and Blommaert, 2015: 36). Recent studies in translanguaging are, therefore, more inclusive and all-encompassing than many other bi/multilingual studies, as they seek to account for modern language use, including the use of linguistic features, modes, registers, genres and styles (Dovchin, 2018).

While translanguaging is manifested in the (re)mixture of languages, registers, styles and symbols, it can also become ‘a product of their sociohistorical trajectories through a multitude of interactions across space and time’ (Hawkins and Mori, 2018: 2–3), which may play an ideological role in reforming and sustaining sub-cultural affiliations of identities, aspirations, class, gender, religion, demographic background, desires and so on (Dovchin and Oliver, 2021; Parra and Proctor, 2021). It treats language not as a separate code or self-standing product, but instead a gathering of meanings both spatially and temporally, within and across past and present contexts in their historical, local, discursive and interpretive elements, considering language beyond its observable linguistic features. While addressing the continual flow and location of meaning within the layers of complexity of their relations, translanguaging also puts forward social semiotics in which signs need to be understood productively, contextually and discursively (Li and Zhu, 2019).

In order to understand the fluidity in translanguaging created by the mixed codes, modes and genres and its social dynamics caused by the social, political, historical and ideological associations, we seek to understand the concept of translanguaging not so much through separate linguistic codes (though they remain useful for analysing the belonging of languages), but rather by unveiling ‘the absorption and transformation of texts by texts’ (Lesic-Thomas, 2005: 6), or ‘the voices within a voice’. We also interrogate those processes by which language learners and users engage with translanguaging to reflect their own personal, social and historical ideas, identities and identifications in relation to others’ contradictory and conflicting ideas (Liu and Fang, 2020). By doing so we can come to understand language learners’ and users’ inclination towards recycling linguistic and semiotic resources from available resources, their dexterity in bringing several voices, genres, styles into one single utterance, and their sophisticated ways of connecting the past with the present with sociohistorical references, particularly if linguistic forms are transcended through translanguaging.

Lastly, recent arguments have further insisted that the participants are indeed quite normal, unremarkable, and ordinary in their language use (Oliver and Nguyen, 2017), and that translanguaging is by no means a ‘new’ phenomenon (Dovchin and Lee, 2019). Instead, it should be understood as reflective of everyday, quotidian, basic, mundane, unremarkable and ordinary practices, rather than of peculiar, exotic, strange or unconventional practice. It is ‘neither to celebrate nor to deplore, but something to observe and examine with interest like anything else’ (Sarkar and Low, 2012: 412). Despite these claims, translanguaging has been overlooked for pedagogical assessment purposes, as discussed in the next section.

Validity and Fairness in Assessment

The integrity and quality of assessment practices are underpinned by the validity they hold. Despite the centrality of the concept of validity to effective assessment, there are various interpretations of its meaning, which have been expanded over time (Newton and Shaw, 2016). From a sociocultural perspective, principles of fairness are deeply embedded in the notion of validity and have been used to reveal the presence of cultural and linguistic biases across a range of assessment practices and processes.

Validity theory in assessment has developed over the last 100 years. Initially, the focus was content validity, which is useful for narrow, content-specific domains and is represented in most standardised achievement tests (Kane, 2013). Criterion validity was developed to measure relationships, or correlations, between the test and the criteria, and in educational assessment there was an important shift from norm-referenced testing to a criterion-based approach (Glaser, 1963). The notion of construct validity was introduced to investigate the degree to which the construct itself influences task performance and the extent to which it can measure what it is intended to measure (Cronbach and Meehl, 1955). Compared with content and criterion validity, construct validity is concerned less with the evaluation of the test than the evaluation of the interpretation of the test scores (Cronbach, 1971). Bringing these approaches together, Messick (1989) proposed a ‘unified’ understanding of validity as a combination of content, criterion and construct validities with consideration of the consequences, both intended and unintended. From this view, validity is a holistic concept that cannot be met in part: ‘To validate an action inference requires validation not only of the score meaning but also of value implications and action outcomes, especially of the relevance and utility of the test scores for particular applied purposes and of the social consequences of using the scores for applied decision making’ (Messick, 1989: 5). Thus, validity extends beyond the inferences drawn, to how they are used. Messick’s (1989) perspective differs to other propositions for measuring validity, such as Kane (2013, 2016) who advocates for an argument-based approach to validity: if the aims of an assessment are clearly stated and the evidence collected aligns with the aims, the assessment and its interpretations are considered valid (Kane, 2013).

