Pre-service teacher investment through dialogic action learning

and the literacy practices they are expected to take on (Boughey & McKenna 2016). Not unlike many other academic literacy courses in South Africa, the course under discussion is increasingly plagued by a lack of resources, such as experienced teaching staff for approximately 400 to 600 education students in the Senior Phase (SP) and Further Education and Training (FET) phases annually, as well as time available for the subject on the timetable. The challenge, therefore, was to design the course in a way that allowed for a rich learning environment, despite the lack Background: A lack of student ‘buy-in’ and engagement are often major obstacles in academic literacy courses. To create a dialogic learning environment which encourages student investment and challenges traditional student-lecturer hierarchies of power, the curriculum of a first-year academic literacy course at a South African university was reconceptualised around an action-learning project. Objectives: The aim is to determine whether the reconceptualised course enabled dialogic learning that fostered a sense of investment in students. Method: An interpretivist paradigm was followed, drawing on a qualitative research approach. To explore the nature of student investment, discourse analysis was used to analyse group reflections submitted at the end of the 21-week course. Student reflections were coded thematically using an inductive approach. Results: These reflections indicate that specific mechanisms need to be in place for effective dialogic engagement. If in place, findings suggest that the dialogic approach could encourage critical thinking, help students to develop problem-solving skills, lead to cognisance of multiple perspectives, deepen understanding of course material and expectations, promote inclusivity, and encourage reflection on the learning process. Conclusion: This study indicates that purposefully embedding a dialogic approach into a curriculum through purpose-driven group activities, can lead to more engaged learning. Contribution: The paper contributes to the field of academic literacy studies by showing how academic literacy practitioners may use the now-established pedagogies of action-learning and dialogic teaching and learning to design courses that create an enabling environment for students to draw on deep approaches to learning.


Introduction
A significant challenge in an increasingly 'massified' or bloated higher education system, with dwindling resources (Msiza, Ndhlovu & Raseroka 2020), is creating effective learning environments which are likely to foster deep approaches to learning (Biggs 1999).Generic academic literacy courses which are aimed at empowering students with the academic literacy competencies needed to successfully engage with their studies are particularly plagued by low student motivation (or, as will be argued later in this paper, investment), with students failing to see how these courses relate to the rest of their studies (Van de Poel & Van Dyk 2015).This, in turn, is more likely to result in surface approaches to learning and a lack of transference of these abilities to students' other subjects, because they do not integrate these practices into their current identities (Darvin & Norton 2018;Donovan & Erskine-Shaw 2020).Even discipline-specific academic literacy courses often still fail to engage students, who might display low levels of investment (Darvin & Norton 2018), because of the incongruence between students' identities and the literacy practices they are expected to take on (Boughey & McKenna 2016).
Not unlike many other academic literacy courses in South Africa, the course under discussion is increasingly plagued by a lack of resources, such as experienced teaching staff for approximately 400 to 600 education students in the Senior Phase (SP) and Further Education and Training (FET) phases annually, as well as time available for the subject on the timetable.The challenge, therefore, was to design the course in a way that allowed for a rich learning environment, despite the lack Background: A lack of student 'buy-in' and engagement are often major obstacles in academic literacy courses.To create a dialogic learning environment which encourages student investment and challenges traditional student-lecturer hierarchies of power, the curriculum of a first-year academic literacy course at a South African university was reconceptualised around an action-learning project. of resources.The aim is twofold: firstly, to equip pre-service teachers with a working knowledge of literacies, literacy practices, identities, and of how teachers should be cognisant of these when developing lessons; and secondly, to help these student teachers expand their own academic literacy competencies so as to navigate their own studies more successfully.These competencies include writing wellstructured paragraphs around main ideas, writing effective introductions and conclusions, employing reading strategies when engaging with academic texts, interpreting task words, improving the coherence and cohesion within writing, synthesising information, critically engaging with the literature, developing descriptive and argumentative writing, and applying appropriate referencing conventions.Therefore, the aim with the course is to impact both students' current identities as pre-service teachers who can participate effectively in academic discourse, and their future identities as practising teachers.The focus of this paper is not on these academic literacy competencies per se, but rather on determining which aspects of the course pedagogy (description follows) facilitated dialogic engagement, and on investigating the nature of this dialogic interaction.
In 2022, I reconceptualised the curriculum of this course around an action-learning project, previously described in Fouché (2022).I did this to promote engagement, foster an environment that encourages deep approaches to learning, challenge traditional student-lecturer hierarchies of power, and create a space that might lead to greater student investment.The pedagogy of action-learning implies 'learning from and with each other in small groups or "sets" from action and concrete experience in the workplace or community situation' (Zuber-Skerritt 2011:5).It entails the principles of the empowerment of students; student-driven learning with the minimum interference of facilitators, addressing real and complex problems; working in unfamiliar settings; and reflecting on experiences (Dilworth 2010).Zuber-Skerritt, Wood and Kearney (2020) added additional principles of action-learning, namely communication between participants to build relationships, commitment to the actionlearning project, gradually emerging competence, compromise, and negotiation, collaboration, and critical self-reflection.
The action-learning project around which the pertinent curriculum was centred, worked as follows: in groups of approximately four, students were required to plan for (in the form of a proposal) and create a YouTube tutorial on a topic in one of their major teaching subjects (for example mathematics, science, life orientation, isiZulu or English), aimed at South African high school learners 1 .Figure 1 indicates how the course was structured from Term 1 to Term 4. As illustrated in Figure 1, the proposal was the assessment artefact used to assess students' understanding of literacy-related concepts at the end of the second term of their studies, and their application of academic literacy 1.Note that in the Fouché (2022) paper, the action-learning project was conceptualised as students doing workshops with groups of students.After piloting this, it became evident that it was logistically unfeasible, and that replacing the workshops with YouTube tutorials had the affordances of reaching a wider audience over a longer period.
competencies, as discussed earlier in this section.Term 4 took the form of a reflection report: during this term, students were guided to reflect on various aspects of the actionlearning project, while simultaneously applying all the academic literacy competencies that had been introduced throughout the year.The course purposefully moved away from assessing by means of the traditional academic essay for two reasons: firstly, assessment in higher education increasingly favours a genre approach according to which students are required to write across a range of the genres in which they are likely to come across later in their studies and beyond.Secondly, both genres (namely the proposal and reflective report) drawn on, still require skills similar to those needed in essay writing, including writing effective introductions and conclusions, writing coherent and wellargued paragraphs, synthesising information, and integrating secondary literature (related to core course concepts, but also to justify various pedagogical choices in their proposals and support statements in their reflections).
The aim with this paper is to answer the following research questions: 1. Which aspects facilitated dialogic engagement in groups?2. What evidence emerged from student reflections about the nature of learning through dialogic interaction during the action-learning process?
Although dialogic teaching and learning can happen at various levels, this article focuses specifically on the dialogic teaching and learning that happened between students in their small project groups (thus, how they taught each other, and learned from each other).Evidence is drawn from students' reflective reports (see Figure 1, Term 4), which were analysed using a critical discourse analysis.
As introduction, this paper is situated in a social constructivist theoretical framework, complemented by theories on dialogic teaching and learning.Thereafter, the key concepts of surface/deep approaches to learning are investigated, as well as investment juxtaposed with motivation.Then the methods and research design are discussed, followed by a description of the data analysis.Thereafter follows a combined findings and discussion section.The paper is concluded with a reflection on the implications of this study for praxis.

