Working Paper No. 99 CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS) Nafisa Ferdous Christine Jost Patti Kristjanson Working Paper Closing the Relevance Gap: Lessons in Co-Developing Gender Transformative Research Approaches with Development Partners and Communities Closing the Relevance Gap: Lessons in Co-Developing Gender Transformative Research Approaches with Development Partners and Communities Working Paper No. 99 CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS) Nafisa Ferdous Christine Jost Patti Kristjanson 1 Correct citation: Ferdous N, Jost C, Kristjanson P. 2015. Closing the Relevance Gap: Lessons in Co-Developing Gender Transformative Research Approaches with Development Partners and Communities. CCAFS Working Paper no. 99. CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS). Copenhagen, Denmark. Available online at: www.ccafs.cgiar.org Titles in this Working Paper series aim to disseminate interim climate change, agriculture and food security research and practices and stimulate feedback from the scientific community. This document is published by the CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS), which is a strategic partnership of the CGIAR and the Earth System Science Partnership (ESSP). CCAFS is supported by the CGIAR Fund, the Danish International Development Agency (DANIDA), the Government of Australia (ACIAR), Irish Aid, Environment Canada, Ministry of Foreign Affairs for the Netherlands, Swiss Agency for Development and Cooperation (SDC), Instituto de Investigação Científica Tropical (IICT), UK Aid, and the European Union (EU). The Program is carried out with technical support from the International Fund for Agricultural Development (IFAD). Contact: CCAFS Coordinating Unit - Faculty of Science, Department of Plant and Environmental Sciences, University of Copenhagen, Rolighedsvej 21, DK-1958 Frederiksberg C, Denmark. Tel: +45 35331046; Email: ccafs@cgiar.org Creative Commons License This Working Paper is licensed under a Creative Commons Attribution – NonCommercial–NoDerivs 3.0 Unported License. Articles appearing in this publication may be freely quoted and reproduced provided the source is acknowledged. No use of this publication may be made for resale or other commercial purposes. © 2015 CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS). CCAFS Working Paper no. 99 DISCLAIMER: This Working Paper has been prepared as an output for the Linking Knowledge with Action Theme under the CCAFS program and has not been peer reviewed. Any opinions stated herein are those of the author(s) and do not necessarily reflect the policies or opinions of CCAFS, donor agencies, or partners. All images remain the sole property of their source and may not be used for any purpose without written permission of the source. 2 Abstract The CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS) Theme Linking Knowledge with Action (K2A) engaged in a two-year social learning process to develop tools, best practices and capacity building trainings around gender sensitive and participatory research for climate change. The output was the resource guide, “Gender and Inclusion Toolbox: Participatory Research in Climate Change and Agriculture”. This paper documents the social learning process of co-designing and co-testing the toolbox with various development partners, CGIAR scientists, technical officers and local communities and offers key reflections and learning on the challenges and entry points to promoting a participatory and gender sensitive research agenda across upstream and downstream stakeholders. Keywords Social Learning; Gender; Participatory Research; Climate Change; Agriculture; Food Security 3 About the Authors Nafisa Ferdous is a gender specialist and consultant working with the World Agroforestry Centre and CCAFS based in Nairobi, Kenya. She has worked in INGO as well as grassroots organizations in India, Bangladesh, Kenya and in her own community in New York, US. Contact: nafisaferdous@gmail.com. Christine Jost is a veterinarian and Participation and Impact Scientist at the World Agroforestry Centre in Nairobi, Kenya. She specializes in the relationship between climate change, gender and equity, socio-ecological health, emerging and transboundary diseases, innovation and institutional change. She has two decades of experience leading transdisciplinary research and development projects in Africa and Asia in collaboration with academic, private, NGO, public and international partners. Contact: c.jost@cgiar.org. Patti Kristjanson is an agricultural economist and senior research fellow with the World Agroforestry Centre in Nairobi, Kenya. Her expertise includes poverty and livelihood analyses, impact assessment, agricultural system analyses and innovative research approaches for linking knowledge with action. She has 25 years of experience leading and managing multidisciplinary teams from international and national agricultural research centres, in collaboration with universities, donors, private and public sectors across Africa and Asia Contact: p.kristjanson@cgiar.org.   4 Acknowledgements This working paper along with the Gender and Inclusion Toolbox was guided by the valuable feedback and recommendations of key practitioners, scientists and subject experts at Food and Agriculture Organization of the United Nations (FAO), World Agroforestry Centre (ICRAF), CGIAR, CARE International and many local development partners. We also benefited greatly from the many workshops, co-design events, farmer reflections, and testing sessions during the development of this toolbox. We extend our gratitude to all those individuals who contributed to this collaborative effort. 5 Contents Introduction .................................................................................................................... 8   Social Learning and the Relevance Gap .................................................................... 9   Development of the Toolbox ....................................................................................... 12   Zero Draft Process ................................................................................................... 12   Testing Process ........................................................................................................ 18   Repeating the Cycle ................................................................................................. 23   Conclusion and Recommendations .............................................................................. 26   References .................................................................................................................... 