Global Knowledge

Development matters: the

Institute potential
Working Paper contribution of the
Series
co-production of
2019-039 research to urban
October 2019 transformation
Diana Mitlin1, Jhono Bennett2, Philipp Horn3,
Sophie King3, Jack Makau4, George
Masimba Nyama5

1 University of Manchester/IIED, UK

2 1to1, South Africa

3 University of Sheffield, UK

4 SDI Kenya

5 Dialogue on Shelter, Zimbabwe

ISBN: 978-1-909336-74-2

Cite this paper as:

Mitlin, D., Bennett, J., Horn, P., King, S., Makau, J. and Masimba Nyama, G. (2019)
Knowledge matters: the potential contribution of the co-production of research to urban
transformation. GDI Working Paper 2019-039. Manchester: The University of
Manchester.

www.gdi.manchester.ac.uk

Electronic copy available at: https://ssrn.com/abstract=3470133

Abstract

The purpose of this paper is to share our experiences – as academics and

professionals – in co-producing knowledge to improve urban development outcomes in
the Global South. The focus of the paper – urban research and practice – is a context
in which academic work influences policy and programming, and professional
knowledge – validated and certified by academic institutions – forms the basis for
urban planning and management. Collaborative research – co-produced with social
movement activities – suggests that four issues need to be addressed to establish
more equitable relations. First, alternative theories of change about how research leads
to social transformation must be recognised, even if they cannot be reconciled.
Second, the contribution of social movement leaders to university teaching needs to be
institutionalised. Third, the relative status of academics vis-à-vis non-academics must
be interrogated, and better understood. Fourth, researchers’ accountability to the
marginalised needs to be established. We argue that academics are insufficiently self-
critical about the power dynamics involved in knowledge production with social
movements, and that long-term relations enable understandings to be built and some
of these tensions to be alleviated. Our conclusion highlights the unequal power
relations that lie behind these challenges, and summarises key measures to address
inequalities and their negative consequences.

Keywords

Participatory research, coproduction of knowledge, urban development, learning

www.gdi.manchester.ac.uk 2

Electronic copy available at: https://ssrn.com/abstract=3470133

1. Introduction

The significance of knowledge is widely recognised. We are said to be living in an

information age (Castells, 1996, 1997, 1998; Madden, 2015) and governments (local
and national) seek to build a knowledge economy. Within and beyond academic
institutions, the potential contribution of multiple types of knowledge has long been
recognised, particularly in the context of indigenous knowledge (Tuhiwai Smith, 1999).
These understandings, combined with motivations to be more relevant to the needs of
non-academic communities, have led to a wide range of efforts to co-produce
knowledge. Co-production has most commonly been used to describe the joint
production of services between state agencies and organised communities (Mitlin,
2008). In the case of research, the ‘co-production of knowledge’ is a term that
recognises the essential contribution of non-academics to generating knowledge.

There has been considerable recent interest in the co-production of knowledge (Simon
et al, 2018; Culwick et al, 2019; Osuteye et al, 2019). This reflects recognition of the
need to go beyond the participation of non-academics in research processes with an
acknowledgement of the significance of alternative ontologies and epistemologies.1 As
academics accept the significance of demands to decolonise and democratise the
academy, and to reframe a radical knowledge agenda, the need for more equitable
research processes shifts from the periphery to centre stage. The concept of co-
production – widely used in the service sector – recognises the significance of multi-
agency inputs into the conceptualisation, planning, implementation, resources and
evaluation of activities. More substantively it acknowledges the central role of users of
services in the production of those services; and hence supports substantive inclusion.
The relevance of co-production to research is now more evident, and in this paper we
present one long-standing experience with the co-production of knowledge to a wider
audience.

The paper provides a platform to share observations about the experiences of

academics and urban social movements engaged in the co-production of knowledge to
secure justice and equity in towns and cities of the Global South. Our objectives are to
explore these relations, and to identify and elaborate on the critical issues that need to
be addressed to create more equitable research processes, and hence achieve social
justice. Social justice is primarily considered here through a focus on more equitable
processes for the co-production of knowledge; however, we believe that more equitable
knowledge processes are required for transformative urban outcomes. We are
motivated in this endeavour because of our shared experience that insufficient
attention is given to significant tensions that need to be shared and discussed for the
co-production of knowledge to be achieved.

1
Hence the theme of the Development Studies Association conference in 2019 was ‘opening up
development’. And the theme of the Royal Geographic Society/Institute of British Geographers
conference in 2017 was ‘decolonising geographical knowledges’.

www.gdi.manchester.ac.uk 3

Electronic copy available at: https://ssrn.com/abstract=3470133

To date, many of the development studies debates related to social justice and
knowledge have focused on indigenous knowledge; our contribution considers the
engagement of broadly Western academic knowledge with the realities of excluded and
marginalised urban residents. Most of the global population is now living in towns and
cities (UN Population Division, 2018), with an estimated 880 million living in unsafe
houses and without basic services (UN-Habitat, 2014), with very high percentages of
urban employment being in the informal sector with associated precarity (Chen, 2014).
Outcomes are increasingly adverse, with low-income residents struggling against
displacement and the difficulties of securing essential services at affordable costs,
while elites manage urban land and commodity production processes to accumulate
wealth and extract profits. Urban-focused research is required to understand the
possibilities for, and constraints on, reforms, urban policies and programmes. This is
the context in which we engage with the challenges of co-producing relevant
knowledge.

This paper draws particularly on our insights as academics and professionals working
with SDI, a network of social movements and support NGOs that works in the informal
settlements of 33 countries in the Global South and has sought to advance equitable
urban development. Professionals working with SDI-affiliated movements are co-
authors of this paper.2 The research processes used for the paper are summarised in
Section 2, which elaborates on the contribution that SDI-affiliated federations of shack
and slum dwellers has played. The focus on this movement adds depth and enables an
interrogation of the issues. However, the findings are relevant to other efforts to co-
produce knowledge with multiple agencies.

As elaborated in Section 3, the themes this paper engages with are long-standing.
Participatory research and action-research have explored the terrain within which the
co-production of knowledge now engages, and all three approaches share key
principles related to collaboration between academic and non-academic researchers.
These methodologies all engage with academic contributions to social justice through
strengthening the voice of disadvantaged populations. This discussion reflects our
shared belief that there is no formula to secure the co-production of knowledge. We
share perspectives from our collective experiences to help both academics and non-
academics build relations that reflect shared aspirations to generate knowledge that
advances social justice and an equitable urban future for all. This is – in its present
form – an academic working paper. But it has been co-authored by non-academics and
we recognise that its themes speak to the interests of non-academics. To widen our
reach, we plan further dissemination once our ideas have been refined through this
working paper.

Section 4 is structured using four themes. The first is the underlying theory of change
used by academics and social movement activists, and how theories of change link to
methodologies of knowledge production and use. The second is that of education, and

2
The SDI co-authors are working within the professional support agencies. We also quote
leaders of grassroots federations to acknowledge the contribution they have made.

www.gdi.manchester.ac.uk 4

Electronic copy available at: https://ssrn.com/abstract=3470133

the opportunity for collaboration to educate students in the central contribution of low-
income community organisations to transforming urban development practice. The third
is that of academic status, and the challenges that status hierarchies mean for those
making efforts to co-produce knowledge between academics and non-academics. The
fourth theme is that of institutional engagements and the development of longer-term
relations as a process that changes outcomes. Section 5 concludes with a summary of
the discussion and a reflection on knowledge and power.

2. Methodology

Empowerment is, in part, the “exercise of informed choice within an expanding

framework of information, knowledge and analysis … a process which must enable
women to discover new possibilities, new options …. a growing repertoire of choices”
(Batliwala, quoted in Rowlands, 1997, p 23, original emphasis).

This paper has been a long time in the making. It draws specifically on the
experiences of all the authors through both formal research and experiential
engagements that have taken place over the past six years. Its topic – how to co-
produce knowledge – is one theme within an ongoing research network which
combines academics from the UK and those from Kenya, South Africa and Zimbabwe,
along with SDI affiliates in those three countries. Funded by the Leverhulme Trust to
understand how participatory planning in African cities can be scaled, refining the co-
production of knowledge is an essential component of the work of this network.

SDI – formally known as Shack/Slum Dwellers International – is a transnational

network of women-led savings groups based in informal settlements that come
together into city and national federations. SDI affiliates share organising practices,
including savings-based organisation, community-to-community (peer) exchanges, and
community-led enumeration (mapping and profiling settlements, and household
surveys). SDI both develops and recognises its own knowledge production processes,
and reaches out to local academic departments to build collaborative alliances in order
to change practices. SDI affiliates recognise the significance of academic knowledge,
both in defining problems and solutions and, more generally, in conceptualising and
theorising urbanisation, urban economic growth and urban development. However,
they recognise that research – as defined by academics – may not align with their
experiences and goals. Hence, they have sought to develop their own knowledge
capabilities around specific strategies and agendas (see, for example, Patel et al,
2012; Appadurai, 2012). SDI’s Indian Alliance (SPARC, 2014, p 7) frames its own
perspective on both academic and non-academic research thus:

Clearly, knowledge creation and its ownership and the right to research remain
foremost the right to research, to explore, to classify, to analyse, to verify and to
extract learning and knowledge from that process remains a critical survival
challenge. Many definitions regarding what is poverty, what is chronic

www.gdi.manchester.ac.uk 5

Electronic copy available at: https://ssrn.com/abstract=3470133

 deprivation, and many seminal definitions that drive development investments
get formulated on the basis of global discourses that leave the poor, the very
subjects of this discussion, outside the process.

This quote highlights the frustration felt by a social movement whose members
experience the deprivations that others study and then define. This is manifest in
relational tension as activists collaborate with academics who are needed to engage
with these debates, but who are embedded in processes that are – by language, place
and status – exclusionary. As both academics and non-academics, we recognise the
vulnerability of urban social movements towards academics who come and do
research on them, rather than with them. SDI, and other civil society organisations, fear
academics misrepresent their realities, generate knowledge that adversely affects their
work and mission, and are insufficiently respectful of the relations they have with a
range of stakeholders. Hence our motivation as academics and professionals in
exploring alternative research practices.

The academic authors come from a range of disciplines (architecture, development

studies and planning) and are well versed in critical social science, with research
experience in a range of geographical contexts. The non-academic authors are
professionals working with the NGO support organisations for SDI affiliates. All the
academic authors have been involved in research projects that have sought to co-
produce knowledge with social movements. The discussion below is informed by the
research relations that we have been directly involved in and those we have observed.
We have not sought a tight definition of the co-production of knowledge that would
restrict the research relations that we explore. Rather, we consider relations that have
used the language of knowledge co-production, participation and partnership.

The discussion uses information obtained through semi-structured interviews about the
co-production of knowledge and academic and social movement relations,3 through
participation in academic engagements by SDI (and other social movements), through
informal consultations about how to resolve tensions in SDI and academic relations
(partially captured through email exchanges), and through participation (sometimes
jointly between authors) in research projects with SDI. While the focus on SDI is
limiting, as it is only one example of a transnational social movement, this is mitigated
by the fact that concentration on these experiences offers depth to our analysis. As we
argue below, it is the longevity of an engagement that offers insights into research
processes.