Sociocultural and social justice perspectives in educational assessment have developed from Messick’s (1989) unified approach to validity, and highlight that many assessments do not hold validity for students of culturally and linguistically diverse backgrounds (Gipps and Stobart, 2009; Klenowski, 2009, 2014; Stobart, 2005). For Gipps and Stobart (2009) validity is founded in the notion of the ‘fairness’ of assessment practices. Assessment cannot be separated from the social, cultural and political contexts in which they were created, ‘assessment is a socially embedded activity that can only be fully understood by taking account of the social and cultural contexts within which it operates, alongside the technical characteristics’ (Gipps and Stobart, 2009: 107). The social, cultural and political contexts in which assessments are embedded advantage some groups in society and disadvantage others and these biases are inherently unfair and, therefore, threaten the validity of the assessment. Wigglesworth et al. (2011: 325–326) provide an example from Australia's national standardised NAPLAN ² test for Year 3 Reading in which students were asked to interpret a graphical representation of a poster for a film to be screened at a cinema. They argue that this question disadvantages Australia's Aboriginal children living in rural and remote communities who are unlikely to have access to cinemas. Additionally, they point out that the Standard Australian English (SAE) grammatical constructions used in the questions, such as passives, are not present in many Indigenous languages and creole varieties spoken, thus it is grammatical and cultural knowledge that is being tested rather than reading comprehension (Wigglesworth et al., 2011: 326). Another question relating to a paperboy delivering newspapers, a completely foreign concept in remote communities, serves to consolidate the point being made (Wigglesworth et al., 2011: 326). The cultural and linguistic bias present in these questions challenges our sense of fairness and poses a threat to all aspects of validity. Content validity is not met due to the unfamiliar cultural content, and the influence of language in the question does not uphold construct validity, and importantly in the context of widely publicised high-stakes testing, the social consequences are considerable. The reporting of NAPLAN results frequently positions Indigenous children in remote Australia as ‘failing’, fuelling deficit perspectives, pervasive in educational discourse, on the premise of unfair and invalid testing (Dixon, 2013; Freeman, 2013; Vass, 2012).

The negative social and political consequences of large-scale standardised assessment are not limited to the test itself and its interpretation as high-stakes testing can significantly impact on classroom practices. This occurs in the form of a ‘washback effect’, that is when the presence of the test influences the nature of teaching to align with the test (Messick, 1996; Shomany, 2011). When English-dominant monolingual perspectives underpin the testing regime, then these views become entrenched in classroom teaching and assessment practices as teachers seek to improve student performance. As a consequence, in Australia teaching and learning has become focused on SAE and, moreover, the development of literacy in this dialect to the detriment of other languages, knowledges and skills. In this way, the assessment reflects the social values of the sociopolitical context (McNamara, 2012). The negative social consequences are reinforced in the classroom as teachers seek to align their teaching with tests that perpetuate the disadvantage that is experienced in the initial testing. That is, teaching becomes aligned with testing rather than the context of the students. An example from the Australian context clearly illustrates the far-reaching negative consequences of high-stakes standardised assessment for Aboriginal students and their language practices. In 2008, following the first NAPLAN results, which showed that Aboriginal students in the Northern Territory (NT) of Australia performed well below average, the NT minister for education, in what has been described as a ‘knee-jerk’ reaction, announced that all schools would run ‘first four hours in English’ programmes, effective immediately (Devlin, 2017: 207). This drastic change was implemented after 35 years of bilingual programmes in Aboriginal languages. After the ‘first four hours in English’ ended in 2012, there was little left in terms of Aboriginal language education (Wilson, 2014).

Gorter and Cenoz (2017) argue that policies and practices must change to reflect the language practices of populations in fair and accurate ways through the adoption of multilingual assessments. Currently, the language of assessment in the Australian schooling system is limited to SAE, reflecting the nation's monolingual mindset (Clyne, 2005), which stands in juxtaposition to the cultural and linguistic realities of its population. Thus, current assessment practices, even classroom-based assessments that have the scope to ensure inclusive practices, become tests of language and in this way do not hold construct validity for those who are not proficient in the language of the test (Shomany, 2011).

