The development of the European integration project affirmed the narrative about the obsolescence of major war on the European continent, through the profound transformation of relations among European states in the second half of the 20th century. The concept of ‘Europe whole, free, and at peace’ (Bush, 1989) dominated much of the public and academic debate in the aftermath of the fall of the Berlin Wall. The idea of an expanding ‘security community’ promised to bridge the east-west divide and enhance the security of the whole continent. Yet, just over three decades later, the belief that war had become unthinkable as a means of resolving political differences on the continent was shaken by Russia’s full-scale invasion of Ukraine, prompting the EU’s top diplomat Josep Borrell to warn in 2024 that ‘[a] high intensity, conventional war in Europe [was] no longer a fantasy’ (Foy, 2024). The degree of uncertainty facing the European security order today has never been so heightened throughout the post-Cold War history of the EU. 

Most studies on European security in the last decades have evolved around the changing perceptions of security risks and the subsequent policy responses centred on capability building and deepening institutional integration at EU level. This has given rise to the burgeoning literature on EU crisis management (Emerson and Gross, 2007; Blockmans, 2008; Popescu, 2011; Gross and Juncos, 2011; Juncos and Blockmans, 2018), the academic debate on resilience (Juncos, 2017; Tocci, 2019; Korosteleva, 2019) as well as to various conceptualisations of the EU’s external power. In the latter vein, the EU has been seen as a civilian power (Duchêne 1972), a normative power (Manners, 2002), a market power (Damro, 2012), a liberal power (Wagner 2017) or a superpower (Moravcsik 2017). What is characteristic of these conceptualisations is their reliance on traditional notions of power related to an actors’ ability to calculate risks, assign probabilities and design policy responses meant to control outcomes, i.e., they are different conceptions of control power (Katzenstein and Seybert, 2018).  

In the domain of security, EU policymakers have faced unpredictable security shocks in the last three decades. These uncertainties have conditioned the framework within which EU enlargement and neighbourhood policies have evolved in the aftermath. These include the Balkan wars of independence of the 1990s after the break-up of Yugoslavia, the lingering territorial conflicts in the post-Soviet space throughout the 1990s and the 2000s after the break-up of the Soviet Union as well as the 2008 Russo-Georgia war, the Russian annexation of Crimea in 2014, the Nagorno-Karabakh dispute and the full-scale Russian invasion of Ukraine in 2022. The EU’s security posture was also affected by the surprising election of Donald Trump to the US presidency and the departure of the United Kingdom from the EU in 2016. Both events triggered concerns about the ability of EU Member States to guarantee their own security and project security to the neighbourhood. In all these instances, the EU has been pushed by circumstances to improvise, innovate and transform itself on the spur of the moment and without much planning in advance. In so doing, it has demonstrated power best captured by the notion of protean power, or ‘the effect of actors’ agility as they adapt in situations of uncertainty’ (Katzenstein and Seybert, 2018, p. 80).

The concepts of risk and uncertainty are fundamental in understanding the complexities of global politics and the interactions among various international players. Although these terms are often used interchangeably (for a discussion on the terms, see Knight, 1921), they embody distinct characteristics that are important for scholars and policymakers to recognise. The complexities in the international environment involve ‘unknown and/or uncertain attributes’ of opponents that push policy makers to operate in a foggy situation involving ‘high-risk calculation’ (Jarvis, 2011: p. 297). Donald Rumsfeld highlighted this policy dilemma when stating that ‘there are things we know, and we know we know them – the known knowns. There are things we know that we don’t know – the known unknowns. And there are unknown unknowns; the things we do not yet know that we do not know’ (Rumsfeld, 2001). 

The terms ‘risk’ and ‘uncertainty’ can mean different things to different people. For Knight, risk can mean ‘a quantity susceptible of measurement’ (Knight, 1921, pp. 19–20). Uncertainty, in contrast, has an unmeasurable quality to it (Knight, 1921, p. 20). So uncertainty can be limited to non-quantifiable cases in comparison to risk that can be more quantifiable (Knight, 1921, p. 20). Katzenstein and Seybert (2018) use the terminology of calculable versus incalculable to capture the distinction between the domain of risk characterised by calculable expectations about the future and the domain of uncertainty defined by its unforeseeable qualities (Katzenstein and Seybert, 2018, p. 85).

The differentiation between risk and uncertainty has significant implications for policymakers as it influences their ability to handle foreseeable shocks versus unpredictable events. Under conditions of risks, policy makers and relevant stakeholders operate in the domain of the expected and predictable. They are aware of the consequences of certain occurrences and can attach probabilities to different eventualities. In such scenarios, they find themselves in an environment where they have adequate information that can help them estimate the risks and plan accordingly the resources at their disposal. In complex but predictable environments, policy makers act with the intention of exerting control over future outcomes, i.e. they exercise what Katzenstein and Seybert (2018) refer to as control power.

Under conditions of uncertainty, policy makers must be creative and use the tools that they have at their disposal in innovative ways to handle unpredictable scenarios. The concept of protean power has emerged as a framework to understand how individuals, organisations, and states navigate such dynamic and uncertain environments. Protean power, as conceptualised by Katzenstein and Seybert (2018) refers to ‘practices of agile actors coping with uncertainty’ (Katzenstein & Seybert, 2018, p. 80). Protean power stems from the ability of actors to shape their environments by leveraging a combination of resources, strategies, and networks and by innovating and improvising in situations of unexpected developments. This is reflected in flexibility, adaptability, resilience and transformation in the face of sudden shocks to the status quo. Power in such contexts is generated through the surprising actions and self-transformation of agile actors who try to steer the course of uncertainty. 

Our analysis of control and protean power starts with acknowledging the distinction between risk and uncertainty in decision-making as suggested by Katzenstein and Seybert (2018). The former is connected to the realm of predicable and foreseeable occurrences whereas the latter is experienced because of the potentiality of unpredictable and unexpected change. Classifying events and situations as representing risks versus uncertainties is not easy. In retrospect, we can make relatively safe assumptions about political life as risky or uncertain, but we cannot be sure that policy makers at the time have experienced the environment as risky or uncertain in the same way as we describe it years later. We therefore try to contextualise the events and occurrences in the Western Balkans and the Eastern neighbourhood that have spurred the EU into action over the course of the last 30 years, keeping in mind ‘the fluidity of real-life situations that often oscillate between risk and uncertainty’ (Katzenstein and Seybert, 2018, p. 85) and providing an expert reading of predominant perceptions of risks and uncertainties at the time of the events.

Likewise, when we distinguish between the effects of control power, linked to the domain of risk, and the effects of protean power, generated in the context of radical uncertainty (Katzenstein and Seybert, 2018), we are cognisant of the interplay between the two types of power and their interdependent and even reinforcing qualities, with protean power often leaning on control power capabilities and control power resources often necessary for generating protean effects. Our analysis is in this sense both guided by the main conceptual framework offered by Katzenstein and Seybert and sensitive to the complexity of the empirical contexts that we deal with.  

This study examines the interaction between control and protean power in the evolution of the EU security policy in the Western Balkans (WB) and the Eastern neighbourhood (EN). It starts by outlining the security risks and uncertainties that the EU has faced in and in relation to the two regions since the fall of the Berlin Wall. It then investigates EU innovations and improvisations developed in concrete situations of uncertainty, giving rise to protean power practices as well as the control power tools employed by the EU in response to security risks in the two regions.

Download the full document