A causal model for creating public value (in institutional voids)—the case of Lesvos

The Greek island of Lesvos is a holiday destination that became the main entry point for forced migrants to Europe during 2015 and 2016. These circumstances of disruptive societal change are used as the basis for developing a causal and dynamic model that creates public value in institutional voids. The clash between a holiday location and a humanitarian-ecological disaster and the fact that volunteers were the frontline response at the border of the EU in a politically complex situation made this, at the time, a very special case. The attention of celebrities and the media, Lesvos’ geographical proximity to Turkey, the historical conflicts between the two neighbouring countries, the lack of interaction between local people and the involvement of international NGOs spurred conflicts between various actors, causing social fragmentation on the island. The model identifies the core aspects of strengthening social capital and building trust to counter such societal fragmentation. Following Archer’s morphogenetic approach as a meta-concept, and the identification of causalities for understanding the social dynamics on Lesvos as a starting point, an extended version of Moore’s public value concept was used to build the empirical basis for the development of the model. The model presented can be used to distinguish between three very separate pathways of creating public value. I call this public value (PV) outcome creation. These pathways are described separately by using causal loop diagrams (CLDs). When aligned, they suggest a causal understanding of how adaptive leadership and collaborative governance can be a core element and a generative mechanism of social change that creates public value. Although the paper draws on a case study that is relevant to migration studies, it is not a migration study, but rather an attempt to use and improve “social-theory reflection” by applying a systems thinking approach as we embark on addressing public value creation, social problems and social-systems transformation (in institutional voids).


Introduction
In 2015, the year of Europe's refugee crisis, hundreds of thousands of people fled across the Mediterranean Sea, looking for a better future in Europe. The island of Lesvos, a Greek holiday destination, became one of the main entry points for forced migrants to Europe in 2015 and 2016 and the location of an unfolding drama.
The challenge of Lesvos and its prevailing institutional void, and often of complex problems in general-and the research question of this paper-is how to address the quandary of exploring and aligning people on what precise public values to pursue, as these are often conflicting.
Even though the number of forced migrants arriving is now far lower [84] than during the calamitous autumn and winter of 2015 and 2016, the absence of a good reception, speedy and fair processing of applications, and an orderly return back for those whose asylum applications were rejected led to desperation among the forced migrants who were stuck on the island. This meant that asylum seekers who were trapped in terrible conditions in overcrowded camps continually needed to share their limited space with newly arriving migrants. Close to 6000 people are being held at Moria (December 2017), a former military camp which was initially established as a temporary measure to house no more than 2000 people. Keeping people in shocking conditions such as those that exist in Moria for months and even years, and without any information, hope and occupation, leads to despair, which in turn leads to unrest in the camp, leading to increasing tensions. All this contributes to the perception in the international press of a "refugee island", and this destroys the islanders' opportunity to rebuild Lesvos as an attractive tourist destination, which is a source of income for many.
The circumstances of this disruptive societal change are used as a basis to develop a causal and dynamic model explaining the emergence of creating public value in institutional voids. The suggested model identifies the root causes that can counter societal fragmentation. The paper draws on a case study that is relevant to migration studies; however, it is not a migration study but rather an attempt to use and improve "social-theory reflection" by applying a systems thinking approach as we embark on addressing public value creation, social challenges and social-systems transformation (in institutional voids).
Often, these values can be conflicting. This is especially valid in institutional voids. In this paper, an institutional void is seen as the vacuum left by the absence of existing formal institutions to address a specific problem. It is an absence of the traditional sources of power [45]. These spaces can also, in the case of power shifts, be reconfigured and can enable new structures to evolve. The institutional void referred to in this paper relates to how the arrival and processing of forced migrants is handled. This subsequently had an effect on the whole island. Despite the existence of institutions which were responsible for the migrants, they were not doing enough. De Largy [23] has argued that the refusal to provide basic humanitarian support to refugees, thus producing the conditions that ensue, aims to deter potential migrants, hence preventing more people from entering Europe through Lesvos. If this is true, the situation on Lesvos is not a management failure, but a strategic choice.
To address this challenge, Moore's classical public value triangle is used as a framework to develop a causal model for creating public value. Although it is a disputed concept, it is an approach that is increasingly being used and discussed in the context of public management and policy. Public value theory was developed for properly functioning, democratically legitimised societies. Thus, when we apply its concepts to situations of institutional voids, extensions to the theory are necessary. Such extensions are suggested and discussed in this paper.
The application of CLDs (Causal Loop Diagrams) and systems thinking emphasises the important nature of explaining the underlying system through stories, explained on the basis of CLDs [76]. These narratives are vehicles for interpreting change, making attributions, justifying and explaining behaviour, and summarising complex scenarios [57]. They assist inferences and decision-making processes as part of the process of sense making, which influences how we act in the world [85]. By addressing the concept of public value with a systems thinking approach, a "social-theory reflection" is enabled. Systems theory [10] offers, according to De Haan [22], a conceptual lens to analyse and understand both societal and governance complexity, and to identify and theorise about the endogenous mechanisms in the processes of (transformative) change.
The purpose of this paper is to identify possible solutions to disruptive societal change in a situation of institutional void. This paper uses the situation of the migration surge on the island of Lesvos, Greece, in 2015 and 2016 with its vast implications for forced migrants and the host community as a stark example of disruptive societal change in a situation of institutional void. Responding to such disruptions, if adaptive leadership is either in place, or emerges, can cause social innovations to emerge under this crisis. The paper proposes an integration of relevant theoretical frameworks and suggests an extension to Moore's public value concept for realising the social dynamics necessary to deal with disruptive societal change in a situation of institutional void. The goal of the paper is to contribute to "social-theory reflection" in such cases.
The paper is structured as follows: in Chapter 2, the background of the theoretical concepts (e.g. public value theory, adaptive leadership, and morphogenesis) are described, while in Chapter 3, the methodological approach adopted is explained. Chapter 4 presents the results and analysis. The proposed model is a hypothesis of how social capital in a territory can be built and sustained, in order to deal with disruptive change in institutional voids. Chapter 5 provides the conclusion.

