Power and Empowerment in Transdisciplinary Research: A Negotiated Approach for Peri-Urban Groundwater Problems in the Ganges Delta

. The co-creation of knowledge through a process of mutual learning between scientists and societal actors is an important avenue to advance science and resolve complex problems in society. While the value and principles for such transdisciplinary water research have been well established, the power and empowerment dimensions continue to pose a 20 challenge, even more so in international processes that bring together participants from the global north and south. We build on earlier research to combine known phases, activities and principles for transdisciplinary water research with a negotiated approach to stakeholder empowerment. Combining these elements, we unpack the power and empowerment dimension in transdisciplinary research for peri-urban groundwater management in the Ganges Delta. Our case experiences show that a negotiated approach offers a useful and needed complement to existing transdisciplinary guidelines. Based on the results, we 25 identify responses to the power and empowerment challenges, which add to existing strategies for transdisciplinary research. A resulting overarching recommendation is to engage with power and politics more explicitly and to do so already from the inception of transdisciplinary activities, as a key input for problem framing and research agenda-setting. institutional research used local reports of the negotiated approach meetings as a source of data on the community’s problem 550 perceptions.


Introduction
Sustainable groundwater management faces various challenges that lend themselves well for transdisciplinary research, including the challenge of social participation and coordinated approaches between multiple actors such as scientists, government agencies and groundwater users (Barthel et al., 2017). This is also true for groundwater management in peri-urban 35 areas. Peri-urban areas are spaces in transition that connect urban and rural environments and that show features characteristics of both (Allen 2003;Mc Gee 1991;Singh and Narain 2020). Here, rapid urbanization often results in an increasing pressure on groundwater resources as a source of water for both local livelihoods and household uses, industrial activities and various urban needs. As dynamic spaces in transition, peri-urban areas there featureis a large diversity and heterogeneity in the actors and interests in peri-urban areas, combined with institutional overlaps, voids and ambiguities (Allen, 2003;Gomes and 40 Hermans, 2017;Narain and Roth, 2022).
In peri-urban areas, water-dependent livelihoods such as farming and fishing may still abound. Proximity to urban and industrial centres may create a spike in real-estate development, and new actors enter the scene. Migrants from more remote rural areas may be attracted by the proximity of urban centres for employment and opportunity, while urban residents and developers may be attracted by available spaces and land. These actors compete for, or threaten the quality of, existing (ground) 45 water resources, such as larger industrial or agro-industrial users, urban water users, and waste (water) disposal activities (Narain et al., 2013;Gomes, 2019). Increased climatic variability, degrading surface water sources, land use change, coupled with unequal power structures, rules, norms and practices, create pressure on already stressed water resources and lead to uncoordinated overexploitation of groundwater aquifers (Narain et al., 2013;Hasan et al., 2019;Banerjee and Hermans, 2020).
These increasing demands and pressures, for different users and purposes, are combined with often limited information and 50 knowledge about the actual state of groundwater quantity and quality (Olago, 2019).
Power differences play a large role in the groundwater management in peri-urban areas. As highlighted by political ecology analyses around water governance, power is a key factor shaping differential access to resources (see Swyengedouw 2009;Bryant and Bailey, 1997). PeriurbanPeri-urban water resources tend to be reappropriated and reallocated, whereby some water users tend to get deprived of access to the resource (Banerjee and Hermans, 2020;Narain and Roth, 2022). The resulting lack 55 of access to groundwater during critical periods affects the livelihood securities of the vulnerable and contributes to the incidence of poverty (Banerjee and Jatav, 2017;Butsch et al., 2021).
These combined features around of groundwater management in periurbanperi-urban areas result in complex situations that match the classic definition of a a "wicked" problem situation, at the juncture where conflicting goals and equity issues meet with knowledge limitations and contested problem formulations (Rittel and Webber, 1973). Such complex or wicked problem 60 situations are typically what transdisciplinary research hopes to engage with. Transdisciplinary research has been on the rise as a process of co-creation of knowledge by science and society to offer solutions for complex problems in human-water systems (e.g. Steiner, 2015a, Krueger et al., 2016;Ferguson et al., 2018;Ghodsvali et al. 2019;Sapkota, 2019;Pohl et al., 2021). In this co-production of knowledge, stakeholder participation and empowerment, as well as dealing with Ghodsvali et al., 2019;Van Breda and Swilling, 2019).
In transdisciplinary research, the differences in the types of knowledge and experiences that different groups bring to the table, are mixed with established structures for social interactions and the associated power and political dimensions (Jahn et al., 2012;Krueger et al., 2016;Brown, 2018;Pohl et al., 2021). Who is participating in the joint problem articulation and the research efforts, how are these participants selected and how do they report back to their fellow community members? What 70 is needed for these various groups to effectively communicate with each other, and to appreciate the depth and breadth of each other's knowledge and experience? Especially when dealing with relatively vulnerable communities, who are not usually involved in research or decision-making, such as is the case for peri-urban communities, these issues of power and empowerment cannot be ignored. Glossing over critical power inequalities may not always be critical for researchers and the production of new scientific knowledge, but it will not help to resolve wicked problems in ways that are scientifically sound, 75 equitable and socially sustainable.
In this paper, we lookThus, there is need for strategies that help to address deal with power differences and empowerment issues in transdisciplinary water research, for local groundwater management in peri-urban communities. In 2013, we started a multi-year transdisciplinary water research project that aimed to support groundwater users in peri-urban communities in Bangladesh and India. An international team of researchers and non-governmental organizations worked to develop new 80 scientific knowledge and approaches, while at the same time develop the capacity of local stakeholders to improve groundwater management. From the start of our activities, we were aware of the need to navigate power differences and of the difficulties of combining meaningful societal activities with scientific research. At the same time, we did not find clear-cut recipes to cope with all those challenges in existing accounts of transdisciplinary research.
In this paper, Wwe contributereflect on our insights and experiences from case experiences with transdisciplinary water 85 research and stakeholder empowerment in peri-urban communities in Bangladesh and India. This is done by complementing insights from the literature on transdisciplinary water research with a so-called negotiated approach for stakeholder empowerment (Leeuwis, 2000;Koudstaal et al., 2011). We contribute our insights from case experiences with transdisciplinary water research and stakeholder empowerment in peri-urban communities in Bangladesh and India. This negotiated approach accepts that social learning is characterized by power differences and strategic behaviour, rather than presuming a neutral 90 dialogue among equals. It uses principles from negotiation literature to support a transformative change towards more local self-governance of natural resources, while also seeking to use and enhance the joint knowledge base. The next section summarizes the relevant literature on the combination of transdisciplinary research and approaches that help deal with power, empowerment and conflict. This is followed in subsequent sections by case experiences with peri-urban groundwater management in the metropolitan areas of Khulna and Kolkata. The findings from these experiences result in practical lessons 95 and suggestions for a more power-sensitive transdisciplinarity, after which we conclude with some final take-aways.
2. Transdisciplinary research and stakeholder capacity development for peri-urban groundwater management 2.1 Transdisciplinary research

