Reading the International Crisis Group or why think tank reports have to be taken with a pinch of salt, by Berit Bliesemann de Guevara

Hillary Clinton at ICG Annual Award ceremony (Picture borrowed from

Hillary Clinton at ICG Annual Award ceremony (Picture borrowed from

Recently the New York Times caused turmoil among prestigious and influential US think tanks when it published an investigative article about concealed connections between these non-profit research organisations and a broad range of foreign countries, among them Norway, Japan, Saudi Arabia and Qatar (NYT online, 6/9/2014). The article revealed that leading think tanks like the Center for Global Development, the Brookings Institution and the Atlantic Council are receiving funding from foreign governments in exchange for influence on the organisations’ recommendations to US policy-makers. The article criticized this practice for its lack of transparency and the supposed loss of intellectual freedom and objectivity. ‘The money is increasingly transforming the once-staid think-tank world into a muscular arm of foreign governments’ lobbying in Washington’, the authors cautioned. The attacked think tanks rejected the accusations, pointing to ‘credibility’ and ‘scholarly independence’ as their major ‘currency’, although one interviewee admitted that self-censorship could be a future problem because in times of dwindling funding sources, saving one’s job could well trump the urge to be critical.

The article is captivating because it provides evidence for practices many had suspected existed. Yet its critique is somewhat artificial. The authors base their arguments on the assumptions that think tanks can be truly independent and that their products – policy analysis and advice – can be objective and free from influence. Actually, there are very good reasons to always be cautious when it comes to think tank reports. The mechanisms at work, however, which put into question independence and objectivity, are much subtler and much more commonplace. While there is a healthy reflex to assume that money flows equal dependency, the reality is often much more complex and money transfers as such are not a sufficient indicator for influence on policy recommendations, let alone outcomes.[1]

Imagine the think tank landscape as a marketplace of ideas, in which a major concern for policy advisors is to have access to, and influence on, policy-makers. This simple aim already shapes the type of knowledge they produce. In order to become an influential player, the knowledge offered needs to be ‘marketable’, that is, politically feasible, compatible with current policy debates and political colours, and reflecting the zeitgeist. It is for this reason, among others, that think tank reports always have to be taken with a pinch of salt, as they never present ‘bare facts’ and ‘untainted analysis’.

In order to develop my argument further, I will draw on recent research by an international network of colleagues who have explored in detail one of the most notable and widely referenced producers of knowledge about conflict areas used extensively by policy-makers, the media and academics: the International Crisis Group (ICG).[2] The ICG describes itself on its website as ‘an independent, non-profit, non-governmental organisation committed to preventing and resolving deadly conflict’. It currently covers ‘some 70 areas of actual or potential conflict’.

Reading instructions for International Crisis Group reports

If you are:

  • a policy-maker concerned with questions of violent conflict and intervention in Africa, Asia, Europe, the Middle East or Latin America;
  • a so-called practitioner in an international peacebuilding or development intervention;
  • a journalist or dedicated news consumer keen to stay on top of the world’s war-related events;
  • or a conflict researcher sponging up every piece of information about your research region, …

…then you have very likely come across ICG reports, briefings, CrisisWatch bulletins or op-eds. Yes, we all read them, and depending on which conflict we are interested in, we may even have little or no informational choice when it comes to sources of field-based analysis.

In general, ICG reports are known to be timely, detailed, useful and above all based on the expertise of analysts who reside in the conflict regions they write about and have access to sources ‘on the ground’. As a former ICG analyst put it:

‘ICG presents itself as unlike “armchair” think tanks in DC and other Western capitals by way of its presence in “the field”. [T]his needs to be emphasized as it leads (policy) audiences to attribute (rightly or wrongly) much more authority to ICG’s reports than to others’.’[3]