Social constructivism
In my exploration of the power of dialogic teaching and learning between students, I drew on the social constructivist theoretical framework, as conceived by Vygotsky.Beck and Kosnik (2006) argue that social constructivism is built on five key principles (the first two relating to constructivism in general).Firstly, students construct their own learning, or, as argued by Dewey (1916:46), 'education is not an affair of "telling" and being told, but an active and constructive process'.Secondly, for knowledge to become concrete, it must be experience-based.Thirdly, and of particular importance to social constructivism, learning is a social process.Vygotsky (1962) saw learners as social beings who are active in their own learning within specific historical, social, and cultural contexts.He was particularly interested in the relationship between language and thought processes and noted the importance of dialogue with others when constructing knowledge.Well-developed class communities have been cited as having a positive influence on learning such as creating opportunities for democratic dialogue (Brophy 2002).Fourthly, social constructivism assumes that all aspects of a person, including attitudes, emotions, values, and actions, are connected, and intricately linked to knowledge construction.Finally, this theoretical framework posits that learning communities must be inclusive and equitable.Students must 'develop a sense of dignity and their own ideas and way of life' in a way that does not oppress non-mainstream groups, but rather forms 'inclusive communal groups with a strong sense of solidarity' (Beck & Kosnik 2006:14).Hence the pedagogy of action-learning is well suited to this research project.In line with social constructivism, it draws on the social nature of learning, emphasises the role of the student in knowledge construction, stresses the importance of communication, and assumes the gradual construction of knowledge and the resultant emergence of competencies.

