In January 2011, the anti-government protests which had started in Tunisia in the previous month were spreading to Egypt. To the astonishment of decision-makers and commentators alike, the protests could not be suppressed by the long-standing autocratic leaders in both countries, culminating in the downfall of both Ben Ali of Tunisia and Mubarak of Egypt within the span of a month. Not long after, the protests spread across many parts of the Middle East and North Africa (MENA) in another unforeseen development. Three years later, a chain of events that had erupted after the Euromaidan protests in Ukraine in 2013 evolved into a full-blown violent conflict after Russia violated Ukraine’s sovereignty and territorial integrity by annexing Crimea in March 2014. This move caught both experts and decision-makers in the West by surprise, with Putin openly admitting in April that Russian servicemen had indeed backed the ‘little green men’ in Crimea, fighters without military insignia that had initially caused confusion in the West. Weeks later, in June 2014, the so-called ‘Islamic State of Iraq and al-Sham’ (ISIS) seized Mosul and its international airport while Iraqi security forces failed to counter the offensive and withdrew. The collapse of the Iraqi army at Mosul and the fall of the city to ISIS surprised many expert observers and reportedly even ISIS. Similarly unforeseen was the group’s rapid expansion beyond Mosul, which cemented ISIS’s rise as a powerful and destructive actor in Iraq and Syria.

Each of these events represented a moment of ‘peak surprise’ for Western professional analysts and decision-makers in three partly overlapping crises erupting in the European neighbourhood in the first half of the 2010s. In the aftermath, intelligence communities and policymakers were accused of failing to anticipate, warn, listen or prepare for these eventualities. In response, some intelligence professionals claimed that some of these events had been ‘inherently […] unpredictable’ because ‘there were no sort of secrets there which could have told us they were going to happen’ as the British Chief of SIS (MI6) argued in relation to the Arab uprisings. Strategic documents and reviews issued in Washington, London, Berlin or Brussels in subsequent years painted the picture of a new era of uncertainty.

Introduction

The core premise of this book was that we can learn more about estimative intelligence and the prospects for anticipatory foreign policy through a double comparison of cases and actors rather than single actor case studies. What are the distinctive features of each of these cases so that we can avoid misapplying potential lessons learned to future crises that may only superficially look similar? Or is it possible to discern also common and potentially novel challenges across all three of these quite different cases? Might such challenges constitute more enduring characteristics of future threats and opportunities in foreign affairs? To what extent are these characteristics novel, quasi-structural features of contemporary security threats that together constitute an era of surprise alluded to in the title of the book? Or conversely, are most of the diagnostic challenges in these recent cases well known from previous case studies and enquiries going back to the Cold War era, with the real problem being the inability to remember and update previous lessons? Which aspects of intelligence production and use are the most challenging for all three polities at the heart of our study? And what does this tell us about the most important lessons yet to be learned, the failures of previous attempts to reform and internalise lessons, or indeed the nature of common new challenges facing the three polities? Alternatively, can we discern significant differences between these three polities in how they handled some of the challenges in estimative intelligence production and use? If so, are any weaknesses and strengths identified unique to these polities given the way they organise, resource and target their intelligence and foreign policy? Or could practitioners in Brussels, Berlin and London benefit from learning innovative lessons from each other or mitigate each other’s weaknesses through closer collaboration?

This chapter seeks to answer these questions by drawing on the evidence and arguments presented throughout this volume and in closely related publications of the underlying INTEL research project, including the case timelines based on open sources. It builds on the theoretical discussion in Chapter 1, in which we have outlined the normative model of anticipatory foreign policy, the taxonomy of surprise, the overview of performance criteria together with mitigating or aggravating factors, and the discussion of underlying problems and challenges of organisational learning.

