Background The CGIAR Research Initiative on Low-Emission Food Systems (Mitigate+) will help low- and middle-income countries reduce food systems emissions without impeding progress toward development goals. Thereby, Mitigate+ uses an innovation systems approach. In close collaboration with partners, CGIAR researchers will develop and deploy evidence and tools to understand and untangle the complexities of the political, economic, cultural, and biophysical components of the food system’s greenhouse gas problem. To generate more impactful and demand-driven knowledge, the Initiative’s “place-based” participatory research will focus on working with stakeholders, including farmers and other actors of the local food systems, to collaboratively design, test, and deploy emission-reduction strategies. This approach is fundamentally different from top-down development models where “solutions” are designed and delivered without adequate participation from intended beneficiaries. A critical tool in this participatory research for development (R4D) strategy are living laboratories for people (LL4P). We understand them here as actor-centered innovation spaces facilitating co- creation of social and technical innovations in a real-life landscape setting. This document describes the rational and operationalization of one step of the participatory R4D process. In this particular step, we intend to co-create with stakeholders a theory of change oriented towards stakeholders’ development visions while exploring opportunities to reduce food systems emissions. It remains a critical challenge for researchers and practitioners to combine innovations and interventions in system changes to help achieve sustainable development goals. Ideally, theories of change establish this link by working backward from the outcomes to identify all the conditions that must be in place for the desired change to occur. Theories of change are based upon assumptions that are often implicit and not transparent. For instance, assumptions about which actors need to change which behavior and what drives their behavior are critical. The process described in this guide breaks down ambitious system change goals into strategies and interventions to stimulate change. It will require process participants to reflect on underlying behavioral change assumptions and deeper structures in the network of actors. Before proposing a structure for a workshop process, we describe the underlying concepts. A conceptual framework for a behavioral perspective on theories of change We elaborate the discussion of system transformation and related research by focusing on the determinants of the individual behaviors that interact in the system. System changes are essentially about people and their decisions and actions determined by their mindsets and psychological factors, social relations, their action resources, the bio-physical as well as institutional environment (Falk, 2008; Michie et al., 2011; Abercrombie et al., 2018; Figure 1). It is not the intention of this document to dive deeper into the different factors. Petit (2019) provides a very comprehensive overview of behavior drivers and unpacking their dimensions. We will highlight a few aspects which we consider critical in the context of agricultural R4D. In the domain of action resources, people’s production capacity is of critical importance (Schultz, 1971). Technological knowledge is a related aspect that has received much attention in R4D (Wigboldus & Brouwers, 2016). The understanding of systems is another human capital dimension that determines people’s innovation behavior (Meyfroidt, 2013), as they develop mental maps as internal representations of Figure 1: Key drivers of behaviour prototypical situations (Meyfroidt, 2013, Levine et al., 2015, World Bank, 2015). Relevant are also assets such as access to inputs and materials, such as seeds, fertilizer, machinery, or draft animals (de Roo et al., 2019). Another common assumption is that financial incentives can motivate behavioral change. Therefore, many projects integrate market or micro-financing mechanisms into their theories of change. The broader view of action resources is an invitation to reflect critically on specific actors’ key constraints in a particular innovation context. In the domain of mindsets, we want to highlight the role of motivation. Often, a constellation of emerging needs and interests initiate transition processes (Wiek et al., 2006). Motivation energizes and directs behavior (Michie et al., 2011). We acknowledge that people act to satisfy their various needs. Examples of the classifications of needs can be found in Maslow (1987) and Max-Neef (1991). Maslow (1987) categorized physiological, safety, love and belonging, esteem, and self- actualization needs. Max-Neef (1991) distinguished between subsistence, protection, affection, understanding, participation, idleness, creation, identity, and freedom needs. Assumptions about what determines actors’ behavior are often implicit and we can broaden our view to ask ourselves which of the need categories may be important for an actor in the context of the specific system at a particular point in time. Coordinating synergies and conflicts of interest is an important function of the governance system. In this context, we understand governance as the combined societal processes that coordinate actors’ relations and their experience of their actions’ costs and benefits (inspired by Ostrom, 2009; Woodhill, 2010; Loft et al., 2015). Key coordinating mechanisms are 1) the state/hierarchy; 2) markets/contracts; 3) self-organization/community (Ostrom, 1990). Diverse studies have shown that value chain governance (Barrett, 2008), horizontal coordination among smallholders (Develtere et al., 2008; Markelova & Mwangi, 2010), as well as government regulations (de Roo et