TOWARDS THE INSTITUTIONALISATION OF COMMUNITY ENGAGEMENT IN HIGHER EDUCATION IN SOUTH AFRICA

Although discourse related to the salience of community engagement in higher education has proliferated, little has focused on the process of institutionalisation of community engagement in South Africa. This paper presents findings from a national study that explored how community engagement could be institutionalised at higher education institutions locally. A qualitative research approach was used to guide the study and semi-structured in-depth interviews were conducted with members of executive management, members from the directorate and office of community engagement and academics from six universities nationally. Data revealed that there were eight key factors deserving of consideration in order to institutionalise community engagement through teaching and research, student involvement, transforming institutional infrastructure and through community partnerships.


INTRODUCTION
Kofi Annan, former Secretary-General of the United Nations, said that, "the university must become a primary tool for Africa's development in the new century. Universities… enhance the analysis of African problems; strengthen domestic institutions; serve as a role model for human rights, and enable African academics to play an active part in the global community of scholars" (Bloom, Canning & Chan, 2006: 2). It is within this context that higher education institutions in South Africa must respond to the needs of its local communities, become more socially responsive, but more importantly, join the global scholarly community that has increasingly nurtured community engagement (CE) within its teaching and research initiatives.
South Africa, through the inheritance of the apartheid legacy of poverty, poor human socio-economic development and a growing health burden, faces greater responsibility, which challenges universities locally to restructure their http://dx.doi.org/10.18820/2519593X/pie.v38.i2.03

Bhagwan
Towards the institutionalisation of community engagement in higher education institutions, to become more socially responsive to these problems. Moreover, it compels South African universities to reconsider how their teaching, research and their internal structures and policies can become more enabling of community engagement. In the transition towards inclusive democracy in 1994, South African universities endeavoured to undo the historical threads of social justice, by heeding the call made by the Department of Higher Education (DHET, 2014:10) that universities redress past inequalities and "transform the higher education system to serve a new social order," in a way that redirected its purpose towards society and development. Community engagement and service learning became entrenched within documents such as the Green Paper on Higher Education and Transformation (1996), the White Paper on Higher Education (1997), the National Plan for Higher Education (2001), the Founding Document of the Higher Education Quality Committee (HEQC) of the Council on Higher Education (2001), the HEQC Criteria for Institutional audits (2004) and the HEQC Criteria for Programme Accreditation (2004). More recently the White Paper for Post-School Education and Training (2013) noted that many of the community engagement initiatives at South African universities are fragmented and not interrelated with the teaching and research functions of these institutions, which suggests the need for it to be rigorously interweaved into the core business of teaching and learning, but moreover institutionalised into the broader academic culture of the university.
Despite the calls made within the aforementioned policy documents, community engagement has emerged slowly within the local literature (Albertyn & Daniels, 2009;Maistry, 2012). Several local scholars have however written around the theorisation and conceptualisation of community engagement locally (Bhagwan, 2017;Slamat, 2010;Muller 2010). Despite this, Favish (2010) lamented that CE remains under theorised in South Africa and argued the need for greater interconnectedness between CE and the other core activities of the university. The Council on Higher Education (CHE) (2008) made similar observations following a perusal of institutional audit reports, arguing that whilst the conceptual continuum appears evident at a superficial level, there are dissonances in the understanding of CE.
In contrast, the theorisation and implementation has burgeoned globally, particularly in the United States and scholarly work has evolved around its institutionalisation in higher education. Accordingly, almost a decade ago twenty-one new initiatives emerged within higher education networks, in the United States, which began organising themselves around advancing engagement (Battistoni & Longo, 2011). Concomitantly universities began transforming themselves into socially responsive civic institutions by engaging in service learning, community based participatory research, community oriented primary care and community service and outreach (Bhagwan, 2017).
It is against this backdrop that the current study sought to understand how community engagement was being strengthened at universities that were flagged as best practice examples of CE and what was needed to further entrench CE within higher education institutions nationally.