Public value theory
Public value theory was originally developed by Moore [54] to help public managers perform their many, and often conflicting, obligations [15]. As with many social theories, public value theory branched out into many different "streams", both normative, such as Bozeman's values account [12] and Meynhardt's non-normative approach [51,52]. Apart from being used in public administration and public management [1,8], it is approached from various viewpoints, such as ecology [77] and philosophy [57].
Public value theory has been criticised [21,65] for being too fuzzy, for reinforcing coercive powers, and for undermining democracy rather than empowering civil society. This paper takes no stand with regard to that criticism, but argues that an extension of the public value triangle is necessary and will, in the situation of an institutional void, make the role of civil society more explicit. The paper asks what concepts from Moore's original ideas can be used to build a causal model for creating public value in situations of institutional voids. The need to extend public value theory to a more complicated world, building a more general theory of public value creation, has also been raised by other authors [14,16].
According to Castells, "[t]he public sphere is the space of communication of ideas and projects that emerge from society and are addressed to the decision makers in the institutions of society" [17]. It is a shared space, but also a contentious place, where hierarchies of public values or constellations of competing values [40] are disputed.
The applicability of public value in complex situations has been examined by Geuijen et al. [29]. They examined whether public value theory would be useful in guiding analysis and action with respect to serious global issues like forced migration. The conclusion was that it is useful; however, no guidance was given as to how to address the "muddling through" to create public value.
Moore argues that in order to create public value, legitimacy and support for the action need to exist, as well as the operational capacity to deliver [54,56]. Legitimacy and support are sometimes called the "authorising environment". When both exist, public value can be created. However, Moore goes on to say that once public value has actually been created, it increases both legitimacy and support, as well as operating capacity, to create even more public value later on. In systems language, a virtuous cycle is established.
In the public value literature, this cycle has been called "Moore's strategic triangle". Moore starts his deliberations in situations where existing functioning democratic structures are a given and provide the framework for legitimate coercive and calculative power by the public manager, who has the crucial role of defining who should be engaged in the process of defining, as well as judging, the public value created.
In situations where these functioning democratic structures, in relation to the public value to be created, do not exist, the strategic triangle needs to be extended, because "reaching agreement" must be made explicit. A multi-actor perspective of public value creation [16], in partnership with other levels of governments and organisations, and with the active involvement of informal associations, community groups and individual citizens, is required. Hence, situations of institutional voids place civil society at the centre of debates about democracy, the public sphere and public value.
Public value exists at a time when the connection of citizens to government and public life is frayed. There is recognition among proponents and practitioners of deliberate democracy and civic engagement that the time is right for a new approach to governance issues, including questions of equity and democratic values. Despite the weaknesses of public value, Bryson et al. [15] believe that the new approach of governance will be the most robust and influential if it is infused appropriately with the language, the concepts and the methodologies of public value. Public value is gaining traction, but scholars developing the concept are still at the stage where they work with different approaches.
Benington [8] argues that neither the public nor the public value is given. In the case of Lesvos, the public value to be created suggests itself to be the humane treatment of refugees. However, the first dimension of public value (as an outcome) to be created is a choice, and it can change over time. At times, as was the case on Lesvos, in an emergency situation, the public value outcome needed is obvious because the situation is so dire and catastrophic that one does not need long and complicated deliberative processes to arrive at a concrete manifestation of the particular public value. There was a reality that needed to be dealt with immediately, which was due to the dramatic increase in the number of refugees arriving on the island. But there was also a second reality, one that was unfolding over a longer time period than the emergency in 2015/2016. It was the reality of a small island depending on tourism for its livelihood, as well as the unbearable conditions in the registration camp, Moria. Benington [8] argues from a public value perspective that the primary regulation of civil society is through trust and loyalty. On Lesvos that trust has been severely eroded, so the driving questions are "How can trust be restored" and "Who can initiate such a process?"

Adaptive leadership
An adaptive leader deals with the immediate needs of a crisis, but not at the expense of long-term objectives [35]. Hence, emergency situations, which are often technical problems, and adaptive challenges, which emerge over time, need to be addressed simultaneously. Thus, the key task for adaptive leadership is to identify the adaptive challenge and frame key questions and issues [34] for the longer-term wellbeing of the organisation, the region or the territory. The leader needs to pick up on the weak signals of the larger, longer-term issues. Heifetz et al. [34] call that "going on the balcony". He also urges for "voices of leadership from below", as he identifies that a key insight in tackling wicked problems, i.e. adaptive challenges, is that the most powerful momentum for change and subsequent solutions may well come from people in unexpected places, independently of their position in society.
The second task of adaptive leadership is to prioritise the weak signals into those that matter less and those that need attention. This is the ability to recognise the complexity of the crisis and to diagnose it appropriately. Heifetz et al. [34] claim that the skill of diagnosing is the most important, but also the "undervalued capacity" of adaptive leadership. The challenge is to maintain a sustained focus on the adaptive challenges which the group itself has to face, rather than allow attention to be drawn to more comfortable but less important issues.
Finally, adaptive leaders need to make sure enough resources are spent on addressing the serious problem. In relation to organisational crises, Snowden and Boone [72] state: "The minute you encounter a crisis, appoint a reliable manager or crisis management team to resolve the issue. At the same time, pick out a separate team and focus its members on the opportunities for doing things differently. " A key aspect when it comes to jointly finding new solutions that are able to solve complex problems is the concept of social capital. The concept is used in a variety of disciplines that tend to stress the different meanings and contexts of social capital. But a common denominator is that they all emphasise the fact that social capital can play a decisive role in the development process of a society [26,42]. There are different ways to detect, measure and analyse social capital, but many researchers have used participation in networks and voluntary groups as proxies (references), while others find trust in its core [25,41,81].
Putnam [64] calls social capital the "currency" of civil society, enabling a society to function: "The glue that binds members of neighbourhood associations, villages, community groups, civic organizations, churches, clubs and other kinds of voluntary activity is the relationships of reciprocity, trust and loyalty together; it is the 'social capital' . " However, there is a distinction between bonding, bridging and linking social capital [38]. The orchestration of these different levels requires an adaptive leadership approach.