As we have shown, language and assessment practices are both ‘socially and politically embedded’ and have been systematically used in education to gain social and political advantage largely for the white, English-speaking, upper middle class (García and Otheguy, 2017). While many argue that it is impossible to completely remove cultural and linguistic bias from assessment (e.g. Klenowski, 2009), improvements can be made. We propose that the concept of translanguaging provides such an opportunity. Translanguaging recognises the entirety of students’ linguistic repertoires, which can be actively employed in the classroom to promote learning (see Oliver et al., 2021 and the edited collection by Shay and Oliver, 2021). When it is extended to assessment, students are able to use all the linguistic resources available to them to demonstrate the full extent of their knowledge and understanding. In this way, assessment can better meet the criteria of validity; that is, it becomes fairer and more accurate. In this way translanguaging can serve as a theory for transformative practice in both teaching and assessment.

Translanguaging and Assessment

In this section, we argue for the adoption of a translanguaging perspective in assessment and propose two changes to current assessment practices in Australia. First, assessment practices need to provide opportunities for students to translanguage, that is, to employ all their linguistic resources to demonstrate their knowledge, understanding and skills (Menken and Shohamy, 2015; Otheguy et al., 2015). Second, incorporating translanguaging practices alone will not suffice: the modes of assessment must also change. Literacy-dominant assessment practices do not provide adequate room for translingual students to express themselves using all their linguistic resources. Therefore, the modes of expression must be realigned to allow for this to occur. We suggest transmodal assessment is required to match the realities of language practices and to uphold the principles of fair and valid assessment.

Monolingual, literacy-based assessment currently dominates assessment practices. ‘One country, one language’ ideologies permeate assessment practices and are reinforced by the ‘one test, one language’ principle. As we have shown, such practices are discriminatory and unfair and do not meet the criteria for valid assessment. Limited to the confines of one language, in this paradigm, students’ knowledge is actively silenced as they are not able to access and use all the linguistic resources available to them. As argued by García and Otheguy (2017: 61), in assessments a monolingual student can enlist the full extent of their linguistic system, whereas a bilingual is only permitted to use half, concluding that such assessment is ‘deeply and inherently biased’. They provide the example of ‘Paco’, a child who has a large linguistic repertoire of both Spanish and English, but who is judged in each of these languages separately and so his knowledge is considered deficient (García and Otheguy, 2017: 61–62). The assessment does not accurately judge Paco's knowledge, understanding and skills and, as such, does not meet the purpose of assessment. Additionally, to impose monolingual requirements in assessment stifles creativity and bounds the imagination. The large and growing field of translanguaging has described how speakers simultaneously bring together multiple linguistic resources in innovative ways, such as the creation of new words as part of the process. One example from the Australian context acts to highlight this process: the word ‘Noongaroke’ was coined to describe the adaption of ‘karaoke’, popularised in Japan, into the contemporary cultural practices of Noongar people, who are the Aboriginal traditional custodians in the south-west region of Western Australia (Haebich and Morrison, 2014). This word does not belong to the boundaries of one language and cannot be easily ascribed to one language. It is used in English and Aboriginal English, but does not bear any features of these languages, instead it represents a fusion of the traditional Aboriginal language, ‘Noongar’ and ‘-oke’ borrowed from Japanese. This serves to show how translanguaging practices can be used to more accurately describe a concept or phenomena and does so in creative, playful and innovative ways. It is envisaged that when assessment practices provide scope for translingual freedom of expression, the quality and creativity of student responses will increase.