Dialogism and dialogic pedagogy
Closely related to the social constructivist framework, I draw on Bakhtin's conception of dialogism and Freire's work on dialogic pedagogy.Holquist (1990:xlii) writes, 'Dialogism conceives knowing as the effort of understanding, as "the active reception of speech of the other".'Bakhtin (1984) states that: Truth is not born nor is it to be found inside the head of an individual person, it is born between people collectively searching for truth, in the process of their dialogic interaction.(p.110) Therefore, it is in interpersonal discourse that a deeper understanding of reality is conceived.
Freire's conception of dialogic pedagogy is equally relevant to this discussion.Although Freire focuses on dialogue between the teacher and the student, the tenets of this pedagogy are equally applicable to the dialogic teaching and learning that could happen between students.Shih (2018) identifies three premises of dialogue in Freire's (2005) work, namely that: (1) each person must have the right to speak, (2) dialogue is not one person 'depositing' ideas into another, and (3) dialogue must not be a polemical or hostile argument, in which it becomes 'a crafty instrument for the domination of one person by another'.Dialogic pedagogy, as conceived by Freire, therefore relies on equality of the participants, and on the breaking down of power relations that lead to the dominance of one participant over another -intricately linked to the inclusive and equitable communities to which Beck and Kosnik (2006) refer in a social constructivist framework.Another important aspect of Freire's (2005) work is the emphasis on problem-posing education to oppose banking education (the second point above), in which students become 'critical co-investigators' in the learning process -attributes shared by an action-learning pedagogy.Lyle (2008) points to considerable evidence which indicates how dialogic engagement in collaborative groups benefits cognitive development.Co-investigators who become the subjects of their learning journey, therefore benefit significantly from this process in terms of their own (deep) learning.
In designing this academic literacy course around an actionlearning project which drew strongly on dialogic pedagogy, I attempted to create an environment which would encourage the disruption of 'existing hierarchies of knowledge, repositioning students as knowers and reconfiguring what counts as knowledge' (Mendelowitz, Ferreira & Dixon 2022:59-60).Like the narrative pedagogy these authors follow, an action-learning project also has the potential of creating a multi-voiced space that fosters engagement.It is important to note that for this iteration of the course, dialogic learning was not scaffolded; rather, its use was built into the design of the course, in that students were required to meet, interact, and negotiate.
Deep and surface approaches to learning Marton and Säljö (1976)

Course structure
Term 4: The end-goal of this term is a report in which each group reflects on its learning process.This includes the challenges and successes of creaƟng the tutorial and of the acƟon-learning process, linked to relevant literature.Students are required to reflect on the role of literacies in the challenges the learners experienced in the selected topics.The report should reflect an understanding of how the outcomes of students' other first-year subjects relate to each other, and should showcase the implementaƟon of the academic literacy abiliƟes acquired throughout the year.In addiƟon to this, students will be required to individually reflect on their own contribuƟon to the project.
Term 2: Students are divided into groups based on their elecƟve subjects.The end-goal of the term is for groups to submit an academic proposal for their planned YouTube tutorial, including theoreƟcal reasoning about why they selected a specific topic for the tutorial, and what pedagogy they will draw on.They will also have to consider the literacies and cultural capitals they will bring to the classroom, as well as those of their leaners.
http://www.rw.org.zaOpen Access sharp contrast to surface approaches, which are characterised by the passive acquisition of knowledge, memorisation of disconnected information, and unreserved reliance on the instructor as the ultimate authority (Biggs 1999;Entwistle 2017).In contrast, students embracing deep approaches to learning embark on a journey of inquiry, critical examination, and active engagement with knowledge, which brings about 'conceptual change in students' understanding of the world' (Biggs 1999:60).Entwistle (2017:3) contends that 'effective university teaching hinges on establishing a connection between subject matter and the strategies employed to encourage students to grapple with ideas and cultivate their own comprehension'.This aligns with Biggs' (1999) assertion that although certain students may naturally gravitate toward either surface or deep approaches to learning, educational instruction can be intentionally structured to promote and facilitate deep learning.That was the intention behind the design of the course under discussion.
In a previous article, I illustrated how the academic literacy course was designed to encourage deep approaches to learning (Fouché 2022).In the current paper the interest is to see whether evidence of such deep learning, linked to student investment, could be found after implementing the curriculum, or, as hypothesised in Fouché (2022:78), whether our students engage with the course as 'doing teaching, rather than [just] having knowledge'.The actionlearning project around which the course is designed, requires students to extend their knowledge into different contexts (new genres of writing, e.g.report writing, as well as new modes, e.g.YouTube videos), while also drawing on other subjects, and simultaneously integrating those particular knowledges into the context of the actionlearning project.
In the next section, I argue that drawing on deep approaches to learning, and transferring knowledge, are intricately interlinked with the concept of student investment.