The ICG also claims to be highly successful in its policy advice, and long lists of lauding quotations from world leaders seem to confirm this. The organisation attributes its influence on policy-makers to ‘key roles being played by senior staff highly experienced in government and by an active Board of Trustees’, whose composition of former high-level statespersons and other influential personalities resembles a ‘who’s who of influential power brokers’ in international politics, as a praising 2005 Time Asia article described it. As a crucial part of the ‘ICG approach’, advocacy offices in political centres such as Brussels, New York and Washington D.C. ‘translate’ field findings into practical policy advice and lobby the highest echelons of the policy world, thereby cutting out the information-distorting effects of the usual middlemen. According to the organisation’s public self-evaluation, this methodology works a treat. The ICG’s annual reports are replete with ‘evidence’ of its impact on conflict-related policy-making by national governments and international organisations. And its own history-writing, especially its anniversary brochure ‘15 Years on the Front Lines’, represents the think tank’s career as a story of rise and success – not least thanks to its dedicated staff striving to make a difference on behalf of the civilian victims of violent conflict.[4]

So why are ICG reports problematic? In the end, aren’t we all – policy-makers, practitioners, journalists and academics alike – reading and using them?

Four reasons why ICG reports[5] are to be read with caution, and a caveat

  1. The ‘ICG approach’: what do we really know?

The first reason why we should be cautious when relying on ICG analyses is the simple fact that we know so little about the organisation and its work, its self-description apart. The organisation does not even reveal the names of reports’ authors – mainly for public branding reasons and because reports are seen as a collective product that goes through an intensive internal discussion process, an ICG advocacy officer explained to me. The ICG describes its field presence in terms, which imply the possibility of an independent outsider position for analysts looking at clearly identifiable problems, which then crystallize in unambiguous reports:

‘Our analysts are based in or near many of the world’ s trouble spots, where there is concern about the possible outbreak of conflict, its escalation or recurrence. Their main task is to find out what is happening and why. They identify the underlying political, social and economic factors creating conditions for conflict, as well as the more immediate causes of tension. They find the people who matter and discover what or who influences them.’

The reality in the field is not that easy. A former ICG field analyst described how she became both the target of other actors’ versions of the conflict and peace process story and a mediator between different stories:

‘In the process [… ] I was accused of being close to people on the whole spectrum – from the [ethnic] rebels to the president of [the country], the whole spectrum of positions [… ] They instrumentalise. But at the same time they keep talking to me, because [… ] all the people appreciate the fact that I am faithful to it [… ] They knew also that I had access to the other side, to all sides, so every party would talk to me [… ] it was in their interest also to talk to me.’[6]

Fact-finding on the ground is not straightforward. What we as readers do not learn is how information in a report was gathered, selected and interpreted, who took part and decided in the process from report drafting to final product to policy recommendations, and which quality controls, if any, existed.[7]

Second, we also know very little about the ICG’s role in international politics, about how it tries to ‘impact’ on political perceptions, processes and outcomes, and how it might be manipulated or used by other actors. Again, the ‘ICG approach’ seems easy and straightforward:

‘All too often the missing ingredient is the “political will” to take the necessary action. Crisis Group’ s task is not to lament its absence but to work out how to mobilise it. That means persuading policy-makers directly or through others who influence them, not least the media. That in turn means having the right arguments: moral, political, legal and financial. And it means having the ability to effectively deploy those arguments, with people of the right credibility and capacity.’

Yet this description leaves many questions unanswered. If the ICG’s access to policy-makers depends on its staff’s connections, then how does the organisation ensure its independence? In other words, what formal and informal relations exist between ICG experts, local stakeholders and international decision-makers? What different roles do the organisation or its representatives play in conflicts and peacebuilding processes, for instance, and what happens during advocacy ‘behind closed doors’?

The description of the ‘ICG approach’ is so vague that it leaves most of these questions unanswered. For some strange reason, however, ‘Know your sources!’ – the imperative of text analysis that we all learned in history class – does often not seem to apply to the use of ICG knowledge.

  1. The ICG’s conflict analysis: one story among others

The second observation that should caution us against an uncritical reading of ICG analyses is that the ICG’s interpretation of a conflict, a crisis situation or an intervention is just one story among many possible others[8] – and not always the most comprehensive or differentiated one, not least because the format of the report (its size and scale, for instance) sets limits to analysis. In the words of Nikolas Kosmatopoulos,

‘the report presents itself as an assemblage of a series of technical characteristics that help to shrink the world and make it fit the model format of the crisis expert.’

Socio-political problems, and especially those in protracted conflict contexts, do not just ‘reveal their nature’ to the beholder – such ‘wicked problems’ are always rendered manageable in the policy process through means of interpretation, framing and labelling and thus put into a certain drawer, which contains a limited range of legitimate tools for their solution. Storytelling is a crucial part of determining a conflict’s causes and courses, and dominant stories can become highly efficacious.