LITERATURE REVIEW
Post the new democracy, universities were urged to "create and advance economic, social and cultural opportunities and development" (Higher Education South Africa, 2007:15). The Higher Education Quality Committee of the Council on Higher Education also called for knowledgebased community service and argued for the co-construction of relevant knowledge that would benefit communities (HEQC, 2001). The UNESCO Co-Chairs for Social Responsibility have argued that knowledge-based community service has since become a criterion for accreditation and quality assurance and community engagement needs to be reported on in institutional audits (Hall, Tandon, Tremblay & Singh, 2015).
Literature related to community engagement (CE) in South Africa reflects a lack of conceptual clarity around CE (Favish, 2010;Nongxa, 2010;Pienaar-Steyn, 2012;Bhagwan, 2017) and may be a factor challenging its institutionalisation at local higher education institutions. Some of the more salient definitions that are offered by the Centre for Higher Education Transformation (CHET, 2003 :4) which defined CE as "... a systematic relationship between Higher Education and (their) environment (communities) that is characterised by mutually beneficial interaction in the sense that it enriches learning, teaching and research and simultaneously addresses societal problems, issues and challenges". This definition views CE as being embedded within the knowledge exchange between universities and communities through co-inquiry, joint research initiatives, co-learning, interdisciplinary and use of knowledge that benefits academia whilst endeavouring to solve real world problems (Bender, 2008).
However, several papers recently endeavoured to enrich an understanding of CE locally. Bhagwan (2019) detailed the process of community engagement and how universitycommunity partnerships can be nurtured. In her paper she emphasised the need for institutional preparedness, a greater commitment to understanding the local contexts surrounding universities by building respectful community partnerships that focus on capacity building and sustainability, working towards transdisciplinary research and the co-production of knowledge with community partners. Through a research initiative undertaken in a rural town in South Africa, Albertyn, Botha, Van der Merwe, Le Roux and Coetzee concluded that teaching and learning related to community development principles, planned activities, appropriate research methods, quality, ethical issues, sustainability and reciprocal benefits were the criteria to ensure an integrated approach to community engagement. Olowu's (2012) article provided greater insights into developing a systematic approval for benchmarking universitycommunity engagement in South Africa and is relevant to enabling institutionalisation of CE.
Internationally, efforts to nurture community engagement activities that could "enrich scholarship, research, and creative activity; enhance curriculum, teaching and learning; prepare educated, engaged citizens; democratic values and civic responsibility; address critical societal issues and contribute to the public good" have also received dedicated attention (CIC Committee on Engagement, 2005:2). Community engagement has been defined as the "strenuous, thoughtful, argumentative interaction with the non-university world in at least four spheres: setting universities' aims, purposes and priorities; relating teaching and learning to the wider world; the back-and-forth dialogue between researchers and practitioners; and taking on wider responsibilities as neighbours and citizens" (ACU, 2001:i). Scholars have argued for "nothing less than a transformation of contemporary academic culture," and a transformation of higher education institutions into "engaged" universities (Zlotkowski 1995:130).
It was the Talloires Network (2020) that devoted itself to advancing the civic roles and social responsibilities of universities in multiple ways. The Talloires Network is an International Association of institutions committed to strengthening the civic roles and social responsibilities of higher education, that have worked together to implement the recommendations of the Talloires Declaration and build a global movement of engaged universities. The Talloires http://dx.doi.org/10.18820/2519593X/pie.v38.i2.03

Bhagwan
Towards the institutionalisation of community engagement in higher education Network, through this Declaration, agreed to embed public responsibility through the policies and practices of higher education institutions by creating institutional frameworks that encourage, reward and recognise good practice in social service by students, faculty and their community partners and by fostering partnerships between universities and communities to enhance economic opportunity, and by strengthening the relevance of university education and research (Watson, Hollister, Stroud & Babcock, 2011:xxiv). The global literature has consequently mirrored the shift towards the "engaged" university and a change in institutional vision and culture was articulated within Boyer's (1996:7) "scholarship of engagement". Aligned with this, a set of indicators of engagement evolved that included administrative and academic leadership, internal and external resource allocation, faculty roles and rewards and professional development and community voice as key markers for institutionalisation (Hollander, Saltmarsh & Zlotkowski, 2001). This led to the development of multiple institutional assessment tools that reflect institutional commitment to community engagement through the university's mission, structures, leadership and through, faculty and community involvement (Furco, 1999;Holland, 2006). The development of effective benchmarking tools for university community engagement is still however, in its infancy in South Africa (Olowu, 2012).
Major shifts are required within an institution's culture, related to institutional behaviours and processes, in order to institutionalise engagement (Eckel, Hill & Green, 1998). Universities that have been successful in terms of the criteria set within the classification for community engagement are those that reflect deep and pervasive changes in practices, structures and policies, thereby enabling an institutional culture of community engagement (Saltmarsh & Johnson, 2018). Whilst depth is critical to transformation, it is insufficient as deep changes are not necessarily broad (Eckel et al., 1998:4). Hence, it is possible for deep changes to occur within specific units or academic departments without being widespread throughout the institution, meaning that no pervasive organisational transformation has occurred (Eckel et al., 1998). Pervasiveness refers to the extent to which community engagement "crosses unit boundaries and touches different parts of the institution" (Eckel et al., 1998:4).
The shifts and transformation being called for should be deep and pervasive, such that it affects the entire institution, is intentional and occurs over time (Saltmarsh, Janke & Clayton, 2015). Although some institutions have begun to prioritise engagement, Mugabi (2015) expressed that community engagement remains marginally institutionalised and most South African universities have not fully integrated engagement into their agenda. Furco (2014) argued that the institutionalisation of community engagement requires an intentional agenda for embedding it into the institution's academic culture. He added that unlike marginalised practices that are episodic, isolated, at risk and lack status, institutionalised practices are widespread, legitimised, expected, supported and resilient to changes at the institution (Kramer, 2000). Individualised engagement programmes that are strong are large, deep and permanent (Kiely, Kittelberger & Wittman, 2018).
Transformational change occurs only "when shifts in the institution's culture have developed to the point, where they are both pervasive across the institution and deeply embedded in practices throughout the institution" (Saltmarsh & Johnson, 2018:9). Institutionalising community engagement also occurs by focussing on how engagement affects the educational experience of students; how it is integrated into faculty scholarly experience i.e. teaching, research and creative activity and service, and the creation of engagement specific budgets, professional staffing and other infrastructure (Saltmarsh & Johnson, 2018: 1 Hundley (2018) argued that community engaged scholars enact their passion by being instruments of change within teaching, research or service for the collective good.
The institutionalisation of CE moves through crucial phases of development that include critical mass building, quality building and sustained institutionalisation, which occurs over five to seven years (Saltmarsh & John, 2018). Empirical inquiries have in fact documented that it takes 15-20 years to achieve full institutionalisation of CE (Sandmann, Thornton & Jaegar, 2009;Furco, 2010).

