Guest article: Complexity and environmental policy

Contact

Gary Kass

What is complexity?

In common usage, complexity is synonymous with situations that we see as ‘messy’ or ‘complicated’. More formally, complexity describes something with many parts that interact with each other in many different ways, leading to the ‘emergence’ of properties that are greater than the sum of its parts, and that cannot reliably be predicted from studying the individual parts. Emergence has been described as “the stuff of systems” (Hoverstadt, 2025).

However, there is no single definition of complexity, and scientists from many disciplines have pursued multiple avenues to describe, understand, and explain complex systems for over 50 years – with the Santa Fe Institute founded in 1984 as the first research institute focused on complex systems.

Complex systems can be biological: cells, organs, organisms, ecosystems; or to do with infrastructure: power grids, transport, computer and communication systems; or social and economic organisations: cities, businesses, charities, and public institutions.

Complexity is neither inherently ‘good’ nor ‘bad’: it is necessary for some systems to form and remain viable, but it may also cause problems, such as the exponential spread of a disease resulting from multiple interactions between infected individuals. “Problems arise not from too much or too little complexity (at any scale) per se but rather from mismatches between the complexities of a task, such as ensuring a reliable supply of electricity, delivering mass vaccinations, or tackling climate change, and the complexities of the systems performing those tasks, such as a power grid, a mass vaccination programme or rolling out renewable energy.” (Siegenfeld and Bar-Yamr, 2020)

Why does complexity matter for environmental policy?

We can address this by exploring how ‘policy’ relates to ‘the environment’. These are already two ambiguous terms, each relating to complex ideas.

Policies are complex – there are many moving parts that interact with each other, sometimes creating synergies (positive feedback) and sometimes clashing, reducing coherence and effectiveness (negative feedback). Policies also often create unintended outcomes from the emergence of behaviours that could not be predicted in advance. Paul Cairney (Professor of Politics and Public Policy at the University of Stirling) has documented the relationships between complexity and policies, noting that complexity is embedded in policy making processes, policy issues, and policy mixes, such as combining regulation, voluntary initiatives, funding and market-based incentives (Cairney, 2020).

The environment also has huge numbers of interacting parts that give rise to multiple outcomes and feedback, over multiple spatial and temporal scales.

Bringing policies into the sphere of the environment really multiplies the complexity. In climate policy, the huge range and depth of complexity can generate multiple and pervasive uncertainties, such as in climate data, climate models, understanding impacts and assessing the effectiveness of mitigation or adaptation measures. Such uncertainties generate many consequences, as described in the IES’s 2025 ‘Speaking Up for Science’ collection of case studies (pp. 36-41).

How can we handle complexity in environmental policy?

Since the IES is led by science, we are interested in how scientific approaches and methods can help us to make sense of, and respond to, complexity arising from different sources and manifesting in different forms. To this end, ‘complexity science’ (or complex systems science) has evolved as an interdisciplinary endeavour spanning natural and social sciences, arts and humanities.

Complexity science includes academic fields such as system dynamics, cybernetics, evolutionary dynamics, critical systems theory, network science, fractals and scaling, nonlinear dynamics and chaos. Each field has its own suite of quantitative and qualitative methods, such as stocks and flows modelling, soft systems methodology, agent-based models, and network structure analysis.

Bringing these approaches into environmental policy requires a clear understanding of the specific needs of the stakeholders and affected parties involved in the context of the issues being explored. Only once the objectives are clear should methods be considered, recognising that none has the monopoly on wisdom, and that multiple methods might be required, combined in appropriate ways and in appropriate order. For example, when considering scaling up a circular economy system, network analysis can help to map and characterise the system as it is, while soft systems methodology can help understand stakeholders’ perspectives on what goals should be pursued and how to move ahead.

Assuming this approach has been followed, what might a complexity-sensitive and systemic approach to environmental policy look like? Essentially, the most significant aspect of complexity in relation to policy is the impact it has on the ability to predict the outcomes of the measures being proposed and implemented. With multiple moving parts, interacting in many different ways, and the situation adapting to change as the policy is put into effect, it becomes increasingly difficult to anticipate what will happen with any great reliability.