Particularly notable exchanges took place with SDI’s Indian affiliate in 2012, and during
a network meeting with SDI participants at the World Urban Forum in 2014 (preceded
by interviews with both Zambian and Zimbabwean SDI participants in the Forum). A
workshop on impact at the Global Development Institute in Manchester (2017) brought
together civil society scholars (including SDI professionals) and academics to explore

3
While most interviewees have been federation leaders, we have also drawn on the
experiences of a small number of local government officials.

www.gdi.manchester.ac.uk 6

Electronic copy available at: https://ssrn.com/abstract=3470133

relevant issues. Three research team reflections took place in Uganda in 2016
(Kampala, Mbale and Kabale) between academics, NGO professionals and community
activists involved in co-producing knowledge about one national-level urban
development programme. Two recent group discussions, one with a Nairobi-based
team of three (academic, professional, community leader) in March 2017 and the
second involving a larger combined team from four cities (three academics, four
professionals, three community activists) in March 2018 added experiences and
analysis. Conference panels that have focused on this and related themes include two
sessions at the RC21 conference in Leeds (2017); a double panel on participation in
towns and cities of the Global South at the Development Studies Association in
Bradford (2017); and a special session on the co-production of knowledge with
academic and SDI presentations at the Royal Geographic Society/Institute of British
Geographers conference in 2014. A first draft of this paper was prepared in September
2018 and was shared with all the authors, who added and edited text.

Before beginning with the review of the literature and discussion of challenges, we
acknowledge the benefits of collaboration. While the focus on challenges is necessary
to move forward this methodological field, our findings are not overwhelmingly
negative. SDI activists recognise that academics have given them useful tools and
skills, and represented local realities in ways that have helped to legitimate support
from government agencies. Before beginning on the literature review and analysis of
tensions, we illustrate some of these benefits for one of SDI’s affiliates, the Muungano
Alliance in Kenya (Box 1).

Box 1: The benefits of academic collaboration for the Muungano Alliance

The Muungano Alliance in Kenya has acknowledged expertise on Geographic Information

Systems (GIS). Staff at SDI Kenya (the support NGO for the Muungano Alliance) explain that
they learned many skills from university students. Skills around settlement profiles,
enumerations and mapping using GIS have drawn on academic expertise and produced
community cadres with critical skills that contribute to settlement upgrading processes. Besides
building local capacity in informal settlement neighbourhoods, this process has also helped –
across the SDI network – to challenge the ‘expertisation’ of urban planning.

SDI Kenya recognises the benefits of academic support on the zonal plan for Mathare informal
settlement; academics helped to conceptualise forms of neighbourhood development and
understand how to connect local infrastructure into city networks. At that time, residents and
Muungano members felt that tenure issues were too sensitive to be discussed within the
settlement. Working on infrastructure led to strategic links with utilities and local government
and, with academic help, the Muungano Alliance built successfully on existing practices to
advance access to essential services. Most recently, with support from the Strathmore Business
School, they have been able to articulate the ’poverty penalty‘, or the additional cost paid due to
the inaccessibility of formal services. Hence SDI-Kenya recognises the role of academics in
building skills (GIS), changing an understanding of what is required within the Alliance (city-wide
infrastructure connections), and external representation of local realities (organic and inorganic
settlement forms and the ’poverty penalty’).

www.gdi.manchester.ac.uk 7

Electronic copy available at: https://ssrn.com/abstract=3470133

Analysing the co-production of knowledge is inherently difficult (Anderson et al, 2013).
Relations inevitably involve complex interactions between those involved, and are
influenced by the external context. As opportunities open and close, and as outcomes
change and understandings deepen, the fallacy of measuring success and failure at
any moment it time through ‘snapshots’ is exposed. We seek to elucidate issues, share
experiences and suggest ways forward based on long-standing engagements. The
section below shares ideas from the broader literature before our findings are
presented in Section 4.

3. Methodology

Understanding academic and community knowledge relations: what the literature

contributes

There is a politics of urban knowledge because urban knowledge is political.

(Madden, 2015, p 300)

Knowledge democracy is about intentionally linking values of justice, fairness

and action to the process of creating and using knowledge. (Tandon et al, 2016,
p 23)

The co-production of knowledge has a substantive heritage linked to debates about

participation and research (Gaventa & Cornwall, 2015; Tandon et al, 2016; Kara,
2017). There is a strong overlap between earlier traditions of both action research and
participatory research, and evolving practices of the co-production of knowledge.
Participatory research has been motivated by academics committed to processes of
inclusion; it has also been catalysed by social movement activists and academics who
recognise that action is a catalyst for learning and that securing social justice requires
rigorous learning and knowledge building (Burns et al, 2012; Gaventa & Cornwall,
2015). These practices have gone beyond drawing disadvantaged communities into
academic research processes and have recognised the insights provided by alternative
forms of knowledge and the potential, particularly of indigenous knowledge, to advance
understandings of the natural world and of human–nature interaction (Tuhiwai Smith,
2012). Traditions of action research and participatory research share with the co-
production of knowledge the commitment to move beyond the academy and to
challenge and reform research practices (Gaventa & Cornwall, 2015). This is
particularly aligned with the academic discipline of development studies, which takes a
normative standpoint and is motivated by values of social justice, equity, equality,
empowerment and associated resource redistribution. In summary, these research
approaches directly address social justice by providing more equitable spaces for
knowledge generation, they offer the potential for new knowledge (methodological,
theoretical, conceptual, empirical) about social justice, and promote the development of

www.gdi.manchester.ac.uk 8

Electronic copy available at: https://ssrn.com/abstract=3470133

new capabilities among all participants in the process (Gaventa & Tandon, 2010;
Oldfield, 2015).

The co-production of knowledge builds on shared research traditions to advance our

understanding about how disadvantaged groups can be active partners in
understanding the city (Van Ewijk & Baud, 2009; Beebeejaun et al, 2015; Jacobs et al,
2015; Openjuru et al, 2015). For us, the co-production of research begins with a
recognition of the value of different forms of knowledge and the right of disadvantaged
groups to be centrally involved in research on urban poverty and inequality. Knowledge
that is co-produced engages with alternative understandings, builds new research
capabilities, jointly tests different explanations, and negotiates research processes. We
recognise that knowledge generation has the potential to exacerbate as well as
address inequalities and injustices.

The uneven power relations between types of knowledge producers are immediately
apparent from any engagement with this literature. Freire (2000) gained a global
audience with his volume, The Pedagogy of the Oppressed, in which he sought to
define a learning space that challenged exclusionary knowledge institutions, and
realised a pedagogy for liberation. Academics have responded to these critical
reflections by developing new relations and methodological approaches that challenge
their own dominance (Tandon et al, 2016; Bell & Pahl, 2018). Recently, in the context
of the need to ‘decolonise knowledge’, they have critiqued the dominance of academic
knowledge (Baldwin, 2017) and recognised the need to ensure that multiple forms of
knowledge are acknowledged within universities (Noxolo, 2017) as well as beyond.
There are therefore now multiple and long-standing efforts to secure alternative
knowledge production processes. Gaventa and Cornwall (2015, p 465) argue that
“understandings of the relationship of knowledge and power in the participatory
research process have become more nuanced, taking into account the complexity and
contingency of power relations”. However, this appears optimistic. Whatever the quality
of understanding about issues of power, outcomes are acknowledged to be mixed
(Openjuru et al, 2015), with academics and their institutions dominating processes of
knowledge generation (Standing & Taylor, 2016; Tandon et al, 2016).

Gaventa and Cornwall (2015, p 466) and McFarlane (2006) suggest tensions are
grounded in debates between a positivist social science methodology seeking to
establish objective universal fact, and methodologies that value multiple perspectives
and voices. That is, the power of academics to dominate is based on specific research
methodologies. This conclusion is challenged from two directions. Holland (2013)
argues that those generating participatory statistics need to meet the challenge laid
down by positivist quantitative academic researchers and adopt principles of
standardisation and comparability. While recognising the tension between breadth and
depth, Holland argues that participatory statistics can work within the boundaries of the
‘representative sample’, by implication avoiding marginalisation. The second challenge
is that, as discussed below, methodological approaches such as interpretive sociology
and anthropology, while recognising the significance of alternative perspectives, may

www.gdi.manchester.ac.uk 9

Electronic copy available at: https://ssrn.com/abstract=3470133

not necessarily be participative in any meaningful way (Jacobs et al, 2015). The
alternative nature of social movement – and specifically SDI – knowledge has been
recognised but, as McFarlane (2006, p 289) notes, little attention has been paid to “the
ontological and epistemological basis of [alternative forms of] knowledge”.

In summary, the inequalities in power relations have been raised but not resolved.
While such inequalities may be in part related to an implicit hierarchy of research
methodologies, they go beyond this distinction.

Participatory research appears to be the route through which community knowledge is

valued as equal to academic knowledge. However, SDI’s Indian-affiliated NGO, the
Society for the Promotion of Area Resource Centres (SPARC) (2014), argues that even
the label of participatory research is misused and that frequently the right of all groups
in society to be involved in, and set the terms for, joint knowledge generation is not
realised. While participatory research implies the involvement of those on whom the
research is focused, as a term it says little about how substantive the involvement is
(Brown-Luthango, 2013; Bell & Pahl, 2017; Simon et al, 2018). SPARC (2014, p 7)
elaborates on its experience:

Informing the poor that a research is being ‘conducted’ for ‘their good’ by others
is often called ‘participatory’ research. We clearly have to move towards being
proactive and leading the process based on defining our own needs. We
continue to have to constantly defend the rights of the poor to ‘the right to
research for change’, for assessment of what is being done and for these rights
to not be treated as objects of research by others…. The interpretation of
participatory research for us accompanies the right to research where the poor
define, own and execute the research, test its findings and create knowledge
that gets embedded in their development processes.

Appadurai (2012), writing on the experiences of SDI’s Indian Alliance, highlights the
potential of its data collection to address exploitation and dispossession, empowering
communities by enhancing self-knowledge and therefore conscious identity. A tool for
organisation as well as documentation, community-led data collection takes ‘power
away from external agencies such as the state and puts it back where it truly belongs,
which is within the community itself’ (p 640). What is also notable about Appadurai’s
discussion is the absence of academic relations and academic knowledge. The co-
production of knowledge – with its explicit recognition that multiple parts of the research
process have to include non-academics on an equal basis, appears to be a useful
advance on practices of participatory research.