al., 2019) affect smallholders’ technology choices. The concept of governance includes institutions that mediate individuals’ access to, and benefit from, action resources, and hamper and/or facilitate individuals’ ability to choose certain behavioral options, both at the local and national levels (Ellis & Allison, 2004). The governance framework influences behavior in multiple ways through conditional consequences associated with particular actions (Crawford & Ostrom, 1995; Falk et al., 2012). In this context, institutions are essential determinants of decision making because of their effect on capital availability and need satisfaction. Reconstructions of transitions often reveal that a bundle of latent and fuzzy incentives initiated at least the beginning of the process (Wiek et al., 2006). Institutions affect the way people access knowledge, inputs, and connections (de Roo et al., 2019). Keeping the individual behavior in mind, we move to the multi-actor network perspective. Figure 2 illustrates this view by showing a fully hypothetical actor network constellation. We assume that all individual and organized actors carry their action resources and mindsets as they enter relations (illustrated by arrows between actors inf Figure 2) as social spaces where actors learn, take decisions, and interact to achieve outcomes (Ostrom 2011). The social relations are framed by governance mechanisms as the third behavioral driver domain we highlighted. Identifying and understanding what drives actors’ behavior, their relations and related institutions is pivotal to initiate change at scale (Abercrombie et al., 2018). This understanding can be the starting point to influence behaviors strategically (Woltering et al., 2019). Figure 1: Individual decision making in a multi-actor actor network—A hypothetical example The facilitation processes We propose a multi-actor workshop process to develop theories for stimulating systemic change. The purpose of the workshop process is to allow LL4P stakeholders to co-create a common vision of change and develop strategies to pursue such change. We thereby build on other participatory facilitation processes which are conducted in the frame of MITIGATE+’s support to the LL4P. Most notably, we will build on net-mapping and mental model exercises as outlined elsewhere. Selection of participants In the earlier conducted net-mapping exercise, key actors have been identified. Organizations and individuals which most strongly influence the food system and the ones who are most strongly affected would be given priority to participate in the Theory of Change reflection process. This means they have been selected because of their role in the food system of the LL4P, from government, non-government and private sector and civil society. Farmers will be represented in the cooperatives for the different commodities. Special effort will be made to include female representatives of different actor groups. 1) Workshop preparations We advise having at least two facilitators to run the various steps, supported by documenters and observers. In this way, the facilitators can engage with the participants actively, while the discussions can be documented. Brainstorming observations by multiple facilitators and documenters afterwards is helpful to interpret the contents and understand group dynamics. Materials required: • A venue for the plenum and space for break-out groups (depending upon the number of participants) • White boards and/or pin boards • Blocks of white paper • Facilitation cards of different colors • Pens of different colors • Sticky notes (also small points for voting) • Lego-blocks The process requires a full day, which should be arranged according to participants convenience. The data will be used as input to update the MITIGATE+ intervention strategy. It will be condensed in spreadsheets and illustrations and then summarized in a report. At a later stage, a publication potentially using data from different countries is envisioned. While we will record who participated in the workshop, we will not record which participant made which specific contributions. Participants lists will not be included in publications. We prepare participants lists including the names, affiliations, key expertise and contact details of all participants who agree to share this information with the other participants. We print one copy of this list for each participant. We will also need one oral consent form for each participant. We want to use the net-map developed in a previous workshop process as starting point for this workshop. Therefore, please have the documentation of the net-mapping at hand. As you set-up the venue, we recommend a circular sitting arrangement. 2) The workshop process Step 1: Setting the scene (max. 60 minutes) The workshop facilitators welcome the participants. One of the participants will lead a prayer. The facilitators will then provide a brief introduction to the workshop purpose, MITIGATE+, and the ALL4P. This introduction will also include how the information co-created during the workshop will be documented, stored, and used. After these introductions, we ask the participants to sign the oral consent forms. Participants will then be asked to note on moderation cards what motivated them to join the workshop and what do they expect from it. The facilitators will keep the cards and stick them onto a pinboard. From there we will continue with some brief introductions. We will prepare and share participants lists beforehand to avoid lengthy explanations of people’s affiliations and key areas of expertise. Introducing each other will then be done in groups of four participants. The participants are asked to answer the following questions to each other in the groups: What do I want to bring into the world? What really matters to me? Step 2: Reviewing