Study design and setting
This inquiry was guided by qualitative research methodologies, specifically a multiple case study design (Yin, 2001) as selected universities across South Africa were chosen. The study sought to distil rich descriptive information from key stakeholders namely Executive management that included 1 Vice Chancellor, 2 Deputy Vice Chancellors, academics, middle leadership and administrative staff involved in community engagement at their institutions, with regards to how community engagement was being strengthened at their institutions and what was necessary to institutionalise community engagement at higher education institutions in South Africa. One of the aims was to understand how CE was being advanced at these local universities and how it could be further strengthened at local universities. The research questions then were: What factors contribute to advancing CE and what can be further done to strengthen CE at local universities.
Six universities were involved in this study from the Western Cape, Eastern Cape, Limpopo and Gauteng. Two primary criteria were used to purposefully select the case study sites for this inquiry. The South African Higher Education Community Engagement Forum list of best practice institutions was used and from this those universities with best practice community engagement activities were identified. Moreover, those institutions that offered diversity in the forms of community engagement being undertaken namely community outreach, volunteerism, community-based research and service learning formed the criteria to purposefully select them for inclusion in the study. Finally, all those who voluntarily agreed to participate were included in the study.

Sample
In accordance with qualitative inquiries non-probability sampling methods, specifically purposive sampling, was used to recruit the participants. The primary criteria for selection related to their administrative management of CE as well as academic involvement in CE initiatives at their institutions. In total 33 academic and administrative staff participated in relation to the data derived for this paper.

ETHICAL CONSIDERATIONS
Ethical clearance for this study was obtained from the Institutional Research Ethics Committee (IREC) of the Durban University of Technology (DUT). The study was allocated ethical clearance number IREC 088/15. Permission was thereafter obtained from all participating universities. http://dx.doi.org/10.18820/2519593X/pie.v38.i2.03

Bhagwan
Towards the institutionalisation of community engagement in higher education

DATA COLLECTION PROCESS
After receiving permission to conduct the study at the respective institutions, dates to visit these sites were arranged. Academics who were involved in best practice projects at their institutions, as well as administrative staff and management were interviewed. An interview guide was used to facilitate these semi-structured in-depth interviews. Semi-structured interviews have a set of predetermined questions on an interview guide, but the interview will be guided rather than dictated by the schedule (Greeff, 2011). The guide was pilot tested with a group of similar participants and reviewed by the ethics committee at DUT, before use. It consisted of broad questions that focused on how CE was conceptualised, the values and principles underpinning CE partnerships, how CE was being strengthened and recognised at their respective institutions and what could be done to advance the CE mandate at local universities.
Whilst in-depth interviews and focus groups were also conducted with students and community partners, they will not be reported on within this paper. In addition, document analysis was also undertaken in relation to the university's mission statement. Holistically this data served to understand what efforts were being made to embed CE into institutional culture and to understand from these CE managers, academics and practitioners what was required to advance and institutionalise CE in South Africa.