Bell and Pahl (2017, p 105) argue that the co-production of knowledge both advances
social justice and ‘destabilizes academia as a privileged site for the production and
dissemination of knowledge’. However, their optimism about the radical nature of
knowledge co-production is not evident in other experiences. For example, Brown-
Luthango (2015, p 316), describing a collaboration in Cape Town between academics
and informal settlement communities, argues that the process was ‘a university

www.gdi.manchester.ac.uk 10

Electronic copy available at: https://ssrn.com/abstract=3470133

conceived and driven process’. The experience of being peripheral regularly emerges
from social movement activist accounts of participation in academic activities.4

The nature of urban development and its relations with academia and professional
training is relevant here. Urban development is highly professionalised, with architects,
planners and engineers setting standards for urban development and influencing
outcomes, frequently with negative impacts (Songsore & McGranahan, 1998; Myers,
2003). Traditional academic roles are to advance understanding – through research –
and to train, accredit and legitimate these professional cadres. The rule-based,
regulation-driven nature of urban development under capitalism (Escobar, 1992) is
motivated both to manage the agglomeration of economic activities taking place in
urban areas and to control exploited and disadvantaged urban populations. This
professionalised control over urban space means that social movements have to
navigate the ‘expertise’ when negotiating for improved access to tenure security and
basic services (Mitlin, 2013; Oldfield, 2015). The urban context is deeply political and
that includes the ways in which contractions and exclusions are understood, and how
interventions are designed and implemented (Oldfield, 2015). Upgrading, for example,
involves changes in land allocations and potentially leads to the allocation of significant
assets to at least some households. Urban social movements are significantly
disadvantaged when excluded from research and knowledge generation processes
that set standards (SPARC, 2014). In this context, community participation in project
implementation (for example, informal settlement upgrading) will not secure more just
outcomes; rather, new models of urban development are required, and if these new
models are to be relevant to low-income informal communities, the latter’s involvement
in programme design is essential (Jacobs et al, 2015; Burra et al, 2018). Research
partnerships need to take both the difficulties in and opportunities for transformative
outcomes into account. Drawing from the experiences of one network of academics
and non-academics in Kenya, South Africa, Sweden and the UK, Simon et al (2018)
argue that their outcomes were limited by a lack of influence over local political
processes, and the difficulties of institutionalising progress towards transformative
cities. But movement involved in research – in co-producing knowledge – is intended to
challenge political processes and outcomes, and to enable new approaches to urban
development that disturb and contest present exclusionary outcome (King & Kasaija
2018; Burra et al, 2018).

To improve academic contributions to transformative knowledge, critical theorists such

as Escobar (1992, 2018) have challenged academics to reflect on the ways in which
knowledge about development reinforces existing power hierarchies and excludes low-
income and disadvantaged groups from being able to develop and legitimate their own
options. Chambers (1995, p 199), a leading protagonist in rural participatory research,
argues that professionals should be enlightened such that they give up their power and

4
http://www.rioonwatch.org/?p=46223. In a conference in Rio de Janeiro, only eight favela
residents were invited to take part. One of the movement leaders explained that the lack of
collective participation reinforces their understanding that grassroots knowledge is not valued by
academic institutions.

www.gdi.manchester.ac.uk 11

Electronic copy available at: https://ssrn.com/abstract=3470133

create space for less powerful community knowledge providers with new and better
means of research. Anderson et al (2013) analyse one recent experience with the co-
production of knowledge, the CityLab process in Cape Town that sought to “broker
inter-disciplinary engagement, both across academic disciplines and between the
academy and broader society” (p 2). They conclude that “success is reliant on the
development of mutual respect, trust and reciprocity” (p 8), emphasising the
significance of the personal qualities of those involved. Personal commitment may not
be enough, however; Chambers (1995, p 203) himself notes the importance of
downward accountability and the need for systemic reform, as well as improved
personal orientation.

Despite the significance of individuals, it is an institutional response that is required.

Co-producing knowledge must move beyond an engagement between committed
individuals to sustained relations between agencies (departments, movements) (see,
for example, Oswald et al, 2016) and a change in institutions of knowledge production
(defined as norms and practices of behaviour). Universities need to change their
objectives and the expectations and incentives placed on staff to engage more
equitably with local communities (Brown-Luthango, 2013), and deal with issues of
accountability (Oldfield, 2015). Relatively little attention has been given to the nature of
that institutional response in the context of urban development in the Global South
(Tandon et al, 2016).5 SDI’s Indian Alliance works with an NGO called Participatory
Research in Asia (PRIA).

PRIA’s director, Rajesh Tandon, acknowledges that Indian higher education has not
supported processes of participatory knowledge generation (Openjuru et al, 2015), and
argues that universities need to engage with disadvantaged communities in research,
teaching and service (Tandon et al, 2016). As social movement activists engage more
with the academy, the potential contribution of such alliances becomes more evident.

Moving from research into collaboration in teaching identifies new complementarities,

although academic needs may dominate here also. Tensions may be particularly acute
when the teaching programme is designed to fulfil the requirements of professions.
Winkler (2013) discusses tensions between the objectives of academics and
community members through ‘service learning’ in South Africa,6 while Brown-Luthango
(2013) references similar difficulties. Tensions in relations also emerge through an
analysis of dissemination practices. These need to move beyond peer-reviewed
publications and be jointly determined by those involved in the research. Researchers
concerned to co-produce knowledge need to be sensitive to accessibility of meeting
locations and publication outlets, and need to choose venues that encourage debate
(Burns et al, 2016; Bell and Pahl 2017). A second issue is that of authorship: the
contribution of non-academics must be recognised (Openjuru et al, 2015). Agreements

5
Although the planning community has reflected on this in the context of the Global North. See,
for example, Baum (2001).
6
Service learning requires university students to participate in community projects in exchange
for course credits (Brown Luthango, 2015, p 314).

www.gdi.manchester.ac.uk 12

Electronic copy available at: https://ssrn.com/abstract=3470133

about who is credited and how they are credited are critical (Oswald et al, 2016).
Despite these challenges, such activities offer the potential for locally grounded
engagement with non-academic knowledge production, and for the contribution of more
place-based and collaborative approaches put forward by social movements and
community-based organisations (Escobar, 2018).

While recognising the significance of institutionalisation, we acknowledge that

academics may have their own experiences as movement activists. Academics may
also have their own experiences of marginalisation and exclusion. They may be
involved in other struggles (such as gender or the environment) in which they have
activist roles. Equally, social movement activists may be, or aspire to be, scholars
and/or professionals themselves. Moreover, irrespective of the ‘starting point’,
individuals – activist and academic – change through the processes of co-producing
research (Oldfield, 2015). As noted by Gillian and Pickerill (2012), this can make an
understanding of positionality more complex for both researchers and activists, but it
offers the opportunity to strengthen processes of knowledge co-production as personal
capabilities and understandings change.

The broader context can also be significant. In the UK, the Economic and Social
Research Council and the Arts and Humanities Research Council have responded to
requests from the research community with programmes to support the integration of
non-academics into research, while the recent reorganisation of the Research Councils
itself reflects the government’s objective for research to be useful beyond the academic
field. The UK Research Council recognises the potential significance of the co-
production of knowledge, but it is not evident that the grant-making process is fit for
purpose (Bell & Pahl, 2017).7 The emphasis – at least in UK academia – on impact has
encouraged academics to take the co-production of knowledge more seriously and has
incentivised these efforts (Green, 2017).8 Alongside these changes, greater emphasis
is being placed (at least in the Global North) on ethical research. While some
associated measures have been focused on risk mitigation and compliance with the
demands of insurers, these debates raise questions about ethical frameworks that are
appropriate for a range of research collaborations (Gillian & Pickerill, 2012).

Underlying the co-production of knowledge is a conceptualisation of knowledge

democracy in which knowledge is acknowledged “as a shared resource, jointly
generated and publicly owned” (Miller et al, 2006, p 14). However, despite academic
commitment, there are multiple challenges to the realisation of this goal, ranging from
the fundamental (theories of how individuals and agencies catalyse change, deeply
stratified urban relations), to the operational (how the remuneration and authorship are
decided).

7
Bell and Pahl (2017) suggest that funding bodies need to give greater attention to the quality
of established relationships when assessing bids.
8
Funders of development research increasingly want to see and are willing to fund ‘uptake’
activities to maximise the chances of projects achieving impact. This does tend to incentivise
that project-focused impact, which may undermine the prospects for broader, transformative
change in the longer term. See Jordan (2017).

www.gdi.manchester.ac.uk 13

Electronic copy available at: https://ssrn.com/abstract=3470133

4. Co-production and its realisation in urban research
The university – they have no impact – they just do their own. The university does
research; it is of no use to us.9

These words were spoken by Felitza, community leader from Nairobi, when she was
asked about her experience of working with academics. Felitza went on to identify
positive benefits emerging from collaboration, but in her eyes this collaboration was
different from academic research. Building on this and other interviews, this section
explores the practice of co-production.

4.1. Knowledge and activism: the theory of change

The goal of research is not the interpretation of the world, but the organisation
of transformation.10

Most of the academics who engage with SDI and other social movements do so
because they wish to support more equitable and inclusive cities that are more
effective in addressing the needs of low-income and disadvantaged residents. They
believe, for the most part, that their research contribution is to promote change through
research projects and associated documentation, including learned papers, policy
briefs, etc. That is, their theory of change is that rigorous evidence of established
quality will change adverse outcomes, either by identifying contradictions (that
potentially catalyse action), and/or by elaborating problems (and solutions to those
problems) that politicians and/or officials have not previously accepted, because of the
lack of such evidence. This requires that the methodology be adequate for the purpose
according to academic criteria, which leads to their findings being accepted as an
accurate summary of conditions and/or needs in these neighbourhoods and urban
areas. Academics both work with social movements to produce these knowledge
findings, and build the capability of movement activists to participate in this research.

This theory of change is very different from that of social movements (including when
they engage with data collection and analysis). Social movements secure change by
building mass organisations that gain influence because of the implicit or explicit threat
of disruption and/or electoral opportunity. Knowledge is an important legitimator of the
redistribution of resources, and of development models that address poverty and
inequality. While SDI movements have placed a considerable emphasis on the co-
production of models to secure tenure and deliver basic services with local
government, this strategy does not assume that the state is committed to pro-poor
development. Rather, it is premised on the understanding that low-income groups gain
little from confrontation, because they have limited protection from the worst abuses of
power, and/or because negative representations of informal settlement dwellers have

9
Felitza, April 2017
10
Antonio Conti, quoted in Castree et al (2009).

www.gdi.manchester.ac.uk 14

Electronic copy available at: https://ssrn.com/abstract=3470133

been used to undermine their claims to resources by de-legitimatising their right to
entitlements (Patel & Mitlin, 2009; Mitlin, 2018). Co-production helps to build positive
relations with individuals in government who are willing to support more inclusive urban
development; however, the need to mobilise large numbers of disadvantaged citizens
to engage with solutions remains key (Gaventa and Cornwall, 2015; Lines & Makau,
2018; SDI, 2018). Hence knowledge generation is focused on citizen mobilisation.

Academics may, of course, subscribe to the significance of mass mobilisation and be

cautious about their capacity to advance change through their intellectual contribution.
It is not unusual for academics to be a part of social movements – for example, those
concerned with gender and environmental justice – alongside their academic
employment. They may seek to elaborate the contradictions of urban development
processes or to support mobilisers with facts that support their position. But their
contribution – at least in terms of social justice for informal settlement residents – is not
that of being a mobiliser. Irrespective of which underlying theory of change is applied,
academics may build the understanding of movement activists with respect to current
government policy and programming perspectives, and hence enable them to position
their issues and better organise to amplify their voice. And they may build their
technical skills to collect data, develop projects and undertake other development
activities (Bennett, 2018a).

Note this distinction is not between positivist ‘rational’ scientific knowledge and more
qualitative and interpretive social science and/or community knowledge (see Section 3
above). Academics may argue that qualitative methods, such as life histories, are
needed to strengthen the voice of activists. However, engagements with SDI identify
the difficulties with such assumptions. Life histories are not necessarily part of a co-
produced research project. Indeed, life histories may reinforce the role of the
movement activist as a research object, whose history is to be extracted and
presented. Methodologies and methods are less significant than the shared research
processes, and the shared understanding of how knowledge changes outcomes.