the net-map (30 minutes) Please re-create the net map using a white-paper, sticky notes, and Lego blocks. Place the net map in the middle of the room. The facilitators explain the symbolic of the blocks and arrows and then briefly summarize the specific net-map. Give the participants some 5 minutes to look again at the net-map. Then the participants are asked: From your perspective today, is there anything that deserves to be revised? Are actors missing? Should the power of actors be adjusted? Is there a need to revise some relations? If changes are proposed, document them and try to incorporate them into the net-map. Step 3: Deep system reflection (60 minutes) Using an embodiment practice, participants can intuitively experience the current state of the social-ecological-technical system, reflect on the potentials and blind spots, and sense a desired future. This step can help to think outside of the box in terms of actor relations and system dynamics. 4D mapping is a tool in the Social Presencing Theater toolbox the scholars in the Presencing Institute developed, most notably Arawana Hayashi and Otto Scharmar. 4D mapping makes the current reality in a social system visible and allows workshop participants to explore more visionary system states and the transition process that leads to it. In 4D mapping, the workshop participants adopt the role of system actors. They explore the particular actors by finding a place in the room and a shape that embodies the experience of its role in the system. Once all players have entered the space, the facilitator invites them to adjust the place and shape in a way that represents the system’s actors intuitively. The outcome of this step is referred to as Sculpture 1. In the last step of the exercise, the participants are invited to move again and continue to move until a desirable shape of the social body comes to a stop in what is referred to as Sculpture 2. The transition from Sculpture 1 to 2 can represent one pathway from the current system state to an emerging state. Reflecting on the actor constellations in the two sculptures may provide hints about potential directions, blockages and barriers, and opportunities for pathways to change (Hayashi 2021). A detailed description of the process and its facilitation will be provided in a separate document. Instructional videos can be accessed here: https://www.youtube.com/watch?v=cIHyD6vDTzw https://www.youtube.com/watch?v=-IGN7s-4knI A recording of a process Arawana Hayashi facilitated can be viewed at: https://www.youtube.com/watch?v=TUiGvHJYcXw Step 4: Enquiry of a joint vision (60 minutes) The next step has the intention to formulate a common ground for where the focal food system should move. This acknowledges that different actors may have different interests and aspirations. The purpose of this step is to find a common intention which can give direction to the actors involved to develop transformative system innovations. Once a vision has been formulated, alternative pathways towards moving in its direction become the core of the innovation process. The facilitators ask the participants to sit back, relax, and think about their current situation and the situation of the food system that is subject to the process. Allow them some time to be in silence. Then ask them: What would be the most important aspects of a future food system you would like to see. Be ambitious and courageous! Please note down each aspect on a separate moderation card. Next, the participants briefly share the key aspects which are part of their image of a desired future food system. The facilitators group the aspects and try to propose a condensed formulation which includes especially the most frequently mentioned aspects. Once a draft vision has been formulated, the participants can revise it. Write this vision prominently on a white paper. As Mitigate+ has the explicit agenda to reduce food systems emissions without impeding progress toward development goals, it will be important to connect the formulated vision to climate change mitigation and adaptation. Therefore, ask the participants: Is there a potential to reduce GHG emissions in the frame of the formulated vision? Is there scope to capitalize on the global interest to mitigate climate change in moving towards the desired future? Let participants share their views. Based on the input provided, add one or two additional sentences to the vision which are related to the reduction of food systems emissions. Step 5: Identification of entry points for system change (90 minutes) The next step of the process outlines pathways to change. This step builds on the previous ones. Proposed system changes should be directed towards the jointly formulated vision. They should be based on the actor network assessment as expressed in the net map as well as the insights gained during the deep system exploration. The facilitators ask the participants to keep all this in mind when answering the following questions: Which actors could most strongly influence a move towards the desired future? Avoid discussion but only collect a list of actors on a white paper. Then distribute the participants in three to four groups. Let each group select two to three actors which they would like to focus on. Avoid that different groups work on the same actors unless the participants strongly demand the same. The task of the groups will be to answer: What would these actors need to do (differently) to trigger a change towards the desired future? The actor should be named as explicit as possible. E.g. it may still be ambiguous to talk about what a ministry would need to do. Better is to tell who in a ministry would need to do what. This does not have to be an individual and should not be confrontational. In a similar way, also the behavioral change should be as explicit as possible. For