Morphogenesis: the duality of agency and structure
In situations of social-systems transformation, or in times of radical change, we need to reach (some level of ) agreement on, generally, how we want to live, what a good life is, and what values, societal narratives, frames and structures we want to use to guide our lives. Having found answers to the above, we can develop a momentum for agreeing on what public value we want to create, for whom, with what resources, at what cost, and what the time frame will be.
These questions are usually answered through an interplay between individuals and the society that they form part of, and its norms. Charles Taylor calls this interplay the moral map of transformation [80], in which each individual must gain orientation for his or her life and actions. In this space, a person finds his or her way around "with the help of a culturally constituted, partly linguistically, partly implicitly in social practices 'moral' map" ( [68] [2016]). This topography of self describes what justice means, what is important and unimportant, what is good and bad, and what comprises a good life-questions which date far back to Plato's and Aristotle's Virtue ethics and the 5th century BCE, and all of which are value-laden questions. Answered on an individual basis, these notions need to be shared (and lived) in the public sphere. Much has been theorised about where the primacy of agency in this interplay lies: with the individual or with society, its norms, cultures and its institutions.
Archer [5] suggested a resolution of the debate about the primacy of agency by arguing that the question of primacy is wrong, since both the individual and society influence each other, separated by time. At any point in time, individuals are free to create change within the society in which they find themselves. But by doing so, society, its norms and its institutions change, creating a different framework for individuals to instigate change the next time around. Archer's name for this process is morphogenesis. Her perspective offers the possibility of identifying intervention points, and of intentionally making changes in the real world-whether in the personal, natural or societal spheres. It enables relational social subjects to jointly create public value [24] and align powers. It is an empowering model, where people have emergent powers to act and to change constraining or enabling structures, also in situations of institutional voids.

Systems theory
Systems theory [10] offers, according to De Haan [22], a conceptual lens to analyse and understand both societal and governance complexity, and to identify and theorise about the endogenous mechanisms in the processes of (transformative) change. Uncertainties, nonlinear processes of change and innovation, and emergence are important features of societal change and can be readily conceptualised, presented and understood with the tools of system theory [20,61,76].
Bryson et al. [16] suggested a whole-systems approach for understanding and analysing serious problems, stressing the fact that absent these kinds of understandings, the chances are very high that problematic situations will be misunderstood and that public concerns will be poorly addressed or made worse [69,71].
A system is an interconnected set of elements that is coherently organised in a way that achieves something [49]. It consists of three kinds of things: elements, interconnections, and a function or purpose. A well-functioning system, according to Meadows [49], has three characteristics: resilience, self-organisation, and (over time) hierarchy. In a situation of institutional voids, the system can, in the case of power shifts, reconfigure. This capacity to make its own structures more complex is called self-organisation. Self-organisation produces heterogeneity and unpredictability and is likely to develop whole new structures and whole new ways of doing things, which can be threatening to established power structures. It is a space for experimentation and innovation. However, sustaining such self-organising systems needs stable intermediate forms, which, according to Meadows [49], will be naturally hierarchical. However, a system hierarchy does not necessarily imply a classical top-down "leadership"; what it needs is alignment. As already mentioned, is through an adaptive leadership approach, and the second dimension of public value as a means/process, enabling conflicting debates where the complexity of the situation needs to be addressed and a sense of direction needs to evolve.
In a system, there are no independent or dependent variables; all variables are at times independent and, at other times, dependent. This echoes Archer's process of morphogenesis. Systems theory involves considerations of how the problem evolved in the past to manifest itself in the present, where one might go from here, and how its development can be influenced. Cederquist [18] provides a thorough description, based on interviews in the local community, of how the situation in relation to the forced migrants arriving on Lesvos in 2015 and 2016 developed over time. By looking at the behaviour over time, the underlying dynamics are hypothesised in the model (Fig. 3a-e). This approach builds on and is presented through the concepts of systems thinking and translated into the visual language of systems dynamics [49,66,67,74,76].

Methodological approach
The scientific enquiry in this paper uses abductive and/ or retroductive reasoning, which reinterprets empirical observations in the context of more general ideas presented as models. Those observations are made by "taking a set of empirical observations and proposing hypothetical mechanisms that, if [they] existed, would generate or cause those observations" [53].

Systems thinking and causal loops
In this paper, a conceptual model based on qualitative data is presented. The model will visually display perceived patterns or relationships based on the information collected through the interviews, synthesised by the author, and informed by theoretical aspects of public value creation, agency and social change. The boundary of the system is the geographical area of Lesvos with the people on the island. The models presented are qualitative and there is no goal of simulating them at this stage.
Before describing how the models are built, a short introduction is given to some critical elements of the system dynamic model. By using a causal concept of time (morphogenesis) embedded in causal loops and stories of social change, I hope to make a small contribution to Howaldt and Schwarz's [37] efforts to develop what they call "middle-ranged models" of social change. In its current form, the model can be used to inform various discussions. To show how this can be done with the model, five examples are described as causal stories through causal loop diagrams (CLDs) [76].
CLDs connect variables through arrows. Each arrow represents a causal influence: the variable at the tail of an arrow causally influences the variable at the head of an arrow [43,67]. The direction of influence is indicated by a plus or minus sign at the head of the arrow. A plus sign means a causal influence in the same direction: if the variable at the tail of the arrow increases, then the variable at the head of the arrow will also increase. Similarly, a decrease at the tail causes a decrease at the head. A minus sign means a causal influence in the opposite direction: if the variable at the tail of the arrow increases, then the variable at the head of the arrow will decrease. Similarly, a decrease at the tail causes an increase at the head. To avoid cluttering a diagram, if there is no sign at the head of the arrow, then a plus sign is assumed.
Although not explicitly indicated, but mentioned in the text (Fig. 3a-e), is the distinction between flows and stocks. Stocks possess four characteristics that are crucial in determining the dynamic behaviour of systems. Stocks (1) have memory, (2) change the time shape of flows, (3) decouple flows, and (4) create delays. If one were to take a snapshot of a system at an instant in time, it is the stocks one would see. Hence, when building a model, it is crucial to identify the stocks in the system. Stocks are often identified as being nouns. They do not disappear if time is (hypothetically) stopped (i.e. if a snapshot were taken of the system), as opposed to flows that do disappear if time is (hypothetically) stopped. This is because all flows have the unit of something per time. The arrival rate of refugees on Lesvos is a flow, and its units are people per time (which could be years, months or days). Forced migrants trapped on the island is a stock. At any one time, one can count the number of asylum seekers trapped and report that number. Stocks send out signals (information about the state of the system) to the rest of the system. The flows represent actions or activities that fill or drain the stocks (accumulations) over time.
Example of flows in my model are building or destroying trust, courageous conversations, creating public value outcomes, acting to solve a problem etc., all of which are activities that need to time to unfold. It is important to keep in mind that the choice of parameters and their categorisation, i.e. the information which in this model is identified as a stock, can in another model be identified as a flow. This choice is made by the model builder(s) and is dependent on the model focus and boundary, the research question, and the time horizon that the researcher wants to focus on.