Seen through this lens, the ICG is in the business of creating and disseminating powerful stories. A central question for readers should therefore be in how far ICG reports are shaping perceptions of conflicts and legitimate solutions in specific ways – and what plausible alternative stories are out there.

In his research on ICG reporting about the Democratic Republic of Congo (DRC), Kai Koddenbrock shows that ‘the logics of government and the dilemmas of rule in a country with the size, geography and history of the DRC receive hardly any attention in ICG reporting’. Based on his analysis of the manifold dilemmas of rule in the DR Congo he provides an alternative reading, namely that:

‘President Joseph Kabila has in fact responded skilfully to the dilemmas of elite inclusion across the different hubs of power and wealth from the Kivus to Katanga to the capital Kinshasa. While his political and human rights records are by no means impeccable, not all is rotten in the state of Congo, and the Kabila government deserves more analytical rigor and openness than is offered by the pathologising modes of analysis used by the ICG.’

Koddenbrock’s alternative story also suggests tongue-in-cheek that, in his urge to stay in office, Kabila resembles Germany’s Angela Merkel more than commonly thought.[9]

A similarly reductionist and pathologising story, which omits the long-term conflict history in the Mano River Basin, is what Morten Bøås criticises in his article on early ICG reporting on the wars in Liberia and Sierra Leone. Not only did the first ICG reports on these wars plug into the simplifying ‘greed and grievance’ explanation of so-called ‘new wars’, according to which the post-1990 wars were waged over nothing but economic resources, in this case ‘blood diamonds’. The reports also contributed to a personalisation of the ‘root causes of war’ by blaming them mainly on the figures of president Charles Taylor, described as ‘brutal and unscrupulous’, and rebel leader Fodoh Sankoy, who is characterised as ‘psychopath’. Bøås argues that:

‘Reducing this complex history to a narrative of greed, criminal intent and “bad men”, however, is not a good way to “hunt ghosts of a difficult past”, as it leads to policy approaches that gloss over difficult historical trajectories and challenging structural issues. Instead of helping to create the vantage point of a break with past practices, the only thing that this achieves is the creation of “new ghosts of the past”, who sooner or later will come back to haunt Liberia and Sierra Leone until those hard and painful issues are handled internally.’

I could add further examples from Bosnia-Herzegovina, Indonesia, Lebanon, Mexico and Uganda,[10] but I think for the purpose of my argument I can leave it at this. If there is one central lesson here, it is that the ICG tells one conflict story among many other possible ones, and while it may have good access to local stakeholders, its story is therefore not necessarily ‘truer’ than others. As the authors quoted above have suggested, telling a different story based on similar evidence may well show ‘problems’, and for that matter ‘solutions’, in a very different light.

  1. The ICG’s policy advice: a western/liberal mantra

My third point of critique concerns ICG’s policy recommendations. Of course I am not the first to moan about how simplifying and formulaic this part of ICG reports is – having a rant about the executive summary/recommendations part of ICG reports is common courtesy among conflict researchers. So why are they problematic?

Sonja Grigat has explored the rhetoric of ICG executive summaries in the case of Indonesia, where ‘the ICG mantra-like recommends measures to reform the security sector, notably the police’, no matter which issue it has been reporting about over the past 15 years. Her explanation for this finding is that:

‘ICG reporting fulfils a function that transcends the immediate contribution to preventing and resolving violent conflicts. ICG publications essentially aim at discursively disciplining their audience through practices and procedures characteristic of liberal governance into this specific form of social action and corresponding mind-sets, thus perpetuating liberalism as the global ‘regime of power’.

In other words, the ICG is engaging in what could be termed a ‘mass-scale education programme for unruly societies’ based on the liberal norms of democracy, rule of law and market economy. This may not seem like a problem as such, were it not for the depoliticising and criminalising effects on violent behaviour of this educating rhetoric:

‘By building community around the liberal governance programme and its underlying value of non-violence under recourse of epistemic and moral authority, legitimate opposition is largely eliminated. Second, [this rhetoric] contributes to obscuring the historically specific conditions and complex social circumstances which actually led to the events depicted in the ICG’s reports in the first place. […] [T]he ICG claims the role of a teacher, transgressing not only the borders of science and politics, but also those of science and domination […]. The long-term effect of this approach is the legitimisation of a restrictive understanding of violent political conflicts, which above all contributes to a non-reflective (re-)production of liberal forms of governance.’