Core concepts and known challenges in transdisciplinary research
Transdisciplinary research is a process of mutual learning among scientists across disciplines and societal actors aimed at 100 creating knowledge that benefits both scientific praxis and discourse, as well as societal problems (Jahn et al., 2012;Lang et al., 2012;Scholz& Steiner, 2015a). There are various conceptualizations of transdisciplinary research, which describe transdisciplinary research as a process of mutual learning whereby science and society interact to develop new knowledge (Max-Neef, 2005;Jahn et al., 2012;Lang et al., 2012;Brandt et al, 2013;Seidl et al., 2013;Scholz & Steiner 2015a;Brown, 2018;Cundill et al, 2018;Djenontin & Meadow, 2018;Fam et al., 2018). With its emphasis on co-creation of knowledge 105 between scientists and local actors outside academia, it is closely related to, and for many practical purposes often indistinguishable from participatory action research (Whyte et al., 1989;Bradbury, 2015) and other participatory, interactive and community-based approaches (Lang et al., 2012). When it comes to human-water systems, transdisciplinary water research has been explored by Krueger et al., (2016) to see where and how water knowledge is produced in society. Transdisciplinary water research has been used for instance as means for more systemic learning on water security issues (Steelman et al., 2015) 110 and of for stakeholder engagement for broaderand impact of water scarcity modelling (Ferguson et al., 2018). Transdisciplinary water research has also been studied for its role in food-water-energy nexus research to support the achievement of sustainable development goals (Ghodsvali et al., 2019).
All these approaches use a systematic method of inquiry to assist societal actors in improving their actions for addressing societal problems (Bradbury, 2015), while generating methodological innovations and new empirical and theoretical 115 knowledge related to the problem field (Lang et al., 2012). In this interaction, different actors bring in their own perception of reality, thought-styles, roles and practices of communication, whereby (scientific) knowledge is combined with understanding rooted in deep experience (Max-Neef, 2005;Jahn et al., 2012;Pohl et al., 2021). In this process, three types of actors play a key role: i) Stakeholders such as local water users and other people directly related to the water resource, but also NGOs or companies; (ii) legitimized decision-makers such as policy advisors, government officials and elected political representatives; 120 and (iii) the science community with scientists from academia, applied research institutes and think-tanks (Seidl et al., 2013;Scholz and Steiner, 2015a).
Transdisciplinary science generally distinguishes three main phases, each of which has various challenges: problem framing; co-creation of solution-oriented knowledge; and re-integration of knowledge with scientific and societal practice (Jahn, et al., 2012;Lang et al., 2012;Brandt et al, 2013;Scholz & Steiner 2015b;Steelman et al. 2015). Table 1 shows an illustrative list 125 of these phases and their challenges, based on Lang et al. (2012) and Steelman et al. (2015).
INSERT  Table 1 shows that many of the key challenges relate to the interactions between the different types of actors and the representation of their interests. It starts from the very first phases, with a potential lack of awareness, ownership and legitimacy. This is also is in line with Jahn et al. (2012), Klenk and Meehan (2015) and Pohl et al. (2021), who reflect on integration in transdisciplinarity. Withoutobserve that without further scrutiny, the concept of integration in transdisciplinarity easily conceals problems with differences in values, knowledge and power (Klenk and Meehan, 2015). Ghodsvali et al. (2019) 135 observe also note the apparent challenges involved inthat fewer papers report stakeholder engagement that goes beyond instrumental levels, in transdisciplinary water nexus research beyond the instrumental levels, and take this as an indication of the challenges involved.
Some noteworthy exceptions are present though. Brown (2018) describes experiences with collective learning to enable local communities to cope with sustainability challenges. Process structure and open learning attitudes are identified as the two 140 critical ingredients The used transdisciplinary approach was modelled after the experiential learning cycles of David Kolb, modified for use in these collective social learning processes (Brown, 2018). Process structure and open learning attitudes are identified as the two critical ingredients. Krueger et al. (2016) discuss fairness and competence as two important criteria for participation in transdisciplinary co-production of knowledge. Fairness signals the need for everyone with an interest to participate, and to be recognized as valid voices in the process. Competence emphasizes the use of clear rules and procedures 145 in the participation process (Krueger et al., 2016). Cundill et al. (2018) similarly stress the importance of careful process design in their experiences with a global transdisciplinary research initiative, taking into account the influence of legal agreements, power asymmetries and institutional values and cultures.
Thus, a clear process design, fairness and open attitudes are known principles for stakeholder engagement in transdisciplinary water research. However, applying these principles can be difficult. When it comes to complex and wicked societal problems, 150 knowledge, learning, capacity and power are intertwined (Rittel and Webber, 1973;Brown, 2018: 285). This limits and complicates joint problem solving (Jahn et al., 2012;Klenk and Meehan, 2015) and makes open dialogue, participatory modelling and scientific knowledge limited as source of undisputed solutions (Barnaud et al., 2010). The questions, assumptions and scenarios included in scientific studies will need to reflect those of societal stakeholders, making them inherently subjective and suited for some problem framings but not others (Godinez-Madrigal et al., 2020). Therefore, 155 transdisciplinarity requires approaches for collective learning that help navigate the dimensions of power and fairness in the interactions within and between the various groups of scientists, government agencies and societal water users.