The alternative theories of change underpinning academic and non-academic

perspectives on research can be seen in their comparative approaches to data
collection and sampling. Academics use sampling, both in terms of quantitative and
qualitative analysis, to ensure that resources are used to best effect. They draw on
probability theory (in the case of quantitative work) and alternative techniques, such as
snowballing (for qualitative studies), to ensure that research funding is well allocated.
For SDI, the objective of data collection is both to inform a significant number of
citizens about the situation and to use that knowledge to mobilise them into the
movement, ie to encourage them to become actively engaged in pushing for political
change. Sampling is anathema to mobilisation. By excluding some citizens from
interviews, sampling perpetuates externally determined divisions within communities
and potentially leads to mistrust. SDI data-collection processes involve meetings
alongside data collection, sharing findings about the lack of tenure and services in the
locality, and engaging residents in action. They are designed to address a context in

www.gdi.manchester.ac.uk 15

Electronic copy available at: https://ssrn.com/abstract=3470133

which, in the words of one SDI community leader, “People living in informal settlement
[are] misrepresented – voiceless and faceless. For us – the question is how to use
data to transform, and how to bring that transformation into their lives and
settlements.”11 We are not arguing that all knowledge co-produced between academics
and social movements must involve surveys with 100% coverage. Rather, we are
highlighting differences in understanding as to which knowledge processes catalyse
change. In a context of unequal power, the perspectives of social movements are not
considered equally. Acknowledging that research activities have implications for social
movement mobilisation is a step towards inclusion and recognition.

Further tensions that highlight the centrality of interactions between information and
mobilisation are illustrated by three recent Kenyan experiences. All three examples
involve academics who consider themselves to be co-producing research with the
social movements. The first example concerns a discussion about land sharing for an
SDI Federation group in Kenya seeking secure tenure of the land they occupy. An
academic proposed that the whole group move onto half the land, selling the other half
to finance the development. While this appeared a clever way forward for the
academic, the Federation cancelled a community meeting to discuss this option. The
Federation did this because the tenants within the settlement were not yet strong
enough to negotiate their inclusion; they needed more time to strengthen their local
organisations and ensure that their claim for inclusion would be successful.

The second example involves a proposal for a research project to take soil samples
within a Nairobi neighbourhood to demonstrate the health risks and possible toxic
contamination from adjacent factories. The local Muungano Alliance feared that
evidence of contamination would strengthen the position of those wanting to evict the
local community. Given that the residents did not wish to move, but were not yet strong
enough to secure a clean-up, the social movement suggested that the research be
postponed.

The third example concerns academic research in one neighbourhood, which reported
that only 4% of residents were interested in upgrading. The Alliance wanted to delay
sharing this information because, they argued, it reflected a context of extreme
insecurity. Once the community was strengthened – they argued – then it would be
more confident of its ability to negotiate access to land and would report to interviewers
that it intended to stay. Tabling this information now would simply increase pressure for
displacement.

In all three cases, the Muungano Alliance negotiated with the academics involved to
change their perspective. These examples point to the significance of the timing of
research, presentation of knowledge and the development on new options that respond
to the needs of informal settlement residents. Research needs to be sensitive to the
possibility of adverse information and the requirement that movements defend the

11
Joseph Muturi, SDI Core Group member and leader of Muungano wa Wanavijiji, speaking at
‘Leaving no-one behind: how can we better monitor progress in “slum” areas?’. Overseas
Development Institute networking event at the World Urban Forum, 8 April 2014.

www.gdi.manchester.ac.uk 16

Electronic copy available at: https://ssrn.com/abstract=3470133

needs and interests of their members. If the goal of research is to organise the
transformation of society, then the co-production of knowledge requires attention to the
relative contributions of, and interactions between, knowledge and mobilisation. Close
engagement between movement activists and academics helps to reconcile alternative
perspectives about the contribution of knowledge generation to pro-poor change. We
discuss modalities of long-term collaboration in the sub-section on institutional
solutions to accountabilities below.

Alternative perspectives on how pro-poor change can be achieved are partially

addressed through joint identification of research objectives and other parts of the
research process. As elaborated below, academic involvement in community-led data
collection helps to legitimate the work of SDI affiliates. Working with academics helps
SDI affiliates to engage with local and national governments, who consider academic
validation of community-led data to be significant. When technical skills are required,
academics (staff and students) may offer these skills at relatively low cost and/or with a
sensitivity to the communities’ contribution. Co-learning about government policies and
programmes offers a way to extend the understanding of Federation leaders and NGO
professionals. And this points to the importance of structured collaboration between
academics and movements, and clearer accountabilities within these relationships. We
return to these themes after discussing collaboration through teaching and the unequal
social status of academics and activists.

4.2. Improving education: redefining professionalism

One of the major challenges for SDI affiliates as urban movements is the entrenched
way in which professionals guide urban development processes. Professional visions
and practices influence state programmes, regulations and standards and frequently
lead to unequal and anti-poor outcomes. Hence a major motivation for SDI
engagement with academics is the chance to influence the teaching programmes that
train urban professionals. As a strategy for change, such activities influence future
generations of professionals and academics to make contributions that they would not
otherwise have done with respect to participation, inclusion and poverty reduction.12 As
SPARC (2014, p 10) states: “With academics and research agencies, the main
purpose … is to embed knowledge that works for the poor into mainstream education
processes.” Engaging with universities and colleges offers SDI affiliates the opportunity
to add to students’ skill sets and prepare them to work with communities in the future.13
The value that SDI affiliates place on their contributions to education is evidenced by
their partnership with the African Association of Planning Schools (AAPS).14 Teaching-
based partnerships are also a way in which academics seek to influence urban

12
Kanbur (2012) describes the use of exposure programmes to educate development
professionals.
13
Sekai Chirembe, Medellin, 10 April 2014.
14
SDI blog, by Peoples Process on Housing and Poverty in Zambia (PPHPZ) and the University
of Zambia (UNZ), ‘Shaping human settlements through partnerships between slum dwellers and
academia’, 2 May 2014. .

www.gdi.manchester.ac.uk 17

Electronic copy available at: https://ssrn.com/abstract=3470133

transformation; some academics build partnerships with social movements to validate
their work to advance social justice and tackle inequality, and to educate students in
this regard. Box 2 illustrates some immediate benefits in terms of generating useful
knowledge.

Box 2: The benefits of engaging in teaching

The SDI Alliance in Zimbabwe began to expose students to their work many years ago.
They recognise that there are immediate benefits. Students began to generate
dissertations topics that address gaps in the academic literature emanating from limited
academic understanding of the realities faced by low-income urban residents. These
dissertations also documented the key role of the Zimbabwe Homeless People’s
Federation in transforming urban environments.

In addition to achieving the dual goals of changing the understanding of students about
what needs to happen in urban development, and capacitating them to work with
communities, such engagements can potentially lead to new ventures. In South Africa,
the inception of 1to1 (now a social enterprise) developed from a student project to
support the grassroots development of South African cities.

SDI affiliates have seen students whom they have taught go on to take up significant
positions in government. Students keep in touch and SDI affiliates are confident that
many have developed a good understanding of, and support for, a community-led
approach. For example, a senior official in planning within Nakuru County, who worked
with SDI Kenya as a student at the University of Nairobi, waives inspection and
approval fees for Federation-led greenfield developments. More generally, federation
leaders appreciate being recognised for their expertise and are positive about these
experiences.

However, SDI’s experiences in educating students raises challenging issues. Studios

and exposure visits may not be part of the formal learning process and hence may not
be accredited; this immediately downgrades them in the eyes of both staff and
students, making them optional extras. On some occasions, planned AAPS studios did
not run because of a lack of finance.15 This is a further indication that their value is not
recognised by departments, although it may be recognised by individual academics. In
terms of student contributions to community planning needs, academics may determine
the timing according to the academic programme; hence the contribution may have
limited value. If not appropriately facilitated, students may appear as experts and
unduly influence local processes, and/or may exhibit inappropriate attitudes and
struggle to deal with issues related to their own position of social privilege (Bennett,
2018a). When professional training requires some level of ‘practical’ work and/or an
optional component that may be selected by interested students, it is particularly
difficult to ensure that local organisations and their members benefit. The problem is
15
Notes from exchange with Gobabis, May 2017; prepared by Guillermo Delgado.

www.gdi.manchester.ac.uk 18

Electronic copy available at: https://ssrn.com/abstract=3470133

not simply one of timing; academic needs may dominate the design of the engagement
(Bennett, 2018a; also see Winkler, 2013). The temporary nature of student inputs
raises challenges for the staff managing this process (Bennett, 2018a). At the same
time, the emphasis of service learning on accreditation may prevent a holistic response
to local needs (Bennett, 2018b). This points to the need for more radical redesign of
teaching and an emphasis on ‘un-learning’ negative practices, as well as developing
more positive approaches.

Reflections in the literature reflect experiences in South Africa, where high levels of
professionalisation combine with the commitment of some scholars to engage with the
needs of low-income communities and, in the context of decolonisation debates, with
radical critiques of academic contributions. There is a need to bring critical social
science into other disciplines, including those that focus on spatial development
(architecture, planning, spatial design) (Bennett, 2018a). The significance of long-term
relations to enable engagement with community priorities is also noted (Bennett,
2018a).

Experiences in Manchester identified a deficit in vision and developed an alternative

approach.

Box 3. Community-led postgraduate teaching in Manchester

Since 2010, community leaders from SDI affiliates have visited the Global
Development Institute (GDI) at the University of Manchester for a week-long
contribution to a Masters class in Citizen Led Development. Community leaders
deliver 60% of the lectures. The class begins with the students being introduced to
academic literature on urban poverty and informal settlement upgrading. Then the
community educators arrive. Several days of lectures elaborating the SDI local process
expose students to the realities of urban poverty and the work of SDI activists. After
developing the course unit with the SDI Alliance in Zimbabwe, community leaders from
South Africa and Kenya (plus video-conferencing with Ugandan and Kenyan leaders,
as a result of visa refusals) were willing to share their expertise.

GDI staff consider these contributions essential for a Masters education in

development studies, and beneficial to the co-production of knowledge. As community
members and academics engage to share perspectives from experience in the context
of the academic literature, theories and conceptual frameworks are challenged and
unpicked. This demonstrates the potential to align teaching and knowledge generation.
Students are exposed to the mismatch between abstract theories and grounded
realities, and the gap between academic generalisation and local specificities. Such an
exposure challenges students and academic staff alike. Students become active
participants in the co-production of knowledge as they share their own reflections. As
tensions in the literature are exposed, community leaders contribute their own
experiences and in so doing demonstrate their capability to contribute further to
advancing knowledge.

www.gdi.manchester.ac.uk 19

Electronic copy available at: https://ssrn.com/abstract=3470133

SDI affiliates recognise that planning education in many African universities is not fit for
purpose, with little attention paid to the challenges of informal settlements. This implies
curriculum change.16 However, the affiliates’ experience is that academics are not be
willing or able to take on their university’s institutional process and change the
curriculum to accord recognition to community lecturers as knowledge providers (see
also SPARC, 2014).17 In Zimbabwe, building on the experience with the University of
Manchester, the SDI Alliance has sought to advance this agenda and replicate this
course unit. Efforts with the University of Bulawayo have exposed challenges. While
the name was kept when a new course unit was set up, there was no platform for
community activists and educators; some academics have been reluctant to give up
their role at the front of the classroom. Further engagements have been successful and
have resulted in plans to set up a new degree programme. Efforts with the University of
Nairobi have highlighted that curriculum change needs to be a priority for senior staff
members. This points once more to the importance of longer-term established relations
that extend beyond individual academics.

