Tag Archives: conflict forecasting

Sri Lanka: Citizen-based Early Warning and Response

Colleagues at the Foundation for Coexistence (FCE) in Sri Lanka just shared their report on “Citizen-based Early Warning and Response” with me (PDF). I’ve been following the Foundation’s work for the past five years so I was keen to get an update on their work in the field of conflict early warning and rapid response.

What follows is a brief review of the report and the FCE’s conflict early warning and rapid response initiative. I conclude with some of my own thoughts based on my early warning experience with FAST, CEWARN, ECOWARN, WANEP, MARRAC, EC, OSCE, OECD, UNDP, UNEP, OCHA, UNICEF,  WFP, USAID, IFES, BELUN, ICG, JRC, International Alert  and Ushahidi.

Introduction

The Foundation takes a human security approach to early warning, which focuses both on protection and empowerment. They note that in most standard definitions of early warning, e.g., “the systematic collection and analysis of information,” do not actually include giving a warning—a point which certainly resonates with my experience.

Evolving Generations

One of the conceptual innovations that FCE contributed to the field of conflict early warning is the notion of first, second and third generations early warning systems. A first generation system monitors and analyzes conflict from outside the conflict regions; they are typically based in the West. “The problems of the first generation are consistent with those of quantitative approaches; they use limited secondary sources which do not provide any certainty about their accuracy and they have difficulty in predicting eruption of armed conflict accurately.” In other words, they focus exclusively on prediction and “do not have effective procedures to communicate with [Track 1] decision-makers for early response.”

Second generation early warning systems conduct monitoring within conflict countries and regions. “However, analysis is still conducted outside conflict countries (in the West).” Second generations systems entail field-based monitoring, risk assessments and active lobbying. “The advantage of qualitative approaches is that [they offer] vastly more content-rich and contextual information than quantitative statistical analysis.” However, the actors engaged in “second generation” early response are no different from those of the first, they are strictly Track 1 actors.

Third generation early warning systems are created by people in conflict areas for themselves. “It can be referred to as ‘Early Warning and Early Response system of citizens, by citizens and for citizens.'” Unlike first and second generation initiatives, monitoring and analysis is conducted on-site. “The logic behind them is that closeness to the conflict area enables one to understand the situation better and intervene rapidly and appropriately.” According to FCE, third generation systems thus have a stronger link between early warning and rapid response.

Early Warning

The FCE uses an events-data software program called FCEWARN for early warning; the unique feature of which is “that it can be utilized to monitor conflicts at the ‘micro’ level, especially at the village level.” The software basically quantifies conflict and peace indicators to display them as descriptive statistics such as tables and graphs. The FCE combines this software with “Geographic Information Systems (GIS) software that visualizes spatial dimensions of conflict and peace indicators.”

The information fed into the software program is collected by the Foundation’s 37 field monitors operating in teh conflict zone.

“They are organic members of the communities they represent. They collect information on peace and conflict indicators and send it to the information center in Colombo in a specific format on a daily basis. […] The field monitors collect information through co-existence committees, state and non-state actors, local media and interpersonal relationships

As a result, the information centre in Colombo […] receives 30 event data forms in the least a day. In total this amounts to 600 event data on average per month.  This density of first hand information allows for adept trend analysis at the early warning stage.”

The FCE draws on the software and data to generate early warning products that “support the early response functions in the conflict zone by teh field monitors […].” In addition, the Foundations makes use of SMS alerts. The FCEWARN software program has the flexibility to integrate a functionality for SMS alerts.

Rapid Response

The FCE claims that “the development of computer software (FCEWARN) for early warning” is their “key achievement” vis-a-vis their “venture into conflict early warning during the past five years.”

However, I would point to their success in responding to 156 cases of conflict as their key achievement. According to the Foundation, their early warning initiative has “intervened in a recorded number of 156 cases of conflict.” The Foundation nodes that “four independnent evaluations by international experts in the science of conflict resolution have attested that this system has prevented or mitigated or contributed to resove conflicts.”

