The leadership for sustainable development


Draft for review, date last revision: 17 October 2022 (3000 words)


In this blog I summarized how leadership may promote sustainable 'legitimate cocreation' by hybridizing governance and to that end use tools like impact assessment, the right kind of polls, and citizen's juries. Below, I elaborate more on this kind of leadership. We need to understand the agency-side of hybrid governance better. Here I focus on Complexity Leadership Theory, with the CAS at its heart.


Resilient Complex Adaptive Systems (CAS)

In our time, every organization is dependent on clients and on other organizations: suppliers, financiers, implementers, unions, partners, and competitors with whom they must collaborate and agree on the rules of the game, and governments can enforce the rules. An organization may look well-organized, but it is impossible to track who is collaborating and competing with whom inside and on the outside. Yet, the survival, health, vitality or resilience of the CAS and each organization in it depends on this myriad of interactions when changing circumstances require adaptation.


Any group of interdependent organizations, organizational units, or less organized stakeholder groups, jointly depending on scarce natural resources, is referred to in complexity theory as a (social or socio-ecological) Complex Adaptive System or a CAS (e.g., Preiser et al. 2018). A hierarchical organization such as a hospital is therefore a CAS. At another scale, the same goes for the global society as it depends on shared and finite natural resources. An insight from complexity theories ('Ashby's Law') is that if a CAS is to be able to deal successfully with the diversity of challenges that its environment produces, then it needs to have a repertoire of responses which is (at least) as nuanced as the problems thrown up by the environment.


We may intuitively understand Ashby’s Law and see how resilience depends on the myriad of interactions nobody can oversee. Then we also understand that such a repertoire of responses does not emerge automatically: for the adaptability of a CAS, the behaviour of the individual people in the CAS is more decisive than the precise structure of the organization and the 'orders from above'.


What does this mean for those who ‘manage’ these people? How to they intervene in self-organising social interactions to stimulate hybrid governance? According to CLT, the answer is ‘enabling leadership’.


Complexity Leadership Theory (CLT)

Uhl-Bien et al (2007) proposed Complexity Leadership Theory (CLT). The central proposition of CLT is that effective leadership in CAS is exercised jointly by several leaders having no hierarchical mutual relationships - no control over each other. These leaders self-organize into networks criss-crossing through the CAS, which enables them to combine their knowledge and influence, devising and initiating gradual changes of the CAS. As together they have knowledge about the whole CAS, their arguments are strong and their whole is more than the sum of the parts. With these gradual changes, they try to pave the way for bigger changes.


Leader networks cannot force or plan these major changes. But they can think of ways to destabilize the CAS so that it may move to another equilibrium. In systems theory, CAS are living ('autopoietic') systems that can only be distorted - not steered - by some intervention from the outside. The CAS can adapt to that change in its environment by moving to a new equilibrium. Those who intervene can hope that the new equilibrium will be more resilient (or healthy, vital) than the previous one, so that the CAS can survive better in some adapted form with a view to future changes in its environment. (Vaccination is an example of distorting living systems to provoke adaptation to resilience).


By imagining how the 'autopoietic' CAS might adapt to their distortions and make them constructive (i.e., for the greater good of all), the leaders try to steer after all. Complexity leaders can try, in a coordinated effort, to bias development by influencing the factors determining to which new equilibria the CAS may move, perhaps in small steps cascading toward an even greater change (i.e., they are ‘organising criticality’; ref. Kron & Grund 2009).


Such complex changes, supposing they even can be deliberate, can never be attributed to one person.


How complexity leadership is 'networked' and 'transdisciplinary'

Digging deep, when proactive change happens, researchers may be able to reconstruct who were part of leader networks having some influence. Such reconstructive research shows that the leaders participating in complexity leadership are in contact with each other so that a network is formed - which does not necessarily mean that they all need to know each other personally. The network operates on trust in each other's intentions and in each other's skills. That intent consists of a shared analysis of what the CAS is – jointly making somewhat arbitrary and fuzzy boundary judgments to define the CAS (e.g., Ulrich 2022) - and why it needs to change. Those skills include being able to contribute to joint systems analyses – how does the CAS work? How resilient is it? If not so, is it worth saving? Which small measures that contribute to vitality can leaders in the network initiate and are politically safe enough for them to take? (cf. the idea of ‘small wins’ by Termeer & Dewulf 2019). To which larger changes may this build up?


Ample small group techniques exist to perform such 'integrated' systems analyses and imagine constructive distortions, like Adam Kahane's Transformative Scenario Planning (2012). And small groups can engage with large groups to probe for the understanding and support of their emerging ideas. Complexity will foreclose full integration as everything on our globe is connected to anything else. But imagining the boundary judgements the public sector will make when looking for legitimacy (see above), cocreators may look for  more integration than before. They become increasingly 'transdisciplinary'.