Power and empowerment in transdisciplinary research
Transdisciplinary scholarship is not blind to the issues of power and fairness. For instance, it recognizes the need for, and 160 difficulties in, establishing a safe platform for joint learning and discovery (Jahn et al., 2012). It also recognizes the importance 6 of representation of different types of stakeholders, including local water users and community stakeholders (Seidl et al., 2013;Scholz and Steiner, 2015a;Dyer et al., 2014). And tTransdisciplinary research in an international and developing world context recognizes the importance of dealing with institutional cultures (Cundill et al., 2018), institutional ambiguity and informality (Van Breda and Swilling, 2019). What the transdisciplinary literature does not yet offer, is guidance on how to enable a process 165 and platform for reflexivity and joint learning in a context of power differences, conflicting interests and institutional diversity, ambiguity and informality.
Current guidance and experience is shared only through fairly abstract phrases such as the need for "mechanisms to support mutual learning" and taking the necessary time (Raymond et al., 2010). However, in many cases participation requires not just taking the effort and time to invite stakeholder representatives and raise their problem awareness, but also requires empowering 170 and capacitating different types of stakeholders to participate and collaborate effectively (Richards et al., 2004;Krueger et al., 2016: 380).
In a context of power differences and competing interests, transdisciplinarity requires two types of capacity building and empowerment. It is not just the capacity of all actors to participate in the knowledge and learning process on an equal footing, but also the capacity to influence and act more effectively in processes of problem solving for water management. Since 175 transdisciplinary water research seeks to combine scientific knowledge development with societal problem solving, those two types of empowerment are of equal importance. Truly engaging with this dual empowerment dimension is relatively novel (Massuel et al. 2018;Steelman et al., 2015: 596).

A negotiated approach to empowerment and transdisciplinary problem solving
The need to address power dimensions in stakeholder participation has been recognized by development practitioners (e.g. 180 Bebbington et al., 2006;Sneddon and Fox, 2007;Barnaud et al., 2010). This has led to different approaches, including a negotiation negotiated approach, starting from the shortcomings of participation models such as social learning or participatory decision-making to deal with conflict (Leeuwis, 2000). Rather than approaching participation as collective decision-making or knowledge co-development, participation should be approached as negotiation: "If in practice participatory projects emerge as `arenas of struggle', and if stakeholders tend to act strategically, rather than communicatively, then why not base 185 methodological approaches on these assumptions?" (Leeuwis, 2000: 946). Including more explicit attention for strategic behaviour would also provide better outcomes of negotiation and participation process for disadvantaged groups (Edmunds and Wollenberg, 2001). Building on work in relevant fields such as consensus building (Susskind et al., 1999), network management (De Bruijn and Ten Heuvelhof, 2008) and negotiation analysis Fisher et al. (2011), different tasks for an integrative negotiation process were thus identified (Leeuwis, 2000). 190 In parallel to, and interaction with, this academic development, civil society organizations had similar observations and experiences, reaching similar conclusions. Their experiences and the academic reflections transpired into practical guidelines for a so-called negotiated approach over the past years (Koudstaal and Paranjpye, 2011). The earlier participation and coproduction activities such as ensuring access to knowledge development for local platforms, continuous learning, and recognizing community knowledge as well as rigorous and innovative science, are remain still important pillars: Access to 195 knowledge development for local platforms and continuous learning, recognizing community knowledge as well as rigorous and innovative science. However, the negotiated approach uses this as part of a larger aim, which is a transformation of governance, i.e. moving towards self-governance of local communities. For this, it follows the tasks proposed by Leeuwis and Van den Ban (2004) and the notion of 'principled negotiations' as described and popularized by Fisher et al. (2011). In principled negotiations, parties focus on their underlying values and interests, rather than on specific positions regarding 200 preconceived specific negotiation outcomes. This is somewhat similar to the difference in negotations between "creating actions, designed to build a bigger pie, and claiming actions, designed to obtain a larger share of the pie so created" (Raiffa, 2002: 2).
The negotiated approach offers eight tasks as guidance, and, as can be seen from Table 2, these tasks connect well to some of the challenges identified for transdisciplinary research. This is especially visible for the transdisciplinary research challenges 205 related to participation, joint ownership and legitimacy of the process and its outcomes.

Methodology and data
An open question is how to combine these empowerment processes with transdisciplinary knowledge co-production. We 210 investigate this question, using the main phases for transdisciplinary research as described in Table 1, combined with the main tasks for a negotiated approach as provided in Table 2. In the subsequent sections, we share our experiences with combining transdisciplinary research with the negotiated approach to address the challenges in groundwater management in peri-urban Council under its Urbanizing Deltas of the World programme and had an explicit focus on transdisciplinarity, combing scientific research with sustainable development. In the project, team members from Bangladesh, India and the Netherlands cooperated to enhance understanding and build capacity with local stakeholders to support sustainable groundwater management in peri-urban Kolkatta and Khulna. Project partners in India consisted of staff from SaciWATERs, a consortium 220 for interdisciplinary water research with expertise in socio-economics and peri-urban water governance, and The Researcher, a civil society research organization that supported the community and stakeholder engagement activities in Kolkata in India,.
In Bangladesh the Institute of Water and Flood Management of Bangladesh University of Engineering and Technology In the description of our experiences, we follow the main phases, tasks and activities as identified in Tables 1 and 2 above. In 235 doing so, we pay specific attention to the interactions and interfaces between researchers, local communities, and state/government actors. Although tThe three main transdisciplinary research phases of problem framing, co-creating knowledge and re-integrating knowledge help to structure our account, together withand the eight negotiated approach tasks help to structure our account,. However, it is important to note that activities often overlap and that the process always features various iterations, going back-and-forth between phases and activities. It is less of a linear and more of an interative and 240 circular process.

Case introduction: The Shifting Grounds project and its early project design and problem framing
The Shifting Grounds project was jointly formulated in 2013 through international workshops of researchers in collaboration with government stakeholders and local community representatives. The aim was to combine research, capacity building and development activities to address peri-urban groundwater problems in cities in Bangladesh and India. Khulna and Kolkata 245 were selected as project cities, being both located in the Ganges delta, sharing some key hydrological and geophysical features, but being located inwith different institutional contexts. The international project team sought a conscious mix between a research-initiated process and a community-initiated process to enable a balanced effort of co-creation of both scientific knowledge as well as practical solutions.
The project started with the ambition to combine transdisciplinary research and the negotiated approach, given the expected 250 differences in groundwater access, dependence and power within peri-urban communities. The consortium benefited from earlier research cooperation on peri-urban water security between partners in India and Bangladesh, and from extensive experiences of civil society partner, Both ENDS, with the negotiated approach. The initial project design targeted peri-urban villages near each of the two cities. Site selection criteria included scientific suitability as well as willingness and (basic) abilities of village stakeholders to engage with the project. For the latter, we looked at the existence of a nucleus for self-255 organization, such as the presence of an active community-based organization or local village committee that had also identified groundwater-related problems as an important issue for village development. The latter was used to ensure a workable fit with the initial problem framing around groundwater issues, which had been decided early on by the core project team members as a key research gap for peri-urban water security in the region.
The project was designed around three distinct research activities, along with community empowerment. Two PhD researchers 260 and one postdoc researcher were engaged: the first two, to study physical groundwater systems and local institutions, respectively, and the third, to study socio-economic and livelihoods dynamics. Community empowerment focused on capacity building within the peri-urban communities and on strengthening links of community actors with external government processes and state actors. The community empowerment was led by civil society partners in the project consortium and was referred to as the negotiated approach process; the research process was led by the research organizations. Both functioned 265 together as one team, with joint problem formulation and frequent project team meetings. Key policy-makers and local experts were represented in a Project Advisory Group.
This team constellation was purposefully designed to allow civil society organizations to use their experience and expertise in facilitating (sensitive) processes within the community, while enabling researchers to bring in their research expertise and knowledge. The frequent meetings within the project team helped provide shared understanding on problem framing and 270 process design, as well as a space where different team members could benefit from each others' strengths, expertises and positions within local and national networks. This also brought sometimes tensions, dilemmas and power differerences inside the project team. Through clear arrangements and agreed responsibilities, combined with frequent meetings, we have tried to navigate those.