Motivation versus investment
The term 'motivation' is widely used in education research as a variable which influences student learning.In the mid-1990s, Norton suggested the concept of 'investment' as an alternative lens through which we may consider student engagement.While the term 'motivation' derives from cognitive and psychological theories which see students as being either inherently motivated (or unmotivated) due to a variety of factors, the term 'investment' is proposed as a sociological construct which is linked to students' current and future identities (Darvin & Norton 2018).These identities are multiple and ever-changing, and are not exclusively inherent, but are also shaped by external forces such as race, social class, and ethnicity.Darvin and Norton (2018) propose that investment signifies students' commitment to the goals, practices, and identities within the language learning process.Investment highlights the complexity of learners' engagement with language learning and their negotiation of power dynamics (Darvin & Norton 2018).The distinction between motivation and investment in language learning underscores the dynamic interplay between learners' identities, power relations, and their engagement with language learning practices.
In this paper, and in line with scholars like Boughey and McKenna (2016), I take the stance that if students are not invested in acquiring academic literacy practices and conventions, they are unlikely to follow deep approaches to learning, and that the result will invariably be surface approaches to learning.For them to become invested in integrating the academic literacy practices they acquire in the academic literacy classroom, these need to be integrated and aligned with their current identities as students, and their future identities as teachers.Generic academic literacy courses make such an alignment exceedingly difficult -if students do not see how these academic practices relate to the rest of their studies, and their future careers, they are unlikely to integrate them into their identities.In an ideal world, academic literacy would be taught in an integrated manner, in the same classroom (and even by the same lecturer) who teaches other content subjects (Wingate 2018); this integrated approach would lead to greater investment from students.However, logistical institutional constraints make this type of academic literacy instruction impossible in various contexts, which is why many universities aim towards academic literacy courses which raw on the content and conventions of students' respective disciplines (Carstens 2014).I argue that this must be supplemented with pedagogies that create an environment which encourages deep approaches to learning.Purposefully integrating dialogic learning into the design of the course, as shown in this paper, can go far towards reaching that goal.
Next, I discuss the research methods and design, as well as the results and the discussion of the data collected for this study.

Research methods and design
Following an interpretivist paradigm, this study draws on a qualitative research approach.As indicated in Figure 1, the final term of the course was built around a range of reflective exercises, culminating in what we termed a reflective report which was then assessed.Students were asked to reflect, as groups, on the feedback they received on their research proposals, the effectiveness of their YouTube tutorials, and how they experienced the group dynamics of the actionlearning project.They were asked to highlight aspects that they were proud of, or that worked well, as well as the challenges they experienced.Assessment criteria, which were provided at the onset of this term, made it clear that students would be assessed on the quality of their reflections, as well as their ability to apply academic literacy conventions to their writing, and students were encouraged to provide honest reflections on their experiences.
Students were not directly questioned about dialogic learning.However, upon reading their reports, I realised that dialogic learning emerged as a valuable learning mechanism in students' reflections.Therefore, I extrapolated from students' responses to make deductions about their experience of the dialogic teaching and learning occurring within groups.Accordingly, reports were analysed inductively to determine emerging themes related to dialogic learning unfolding from students' reflections.
The study was conducted in a B.Ed. programme at a South African university, on an academic literacy subject aimed at first-year teachers-in-training in the SP or FET Phase.Approximately 90% of this cohort consisted of Black students, with the remaining 10% consisting of Coloured, 2 Indian and White students.The 2022 cohort, which is analysed, consisted of 620 students, divided into 12 tutorial groups (referred to as TGs going forward) of between 50 and 52 students.These TGs were formed around combinations of students' two major subjects, to ensure that students from similar disciplines were grouped together.Each of these groups of students was further divided into project groups (referred to as PG going forward) with an average group size of four studentsstudents chose their own group members.
Because of the vast amount of data generated, data analysis was restricted to five of the 12 TGs.Groups were selected to ensure a spread in: (1) disciplinary focus in the TGs, and (2) tutors tutoring the groups, to ensure that reflections from TGs taught by a range of tutors were analysed.The four tutorial groups consisted of a range of students from the disciplines of Social Sciences, Languages (English, Sesotho, and isiZulu), Natural Sciences, and Mathematics, and were taught by four different tutors.Ethical clearance was applied for and approved by my university.Across the five TGs, 44 PGs gave permission for their responses to be analysed.Although marked differences between students who gave permission for their writing to be analysed, and those who did not, were not noticed when feedback was provided on these submissions for course assessment purposes, it is possible that PGs in which all members provided permission for their reflective reports to be analysed, might have been more cohesive and effective groups, which in turn might have skewed the analysed data.This is a limitation in the current study.

Data analysis
Data were only analysed after students had received their final marks for the course, to ensure that no student felt coerced.Only PGs in which all participating group members gave written consent were considered for analysis.No student was identified by name, and both TGs and PGs were re-labelled to further ensure anonymity.Direct quotations are verbatim unless otherwise indicated.Though most quotations are contained in figures, they are indicated in italics where included in the text.
2.The term 'Coloured' as a racial categorisation is commonly used in South Africa to refer to mixed-raced groups, indigenous South African (Khoi or San) people, and South Africans from Malay descent (Andrews 2018).It is not a derogatory term within the South African context.
Trustworthiness (Elo et al. 2014) was ensured in various ways.Firstly, I have attempted to clearly outline my methodology and data, and an audit trail is, within the confines of the conditions of the ethical clearance, available upon request.Secondly, my research assistant and I had coded the data separately, after which we had discussed our coding, to ensure inter-coder reliability.Thirdly, both coders agreed that data thematic saturation had been achieved after analysing the data from PGs who had given permission for analysis.Finally, I made a concerted effort to constantly maintain reflexivity, and my own positionality in the research.
There were two main categories into which the themes (shown in bullets) were organised.
Category 1: The mechanisms that students put in place to navigate effective dialogic learning: • the use of online group chats for effective communication • an overt commitment to participation.
Category 2: The positive outcomes of dialogic learning: • affordances of drawing from multiple perspectives • the whole being more than the parts • development of multiple competencies • the impact on students' future identities.