Exploring the ICG’s recent ‘Mexican turn’, Markus Hochmüller and Markus Michel-Müller have also shown how western-centric and technocratic ICG reporting about Mexico is. Looking at its knowledge production about drug trafficking, Mexican statehood and indigenous justice, the authors come to the conclusion that,

‘[…] the ICG’s apolitical crisis knowledge production ignores idiosyncratic features of the Mexican political context, including transnational entanglements, local power dynamics, and centre–periphery relations and their historical legacies. While we sympathise with the ICG’s overall interest in making Mexico a safer and more secure place, on the basis of our findings we would nonetheless conclude that, because of its Western-centric assumptions and apolitical perspectives, the group’s policy proposals provide minimal solutions that are sustainable in this regard.’

The naturalisation of certain ‘crisis’ imaginaries and subjectivities is what Nikolas Kosmatopoulos’s contribution on the ICG in Lebanon is interested in. He concludes that ‘crisis report’ writing is a form of techno-politics, which creates a binary ‘imaginary geography’ that distinguishes the field of crisis (the Lebanon) from that of order and decision (the western capital/bureau):

‘[…] scale-related techno-politics attach the ‘field’ […] to the space of crisis, while they attach the bureau – with the accompanying element of the advocacy manager, etc – to the space of decision. […]

[T]echno-politics attaches the urgent sense of the unexpected to the space of crisis, while they treat the space of decision as a structured and organised system of Realpolitik. […]

In the space of crisis rebel entities are placed under overt or covert forms of surveillance on the grounds of universal values and priorities. In the space of decision, on the other hand, the observer’s subjectivity is merged and produced as part of a quasi-scientific system of alert-and-response to international threats, which increasingly resemble natural phenomena such as epidemics, volcanic eruptions, or extreme meteorological events.

The new ontology, then, is successful because through it crisis bears a resemblance to a quasi-natural phenomenon.’

Taken together, these different findings should caution us against simply buying into the interpretations and ‘solutions’ that ICG reports offer us about a conflict area. There is more to the policy recommendations that a mere look at the surface would suggest. What lies underneath is a powerful image of the world that perpetuates simplifying and denigrating ideas of the ‘good West’ and the ‘unruly rest’ and diverts attention from the structural and active role ‘the West’ plays in areas of violent conflict.

  1. ICG staff: the functions and power of flex-nets

The fourth reason to treat ICG reports with caution has to do with the socio-political networks ICG staff are part of. True, their belonging to professional networks, established during work at other organisations such as the UN, governmental agencies, media outlets or other NGOs, provides the ICG with the connections needed to carry out its advocacy work. At the same time, however, these ‘flex-nets’, in which individuals (the ‘flexians’) shift roles on a frequent basis, can be a severe cause for concern, as Roland Kostić shows in his piece on the ICG’s role in struggles among different intervention agencies in Bosnia and Herzegovina around the years 2000-01. His findings are surprising, to say the least, and put the content of the reports produced during this time in a completely different light:

‘[…] peace-building interventions seem to be true battlefields of ideas. […] [I]n order to understand a specific incidence of knowledge production and policy framing, one needs to understand the broader network beyond the institutional boundaries of the think-tank. In this case, though not visible to an ordinary observer, the ICG’s work in the early 2000s in BiH was seemingly part of a broader knowledge production flex-network united by a common effort to promote the position of the US Department of State. It seems to have consisted of US military and intelligence representatives […], US diplomats […], and [the] ICG Balkans director […]. Combining access to privileged micro-level information, analysis and internal policy debates among internationals allowed the ICG-linked flexians to cut through the international bureaucracy and connect different levels of international policy making.’

What this example shows is that it does not require money flows, as those unveiled by the New York Times article, to thwart the ‘independence’ and ‘objectivity’ of think tank work. Networks can do the same, while at the same time contributing to the think tank’s immense success in certain policy circles.