TDR Transdisciplinary Research Phase A: Problem framing and team building
Negotiated Approach NA Task 1: Preparing the process and Task 2: Reaching agreement on process design

The research-government interface
At the government interfaceFor the activities in Kolkata, the project worked with the two distinct systems in place for decisionmaking processes in the State state of West Bengal: an administrative and a political system. The administrative government 280 system was run from the State state level, via Districtsdistricts, to provide important services to the communities. This administrative system had a hierarchical structure, with an important role for the District Magistrate that operated from Kolkata, and the local Block Development Officer at the block local level.
In the preparation phase of the project, connections with this administrative system were established via connections contacts with the formal decisionmakers and state-level water agencies. Representatives of some of these agencies were invited as 285 members of the project advisory committee. To gain access to these state representatives, the personal network of one of the Indian researchers proved to be essential. The research components in the Shifting Grounds project were highlighted, whereby especially the groundwater research and hydrogeological modelling had the interest of the government actors. The physical science, a cross-country study in Bangladesh and India on groundwater, turned out to be the main selling point in initiating the contacts with the formal government representatives. At the start of the project implementation, this support from different 290 state government officials also made it possible to get support from the District Magistrate in charge of the district in which the project village was located. Given the relatively hierarchical formal institutional setting and large power distance between District district and Statestate-level officials and local level stakeholders, this support was essential to undertake activities with government officials and stakeholders at the local block and village levels.
This created a supportive atmosphere, including state-level experts in the project advisory group, but the longer-term ownership 295 at the district and state levels for the Shifting Grounds project remained limited. Although the groundwater problems in the peri-urban areas were acknowledged as important issues, the project itself was too much focused on one specific local area, with relatively limited resources, to spark a more intensive involvement from the higher levels of administration.

The research -community interface 300
In parallel to the administrative system, there is a political system with elected representatives at various levels. At the community level, these are the panchayat members and the gram panchayat. These are local self-governing bodies, with are the village councils (, panchayats) , which arebeing the lowest elected official body bodies in rural areas in India, . Grand gram panchayats consisting of a number of village councils.
In the beginning, the project team had visited various peri-urban villages to select a suitable project site. In this selection 305 process, we looked for visible signs of groundwater management problems, for willingness of local stakeholders to work with researchers to address these issues, and for the presence of a certain nucleus of self-organization as sign of a certain level of competence within the village community that our project could build on. The peri-urban village that was eventually selected for this project was located alongside a canal of historic importance, south-east of Kolkata. It is part of the East Kolkata Wetlands, a Ramsar site. Recent developments included a growth in aquaculture, profitable with rising demand for fish in 310 Kolkata and its suburbs, as well as an increased reliance on groundwater for aquaculture and rice paddy fields.
The project team benefited from the existence of a receptive village leadership. Certain members from of the local panchayat shared their knowledge and support and actively participated in project activities from early on. Support from informal local community groups was present through a local youth club and various smaller women self-help groups, who were mobilized with the help of a local panchayat member. An initial informal community meeting was facilitated through the involvement of 315 a youth club, which was asked to bring people from different occupational groups to ensure diversity in participation.
Access to safe drinking water was a critical issue, identified at the first stages of engagement with the village community in 2015. The existence of a private water-bottling plant inside the village was a controversial issue. The bottling plant was set up on purchased village land and had a bore well installed as the source of bottled water supply. Given these investments and operations, the owner of the bottling plant was a locally powerful figure. In the first project community meetings we discovered 320 noticed two distinct interest groups, divided in a pro-and anti-bottling plant lobby. One group supported the activities at the bottling plant, sometimes because they would benefit from those, as water vendors or workers, while another group considered it an illegitimate appropriation of local groundwater resources in the village.
The local water bottling plant proved to be a very sensitive issue, as it was closely linked to the village power structures and politics. Even before any choices on problem framing were made, the ability to continue within the community was threatened 325 by the sensitivities over the bottling plant. Therefore, as more information on village problems emerged, the project continued with a more specific focus on what was not the most contentious, but the most crucial issue, shared by groups across the village: access to safe drinking water, free from arsenic risks. This choice was informed by village concerns, later on combined with and later on confirmed by groundwater research information. In later stages, providing visible contributions to help the villagers cope with the arsenic problems, helped us to build confidence with them villagers and their social and political leaders. 330 Gradually, the project team realized that the village was very much divided on political lines, a common feature of rural society in the state of West Bengal. The water bottling plant was one issue of contention, but not the only one. This put us in a difficult position. Already from the start, we realized the importance of remaining neutral as a project team, avoiding reliance on current political leaders who might represent one political faction only. At the same time, the village leadership and the officially elected local bodies could not be by-passed, in order not to compromise the participation process and the safety of its 335 participants. As a result, politics and associated legitimacy questions affected the further stages of the project.

Experiences within the Shifting Grounds project team
The researchers of SaciWATERs (Hyderabad) and TU Delft (the Netherlands) had easier access to the State state and District district level government officials than the local project organization, The Researcher, in Kolkata. SaciWATERs and TU Delft 340 were recognized as research institutes of national and (inter)national importance, which enabled them to access to the stakeholders experts and officials at these levels. The local partner in Kolkata, The Researcher, cultivated a good rapport with the local community representatives. At the same time, across the project team, there was a steep early learning curve on the mechanisms and particularities of the negotiated approach. Even if some guidelines were available, these were fairly generic, and their application in this the specific setting in West Bengal, brought its own challenges and questions. During the first two 345 larger project workshops in Kolkata, the presence of professor Paranjpye, one of the original developers of the negotiated approach for water management in India, proved essential to support the team in the early phases of process design for stakeholder empowerment.