instance, “creating an enabling environment” may not support innovative thinking. Better is to express what actions could support a more enabling environment. The groups should use a white paper and write on top the identified behavioral change and below the actor (see Figure 3). Leave at least the lower half the paper open for additional content to be added in Step 6. Use one piece of paper per behavioral change. Let the groups very briefly present the results of the group work. Step 6: Brainstorming about behavioral determinants (60 minutes) During this step, we enquire what could motivate, prompt, or enable the actors identified in Step 5 to take the desired actions. The facilitators broaden the view of participants by briefly introducing the three categories of behavioral drivers as outlined in the conceptual background section above (mindset, action resources, governance): What could motivate, prompt, or enable the actors identified in Step 5 to take the desired actions? Let the same groups as in Step 5 continue working together and ask them to write on their white papers they created in Step 5. An intermediary question to ask in this step may be: What prevents the actor from taking this action at this stage? Keep different behavioral drivers in mind! There is the risk that fast assumptions will be made about what determines an actor’s actions, or that political or power influences intimidate open discourse. To support the dialogue and generate refined suggestions it is critical that the facilitator ensures relevance and plausibility of the discussion. To encourage participants to think outside the box, we recommend that the facilitators join each group after approximately 10 minutes and ask them: Without doubting what you have identified, are you really sure that this is what would primarily motivate, prompt, or enable this action to be taken? Participants should list and briefly explain whatever could motivate, prompt, or enable the actor to change her behavior. Figure 3 provides Figure 3: An example of a group’s an exemplary white paper structure. whiteboard after Step 6 Step 6b(optional): Homework (introduction of homework app. 15 minutes; in workshop formats which allow more time, this could again be done in group work taking another 60 to 90 minutes) We can invite participants to continue the reflection process on their own after the workshop. Based on the motivating, prompting, and enabling behavioral factors identified in Step 6, participants could think: Who could motivate, prompt, or enable the actor to take the desired action? What could this actor do to allow or motivate the change in behavior? The answers to these questions can generate additional ideas for how changes in the system can be triggered. The participants can enter their answers into an Excel which has a structure like Table 1 below. The participants can share their tables with the facilitators after the workshop. Table 1: Summary of Actor Relations Primary actor What do they What could Which What could What can the (PA) who can need to do motivate influencing the IA do to project, etc., change critical differently to them, what actor (IA) enable or do to enable or behavior bring about could enable could take motivate the motivate the the desired them to make actions that PA to change PA or IA? system the change? enable or behaviour? Please be change? motivate the specific about PA to change who would behaviour? intervene. e.g., farmer e.g., avoid e.g., knowledge e.g., extension e.g., better e.g., develop clearing of forest about better soil services knowledge, capacities of management facility to co- extension staff develop practices, resources to visit farmers Step 7: Reflection and closing (30 minutes) The purpose of this step is to let participants step back one more time from all the details that have been discussed sense into the complexity of the system and its challenges. We want them to detach for a moment from all the confusing and contradicting signals and facts, some of which are related to each participant’s identity. We use the image of a wise council, which is a group of women and men with great knowledge and experience. The members look at the issues with an open mind and heart and without any agenda and independent on where they come from (sex, age, education, sector). We place chairs around netmap and the white papers prepared by the groups and ask the participants to take a seat in silence: Please take a seat on your chair of the council! Take a deep breath! Note all the complexity, the challenges and the opportunities which emerge from the picture in front of you. And for now, just sense what you see and feel. In this moment, you don’t have to fix it! Please speak out to the group by completing the sentences: I see… I feel… It is important to allow silence in this process. We have the reflex as facilitators to jump in when there is silence in the room. Having said this, it has helped in the past if one person from the team can start with sharing (after some time) statements completing I see…; I feel… Whatever is spoken should not be commented or discussed. Just let it sink in. Step 8: Closure (15 minutes) The facilitators will explain how the results of the workshop will be shared and used in next steps of the LL4P process. This will depend on the countries and stage of LL4P implementation. The facilitator thanks the participants for their time and imput. One of the participants will lead a prayer. Transferring the Results into a Theory of Change The results of this process can be translated into a theory of change in a logframe format. To harvest the results of the reflection process best, we recommend creating an illustration of the larger social network identified above. This can be done on a large paper; we used the VUE Tool to perform this digitally. The starting