Patrick Meier (PhD)NB: THIS BLOG HAS MOVED TO iREVOLUTION.net
Born and raised in Africa. Director of Crisis Mapping at Ushahidi and co-founder of the International Network of Crisis Mappers.Previously co-directed Harvard's Program on Crisis Mapping and Early Warning. Consulted extensively for international organizations in Africa, Asia and Europe. PhD from The Fletcher School and Doctoral Fellowship from Stanford University's Program on Liberation Technology. MA from Columbia University. BA from York and EAP at UC Berkeley.
[Cross-posted on Ushahidi blog]
CeaseFire began operations 10 years ago in one of Chicago’s most violent neighborhoods. Within a year, the project reduced shootings by 67%. These results have since been replicated dozens of times in Chicago, other US cities and abroad including Iraq. An extensive three-year independent evaluation funded by the US Department of Justice provided statistical proof that CeaseFire’s model was directly responsible for the dramatic decline in killings. Indeed, “this evaluation scientifically-validated CeaseFire’s success in reducing shootings and killings by 41% to 73% and demonstrated a 100% success rate in reducing retaliatory killings in five of the eight communities examined.”
Thanks to PopTech, I was able to meet with the CeaseFire team in Chicago on two occasions this year. This is truly one of the most stunning projects I’ve come across in a long time. Members of the CeaseFire team are true professionals. They prove that preparedness is fundamental to successful tactical early warning and response. Having worked in the field of conflict early warning and rapid response for years, I’ve always known that a people-centered approach to violence prevention would be far more effective than a state-centric approach. CeaseFire demonstrates this beyond the shadow of a doubt. The model itself is based one how diseases are controlled, which is fascinating.
I’m incredibly excited to be part of PeaceTXT, a joint PopTech, CeaseFire, FrontlineSMS:Medic and Ushahidi initiative to explore how technology might help the team leverage and scale their incredible work. From the PopTech website:
The PeaceTXT project brings together some of the world’s best technologists and social innovators to explore how mobile tools and mobile messaging might further accelerate CeaseFire’s ability to engage communities, change social norms, improve its efficacy and find new paths to scale. Our investigation is broad but intensive, exploring and rigorously evaluating several possible applications.
While the initial work will focus on CeaseFire’s efforts in its model sites in Chicago, there is clear potential to apply key insights and methods to the global context. Experience and learning gained from this project is expected to prove invaluable to conflict resolution and violence prevention efforts, nationally and globally.
The project was publicly launched and announced last week at PopTech 2010 thanks to seed funding from the Rita Allen Foundation. My colleague Josh Nesbit and I have already learned a lot from the conversations we had in Chicago, but we know that we’re just scratching the surface. So I’m eager to get back to Chicago and pilot some projects with the CeaseFire team.
Thank you PopTech for bringing us together!
By Tadakazu Kanno (PhD candidate, the Department of War Studies, King’s College London)
Patrick has already written about the impending violence regarding the coming elections in Sudan in this blog (“Sudan: An Early Warning of Impending Conflict”). I also came across a similar article posted on Alert Net, which is titled “West silent over Darfur crisis despite risk of spiraling violence”. In fact, this is not the first time I came across early warnings of election violence. The BBC’s article “Kenyans rearming for 2012 poll” (dated 07/October/2009) warns that “rival ethnic groups in Kenya who fought after the 2007 election are rearming in readiness for violence at the 2012 poll.” The International Crisis Group’s report (dated 01/October/2009) “African Peace-building Agenda: “Nigeria Needs to Prevent another Electoral Debacle” says that “Nigeria and its international friends need to act urgently and concertedly to prevent another electoral disaster in 2011.” The significance of the prevention of election violence is obvious if we look at the 2007 Kenyan post-election violence, which claimed over 1,000 lives and 300,000 people were displaced.