4.3. Knowledge and status: the siren call

The thing is that it is not that they do anything special. It is … like putting
powder on someone’s face. Does not change who we are, but changes the look
of it.18

SDI Federation members are clear that one of their key motivations for collaborating
with academics is their status as acknowledged ‘experts’; this is considered to enhance
the legitimacy of SDI Federation data and the knowledge generation process. The
contribution of academics to their data collection – according to one community leader
from Zimbabwe – “gives a punch”.19 The circles in which academics operate–
particularly with local government – open “an avenue to the city that was different from
the traditional ways that we spoke to the city”.20 Unequal status in the context of
academia also reflects itself through micro-level engagements. One of the community
lecturers visiting Manchester in 2011 spoke about her mother cleaning at the University
of Harare to indicate her own pride in lecturing at Manchester. This first cohort of
community lecturers asked for certificates of their contribution from the university; they
later spoke about having these framed on their walls, and telling the City of Harare
council members and officials of this experience.

However, status inequalities can be problematic. And in many engagements (both

public meetings and private interaction), status is reinforced by the particular use of
language, personal title and familiarity with technologies. A community leader from the
informal settlement of Denver (Johannesburg) identified some of the negative
consequences: “Sometimes as communities – we are waiting for someone to come to

16
See note 13.
17
Interview with Beth Chitekwe-Biti, Harare, 2018.
18
Nelson Ncube, Medellin, 9 April 2014.
19
Sekai Chirembe, Medellin, 10 April 2014.
20
SDI focus group, World Urban Forum, Medellin, 10 April 2014.

www.gdi.manchester.ac.uk 20

Electronic copy available at: https://ssrn.com/abstract=3470133

the settlement to bring something … When you behave like a professional or
intellectual – then you give them this impression – that this person must be right on
those points.”21 For him, the solution was to invest in a different kind of relationship that
ensures that community leaders and residents recognise that academics do not come
with answers. If this complementary relationship is identified, then knowledge co-
production is possible. He feared that academics may hold up processes of learning,
because local leaders believe that they come with solutions. The example highlights
the need for the ‘unlearning’ of some professional practices (Bennett, 2018b).

One example of the complexities of what are often multi-scalar research relationships
within academic–social movement partnerships is explored in Box 4.

Box 4. Urban development research in Uganda: fostering complementary

relationships?

An attempt to engage with status inequalities was made during a research partnership
in 2016 between the Ugandan SDI Alliance, academics at The University of
Manchester and a local postgraduate consultant linked to Makerere University. This
partnership sought to co-produce knowledge about the Transforming Settlements of
the Urban Poor programme in Uganda (TSUPU). The research aimed to generate
academic knowledge about the political effects of basic services

Co-production in Kabale and Mbale municipalities, while enabling NGO professionals

and national and local Federation leaders to generate useful knowledge about
outcomes for community mobilisation, poverty reduction and the development of more
inclusive urban governance.

During research design discussions, reflective conversations took place about how to
ensure that the research built confidence and capacity among local research teams
and Federation memberships, rather than reinforcing status inequalities. It was agreed
that mixed research teams would be formed in each locality, made up of two
academics, two NGO professionals and locally nominated members of regional
Federation executives. One national leader also joined the research team in Kabale.
Before fieldwork began, research planning meetings were held with the Federation
executive committees, then reflective discussions were held with the newly formed
local research teams. These discussions focused on identifying the different skills and
knowledges that each member of the team brought to the research, as well as the
challenges the team might encounter based on its make-up.22 These different facets
included the rich diversity in perspectives brought by the involvement of people who
had played different roles in the programme under investigation at both national and
local levels, local cultural awareness and language expertise, and the view from an
‘external’ lens.
21
Dominsani, group discussion, Nairobi, March 2018.
22
See also ‘Co-producing knowledge about government vision, commitment and capacity to
reduce urban poverty in Uganda’. ESID blog, 28 June 2016. Available at http://www.effective-
states.org/co-producing-knowledge-about-government-vision-commitment-and-capacity-to-
th
reduce-urban-poverty-in-uganda/. Accessed 13 March 2019

www.gdi.manchester.ac.uk 21

Electronic copy available at: https://ssrn.com/abstract=3470133

The reality of time and workload pressures and all the complexities of in-depth
qualitative research meant that little reflection took place on questions of status within
the first field site. During the second period of fieldwork in Kabale, a final reflection was
held by team members, who analysed their respective roles within the research
process. This approach contributed to positive and supportive working relations within
the local research teams, and deeper learning from the process for future
engagements; however, it had a limited effect on the perceptions of the research team
held by wider members of the local Federations and in some ways reinforced status
hierarchies between local leaders (inevitably those with stronger English language
skills), and the wider Federation membership. Conversely, and rather uncomfortably,
the act of Federation members interviewing government officials side-by-side with a
British researcher from a respected university increased the legitimacy of the
Federation among local government staff, and resulted in stronger relationships
between the Federations locally and the municipality. Co-produced individual and
group interviews with Federation members and local community stakeholders and final
reflections with regional Federation executives also resulted in status hierarchies being
revealed, and disjunctures between different layers of the local Federation structure
encouraging stronger reflection on questions of inclusivity within the local movements.

These experiences in Uganda reflect discussions within multiple SDI alliances on

questions of expertise, status and movement capability development internal to the
Alliances and Federations. Recognising that academic engagement with social
movements may undermine the growth of movement learning and capability
development led us to a deeper exploration of unequal status, and the stratification of
status based on expertise. Discussions with community leaders highlighted the way
these power dynamics go beyond those involving the academics and communities and
are internal to SDI Alliances and Federations. A professional in the Zimbabwe Alliance
described how power based on expertise is such a potent form of social stratification
that it permeates internal Federation processes and the relations between the
Federation and NGO, as well as those between the SDI alliances and academics:

You have teams that are super good with profiles. Enumerators coming from
outside can end up taking over a process in a given slum community. That is
where most of the challenges are located. We need tools and processes that do
not take away power from the community that we want to support.23

Experiences in Zimbabwe suggest that the solution to over-dependence on those seen

as ‘experts’ lies in SDI modalities and, particularly, in community exchanges that build
the confidence of community residents that everyone – including them – can be
teachers, with relevant experiences to share. But regular reminders are also required.
A Federation leader from Bulawayo emphasised the importance of allowing the local
23
Team group discussion, Nairobi, March 2018.

www.gdi.manchester.ac.uk 22

Electronic copy available at: https://ssrn.com/abstract=3470133

community – the residents of the settlement – to “own the data collection and
knowledge generation process”.24 She suggested that “external people – other
Federation members or academics – should wait to be asked to come and help – then
the tensions are much less”. The sequencing of local ownership and invitations for
external contributors is what makes the difference, in her experience.

A second challenge related to unequal status is that of unequal power when

determining the allocation of financial and reputational resources. One of the key
challenges faced by SDI affiliates is that academics draw them into research projects
as ‘case studies’; acknowledgement of the significant of the co-production of
knowledge requires an engagement beyond such case studies. The financial resources
allocated to such studies tend to be small and there is a poor process of co-planning
and budgeting. One SDI professional wrote to a colleague after one such invitation:
‘This is appropriation not partnership. There is no role here for the community except to
be research objects for the university. I do not know what to propose … There is no
time to generate a counter-proposal that – at the very least – incorporates action-based
learning, drawing extensively on Federation capacities and knowledge.”25 Less than
5% of the research budget was allocated to the SDI Alliance for the case study;
moreover, the Federation did not believe the planned outputs would address their
needs. The SDI Alliance sought the central participation of the Federation in the design
and management of the content of the research (questionnaire, data collection and
data analysis) and 20% of the research budget. The eventual outcome was that their
share doubled to 10%, with no evident shift in the research process. The academic
researchers themselves identified this research project to be about the co-production of
knowledge with local communities.26 But this approach is not consistent with the
understanding of co-production of knowledge discussed here. Such an outcome
highlights the power inequalities.

A third challenge facing SDI affiliates is their own considerable data resource collected
by volunteers and capacitated community researchers. Academic researchers
frequently ask for access to this data with little evident consideration being given to the
scale of investment. While academic researchers generally receive salaries,
community members typically receive minimal amounts for transport and food. When
academics request access to their data this needs to be considered. A related issue is
ensuring that community ownership is respected. Community members were shocked
to turn up to one consultation with a local authority only to find that the consultant
involved in the urban development was an academic (with whom they had previously
collaborated) who had offered their data to the local authority as a part of the
commercial services that the consultant was providing. Their own data was presented
back to them without acknowledgement. Affiliates are now thinking about how they can
protect their data.

24
Sazini Ndlovu, team group discussion, Johannesburg, March 2017.
25
Email correspondence with NGO professional, 29 June 2015.
26
The relevant website is not referenced here, because this is seen as a systemic problem; it does not
appear fair to pinpoint a problem case.

www.gdi.manchester.ac.uk 23

Electronic copy available at: https://ssrn.com/abstract=3470133

Resource issues are also related to authorship and attribution. Authorship is a choice
determined according to multiple factors, which are frequently neither discussed nor
agreed during the research process. This silence both reflects and exacerbates power
imbalances. The ability of academics to claim authorship is a testament to their more
powerful position within these relations. The denial of authorship to non-academics fails
to legitimate their contribution and may add to their own uncertainty about their value to
knowledge production. SDI affiliates wish to emphasise the community researchers
who are the closest – in a social sense – to disadvantaged communities. Being listed
as authors emphasises the legitimacy of the community researchers as equal partners
in the process. Academics, educated in other attribution traditions, typically list authors
alphabetically or by the order in which they have contributed to the written
documentation. But community researchers’ expertise may not be explicitly related to
the conceptual framework or theoretical insights, and they are unlikely to have drafted
much of the academic text. Their contribution is their understanding of their own
experiences of poverty, inequality and other forms of disadvantage.

In our experience, agreements about resources must be negotiated. Open dialogue

about what is required is helpful here. The better the understanding about different
objectives, the more easily this can be reconciled. The more open the discussion about
available resources, about how the resources are shared and about how outputs and
outcomes are identified, achieved and tested, the easier the processes of
accountability. Such discussions open new possibilities for resource acquisition and
use. In South Africa, this led to innovations in research dissemination with students
developing a handbook that could be read like a diary to share findings with community
members. IIED, for a further example, uses its ownership of the journal Environment
and Urbanization to enable grassroots activists to author papers alongside academic
contributions. Discussions between the Uganda Federation and academic researchers
led to the understanding that authorship may not be wanted when information is
sensitive and the status of the individual sharing their knowledge is weak relative to
that of others (King with Goretti, Kasaija & Owere, 2016).

Negotiations about research plans may be seen by some academics to be a threat to

the ‘objectivity’ of research processes. However, recognising that objective facts are
always subjectively observed and recorded, and that engagement with local activists
and residents results in greater information and new perspectives that can be
triangulated and jointly tested, helps to resolve such tensions.

4.4. Institutional solutions to accountabilities

You will come and do your research, but when you have gone, it is we who will
remain.27

You can stay as long as you like; my only problem is if you don’t spend much
time with us. Not enough time to understand our work.28

27
See note 23.

www.gdi.manchester.ac.uk 24

Electronic copy available at: https://ssrn.com/abstract=3470133

Relations between academic departments, universities and social movements may
benefit from being institutionalised. This may help social movements respond to
academics who come promising processes of knowledge co-production and who
perceive themselves as being supportive of movement processes, but whose actions
have consequences that concern movements. Several of the examples of tensions
noted above could be prevented through greater accountability over, for example, the
ways in which resources are allocated, the timing of research (and teaching) activities
and the use of research findings. We explore one further example that highlights the
way in which methodology is used by academics to manage activist challenges about
the orientation of academic research. We then reflect on the value of longer-term
relations to provide a platform for greater accountability and ethical engagement.