Mechanisms that students put in place to navigate effective dialogic learning
A factor that strongly emerged from students' writing was the need to put mechanisms in place to regulate communication and interaction between group members.Two themes were identified: students' use of online group chats, and an overt commitment to participation.

The use of online group chats for effective communication
Fourteen of the 44 analysed groups specifically commented on their use of online group chats for effective communication, or as PG 3D puts it, as an additional forum for [face-to-face] meetings with peers.
Unsurprisingly, most students used WhatsApp groups to remain in contact with group members.Group chats were particularly useful for facilitating effective communication, and for collaboratively constructing knowledge (cf.Vygotsky 1962) by discussing and completing tasks outside of in person meetings (see Examples 1, 2, and 4 in Figure 2).The affordance of this forum as a mode for frequent communication was highlighted by several groups, as illustrated by Examples 3 and 5 in Figure 2.An interesting aspect that emerged, was that for many students the forum created a democratic space (cf.Brophy 2002), as in Example 5 in Figure 2.
This forum for addressing 'difficult' things (Example 3, Figure 2) seemed to work particularly well because on the WhatsApp group, where they interacted daily, we all had an equal voice when it comes to decision making and contributing ideas.Everyone was allowed to voice out their ideas (…).Because we were correcting each other's mistakes, that showed social capital (Example 5, Figure 2).It is interesting to note that students can use core concepts from the course, such as 'social capital', when speaking about the interaction on the WhatsApp group, indicating a deep engagement with course concepts, and the ability to transfer these into other contexts.
However, this mode of communication was not enough to facilitate effective communication for all groups, as indicated in Example 6 (Figure 2): There was a lack of communication between us because sometimes we used to meet online on WhatsApp and some of the members would read the messages late and miss some of the important things, and this was challenging because our work was affected in some way.Although we had challenges as a group, we developed ways of overcoming them.Some of these ways of overcoming challenges are discussed next.

An overt commitment to participation
Another theme that strongly emerged was that some kind of overt commitment to participating in the group-learning process was necessary to facilitate effective dialogic engagement in groups; altogether 29 of the groups brought this up.Example 1 in Figure 3, for instance, indicates how this could lead to an adherence to the goals of the group.This commitment often took the form of agreeing to meet inperson, as was the case with Examples 2 and 3 in Figure 3.
The importance group members placed on meeting in-person highlights the social constructivist nature of learning (cf.Vygotsky 1962), and students' need for this within our context, as well as the need for an embodied presence to supplement other modes of communication, like the online group chats discussed above.Example 2 (Figure 3) indicates how the discourses, including behavioural conventions, of teachers in training started emerging through such a commitment, teaching the students how to manage their time, to be professional, and to treat colleagues with respect.By attaching value to such behaviour, students started investing in what Darvin and Norton (2018) refer to as their current and future identities, in this case as pre-service and future teachers.Example 3 (Figure 3) emphasises the importance of a commitment to equality in agreeing to be available in-person to work collaboratively in the same space and time -thus, what Brophy (2002) calls a well-developed class community which creates opportunities for democratic dialogue being established.
In-person meetings often resulted in some strife between group members, as illustrated by Example 4 (Figure 3).A mechanism used to facilitate disagreements (and to assist students in not only letting their writing 'form a single entity', but also the group as a whole), which was encouraged by the course and implemented by many groups, was that of ground rules, as indicated in Example 5 (Figure 3).This seemed to lead to the 'inclusive communal groups with a strong sense of solidarity' to which Beck and Kosnik (2006:14) refer, and to engaged dialogic interaction instead of 'polemical or hostile' arguments (Shih 2018).Clearly defined responsibilities, as indicated in Examples 3 and 6 (Figure 3), created a space in which students felt enabled to voice their opinions and which allowed for equitable work distribution.Some groups supplemented this with a commitment to regular reflections, as seen in Example 7 (Figure 3).A democratic environment, as is required in social constructivism, allows for such reflection, which is a hallmark of deep approaches to learning (cf.Biggs 1999).What is more, in the commitments students make to the group work process, we see a negotiation of power dynamics (cf.Darvin & Norton 2018) which, when effectively navigated, could lead to increased student investment.
FIGURE 2: The affordances of group chats.

6
We were using WhatsApp as a plaƞorm for effecƟve communicaƟon.
1 2 [We used] telephones when we couldn't meet in person, using WhatsApp to communicate, and holding calls to discuss and complete the task.