Of note is that the FCE’s early response system is “based on the application of multi-track diplomacy,” unlike first and second generation systems. The Foundation “emphasizes making citizens a major stakeholder in the process of transforming the conflict.” They also recognize the need to build “sufficient capacity and power of mobilization to solicit substantial amount of stakeholder effort from different vantage points.”

To this end, the field monitors are “the primary coordinating hubs of information and early response interventions in the conflict zones. They collect and analyze information and initiate early response processes to prevent conflicts.” Field monitors should therefore have “substantial influences on the masses and/or stakeholders in the conflict zone.” In sum, the rapid response component has to “assume the role of a ‘near’ mediator.”

Unlike the vast majority of conflict early warning initiatives, the FCE actually “reviews the outcomes of one instance of intervention and builds analysis and prognosis for another phase of intervention. This cycle continues until the conditions to the precipitating event are transformed or diluted to a satisfactory level.”

Conclusion

The FCE continues to make important contributions to the field of conflict early warning by demonstrating what an alternative, third generation approach can accomplish compared to top-down first-generation systems. Perhaps what is missing from the report is a stronger emphasis on preparedness and contingency planning. In other words, it would be beneficial to many of us if we could read more on the pro-active and preventive operational measures taken by FCE field monitors beyond conflict resolution excercises.

I would also suggest the notion of “fourth generation” early warning systems. While third generation systems are supposed to be “of the people, for the people by the people,” I think a direct focus on empowering local communities to manage and prevent conflict themselves (as opposed to “external” field monitors) would constitute a fourth generation system. A partial example is Ushahidi, which allows villagers to report alerts by SMS and to also subscribe directly to SMS alerts of incidents taking place in their vicinity.

Advertisements

Conflict Early Warning of Mumbai Attacks (Updated)

Update: Us Warned India Before Mumbai Attack

I’ve been spending the past few days talking to fellow colleagues in the conflict early warning community about the recent carnage in Mumbai. Were there any credible early warnings of the terrorist attacks? Macro-level conflict early warning models forecast specific “events of interest” but rather assess structural risk over longer time spans. So these models did not forecast the attacks. Any likely warning would have to originate from intelligence sources, just as occurred in Kenya.

News is now just coming in that warnings had been communicated to The Taj Mahal Hotel. The chairman of the company that owns the hotel noted how ironic it was that “we did have such a warning, and we did have some measures” but he did not elaborate on the warnings or what security measures were enacted (1). While I recognize that the warnings may not have been particularly specific, what surprises me is why the residents of Mumbai themselves were not alerted about the increased security risk?

Given that 75% of Mumbai’s residents have mobile phones, it would have been feasible to set up a dedicated phone number for residents to send text messages in case they saw something suspicious. The intelligence community tends to be highly hierarchical and centralized, which limits the number of “sensors” or “feelers” it has access to. We’ve been talking about crowdsourcing conflict information, why not crowdsource intelligence since 96% of all intelligence information is open source to be begin with? Especially since the first news of the attack was disseminated on Twitter?

Muhajeriya, Darfur, under threat of militia attack

Just to emphasize once more the importance of on-sight observation and local contacts for conflict early warning, The Aegis Trust recently released this early warning:

The Aegis Trust has received credible eyewitness reports that Janjaweed militia from the Maalia and Rizeigat tribes, estimated to be 300-strong, travelling in jeeps and armed with kalashnikovs, are massing outside the Darfur town of Muhajeriya. Reliable sources on the ground confirm that these are the same militia responsible for the attack on Muhajeriya in October 2007. According to eyewitness testimony, in the past week they have destroyed the village of Sinet, together with several smaller villages in its vicinity, to the northeast of Muhajeriya. On Monday, a UNAMID patrol in the area where the Janjaweed has been sighted was fired upon and one peacekeeper was killed. Sources on the ground are anticipating an assault on Muhajeriya by the militia in the next few days.