Current assessment practices are not only monolingual, but there is also a preference for the use of words in written or spoken modes, which means for speakers of some languages the full message cannot be communicated. For example, in some Australian Aboriginal languages meaning is communicated with a combination of words and hand, body and facial gestures, which work together to create the full story. Haviland (1993) describes Guugu Yimithirr (GY), an Aboriginal language in north-east Queensland that uses cardinal directions in contrast to a left/right system centred on personal location; he explains, ‘Rather than calculating location relative to inherent asymmetries in such local objects, the GY system apparently takes as its primitives global geocentric coordinates, independent of specific local terrain, and based instead on absolute horizontal angles’ (Haviland, 1993: 6). Haviland (1993) compares video footage of a man retelling a story in 1980 and again in 1982 to demonstrate the extent to which cardinal direction is embedded in the GY systems of communication. In the first footage, the man retelling the story of the boat flipping over rolls his hands over one another to communicate an east to west direction. Footage of the same person telling the story two years later, this time in a different location, showed an outward circling movement. Haviland (1993: 16) concludes that the man adjusted his motions to align with his geographical location in order accurately depict the direction the boat flipped. This information was not spoken, instead the combination of hand gestures and body position was used to communicate this information, which, in this cultural context where cardinal direction is highly valued, can be viewed as vital information. The interactional significance of gesture in Aboriginal communities has been widely described and is often used to replace verbal information in ‘selectively covert’ (Haviland, 1993: 17) and meaningful ways (Eades, 2013; Malcolm, 2018). The dominance of SAE and, moreover, written literacy practices in schooling, assessment and society generally, threaten the unique linguistic diversity of Australia, which provides insight to the inner workings of the human mind, through the development of language and culture.

Therefore, it is not just the language of the assessment that needs to be considered, but it is also the modality of the assessment. Current assessment modes reflect the communicative preferences of the language, which may be informed by social and political values. In Australia, the dominant assessment modes of reading and writing, or ‘literacy’, reflect the current social and political views of the importance, and even superiority, of SAE. The monolingual linear equation previously described can be extended to include ‘one language, one test, one mode’ assessment designs. The same assessment cannot be rewritten to become translingual because the meaning-making practices of other languages may not fit within the confines of the assessment mode. Nor is multimodality an acceptable response to translanguaging. As a concept, it is premised on the same monolingual thinking as multilingualism: that each exists as a separate unit rather than as one expansive repertoire (Dovchin, 2018; Li, 2018). As monolingual assessment constructs limit students’ ability to demonstrate their knowledge, multilingual constructs limit students’ ability to demonstrate their understanding as they are forced to express themselves in the confines of prescribed language boundaries. For complete freedom of expression, and to be a fair and valid assessment of students’ performance, assessment must be designed to enable students to access their full semiotic system. Adopting a translanguaging perspective in assessment requires an expansion of forms and modes of assessment beyond those in which print literacy dominates, to include diverse modes of expression – those that are not present in texts and cannot or should not be replicated in other forms. In this way, assessment could become inclusive of all knowledge forms and their modes of communication. Transmodal assessment is required. We suggest that classroom-based assessment is the main vehicle that can drive this shift in perspectives.

Transmodal Assessment

Current assessment practices attempt to fix students’ knowledge or skills into a discrete singular measurable set, much like languages have been confined to distinct boundaries. However, as we have seen through a translanguaging lens, this is not how language systems or knowledge systems operate. Assessment practices need to open up and embrace transmodal freedom of expression to capture students' knowledge more fully. Building upon multimodality and multiliteracies (Kress, 2009; New London Group, 1996), we adopt the term ‘transmodal’ or ‘transmodality’ to accurately capture, and align to, the concept of translanguaging by expanding the notion of modality from multiple unitary modes to the fluid movement through modes (described in the examples below) as other scholars have (Hawkins, 2018; Stornaiuolo et al., 2017). Transmodality has been discussed in a range of contexts in relation to linguistic expression, especially in relation to translingual practices (Alim et al., 2009; Pennycook, 2007). Traditionally, assessment design has been represented as multimodal, although there is growing interest in the conceptual development of ‘transmodal’ practices (Govender, 2020; Newfield, 2014; Tomlinson, 2015). Transmodality is not only deployed as ‘a way of thinking about language use as located within multiple modes of semiotic diffusion’ (Pennycook, 2007: 44), but it also suggests that certain modes cannot be viewed as discrete items outside other meaning-making practices, for example, ‘bodies, texts, contexts and histories in which they are embedded’ (p. 49). Transmodality thus points to the ways in which translanguaging meaning occurs across fluid modes of meaning-making in ways that transgress fixed ideologies in discrete channels.