29   6 Acronyms ALP CATIE CBA CCAFS CCSL CIS CRA CSA CSV FAO ICRAF IFAD K2A MAP MDP NGO Adaptation Learning Programme Centro Agronómico Tropical de Investigación y Enseñanza Community Based Adaptation Climate Change, Agriculture, Food Security Climate Change and Social Learning Climate Information Systems Climate Resilient Agriculture Climate Smart Agriculture Climate Smart Villages The Food and Agriculture Organization of the United Nations World Agroforestry Centre International Fund for Agricultural Development Linking Knowledge with Action Mesoamerican Agro-environmental Programme Masters in Development Practice Non-Governmental Organization 7 Introduction Starting in 2013, the Linking Knowledge with Action (K2A) team of the CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS) engaged in a social learning experiment that produced the “Gender and Inclusion Toolbox: Participatory Research in Climate Change and Agriculture”1 – a resource for concepts, learning activities and field tools for implementing gender transformative research in climate change programming. The toolbox was co-designed using multiple field-tests, consultation, and capacity and design workshops engaging dozens of scientists, development experts, and over 50 local partner organizations in West and East Africa, South East Asia and South America. The toolbox is a way of exploring a larger question around knowledge production within the epistemology of climate change and its links to development impact. Climate change has been described as a “wicked problem” – one whose complexity and discourse continuously changes and involves the interests of multiple actors (FitzGibbon 2012, Collins and Ison 2009; Pahl-Wostl et al. 2008). The production of knowledge around such a dynamic issue suggests the need for the research community to move beyond expertdriven science to one of co-production and social learning. Changing ‘research as usual’ in this way opens up the possibility of a more equitable climate science attuned to the needs and opportunities of communities, where the potential of scientific solution-making is limited only by openness to new ideas (Kristjanson et al. 2009, Snowden and Boone, 2007). Central to this change is the need for social and gender transformative research that encompasses and addresses the underlying structures of marginalization (see Box 1). On one hand, this means a greater accountability to the social contexts that climate change science enters into. Extractive research often falls short of providing solutions that can take hold in real life because the underlying causes and structures of social and gender inequality are not addressed. Gender transformative approaches on the other hand, which seek to improve access, power, norms, and relations, often provide greater opportunities to achieve lasting impact, particularly with marginalized groups. 1 Download Gender Toolbox: http://hdl.handle.net/10568/45955 8 Box 1: Aquatic Agriculture System’s Definition of Gender Transformative Approaches (2012) A gender transformative approach is distinguished from the following common approaches to gender integration as follows: § § § Gender neutral: programmes that distinguish little between the needs of men and women, neither reinforcing nor questioning gender roles Gender sensitive: programmes that recognize the specific needs and realities of men and women based on the social construction of gender roles Gender transformative: approaches that seek to transform gender roles and promote more gender-equitable relationships between men and women   One of K2A’s pathways for achieving a science for development is using elements of social learning, or research approaches that seek out collective understandings and collective responses by involving diverse partners and perspectives. Since an individual’s knowledge is limited to their unique experiences, learning with many people across many experiences may provide a more comprehensive and informed picture. A social learning approach to science also calls for iterative loops of action and reflection by participants, with each loop integrating new or different knowledge, values, norms and perspectives. This applies to learning approaches in general, although not all intentionally plan for more than a single learning loop, whereas ‘triple loop’ learning may be optimal (Shaw and Kristjanson, 2014). What the ‘social’ refers to in the social learning approach that CCAFS is particularly interested in is the common desire of widely ‘scaling out’ results of research for development– in our world, this implies reaching millions of smallholder farmers with lessons on climate smart agricultural options rather than hundreds or thousands. Another critical component to this is the potential to promote a more inclusive scientific approach that builds the capacity of researchers and farmers, particularly those belonging to marginalized groups, in joint sense-making and problem solving. Ultimately, this process of joint learning using meaningful participation can help support ‘a change in understanding that goes beyond the individual to become situated within wider social units or communities of practice’ (Reed et al. 2010). Social Learning and the Relevance Gap Current research practices are often criticized for operating through linear exchanges between researchers and communities and therefore reinforcing hierarchical power relations. Coupled with a lack of awareness around gender and social analysis, there are real research gaps that exist within climate change literature. For organizations like the CGIAR that operate in a highly networked and decentralized system, political will and institution-wide attempts to mainstream more participatory and gender-sensitive research practices may be particularly 9 challenging (Harvey et al. 2012). In some ways, co-production proposes both practical and theoretical challenges to traditional research. One of the emerging scholars on co-production, Catherine Durose (2011), points out that co-production places a high priority on social change and community-benefit and therefore can represent a trade-off for researchers when compared to publishing and other traditional forms of academic reward. The question of rewards and incentives as built into academia today can produce a ‘relevance gap’- where outputs in the form of development impact and understanding ‘usefulness’ from the perspective of climate vulnerable communities becomes peripheral to the research process. The relevance gap is not a new problem and scholars in critical anthropology, sociology, and development studies have discussed the need for democratizing knowledge production at large. Development thinkers like James Fergeson (1994) have pointed out the dangers of “expert knowledge” and reproducing inequality in development. Fergeson interprets knowledge-power in the tradition of Foucault, where discourse, information and knowledge is produced, legitimized and circulated through a political process of bureaucratization. In what Fergeson calls “the anti-politics machine”, the bureaucratic power of governments, nongovernmental organizations (NGOs) and other development actors have the capacity of creating and validating particular narratives over others, in effect directing what is considered a success or failure, priority or non-issue. This process of bureaucratization easily falls into the trap of reducing the heterogeneous voices and interests of real people and can strip development of its socio-economic, cultural, and political context. Paulo Friere (1970) highlights a similar idea in his concept of the ‘banking model’ of knowledge, where the oneway dissemination of knowledge which is often found in science, when practiced in any social context or institution with existing hierarchies can exacerbate or increase knowledge ‘monopolies’. Without addressing power, in other words, by taking into account gender and social analysis, the means of producing, controlling and using knowledge stays in the hands of the privileged few and in fact, prompts bias. Fergeson, Durose, Friere and many others point to the need to recognize how learning and knowledge production is practiced in different institutional and political contexts. Within the CCAFS context, the K2A team works under the umbrella of social learning as a way to shift research as usual in CGIAR. Some of the major benefits of these approaches to research have been highlighted by the CCAFS knowledge integration team in Box 1 (CCAFS 2013). 10 Box 2: Opportunities for Using Social Learning in CCAFS Research 1. People can be amazingly resourceful and innovative, given a supportive environment. Social learning draws on this untapped potential. 2. Unlocking this potential requires going beyond ‘business as usual’ and making use of diverse partnerships, multi-way communication, cooperation and collaboration. 3. If we want food security in a hotter world, we need to seize new opportunities for doing research differently. We can take participation further, to more proactive social learning through action and reflection — leading to changed behaviours. 