point to develop the theory of change can be the acknowledgment that project outcomes are in their essence changes in any actor’s behavior that result from the project’s activities and outputs. Following this logic, we propose to start building the theory of change with different actors’ (e.g., farmers) behavioral changes (e.g., use of a particular technology) that were identified in Step 5. Once project outcomes are identified in this way, it is important to connect them back to the visions articulated in Module 4. And it may be helpful to connect the vision to high-level targets, e.g., the Sustainable Development Goals (SDGs). Can achieving the vision be associated with impacts on the SDG level? Next, we propose to proceed from the outcomes to the outputs. This step would need to be linked to an initiative which intends to influence the system. The stakeholders co-innovating in the LL4P could initiate such an initiative. The MITIGATE+ Initiative can be another vehicle to trigger change. Outputs would then need to be identified at the interface of the motivating, prompting, and enabling factors identified in Module 6 and the capacities and capabilities of the ones driving an initiative. This step could either be subject to a next innovation workshop or it could be conducted by the MITIGATE+ team to identify what MITIGATE+. References Abercrombie, R., Boswell, K., & Thomasoo, R. (2018). Thinking big: How to use theory of change for systems change. UK: Lankelly Chase Foundation. Barrett, C. B. (2010). Smallholder market participation: Concepts and evidence from eastern and southern Africa. Food security in Africa. Food Policy, 33(4), 299-317. Crawford, S. E., & Ostrom, E. (1995). A grammar of institutions. American Political Science Review, 89(3), 582-600. de Roo, N., Almekinders, C., Leeuwis, C., & Tefera, T. (2019). Scaling modern technology or scaling exclusion? The socio-political dynamics of accessing in malt barley innovation in two highland communities in Southern Ethiopia. Agricultural Systems, 174, 52-62. Develtere, P., Pollet, I., & Wanyama, F. (2008). Cooperating out of poverty. The Renaissance of the African Cooperative Movement. ILO, Geneva, Switzerland Ellis, F., & Allison, E. (2004). Livelihood diversification and natural resource access. Overseas Development Group, University of East Anglia. Falk, T. (2008). Communal farmers’ natural resource use and biodiversity preservation: A new institutional economic analysis from case studies in Namibia and South Africa; [Biodiversity Transect Monitoring Analysis in Southern Africa, BIOTA]. Cuvillier Verlag. Falk, T., Vollan, B., & Kirk, M. (2012). Analysis of material, social, and moral governance in natural resource management in southern Namibia. International Journal of the Commons, 6(2): 271–301. Hayashi, A., & Dutra, R. (2021). Social presencing theater: The art of making a true move. Presencing Institute. Cambridge, USA Levine, J., Chan, K. M., & Satterfield, T. (2015). From rational actor to efficient complexity manager: Exorcising the ghost of Homo economicus with a unified synthesis of cognition research. Ecological Economics, 114, 22-32. Loft, L., Mann, C., & Hansjürgens, B. (2015). Challenges in ecosystem services governance: Multi-levels, multi- actors, multi-rationalities. Ecosystem Services, 16, 150-157. https://doi.org/10.1016/j.ecoser.2015.11.002 Markelova, H., & Mwangi, E. (2010). Collective action for smallholder market access: Evidence and implications for Africa. Review of Policy Research, 27(5), 621-640. Maslow, A. H. (1987). Motivation and personality, 3rd ed. Longman, New York. Max-Neef, M. A., Elizalde, A., & Hopenhayn, M. (1991). Human scale development: Conception, application and further reflections. Apex Press, New York, London. Meyfroidt, P. (2013). Environmental cognitions, land change, and social–ecological feedbacks: An overview. Journal of Land Use Science, 8(3), 341-367. Michie, S., Van Stralen, M. M., & West, R. (2011). The behavior change wheel: A new method for characterising and designing behavior change interventions. Implementation Science, 6(1), 1-12. Ostrom, E. (2009). Understanding institutional diversity. Princeton University Press. Ostrom, E. (2011). Background on the Institutional Analysis and Development Framework. Policy Studies Journal 39(1):7–27. Petit, V. (2019). The Behavioural Drivers Model: A Conceptual Framework for Social and Behaviour Change Programming. UNICEF Schultz, T. W. (1971). Investment in human capital. The role of education and of research. Free Press, New York. Wiek, A., Binder, C., & Scholz, R. W. (2006). Functions of scenarios in transition processes. Futures, 38(7), 740-766. Wigboldus, S., & Brouwers, J. (2016). Using a Theory of Scaling to Guide Decision Making: Toward a Structured Approach to Support Responsible Scaling of Innovations in the Context of Agrifood Systems. Wageningen University & Research. Woltering, L., Fehlenberg, K., Gerard, B., Ubels, J., & Cooley, L. (2019). Scaling–from “reaching many” to sustainable systems change at scale: A critical shift in mindset. Agricultural Systems, 176, 102652. Woodhill, J. (2010). Capacities for institutional innovation: A complexity perspective. IDS Bulletin, 41(3), 47-59. World Bank. (2015). World development report 2015: Mind, society, and behavior. The World Bank. doi: 10.1596/978-1-4648-0342-0 International Food Policy Research Institute CGIAR Initiative on Low-Emission 1201 Eye St. NW Food Systems Washington DC 20005 USA https://www.cgiar.org/initiative/low- Tel. +1 202 862 5600 emission-food-systems © 2022 International Food Policy Research Institute (IFPRI), CGIAR and Wageningen Food and Biobased Research. This publication is licensed for use under a Creative Commons Attribution 4.0 International License (CC BY 4.0). To view this license, visit https://creativecommons.org/licenses/by/4.0. .