One Masters student from a European university studied SDI data collection processes
and explored the hypothesis that data collection would lead to formal tenure and hence
financial investment. The research concluded that this was not the case. A professional
associated with the affiliate asked the university not to publish the thesis, because of its
potential to undermine community data collection. The professional argued that the
student’s critique was based on unrealistic expectations about what the specific
process could achieve, and that these expectations were not those held by those
developing the process. This was not explained in the thesis. Furthermore, the
research was a ‘snapshot’ of the process at a specific point in time; at no point did the
researcher discuss this shortcoming. Third, the research oversimplified the
complexities of issues such as land tenure and did not engage with relevant expertise.
University staff argued that, as the methodology met academic standards, the thesis
met the standard for publication. The SDI professional commented to the academic
involved:

We observe that a premium is placed on the methodology that the researcher

used and that the research is further presumed to be ethical. We see that the
research question [is] completely misplaced …. the contribution of research is
deeply important to the endeavours of slum dwellers, and we hope to steer this
in a positive way …. [However] our partners and friends take advantage of, and
undermine the efforts of, poor communities. And do so purely for the purposes
of academic pursuit. The damage this does is very real for us.

SDI’s solution, both in terms of student interest and academic relations, has been to
deepen engagements and so secure greater accountability. Box 5 describes how the
SDI affiliate in Zimbabwe is managing increasing levels of student interest to avoid the
problem discussed above; this exemplifies greater institutionalisation of collaborative
relations.

28
Perween Rahman, then director of the Orangi Pilot Project.

www.gdi.manchester.ac.uk 25

Electronic copy available at: https://ssrn.com/abstract=3470133

Box 5: SDI Zimbabwe: managing student engagement

In Zimbabwe, the SDI Alliance has responded to this interest with a requirement that
students engage the community when setting their research objectives. This is to
ensure that the process adds value to the work of Federation activists and to minimise
potential tensions that might arise. The NGO staff and Federation leadership recognise
that this can be a tricky process, as they do not want to influence the research process
itself. Rather, they want to make sure the orientation of the research reflects issues that
are priorities for local groups, and they want to ensure that the research is possible in
that local context.

Over time, this has evolved into a structured process. The students now have meetings
with the SDI Zimbabwe Alliance to explain what they are interested in, and then they
agree the research process prior to beginning their work. This dialogue also helps to
diversify the ways in which findings are shared, as the Federation explains how
students can reach their members with the research findings, which would otherwise
just remain in a thesis.

Federation leaders believe that student researchers “should be referred by the

academics we are working with. We cannot just take any students from anywhere”. 29
Working with staff and students from a university with which they have a formal
relationship ensures that research processes and potential outcomes are discussed,
and the different interests of all parties considered.

While the term ‘co-production of knowledge’ is widely used to describe academic–

movement knowledge relations, in practice there is a continuum of engagements. At
one end of that continuum is Federation engagement in academic research for reasons
of income generation; at the other are projects that are jointly developed and which
emerge from shared values. Close relations – built over long periods and with
considerable trust – enable projects to begin, even when it is unclear where they are
placed on this continuum. Hence long-term relations emerge as a solution to resolving
some of these tensions, enabling relations to move from an exploratory stage, in which
beliefs and principles are shared and tested, through to shared research projects with
common goals and jointly identified roles (see also SPARC, 2014). It is deeper long-
standing institutional relations that change the potential of the co-production of
knowledge. Over time, partners better understand their own motivations and those of
others, and hence what joint activities are possible and how tensions can be

29
Sazini Ndlovu, Bulawayo, 29 August 2018.

www.gdi.manchester.ac.uk 26

Electronic copy available at: https://ssrn.com/abstract=3470133

managed.30 The confidence that comes with such relations produces many benefits.
Shared values open up new possibilities for engagement. However, there is an
insufficient acknowledgement of this reality in many academic departments. A
workshop with academics and leading civil society activists from the Global South,
including those from SDI, highlighted that “most universities are woefully under
resourced to identify and support the longer-term research agendas that have the most
transformative potential – and yet this is their comparative advantage over think tanks
or consultants” (Jordan, 2017).

Our discussion on the period required to build relations of trust suggested at least five
years is needed to build effective relations with an institutional commitment in place.31
But this metric-driven approach may be misleading, because it conflates time and
formality with a value-based alignment and the acknowledgement of interdependency
required for partnership. Agencies such as the Institute for Development Studies
(Oswald et al, 2016) and the International Institute for Environment and Development
(IIED) have developed long-term relations with a range of civil society groups
interested in knowledge generation. In the case of IIED, three decades of collaboration
with SDI affiliates has deepened and broadened to include formal research
programmes, action-research projects, advocacy, documentation and dissemination,
and meetings that bring together new groupings of agencies and activists. This work
has included IIED acting as a conduit financier for $30 million from US and UK trusts
and foundations, while the confidence of Northern donors in the ability of Southern
social movements to manage large grants was established.

There is a lack of demonstrable examples of what such practices of engagement look

like for academic departments, and how ethical principles can be embedded into
academic practices. 1to1 – Agency of Engagement, an NGO working in South Africa
that emerged from student fieldwork – sees a principled code as a way of engaging
communities living in low-income neighbourhoods, ensuring good practice and clarity of
expectations (Bennett, 2018a). Such a code may be a useful starting point for
discussions between professionals and community organisations; however, there are
also evident limits (Bennett, 2018a; Banks et al, 2013) A code cannot be adapted to
cover the breadth of professionals and the malleability of urban development practice:
rather than a ‘silver bullet’, it has to be understood as a contribution. A further approach
may be a memorandum of understanding (MOU) to define the terms of engagement
between social movement organisations and academic institutions (see Bennett, 2018a
for a South African example). SDI’s Zimbabwe Alliance explored such an MOU with the
University of Zimbabwe. An immediate challenge was that their academic MOU, in its
existing form, only recognises formally registered organisations such as NGOs, thereby

30
One benefit of long-term relations is that Federation members learn how to deal with
unexpected academics coming to research in informal settlements without engaging them from
the outset. Federation leaders are concerned that many of the initial contacts these academics
establish are with individuals seeking personal benefits, but who may be vulnerable, because of
their lack of understanding. With a better understanding of academia, Federation leaders
believe they respond more effectively to these situations.
31
Group discussion, Nairobi, March 2018.

www.gdi.manchester.ac.uk 27

Electronic copy available at: https://ssrn.com/abstract=3470133

excluding informal communities. The Zimbabwe Alliance has delayed the signing of the
MOU until the urban poor are recognised as equal parties and signatories. The Alliance
did sign an MOU (in October 2014) with the National University of Science and
Technology in Bulawayo, as this university was prepared to meet that condition.

Rather than place reliance on ethical codes, MOUs or other mechanisms, we need to
see these as tools that help issues to be explored and ethical problems to be resolved.
Long-term engagement with an effort to create space for grassroots activists to share
their opinion and amplify their voice builds and deepens understanding. It is the quality
of engagement that is critical to the efficacy of the research and the ethical standards
that are followed – rather than any specific tool (see also Gillian & Pickerill, 2012, p
136). Our experience is that, over time and in differing contexts, perspectives (positive
and negative) reveal themselves. Longer-term commitment means that academics
invest time to engage with the realities of urban social movements. It is not just a
question of how much time is spent, but the quality of that time. Staying overnight in
informal settlements is invaluable, as it offers time for discussion outside of formal
meetings, and opportunities for everyone to relax and chat. Familiarity enables mutual
learning, including sensitivity about what knowledge is placed in the public arena and
when it is placed.

As participants get to know each other, there are new opportunities. As the contribution
of community activists is validated through positive engagements, there are
progressive gains that accrue. Departmentally institutionalised academic–movement
relations enable regular diverse interactions that build iteratively to greater
understanding. Opportunistic engagements become more strategic with this familiarity,
which is why there is a need for institutional processes that maintain an engagement,
with room for exploration. The visits of community leaders to teach the Masters-level
course unit at The University of Manchester led to the sharing of SDI modalities with
local communities in Manchester from the first teaching programme in 2010 (see Box
3). When Sophie King, an academic working on issues of exclusion in both the UK and
Uganda, learned more about SDI, these exposure visits became an opportunity to
develop a sustained exploration. Sophie’s knowledge built on her research with the
Ugandan Alliance in 2016 (a project made possible through the same departmental
links). From 2015, there has been a more consistent and therefore deepening
engagement between South African and Kenyan affiliates and low-income women’s
groups in Manchester and Salford. These exposure visits developed a three-year
action research programme supported by the University of Sheffield. Manchester
groups are now emulating SDI modalities, now including contributing themselves to
teaching on other courses within the department.

In the experience of SDI, academic participation in community exchanges is a

significant opportunity to improve mutual understanding of the perspectives of all
stakeholders. Such exchanges – generally taking place over a week and based in an
informal settlement – provide a platform for mutual learning, shared observations of
new realities for the guests, and joint presentations about work ‘at home’. The value of

www.gdi.manchester.ac.uk 28

Electronic copy available at: https://ssrn.com/abstract=3470133

community-to-community exchanges in building the confidence of grassroots residents
has been recognised (Patel & Mitlin 2002; ACHR 2000). These exchanges challenge
assumptions about the flows and uses of knowledge. The very informal locations in
which exchanges take place disrupt power relations between academics and the
residents of low-income communities, as participants sit on the floor in large rooms or
outside under trees. As more community members begin to talk – in settings that are
more familiar to them – then the balance of the discussion shifts. Thinking about the
setting within which dialogue and then the co-production of knowledge can take place
most effectively has been a key consideration when SDI affiliates design interactions
with local academics. While long-term engagements change the relational context
within which negotiations take place, the everyday encounters also need to change for
activists to be empowered and share their perspectives.

5. Conclusion
Power is money and knowledge.32

Historically, universities have not only produced knowledge but have also been the
arbiters of which knowledge is ‘good’ and ‘valid’, establishing the very frameworks
by which such assessments are made. Tautologically, universities have long
considered knowledge produced by universities as the best and most legitimate.
(Tandon et al, 2016, p 29)

Efforts to bridge the gap between academic and non-academic forms of knowledge
generation are long-standing, and there is a continuing interest by academics in a
substantive engagement with non-academic stakeholders. The co-production of
knowledges renews interest in the inter-dependency between academics and non-
academics to achieve joint values of social justice and equity, with specific objectives
related to inclusive and equitable urban development. The collaboration offers
opportunities to secure redistribution, negotiation reform, and to build relations that
acknowledge the substantive contribution of non-academics to knowledge generation,
knowledge democracy and the potential of long-term relations. These efforts are driven
by shared values and a belief in the potential of knowledge generation orientated
towards social justice, as well as the belief that the inclusion of low-income and
disadvantaged groups in research is itself an issue of social justice, as well as
producing outcomes that advance that goal.

However, as discussed above, whatever the written commitment to this approach,

considerable effort is needed to overcome tensions which, for the most part, stem from
unequal power differentials. While this reality is recognised, it is rarely fully explored.
The need to reconstruct knowledge generation and professional education is acute in
an urban context in which academic experts and/or academically certified professionals
determine urban regulations (planning, environmental health, construction), welfare
services (poverty programming, social provision), security (criminology, planning),

32
SA Federation slogan.

www.gdi.manchester.ac.uk 29

Electronic copy available at: https://ssrn.com/abstract=3470133

infrastructural investments (engineering, planning), governance systems (politics,
government administration) and employment opportunities (economics, business
studies, planning).

Investing more time in the relationship offers better communication and potentially
greater accountability. Greater familiarity with and awareness of the benefits of
collaboration have led to an expanded range of joint activities. Longer-term
engagements enable interactions to deal more explicitly with the ways in which
activities are influenced by the underlying unequal distribution of power. They also
recognise the interdependency between academics and social movements seeking to
transform urban outcomes. While longer-term engagements may be assisted by formal
arrangements such as MOUs and codes of practice, it is the improved dialogue and
enhanced understanding that makes the difference. Familiarity builds trust in values
and the ability to address issues of power that lie at the centre of many of the tensions
discussed above. It is fitting, therefore, that we close this paper with a reflection on
power and the co-production of knowledge.

Power is manifest through one group – academics in this case – having more options
and using their options to exclude or not sufficiently include social movements. Power
is recognised to be manifest explicitly, implicitly or to be invisible (Gaventa & Cornwall,
2015). In all three forms, it may reinforce, or undermine, hierarchies through relations
and associated interactions. Miller et al (2006, p 6) argue that ‘power is dynamic,
relational and multidimensional, changing according to the context, circumstance and
interest’. However, inequalities in power are not infinitely or easily malleable. As shown
above, alongside efforts to undermine the existing distribution of power within and
through the co-production of knowledge, are forces that reinforce existing inequalities.
How can individuals and collectives assert themselves through such interactions and
change outcomes?

Power is realised through relations. Rowlands (1997) elaborates alternative forms of

relational power in the context of gendered experiences of disadvantage, suggesting
that while power over – ie one individual or group being able to dominate – is inherently
limited in its transformative potential, other forms of power – power with ( the power of
coalitions), power from within (the power of a self-believing individual) and power to
( the power that a capacitated group or individual can achieve) – may create and reflect
new potentialities. Rowlands emphasises that empowerment must be “more than
participation in decision making; it must also include the processes that lead people to
perceive themselves as able and entitled to make decisions” (Rowlands 1997, p 14,
original emphasis). That conclusion resonates with the experiences discussed here.
Engagement with academia legitimates knowledge outputs of communities (data),
building the confidence of social movement activists alongside building capabilities and
changing the way in which movement activists are perceived by universities and
governments. While Appadurai (2001) describes how Federations build a capacity to
aspire to development goals within their membership, the discussion in Section 4
shows that successful efforts to co-produce knowledge build the capability to manage

www.gdi.manchester.ac.uk 30

Electronic copy available at: https://ssrn.com/abstract=3470133

relations by identifying shared values and complementarity of actions towards shared
goals. However, that requires a joint commitment to overcoming practices that reinforce
relational inequalities. Specifically, there needs to be joint reflection on the purpose of
knowledge production and the reconciliation of alternative theories of change. This
reflects on the purpose of collaboration and what each group wishes to achieve, and
how different kinds of activities – specifically research and teaching – can support the
deepening of collaboration. It also requires a specific reflection on issues of power –
particularly the way in which status validates some voices and diminishes others – and
how the contribution of the most disadvantaged and marginalised can be valued.

Table 1 summarises the issues discussed above, identifying the significant dimensions
of power associated with each issue and drawing on a four-fold categorisation of
dimensions of power: power over, power with, power to, power in (Rowlands, 1997;
Miller et al, 2006). While ‘power over’ is viewed negatively, because of the underlying
conception of hierarchies of control, other forms of power are considered to have more
potential (Miller et al, 2006). However, this should not be assumed. ‘Power with’
requires relationships that may, as noted above, involve inequalities in power. Also
significant appears to be ‘power through’, by which is meant the power gained as
capabilities to do tasks that could not previously be achieved are developed. These
capabilities involve managing relations – in this case with academics – and gaining
analytical and conceptual skills through such engagements. Table 1 also suggests
ways in which the challenges might be addressed.

Table 1: Challenging relational inequalities through the co-production of

knowledge
Power realigned Potential solutions

Negotiating Need to recognise differences in how Better discussion of alternative theories of change,
theories of change influence is achieved and align strategies ensuring good understanding of the nature of each
where possible, maximise theory of change, and identifying tensions and
complementarities and minimise watering complementarities. Quality of dialogue is essential:
down. Maximising power with new learn about each other, and how joint action can
alliances, but also power through building achieve shared goals. The time invested is
new capabilities for action and working indicative of the commitment to understanding.
together. There may be discomfort as individuals participate in
new social situations.

From knowledge While academics tend to place greater As partnerships are more longstanding and have
generation to emphasis on new knowledge, urban greater depth, there are more possibilities to
education social movements tend to place greater develop new activities. There are opportunities for
emphasis on using existing knowledge to social movements to take on roles in teaching,
change outcomes by educating legitimating their knowledge and contribution to
professionals. Movement activists multiple activities. Students may be informed
recognise the power within after having through engagement with marginalised and
taken on a public role as teacher, and the disadvantaged populations and may develop new
power to validate ideas and approaches understandings supported by academics. Officials
from a position with this status. realise through joint teaching activities that citizens
also have important knowledge to share.

Greater equality of Status is a very significant from of power. Co-production of knowledge must recognise the
status Aspirations and confidence are inculcated contribution of grassroots leaders and community
into high-status individuals. Power within groups to addressing urban problems and

www.gdi.manchester.ac.uk 31

Electronic copy available at: https://ssrn.com/abstract=3470133

 new identities emerges through positive strengthen public recognition of their value. Activities
experiences: teaching as well as need to be designed to achieve this. Individual and
research can be important here. collective capabilities need to be recognised through
Academics and professionals with high public presentations, joint authorship, video blogs
status may use this status to authenticate (as well as written work), and other ways that
low-status knowledge generators. authenticate the knowledge contribution of
movement activists.
Most research funders expect academic
scholars to lead research bids. Unequal A more equal share of both resources and
and non-negotiated resources attribution is needed. This challenge cannot be
distributions mean that academics and solved by research partners alone. Greater
professionals have power over grassroots awareness and action by donors is needed,
leaders. Resolving these resourcing including requirements for long-standing
issues opens up new approaches to relationships, sharing of financial details, and
knowledge generation. enhanced processes to ensure that resource issues
have been discussed and agreed.

Sharing Accountabilities involve shared For urban social movements, what appears
accountabilities information about commitments. The important is that academics make an investment in a
ability to hold individuals and agencies to relationship. That investment offers them leverage
account requires such information. Such and/or protection; there is both a commitment and
information creates mutual ‘power over’. cost to academics who build such a relationship.
All parties can see issues as they emerge The investment is a demonstration of commitment.
and engage with the others to resolve This commitment opens the potential to build
them. In a context in which urban social understanding – and invites the movement to
movements may have their perspectives, articulate the frustrations they experience. It invites
actions and resources misrepresented them to be honest about what is required to make
and their contribution inaccurately the relationship work for that social movement, with
attributed, such information is essential to the understanding that the relationship is significant
ensure that they are not disadvantaged. for the academics.

Co-producing knowledge can be empowering for social movements; through such

empowerment and associated alliances many things are possible. This working paper
has highlighted the importance of collaborations between academics and urban social
movements to achieve justice both in knowledge generation and in urban outcomes. It
highlights the potential that can emerge as movement activists are supported – through
long-term relations – to be aware of their own contributions and to amplify their voices
for academics who wish to understand and engage with their perspectives. However,
this discussion also highlights the need for much greater awareness about the
obstacles that must be overcome. It is not enough to build relations and understanding.
Consciousness about the unequal distribution of power has to be ‘centre-stage’.
Objectives and methodologies need to be determined collaboratively. Academics have
to be willing to use – rather than benefit from – their status, even if it reduces their own
public role. It is through a focus on interdependency and complementarity that
academics and movements can best advance their shared goal of an equitable urban
future.

www.gdi.manchester.ac.uk 32

Electronic copy available at: https://ssrn.com/abstract=3470133

 References

ACHR (2000). Face to Face. Bangkok, Asian Coalition for Housing Rights (ACHR).

Anderson, P. M. L., Brown-Luthango, M., Cartwright, A., Farouk, I. and Smit, W. (2013).
‘Brokering communities of knowledge and practice: reflections on the African Centre for
Cities’ CityLab programme’. Cities 32, 1–10.

Appadurai,A. (2001). ‘Deep democracy: urban governmentality and the horizon of politics’.
Environment and Urbanization 13(2): 23-43.

Appadurai, A. (2012). ‘Why enumeration counts’. Environment and Urbanization 24, 639–641.

Baldwin, A. (2017). "Decolonising geographical knowledges: the incommensurable, the

university and democracy." Area 49(3): 329-331.
Baum, H.S. (2000). ‘Fantasies and realities in university–community partnerships’. Journal of
Planning Education and Research 20, 234–246.

Banks, S., Armstrong, A. with Carter, K., Graham, H., Hayward, P., Henry, A., Holland, T.,
Holmesh, C., Leeh, A., McNulty, A., Moore, N., Nayling, N., Stokoe, A. and Strachan,
A. (2013). ‘Everyday ethics in community-based participatory research’. Contemporary
Social Science 8, 263–277.

Beebeejaun, Y., Durose, C., Rees, J., Richardson, J. and Richardson, L. (2015). ‘Public harm
or public value? Towards coproduction in research with communities’. Environment and
Planning C 33, 552–565.

Bell, D. M. and Pahl, K. (2018). ‘Co-production: towards a utopian approach’. International

Journal of Social Research Methodology 21, 105–117.

Bennett, J. (2018a). ‘Un-learning “community”: reflections on socio-technical spatial design

support with Slovo Park’. Out of the Box. Council for Scientific Research, Pretoria.

Bennett, J. (2018b). Design Praxis in a post-Rainbow Nation City. Addis Abba: Through Local
Eyes.

Brown-Luthango, M. (2013). ‘Community–university engagement: the Philippi CityLab in Cape

Town and the challenge of collaboration across boundaries’. Higher Education 65,
309–324.

Burns, D., Harvey, B. and Aragon, A.O. (2012). ‘Introduction: action research for development
and social change’. IDS Bulletin 43, 1–7.

Burra, S., Mitlin, D., Menon, G. with Agarwal, I. Banarse, P., Gimonkar, S., Lobo, M., Patel, S.,
Rao, V. and Waghmare, M. (2018). Understanding the Contribution of the BSUP
(JNNURM) to Inclusive Cities in India. Effective States and Inclusive Development
(ESID) Research Centre Working Paper 97. Manchester: ESID Research Centre,
University of Manchester.

Castells, M. (1996). The Rise of the Network Society. Oxford: Wiley Blackwell.

Castells, M. (1997). The Power of Identity. Oxford: Wiley Blackwell.

www.gdi.manchester.ac.uk 33

Electronic copy available at: https://ssrn.com/abstract=3470133

Castells, M. (1998). End of Millennium. Oxford: Wiley Blackwell.

Castree, N., Chatterton, P., Neynen, N., Larner, W. and Wright, M.W. (2009). ‘Introduction:
the point is to change it’. Antipode 41(S1), 1–9.

Chambers, R. (1995). ‘Poverty and livelihoods: whose reality counts?’. Environment and
Urbanization 7, 173–204.

Chen, M. (2014). ‘Informal employment and development: patterns of inclusion and exclusion’.
European Journal of Development Research 26, 397–418.

Culwick, C., Washbourne, C.-L., Anderson, P.M.L., Cartwright, A., Patel, Z. and Smit, W.
(2019). ‘CityLab reflections and evolutions: nurturing knowledge and learning for urban
sustainability through co-production experimentation’. Current Opinion in
Environmental Sustainability 39, 9–16.

Escobar, A. (1992). ‘Planning’. In Sachs, W. (ed.), The Development Dictionary: A Guide to

Knowledge as Power, pp 132–145. London: Zed Books.

Escobar, A. (2018). Designs for the Pluriverse: Radical Interdependence, Autonomy and the
Making of Worlds. Durham NC: Duke University Press.

Freire, P. (2000). Pedagogy of the Oppressed. New York: Continuum.

Gaventa, J. and Cornwall, A. (2015). ‘Power and knowledge’. In Bradbury, H. (ed.), The Sage
Handbook of Action Research, pp 465–471. London: Sage Publications.

Gaventa, J. and Tandon, R. (2010). ‘Citizen engagements in a globalizing world’. In Gaventa,

J. and Tandon, R. (eds), Globalizing Citizens: New Dynamics of Inclusion and
Exclusion, pp 1–31. London: Zed Books.

Gillian, K. and Pickerill, J. (2012). ’The difficult and hopeful ethics of research on, and with,
social movements’. Social Movement Studies: Journal of Social, Cultural and Political
Protest 11, 133–143.

Green, D. (2017). ‘The NGO–academia interface: realising the shared potential’. In

Georgalakis, J., Jessani, N., Oronje, R. and Ramalingam, B. (eds), The Social Realities
of Knowledge for Development, pp 20–31. Brighton: Institute for Development Studies.

Holland, J. (2013). ‘Introduction: participatory statistics – a “win-win” for international

development’. In Holland, J. (ed.), Who Counts? The Power of Participatory Statistics.
Bourton on Dunsmore, UK: Practical Action Publishing.

Jacobs, F., Jordhus-Lier, D. and Tsolekile de Wet, P. (2015). ‘The politics of knowledge:
knowledge management in informal settlement upgrading in Cape Town’. Urban Forum
26, 425–441.

Jordan, C. (2017). ‘How can development academics help change the world?’. GDI blog.
University of Manchester.

Kanbur, R. (2012). ‘Exposure and dialogue programs in the training of development analysts
and practitioners’. In DeMartino, G. and McCloskey, D. (eds), OUP Handbook on
Professional Economics Ethics. Oxford: Oxford University Press.

www.gdi.manchester.ac.uk 34

Electronic copy available at: https://ssrn.com/abstract=3470133

Kara, H. (2017). ‘Identity and power in co-produced activist research’. Qualitative Research 17,
289–301.

King, S. and Kasaija, P. (2018). State–Movement Partnership in Uganda: Co-producing an

Enabling Environment for Urban Poverty Reduction? Effective States and Inclusive
Development Working Paper No. 98. Manchester: ESID Research Centre, University of
Manchester.

King, S. with Goretti, K., Kasaija, P. and Owere, S. (2016). ‘Coproducing knowledge about
government vision, commitment and capacity to reduce urban poverty in Uganda’.
ESID blog. Effective States and Inclusive Development Research Centre, University of
Manchester.

Lines, K. and Makau, J. (2018). ‘Taking the long view: 20 years of Muungano wa Wanavijiji,
the Kenyan federation of slum dwellers’. Environment and Urbanization 30, 407–424.

Madden, D. (2015). ‘There is a politics of urban knowledge because urban knowledge is

political: a rejoinder to "Debating urban studies in 23 steps"’. City 19, 297–302.

McFarlane, C. (2006). ‘Knowledge, learning and development: a post-rationalist approach’.

Progress in Development Studies 6, 287–305.

Miller, V., VeneKlasen, L., Reilly, M. and Clark, C. (2006). Making Change Happen: Power.
Washington DC: Just Associates.

Mitlin, D. (2008). ‘With and beyond the state: coproduction as a route to political influence,
power and transformation for grassroots organizations’. Environment and Urbanization
20, 339–360.

Mitlin, D. (2013). ‘A class act: professional support to people’s organizations in towns and
cities of the global South’. Environment and Urbanization 25, 483–499.

Mitlin, D. (2018). ‘Beyond contention: urban social movements and their multiple approaches
to secure transformation’. Environment and Urbanization 30, 557-574

Myers, G. (2003). Verandahs of Power. New York: Syracuse University Press.

Noxolo, P. (2017). ‘Introduction: decolonising geographical knowledge in a colonised and re-

colonising postcolonial world’. Area 49, 317–319.

Oldfield, S. (2015). ‘Between activism and the academy: the urban as political terrain.’ Urban
Studies 52, 2072–2086.

Openjuru, G. L., Jaitli, N., Tandon, R. and Hall, B. (2015). ‘Despite knowledge democracy and
community-based participatory action research: voices from the global south and
excluded north still missing’. Action Research 13, 219–229.

Osuteye, E., Ortiz, C., Lipietz, B., Castán Broto, V., Johnson, C. and Kombe, W. (2019).
Knowledge Co-production for Urban Equality. KNOW working paper 1. London:
Bartleett Development Planning Unit, University College London.

www.gdi.manchester.ac.uk 35

Electronic copy available at: https://ssrn.com/abstract=3470133

Oswald, K., Gaventa, J. and Leach, M. (2016). ‘Introduction: interrogating engaged excellence
in research’. IDS Bulletin 47, 1–18.

Patel, S. and Mitlin, D. (2002). ‘Sharing experiences and changing lives’. Community
Development 37, 125–136.

Patel, S. and Mitlin, D. (2009). ‘Reinterpreting the rights based approach: a grassroots
perspective on rights and development’. In Hickey, S. and Mitlin, D. (eds), Rights-
based Approaches to Development: Exploring the Potential and Pitfalls, pp 107–125.
Sterling VA: Kumarian Press.

Patel, S., Baptist, C. and D'Cruz, C. (2012). ‘Knowledge is power – informal communities
assert their right to the city through SDI and community-led enumerations’.
Environment and Urbanization 24, 13–26.

Rowlands, J. (1997). Questioning Empowerment: Working with Women in Honduras. Oxford:

Oxfam.

Simon, D., Palmer, H., Riise, J., Smit, W. and Valencia, S. (2018). ‘The challenges of
transdisciplinary knowledge production: from unilocal to comparative research’.
Environment and Urbanization 30, 481–500.

Songsore, J. and McGranahan, G. (1998). ’The political economy of household environmental

management: gender, environment and epidemiology in the Greater Accra
Metropolitan Area’. World Development 26, 395–412.

SDI (2018). Annual Report 2017. Cape Town: SDI (Shack/Slum Dwellers International).

Society for the Promotion of Area Resource Centres (SPARC) (2014). Annual Review 2013–4.
Mumbai: SPARC.

Standing, H. and Taylor, P. (2016). ’Whose knowledge counts? Development studies

institutions and power relations in a globalised world’. IDS Bulletin 47, 169–178.

Tandon, R., Singh, W. Clover, D. and Hall, B. (2016). ‘Knowledge democracy and excellence
in engagement’. IDS Bulletin 47, 19–36.

Tuhiwai Smith, L. (1999). Decolonizing Methodologies: Research and Indigenous Peoples.

London: Zed Books.

UN-Habitat (2014). ‘Background paper: World Habitat Day October 6’. UN-Habitat, Nairobi.

UN Population Division (2018). World Urbanization Prospects: The 2017 Revision. New York:
Department for Economic and Social Affairs, United Nations.

van Ewijk, E. and Baud, I. (2009). ‘Partnerships between Dutch municipalities and
municipalities in countries of migration to the Netherlands: knowledge exchange and
mutuality’. Habitat International 33, 218–226.

Winkler, T. (2013). ‘At the coalface: community–university engagements and planning

education’. Journal of Planning Education and Research 33, 215–227.

www.gdi.manchester.ac.uk 36

Electronic copy available at: https://ssrn.com/abstract=3470133

Author biographies

Jhono Bennett is an architectural urbanist based in Johannesburg. He is a co-director and co-

founder of 1to1 – Agency of Engagement, a design based social enterprise that has been
developed to support the positive re-development of South African cities. Jhono holds a
research position at the University of Johannesburg’s DSD Desis Lab and has a focus on the
intersectional role of design in how South African cities are seen, made and managed. He is
preparing to undertake a PHD in this sector currently titled a Visual Narrative of the post-post-
Apartheid City which plans to focus on the future of urban centres in Southern Africa.

Philipp Horn is a Lecturer in the Department of Urban Studies and Planning at the University of
Sheffield. His research interests centre around rights-based approaches to urban development,
indigenous rights to the city, models of knowledge co-production, and citizen-led
planning. Working collaboratively with urban grassroots movements in Bolivia, Ecuador and
Kenya, his research seeks to identify pathways for integrating specific interests, demands and
rights-based claims of historically marginalised groups into urban policies and planning
interventions.

Sophie King's research and practice has focused on modes of community organising among
marginalised social groups and understanding how collaborative approaches to local
governance and service delivery can contribute to reducing poverty and inequality. Her
experience of knowledge co-production includes the co-production of research into collaborative
service provision and inclusive governance in Ugandan municipalities with ACTogether and the
National Slum Dwellers Federation of Uganda. Sophie is currently working with residents’
groups in low income neighbourhoods of Greater Manchester to co-produce knowledge and
action in response to learning exchanges with Kenyan and South African affiliates of
Slum/Shack Dwellers International.

Jack Makau is coordinator of Slum Dwellers International (SDI) Kenya, the NGO that provides
professional and technical support to Muungano wa Wanavijiji, the Kenyan federation of slum
dwellers. He has been a support professional for Muungano since 2000. He has worked closely
with SDI’s secretariat to support the development of SDI affiliates across the network,
particularly in their engagement with professional agencies such as the Global Land Tools
Network. Jack has also collaborated with academics from both the global North and South in
research and teaching activities.

George Masimba is Director of Programme for Dialogue on Shelter (Zimbabwe), the support
NGO for the Zimbabwe Homeless People’s Federation, an affiliate of Shack/Slum Dwellers
International. George has managed slum upgrading projects in informal settlements where
participatory strategies such as community-led enumerations and mapping have been employed
as a tool for enabling the upgrading processes. He has actively contributed towards the
formulation of the Harare Slum Upgrading Strategy and the Harare Slum Upgrading Finance
Facility through his affiliation with Dialogue on Shelter. He has extensive experience of
collaboration with academics, both through his PhD studies and through research programmes
with Dialogue on Shelter and the Zimbabwe Federation.

Diana Mitlin is an economist and social development specialist working at the Global
Development Institute (University of Manchester) and the International Institute for Environment
and Development (IIED). She has worked with SDI and its affiliates since its inception in 1996 in
a variety of roles and supports its efforts to secure transformative urban development that
recognises the rights and capabilities of grassroots organizations to design and manage
interventions. She has led and supported research programmes with SDI affiliates in India,
Malawi, Namibia, South Africa, Tanzania, Uganda, Zambia and Zimbabwe.

www.gdi.manchester.ac.uk 37

Electronic copy available at: https://ssrn.com/abstract=3470133