DATA ANALYSIS
The data was analysed using thematic analysis as per the steps outlined by Braun, Clark, Hayfield and Terry (2020). Thematic analysis allowed the researcher to make sense of collective meanings and experiences. This helped to organise and reduce the data into broad themes. A preliminary coding scheme was generated that served as a template for the data analysis (Tutty, Rothery & Grinnell, 1996). Similar themes and recurring patterns in the data were linked together and the contrasts and differences identified (Liamputtong & Ezzy, 2005).

Trustworthiness
To enhance the trustworthiness of this study, the researchers used Guba's model (1994). This model provides four criteria to ascertain rigour in qualitative studies, namely credibility, dependability, conformability and transferability. Strategies used were multiple interviews with the participants, member checking, using a triangulation process during data analysis and maintaining an audit trail and a reflective journal (Buetow, 2019). This ensured trustworthiness of data and reflexivity of the researcher. Although not reported on within the purpose of this paper, data was also derived from students and community partners that validated the findings and served as important within the triangulation process.

DISCUSSION OF FINDINGS
Analysis of the data revealed that there were eight salient themes that were critical to enabling the institutionalisation of community engagement. These are discussed in the sub-subsections that follow:

Mission, vision and policies
Participants expressed that institutions seeking to advance engagement should state that it will be "an engaged university … in its vision, mission and policies." http://dx.doi.org/10.18820/2519593X/pie.v38.i2.03 Perspectives in Education 2020: 38 (2) Create "an ethos of engagement and speak to social justice and community-based learning." Talk "about communities" and use the "social justice and engagement and communitybased needs," in their mission, vision and policies.
"Develop a policy for community engagement. The policy must be infused into teaching and learning." The data reflected the importance of embedding CE within the mission statement and teaching and research policies of the university. An analysis of higher education mission statements in the United States reflected citizenship, democracy and social responsibility as being the key themes within these documents as well (Jacoby, 2015). Mission statements should reflect the purpose and goals of engagement and serve as a basis for identifying potential partnerships (Bringle & Hatcher, 2002). Moreover, universities should also explore how their strategic plan incorporates specific goals to achieve engagement and whether these goals are communicated throughout the university (Beere, Votruba & Wells, 2011).
Local scholar, Mugabi (2015:23) suggested that each university pay attention to its surrounding "institutional context, for example, history, disciplinary focus, location, ownership, mission, culture values and priorities, and national policy agendas when drawing up its mission and vision… because universities, even those in the same country, cannot have the same institutional environments, the focus, forms and organization of community engagement." Each university should therefore support its own unique vision in relation to engagement. One example of this is the mission statement of Nelson Mandela Metropolitan University 1 , which reads "we create and sustain an environment that encourages and supports a vibrant research, scholarship and innovation culture and we engage in mutually beneficial partnerships locally, nationally and globally to enhance social, economic, and ecological sustainability." The NWU Institutional Plan 2011-2013, prioritised engagement as one of its five values by describing its relationship with the "cultural, social and economic life" of its localities, region and nation; "with international imperatives; and with the practical, intellectual and ethical issues of … partner organisations" (North-West University, 2010:4). Other good international examples are of Northern Kentucky University, which states it will be nationally recognised as the university "that prepares students for life and work in a global society and provides leadership to advance the intellectual, social, economic, cultural, and civic vitality of its region." Its mission statement states it "supports multidimensional excellence across the full breadth of its work; teaching and learning, research and creative activity, and outreach and public engagement … with the emphasis on active learning, including student research, internships, co-op programs and service learning" (Northern Kentucky University, 2009:3). Civic engagement and social responsibility are emphasised within the University of Melbourne's strategic plan, under "knowledge transfer." Knowledge transfer is conceived as fostering partnerships to advance teaching, learning and research and meeting responsibility for greater good" (Watson et al., 2011: 49).
Some scholars have argued that whilst CE should be mirrored within the mission and institutional strategic plans, it is important that a unified definition exists that incorporates elements of the Carnegie definition (Manok, 2018). The Carnegie definition of community http://dx.doi.org/10.18820/2519593X/pie.v38.i2.03