Introduction

The annexation of the Crimean Peninsula by Russia in March 2014 was a strategic surprise for the EU of the most negative kind, not least as it was followed by Russia’s increasingly direct military interference in eastern Ukraine that sparked open warfare with thousands of people being killed. The events marked a fundamental reassessment for the EU of the threat Russia posed not just to its immediate neighbours, but also to current EU member states with substantial Russian minorities and a history of being occupied or controlled by the Soviet Union. For the wider EU, it removed the already rather minimal basis for cooperative relations with Russia as fundamental international laws and diplomatic norms were broken in the most blatant way and it raised concern over whether the EU’s approach to the region was still fit for purpose. When taken together with other Russian coercive and aggressive actions vis-à-vis some EU member states, it signalled for many observers nothing less than an undeclared new ‘cold war’. In contrast to the Georgian-Russian seven-day-war of 2008, the Ukraine-Russia conflict of 2014 triggered a number of post-mortem studies by governments or parliamentary committees, individual academics, think tanks and, of course, journalists to assess the performance of ‘the West’, the EU as a whole, or specific EU member states, and what lessons they should take away from this crisis. The overwhelming majority of these reviews focus on decisions and policies vis-à-vis Russia and Ukraine. The question of intelligence is typically treated as part of the wider question of whether ‘the West’, ‘the EU’ or particular states should have been surprised by what happened and whether or not these actors bear some responsibility for causing these Russian actions. Those who are more critical of the EU’s policies vis-à-vis Ukraine and Russia also tend to claim that Russian military actions should not have come as a surprise. Other scholars claim that the EU at least played a causal role and should shoulder some responsibility for the crisis, while Richard Sakwa is even more critical of EU and NATO.

Introduction

What can we realistically expect from estimative intelligence and anticipatory foreign policy? What are appropriate yardsticks to use when retrospectively assessing the performance of governmental and external analysts, policy planners and decision-makers? To what extent and when is being surprised to be expected and excused? When do performance shortcomings point to underlying issues that might be avoided or addressed through learning the right lessons without creating great problems elsewhere? And how can obstacles to lesson learning and remembering in intelligence and foreign policy be overcome? This chapter will try to engage with these questions as it sets out a common conceptual and theoretical framework for the post-mortems analysis and identification of lessons to be learned in this volume.

Even though the intelligence studies literature has engaged with some of these questions, there is no suitable framework to take off the shelf that provides a persuasive normative grounding and one that works for the three European polities at the heart of our study – rather than the frequently studied US context. This chapter draws not only on the relevant literature in the core areas of strategic surprise and post-mortems in intelligence studies and foreign policy, but also considers insights from foresight and forecasting studies, crisis, risks and emergency management, and public administration about the role of experts, expertise and learning. We first develop a normative model of evidence-sensitive anticipatory foreign policy within which we situate intelligence and political receptivity to it. A second section looks at the specific challenges for estimative intelligence when seeking to minimise surprise in foreign affairs. We provide a taxonomy of different degrees and types of surprise and discuss when being surprised might be condonable or expected. Thirdly, we investigate how to identify the most important causes of any performance problems in intelligence-policy nexus. Finally, we look at the specific challenge of identifying and learning the right lessons and how to prioritise among recommendations for change and reform.

Normative Expectations towards Intelligence in Anticipatory Foreign Policy

It is helpful to reflect on whether and to what extent we can learn from the role of experts and expertise in fields beyond foreign policy, such as migration policy or public health.

The chapters by Ben Rosamond and William Outhwaite stimulate new thinking about the causes, manifestations and trajectories of de-Europeanisation, differentiation and disintegration in Europe in different ways. While Rosamond focuses on what we can and cannot learn from (neofunctionalist) integration theories about the manifestations and causes of disintegration, Outhwaite draws our attention to the significant differences in how (member) states have related to Europe while also discussing key pathologies and problems of the Union’s constitutional order. Outhwaite’s piece returns frequently to the British case and the country’s elite’s relations to Europe, whereas Rosamond is more interested in exploring larger forces at play in the systemic weakening of the ‘democratic capitalism compact’ and ‘permanent austerity’. While neither of the pieces claims explicitly to advance recommendations on how to best respond to the increasing contestation of the European order, Rosamond appears to agree that more of the same kind of Europe should NOT be the answer. In contrast, Outhwaite expresses his worry that disintegration of the centre might means a resurgence of nationalism and xenophobia, being particularly critical of attempts to undermine liberal democracy in Poland and Hungary. He sees the consolidation of a strong eurozone core with a more loosely integrated circle of countries around it as possibly the most realistic option for stemming these forces. In the following I will present some, necessarily selective, responses to and excurses beyond the arguments presented by both authors.