Groundwater research and access to official data
The groundwater modelling, another a key research component, struggled with the acquisition of regional-level data for the Kolkata site, despite early efforts to establish good contacts and obtain the support fromwith key government officials for our Formatted: Font: Not Italic projectin the State Water Investigation Directorate. The groundwater researcher had to work with Aa very limited set of regional level data, combined with some local measurements during from a local field visit, . This constrained the groundwater 355 researcher in modelling and in-depth site-specific knowledge on the local groundwater specificssituation. Nevertheless, the groundwater knowledge that was available suggested that simply demanding more tube wells for local users might not be advisable, as it would lower the water table of that theparticular village.
The presence of arsenic is a known issue in the Gangetic delta regions in India and Bangladesh since the 1980s and 1990s. For India, estimates were that about 6.5 million people were affected by severe health risks, using groundwater from affected 360 aquifers for human consumption (Hasan, 2016). A review of groundwater data that were available, supported the focus on arsenic mapping and awareness, as likely risks also for this particular village. The water quality data that were obtained for the groundwater research indicated the presence of arsenic, which was validated by the Block Development Officer, Gram Panchayat, and Public Health Department Engineer.

Institutional research on formal institutions regulations and water rights
Formal institutions provide a key leverage point for sustaining future interventions and improvement in water management. For these national and state level policies, acts, and ordinances, an institutions brief was prepared by the institutions researcher to support the negotiated approach process. The brief was presented to the community in their own language, Bengali, printed as a brochure with many pictures and illustrations that made it attractive and helpful to understand, also for the illiterate 370 community members. This was useful in respectas a way of imparting knowledge to the community about people's rights to water and the official government acts and departments regulating water in the state. The community had never heard of such rights to water or water governing acts. Not all of this knowledge could be translated immediately into action, but the knowledge remained an element of awareness and empowerment on community water rights. Being aware of your their rights, and the official legal acts and ordinances that are recognized by government bureaucrats and administrators, helps communities 375 become more accepted as partner for dialogue.

Socio-economic research: Synchronizing longer-term research with short-term community needs
The initial idea was that household survey results would be used to prepare an integrated groundwater security index, based on household survey results, which could be shared with the community and wouldto help to prioritize issues to be tackled in 380 the negotiated approach. It would also have been useful to conduct aAn early survey, would also have helped to get a better picture of socio-economic heterogeneity and structures. However, a survey could not be started without initial community engagement and support. As this was initiated, the first community meetings already helped to prioritize local issues and suggested that a lotsome of the issues represented in the scientific groundwater security index might not be relevant locally.
Based on this, more questions towards water quality and water distribution could bewere included and wastewater irrigation 385 was added -something that was not there in the standard set-up for the index survey. Also, conducting the household survey gave the project team a better overview of the problems in the village, especially the differential access to water.

13
Peri-urban spaces are zones of transition and great socio-economic heterogeneity (Allen 2003;Singh and Narain 2020), where the socioeconomic dynamics change very rapidly with regard to status and income. In more remote rural areas it is easier to understand the status of the people as it is more stable. Also, cConducting the household survey gave the project team a better 390 overview of the problems in the village, especially the differential access to water. Here, tThe socioeconomic status and dynamics became clear only during the survey, when we visited the households more intensely for several months. The survey also gave us the idea that there wis as a sizable section of population using groundwater for irrigation. This was not raised in the first community meetings, where the village community had predominantly raised its drinking water problems.
The household survey results eventually were only available only well into the third year of the project. At this time, the 395 negotiated approach team had already started working on the particular issue of drinking water and arsenic. Still, the socioeconomic research did reinforce earlier choices in the process. We came to know that there were over 900 families in that the village and there werewith only ten available potable water sources. This reinforced the focus on drinking water.

Discontinuous participation due to village politics and power shifts 400
In the spring of 2016, State Assembly elections were held, resulting in political schisms reaching new heights between rival groups. The deep political divides meant that some community members who had earlier been in leadership positions and had been very supportive to our activities in the initial project stages, could no longer play a role in support of the negotiated approach process. These political dynamics meant that the project team had to make continued efforts by bringing the various lose threads together, roping in new persons, and assuaging the conflicting interests to the extent possible. After two of the 405 three initial village 'champions' left the stage due to the rising political tensions, we built rapport with the new leadership of the youth club. One re-elected panchayat member who had previously shown support provided a stable factor and enabled us to connect with the community in the subsequent phases of the process and to maintain contacts with the women's self-help groups. We attended several meetings of these self-help groups, urging the participating women to attend also our informal project community meetings. These efforts ensured a good participation of women in the subsequent meetings. 410

Transdisciplinary Research TDR Phase C: Re-integrating and applying produced knowledge in science and social practice Negotiated Approach NA Task 4: Solutions analysis and Task 5: Forging agreement
In a project mid-term review meeting in September 2016, community representatives signalled impatience and dissatisfaction 415 with project progress. Their feeling was that, until then, little direct benefits to the community were visible, endangering their willingness to continue the engagement. They requested the project team to do something concrete in the short-term, to gain confidence of the community and continue the process further. From a pure scientific research perspective, this was difficult to respond to. The research activities were nowhere near finalization and actionable results. Also, the three project researchers by then had differentiated between project sites to focus on in their research, based on access to data, progress in the research 420 and capacity building, and power dynamics. Two researchers were focusing relatively more on Khulna and one researcher was focusing relatively more on Kolkata.
As part of a reciprocal transdisciplinary process, the international project team promised to make an effort to mobilize additional resources to address the pressing issue of arsenic contamination of water sources. This was started in the months after the mid-term meeting and brought in new experts, doctors and equipment to enable visible actions focused on the arsenic 425 contamination of local domestic water sources. An A local arsenic awareness and mapping campaign was started, with arsenic testing of various local water sources and a village health camp. For this, national and local experts were engaged, including a local medical college and water laboratory. This helped to get more detailed information on the local prevalence of arsenic in various water sources, and through the local health camps and workshops, villagers could be checked for symptoms and received medical advice, as well as education about locally developed arsenic removal filters. 430