Bhagwan
Towards the institutionalisation of community engagement in higher education engagement is classified as "the collaboration between institutions of higher education and their larger communities (local, regional/state, national, global) for the mutually beneficial exchange of knowledge and resources in a context of partnership and reciprocity" (Discoll, 2009: 39). The Carnegie Community Engagement classification is therefore a strategic planning tool (Quan, 2018) that SA universities can use to develop their mission and vision statements. Continuous reflection and a re-evaluation of how the university's community engagement definition evolves in relation to new trends however remains critical (Manok, 2018).
Whilst the aforementioned speaks to institutional policies, at a deeper level the data suggests the need to infuse notions of social justice and community-based learning into academic life. Social justice can be integrated into education by teaching students to acquire a vision of a just society that provides for the basic needs of all members of society, the realisation of the potential of society's people and developing an understanding of the power of people to change unjust structures so that people can be empowered to work towards a more just society (Bhagwan, 2017:9). Much of this can be achieved through community-based learning as suggested in one of the excerpts that helps students to acquire, practise and apply disciplinary knowledge and skills whilst acting on concerns that affect communities. The need to interweave the institutional vision and mission and translate policies into action occurs through teaching and research as discussed in the following sub-section.

Embedding community engagement within teaching and research
"It is …about the scholarship of teaching and learning." "I think community engagement should not be looked upon as a third pillar, but rather infused within teaching and learning" "Community engagement should also get involved in work integrated learning, which cuts across all the disciplines, at all the schools all the faculties." It is about being "serious about knowledge construction and dissemination in partnership." As evidenced in the data, the institutionalisation of CE occurs through a deeper understanding that CE remains indivisible from the core missions of teaching and research. Moreover, as one participant suggested, institutionalisation starts to take its roots only when CE is interweaved across all schools and faculties as opposed to sporadic, linear efforts being made by a few academics and departments. Engaged teaching manifests when course or curriculum related learning activities engage students with the community in mutually beneficial ways (O' Meara, 2008). Curricular engagement within the Carnegie classification for community engagement is described as "teaching, learning and scholarship that engages faculty, students, and community through interactions that address community identified needs, deepen students' civic and academic learning, enhance community well-being, and enrich the scholarship of the institution" (Saltmarsh & Johnson, 2018:6). The focus is on the "extent to which community engagement is part of the central academic experience of the http://dx.doi.org/10.18820/2519593X/pie.v38.i2.03

Perspectives in Education
2020: 38 (2) campus, the number of students impacted, the breadth of the courses offered, the depth of the curriculum, the goals of learning outcomes, and assessment of community engagement outcomes." What this translates into is that academics integrate outreach into their classes and curricula, and that students begin participating in co-curricular activities such as service learning that place them in the community, where they can apply their learning to solving real-world problems (Boyer, 1990:7). This supports the arguments made for it to become embedded across all disciplinary academic departments.
Reiff (2018) identified the following criteria for service-learning, which can also be used to guide same in the SA context: • It should involve students in substantive, ongoing community service activity that is directly integrated into the curriculum in a significant way.
• The service component should be developed cooperatively with community partners, so that it meets the needs identified by the community.
• It prepares students for work in the community.
• It requires systematic and regular reflection, reporting and assessment of the service experience with faculty and other students e.g. reading assignments or oral presentations that combine the service experience.
• It assesses the impact of the service-learning component.
Moreover, service learning should reflect mutuality, reflection before/during/after service, by tying service-learning objectives, to learning objectives and goals of community partners (Welch, 2018). Capstone courses with titles such as creating liveable communities for an aging society, women and development, poverty and homelessness, social justice education and environmental advocacy, can be used to compel students from a variety of disciplines to respond to complex concerns for their communities (Watson et al., 2011).
Institutionalisation at a deeper level, however, impacts course design and classroom pedagogy. Designing the undergraduate curriculum to embed experiential, community based and interdisciplinary learning across all departments, is one way to strengthen institutionalisation through teaching and learning (Furco, 2014). When individual academics adopt service learning in a linear way, without departmental, faculty or institutional commitment, the impact on the student is limited (Schneider, 2005). However, if a student repeatedly encounters engagement as they move through every academic level it will become entrenched within their learning. Empirical evidence in fact supports notions that service learning and community-based research is beneficial to student learning (Antonio, Astin & Cress, 2000;Astin, Vogelgesang, Ikeda & Yee, 2000;Passarella & Terenrzini, 2005).
As participants suggested, there is a need to reconsider research approaches such as community participatory approaches that will enable the co-production of knowledge. Community based research focuses on research "with" and not "on" communities and values community members as research partners or co-investigators (Bhagwan, 2017:13).