Rosamond rightly highlights that neofunctionalist writing in the 1960s and 1970s did engage with questions of disintegration (output failure, spill-back, retraction) as well as some of the necessary scope conditions for integration to occur when comparing Europe with other regions, most notably Latin America. Neofunctionalism, despite being famously declared as obsolete by Ernst Haas during a period of stagnation and paralysis in integration, is still helpful for understanding how decisions to pool or delegate authority in one specific policy area can create strong functionalist pressures to go further in response to an unexpected problem. In this conception, problem-solving crises rooted in design and coordination flaws are likely to be addressed by more rather than less integration – as indeed we have seen in responses to the sovereign debt crisis in Europe.

This chapter focuses on media organisations and journalists as sources of outside-in warnings of conflict. Conflict prevention scholars have touched on media’s role in preventive action, but not studied it in any depth. International political communication scholarship offers few relevant insights about the media as warners, as it concentrates on their role during crises or in the build-up to military interventions and wars. This chapter fills this gap by presenting original empirical research on how and when journalists communicate warnings and under what conditions these warnings may gain traction with officials and decision-makers. Our investigation is theoretically grounded in the persuasion framework of Chapter 2, but we also draw on insights from studies about the media as communicators that are in many ways distinct from other outside-in warners such as NGOs. There are three parallel research lines: (1) textual analysis of media coverage showing how news outlets have warned about eight different crises, (2) analyst surveys, and (3) interviews with foreign correspondents and capital-based foreign news editors, officials and NGO staff. Newspapers were selected according to variation across reach, political orientation and nationality; are from countries committed to conflict prevention (US, UK, France, Germany); and cover the whole political spectrum.

The chapter examines warnings relating to violent conflict and massive humanitarian crisis in Darfur, a region of Sudan. As a case of successful ‘crisis warning’ the 2004 Darfur crisis offers important lessons about persuasiveness, especially with regard to the role of senior officials as the chapter focuses on Andrew Natsios, then the administrator of the US Agency for International Development, and Mukesh Kapila, the UN Resident and Humanitarian Coordinator in Sudan. Drawing on extensive original research the chapter shows why these two officials had a variable persuasive impact over time and with different target organisations. The findings suggest that warner capacity and credibility are necessary if not always sufficient for achieving notice and acceptance. Receptivity factors played a facilitating role as they created an environment which was conducive for the warnings being accepted and which reinforced well-tailored warning messages. At the same time, the case shows that warning impact may depend on repeated attempts to get the message across, creativity in exploring alternative channels and routes, and a readiness among sources to take some career risks in order to achieve their intended goal eventually.

The chapter identifies warnings within the EU foreign policy system related to the Russian annexation of Crimea and the destabilisation of Eastern Ukraine in 2014 and examines the reasons for the differentiated impact among EU decision-makers. The Ukraine case was a strategic surprise to the overwhelming majority of EU policy-makers and posed significant diagnostic problems, but senior ministers within the system were warning their peers on the basis of previous cases, historical analogies, expert knowledge and recent indications of changed Russian intent. However, the persuasiveness of the few and late warnings suffered, first of all, from distraction by external crises coupled with limited bandwidth in the EU, shortages in relevant and reliable intelligence, high resistance against divergent policy predispositions of the warning and, finally, significant suspicions of national biases against key warning sources. The supranational part of the system was almost completely blindsided to geopolitical risk, while the newly established EEAS was partly unable and partly unwilling to persuade the Commission. It is fair to say that the system as a whole acted as an impediment to having a holistic discussion of warning as part of foreign policy discourse.

Rwanda has become the paradigmatic case for a missed opportunity to prevent genocide and human suffering on the largest scale, and subsequent analyses of this case have shaped a significant part of the practical and academic thinking. The chapter contests the dominant argument that Rwanda was a straightforward case for heeding plenty, early and high-quality warnings. It argues that indications need to be distinguished from actual warnings and persuasiveness of warnings needs to be empirically studied rather than assumed. The analysis shows that most sources cited in the literature did not contain an actual warning and gave a more ambiguous picture than is claimed by proponents who argue that lack of political will, not warnings, was the problem. It is suggested that hindsight bias partly explains why the availability of warnings has been overestimated, whereas the diagnostic difficulties in this case were underestimated. Contrary to expectations, persuasive warning communication appears to be no less of a problem for preventive policy as the will and ability to respond. The findings suggest that renewed attention is needed to the challenge of making knowledge, relevance and action claims about impending mass atrocities that are clear and persuasive enough.