[PG 2H]
We have a WhatsApp group we were interact every day.In that group, we all had an equal voice when it comes to decision making and contribuƟng ideas.Everyone was allowed to voice out their ideas and correct mistakes from other peoples' submissions before we submit in the portal.

Because we were correcƟng each other's mistakes, that showed social capital. [PG 5E]
There was a lack of communicaƟon between us because someƟmes we used to meet online on WhatsApp and some of the members would read the messages late and miss some of the important things, and this was challenging because our work was affected in some way.Although we had challenges as a group, we developed ways of overcoming them.

[PG 4J]
We were able to communicate as a group not only in tutorials but also in a WhatsApp group chat that was created and could report frequently on things we found difficult.In contrast, groups who did not make such commitments to participation, or who did not honour these, tended to have a negative experience of the action-learning process, and did not experience the advantages of dialogism, or of an environment where knowledge was co-constructed.In Example 1 (Figure 4), for instance, one group expressed its frustrations at not being able to agree on something as elementary as when to meet, which resulted in work being submitted late.The group in Example 2 (Figure 4) experienced not only this frustration, but also found the multilingual nature of their group a barrier to expression.The right to speak, as conceived of by Freire (2005), was therefore constrained by factors like multilingualism, and an unwillingness to commit to participation.Truth, in the Bakhtinian sense, in these cases was not 'born between people collectively searching for the truth' (Bakhtin 1984).Such groups were characterised by inequality and a lack of trust, as was the case in Example 3, at least initially (Figure 4).More guidance on effective group work building on the tenets of social constructivism, dialogism (as per Bakhtin 1984) or dialogic pedagogy (as per Freire 2005) might have aided such groups to become more invested in the action-learning project.

Positive outcomes of dialogic learning
Although some groups experienced conflict they could not resolve, most groups (sometimes only eventually) experienced the action-learning project positively, as indicated by PG 5M: The aspects we found rewarding are collaboration, communication on WhatsApp, face-to-face meetings, and understanding our roles.In addition to these, major themes that emerged from the reflections were the affordances of drawing from multiple perspectives, the whole being larger than its parts, the development of multiple competencies, and the impact on students' future identities.These are discussed below. 1 We could not agree on a Ɵme for meeƟng because some of the group members wanted to do things early then others wanted to do the work late (Freeman, 2006).This was frustraƟng because now we had to call each other and make sure that everyone in the group agreed on the same sƟpulated Ɵme which was even making us write our work late.[PG 1C] 2 The experience (...) was not parƟcularly the best [because of] group-mates not wriƟng their work in Ɵme and we had language barriers as we spoke different languages and had a hard Ɵme expressing ourselves freely.[PG 3I] 3 Despite the good aspects that we faced, working as a group had some of the aspects that we found frustraƟng to all of us.The main aspect is that there was no trust and connecƟon to us as we had different aspiraƟons.It was frustraƟng because one will not trust the other group members' opinions which caused conflict every Ɵme we met as a group (...).This was frustraƟng as before we complete a certain acƟvity we must argue first.[PG 1C] 1 2 3 4 5 6 7 The members were commiƩed to the goals the group has set and they were willing to work on achieving them by aƩending the meeƟng we have planned and by parƟcipaƟng in the discussions we had.[PG 1I] We devise our weekly work internally, and when we meet, we must merge the work to form a single enƟty.Therefore, choosing how to link the paragraphs takes Ɵme because we would disagree on how to do so and what needed to be fixed in each.[PG 5K] We used to solve our problems [with] the ground rules we created (...).There was a rule that stated that every member's opinion maƩers which helped us to tame the heated arguments.We were able to remind each other of that rule which made us listen to different perspecƟves.[PG 2F] Working in a group would be extremely difficult if responsibiliƟes were unclear because we lacked the self-assurance to voice our worries, especially the problem of compeƟng personaliƟes that would affect power relaƟons.[PG 2L] In our meeƟngs, we used to hold discussions to talk about what we did wrong and how we could improve as a group.[PG 4J] Everyone had to be at the meeƟng spot at the agreed Ɵme and this was rewarding as it taught us Ɵme management and how to be professional with the people we are working with.[PG 1D] The lesson we've learned from experience is that if we're given an assignment, we must complete it aŌer class (...).Complet[ing] the assignment at home separately will result in procrasƟnaƟon and sending the work separately will result in inequality because the group administrator will have to do a lot of work.We've decided that going forward, the best approach for students to complete group work is to do it in class; there will be a designated Ɵme for that.This will benefit the group as a whole because everyone will parƟcipate, and the work will be finished even before the due date.[PG 2H] FIGURE 3: An overt commitment to participation.