Involvement of academics
Participants argued that academics play a crucial role in the institutionalisation of engagement in terms of teaching and learning as discussed in the preceding theme. They were urged as follows: http://dx.doi.org/10.18820/2519593X/pie.v38.i2.03

Bhagwan
Towards the institutionalisation of community engagement in higher education Rethink, their teaching and research and to work in participatory ways that mutually benefit the university and the community.
Support transdisciplinary research…(this) could move people out of their disciplines and that's also very much an engagement thought. You can't solve very complicated problems of community through just one discipline.
As discussed under the preceding theme for engagement to move from the margins to a more advanced position as part of institutional culture, requires that academics bring it into institutional culture (Lazarus, Erasmus, Hendricks, Nduna & Slamat, 2008). Hall et al. (2015) found in their study that whilst the middle level leadership, which includes deans, chairs and directors, play a significant role in institutionalising community engagement, it is in fact the professoriate, lecturers and research leaders who are at the heart of the engaged scholarship process. Academics should therefore constantly reflect on how they may transform faculty work roles through their daily practices, priorities and commitments (Rhoades, Kiyama, McCormick & Quiroz, 2008). This requires, as suggested in the data, to rethink how teaching and research is done to embrace community-based participatory approaches. A powerful argument in the data was the argument made for transdisciplinary research. Transdisciplinary research refers to research that transcends the disciplinary knowledge within the university to include knowledge that resides amongst community voices (Bhagwan, 2019).

Student involvement
Student involvement in engagement emerged as the next theme related to institutionalising CE. Participants had the following to say: First step would be to get students invested in the engagement that we're doing.
Students should do an hour of service a week in our community sites and through doing that service, they learn lots of things about the community, they build relationships and the whole way through this course.
We need to cultivate student interest and to grow the number of post graduate students in community engaged scholarship and trans-disciplinary research.
As reflected in the data, participants believed that student involvement was crucial in solidifying engagement within higher education. They asserted that this could occur through involvement in volunteerism, out-reach activities, service learning and community-based research. A review of literature undertaken by Antonio et al. (2000: 374-375) found that student engagement is positively related to "persistence in college, interest in graduate study, the development of leadership skills…greater knowledge of subject matter, ability to apply course concepts, strengthened critical thinking skills (and) civic responsibility." Beere et al. (2011) asserted that one way to ensure that students have a communitybased research experience is to set it as a requirement within a capstone course. They cited the work of Portland State University that stipulates that each student (except Honours and liberal arts majors), complete a community-based research capstone course that is "multidisciplinary, problem focused and community based." This approach can ensure that engagement becomes pervasive throughout the institution and that it involves students in social responsibility initiatives at some point in their academic journey. 2020: 38(2) 7.5 Building an institutional culture "Universities should have divisions of community engagement."

Perspectives in Education
"Embedding community engagement actually into institutional culture." "It should be infused into every discipline." "For engagement to be institutionalised it has to be 'constant,' with little things happening to reach different groups." "We offer service-learning workshops for academics, just to orientate them.
"The opportunities to build partnerships between faculties because it's possible to be in the same university and running projects/programmes in the same area" "We have a community engagement week. Main thrust of it is a community engagement learning symposium, where our academics get given a space to come share what they're doing." We have pushed… for the good scholars … (they) tell you about their community engagement, they can't separate it. It's so embedded in their teaching portfolios for personal promotions" It is 'important to host colloquiums on engagement… (they), showcase best practice projects and strengthen transdisciplinary collaboration." Whilst the data has suggested the need for community-based teaching and learning, other key ways of interweaving into the academic life in the institution also emerged as evidenced in the excerpts above. Workshops, colloquiums, dialogue within the university and faculties were seen as ways to bring engagement scholars together and strengthen awareness of CE through such activities. The Carnegie Foundation classified community engagement into three parts viz. foundational indicators, curricular engagement and outreach and partnerships. Foundational indicators are regarded as institutional commitment and institutional identity and culture. The latter is interlinked with mission and vision, recognition, marketing materials and community engagement as a leadership priority (Saltmarsh & Johnson, 2018) as discussed under the first theme.
Transforming institutional culture however requires "a new set of practices, creating new structures, and revising policies," which collectively requires organisational change (Olowu, 2012:99). Institutional leadership is crucial to transforming institutional culture (Olowu, 2012). Moreover, connecting universities to their communities have the potential for transformational change as it can contribute to reshaping institutional practices and purposes (Boyer, 1990). Whilst depth is critical to transformation, it is insufficient as deep changes are not necessarily broad. It is possible for deep changes to occur within specific units or academic departments without being widespread throughout the institution, meaning that no pervasive organisational transformation was occurring (Eckel et al., 1998:4). Pervasiveness reflects the extent to which community engagement "crosses unit boundaries and touches different parts of the institution" (Eckel et al., 1998:4). Beere et al. (2011:112) suggested the need for "campus conversations," where "dialogue, rather than a lecture or formal talk" is the norm. These conversations although organised at the department or faculty level, should include management as well http://dx.doi.org/10.18820/2519593X/pie.v38.i2.03

Bhagwan
Towards the institutionalisation of community engagement in higher education so that both stakeholders can deliberate upon the value and challenges of engagement and strengthen understanding between administrators and academics. Most important was the suggestion made by one of the participants regarding having an office or division of CE to drive CE initiatives into academic life. This office could implement the workshops suggested by participants, recognise scholarly work that embraced CE and to ensure the "constancy," of CE within the daily space of the institution.