Do we really need sophisticated, data-intensive quantitative conflict models?

Fast and Frugal Early Warning

Bradley Perry, a colleague of mine who just completed his Masters of Science in Applied Intelligence, kindly shared a copy of his excellent MA thesis entitled “Fast and Frugal Conflict Early Warning in Sub-Saharan Africa: The Role of Intelligence Analysis.” Bradley carried out his study to counter the erroneous assumption that only those who have large budgets and operate outside high-conflict regions (e.g., academics) are best placed to engage in conflict early warning analysis.

Instead of drawing on dozens and dozens of indicators like the majority of early warning systems, which necessitates substantial amounts of data (most of which is highly aggregated and/or of poor quality), Bradley takes just three indicators to forecast conflict escalation: income inequality, ethnic fractionalization and political freedom. The results from this “good enough” model suggest that we should question our field’s inclination for data-intensive methodologies. In Bradley’s own words, “the results do argue that both the conflict early warning and intelligence communities should consider the value of fast and frugal analysis.”

The fact that the conflict early warning field has been riddled with data-intensive methodologies for the past 20 years is directly due to the fact that those designing these methodologies are for the most part hardcore academics obsessed with prediction and sophisticated econometric models. To be sure, “most conflict early warning systems rely on resource intensive methods. They often take years to develop, and are built on complicated algorithms that require vast amounts of data.” As Bradley adds, however:

“It is safe to say no one has created a system that has the ability to predict; it is likely that no one ever will. In fact, if prediction were the goal in conflict early warning, the intelligence field would have little to offer. Former US government intelligence analyst and author of Anticipating Surprise: Analysis for Strategic Warning, Cynthia Grabo gives this caveat: ‘Warning is not a fact, a tangible substance, a certainty, or a provable hypothesis. It is not something which the finest collection system should be expected to produce full blown or something which can be delivered to the policymaker with the statement, ‘Here it is. We have it now.’”

Below are excerpts from Bradley’s research that strongly resonated with my experience in the field of conflict early warning:

William G. Nhara, a former advocate for the establishment of an early warning system for the OAU, suggests that an early warning system for the African context should be based on a number of methodologies; rather than detail how their incorporation into one system might appear however, he merely lists general sources of information to include: historical surveys and analyses of events, analyses of the content of documents and reports, comparative analyses of relevant information, physical inspections and field visits, statistical sampling and inference, operations research techniques, economic and econometric analysis, and modeling and remote sensing. This enumeration offers little explanation as to how the analyst might process the information, except to say that the responsible agency should store it in a database.

Robert Mudida, a professor of International Conflict Management at the University of Nairobi, described the AU’s “Situation Room” as merely one set up with CNN TV. According to him, it is not proving to be an effective institution in regards to prediction (R. Mudida, pers. comm.).

Indigenous organizations, those with the most responsibility and the greatest chance for success in conflict early warning, are spending precious, yet scant, resources in research, development, and implementation of these models. However, if it is accurately feasible to avoid the complex set of indicators that accompany most warning models and skip altogether the danger of having otherwise accurate systems fall short in applicability, then the identification of a “good enough” model is worth pursuing.

New Prediction Center Created

The “Prediction Center” is a new joint venture between The Washington Post and Predictify. The service allows readers to vote on possible outcomes for selected stories.

The project goes beyond basic polling systems by integrating discussion features and monitoring a users’ accuracy score across the entire service. While there isn’t currently a way to weight one question more than another, the site’s algorithm does take into account the type of question and the accuracy rate of participants.

To offer an incentive for users to take part in the polls, the site has also implemented a premium program that allows companies to sponsor a poll and reward the most accurate participants with cash. In return, these sponsors are entitled to the demographics data that the service asks for with each vote.

Can crowd-sourcing be an effective way to predict conflict?