What could transmodal assessment look like? Any concept can be explored and represented in different spaces, times and places as expression moves through different modes and across the expanse of an individual's entire linguistic repertoire and/or semiotic system. In assessment, an individual could express their knowledge and understanding using drama, music, singing, narration, art in any medium, gestures and hand signals in written and oral forms in a range of languages, standard and non-standard. The modes could be layered, intwined and infused. Each mode could interact with the other and be used purposefully to communicate and demonstrate the student's knowledge, understanding and skills. The assessment task is an opportunity for students to demonstrate their current knowledge, as well as their ability to use their linguistic and semiotic resources to communicate effectively, that is, to make meaning. Indigenous youth are already engaging in such practices (e.g. within social media – see Oliver and Nguyen, 2017), and current educational technologies make it possible to bring these practices into the classroom by creating authentic transmodal assessments.

Using ethnographic research methods situated outside of institutional settings such as schools, Kral and colleagues (2011, 2014, 2016, 2019) describe how Indigenous youth in remote communities are rapidly adopting digital media technologies to engage in new forms of cultural production across different modes of expression, seamlessly blending and infusing these old and new modes. Kral (2011: 7) describes precontact modes of complex interaction including translanguaging practices of Aboriginal societies:

Prior to contact with Anglo-Australian society, Ngaanyatjarra, Ngaatjatjarra and Pitjantjatjara people had inhabited the Western Desert for thousands of years using a complex of multimodal communication forms and semiotic systems to convey meaning; through language, sign, gesture and gaze, special speech styles and registers, non-verbal communication and the iconic representations found in body painting, carved designs and sand drawings.

Technology is now being harnessed by youth as a vehicle to express themselves using their complex semiotic systems. With exposure to a wide range of Western genres and modes of communication, Aboriginal youth have an expansive meaning-making repertoire at their disposal to express themselves in innovative and creative ways, while at the same time documenting and preserving their culture. Kral (2016: 71) explains, ‘Indigenous youth are the markers of social change and new influences are shaping the multimodal literacy practices they engage in. In these resource rich communicative ecologies young people are employing multiple modes of communication. They are drawing on traditional communication styles integrated with new embedded literacy traditions’. An example of transmodal practices is shown in the description of young girls telling sand stories on an iPad screen that captures their finger movements digitally:

The film recordings we made between 2013 and 2018 with ten young women who transferred the traditional sand storytelling practice to iPads have injected new life into this traditional narrative form. The films burst with colour, energy and originality, and we see traditional iconography merging with contemporary symbols as the young storytellers recount stories of trips out bush collecting traditional foods with humorous memories of flat tyres and seeing scary animals. Other stories reveal the contemporary pastimes of young people playing football, softball and going to the disco. (Kral et al., 2019: 43)

Another example shows youth transforming cultural stories into new modes of communication:

In the contemporary songs of young Ngaanyatjarra musicians we also see a performative process of creation and renewal, and an indication of the enduring importance of the verbal arts in the lifeworlds of young people. Here, direct and indirect references to the tjukurrpa [Dreaming] abound, translating intangible concepts into tangible objects in the form of CDs and music videos. Young musicians are using digital technologies to lay down new narrative forms. (Kral et al., 2019: 43)

With technological development, the locus of control has shifted from institutional settings to communities (Kral, 2014), providing youth with the tools to express themselves as they see fit, in translingual, transmodal cultural productions. This is a picture of rich, purposeful and creative Indigenous literacy practices that are not portrayed in public discourse nor universally celebrated in schools (Kral, 2014), and particularly as a form of assessment. Instead, Aboriginal students continue to be measured by a ‘white stick’ – limited to the constraints of SAE and the written practices of white, often middle-class Australia. As we have argued earlier, policy and practices, particularly the language of assessment, need to represent the linguistic realities of the populations subject to these practices (Gorter and Cenoz, 2017). Street et al. (2009: 195) argue ‘print literacies’ that currently dominate school practices are not keeping pace with the global realities of contemporary communication dominated by digital technologies. To embrace the translanguaging perspective in assessment, to truly capture an individual's knowledge, understanding and skills, transmodal assessment is required to encapsulate students’ communicative realities – their expansive linguistic and semiotic systems, which currently remain a source of untapped potential.