Yet influence does not only come from western sources. Jonathan Fisher shows in the case of think tank knowledge production in/about Uganda why we should not underestimate the agency of governments in the Global South either. Focussing on the Obama administration’s decision to send combat-equipped US forces to Uganda to help ‘hunt down’ notorious LRA rebel leader Joseph Kony, Fisher shows that is was not only the lobbying of western think tanks, but also the Ugandan government’s information management strategies that contributed to this policy decision:

‘Through a combination of diplomacy, the politics of access and information management over several decades […] the Ugandan regime has been able to frame the LRA conflict in a particular way in the minds of many Western policy makers. For many of these individuals this has rendered UPDF-led military solutions putatively more feasible and appropriate than others. Enough [Project] and other groups’ apparent influence over US officials on LRA policy in recent years, therefore, should be seen in this light: influential with regard to policy content but led, overall, by Kampala’s own framing of the crisis and how best to tackle it.’

A caveat: let’s not overstate ICG’s impact

Apart form the last paragraph above, I might have created the impression that the ICG is an omnipotent puppet master, and I therefore feel the need for a brief caveat to end with: the influence of the ICG on policy-making has varied greatly over place and time. Depending on the case and the specific moment in a conflict or peacebuilding process, the ICG has had positive and negative roles; sometimes it has been a crucial actor, while in other places or at other times its role has been negligible.

What seems especially important for the ‘impact’ is how policy-makers make use of think tank knowledge, and this can happen in different forms. Yes, they may use ICG’s knowledge to shape their views and policy preferences; yet often these are already shaped to a considerable extent, and all a think tank report is used for is to substantiate or legitimate a course of action that was already decided before.[11] As a former ICG analyst put it:

‘[…] ICG’s reports can be viewed as a tool in (western) foreign policy bureaucracies’ internal debates and competition over conflicting policy views.’

Berit Bliesemann de Guevara, PhD, is Senior Lecturer in Peacebuilding and Post-war Reconstruction at the Department of International Politics, Aberystwyth University. Her latest research concerns policy-related knowledge production in conflict and intervention contexts. She thanks the Hanse-Wissenschaftskolleg (HWK) Institute for Advanced Study in Delmenhorst for currently hosting her and providing excellent writing spaces in the library’s ‘sky lounge’ and the BIGSSS at Bremen University for co-supporting her research.

[1] Usually, the more sources a think tank can tap into, the more independent it becomes from single donors’ money and potential influence.

[2] The network’s findings have been published in Third World Quarterly (vol. 35, no. 4, 2014).

[3] Quoted at length in ‘Studying the International Crisis Group’.


[4] By contrast, see my critical discussion of the ICG methodology and history writing as organisational myth.

[5] Not unlike other think tank knowledge, I might add.

[6] Quoted at length in ‘Studying the International Crisis Group’.

[7] See Morten Bøås’s article for a deeper discussion of this aspect.

[8] Including academic interpretations, which are nothing but stories that follow specific ‘writing rules’.

[9] See also Judith Verweijen’s blog post ‘Anatomy of a feeble analysis: a critical reading of Crisis Group’s latest report on the DR Congo’. She comes to the conclusion that,

‘Rather than being based on a nuanced analysis that recognizes the multi-dimensionality  and complexity of conflict dynamics and the production of violence in the eastern DRC, ICG’s presentation of the events in the Ruzizi Plain and the Moyens Plateaux of Uvira is moulded around a single narrative-that of “local ethnic and land conflicts” as the “root causes of violence”.’

[10] For a more general discussion, see Greg Simons article on the manufacturing and communicating of crisis.

[11] See ‘Studying the International Crisis Group’ for a more detailed discussion. This is indeed another point where the NYT article’s argumentation falls down. The simple causation chain “foreign donation—influence on policy report—influence on policy” is not that simple in reality.

This entry was posted in International Crisis Group and tagged , , , , , , , . Bookmark the permalink.

2 Responses to Reading the International Crisis Group or why think tank reports have to be taken with a pinch of salt, by Berit Bliesemann de Guevara

  1. Maybe worth mentioning that ICG is “broadly transparent” about who funds it according to Transparify’s 2014 report:

    That’s more than a lot of other think tanks.

  2. Pingback: Weekly Links | Political Violence @ a Glance

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s