Institutional support
The need for institutional support also emerged as being important as follows: "We have community partner of the year so there is recognition." "The VC is incredibly supportive of community engagement." "We have a DVC that is open to service learning to communities, her interest is in community projects." "Our DVC understands why we do it… from a very deep level and so whenever he gets a platform of sorts, whether senior management meetings, other engagements with academics etc. he speaks to it." "It took the process of implementing the infrastructure to basically manage engagement within the institution." "Community engagement projects must be run by a dedicated office." "It's got very good policies, very good administrative system. It's got good report systems." "The role of the community engagement office is to put partnerships and processes in place, to enable engagement…implement workshops related to engagement and to help academics write their philosophies for their portfolios." "The Directorate of engagement should manage a database of information, regarding which communities were involved, what projects were implemented, their location and how many communities were involved." "The Directorate should also monitor activities and provide a report on institution wide activities and scholarly outputs that emerged through engagement on an annual basis." The aforementioned data reflects the importance of management support in order to institutionalise CE, but more importantly a dedicated CE office or division to coordinate, manage and support CE initiatives. The Vice Chancellor or Deputy Vice Chancellor was perceived to be a powerful role player in promoting and supporting the engagement agenda of the Community Engagement office, academics and students involved in community engagement projects.
Institutional policy is equally important to supporting the functions of a community engagement office and to ensure commitment to drive the community engagement agenda institutionally. These organisational structures are therefore responsible for providing greater support to academics involved in engagement and so that community members can easily connect with the university (Beere et al., 2011). Moreover, ccommunity partners have more positive perceptions of institutional engagement when campuses operate centralised offices http://dx.doi.org/10.18820/2519593X/pie.v38.i2.03 2020: 38(2) of engagement compared to decentralised systems lacking a clearinghouse function for engagement activities (Weerts & Sandmann, 2008). The presence of executive management in community spaces was perceived as crucial in building stronger bridges between the university and community and can ultimately serve to validate CE as an equally important mission alongside teaching and research at university.

Perspectives in Education
What also emerged in the data was that some engagement staff worked in decentralised offices within each faculty. This appeared to enable more engagement activities being initiated across departments and disciplines within faculties, as opposed to it being sporadic activities limited to a few disciplinary homes. What was evident from the data was that these faculty representatives are part of an engagement committee who worked with the Directorate of Engagement to conscientise academics regarding engagement and provide capacity building related to engagement.
The institutionalisation of engagement therefore rests on developing university-wide agendas and policies, structures and practices to guide and facilitate the involvement of academic units, staff, students and external communities in engagement (Mugabi, 2015). An approved budget also appears important to supporting engagement, a problem that has challenged universities abroad (Weerts & Sandmann, 2010). An engaged campus is characterised by funding to support its infrastructure, which includes leadership positions, facilitative offices, internal grants to support or relieve academics who pursue engagement, support for major, multidisciplinary initiatives and scholarship support for engaged students (Beere et al., 2011).
It is the institutional context, by way of their policies however that shapes the extent to which academics become involved in community engagement, how they infuse engagement into teaching and research roles and how they are rewarded and encouraged (O 'Meara, 2004;Sandmann et al., 2008). The recognition of engaged scholarly work, as evidenced in the data, is an important way of recognising the importance of CE within institutional culture.

Community partnerships
The active presence of community partnerships was viewed as important to institutionalising CE. Participants said: The transformation and institutionalisation of engagement can only be effected when universities are serious about community partnerships that enable knowledge construction for mutual benefit through this relationship.
It is a two-way engagement…it is from a perspective that the university will possibly come with resources and the community has the indigenous knowledge.
The community provided a real-life context that allowed students, to understand relevant issues not from books but through the stories of people.
Community engagement has an element of sharing knowledge and perhaps mutually constructing or mutually defining an issue …co-production…co-creating knowledge in those spaces.
The institutionalisation of CE is enabled through a tangible presence in the community and through community relationships and partnerships. It is through these relationships that a context is created to facilitate learning about real life issues, through the stories of people. http://dx.doi.org/10.18820/2519593X/pie.v38.i2.03