The source of the photo: the BBC (http://news.bbc.co.uk/1/hi/world/africa/8293745.stm)
Actually I wasn’t looking for these early warnings, I happened to find them. So it is possible to infer that there are plenty of early warnings regarding potential collective violence. In the field of early warning and early response, it is commonly accepted that there are plenty of warnings but the problem is that warnings are not followed by preventive actions (so called “warning-response gap”). About 15 years ago, Michael Lund was arguing in his book “Preventing Violent Conflicts” that:
“early warning occurred in several instances but did not precipitate any action. Strong expectations of potential conflict or even the outbreak of violence were not in themselves sufficient to generate preventive action. For example, fighting and grievous human rights violations in northern Somalia were widely reported by human rights organizations as early as 1988 yet provoked no response; International Alert disseminated a fact-finding report and recommendations regarding the Russia-Chechen disputes to the United Nations and other bodies in 1992, but no action followed; in the Danube River dispute between Hunagry and Slovakia, the disputants issued several requests for mediation assistance long before any third party stepped forward (p.81).”
David Nyheim’s report published in 2009, however, argues that :
“the field has evolved significantly since its initial conceptualization, and early warning has been integrated into the policies of many organizations. Today it cannot be said, however, that the international community is in a position to prevent another Rwandan genocide. Conflict early warning faces challenges similar to those it faced 15 years ago…(p.13)”
Are we going to repeat the same mistakes in those elections? There is good news from Burundi, where elections will also be held this year and people fear that violence regarding the upcoming elections may occur. The African Great Lakes Initiative (AGLI) of the Friends Peace Teams set up a community-based early warning and response project to prevent election violence (Burundi IT Election Warning Project). In the project, a wide network of monitors reports the symptoms of the tension between ethnic groups and violent incidents regularly by using mobile phones and they also plan to avert the tension on the ground through interventions. I’m happy that a civil society organisation started to take an initiative in election violence prevention.
But, we must understand that civil society is not an all-around player. Generally speaking, it may be difficult for civil society to prevent post-election violence by using an early warning and early response mechanism due to the following two reasons. (Although we must think about violence during election campaigns, I would like to focus on post-election violence).
- What civil society can utilize for their interventions is dialogue.
- The results of elections are not negotiable.
As you can imagine, “dialogue for something unnegotiable” is not easy. Most of the disputes between conflict parties are negotiable (it is the case especially at a community-level like the use of land), but the results of elections cannot be changed unless there was violation in election campaigns. If it is difficult for civil society to prevent post-election violence, it may be a good idea to work with law enforcement agencies who possess coersive power to stop violence. However, it is questionable that law-enforcement agencies function rapidly and impartially even if early warnings are wired to them. What is even worse is that they are actively involved in violence in some occasions. The International Crisis Group reports that in the 2007 Kenyan election violence, police officers have taken sides and used terror tactics.
Then, how can we enhance the capacity of civil society in early warning and especially in early response (interventions) to prevent election violence? In this regard, the vision of Burundi IT Election Warning Project is to the point:
“An alternative strategy is to address violence on the level at which it is carried out, building trust and relationships at the grassroots level that would weather possible calls for violence coming from political elites. Such relationships have prevented violence in communities in the past, as is evident from the regional variation in levels of violence as well as from the stories of the prevention of violence by local leaders and citizens who refuse to take part and encourage others to do similarly.
Such interventions are only possible, however, where people have been able to reconcile “ethnic” differences and healed the trauma in their hearts that is the basis of Hutu-Tutsi animosities. Otherwise, the “ethnic” division stirred up for political gain in an election falls on receptive ears and ready hands, motivated by frustration and anger or fear.”
What is significant, therefore, is to create an environment where calls for non-violence are heard and accepted by people even when politicians call for violence. This can be a platform for successful community-based early warning and early response (CBEWER) systems like Burundi IT Election Warning Project. It is a long term approach to create the environment by connecting people from different ethnic groups and enhancing inter-ethnic activities. Although it sounds natural, the greater people’s will for peace is, the greater the chance of success in conflict prevention is. CBEWER systems would function more effectively if they are established on the foundation of long term confidence-building approaches.
Burundi IT Election Warning Project is now collecting mobile phones. The collected mobile phones are going to be used by the field monitors in order to report violent incidents on the ground. By donating mobile phones, you can contribute to the prevention of election violence in Burundi. As I just introduced part of the project, if you like to get the whole picture of the project, please visit their website.
The International Studies Review just published a piece by Sean O’Brien entitled “Crisis Early Warning and Decision Support: Contemporary Approaches and Thoughts on Future Research.” Sean outlines the latest attempt by the US military (ie, DARPA) to develop a crisis forecasting tool. This time, the platform is called ICEWS for Integrated Crisis Early Warning System.