The affordances of drawing from multiple perspectives
The affordances of drawing from multiple perspectives emerged strongly as a theme in students' reflections, as evidenced by 19 of the 44 groups commenting on this, and as illustrated in Example 1 (Figure 5).Example 2 highlights the influence of dialogue in this collaborative process.The social construct of knowledge (cf.Beck & Kosnik 2006;Vygotsky 1962) strongly emerges from these reflections.Listening as part of dialogic pedagogy, which is implicit in Freire's argument that each person must have the right to speak, and is also an important part of the actionlearning pedagogy, was emphasised in Example 3 (Figure 5): we needed to listen to each other and reach an agreement.The social constructivist principle that learning includes all aspects of a person, including attitudes, emotions, values, and actions, was further evident in the observation captured in Example 4 (Figure 5), that: We were able to have members sharing different perspectives, interests, and skills.
Evidence of deep approaches to learning emerges in students applying core course concepts to their reflections.
In Example 5 (Figure 5), for example, the group insightfully reflects on the course concept of 'identities'.Likewise, PG 6L in Example 6 (Figure 5) explicitly link this concept, as well as that of cultural capital, which was a strong focus during the course.Drawing on the concept of social capital within the cultural capital framework, the GP 5E (Example 7, Figure 5) powerfully added: Social capital is the support you get from your peers and other social contacts (Yosso 2005).We were equally allowed to voice out our ideas, and none of us felt left out.As a result, we had a pool of ideas to put in our research proposal.
As Entwistle (2017) argues, for university learning to be successful, students must be able to connect subject matter to their own ideas and comprehension.These groups show how, by drawing on deep approaches to learning, they can transfer theoretical knowledge to real-world contexts.An affordance of working and studying in a diverse, multicultural context such as South Africa is that the multiple perspectives that arise in a group work setting are likely to be far more diverse than against a more monocultural background.
Students therefore have the opportunity for a much richer learning experience than might be the case in many other contexts.

The whole being more than the parts
Closely linked to the affordances of drawing from multiple perspectives, though still worth discussing as a separate theme, is that of the whole being more than the parts, especially as 24 of the 44 groups commented on this.On this theme, the focus is not so much on the ability to draw on students with different life experiences and backgrounds, but more on the value of collaboration, and being able to achieve more in a collaborative context than could be done alone, as powerfully indicated in Example w (Figure 6), that two or more people are always preferable to one (but also in all the other examples in Figure 6).The members of the group in Example 3 (Figure 6) clarify why they believe this to be the case, and implicitly (yet effectively) link the process to both social constructivism and dialogic teaching and learning in their references to breaking down complex activities as groups, and in their focus on group discussions in achieving this.The effective communication referred to in Example 4 acts as a reward in and of itself for the group in Example 4 (Figure 6).A deep engagement with content is FIGURE 5: The affordances of drawing from multiple perspectives.

6 7
We had different idenƟƟes which we managed to bring together as we are from different backgrounds and had different experiences considering that we are from different schools and were taught differently.This helped us work well as a group as everyone gave their different ideas which gave us more informaƟon that helped us with the proposal and to navigate the work given to us.Every group member had different perspecƟves on certain issues, which helped us to idenƟfy some of the mistakes we made and implement our problem-solving skills to those certain situaƟons.[PG 5M] When we were brainstorming-each other's personaliƟes, idenƟƟes, and cultural capital, were important because they enabled us to produce ideas and solve problems.[PG 5L] Social capital is the support you get from your peers and other social contacts (Yosso, 2005).We were equally allowed to voice out our ideas, and none of us felt leŌ out.As a result, we had a pool of ideas to put in our research proposal.3 With all the problems menƟoned above, we now know that for us as a group to get along we needed to listen to each other and reach an agreement.We should also remember that we are different and that it is completely fine to have different views.

The development of multiple competencies
Twenty-five of the 44 groups explicitly mentioned various competencies they believed to have developed throughout the collaborative action-learning process.These include planning, prioritising, time management, communication, as well as the generation of ideas and critical thinking (Examples 1 and 2, Figure 7).The development of problemsolving as an outcome of dialogic interaction that happened during the action-learning project was also highlighted by groups such as the one in Example 3 (Figure 7).These selfreported gains in competencies, strongly linked to academic literacies, add to Lyle's (2008) findings that considerable evidence indicates that collaborative, dialogic talk contributes to cognitive development.Simultaneously, this provides evidence of students' metacognitive engagement with their developing competencies, and of their journey of critical examination, inquiry, and active engagement with knowledge through deep approaches to learning (cf.Biggs 1999).

The impact on students' future identities
Strong evidence of students' investment in the course came apparent in terms of how they linked the dialogic teaching and learning that formed part of this action-learning pedagogy to their imagined future identities (cf.Darvin & Norton 2018), with 22 of the 44 groups commenting on this.These future identities manifested at two levels.Firstly, some groups pointed to a change in nearby future identities, in terms of their future selves as more senior students (as illustrated in Example 1 of Figure 8).Students reflecting on how they imagine their experience of dialogic teaching and learning influencing their future selves as teachers, emerges even more strongly, as illustrated in Examples 2, 3, and 4 in Figure 8.
In this synthesis between current and future selves, we observe investment in how students commit to the practices and identities of a language and literacies learning process that is underpinned by dialogic teaching and learning.