Data sources
Going beyond verbal description or statistical correlations, the usual tools of social theory building, system dynamics builds models to map hypothesized causal connections needed to address a particular problem. Model building may rely on an in-depth review of existing empirical and theoretical literature, the collection of new qualitative or quantitative data, secondary data analyses, and on the experiences and opinions of people who are close to the process of interest.
For this model, the data was gathered through twenty semi-structured interviews, conducted in the summer of 2016 and spring/summer 2017, with individuals from civil society (both Greek citizens and foreigners) on the island of Lesvos. The interviewees included people from international humanitarian agencies (20%), from informal, self-organised local initiatives (45%), some of these initiatives professionalised over time. The last category was individual volunteers and local citizens (90%). Percentages add up to more than 100, since any one individual could be part of more than one group. All interviewees were residents on Lesvos. The interviewees were selected partly based on my knowledge of them and their roles due to previous visits to the island; others were identified through their presence in the media and through theoretical sampling [31]. As new insights emerged during the research process, additional interviewees with new perspectives were identified. Where permission was granted by the interviewee, interviews were recorded, transcribed, and then analysed.
Together with my own experiences from previous visits to the island, an on-site visit in August/September 2015 and subsequent visits, I had access to informal interview situations and acceptance by local communities in order to obtain information and data. This personal experiences also gave me the opportunity to acquire what Polanyi [63] calls "tacit knowledge". McAdam et al. [48] described tacit knowledge as "knowledge-in-practice developed from direct experience and action; highly pragmatic and situation specific; subconsciously understood and applied; difficult to articulate; usually shared through interactive conversation and shared experience. " Additional information was gathered through desk research. The situation on Lesvos was followed in the media-both the mainstream media and various informal and formal social media outlets.
Since systems models are also always causal stories [76], anonymised excerpts from the interviews are used in the results section to support and enrich the hypothesised causal model.

Results and analysis
The setting The setting for the study was the island of Lesvos, and the specific areas of interest were in the northern part of the island (Molyvos, Eftalu and Skala Sikamineas) which, during the peak of the crises in 2015 and 2016, were the main places of arrival for refugees and forced migrants and, at the same time, are the areas on the island most dependent on tourism. The complexity of the situation grew over time as the situation not only affected the forced migrants stuck in limbo on the island, but a secondary crisis evolved for the tourism-dependent local communities. From an emergency in 2015, the situation has grown steadily into a long-term problem, where distrust and conflict within civil society on the island is proliferating, undermining the resilience for coping in an unknown future.
Lesvos has a rich history, and its geographical location has made it a target for the powers since early antiquity. It also has a long refugee history. So, while the scale of the crisis was new, it was not unusual for migrants to be arriving on the island. The 1923 agreement between Greece and Turkey forcibly relocated Muslim Greeks to Turkey and Christian Orthodox Turks to Greece (Fig. 1).
Half the population of this island are refugees. They came from Minor Asia so they know how it is to be the refugee. In our DNA is refugee. Interviewee #2, August 2016

Developing the model Making civil society visible in the public value triangle
What is the role of civil society? This is one of the core questions in the public value context, when applying the framework for promoting the potential of networked governance [75]. Such an approach reflects the idea of a shift in the locus of control of governance away from the state and towards civil society. However, a networked governance implies that there is (a) governance and (b) a structured network. This resonates with the original public value concept which assumed orderly, usually democratic societies, where the need for an agreement about public value outcome is fairly clear, or is reached through existing and functioning social structures [54].
On Lesvos, in 2015 and 2016, in the midst of the refugee crisis, there was neither alignment nor agreement; on the contrary, there was conflict and institutional void.
There was a complete void. The government did not step in to do anything. It was private initiatives and the NGOs who needed to step in to do something. Interviewee # 10, August 2016 The challenge, thus, is to align civil society and bring forth an agreement with respect to the crisis reflecting the diversity of public life, including all those affected. The agreement is not that everyone should be alike and think the same-quite the opposite. Hence it would be more appropriate to call it "the (struggle for) agreement". A recurring challenge is the role of civil society legitimacy [13]. According to Ulrich [82], "[i]n a civil society, the ultimate source of legitimacy lies with the citizen, hence a reflective professional practice that is grounded in an adequate concept of civil society should give citizens a meaningful and competent role to play. Reflective practice then depends on the component citizenship. " This reflective practice takes into account "the interdependence of boundary judgements, observations, and evaluations. The facts we observe, and the way we evaluate them, depend on how we bound the system of concern. Different value judgements can make us change boundary judgements, which in turn makes the facts look different. " The "creation of a well-informed 'public' with the consciousness and the capability to engage actively in democratic dialogues" is at the core of co-creation and public value [7].
Knowledge of new facts can equally make us change boundary judgements, which in turn makes previous evaluations look different [82]. Hence, an effective response has to be grounded in an understanding of the root causes of the problem. Therefore, system thinking and the implicit CLDs provide a structure in which hypothesised causal connections in the system can be discussed. For tackling complex problems, deliberation for alignment is key to identifying joint actions which can lead to solutions fit for purpose. Often joint actions and experiences are the first step in the dynamic process of alignment.