O’Brien gives a brief overview of recent efforts in this space including Bueno de Mesquita’s Policon and Senturion forecasting systems, which are said to be 90%+ accurate. That said, O’Brien notes that Mesquita himself acknowledges that “he is not exactly sure how to interpret this accuracy claim since most of the reported assessments he has were not explicit about how accuracy was measured.” Incidentally, I like how this rather important qualifier is buried at the bottom of a footnote.
Some of Policon’s/Senturion’s supposedly accurate predictions had to do with questions like:
- What policy is Egypt likely to adopt towards Israel?
- What is the Philippines likely to do about US bases?
Keep in mind that millions of dollars were spent on these sophisticated systems and yet I can’t help but think that paying some experts on Egypt and the Philippines a few thousand dollars would have more or less accomplished the same task. Apparently, Senturion accurately predicted the deteriorating disposition of Iraqis toward US forces. Really? Shocking, who would have expected Iraqi public opinion to shift? Yes, that was sarcasm.
O’Brien also references a forthcoming study by Ward, Greenhill and Bakke, which “delivers a serious blow to the predominant way in which most conflict models are evaluated using statistical significance.” These include the predictive models developed by Fearon and Latin (2003) and Collier and Hoeffler (2004). Ward et al. show that these models predict few if any civil war cases at a reasonable probability cut off of 50%. In fact, the Fearon and Latin model “does not even appear to generate a probability of greater than 30%.” In sum, Ward et al. conclude that we cannot correctly predict over 90% of the cases with which our models are concerned.
Many of the most interesting, policy-relevant theoretical questions are also the most complex, nonlinear, and highly context-dependent. They demand consideration of hundreds of massively interacting variables that are difficult to measure systematically and at a level of granularity consistent with the theory. In such cases it is at best impractical and at worst impossible to apply standard regression techniques within the context of a Large N study, short of invoking unreasonable, oversimplifying assumptions. This may in part account for contradictory findings in the literature relative to the validity of alternative theoretical claims.
So lets keep in mind that previous “breakthroughs” have since been largely discounted.
ICEWS phase one of three consisted of a competition between different groups to successfully predict events of interest (EoI) on a set of historical data. The most successful team was Lockheed Martin-Advanced Technology Laboratories (LM-ATL) in cooperation with a number of established scholars and industry partners. The team integrated and applied six different conflict modeling systems, including:
- Agent-based models drawn from Barry Silverman’s Factionism and Ian Lustick’s Political Science-Identity (PSI) computational modeling platforms. The latter is created with “agents representing population elements of various ethnic ⁄ political identities organized geographically and in authority structures designed to mirror the society being studied.”
- Logistic regression models developed by Phil Schrodt and Steve Shellman, which use “macro-structural and event data factors commonly analyzed in the academic literature.” Shellman’s approach uses a Bayesian statistics model.
- Geo-spatial network models built by Michael Ward, which uses “structural factors, event counts, and various types of spatial networks—trade ties, people flows, and ‘‘social similarity’’ profiles—that embody potential EOI co-dependencies between proximate countries.”
- “A final model was developed by aggregating the forecasts from the above mentioned models using Bayesian techniques.
I’m particularly interested in the use of Agent-based models (ABM) for conflict analysis. O’Brien references a very interesting project at Virginia Tech which I was unaware of:
Scholars at Virginia Tech have already developed a 100 million agent simulation that includes synthetic versions of many American citizens, and plan to expand to 300 million agents this year (Upson 2008). Each synthetic agent has as many as 163 variables describing age, ethnicity, socio-economic status, gender, and various attitudinal factors. The simulation is used to assess how different types of pandemics could spread across the United States under different scenarios.
Deadly Ethnic Riots in Nigeria: Need for a Good Micro-level Early Warning and Early Response Mechanism
By Tadakazu Kanno (PhD candidate, the department of war studies, King’s College London)
A deadly ethnic riot took place in Nigeria again. According to the BBC , the riot took place near the city of Jos in Nigeria early in the morning on the 7th of March and claimed over 500 people’s lives. Many of the dead in the villages of Zot and Dogo-Nahawa, largely inhabited by Christian members of the Berom community, are reported to be women and children. It is reported that the attacks were reprisal killings for violence in January, when an inter-communal clash between the Hausa Fulani settlers (Muslim nomadic herders) had claimed at least 326 people’s lives and the most of the victims were Hausa-speaking Muslims.