Limitations and strengths of the study
A limitation of this study is that it was contextually bound to a B.Ed. programme at a specific university in Gauteng aimed at teachers in the SP and FET phases.Results from this study might not be generalisable into other contexts, and similar studies in various contexts would be needed to see whether the type of dialogic teaching and learning, drawn on form an action-learning approach, has similar advantages in other contexts.Furthermore, the study relies on self-reporting by means of reflection and is limited to a qualitative research approach.Future studies might consider triangulating data with students' writing, or embark on longitudinal research.Finally, the reflection happened in group format, and group dynamics might have influenced students' reflections.
At the same time, the strength of this research lies in its use of reflection for assessing pedagogical effectiveness.This allows for meta-cognition -thus, of verbalising students' awareness of their knowledge and thinking (Chang 2019;Langdon et al. 2019), which would otherwise be difficult to measure.
Similarly, the use of group reflections, while having limitations, also has strengths, such as allowing groups to generate and refine ideas through dialogue (Chang 2019), which is suitable for a paper on the affordances of dialogic teaching and learning.

Conclusion
Case (2015) argues that in 'true higher education', there must be a synthesis of students' personal and social identities: [A]n individual who has been able to formulate ultimate concerns and enact projects towards that end; also a student who occupies that role in a way which gives expression to their personal identity.(p.849) In students' reflections on the action-learning project they participated in for the course under discussion, we see such a synthesis.There is clear evidence of students engaging in deep approaches to learning, and in integrating not only the theory discussed in the course, but also pedagogies drawn on in the course, into their current and future identities as students and teachers.
FIGURE 6: The whole is more than the parts.

2
Firstly, each person gives their own ideas that you alone could not have thought of and that helps in having more ideas and soluƟon to the problem.(...) CollaboraƟon and brainstorming are increased when working in groups.As a result, there are more ideas generated, and producƟvity increases.For problem-solving, compleƟng challenging tasks, and fostering innovaƟon, two or more people are always preferable to one.[PG 5M] 3 Working as a group was rewarding as we were able to break down complex acƟviƟes into simpler parts that were not going to be possible if one were to work as an individual.We were able to clarify and beƩer understand the content through discussions as a group.(...) it helped with developing effecƟve communicaƟon skills among each other.6.8 [PG 1.8] 4 Another aspect that we found rewarding is the effecƟve communicaƟon that we applied as we were working together.[PG 1C] 1 The other benefit for working as a team is having different skills that can benefit the whole group.[PG 4K] It is interesting to note that at the onset of the action-learning process, students were very resistant to working in groups.It was only at the end of this year-long project that we saw a change of heart in so many of our students, as evidenced in the reflections included in this paper.Effective dialogism, carefully designed as part of the course pedagogy, would seem to be conducive to a willingness to have beliefs challenged and revised -a hallmark of deep approaches to learning.Integrating dialogism into the design of a course would, however, seem not to be enough in and of itself.To answer the first research question (Which aspects facilitated dialogic engagement in groups?), it is advisable to encourage students to include mechanisms which are likely to facilitate dialogic learning.These might include creating and agreeing to a set of group rules which include commitments to equality, respect, and regular attendance of in-person meetings.Other mechanisms could be creating online group chats to facilitate more frequent communication on a forum that allows for a quick response, as well as clearly defined responsibilities.
Various themes emerged around the second research question, namely, 'What evidence emerged from student reflections about the nature of learning through dialogic interaction during the action-learning process?' Students appreciated the multiple perspectives that emerged from their culturally, socially, and linguistically diverse groups.They further saw value in drawing on other students to complete complex tasks.Learning through dialogic interaction was, therefore, by its nature socially constructed, and evidence emerged indicating that this led to more invested learners who drew on deep approaches to learning.Through this process, students reported the development of various competencies.Finally, we saw evidence of students' investment in their learning in their references to how this would impact their future identities.This paper has accordingly shown how an action-learning pedagogy which, by design, draws on dialogic teaching and learning between students, can result in students adopting deep approaches to learning, and investing more deeply in their learning.Particularly within the diverse South African context, action-learning as an Example of a social constructive pedagogy seems to be particularly well suited to allow our students to draw from the richness of their diversity.Whether these affordances indeed impact students' future identities, however, is a question which remains unanswered, and additional longitudinal research would be needed to determine this.

FIGURE 4 :
FIGURE 4: Challenges encountered in dialogic teaching and learning.
from these reflections, showing an acute awareness of the social construct of knowledge, and of the development of new competencies, which is addressed in more detail in the next theme.