The need for an inclusive agreement
In a Causal Loop Diagram (CLD), several variables and arrows can form closed loops. Figure 2 is a very highlevel model of how to create public value outcomes. To make the agreement (finding) explicit in Moore's strategic triangle, a fourth variable to Moore's three-legged strategic triangle, the public value agreement (PVA) (4) is added, and connected in a causally different way. Public value outcome (PVO) (3) no longer directly impacts the authorising environment (AE) (1), but goes through public value agreement (4) (see Fig. 2). Outcome is strengthened if operational capacity (OE) (2) increases through the authorising environment and a public value agreement is strengthened, all else being equal. For example, in Fig. 2, strengthening PVA creates more public value outcomes (i.e. solutions to a problem), which, in turn, strengthens the agreement. A process that feeds on itself-colloquially known as a virtuous cycle-is established, which amplifies over time an initial change in any of the variables that make up the loop. Note that this closed loop can also become a vicious cycle, namely, when the initial change in any variable is a decrease, or is weakening. The "more" PVO, the "stronger" the PVA. The aggregated PVO is composed by a number of PVO, which need to be collaboratively agreed upon. However, if the PVO favours one segment of a society unreasonably more than another, the polarity flips: the more PVO (for one segment), the less PVA. Conflict ensues. For example, on Lesvos the humane treatment of refugees (PVO-1) came, quickly und due to the enormity of the task, into conflict with the well-being of an island dependent on tourism (PVO-2). Some would even argue PVO-1 came at the expense of PVO-2. In any case, a PVA must be created that accommodates both PVO-1 and PVO-2. Once such a more inclusive PVA exists, AE and OC are similarly extended to serve the needs of all relevant segments of society, delivering both PVO-1 and PVO-2, making the entire causal loop once again unambiguously self-reinforcing.
Therefore, it is important to note that "the agreement" must be an inclusive one, i.e. bridging different groups. Only then can the agreement strengthen both the authorising environment and the public value outcome. If it is not inclusive, i.e. if the agreement is only accepted by part of those affected by the problem, it can become divisive and a source of conflict. Imagine, for example, that an agreement about how to deal with a refugee crisis was made only with the forced migrants without considering the local population-or made only with the local population without considering the forced migrants, or in a sub-group of either, or the decision was made externally, without either. Then the likelihood of a lasting solution for all would be severely compromised.
The public value agreement (PVA) needed is not based on coercive powers, but on a joint transformative experience. They are conflictual spaces, within which people and organisations with competing and conflicting interests can use "agonistic" pluralistic negotiation [58] to create coalitions with a common purpose. However, over time, an alignment for purpose is essential for producing physical public value outcomes that can address the problems. It is not based on a search for the right answer or the best solution, but on a process of agreeing on solutions that address the problems. Those solutions then deliver public value outcomes (PVOs). These processes are context-and time-specific.
Over time initiatives to coordinate the humanitarian response on the island through e.g. weekly stakeholder meetings were efforts based on more traditional aid-oriented leadership models, encompassing neither the local community nor the refugees. These meetings, where some of the NGOs participated as well as the municipality, were not only far more limited in scope, but also their initiation, climate and purpose differed. Neither was the UNHCR program on Lesvos called "communicating with communities", perceived as forum for dialogue. The extended strategic triangle-a hypothesis for the emergence of public value outcomes in an institutional void The model presented in Fig. 3a can be used to distinguish between three very separate pathways of creating public value outcomes. The model is then described in sections and loops (Fig. 3b-e) to arrive at a causal understanding of how adaptive leadership can be a core element and a generative mechanism of social change that creates public value in institutional voids. This is based on the situation on Lesvos identified during 2015 and 2017. Public value is being created when a public problem is being solved. The overall problem that needs to be solved is the development of the dire situation on the island. In the CLD, it is not specifically indicated, but the dire situation is a flow. It feeds into a stock-long-term problems. Stocks accumulate flows over time. Problematic as the dreadful situation is at any time, if not solved, it develops into a long-term problem. The situation is dire, and becoming more so (positive connections are drawn as solid lines with no sign at the head of the arrow) because of the unprecedented numbers of arrivals of forced migrants in a short period of time, as a result of Lesvos' physical proximity to Turkey. Lesvos is an island that depends on tourism. The situation is exacerbated by the deteriorating living conditions in the refugee camps and the rising levels of conflict between various groups [18]. Figure 3b reflects that the only way in which to reduce long-term problems is to solve them, i.e. this is public value outcome creation. This is a flow, leading away from the stock. The better society is at creating and delivering solutions fit for purpose (i.e. public value outcome), the smaller long-term problems become, until, ideally, they disappear.

Unaligned efforts trying to solve long-term problems
Thus, the better the operating capacity, the better problem solving i.e. more public value outcome creation occurs. Operating capacity is fed by solutions fit for purpose and by the authorising environment. The latter means that in a given situation it is easier (more effective) to operate when the operations are supported by an inclusive and accepted authorising environment.
At this point, one could follow the causation backwards in many different directions. First, this could be done when the following solutions fit for purpose back to acting to solve a problem. How do solutions emerge? One way is through trial and error; in other words, by people simply getting up and doing something. This can be triggered both in the short term, when people respond quickly to the dire situation, and in the long term, when people address the long-term problems. The solutions offered may be haphazard and uncoordinated, but some will actually help to solve the problem: a closed loop has been established. Furthermore, a balancing loop is created: triggered by the dire situation as well as the mounting of long-term problems, people spontaneously act to solve problems, thus finding solutions fit for purpose, increasing and improving operating capacity, and creating public value outcomes, thus reducing long-term problems. This is a process that takes place over time. At any time, people act within the situation as it is, thereby changing the situation to a new state, which then becomes the framework in which new actions are possible. And so on. This is a concrete example of Archer's morphogenesis [4,5].
Without the guiding framework of a (collaborative) agreement in the larger system (i.e. without bridging social capital) and a legitimate authorising environment (i.e. in a void), solutions fit for purpose emerge by chance, since individual actors will implement their particular ideas and experiences, which may or may not be appropriate for the situation at hand. If solutions require resources, as they usually do, actions within a void become inefficient and ineffective, at best, and lead to conflicts over resources, at worst.

Increasing the likelihood of finding solutions fit for purpose
Solving problems through doing When looking at Fig. 3c and following the CLD backwards from solutions fit for purpose to the public value agreement, it is indicated that public value agreements can be reached through courageous conversations and/or by "generating agreement by doing (the right thing)". Doing the right thing means solving the problem. The latter is the easier one. If actions taken (through the operating capacity) actually do solve (some of ) the long-term problems, others will take note and do the same: this can be seen as learning by mimicry [79], where ideas are replicated.
In 2017, the Starfish Foundation initiated NeedsHub, a platform for aligning the many organisations working on the island. These were all organisations with a similar mindset, and the initiative was an example of a social innovation as a product, creating an opportunity to support refugees on Lesvos in a streamlined, efficient manner, and ensuring that the most urgent needs were addressed.

Then what will happen now is that then, slowly, other donors who want to donate things can go on there [NeesdHub] and they can see who's working where, and they can research them out a little bit now. Interviewee # 17, May 2017
This alignment by NeedsHub is creating bonding social capital. It refers to "trusting and co-operative relations between members of a network who see themselves as being similar, in terms of their shared social identity" [78]. A self-organised subsystem was developed over time, due to the prevailing institutional void in 2015, where individual persons and civil society groups, where the pre-dominant actors in trying to solve the problem. Hence, over time, self-organisation took place and a structure developed for a specific sub-system. This alignment within a bonding network relates to the notion of Westley [87] that scaling out new solutions, is about engaging more people; it can also cover a larger geographical area, meaning that the international arena of like-minded persons are actors in this process.
Although this process takes time, as indicated in Fig. 3b by the double lines of the arrow connecting public value outcome creation with generating agreement by doing. It is a collaborative process that in comparison to the diffusion explained in Fig. 3d is fairly prompt. If a shared purpose (within the group) is already implicit, then this route of scaling out, as in the example of NeedsHub, is fast. But joint activities and "doing" are also an opportunity for changing relationships and creating bridging social capital, in the dynamic process of enabling a public value agreement. This causal perspective also picks up Weick's [86] argument that people find a way forward because they "begin to act, they generate tangible outcomes in some context, and this helps them discover what is occurring, what needs to be done and what should be done next" [86].
A self-reinforcing closed loop has been established: solutions fit for purpose strengthen operating capacity, thus creating public value outcomes that solve the (longterm) problems and are, over time, picked up by others as the way forward, generating agreement on what works and what does not, and crowding out "solutions" that are not fit for purpose. However, this generates agreement only if the (long-term) problems are solved. Otherwise, it will further erode trust.
There was a point they were up to 80 [76] calls "fixes that backfire", because they are solutions which are well-intended and might solve a problem in the short term (somewhere in a system), but later on there are unforeseen consequences of the solution that either make the original problem worse, or lead to other problems, particularly by undermining trust. In this situation creating PVO-1, can be perceived as creating public value for the forced migrants while undermining the creation of PVO-2 for the local community.
Because there are some organisations that they started handing out food, clothes and heaters to the locals. Who are they to decide who needs it and what about those of us who never got considered? … I want to believe that when all these things happen they do it because they want to do something good. It's with a good heart, but they're missing the target and they are creating, how do you say, it's like they are adding fire to the splitting of people.