The IRIN report on the 1st of February actually gave an early warning by saying that “if the Nigerian authorities fail to punish those responsible for the latest intercommunal violence, they are only paving the way for further bloodshed”. According to the BBC, there was a proximate warning as well. The governor of Nigeria’s Plateau State had warned the army about reports of suspicious people with weapons hours before they attacked, but they failed to take action.
I’m not an expert of Nigeria and the information I have about the situation of the inter-communal clashes is quite limited (the BBC and IRIN only). So, I may not be qualified to talk about the ethnic riots in Nigeria and my analysis may not be really accurate. But, I guess that one of the reasons for the failure of taking rapid actions to prevent the violence was that “who does what” under an emergent situation was unclear. Without preparation and plan, it is difficult to take the right action immediately. Early warning demonstrates its highest ability when it is systematized and well-connected to responders. I was wondering if the latest riot could have been prevented and many of the lives could have been saved if they have had a good early warning and early response system.
David Nyheim’s report on early warning and early response, which was prepared for OECD, shows that now there is consensus of a “good” early warning system, which is one that: (1) is based “close to the ground” or has strong field-based networks of monitors; (2) uses multiple sources of information and both qualitative/quantitative analytical methods; (3) capitalizes on appropriate communication and information technology; (4) provides regular reports and updates on conflict dynamics to key national and international stakeholders and; (5) has a strong link to responders or response mechanisms.
Suppose they have had a good early warning system which meets the criteria. A wide network of monitoring could have caught some symptoms of the attacks much earlier than the governor came to know them. Then, the right information could have been passed to the right person at the right time to take rapid action for averting the tension. Before the eruption of violence, the tension could have been averted through multi-stakeholder dialogue and the plan to attack might have been shattered. If these things failed, as Casey Barrs states, people may need to flee to protect themselves. As a result of a couple of safety nets, the 500 lives – at least some people’s lives – could have been saved.
Micro-level early warning and early response mechanisms, often called “citizen-based early warning” or “third generation early warning”, are a new wave in the field of conflict early warning. One of their characteristics is that they meet all the criteria of a good early warning and early response system (perhaps, the most notable system was the Foundation for Co-Existence’s early warning and early response system in Sri Lanka which operated from 2003 to 2009). They cover specific areas within a country and attempt to prevent riot-type violence.
As the number of civil wars has been decreasing dramatically since the early 90’s (I don’t mean by this that the traditional macro-level early warning systems are less important. They have their own niche.), more attentions need to be paid to the micro-level early warning and early response mechanism as a tool to prevent riot-type violence. Although I don’t have any good statistics regarding the number of deadly riots (If someone has it, please let me know), we have already witnessed the riot in the Moluccas (Indonesia) which claimed 4000 lives in 1999, that of Gujarat (India) in 2002 which claimed 850 lives, that of Kenya which took place from the end of 2007 to the beginning of 2008 and claimed over 1000 lives and the sporadic ethnic riots in Nigeria.
Although a good early warning and early response system is not a panacea for the prevention of ethnic riots (of course, long-term approaches are needed to address root causes along with early warning and early response), this kind of micro-level early warning and early response mechanism will certainly enhance the capacity to prevent ethnic violence in Nigeria.
Many of us in the field of conflict prevention and early warning have argued that there seldom is a lack of warnings on conflict escalation. The problem, we argue, is often response, not lack of early warning. Cue today’s main news story on BBC:
Ten international aid groups say a 2005 peace deal in Sudan is on the verge of collapse and that the world must act now to prevent conflict.
- Agencies blamed a “lethal cocktail” of rising violence, chronic poverty and political tensions.
- Agencies cite disputes over Sudan’s oil resources, national elections in April and the independence referendum as potential flashpoints.
- The BBC’s James Copnall in the Sudanese capital, Khartoum, says the country is clearly at the start of a highly charged and risky 12 months.
- Sudan’s ambassador to London, Omar Muhammad Siddiq, acknowledged that the situation in South Sudan was “deteriorating”.