4. “Why bother?” Doing this can sharpen our edge and help us become better problem solvers of bigger, more complex problems 5. The next steps involve shifting towards partnerships for joint observation, trials, modelling and experimentation. Trying new approaches is how we learn to make this happen. 6. We are asking people to embrace the idea of joint transformative learning and the cocreation of knowledge. Incentives and institutions also have to change. 7. In our vision of success, more scientists are engaged in broad partnerships, producing information that is more useful and more widely used. There is more mentoring of young people, more interactive science, and we all share our knowledge more generously This paper will trace the development of the toolbox from its early stages as an unplanned social learning experiment into what it is at present: a case study on possible bridges between different actors in development, research, and local communities on jointly producing useful knowledge for adaptation and mitigation options for climate change. A particular challenge for research organizations is to include development actors and local communities themselves as equal partners in the process of knowledge co-production. While the toolbox project was successful in prompting changes to the research process, this paper will also document those challenges and learnings. 11 Development of the Toolbox Zero Draft Process The toolbox project was part of the greater CCAFS gender strategy, and fits within the K2A target outcome story for gender and social differentiation2. It was initially conceived as a sister document to the CCAFS and FAO Training Guide, "Gender and Climate Change Research in Agriculture and Food Security for Rural Development“ (2011). Over the course of 2013, several iterative loops of learning were conducted to develop each chapter or “module” into zero drafts. The K2A team began revising the FAO and CCAFS manual in June 2013 by hiring a graduate student intern. The first revisions were made after a desk research phase that incorporated recommendations from testing the CCAFS and FAO training guide in Uganda, Bangladesh and Ghana (Jost et al. 2015). Five tools were improved and two new tools were added to the Climate Resilient Agriculture (CRA) Module of the emerging toolbox. The CRA module was tested in Nyando, Kenya, the following month. Some lessons learned in this early phase of the project included: § Linking with Development Organizations – The K2A team had identified the need to build its own capacity in gender by hiring more subject specialists for projects like the toolbox. Several discussions were initiated with international development partners to co-sponsor a joint position. The idea of sharing personnel as a way of linking institutional learning and exchange was not only difficult to communicate, but also logistically challenging because of highly divergent organizational structures. This early attempt at establishing innovative social learning pathways through shared staffing ultimately failed to take root. 2 This outcome called for, “Research and development actors [to] understand the importance of social differentiation in climate resilience research, and for adoption of climate smart agricultural practices at the household and community levels. Our partners are using our tools to design gender-targeted climate smart agriculture research and development programs, and are measuring the overall benefits for improving the adoption of climate smart agricultural practices”. The toolbox was a key activity to help build the capacity of regional partners and the project’s progress is documented in CCAFS’s monitoring and evaluation throughout. 12 § Linking with Academia - Intentionally partnering with practice-oriented degree programs that emphasize social science and community-based development was also a way of streamlining the orientation of the toolbox towards practitioners. The K2A team partnered with Emory University’s Masters in Development Practice (MDP) program, hosting 2 MDP students with social science backgrounds. Academic partnerships are a standard model within the CGIAR, however by connecting with a professional degree program the incentive to publish was lessened and likely added to a more exploratory social learning approach. The revised CRA tools were tested in the CCAFS baseline villages in Nyando, Kenya, with the World Agroforestry Centre (ICRAF) Kisumu Field Officer from July to August 2013, with the intention of: a) Gaining a deeper understanding of the political, ecological and social systems that enable adaptive capacity of socially-differentiated groups; b) Revising the tools and improving on their participatory and social differentiation analysis content. ICRAF Kisumu, a well-established field office with strong ties to community leaders and villages, was instrumental in establishing legitimacy and trust during pre-testing and testing. The field team came together with ICRAF staff every week to reflect on the testing process, adjusting and changing each tool according to their observations. The field team also communicated with the K2A team periodically on these changes. Pre-testing was potentially the most important component of toolbox development. It was an opportunity to introduce gender and social differentiation and qualitative methods to a field team comprised of quantitative and technical staff. With the majority of rural development workers lacking real exposure to gender analysis or the experience of working with women farmers and marginalized communities, practicing the toolbox methodologies with a field team with this level of experience was a strategic opportunity to simplify and improve the communication and training of gender concepts and participatory approaches. In relation to the farmers in Nyando, this early phase was one of the most extractive components of manual development because the results did not feed into a study or development project. Products of sessions such as village maps, daily clocks and seasonal calendars were left with communities, but since they were not embedded within a project the relevance of the products to the communities themselves may have been lost. 13 Pre-testing resulted in several methodological findings as summarized below: § Sequencing - The team experimented with sequencing activities that asked broad questions on the socio-economic contexts, constraints and priorities of men and women in the villages to establish more of an inductive approach to research. This phase reinforced the importance of building trust, rapport and transparency between researchers and local communities – essential for moving beyond a purely extractive research agenda. Throughout the pre-testing process and in future tests, the team spent significant time establishing “safe spaces” so that opportunities to engage focus group members across gender, social position, age and other factors could be better achieved. The same was true for single-gender focus groups, where the teams focused on creating ground rules and establishing a culture of discussion so that conflicts amongst participants could find space to be addressed rather than ignored. Eventually, early experiments with sequencing and introductory tools evolved into its own module in the toolbox called “Co-Production of Knowledge”, with the objective of stimulating community-defined visions of resilience and development and establishing a culture of discussion during a study. Sequencing activities to support dialogue with the community was also key for supporting socially differentiated knowledge so that biases and hierarchy within a community or focus group could be identified early and