Those small changes of the CAS already become widely visible when powerful leaders – in CLT: administrative leaders - synchronously send synergetic change-oriented signals to staff and the outside world. If they think the time is ripe, they can adopt change-oriented policies following procedures that add to legitimacy.


It is not a conspiracy: shared knowledge does the work, paving the way for legitimate interventions. And it only works if the cocreators do it together: participating leaders must trust that other leaders also do what is necessary, within their capabilities, sharing opportunities to mobilize the CAS in a self-organized, synchronized way.


Complexity leadership needs sub-dynamics

As CAS are complex, the networked leaders must collectively have far-reaching reliable knowledge and persuasiveness. To keep the trust of other leaders whom they cannot control, they cannot resort to coercion, and they must be patient.


Such all-rounders are rare, but not everyone has to be able to do everything. According to CLT, the networks split into dynamics in three sub-networks, each CAS-wide, which support each other. As the networks are unofficial and often volatile, and as one all-rounder can be member of all three dynamics, they are not always easy for a researcher to observe and distinguish. But when asked, a complexity leader may explain, albeit in their own words, to which dynamics her or his actions belong (adaptive, enabling or administrative as explained hereafter); his or her personal transformative scenario or theory of change, so to say.


So, what are these sub-dynamics?

First, there are the adaptive networks, which can tie together knowledge across all corners of the CAS to outline a palette of realistic scenarios for the CAS, including the synchronized small interventions that may gradually move the CAS in desirable directions.


Secondly, there are the enabling networks that give participants of adaptive networks the space - i.e., the time and other resources they need - and that can keep the ideas about adaptation sufficiently 'in sync' with the official agenda of management (Nooteboom & Termeer 2013). Enabling leader networks can have a repertoire of interventions into the structure of governance processes, to increase the quality of governance outcomes in terms of CAS resilience. They are like meta-governors, although enabling networks probably will focus first on nurturing self-organized, subtle, volatile, innovative social structures as vehicles for adaptive leadership dynamics that can pave the way for stronger interventions of meta-governance.


Thirdly, there are the hierarchical administrative leader networks, which, visible to everyone, may give signals that things must change and implement this change if that doesn’t harm their position of power. They can do that as a formal network (like a council of ministers) in which case they first will have a negotiation behind the scenes. Their members can also do it individually, attracted by personal political opportunities. The results of somewhat hidden and shady cocreation come to the surface - which makes these ideas official: the statements by formal leaders create expectations among followers. Followers regard everything a politician or delegate says in public as official. Members of administrative networks take the decisions that set the CAS in motion: this is where joint thinking transforms to visibly separate (but synergetic) acting.


Members of adaptive and enabling leadership sub-networks must be willing to invest for the benefit of the CAS without knowing if they will be personally rewarded for their efforts. They may run risk if they spend their time unmandated or if they develop controversial transformative scenarios which are prematurely exposed (these risks are more likely in a context of political polarisation).


Furthermore, not every participant in complexity leadership is a manager – on the contrary: others can also have important knowledge and persuasiveness to offer and may want to make a personal contribution. Enabling leaders can be managers with formal authority who enable innovative projects, commit to taking the yield of these projects seriously, and defending their space in administrative networks where short-term rationality dominates. Complexity leaders are not only needed throughout the CAS (multi-level, multi-sector), but also at all hierarchical levels of organizations: the workforce of organizations contain unique and crucial knowledge and often is the best place for adaptive leadership.


Adaptive leadership may even become the standard mode of operation if there is enough enabling leadership. It may appear to distant observers as if the quality of development proposals (in view of CAS sustainability) magically improves, as cocreation completely hybridised with legitimacy creation (ref. the idea of superblending I mentioned above).  Without any enabling leadership at all, our world may become Orwellian as administrative networks may become completely self-referential - losing touch with reality in the CAS. They will not be open to ideas cocreated by others.


Consequences for managers

The question arises: what can a manager do to help make a CAS resilient? (Instead of leaving the sinking ship like a rat, so to speak – maybe after benefiting from it first for a while.) The self-organisation into networks of complexity leadership begins if individual managers themselves look with a CAS-lens and gather people around them who do the same. Horizontally: managers may look for colleagues from other corners of the CAS with whom they can share ideas about their joint CAS. They can create partnerships as vehicles for hybrid governance (Lowndes & Skelcher 1998), context for their enabling leadership. And vertically: managers may look for subordinates in their own organizational unit who have more time than they themselves to participate in adaptive networks. They also may assess the room for complexity leadership that enabling leaders 'above them' at the top are offering.