Interviewee # 7, August 2016
Also external events like the well-intended nomination for the Nobel Peace Prize have has not been a solution to the island, but instead further undermined a joint PVA.
The distrust which happened between people, even the distrust that happened because of the Nobel Peace Prize nomination. As soon as that happened it became the "well why not me or why her and There was no connection to the locals. It was the ones they were coming with attitude of salvation of the locals and rescuing them, local society and refugees or only refugees and there was strange feeling. Also, it was the first time that we had so many NGOs. It was so many, more than 150, I don't know, 200 and independent volunteers. There was lack of co-ordination. It was like being in a third world country.

Interviewee # 3, September 2016
In the hypothesised model reaching "the agreement" implies that many different groups are involved negotiating and compromising, so that the emerging authorising environment represents a form of bridging social capital. As mentioned earlier, already in its informal incarnation, the authorising environment supports any operating capacity deployed to create public value outcomes, i.e. solving long-term problems.
Over time initiatives to coordinate the humanitarian response on the island through e.g. weekly stakeholder meetings were efforts based on more traditional aid-oriented leadership models, encompassing neither the local community nor the refugees. These meetings, where some of the NGOs participated as well as the municipality, were not only far more limited in scope, but also their initiation, climate and purpose differed. Neither was the UNHCR program on Lesvos called "communicating with communities", perceived as forum for dialogue.

That's one of the things that it's basically started last year trying to do things for the community as acts of appreciation. This has very frustrated because it's a program and this is like always, I'm very sceptical of programs because what I see is that people understand very quickly that yes communicating with communities is good and right but it's not actually getting different peoples' perspectives and listening to each other. Interviewee # 16, September 2017
Over time-another example of a morphogenetic process-authorising environments become formalised and at that stage become effective enablers for the search for solutions fit for purpose. For example, on Lesvos, antihuman trafficking laws prevented taxis and commercial buses from transporting the newly arrived refugees. It was also forbidden for private individuals to drive asylum seekers.
How do you deal with it by yourself? So, you do what you think is right. I don't know how many laws we broke. I know for sure by just transporting them [the forced migrants] we broke a big law, but we were by ourselves. What are we going to do? Because supposedly it's an illegal immigrant, that you're not allowed to transport them because then you could be accused for trafficking. It's to go against trafficking, basically. I don't know when this law was from, but I'm saying that it's old laws that are binding us today, which means change them.

Interviewee # 7, August 2016
Greece then altered the law, so people helping refugees by giving them a lift in their cars were not criminalised.
The fact that an authorising environment, once it exists, can lose its legitimacy over time, and what it needs to do in order to re-establish that legitimacy has been highlighted again and again in the problem-driven iterative adaptation (PDIA) approach [2]. According to the PDIA approach, the agreement is not fixed for all times, and once people have made a commitment to the espoused purpose and identified leverage points, they need to reassess the extent to which current goals, metrics, incentives, institutions and funding structures support or undermine the achievement of that purpose. Hence, this continual revisiting is needed to ensure that the structures and institutions are still needed. It is important for ensuring that institutions created by the authorising environment are still fit for purpose and also support the structures needed for the operational capacity. In this situation creating the PVO was aligning both the needs of PVO-1 and 2, creating public value for the forced migrants without hampering the PV for the local communities, improving the situation for both. Another example for a PDIA discussion is the general institution of migration camps. "They [the camps] were probably an innovation when they were introduced, some decades ago. At that time, they were indeed a solution. But I am afraid that later, when they became the indisputable solution, they became in reality part of the problem. Unfortunately, instead of a new, the involved and relevant authorities and organizations keep being focused on a model that produces modern concentration camps and arenas where a number of human rights are violated, along with human dignity" [73]. Still "the internment camp", whether in the safe areas policy or detention centres, "has become the routine solution for the problem of domicile of the 'displaced persons'", as Arendt [6] observed 50 years ago. Figure 3e explores how having a public value agreement that was reached through a broader collaborative process, in which boundary judgements have jointly been explored, creates coalitions with a common purpose that increase the likelihood of creating solutions that are fit for purpose. These solutions further enhance the operating capacity (manpower, knowledge and practical public value outputs) to fulfil its ultimate purpose, namely, to solve the long-term problems in acts of public value creation. Besides strengthening the original agreement through a flow called generating agreement in doing, solving long-term problems also initiates a new round of adaptive leadership [36] subsumed in the need to build trust.