adjusted for during testing. § Problem Solving - Many focus group participants were willing to discuss sensitive topics such as conflict over natural resources, gender-based violence and changing attitudes about the role of men and women in agriculture. This was especially true when conducting sex-disaggregated focus groups, where participants often remarked never having reflected on how all these experiences in fact contributed to agricultural change and development. Questions encouraging focus groups to brainstorm potential solutions and strategies were added as a means of stimulating action plans to the tools. Again, adding problem solving components in sex-disaggregated focus groups highlighted the different coping strategies and social networks that men versus women engage in, and gendered strategies behind different social organizations. Action plans were not fully drafted because of ethical issues since CCAFS had not devised a way of following up 14 with them at this stage. There was however great willingness to engage on problem solving around issues suggested by the community focus groups themselves, such as communal management of dams, rivers, and grazing lands, women’s participation in trainings and governance, and levelling women’s access to climate information and technology. § Qualitative Probing - The research team was encouraged to promote storytelling and narratives with focus group participants. As a result, many examples of local knowledge and traditions shaping gender norms and production systems routinely came up. For example, several men’s and women’s groups highlighted the recent resurgence of taboos around women’s practice of agroforestry. A Luo traditional belief stated that women who planted trees would cause their husband to die. This belief was spread by young men in several villages in an attempt to control vegetable and fruit sales exclusively. Further probing of this dynamic revealed that higher awareness about HIV/AIDS helped challenge this taboo (public discourse about male death were linked to HIV/AIDS rather than women’s participation in fruit and vegetables planting). Similarly, focus group participants also identified more active targeting of women and men by local agroforestry projects as a means of spreading agroforestry in a gender balanced way, as well as challenging the gender norms and myths surrounding the practice. This shift away from survey-style questions to a more narrative one was challenging for field technicians in the research team, and thus resulted in developing longer learning activities to learn and promote qualitative probing as a skill. § Gender Sensitivity and Disaggregation - Nyando’s Kobiero village provided a rich experience on perceptions of gender roles and norms, highlighting the importance of understanding vulnerability using properly disaggregated data. The vast majority of households in Kobiero were headed by women whose husbands had died due to high HIV/AIDS prevalence. Households instead had either extremely elderly men or very young males, usually sons, who were de facto heads of household though not operating as breadwinner or decision-maker. Neighboring villages described Kobiero as “cursed” and having strained market, information, and social relations with other villages. It was difficult to prompt discussions around gender norms and agriculture because participants 15 would often obscure details around labor and decision-making within the household, misreport income sources, or choose not to participate. This made understanding the gendered burden of work, potential innovations due to scarcity, and the various coping and risk management practices in these women headed households very difficult. The research team did not probe beyond what Kobiero residents were comfortable with and experienced firsthand the importance of sex-disaggregation and intra-household probing, which calls for identifying women as individual farmers rather than relying on household level data. Once data collection for the CRA module was completed, a “zero draft” was composed that included changes in language and sequencing, and new versions of each activity based on testing and reflection with the field team and K2A team. The K2A team then presented on the testing process at a lecture event at the ICRAF Headquarters in Nairobi. The event helped the team to engage specific scientists to support and review the tools for future modules. The K2A team attempted a different approach to develop the Climate Information Systems (CIS) module. First an attempt to “crowd-source” what the major research questions and priorities for gender in CIS was done amongst subject experts in IFAD, FAO and the CGIAR. The K2A team asked the following basic scoping questions to begin a dialogue: a) What are the main priorities for gender research in climate information work? b) What are research gaps in gender and climate information systems? c) What participatory tools and approaches already exist in climate information systems? d) Recommended literature to help shape the module. This attempt to work across centres and organizations was limited to email-correspondences, but nevertheless helped shape a rough agenda for CIS. Social scientists in the CGIAR highlighted the importance of understanding indigenous knowledge systems and locally defined concepts of weather, climate, probability and forecasting. Subject specialists also highlighted the importance of building instruments that captured the information demands and modes of communication preferred by men and women. A lesson from the crowdsourcing phase was: § Outsider/Insider Sharing - While scientists were interested in the toolbox approach, establishing platforms for sharing were stunted within the CGIAR. CCAFS underutilized tools like the Climate Change and Social Learning (CCSL) group, Yammer and 16 Sandbox networks. There was more traction with scientists already working in social differentiation and particularly with new or younger researchers. One of the CGIAR’s strengths is in its branding, making partners inclined to collaborate and share as a means of being better associated with the CGIAR’s work. In this sense, those who are peripheral (perhaps because of lack of communication across projects and centres, lack of social capital, etc.) may be more inclined to collaborate. This good will and interest for sharing exists as exemplified by the many collaborators on the toolbox, however creating more pathways or incentives to share is needed. After crowdsourcing, the K2A team decided to experiment in co-designing. Working with ICRAF Kisumu, a week-long capacity building and co-design workshop was organized in Nyando, Kenya, with diverse participants from PhD students, technicians and farmers, to partner NGO field staff. New group-based capacity building activities on adaptive capacity, gender and sex, social differentiation and gender analysis, and theories on participation were adopted. Based on the experience of testing the CRA module, social learning approaches to reflection, discussion and eventually collective definitions of concepts were made into a basic training curriculum. These activities became the base for the Learning Activities chapter of the Toolkit, and were further refined with the help CARE International at a later phase. In the same workshop a thorough review of the CIS tools was done through roleplaying and practice, until each activity was improved. These tools were then pre-tested in separate men’s, women’s and youth groups with a local community in Kisumu. Every pretest day was followed by a reflection and evaluation session, and the seminar participants revised the tools according to farmer feedback as well as by discussing their own observations and experiences as a group. As a result of the co-design workshop, four qualitative tools were refined that linked the crowd sourced expert suggestions with qualitative methods shaped by hands-on testing and revision by the seminar team. In terms of capacity building, each participant visibly improved their skills in probing and asking open-ended questions as well as their confidence in concepts. Several key activities were also attempted to address power amongst the male and female, highly experienced versus newcomer, young versus old members of the ICRAF team. Participants later remarked that these activities challenged the culture of the ICRAF Kisumu office in a way that had not been before. The CIS module was finalized when the revisions from the Kisumu team were re-circulated to experts from IFAD, CARE, University of Florida and CGIAR scientists for comment. 