This chapter looks at the persuasion strategies and successes of INGOs. Building on the theoretical framework of Chapter 2, we examine to what extent the most capable INGOs engage in conflict warning (warner capacity), when and how they communicate warnings (communication and message factors) and how they are perceived by key recipients in media and policy-making (credibility and persuasiveness). Our investigation centres on Amnesty International, Human Rights Watch and the International Crisis Group. These INGOs are atypical insofar as they belong to a small group of international ‘advocacy superpowers’. We draw on three sources of data: first, interviews with 154 practitioners from international and local NGOs and think-tanks plus a further 127 interviews with foreign affairs journalists and government/IO officials in the field of conflict prevention; second, a text analysis of the public warning output of these NGOs across six different conflict cases; and third, a recipient survey among analysts and officials active in different organisational contexts, most notably the EU, the United Kingdom and the UN.

The chapter introduces the research puzzle of the study: under what conditions do warnings about impending violent conflict in other states succeed in persuading foreign policy-makers to pay more attention, shift their attitudes and mobilise for preventive or mitigating action? Why are some warnings by some sources noticed and largely accepted, while others are ignored, disbelieved or simply not acted upon? The introduction reviews briefly the literature on the warning–response gap in intelligence and peace studies and makes the case for recasting the problem as one of persuasion. It defines and conceptualises what warnings are, their different claims and varied impact. It argues that only by explaining differences in persuasiveness can prospective warners learn the right lessons. To tackle the questions, the chapter introduces the comparative and longitudinal research design, the methods used and data gathered. The chapter concludes by outlining briefly the structure of the book.

This chapter develops the conceptual basis and theoretical framework for the study, relating to relevant literatures in intelligence studies, peace studies and constructivist theories in IR. It discusses normative versus analytical approaches to warnings in foreign affairs and explicates problematic assumptions in the literature. It argues that warnings come in various forms and may contain a variable mix of knowledge, relevance and action claims. They can differ greatly in their senders’ intentions and relationships to recipients. Therefore, warning impact needs to be measured in a more nuanced way than whether warnings led to effective preventive action. The chapter develops a theoretical framework that draws on the most relevant insights from literatures in social psychology, advocacy and political communication to explain when warnings are persuasive. It distinguishes between outside-in warnings from NGOs and news media and inside-up warnings through intelligence analysts and diplomats. The framework elaborates which variables can best account for differences in persuasive success and how they may interact.

The concluding chapter draws on findings from preceding chapters to return to the original question: What makes warnings about conflicts in foreign countries persuasive? It argues that existing research in conflict prevention has tended to over-count warnings and exaggerate their persuasiveness. Most analytical products dealing with the political situation in foreign countries did not clearly announce themselves as warnings, or their warning content was highly hedged, opaque or hidden within conventional reporting or current intelligence. At the same time, inside-up warnings are influenced by outside-in warnings and the broader information environment created by media and NGOs. Six ‘first order’ factors are most significant to explain the findings as they shape persuasion paths: diagnostic difficulties specific to each case, the level of political equity at stake in a given country, individual capacities and the credibility of inside-up warners, the degree of engagement of a small group of highly capable and credible outside-in warners, the degree of agenda competition and, finally, the role of pre-existing policy biases. The chapter examines what these findings mean for prospective warners, which questions they should ask themselves, and how they can increase the chances of their warnings being listened to and, perhaps, even heeded.

This chapter focuses on the case of the Georgian-Russian War of 2008. It concentrates on the question of which warnings were communicated by whom within the European Union system and how were they perceived in terms of notice, acceptance and mobilisation to act by the EU’s member states as the key decision-makers. It considers not only warnings in relation to South Ossetia, but also those from a few months earlier related to Georgia’s other separatist region of Abkhazia that were in fact to some extent successful. Rather than a classic case of warning failure, the analysis shows considerable variation in impact among member states and between the cases of Abkhazia as compared with South Ossetia. It is argued that these differences are due to the relatively low level of diagnostic difficulty in making knowledge claims, differences in receptivity among member states due to threat perceptions and policy preferences, and source credibility given warners’ respective track records as well as suspicions of national biases. The case confirms that many factors work interdependently in producing different outcomes for warnings over time and highlights that warnings should be considered an integral part of evolving and historically contingent discourses on foreign policy.