Solving problems through collaborative governance
The causal links of an adaptive leadership approach are indicated in Fig. 3e through bold arrows or the rest being greyed out and focus on leadership as a reflective, social learning [39] process. This is because solutions that are being developed through co-creation, and in a process where there has been alignment on the purpose across a wide array of different people, are most likely solutions fit for purpose, and most likely to solve the long-term problems. Solutions fit for purpose fulfil the notion of a social innovations in the definition of Nicholls and Ziegler [60]: "The development and delivery of new ideas and solutions (products, services, models, modes of provision, processes) at different socio-structural levels that intentionally seek to change power relations and improve human capabilities, as well as the processes via which these solutions are carried out. " Westley [87] characterise the dynamics and pathways of scaling social innovations, whereby "scaling up" aims to affect everybody who is in need of the solution they offer, or aims to address the broader institutional or systemic roots of a problem. This requires the creation of bridging social capital. It refers to relations of exchange, respect and mutuality between people who see themselves as unalike. In Fig. 3e, the bridging capital which is being created through the courageous conversations is not being explicitly indicated. However, it is key in the long-term process of identifying and assuring solutions fit for purpose. Such a process asks deep questions: Where are we going with the agreement? Who gains and who loses, and by which mechanisms of power? Is this development desirable? [28]. It serves the purpose and reflection and changing boundary judgements, i.e. gaining a broader and deeper understanding of the system. Knowledge of new facts can equally make us change boundary judgements, which in turn makes previous evaluations look different [82]. A core aspect is also to understand the impacts of one's activities and the effects that they have.
The advantage of building a shared purpose is that it makes collaboration and developing solutions fit for purpose easier. It works independently of supporting frameworks like the state, institutions, marketing or charismatic leaders. Thus, this type of development of new solutions is sustainable because it draws its strength from itself; in systems language, it is a self-reinforcing loop. But this is also the potential downside. Unless people in this loop actively seek outside advice, information and viewpoints, there is the real danger of becoming ideologically self-sufficient, with all the problems that accompany this. Hence, it is always important to recall Flyvberg's [27] plea for Phronesis-that only a holistic evaluation of the situation can tell us what is right to do.
Rather than asking "what works?", one can also reach an agreement by asking "what is the 'right thing' to do?" As we shall see, this route is longer, but once established, it is the true source of transformational change. The right thing relates to the outcome of a reflective practice, taking into account "the interdependence of boundary judgements, observations, and evaluations" [82].
Answering the question "what is the right thing to do?" requires courageous conversations because the process tests, questions and extends established (power) structures, habits and behaviours. Engaging in courageous conversations requires trust, which must first be built Trust is part of a self-reinforcing causal loop: the more you trust, the easier it becomes to build further or deeper trust. Note, however, that self-reinforcing loops are also the drivers to destruction: the less you trust, the harder it becomes to trust at all-a downward spiral.
In theory, we know what to do: dilemmas should, according to Hampden-Turner and Trompenaars [33], be reconciled (win-win) rather than compromised (loselose) to achieve synergy. … get people to sit down and have an ouzo or cup of … coffee … or a meal and suddenly they, through that time, you can get people who didn't want to talk together to talk to each other and then they realise they're not -they actually understand each other more than they thought. Interviewee # 16, September 2017 At first glance, a simple activity, such as drinking coffee together, or providing children with chess classes, which is in fact one of the activities organised by volunteers for refugee children, can have profound transformational impact.

… Just finding what is the common interest or what
Chess is education: people who play chess are try-ing to think about next steps. When children learn it -in the beginning they always say it is someone else's fault when they lose, as it is human to give the blame to someone else, but after two months, even if they lose they are thanking you for a good game. They learn how to think ahead, but also to lose. I think children who learn to play chess will be better citizens. Interviewee # 2, August 2016 This often requires the help of a convener. On Lesvos, the local priest took on this role at the height of the emergency in Skala Sikamineas. Over time, one is re-building trust among the people on Lesvos by engaging in debates about what the public value is that needs to be created.
In the model, building trust can occur in three ways: triggered by a convenor [30], if there is a clear need for trust, and if trust already exists. Trust is destroyed if long-term problems fester. In fact, when trust is destroyed, conflict between groups increases, adding to long-term problems which in turn destroy trust: a classic vicious downward spiral (and, technically, a self-reinforcing loop).
Of course, there will be a big scar because this is also an emotional [experience]. You know what happened. There was a lot of people that used to spend Christmas together and to have holidays together. Because of this [the situation on the island], they don't hang out anymore. Interviewee # 13, September 2017 Building trust when (some) trust already exists is another concrete example of morphogenesis: Any existing level of trust is the framework in which individuals can take (smart) action to increase trust, bringing the level of trust to a new stage, which then serves as a new framework for individuals to engage in actions to increase trust. Once enough trust is established, courageous conversations are possible. At the end of those conversations, "an agreement" can be reached about what is needed for the islanders and for the forced migrants on the island.

The longer we keep locals and refugees apart, the longer the misperceptions and media is going to start saying all these negative things because media only reports negative things and the tension just boils, boils, boils and then explosion and it's happening. Interviewee # 7, August 2016
If the long-term problem on Lesvos is to be solved, the refugees and forced migrants on the island as well as the host communities need to be jointly involved in searching for solutions fit for purpose. Locals need to be engaged in a double role: one, as participants in the innovation process to develop new solutions, and two, as beneficiaries of the innovations, along with the refugees. Hence the need to also innovate for and with the local communities. Reaching a public value agreement, especially in situations of institutional void, is a process in which citizen participation and new forms of deliberative, collaborative and participatory decision making are at the core. Once such a more inclusive PVA exists, AE and OC are similarly extended to serve the needs of all relevant segments of society, delivering both PVO-1 and PVO-2, making the entire causal loop once again unambiguously self-reinforcing. To manoeuvre within that space and to orchestrate a solution in this space requires adaptive leadership, which is one established form of collaborative governance [34].