Bhagwan
Towards the institutionalisation of community engagement in higher education Moreover, it creates the context for community-based research and the opportunity to build partnerships with community members who can advance the university's capacity to conduct rigorous scientific research that has benefits for the society (Furco, 2014). Communities however should not be viewed as objects for study but instead as partners. Those partnering with the community should therefore understand its history, its history of the relationship with the university, its different sub-cultures, its needs and priorities and the sources of power and influence within it (Torres, 2000). Moreover, the community should be seen in terms of its assets, which includes its strengths, wisdom, knowledge and experience that it brings to the partnership (Beere et al., 2011).
The classification for community engagement recognises relationships between those in the university and those outside the university that embrace the qualities of reciprocity, mutual respect, shared authority and the co-creation of goals and outcomes. These relationships are by their nature "transdisciplinary (knowledge transcending the disciplines and the college or university, and asset based (where the strengths, skills, and knowledges of those in the community are validated and legitimized" (Saltmarsh & Johnson, 2018: 4).
Outreach and engaged partnerships are two different but interrelated approaches to community engagement (Janke, 2018). Whilst outreach focuses on the application and provision of institutional resources for community use, engagement differs due to its focus on "collaborative interactions with community and related scholarship for the mutually beneficial exchange, exploration, and application of knowledge, information, and resources (research, capacity building, economic development, etc.)" (Saltmarsh & Johnson, 2018: 6). The latter are the types of interactions that universities need to nurture with communities.

Rewarding engagement
Rewarding engagement was also seen as a means of institutionalising CE as follows: We have research awards, teaching awards and community engagement awards and the community engagement lecture… (all should be equal), just the way others have been recognized and supported.
We have the Vice Chancellor's award for community engagement.
There are awards for student volunteers as well.
We have an awards evening in October... we have the top ten volunteers of the year, we call that a Gold Award.
As evidenced in the data, several universities reward their academics, student volunteers and community partners alike for their engagement work. Research has indicated that a lack of this recognition inhibits CE at institutions and relegates it to a lesser space in comparison with teaching and research (Driscoll & Sandmann, 2001;Sandmannn, 2006). The institution's rituals, awards and ceremonies should therefore recognise the value of CE and it should be "celebrated" (Beere et al., 2011: 38). As indicated in the data, several universities have rewarded academics for their engagement work at awards functions.
In an African context most universities emphasise publications, supervision of students and teaching experience as criteria for promotion, as opposed to engagement work (Mugabi, 2015). Promotion policies that recognise CE need to be created alongside awards that recognise CE activities at institutions (O' Meara, Sandmann, Saltmarsh & Giles, 2011). The http://dx.doi.org/10.18820/2519593X/pie.v38.i2.03 2020: 38(2) latter will draw greater attention to the institution's CE initiatives, provide material for internal and external publications, showcase the institution and the individual, publicly identify engaged faculty to serve as role models for all faculty, reward those doing the work and motivate those who are considering it. Awards should be made at department, faculty and institutional level and can range from a book to securing the title of "engaged scholar" (Beere et al., 2011:116).

CONCLUSION
Although the institutionalisation of community engagement is still in its infancy locally, several universities have begun making concerted efforts towards entrenching it within their university's milieu. The study has unveiled the importance of community-based teaching and learning, transdisciplinary research, community partnerships and a dedicated CE office to bring these dimensions together. Moreover, the importance of institutional support and the embedding of CE within the mission and policies of a university will undoubtedly serve as the pillars from which CE activities may grow.
Collectively all the themes interrelate in a synergistic way, which suggests that when integrated over a sustained period, it has the potential to enable institutionalisation, as opposed to linear, sporadic efforts. Overall the findings cohere with the literature that the key indicators of engagement are administrative and academic leadership, internal and external resource allocation, faculty roles and rewards and community voice (Hollander et al., 2001). Furco (2014) reflected on the journey made by the University of Minnesota to move the university into a sustained institutionalisation phase, saying that it took 13 years to reinvigorate the engagement agenda by prioritising CE as being integral to research and teaching, becoming more intentionally integrated with academic programming; becoming infused into the work of all and not just the work of those who work in traditional outreach units; creating more mutually beneficial partnerships by acknowledging the assets and knowledge in the community; focusing on working "with" the community and not just doing "for" or "in" the community and moving from discrete, time limited projects to "supporting multi-faceted, interdisciplinary 'partnerships' addressing grand challenges and broad societal issues," such as health, education and poverty (Furco, 2014:264).

ACKNOWLEDGEMENTS
This study was made possible through funding from the National Research Foundation.