17 In October 2013 the K2A team held its first Gender Training and Strategizing Workshop, hosting participants from the five CCAFS program regions, NGOs and government partners. The K2A and partner scientists co-facilitated a two-day qualitative gender training. The curriculum covered qualitative research design, focus group dynamics, best practices and basic concepts in climate change and gender that formed the foundation for parts of the Toolbox’s basic concepts chapters. Group-based learning activities practiced in Kisumu were improved and new activities were added to test with conference participants. Participants varied in ages, ethnicities, organizations, language ability, disciplines and gender, highlighting the importance of arriving at definitions and understandings through group reflection and participatory learning activities. In particular, many participant evaluations highlighted their appreciation for reflection exercises from the Toolbox on identity and power such as the “Identity Wheel” and “Exploring Gender & Sex”. These activities asked researchers to reflect and analyze their own subjectivity, biases, power and privilege when working in communities in a group setting. These activities served both as a team building exercise as well as a personal reflection tool to engender social differentiation analysis, connecting theory to one’s own experiences. The success of these tools and group-oriented learning was integrated into the toolbox. As for the final toolbox chapters, the mitigation and co-production modules were developed concurrently using similar processes as the CRA module: first by conducting desk research and integrating FAO-CCAFS Training Guide when appropriate, followed by sharing with experts, then reflecting and revising within the K2A team. These modules were pre-tested in Nyando, Kenya, in collaboration with the ICRAF Kisumu Office. A component of farmer evaluations and reflection were included during this pre-testing, and then the tools were finalized as a zero draft.   Testing Process Current research practices are often criticized for operating through linear exchanges between scientists and respondents. With a completed zero draft of the toolbox, the K2A team was able to move forward with their gender and social differentiation activities, focusing on capacity building using the toolbox methods. An opportunity to work more closely with CARE International emerged in 2014 with CARE’s Adaptation Learning Programme (ALP) in Ghana, which was also an opportunity to span boundaries across an organization with different interests, norms and competencies. A sixteen-day training was held in Tamale, Ghana, with the ALP team and local partners in June 2014. The three overarching goals of the training were to: 18 a) Build capacity of CARE’s ALP staff (comprised mainly of local partners) in participatory approaches, principles and tools, b) Revise the manual based on the expertise of CARE staff and real-time testing in ALP project villages, and c) Test a real research question using the toolbox. Participants of the training included four CARE staff and fifteen participants from partner organizations ranging from agricultural extension agents, NGO facilitators, to project field staff. While only five were women, the diversity of work experience of this group greatly enriched the learning process. Prior to the training, the main research question for testing the manual was discussed with CARE International and ALP leadership. It was decided that fieldwork during the training would explore the relationship between adaptive capacity and changing gender norms or specifically, testing ALP’s gender indicator 1.4 (DFID 2013): “Men and women in communities implementing community based adaptation (CBA) who report 1) equitable shifts in gender dynamics which contribute to improved adaptive capacity for climate-vulnerable people 2) gender-equitable implementation and benefits of adaptation.” The K2 and CARE Ghana team co-facilitated segments of the training, and wove in CARE frameworks on adaptive capacity and adaptation learning. The training in Ghana built on the Kisumu co-design workshop, using several of the improved learning activities from Kisumu. The Ghana training format included influence from CARE concepts, and was carried out in five main parts: 1) Training; 2) Pre-testing; 3) Research Design; 4) Testing; 5) Co-Analysis. Key aspects of the improved training were: § Establishing a Supportive Learning Culture - The trainings hinged on participatory sessions where the group set expectations and ground rules, and used energizers and discussion-based activities to learn and practice concepts. The training focused on group reflection and trust – qualities that participatory research teams need in order to successfully work together. There was a tone of “playfulness” and camaraderie throughout the training that helped navigate serious topics. The “Exploring Gender and Culture” tool pushed the group to deconstruct sex and gender binaries, moving even into transgender rights and the plurality of sexual and gender identities. While some of these topics touched upon 19 deeply personal and religious beliefs, the group overcame these challenges because of the open culture establish in the training. § The “Empathy Mapping” tool was also invited the group to consider their own biases regarding the communities they work with, and the subjectivity of experiences around climate vulnerability. This activity particularly helped male participants reflect on their own limitations for understanding women’s experiences, despite years of ‘gender and empowerment work’. § Addressing Conflict – During the training, issues with timing and stipends for partner NGO participants became an issue. Rather than ignoring these as peripheral to the training, the K2A team addressed the conflicts through discussion, challenging perspectives about their relevance and breaking down boundaries between what is considered within versus outside the research process. Without addressing expectations and stipends, the training would not have been conducive to learning. The training eventually led to pre-testing of toolbox in a community participating in the ALP program. By this time, the K2A team had stepped away from facilitation and handed over more responsibility to the participants to be accountable to the research goals. This worked really well as each team, having already practiced several times in workshop spaces, began to work with ALP farmers. It became apparent that the success of building group accountability early during training sessions allowed for much more ownership over tools and the research process. The participants began experimenting with each tool, based on the relevance of the research questions, adding new variations on tools, working more as a team and tweaking to the research process to run more smoothly. Following pre-testing participants reconvened for study design and planning, to review sampling, tool selection, and analysis based on the ALP research