Any project that an adaptive network criss-crossing the CAS proposes as a vehicle for their work needs multiple managers who want to support that project. Those managers must therefore understand each other well, especially since all must not only enable adaptive leadership, but also - in their role as administrative leaders - implement part of the yield. If the environment of the CAS changes too much before complexity leadership emerges and helps change to emerge, that change is then reactive rather than proactive. The required adaptations then may have to be drastic or the CAS may collapse.


Much better would be to be really proactive and to facilitate continuous change.


Facilitating continuous change without changing the hierarchy all the time

Adaptation can manifest itself when the hierarchical structure of organizations changes. But every new structure creates its own problems. In complex CAS hierarchical fragmentation in siloes is therefore unavoidable, yet the restructuring of institutionalised hierarchies is painful and costly, and should serve as a last resort. Complexity leadership at early stages may prevent the need for major restructuring by devolving adaptation to the operational policy-making or policy implementation level.


Continuous adaptation is already daily fare for the policy professional or street-level bureaucrat who works with stakeholder groups and clients in conjunction with other organizations that in turn have their own professionals. In complex conditions, managers cannot guide their subordinate professionals in detail. Professionals need some freedom to act according to (i.e., adapt to) their situations, to keep relations with other parts of their CAS – as it transcends their organizations - healthy or vital.


The role of managers becomes threefold. Their first role is to self-organise into enabling networks and to facilitate the sharing of knowledge and the synchronizing of action of professionals across organizational borders; for example by removing barriers like internal operating rules or unnecessary confidentiality. The second role of managers is to provide networked enabling leadership where the professionals from several organizations ask for confirmation of the legitimacy of what they are doing or what they propose to do. For dialogues with citizens they can apply institutional blending instruments. Their third and last role is to manage, as administrative leader, the dialogue about that same legitimacy between their organization and its stakeholder group at large. All these acts can be seen as acts of metagovernance.


So, where does this take us?


Complexity leaders as adaptive cocreators separate from legitimate administrators, linked by enablers

These cross-organizational networks of leading professionals cocreate change, and networks of leading managers enable this cocreation despite it being at odds with the mainstream - and therefore in need of legitimacy. If legitimacy of results needs to be confirmed, enabling leaders offer these to the mainstream of administrative leadership to engage in dialogue with large stakeholder groups and legal experts. In this way, legitimacy of proposed changes becomes an emergent property of the CAS. It makes the CAS more resilient. It is only possible if adaptive leadership is separate and safe from administrative leadership, but then again connected by enabling leadership: it is a form of hybrid governance.


How does CLT relate to other approaches to hybrid governance?

Above, I wrote about metagovernance and meta-regimes, but I have not made a complete survey of these approaches or compared them in detail with CLT. It seems probable that most approaches assume that structures largely determine agency and that, therefore, sustainable outcomes depend on the right design of structures. This includes the core of societal structure like the structure of markets and of government (Meuleman & Niestroy 2015), but also the more refined structures, like collaborative projects or transition tables that design development programmes and their governance structures.


Metagovernors may reflect on what would be the checks-and-balances and the collaborative structures that would engender the most sustainable outcomes in a specific situation. The interventions they make to change the structure according to their ideas may be called metagovernance. However, no metagovernor can design and implement such a complex restructuring by his own. Metagovernors from throughout the CAS need to do it together, but help doesn’t come if the required agency does not emerge from the obsolete structure that needs to change in the first place. The chicken-and-egg problem, if the CAS is not too complex, can only be overcome by jointly reflecting without the structure yet being in place. This coevolution between actor and structure requires complexity leadership.


Full circle: complexity leadership to enable superblending of hybrid governance

Might cocreating alliances create legitimacy, in a single process - i.e., by themselves ensuring inclusion and checking lawfulness? In theory this is possible; it might be called superblending as cocreation and legitimacy creation are then blended so well that they can hardly be separately observed anymore. There are then numerous 'bridging frontstages' enabled by 'bridging backstages', increasing the problem-solving capacity of the CAS. But when the stirring stops, cocreation and legitimacy creation would still separate: oil would only float on water.


So, who is superstirring then? That would require that public governors support every step of the cocreation process constantly checking if their electorate will likely give sufficient support in the next elections, and that lawyers are closely involved to predict how judges will rule if the odd citizen who is not included in the cocreation were to appeal to a court. Any proverbial banana peel on which the cocreated results may slip can then be removed in the cocreation itself. And this would need to be done by all required sectors and at all levels of government simultaneously. In theory they can be connected in enabling networks.


These enabling networks would have to be linked sufficiently to power over resources. Resources are often allocated to the destinations of the past. A politician adjusting his or her budget allocations or spatial allocations will often make enemies and may have to follow lengthy procedures which have to run parallel to the cocreation. The enabling networks would have to be able to deal with both, all the way ensuring wide support.