Conclusion
This paper has argued that long-term problems that occur in institutional voids can be addressed only through adaptive leadership that originates in civil society. To start, it is essential that trust that has been lost or destroyed must be rebuilt in order to increase social capital, so that agreements can be made that enable solutions that are "fit for purpose", meaning that they contribute to solving long-term problems, enabling the creation of public value.
The presented model suggests that "solutions fit for purpose" that have emergent transformative potential, as opposed to being a (series of ) products, are the ones including what Westley [87] call scaling-up and scaling-out, and are process-oriented. Essential in these processes is the building of bridging social capital. The model for creating public value (in institutional voids) described takes place along three different routes. The alignment and interconnection of these are the collaborative governance model, and, at its core, the need for courageous conversations. This process is full of tensions, but crucial for creating a shared sense of direction, and to align powers for action. The process is a form of collaborative governance, requiring an adaptive leadership. Its multi-actor perspective of public value creation [16] challenges traditional governance structures of how to "lead"-not only in partnerships with other levels of governments and organisations, but also with the active involvement of informal associations, community groups and individual citizens. Hence, it places civil society at the centre of debates about democracy, the public sphere and public value.
In this paper, the public value approach developed by Moore [55] provides a frame for looking at the problem from an action perspective and points to the normative issues related to reaching an agreement. The traditional managerial public value literature pays very little attention to the agreement necessary to create authorising environments. But, once the public value approach is applied in situations of institutional voids, it becomes critical to pay special attention to the question of an agreement. The question of an institutional void must always be seen in relation to a particular serious problem. There are no general, or universal, voids. Thus, the migration issue on Lesvos revealed an institutional void only in relation to the need to deal with hundreds of thousands of refugees and migrants arriving at the island in a matter of months, and the secondary impact that had on the local population and its means of survival. It does not mean that on Lesvos there is a general and complete absence of institutions.
In a situation like the one on Lesvos, where there was and is no clear authorising environment, such as a public manager formulating the public value outcome, the importance of identifying the process of deliberation to bring into existence an agreement can scarcely be overstated. Thus, it is suggested that we make Moore's public value strategic triangle into a useful guide for understanding and for action also in situations of institutional voids an extension to the traditional public value approach-here called the PV agreement approach. Actually, it is more appropriate to call it "the (struggle for) agreement". Thus, the leadership task on Lesvos is to set in motion a process in which an agreement is reached that addresses the complex problem of forced migration on the island and its interconnection with the tourismdependent local communities, and finds a solution for both. The public value agreement needed is not based on coercive powers, but on a joint transformative experience. They are conflictual spaces, within which people and organisations with competing and conflicting interests can use "agonistic" pluralistic negotiation [58] to create coalitions with a common purpose. For this development, building trust is essential, However, over time, an alignment for purpose is essential for producing physical public value outcomes "solutions fit for purpose" that can address the problems. It is not based on a search for the right answer or the best solution, but on a process of agreeing on solutions that address the problems. Those solutions then deliver public value outcomes. These processes are context-and time-specific.
Hence, an emphasis needs should be made on sensitivity to context when it comes to the public value debate, instead of to accounts that presuppose absolute standards of the good.
It also follows Mulgan's [59] notion that value is never an objective fact, but can be and is being disputed. "Agreement" can arise only through processes of negotiation and argument, while it is important to emphasise that reaching an agreement in conflicts cannot be about the "better argument" in the sense of Habermas [32]. It should be about "sense making", i.e. reaching some level of shared understanding of the problem to be solved and how to solve it, taking into account "the interdependence of boundary judgements, observations, and evaluations" [82]. This is partly because universal assumptions can become dogma and close off the potential for sophisticated dialogue about moral problems [44].
Dahl and Soss [21] argue that this dispute is essential as a foundational premise for democracy and a practical source of action and renewal-contestation, difference and conflict in the context of power relations [46,47,62]. Knowledge of new facts can equally make us change boundary judgements, which in turn makes previous evaluations look different [82]. Only when people are in dilemmas, where different positions are prevailing and different values are at stake, is there a need to pave a way for the beginnings of an agreement. This does not start with large deliberate processes, but small actions like having a cup of coffee or an Ouzo together. A key aspect is to find the common interest of people, so that people can work together even if they hate each other.
According to Ansell and Gash [3], "both trust and interdependence are endogenous -they are shaped (in positive or negative ways) through the collaborative process itself ". In the model, it suggested that there are two ways for the agreement-doing and through, which in the adaptive leadership approach are interwoven. In this process, courageous conversations are a space for social learning [39]. If these courageous conversations do not take place, or are not successful in bringing forth an agreement, a further vicious cycle is set in motion. However, the opposite is also true: through an adequate adaptive leadership [36], not only can an agreement be reached, but through public value creation of bridging social capital will be strengthened, enabling collective sense-making, mobilisation and trust-building between networks and citizens taking on the roles and attributes of distributed leadership, and actually achieving momentum for change (through the self-reinforcing casual loop, awareness building, engagement, courageous conversations, and action). The conflicts emerging between the two public values to pursue cannot be resolved by "top-down or technical leadership" alone; a longer, wider and deeper perspective is needed-one provided by "adaptive leadership" [34]. Hence, in the situation of an institutional void, adaptive leadership can fill the void.
It is necessary to address leadership in different sectors, encompassing community (civic) leadership as well as organisational leadership, political as well as managerial leadership, implementation as well as formulation of policy, operational as well as strategic management, and the evaluation of impacts and results. This implies that such initiatives need to focus not only on individual development, but also on how learning is translated into organisational development [9]. Forums where this learning takes place are therefore needed. Using a systems model and its language provides an opportunity to discuss and engage with complex issues and problems. Hence the model presented here could also be used in a more general discussion of the relationship between adaptive leadership in institutional voids and transformative change. Middle-range theories start with an empirical phenomenon (as opposed to a broad abstract entity like the social system) and extract from it general statements that can be verified by data. They are "theories that lie between the minor but necessary working hypotheses that evolve in abundance during dayto-day research and the all-inclusive systematic efforts to develop a unified theory that will explain all the observed uniformities of social behaviour, social organization and social change" ( [50], quoted in [70]). This might seem to be a constraint: as Box and Norman [11] states, "[a]ll models are wrong, but some are useful. " They are wrong because they do not fully represent reality. But that is their purpose: a map of a territory that maps the territory one-to-one to a map would be pointless. So, the questions always become how to extract, what to leave out, what to include and at what level of detail. These questions have no general answer, but only one in relation to the purpose that the proposed model is meant to serve. The model described is a conceptual one, and not a fullyfledged simulation model. This has been perceived as sufficient, as its purpose is trying to understand the societal fragmentation on Lesvos in the aftermath of the forced migration crises, in order to offer suggestions about how to intervene. For changing a system, not only the intervention as such is key, but also its timing. If waiting until the crises is over, often the chance will be gone. Adaptive leadership is about mobilizing people to make significant adjustments in their attitudes and habits, while at the same time learning their way into the creation of something that does not yet exist. Thinking dynamically, i.e. in behaviour over time, involves considerations of how the problem evolved in the past to manifest itself in the present, where one might go from here, and how its development can be influenced. The situation on Lesvos changed over time [18], and where it might go from here, exploring how it can be influenced, and aligning for action requires adaptive leadership. It might sound naïve to try to obtain an agreement in situations of conflict, such as on Lesvos.
However, examples show that deep-rooted conflicts can be entangled in processes of deliberation. Chambers [19] stressed that community leaders, through relationships, can, in fact, initiate the seemingly impossible: "People who can understand the concern of others and mix those concerns with their own agenda have access to a power source denied to those who can push only their own interests. In this fuller understanding, 'power' is a verb meaning to give and take, to be reciprocal, to be influenced as well as to influence. To be affected by another in relationship is as true a sign of power as the capacity to affect others. Relational power is infinite and unifying, not limited and divisive. As you become more powerful, so do those in relationship with you. As they become more powerful, so do you. A virtuous cycle. This is power understood as relational, as power with, not over" [19]. This is the space for adaptive leadership.