I’ll be following the lead up to the April elections and the January 2011 referendum. If violent conflict does erupt, the international community will yet again have failed in the prevention of armed conflict despite clear early warning signals. And will yet again have failed to train local communities about how to get out of harm’s way.
As Öhman notes, “conflict analysis, including early warning, has traditionally not included a women’s rights or gender perspective.” The consequences? “When women are not included in EWS, their opportunities to fully participate in conflict prevention and post-conflict reconstruction are limited and, thus, EWS risk failing to adequately predict or prevent conflict.” Moreover, gender-based indicators can “provide a more complete understanding of the causes of conflict and, as a result, help develop more appropriate responses to mitigating or preventing it.”
Some key excerpts:
- By bringing to light such patterns of structural discrimination, integrating a gender perspective can improve the effectiveness of early warning systems by gathering more specific information and allowing for more detailed and precise analysis. In turn, this can ensure better preparedness and, when necessary, more accurate and measurable responses — as well as preventive mechanisms — that can more directly address some of the underlying causes of a conflict.
- To integrate gender into EWS, both men and women must have the opportunity to report on their security situation. Likewise, the different threats and concerns that impact men and women as a result of any conflict must be taken into account and duly analyzed.
- Existing inequalities and different roles during conflict can lead men and women to perceive threats and vulnerabilities in different ways. The same signs and occurrences will not necessarily elicit the same reaction in both men and women. A UNIFEM EWS programme in the Solomon Islands clearly identified some of the disparities. Men, for example, rated inter-ethnic relations as a high source of tension, while women rated them only as a medium source of tension. This could confirm the suggestion that women were better able to maintain inter-ethnic alliances, even during tense times, while men generated stronger in-group identification.
- Recent developments in the area of EWS are focused on working more with the local community and ensuring its preparedness and resilience, rather than elaborating external actors’ actions. Negotiations are not limited to states but can‚ — and should — be performed by a multiplicity of actors. Ideally, all actors should work together in order to strengthen their collective impact.
- Ushahidi is software allowing organizations to establish EWS using basic information technology. The software allows anyone to report incidents via SMS or the Internet and to receive information by the same means. It offers a wide variety of ways to integrate gender, including by ensuring that both women and men are aware of the systems launched using Ushahidi and know how to use them.
- Women as informants: Conflict-response mechanisms still have a distance to go in developing and applying a broadened understanding of security that does not limit it to military matters or ignore human rights and women’s rights. Such thinking has also defined what constitutes security information and who provides it. This has often restricted women from reporting such information, especially women who live in remote areas. Yet the perspectives of women are invaluable — not only because without them monitors will not capture the entire picture of the situation, but also because women are often socialized differently than men, and thus may experience conflict very differently. As a consequence, women tend to ask different questions and focus on different aspects of the conflict cycle and, therefore, can provide different and enriching perspectives to early warning and conflict prevention. Men’s and women’s contributions complement each other.
- Men and women experience security differently and their needs for feeling safe are different. Location is one key factor; women experience insecurity in different places than men. Knowing and understanding these differences and patterns of discrimination is important when elaborating a response to EWS data in order to be sure that the response does not inadvertently reinforce existing inequalities and that it reaches all rights-holders equally.
- Because threats that are regarded not as public but as private matters (including sexual violence and domestic violence) are often perceived and treated as non-political matters, there is a risk that EWS reporting disproportionally highlights security threats that affect men. Gender analysis in EWS, however, can raise factors related to the violations of women’s rights that cause conflict or are direct results thereof. The lower status that most women generally hold relative to most men may cause them to be among the first to experience the weakening of security levels. Thus, their experiences can potentially serve as foreshadowing of more widespread armed conflict.
- When states move to limit the equal rights of men and women to the enjoyment of economic, social and cultural rights, exacerbating the discrimination women might already be facing, this can serve as an important factor to indicate increased instability. It is therefore essential that a broad range of indicators, which also cover ESC rights, are developed as a part of the EWS. In addition, a more balanced team has a greater chance of ensuring that a more diverse set of indicators is developed to monitor and address rights violations.
The OSCE will follow up with a second EWS brief that will address in greater detail how to formulate indicators that incorporate gender. I very much look forward to reading that brief as well. Both address as aspect of conflict early warning that is systematically ignored.