In not overly complex or polarized contexts, superblending may foreclose a parallel legitimacy-creating transparent process where public managers and politicians from sectors and levels of government make their own deals on what kind of cocreation they will support. Coordination between public authorities may be limited and informal, as risks can sufficiently be dealt with in the public-private-civil cocreation. But if cocreations have to deal with complex societal transitions, it is more difficult to imagine how superblending may work. Public authorities will have to make their own formal coordinated decisions on where cocreation is allowed to go.


It is easy to imagine that the more adaptive and enabling leadership exists in the CAS, the more superblending may emerge and support the formal coordination of public planning processes. Complexity leadership may, to outsiders 'magically' perhaps, remove the very political and legal banana peels that make cocreation so difficult.


The somewhat disrespectful metaphor of the banana peel should be taken serious: every banana peel is a legitimate concern. Yet, complexity forecloses the possibility to connect everything to everything else transparently and to get everybody's approval, before moving on in a secluded cocreation where cocreators discover what they think is best for the CAS, in dialogue with affected groups.


One can hope that self-organised complexity leadership, leading to working integrally and transdisciplinary , will not only emerge in small pockets, but that these pockets coagulate to self-organised governance capacity at global scale. We have also seen some successes, like the 2015 Paris agreement on climate change. Such successes depend on networked leadership to make the global CAS resilient, global resilience may erode in its absence.


Improved with feedback from / after discussion with: Lonneke Dikmans, Geert Teisman, Jitske van Popering, Ytsen Deelstra, Jos Arts .. (Draft for feedback, appreciate if you send any observations to




  1. Ansell, Christopher et al. (2022). Challenges to the Cocreation of the SDGs and the Way Forward. Chapter 14 in Christopher Ansell, Eva Sørensen and Jacob Torfing (2022). Co-Creation for Sustainability. Emerald Publishing Limited
  2. Ansell, Christopher, Eva Sørensen and Jacob Torfing (2022). Co-Creation for Sustainability. Emerald Publishing Limited (Open source)
  3. Duijn Michael & Jitske Van Popering-Verkerk 2018. Integrated Public Value Creation through Community Initiatives—Evidence from Dutch Water Management Soc. Sci. 2018, 7, 261; doi:10.3390/socsci7120261
  4. Lowndes Vivien & Chris Skelcher 1998. The dynamics of Multi-organizational partnerships: An analysis of changing modes of Governance. Public Administration Vol. 76 Summer 1998 (313–333)
  5. Meuleman, Louis & Ingeborg Niestroy, 2015. Common But Differentiated Governance: A Metagovernance Approach to Make the SDGs Work. In: Sustainability 2015, 7(9), 12295-12321
  6. Meuleman, Louis (2018). Metagovernance for Sustainability. A Framework for Implementing the Sustainable Development Goals. Routledge
  7. Mintzberg, Henry (2015). Rebalancing society. 
  8. Nooteboom Sibout, Peter Marks (2010).  Adaptive networks as second order governance systems. In:  Systems Research and Behavioral Science Volume 27(1): 61-69. John Wiley & Sons, Ltd.
  9. Nooteboom, S.G. & Termeer, C.J.A.M. (2013). Strategies of complexity leadership in governance systems. International Review of Public Administration, 18 (1), 25-40.
  10. Nooteboom, S.G. (2006). Adaptive Networks. The Governance for Sustainable Development. Eburon.
  11. Nooteboom, Sibout (2019). Environmental assessment as an institution of liberal democracy. Impact Assessment and Project Appraisal 2019, Pages 109-112
  12. Preiser, R., R. Biggs, A. De Vos, and C. Folke (2018). Social-ecological systems as complex adaptive systems: organizing principles for advancing research methods and approaches. Ecology and Society 23(4):46.
  13. Rodrik D, Walt S. How to Construct a New Global Order. 2021. Copy at
  14. Termeer Catrien J.A.M. & Art Dewulf (2019) A small wins framework to overcome the evaluation paradox of governing wicked problems, Policy and Society, 38:2, 298-314
  15. Thomas Kron & Thomas Grund (2009). Society as a Self-Organized Critical System. Cybernetics & Human Knowing. 16.
  16. Torfing Jacob, B. Guy Peters, Jon Pierre & Eva Sørensen (2019). Interactive governance. Advancing the paradigm. Oxford. (First paperback edition 2019; first edition 2012).
  17. Uhl-Bien, Mary, Russ Marion and Bill McKelvey (2007). Complexity Leadership Theory: Shifting leadership from the industrial age to the knowledge era. In: The Leadership Quarterly 18:4 (August 2007), pp. 298-318; Special Issue on Leadership and Complexity
  18. Ulrich, Werner (2022 acceded) A mini-primer on boundary critique