Four modes of policy making have been proposed and implemented (to varying degrees of success) that seek to account for such shifting patterns. These are: 1) adaptive governance, 2) transition governance, 3) transformation governance, and 4) anticipatory governance.

A review of these modes of governance in the context of climate change (Soininen et al, 2025) found that “all four governance models are applicable to different aspects of climate change mitigation and adaptation.” Further, the authors point out that “one clear result of social-ecological and governance complexity portrayed by climate change is that context matters.” Thus, what is applicable in one place, at one time, and in the face of particular circumstances, is not necessarily appropriate elsewhere, at other times or in different circumstances.

Adaptive governance responds to the context where coupled social-ecological-technological systems operating in a certain situation undergo constant change driven by external factors. Complexity is embraced by recognising the unpredictability of emergent outcomes, so policy and other aspects of governance must address uncertainty and complexity head-on through adjusting regulatory and market mechanisms. Soinonen et al suggest that adaptive governance might require altering the process of environmental impact assessment to an ongoing adaptive process in which learning takes place through evaluation of management actions, underlying assumptions, and knowledge production itself are re-visited (citing Pace and Cosens, 2023).
Transition governance is focused on governing innovation in technology and social institutions. Recognising the need for change, transition governance seeks to define and put into practice necessary societal shifts, towards a desired future state. This is achieved through 'transition management' processes that explore, experiment, and reflect on innovations in governance to address sustainability challenges. Transition governance focuses primarily on markets and decentralised decision making, with government policy and laws acting as facilitators (e.g. funding small-scale demonstrations), until the market can take up the slack. Recent criticism suggests that transition governance has not yet paid sufficient attention to the full range of legal and policy instruments required to ensure fairness and transparency, leading to loss of legitimacy (Köhler et al, 2019).
Transformation governance is aimed at shifting societal structures and mindsets. Noticeably, global science-policy bodies like the IPCC and IPBES have embraced this notion of ‘transformative change’ (e.g. IPBES 2024). Transformation governance seeks to effect change in three spheres: practice, politics and philosophy, drawing on Donella Meadows’ arguments around leverage points, where transformations can occur by intervening at strategic points, with paradigm shifts acting as the most powerful leverage point (Meadows, 1999). Education, strong regulation, increasing the costs of unsustainable production, and encouraging sustainable consumption are seen as the key policy directions.
Anticipatory governance is described as “governing in the present to adapt to or shape uncertain futures”. (Muiderman et al., 2020). Three core practices are involved: anticipation through futures analysis and foresight (particularly using scenario building to avoid more traditional 'predict and plan' approaches); creation of adaptive planning and management strategies; and monitoring and action. Anticipatory governance envisages a strong role for governments in facilitating the forward-looking perspectives involved, and facilitating planning to build flexibility and agility to enhance responsiveness to emergent change. There are few governance mechanisms described, leading Soininen et al to suggest that anticipatory governance plays a supportive role alongside adaptive, transition, and transformation governance.

We need to build our capabilities to govern in an increasingly volatile, uncertain, complex and ambiguous (VUCA) world (Cascio, 2020). To help in this endeavour, the IES’s event on 11 August will open a space for discussion about how we can shift policy and wider aspects of governance to address some of the most critical environmental and sustainability challenges we face.

Find out more

For further details, please contact:

Gary Kass, Professor of Sustainability Science, Policy and Practice, Imperial College (g.kass@imperial.ac.uk)
Joseph Lewis, Head of Policy, the Institution of Environmental Sciences (Joseph@the-ies.org)

The IES discussion on 'Handling complexity in environmental policy' will take place at 12pm-1.15pm on 11 August 2026.

Gary Kass has been a researcher, knowledge broker, educator and consultant since 1986, working with organisations in the private, public and third sectors across the UK, Europe and internationally. Gary is a Visiting Professor at Imperial College and the University of Nottingham and an End-Point Assessor for the Masters-level Systems Thinking Practitioner and Sustainability Business Specialist Apprenticeships. He is Chair of the External Policy Advisory Committee at the Institution of Environmental Sciences, and is a Certified Member of Systems and Complexity in Organisation, the professional body for systems thinking practitioners.