Negotiated Approach NA Task 6: Communication with constituencies and Task 7: Monitoring and continued involvement
Tackling the arsenic drinking water quality issue in the village was only possible with the consent of the panchayat officials.
After the earlier friction, the panchayat officials eventually recognized the importance of our activities, as they invited us to the local Book book Fair fair organized by three Gram Panchayats in early-2017, to make an audio-visual presentation on the 435 water security issue before a larger audience of several hundred people.
As drinking water had been the most crucial issue across the political divide, people belonging to both political sides were involved in the arsenic testing and education process in a more indirect and informal way. During the testing of water samples and the door to door campaign on water quality, political allegiance played no role, and people from the opposition camp were also involved. But iIn the formal process, i.e. in village meetings and workshops, these people from the political opposition 440 camp were not always involved. Even wWhen present in less formal community meetings, they were not so vocal due to fear of being identified.
In the second part of the process, we organized an arsenic health camp and an arsenic awareness workshop involving local health workers where arsenic-affected tube wells were marked in maps of all the villages under the panchayat, as part of an effort to address the specific concerns regarding arsenic contamination of domestic water supply. In addition to the direct 445 practical health benefits, these maps and the knowledge gained through these health camps and workshops, enabled the villagers to better discuss their needs and concerns to government representatives and panchayat bodies.
Monitoring the effects of project interventions and proposed solutions after the project timespan was not possible due to the lack of resources for the project team members to visit the village after the finalization of the project. Although unsatisfactory, it this was had been anticipated in the project design, whereby we have tried to be clear to all stakeholders, and careful 450 ourselves, about our exit from the village after project closure. Part of this exit strategy, for instance, was to steer away from the most controversial issues around the local drinking water bottling plant, in order not to stir up more conflict than we were capable of handling within the time and resources of the our project.

Negotiated Approach NA Task 8: Strengthening capacity to become and remain equal partners in negotiations
A key dimension in the equal partnerships emphasised in the negotiated approach, is gender equality. In the informal meetings 455 and the larger more formal community workshops, women were no less vocal than men. One of the key persons to mobilize the community for us was a local female panchayat member. That she was a woman probably helped the other women in the community to join our programme in good numbers, as well as to speak out. However, if this lady in our project village had not been re-elected again in the panchayat elections that were held later in the project period in May 2018, our effort to involve the women might have been thwarted. This shows that this effect, although visible, also is fragile. 460 The Government of India is giving much importance on to the panchayats and allocates several hundred crores of rupees for water supply. Among the formal institutions supposed to be in place at this level, is the Village Water and Sanitation Committee (VWSC), looking after the water and sanitation problems of the a gram panchayat area and formally chaired by the panchayat pradhan. However, in this the panchayat of our project village, the committee mostly remained on paper. So our aim was to make it function as the sustainability of the negotiated approach process was dependent on the functioning of this VWSC 465 village committee that works for the whole GPgram panchayat, consisting of seven villages. The panchayat pradhan (chairman) gave us permission and during the final project workshop, members of the committee participated and pledged to use the written project reports with the arsenic testing results forward, to improve the situation. It At the time of writing, it remains remained to be seen whether this committee will become andould remain truly active.

Negotiated Approach NA Task 1: Preparing the process
The village for project activities in Khulna was selected for project work based on pre-assessment and pre-scoping visits in the first project phase, reviewing different potential villages as project-sites, similar to the process done for Kolkata. The project activities were then initiated with a community workshop in October 2015. Following this workshop, several smaller group 475 meetings were held in the village to further establish dialogue. Through a series of village level meetings and workshops, people learned about the project and its negotiated approach, while the project team learned more about the village, its stakeholder groups and social dynamics.
Land use change is a common feature of peri-urban environments. This is accompanied by a rise in the price of land and efforts at occupational diversification (Narain 2009;Narain and Nischal 2007). This dynamic was also visible in the project village. 480 Traditional fish farming and agriculture were on the decline. Some people were selling their agricultural land to land developers and others to migrants.
During the first visits, it was observed that the village road acted as a rough division between the groups of migrants and permanent residents. The permanent residents were located mostly on the right side of the road, and appeared to be more homogenous, with less rivalling groups within them. The part of the village on the left side of the road had more migrants, 485 who were not as well organized. This made it easier to start the community engagement process mainly at the right side of the road. This, of course, had implications for village representation in the remainder of the project. Although this was known to be far from ideal, the project timeline and resources did not allow for complete community mobilization and organization, given that activation and organization of the migrant households would have taken significant additional efforts and resources.
Realizing these limitations, at later meetings residents, including migrants, from both sides of the village were included, such 490 as in the gaming workshops (see below section 6.3).

Negotiated Approach NA Task 2: Reaching agreement on process design
In the course of the first year of engagement, farmers and fishermen groups were formed to represent the community in the project's negotiated approach process. The traditional livelihoods of these groups in the community were under pressure, among others as a result of increased selling of land to land developers and migrants. 495 These village negotiation groups would bewere supported in a participatory problem analysis, for which the main steps had been outlined in a local Bengali guide for the negotiated approach. This guide had been developed by the local project partner NGO JJS, after the support received from international negotiated approach experts from the Netherlands and India (see Kolkata experiences). This The local project partner JJS also facilitated the implementation of these steps with the community, to prepare them for a purposeful dialogue with other stakeholders, including government actorsofficials. 500 Contact with government officials had been initiated from early on in the project. Although government resources seemed constrained, the rapport with government officials based in Khulna City, including the Khulna Development Authority and district agencies, was good and no real problems were foreseen for later stages. The good relations with these government officials in Khulna benefited from the existing relations of local NGO partner JJS, which was based in Khulna, and the nationwide reputation of the research partner BUET in Dhaka. 505

Community level activities
The village negotiation group conducted a problem analysis. This problem analysis used that benefits from the results of the survey results conducted in the village for the project and that it included a social map with water sources and water uses in 510 the villages, a stakeholder mapping, and an identification of several water-related problems. Three priority problems were identified: i.
accessible safe drinking water, ii. canal encroachment and water logging, iii. waste dumping by the city corporation. 515 These priority problems followed community needs and priorities, and thus were not all directly related to groundwater.
Nevertheless, all three issues were incorporated in the project, even if the research interface for some of these problems was weak or even absent. This was part of the implicit process agreement between project team and local community stakeholders.
For these three priority issues, some researchers contributed their expertise, and the local team helped the villagers to develop a small-scale management plan to address them. 520 Although migrants were not represented in the smaller village negotiation group, they were part of the research activities and were invited at some of the workshops. This suggested that the drinking water problem was also acutely felt by this section of the community. The group of migrants included both relatively wealthy and relatively poor households. Most migrants in the vulnerable category used one of the three shared tube wells in the village and they (as well as other poor personshouseholds) needed over one hour to collect water. This was especially a problem for the women, who were responsible for water collection. 525