question on gender and adaptive capacity. It was observed that pre-testing really empowered participants to take control over the research design process, connecting tools with themes within the analysis framework on their own. Fieldwork to explore the research question was done in the ALP village of Jawani, at which point most of the focus groups were being run entirely by training participants with little to no intervention by the K2A team. Finally, a two-day co-analysis meeting using the data collected was conducted to clean, 20 categorize, and produce preliminary summaries on findings. The categories, based on the original research framework, were divided amongst the meeting participants for data management. Then each category and related findings were collectively discussed based on evidence gathered. This group sense-making process was very rich, however time limitations were a huge factor in being unable to complete the summaries of findings. The K2A team were tasked with completing a synthesis report remotely with inputs from CARE international to eventually produce a joint-learning publication. Some of the co-analysis processes were the following: § Most components of research from data entry to publishing are outsourced to consultants, despite the vast amount of knowledge that field staff have. Initially, this was one of the reasons there was such strong push back from workshop participants when the consultant explained that they were responsible for data-entry, which led to tension about time commitment, expectations, and stipends. § The lessons learned on how to improve note-taking were an example of how the research process at large can reinforce power relations and lack accountability, especially in terms of the testimonies and experiences of farmers. Many development organizations have expert facilitators who work with project communities in nuanced and empathetic ways. The capacity to wear both hats however, as facilitator and researcher, is a skill that is not demanded as much because of the way research is compartmentalized. Generally, analysis is coordinated by consultants or ‘experts’ with minimal exposure to the focus groups, while field staff are expected to be facilitators. As facilitators, many field practitioners view note-taking as “someone else’s responsibility”. This downward cascading of responsibility ultimately results in a great loss of data quality and missinterpretation of farmer-voices. By the end of the training, the participants were able to identify exactly the types of personal, recall, and other forms of biases that result from poor note-taking. CCAFSCARE challenged the status quo research process, and participants to have ownership over the entire research cycle from design, data collection, cleaning and data entry, to coanalysis. By sharing responsibility and spending time on explaining the analytical framework for the study, participants easily understood that despite great facilitation skills, their efforts to answer the research questions would be wasted if the field team did 21 not capture the details of focus group discussions. This dynamic occurred several times during co-analysis, where participants recalled very rich anecdotal evidence supporting changes in gender roles but could not cite sources from the transcripts and database. In order to acknowledge and address the note-taking problem, the CCAFS-CARE facilitators and training participants problem solved throughout the sixteen days. The group decided to record focus group sessions. Two instead of one note-taker was assigned for each focus group. More coordination around the pace of focus group session was done between note-takers and facilitators. A thirty minute de-brief session was instituted after focus group sessions, where the entire field team sat and collectively recalled each session as they digitized their notes. This de-briefing session was incredibly helpful to collect details, but also for the field team to assess their progress. Issues such as what research questions were emphasized, which groups were less willing to speak up and why, ethical issues that had come up and how to plan for their next session, were discussed. The de-briefings helped the participants view loss of data quality as an ethical issue in misrepresenting the voices and opinions of the communities they work with. § Major revisions were done to the Goal Tree exercise, eliminating most of the appreciative inquiry elements. The steps involved in each tool were pared down. Appreciative inquiry requires much stronger facilitation skills and stronger relationships with focus group participants than what may be expected for most users of the toolbox. § All elements that involved focus group literacy were also eliminated, in order to accommodate the diversity of focus group participants – no activities remained that required writing. § The seasonal calendar was revised to be more gender sensitive, incorporating gender roles formerly in farming systems matrix. § The wealth and vulnerability ranking was changed to be less invasive and gender specific by testing it in gender disaggregated groups and eliminating the use of actual 22 names of community members. § Tools focusing on women’s empowerment in the co-production toolkit were entirely revised to be more visual and interactive, after reflections and recommendations by the CARE team. Repeating the Cycle The toolbox was released to the public on 15 October, 2014. The K2A team envisioned the toolbox as an on-going project or ‘living document’, where continual use and testing would generate improved methods and knowledge in gender and climate change. Following the publication release, two key capacity building partnerships were made between K2A and the South East Asia and Latin America regional teams of CCAFS. These workshops were an opportunity for the K2A team to feed new learning into the design of the tools, approaches and best practices based on the Latin American and Asian regional context. The first workshop was held in Matagalpa, Nicaragua, with CCAFS regional partners in Latin America from the Tropical Agricultural Research and Higher Education Centre’s (CATIE) research program’s Mesoamerican Agro-environmental Programme (MAP). This workshop connected twenty-five participants representing twenty-one different local organizations across Honduras, Guatemala, Mexico, Nicaragua, Peru, Colombia, and Kenya to interact, learn and improve the toolbox content. Building on the training and testing process of previous K2A workshops, participants engaged in a combination of group-based learning activities and field-testing to connect theory with practice. Some observations and learning points from the Latin America training included: § CCAFS scholarship and products are often available in English only. A Spanish translation of the toolbox was prepared for the workshop. As the workshop progressed, bilingual participants were requested to edit the Spanish translation so regional definitions were corrected for wider use. § Having such a broad range of countries and local NGOs represented provided insight on re-arranging the presentation and content of the toolbox. Participants suggested that the modular format be made more flexible, so that each individual tool could also be used in a stand-alone version. Many of the Latin American participants pointed out the wide 23 potential for using the tools beyond research, often integrating their own expertise during pre-testing and resulting in several variations and options that expanded tool function. For example, many community members requested more exchange of experiences and knowledge before and after research sessions. Several research teams took the opportunity to use elements of each tool as a discussion opportunity on a wide range of subjects like women’s