Limitations and Future Directions

The obvious limitations that exist are the education systems that protect and maintain standard monolingual language ideologies, which filter into the classroom, including assessment and teaching practices. Despite this, there has been a recent move to incorporate a translanguaging approach in assessment (Baker and Hope, 2019; Heugh et al., 2016; Lopez et al., 2017; Schissel De Korne and López-Gopar, 2021). However, these efforts remain limited to specific contexts and modes, most commonly reading, speaking and writing, but they have not yet embraced transmodality in their design. A translanguaging approach to assessment requires a significant shift in teacher perceptions and practice. Schissel et al. (2021: 347) report how teachers struggle when attempting to reconcile two divergent themes: the pressure to produce monolingual-like language and the potential to validate students’ multilingualism; Schissel et al. (2021) further highlight teachers’ conformity to national policies and entrenched practices. Kumaravadivelu (2016: 81) argues that these ‘top-down’ approaches act to maintain hegemonic forces and entrench colonial perspectives, but can be countered using ‘bottom-up’ approaches to teaching and learning through the design of context-specific teaching strategies and materials. Teachers need to develop their own agency to decolonise English language teaching and empower local communities (Kumaravadivelu, 2016: 81). We argue that this also needs to extend to assessment. However, a significant gap exists both in research and teaching/assessment practices, and there is a need to explore how a translanguaging approach can be adopted to inform the development of transmodal assessment to ensure that assessment is used in the classroom to evaluate students’ knowledge and understanding accurately and fairly.

Conclusion

Until educational policies and practices fairly represent the linguistic and semiotic practices of Australian Aboriginal students, it is important that practitioners feel empowered to adopt and enact translanguaging approaches in their pedagogy and assessment of their students. Translanguaging approaches move beyond fixed language boundaries and homogenous ideologies to represent the true nature of human communication. The true nature of language use is not only characterised by the verbal expression and limits of a single standardised language code, rather it is designated with different forms of expression through drama, music, singing, narration, and art, using hand gestures and signals, in written and spoken mediums, as well as within both standard and non-standard language forms. Therefore, current ‘socially and politically embedded’ standardised language and assessment practices, which are used to perpetuate sociopolitical advantages for the white, English-speaking, upper middle class (García and Otheguy, 2017), sustain the discrimination of Australian Aboriginal students. Such forms of assessment do not allow teachers to address the needs of their individual students, but rather are simply a pragmatic match for decontextualised assessment policies and standardised tests (e.g. NAPLAN). Although we would like it to, we understand that NAPLAN is unlikely to change, at least not in the short- to mid-term. Therefore, we advocate for changes to classroom-based assessment to be inclusive of students’ entire linguistic and semiotic systems. Using this approach in assessments would enable students’ knowledge and understanding to be fairly and accurately assessed.

With this in mind, we argue for a transmodal freedom of expression in assessment practices to cater for the needs of diverse individual students and to capture their knowledge to its full extent. The assessment tasks should provide opportunities for students to show their current knowledge, and to do so using diverse linguistic and semiotic resources. Furthermore, we suggest that current assessment practices do not hold validity for Australian Aboriginal students (Gipps and Stobart, 2009; Klenowski, 2009, 2014; Stobart, 2005) and perpetuate deficit perspectives leading to negative social and political consequences (e.g. abolition of bilingual education programmes for Aboriginal students and the move to English-only approaches in response to NAPLAN results). Therefore, transmodal assessment practices are much needed to provide students with the tools to express themselves in translingual and transmodal ways, as is appropriate to their diverse backgrounds.

Embracing and acknowledging transmodal assessment practices is particularly important for the diverse context of Australia, where Aboriginal students are continuously being assessed based on Standard Australian English language norms, led by predominantly white cultural values. Such ways of assessment neglect the cultural backgrounds, knowledge, traditions, and living contexts of those students, as they are asked to perform on something that is different from their reality. It is important to emphasise that policy and practices, particularly the language of assessment, need to represent the linguistic realities of the populations subject to these practices. Translanguaging and transmodal assessment allows all students to have an equal opportunity to fully express their knowledge using diverse translingual linguistic and semiotic systems in various modalities.