Research contributions to situation and solution analysis
The groundwater researcher made several field visits to the village for primary data collection. During these field visits, awareness on groundwater issues was raised through discussions with village community members. When first results were available, information on groundwater quality and groundwater over-pumping fed into the village negotiated approach process, 530 among others via a lecture by the researcher on groundwater scenarios to the village water group. Further, researchers assisted with a Bengali translation of key groundwater terminology.
Community-based groundwater monitoring was considered during the project mid-term deliberations, as a way to combine village capacity building development with groundwater research data collection. Eventually, this was not initiated, mostly due to project timelines and research priorities -in which a PhD study was a key element, for which data collection results 535 would come too late -and time and resource constraints.
Research findings from the developed groundwater models indicated that local groundwater abstractions might not have a very large effect on local groundwater availability, which seemed more influenced by regional level forces tied to the river (Hasan et al., 2019). This provided a confirmation of the participatory management plans, reducing the need to focus on local waterdemand management issues for the short-term. 540 The development of a community-based participatory approach for institutional analysis was a core objective for the institutional research component. This approach was developed with the Kolkata and Khulna sites in mind. The steps in the approach were mostly explored and applied with the Khulna village community for the prioritized drinking water issue. During the earlier stages in the project, an institutions brief on water supply and groundwater management was prepared, translated and discussed with participants in the village. The brief outlined the different organizations, rights and responsibilities for 545 water resource management in Bangladesh. It also contained an infographic about the process for tube-well applications in peri-urban areas. This was accompanied by a de-briefing workshop with the village negotiation group, other community members and some government officials, where they reflected on these institutional structures. This supported the village group in its awareness of the situation, and the stakeholder mapping for the solutions planning. At the same time, the institutional research used local reports of the negotiated approach meetings as a source of data on the community's problem 550 perceptions.
Combined, these groundwater and institutional research efforts helped to deepen the knowledge of villagers about the groundwater management situation, in such a way that they were able to talk about this to the authorities. Their increased knowledge and their ability to use officially recognized terminology, empowered villagers in their communication with the government officials. 555

Transdisciplinary Research TDR Phase C: Re-integrating and applying produced knowledge in science and social practice Negotiated Approach NA Task 5: Forging agreement and Task 6: Communication with constituencies
For the direct engagement with the government officials, a specific six-the members of the communityvillage negotiation group was formed by the villagers for negotiation and advocacy with authorities for their water related problems. They were 560 trained by the local NGO (JJS) and at a local university in Khulna on advocacy and strategy development.
The community negotiation group shared their water related problems with the identified authorities and agencies during a workshop meeting. This workshop enabled the community negotiation group to continue discussions with the individual water related authorities after the meeting. During these individual follow-up meetings, there was more time and opportunity to discuss the specific problems and the authorities shared their plans and initiatives for overcoming those problems. Through 565 these follow-up meetings, all three priority problems were taken up by various government authorities. The public health agency in charge of rural water supply committed to test drillings to establishingfor a functioning deep tube well for drinking water in the village, in recognition of the declining water tables and the need for sufficient safe public drinking water supply points. The Khulna City Corporation cleaned the waste dump near the village and selected two new sites for landfilling elsewhere. The local level government administration (called upazilla) took the initiative to remove canal barriers. Linkages 570 with the Bangladesh Water Development Board and an ongoing internationally funded water management project resulted in an effort to further clean up the drainage canal.
The issue of canal encroachment and water logging was caused by clogged drainage canal structures but was exacerbated by local fish farming practices. Although fish farming was decreasing, a few powerful local elites did engage in fish cultivation.
Branches of the drainage canal were captured to put temporary bamboo structures to keepfor fish cultivation. However, these 575 bamboo fences and temporary dykes for fish cultivation reduced the water flow and exacerbated problems with drainage during heavy rains and water logging. The fish cultivators earned a lot of money and shared the benefits with local powerful individuals. This made it difficult for the local open-water fishermen and smaller farmers to deal with them. The village negotiation group first tried to involve these powerful canal encroachers in the project meetings, but they were not interested as they thought they would lose their livelihood. After these initial efforts, the focus was put on capacity strengthening of the 580 more marginalized groups, to help them to negotiate and improve their knowledge. Illegal activities, especially canal encroachment, were condemned by the government officials at the meeting and in later press coverage.

Applying produced knowledge on drinking water management and institutions
The institutional research followed a sequenced design for a participatory analysis process, aligned with the negotiated 585 approach activities in the Khulna village . In the final stage, this resulted in two gamingsimulation workshops, where the analytical results were shared and discussed with participants in a structured role-playing format. An important reason to opt for this format, rather than a formal report or presentation, was the low level of literacy in the peri-urban village community. In addition, gaming simulations are known to be effective means of communication, if well designed and facilitated. One workshop was held with the village community, and a second workshop was held with 590 government representatives from different agencies involved in drinking water and/or groundwater management at the local level in Khulna. The purpose of these workshops was for participants to explore strategies to address drinking water related problems experienced in peri-urban Khulna. The workshops provided a platform for research uptake where the results of the institutional analysis were shared with local stakeholders in the form of a role-play game. The workshops were valued by the community participants with suggestions for future uses to engage more groups (Gomes, Hermans, Islam et al., 2018;Gomes, 595 2019).

Negotiated Approach NA Task 7: Monitoring agreed actions and sustaining societal impacts
At the end of the project period, peri-urban water issues were being discussed at different levels of government, at universities and in the local media. A gaming-simulation seminar and workshop were organized at the local university, as well as more conventional workshops and meetings. A linkage between community and government stakeholders was had been developed. 600 A peri-urban water forum was established with representatives of several communities, (beyond the project village community only), related government authorities and civil society. This forum connected the Shifting Grounds project with similar projects and activities in other peri-urban villages around Khulna City. In this way, the peri-urban water forum could become sustainable.
A small spin-off project after the ending of Shifting Grounds continued work with the approach for participatory institutional 605 analysis, whereby local professionals were trained to develop gaming workshops for other water-related issues, with external support from JJS and Delft-based researchers. Although this enabled a bit more monitoring after the project end, the longerterm monitoring in Khulna suffered from similar limitations as that in Kolkata (see above).

NA Negotiated Approach Task 8: Strengthening capacity of participants to become equal partners in negotiations
Community empowerment for water management in Bangladesh carries a specific gender-challenge. As women were most 610 affected by drinking water problems, they were interested in participating. During initial field visits it was observed that, though women got a voice in village matters, the last word was always with the men. In the community negotiation group that spoke with the government officials, three of the six members were women. In the first workshop, only the men spoke and when we asked women to speak, the men did not allow them. Towards the end of the project, the women had no problem to 20 speak during workshops and meetings. They actively participated in the negotiation role-playing game and during the final 615 project workshop, the women eventually were discussing discussed directly with male senior government officials.
The success on empowerment, fairness and legitimacy was mixed in the project. Although efforts were made, both powerful and powerless groups were eventually excluded from some of the most intensive part of stakeholder participation activities in the village. For the group of migrants, with its large heterogeneity, this was mostly dictated by limited timelines and resources.
For some of the powerful local elites engaged in fish cultivation, their exclusion seemed a willing choice, possibly seeing the 620 process as a potential threat to their business activities. Within the group that was represented, the role of women seemed to grow over time.

Discussion of the Shifting Grounds project experiences with transdisciplinarity and empowerment
The project experiences described for the research and negotiated approach activities in peri-urban villages near Kolkata and Khulna, partly confirm the challenges known for transdisciplinary research trajectories. Project designs had to be continuously 625 adapted and changed, and, in some ways, had been over-ambitious. Project activities had to be tweaked to the site-specific conditions and constraints, and as a result, the activities across countries were not uniform, neither for the stakeholder empowerment, nor for the research components. The resulting process was very intensive and time-consuming, for all parties involved, much more than for comparable projects aimed at either primarily at research, or primarily at directpractical local water management interventions. Nevertheless, there also seem to have been synergies and added values, and, minimally,. the 630 The societal process with community and government stakeholders has shaped research activities and results, and in turn these research activities and results have influenced the societal dialogue within communities and between communities and government officials.
In addition to the confirmation of these prior experiences, the Shifting Grounds experiences also surfaced new challenges and responses, not previously emphasized in reviews for transdisciplinary research. These are summarized in Table 3. These 635 challenges and responses are specific to transdisciplinary research in situations where power and empowerment shape the process of co-creation of knowledge and solutions and building capacity totheir implement theseation. These responses are often context specific, they do not provide cure-alls and often they come with their own dilemmas and limitations. Some of the responses in fact are about accepting limitations or looking for satisfactory rather than optimal solutions. For instance, even if our stakeholder mapping captured the presence of a significant number of migrant households in one of the Khulna peri-640 urban villages early on, their heterogeneity and low level of organization, combined with our limited project resources, did not enable us to enable their effective representation in our project activities.
Challenges and responses similar to ours will be familiar to experts working on community or stakeholder empowerment projects, but so far, remained either invisible or fairly abstract for transdisciplinary research. In our project, we have adopted a negotiated approach to deal explicitly with power in transdisciplinary research, with some 650 practical lessons captured in Table 3. As we have explicitly engaged with the power-dimension, we have seen power structures and inequalities play out and affect our work. For instance: the contentious issue on the water bottling plant and the local level panchayat politics that proved more important than the State state level government administration in Kolkata, the role of fish farming by local elites and the low-level of organization of migrants in Khulna, and the role of women in both locations. The responses in Table 3 illustrate that these power issues could not necessarily be solved, even if they could 655 be observed. A full negotiated approach process will take more than 4 or 5a few years and may at times be more intensive than what we could help organizefacilitate in our transdisciplinary research project. Therefore, it is more accurate to talk about power and politics in transdisciplinary research in terms of empowerment, instead of in terms of solving power inequalities.
Empowerment is a dynamic process that may never fully end. We do have indications that we have been able to make a 660 fruitful contribution to this empowerment process in our project villages, through our explicit engagement with power. These indications include an increased visibility of the heterogeneity within the peri-urban communities, the recognition of different groups such as women, youth and migrants, and their increased participation in, and knowledge of, local groundwater management processes. At the same time, our longer term impacts remain unknown. This may be typical for transdisciplinary research, where research interventions are mixed with, and followed by, other activities that also influence 665 the problems at hand. Attributing longer-term impacts to specific projects or activities constitutes a research challenge of its own.

Conclusion
We have applied explicitly a negotiated approach in our transdisciplinary processwater research, recognizing to do justice to 670 the importance of power dimensions and empowerment, in addition to moreinstead of assuming a neutral co-learning experiences in transdisciplinary researchprocess. Overall, our experiences confirm that, at least when working with relatively vulnerable and underrepresented local communities, employing a negotiated approach is useful, if not critical. It forces researchers to pay much more attention to the social and political realities, and to community leadership and representation, early on in the process. Our experiences further confirm those earlier reports on transdisciplinarity that stress the importance 675 of early and ongoing joint problem formulation, the importance of flexibility, and the struggle to match longer-term ambitions with short-term needs of both researchers and societal stakeholders.

Formatted: Font: Italic
In addition to these insights, we also added a specific list of challenges and responses for transdisciplinary research that seeks to address actively engage with power and empowerment as part of its efforts. This includes the use of careful observation, stakeholder analysis and social surveys to map existing power structures, and the tuning of knowledge co-creation activities 680 to those power structures with an eye for feasibility and longer-term risks and consequences. Also important is flexibility, including the flexibility to accommodate needs of societal actors that may be outside the core domains pre-identified for research contributions. This list, which resulted from our project reflections, will help to build a better articulated set of principles and guidelines for future transdisciplinary water research.
An uneasy conversation that we further will need to engage with more, is to discussone about the limits of transdisciplinarity 685 and the various dilemmas it raises. Whereas many overviews may give the impression ofresult in an ever-expanding list of principles, tools and approaches for an ideal-type transdisciplinary process, the reality will beis served better by a perspective on transdisciplinarity as yet another craft and "art of the feasible" in which tradeoffs between multiple and sometimes conflicting objectives and perspectives need to be made.
An overarching message for transdisciplinary water researchers from this paper, is to engage with power and politics more 690 explicitly, as part of this process. This is critical from the (pre-)inception phase of activities, as a key input for problem structuring and research agenda-setting. Engaging with power and politics is difficult but fundamental to societal change. Even if some researchers will feel uneasy with this dimension, it cannot be ignored in transdisciplinary research. Ignoring itpower and empowerment, is just another way of dealing with it -allowing existing structures and forces of power and politics to coshape transdisciplinary results in an unobserved manner. Engaging with power and politics is difficult but fundamental to 695 societal change.
28 List of Tables: Table 1 Challenges and strategies in transdisciplinary water research (source: Lang et al., 2012;Steelman et al., 2015) Table 2. Activities for the negotiated approach (NA) in transdisciplinary research (TDR) (Source NA tasks: Koudstaal and Paranjpye, 2011) Table 3. Challenges and responses used in negotiated approach for transdisciplinary water research in peri-urban cases 855