empowerment, violence against women, reproductive rights and farming practices after official research sessions had concluded. The use of tools, in other words, was easily amenable to participatory learning, advocacy and extension activities. § The Nicaragua training was a great departure in terms of facilitation. The K2A team was required to condense the workshop format to six days. As a result, many of the modules were done in small group activities, small group discussions and plenaries to save time and increase interaction. Since simultaneous translation was needed for facilitating between English and Spanish, three different translators worked as a team to support the English language facilitators. After two days of pre-workshop planning, six CATIE staff members were delegated as co-facilitators, sharing responsibilities over the training schedule in addition to the K2A staff. This was a testament to the practitioner-oriented content of the toolbox since the activities were easily understood and communicated by the new team. The team met several times during the workshop week to reflect and adjust facilitation technique, adjusting and improve facilitation as needed. § A video documentary was made, incorporating elements of reflection and participatory evaluation. Each workshop participant was asked at different times of the workshop to narrate the day’s events and key learning points, challenges and suggestions for improvement. Male and female farmers were also interviewed to reflect on the new participatory research approaches and suggestions for improvement. § The field practice of tools prompted a planning discussion at the end of the workshop on how to produce gendered knowledge on climate change in the different field sites that partners worked in. This emphasis on praxis was a significant learning moment for 24 program managers and researchers to reflect on the value and potential scaling out of participatory approaches in on-going programs. Shortly after Nicaragua, K2A partnered with CCAFS in South East Asia, bringing together twenty-two participants representing sixteen local organizations engaged in CCAFS research from Cambodia, Laos and Vietnam. The 10-day training, held in Vietnam, built on the format of the Nicaragua training, relying on a team of facilitators instead of the K2A team alone. A similar format of capacity building that combined group-based learning and field practice was used. Some of the observations and learnings from the workshop included: § Both male and female participants remarked that it was difficult to step outside of the traditional researcher roles, attributing this to gender, class norms and power differences in the South East Asian context. This in itself was a good first step in social analysis and reflection for transforming the research process. A female participant from Laos remarked that participatory approaches required a certain type of extroversion that went against the gender norms of Laotian femininity. For K2A, this highlights the importance of expanding the facilitator base so that more facilitators from the Global South are empowered to interpret and present participatory approaches. This would give an opportunity for participants to see alternatives to Global North interpretations and also adapt participatory and gender sensitive approaches to their own context. § The CCAFS South East Asia team wanted to use the training to initiate discussions amongst participants on action planning for integrating gender and participatory methods in their climate smart villages (CSV) sites. The action planning was done in regional and site-specific groups at the very end of the training. It brought up a larger discussion on the challenges of gender mainstreaming at CCAFS, because of the multiple accountability and management structures at play. While a great political will existed with the workshop participants and partner NGOs they represented, the new Flagship model and structure of funding made the implementation of action planning more uncertain. K2A’s strategy for scaling out gender and social differentiation tools and research involves not just building capacity of the field level research staff and partners, but coordinating gendered research upstream as well. 25 Conclusion and Recommendations The first FAO-CCAFS training guide was designed to address gender questions more aligned with the CCAFS themes of research on adaptation, climate information services and mitigation. However, the K2A team observed that climate change development and research partners needed a more generalized resource to help take into consideration social differentiation at the very beginning of the program design cycle. As the toolbox project progressed, working primarily with partner organizations became a priority to enable streamlining content into a development-oriented research resource. By documenting the social learning processes used to develop the toolbox, we were able to observe how changing research power and norms and shifting incentives through partnerships were possible despite the complexity of actors and interests involved. In terms of the Toolbox content and design, the tools did eventually achieve triple loop learning as the feedback from scientists as well as field testing, looped back to the K2A team in re-evaluating and changing their research modules, methods and practices. Further integration of the toolbox into CCAFS research, international partner programs and local partner projects is a key next step to evaluating the potential for participatory and social differentiation research. In this way, triple-loop learning can be engaged in generating new research questions with rural communities that are engaged in doing participatory research themselves. This has not been implemented yet, and this paper therefore focuses on lessons learned through social learning for developing the toolbox. Some of these major lessons learned are summarized below: i. Strategic Partnerships The innovation process in social learning theory lies in partnerships with boundary organizations or when divergent perspectives and competencies enrich one another (Nilsson and Swartline 2009, Pelling et al. 2008). Divergence was one motivation for partnering with CARE International, a global leader on gender and community based development. By embedding good research practices (competencies at CCAFS) with gender and communitybased development experience (competencies of CARE), the research process did not remain isolated as often happens within a purely research setting. Joint learning can happen in planned or unplanned ways, and for most of the toolbox project the exploratory nature of content development may have been a factor in the loss of several 26 partners – notably FAO, the main partner in developing the original training guide. Research interests shifted too autonomously to involve all planned development partners, while taking advantage of new partnerships as opportunities to work with flexible partners arose. In this sense, the triple loop learning that lead to the decision to orient the toolbox as a practitioner’s resource rather than a purely research resource, also led to changes in our partnership needs. ii. Co-Learning and Language The new toolbox attempted to move away from technical language towards a more ‘ordinary language’, to make research processes accessible to groups of varying scientific capacity. The use of ordinary language is a topic development studies and activists see as instrumental for framing research as a public good- shared and used by all. The prominent activist Arundhati Roy describes the relationship between language and development as following: “I think it’s vital to de-professionalize the public debate on matters that vitally affect the lives of ordinary people. It’s time to snatch our futures back from the ‘experts.’ Time to ask, in ordinary language, the public question and to demand, in ordinary language, the public answer” (Roy 2001) While Roy goes further than research organizations like CCAFS in articulating social justice, she nevertheless highlights the interconnectedness of language, discourse and development. As we move away from expert knowledge and technical language, the assumption that communities have the capacity to shape discourse and also their own development visions can better take hold. In this way, the role and sensitization of the practitioner in doing participatory research and social differentiation analysis is of utmost importance if we are to promote social learning across diverse groups. Key changes in approach, language, power relations and perhaps more specifically the task of shifting the value system of practitioners to encourage co-production and co-learning was explored more explicitly throughout the social learning, training and testing process of the toolbox. iii. Capacity Building for Whom? The toolbox project is a departure from many CGIAR efforts, because it discusses sharing control over the research process between upstream and downstream actors. Field-based practitioners are often the most knowledgeable members of a project next to farmers themselves, yet are undervalued and underutilized as resources and experts. This can cause breaks in accountability between field researchers and local communities, and field researchers and project management teams. 27 The K2A team came to the conclusion that shifting the research process to be more participatory and investing in downstream actors can be a critical intervention to improve gender and social differentiation analysis. If practitioners are not expected to share in the responsibility over research outputs, and in many cases have no idea what research objectives and approaches in a study are, conducting gender-sensitive research in particular becomes highly problematic. The K2A team encountered challenges targeting capacity building efforts, as most INGOs are moving away from centrally organized structures to more decentralized ones – where consultants rather than permanent staff are hired for coordinating research and design. Furthermore, implementation is rarely in-house and instead done through existing networks, local partner organizations, or government agencies. Embedding good practices in a development organization becomes more murky and complex in this structure, where downstream actors are increasingly disconnected from the upstream. In this context, partnerships can enrich learning across organizations in principal, but how it trickles down to the community level is still a difficult question to answer. For the CARE-CCAFS collaboration, much more effort was put into creating social learning interfaces, where the entire team gained more control over the design, testing and analysis of questions. The process of the collaboration showed great promise, however understanding how to enable social learning throughout disparate networks (in this case, across CCAFS and CARE) has yet to be explored. 28 References Aquatic Agricultural Systems (AAS). A Transformative Approach. http://aas.cgiar.org/penang-dialogues/building-coalitions-creating-change/gendertransformative-approach [accessed on 28 February 2015] CCAFS, 2013. Unlocking the potential of social learning for climate change and food security: Wicked problems and non-traditional solutions. Copenhagen, Denmark: CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS). Available online at: http://hdl.handle.net/10568/27781 Collins K., Ison R., 2009. Jumping off Arnstein's ladder: social learning as a new policy paradigm for climate change adaptation. Env. Pol. Gov., 19: 358–373. DfID, 2013. Annual Review of Project Title: Adaptation Learning Program for Africa (ALP). Department for International Development (DfID). Available online at: http://iati.dfid.gov.uk/iati_documents/4275930.docx Durose C., Beebeejaun Y., Rees J., Richardson J., Richardson, L, 2011. Towards coproduction in research with communities. AHRC Connected Communities Programme Scoping Studies, London, UK. Arts and Humanities Research Council (AHRC). Available online at: https://www.dur.ac.uk/resources/geography/reframing_state/CCDiscussionPaperDurose2et al.pdf Ferguson J., 1994. The Anti-Politics Machine: Development, Depoliticization, and Bureaucratic Power in Lesotho. Minneapolis, MN. University of Minnesota Press. FitzGibbon, J., Mensah, K.O., 2012. Climate change as a wicked problem an evaluation of the institutional context for rural water management in Ghana. Sage Open Press. Freire P. 1970. Pedagogy of the Oppressed, trans. Myra Bergman Ramos. New York, US. Herder and Herder. 29 Harvey B., Ensor J., Carlile L., Garside B., Patterson Z., Naess L.O., 2012. Climate change communication and social learning-Review and strategy development for CCAFS. Copenhagen, Denmark: CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS). Available online at: http://hdl.handle.net/10568/24456 Jost C., Kyazze F., Naab J., Neelorm S., Kinyangi J., Zougmore R., Aggarwal P, Bhatta G., Chaudhury M., Tapio-Bistrom M., Nelson S., Kristjanson P. Understanding gender dimensions of agriculture and CC in smallholder farming communities. In revision for Clim. and Dev. Kristjanson P., Reid R., Dickson N., Clark W.C., Romney D., Puskur R., MacMillan S., Grace D., 2009. Linking International Agricultural Research Knowledge with Action for Sustainable Development. Proc. Natl. Acad. Sci ., 9(13):5047-5052. Nelson S., Chaudhury M., Tranberg H., Lambrou Y., Tapio-Biström M., Kristjanson P., 2011. Training guide. Gender and Climate Change Research in Agriculture and Food Security for Rural Development. Copenhagen, Denmark: CGIAR Research program on Climate Change, Agriculture and Food Security (CCAFS) Available online at: http://www.fao.org/docrep/015/md280e/md280e00.htm Nilsson A.E., Swartling A.G., 2009. Social learning about climate adaptation: global and local perspectives. Stockholm, Sweden. Stockholm Environment institute (SEI). Available online at: http://www.sei-international.org/publications?pid=1535 Pahl-Wostl C., Mostert E., Tàbara D., 2008. The growing importance of social learning in water resources management and sustainability science. Ecol. and Soc., 13(1): 24. Pelling,M., High C., Dearing J., Smith,D., 2008. Shadow spaces for social learning: a relational understanding of adaptive capacity to climate change within organisations. Env. and Planning A, 40(4), 867-884. Reed M.S., Evely A.C., Cundill G, Fazey, I., Glass, J., Laing, A., Newig, J., Parrish, B., Prell, C., Raymond, C., Stringer L.C., 2010. What is social learning? Ecol. and Soc., 15(4): r1 Available onling at: http://www.ecologyandsociety.org/vol15/iss4/resp1/ Roy A., 2001. Power politics. Cambridge, UK. South End Press. 30 Shaw A, Kristjanson P., 2014. A Catalyst toward Sustainability? Exploring Social Learning and Social Differentiation Approaches with the Agricultural Poor. Sustainability 6(5): 2685-2717. Snowden D.J., Boone M.E., 2007. A leader's framework for decision making. Harvard Business Rev., 85(11): 68. 31 The CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS) is a strategic initiative of CGIAR and Future Earth, led by the International Center for Tropical Agriculture (CIAT). CCAFS is the world’s most comprehensive global research program to examine and address the critical interactions between climate change, agriculture and food security. For more information, visit www.ccafs.cgiar.org Titles in this Working Paper series aim to disseminate interim climate change, agriculture and food security research and practices and stimulate feedback from the scientific community. CCAFS is led by: Fund Research supported by: Fund Strategic partner: