Tag Archives: Early

Big Data: Sensing and Shaping Emerging Conflicts

The National Academy of Engineering (NAE) and US Institute of Peace (USIP) co-organized a fascinating workshop on “Sensing & Shaping Emerging Conflicts” in November 2012. I had the pleasure of speaking at this workshop, the objective of which was to “identify major opportunities and impediments to providing better real-time information to actors directly involved in situations that could lead to deadly violence.” We explored “several scenarios of potential violence drawn from recent country cases,” and “considered a set of technologies, applications and strategies that have been particularly useful—or could be, if better adapted for conflict prevention.” 

neurons_cropped

The workshop report was finally published this week. If you don’t have time to leaf through the 40+page study, then the following highlights may be of interest. One of the main themes to emerge was the promise of machine learning (ML), a branch of Artificial Intelligence (AI). These approaches “continue to develop and be applied in un-anticipated ways, […] the pressure from the peacebuilding community directed at technology developers to apply these new technologies to the cause of peace could have tremendous benefits.” On a personal note, this is one of the main reasons I joined the Qatar Computing Research Institute (QCRI); namely to apply the Institute’s expertise in ML and AI to the cause of peace, development and disaster relief.

“As an example of the capabilities of new technologies, Rafal Rohozinski, principal with the SecDev Group, described a sensing exercise focused on Syria. Using social media analytics, his group has been able to identify the locations of ceasefire violations or regime deployments within 5 to 15 minutes of their occurrence. This information could then be passed to UN monitors and enable their swift response. In this way, rapid deductive cycles made possible through technology can contribute to rapid inductive cycles in which short-term predictions have meaningful results for actors on the ground. Further analyses of these events and other data also made it possible to capture patterns not seen through social media analytics. For example, any time regime forces moved to a particular area, infrastructure such as communications, electricity, or water would degrade, partly because the forces turned off utilities, a normal practice, and partly because the movement of heavy equipment through urban areas caused electricity systems go down. The electrical grid is connected to the Internet, so monitoring of Internet connections provided immediate warnings of force movements.”

This kind of analysis may not be possible in many other contexts. To be sure, the challenge of the “Digital Divide” is particularly pronounced vis-a-vis the potential use of Big Data for sensing and shaping emerging conflicts. That said, my colleague Duncan Watts “clarified that inequality in communications technology is substantially smaller than other forms of inequality, such as access to health care, clean water, transportation, or education, and may even help reduce some of these other forms of inequality. Innovation will almost always accrue first to the wealthier parts of the world, he said, but inequality is less striking in communications than in other areas.” By 2015, for example, Sub-Saharan Africa will have more people with mobile network access than with electricity at home.

Screen Shot 2013-03-16 at 5.46.35 PM

My colleague Chris Spence from NDI also presented at the workshop. He noted the importance of sensing the positive and not just the negative during an election. “In elections you want to focus as much on the positive as you do on the negative and tell a story that really does convey to the public what’s actually going on and not just a … biased sample of negative reports.” Chris also highlighted that “one problem with election monitoring is that analysts still typically work with the software tools they used in the days of manual reporting rather than the Web-based tools now available. There’s an opportunity that we’ve been trying to solve, and we welcome help.” Building on our expertise in Machine Learning and Artificial Intelligence, my QCRI colleagues and I want to develop classifiers that automatically categorize large volumes of crowdsourced election reports. So I’m exploring this further with Chris & NDI. Check out the Artificial Intelligence for Monitoring Elections (AIME) project for more information.

One of the most refreshing aspects of the day-long workshop was the very clear distinction made between warning and response. As colleague Sanjana Hattotuwa cautioned: “It’s an open question whether some things are better left unsaid and buried literally and metaphorically.”  Duncan added that, “The most important question is what to do with information once it has been gathered.” Indeed, “Simply giving people more information doesn’t necessarily lead to a better outcome, although some-times it does.” My colleague Dennis King summed it up very nicely, “Political will is not an icon on your computer screen… Generating political will is the missing factor in peacebuilding and conflict resolution.”

In other words, “the peacebuilding community often lacks actionable strategies to convert sensing into shaping,” as colleague Fred Tipson rightly noted. Libbie Prescott, who served as strategic advisor to the US Secretary of State and participated in the workshop, added: “Policymakers have preexisting agendas, and just presenting them with data does not guarantee a response.” As my colleague Peter Walker wrote in a book chapter published way back in 1992, “There is little point in investing in warning systems if one then ignores the warnings!” To be clear, “early warning should not be an end in itself; it is only a tool for preparedness, prevention and mitigation with regard to disasters, emergencies and conflict situations, whether short or long term ones. […] The real issue is not detecting the developing situation, but reacting to it.”

Now Fast froward to 2013: OCHA just published this groundbreaking report confirming that “early warning signals for the Horn of Africa famine in 2011 did not produce sufficient action in time, leading to thousands of avoidable deaths. Similarly, related research has shown that the 2010 Pakistan floods were predictable.” As DfID notes in this 2012 strategy document, “Even when good data is available, it is not always used to inform decisions. There are a number of reasons for this, including data not being available in the right format, not widely dispersed, not easily accessible by users, not being transmitted through training and poor information management. Also, data may arrive too late to be able to influence decision-making in real time operations or may not be valued by actors who are more focused on immediate action” (DfID)So how do we reconcile all this with Fred’s critical point: “The focus needs to be on how to assist the people involved to avoid the worst consequences of potential deadly violence.”

mind-the-gap

The fact of the matter is that this warning-response gap in the field of conflict prevention is over 20 years old. I have written extensively about the warning-response problem here (PDF) and here (PDF), for example. So this challenge is hardly a new one, which explains why a number of innovative and promising solutions have been put forward of the years, e..g, the decentralization of conflict early warning and response. As my colleague David Nyheim wrote five years ago:

A state-centric focus in conflict management does not reflect an understanding of the role played by civil society organisations in situations where the state has failed. An external, interventionist, and state-centric approach in early warning fuels disjointed and top down responses in situations that require integrated and multilevel action.” He added: “Micro-level responses to violent conflict by ‘third generation early warning systems’ are an exciting development in the field that should be encouraged further. These kinds of responses save lives.”

This explains why Sanjana is right when he emphasizes that “Technology needs to be democratized […], made available at the lowest possible grassroots level and not used just by elites. Both sensing and shaping need to include all people, not just those who are inherently in a position to use technology.” Furthermore, Fred is spot on when he says that “Technology can serve civil disobedience and civil mobilization […] as a component of broader strategies for political change. It can help people organize and mobilize around particular goals. It can spread a vision of society that contests the visions of authoritarian.”

In sum, As Barnett Rubin wrote in his excellent book (2002) Blood on the Doorstep: The Politics of Preventive Action, “prevent[ing] violent conflict requires not merely identifying causes and testing policy instruments but building a political movement.” Hence this 2008 paper (PDF) in which I explain in detail how to promote and facilitate technology-enabled civil resistance as a form of conflict early response and violence prevention.

Bio

See Also:

  • Big Data for Conflict Prevention [Link]

Marketing Peace using SMS Mobile Advertising: A New Approach to Conflict Prevention

I was just in Kenya working on the next phase of the PeaceTXT project with my colleague Rachel Brown from Sisi ni Amani. I’m finally getting to implement an approach to conflict early warning and early response that I have been advocating for since 2006. I came close in 2008 whilst working on a conflict early and response project in Timor-Leste. But I wasn’t in Dili long enough to see the project through and the country’s limited mobile phone coverage presented an important obstacle. Long story short, I’ve been advocating for a people-centered and preparedness-based approach to conflict early warning systems for half a decade and am finally implementing one with PeaceTXT.

Conflicts are often grounded in the stories and narratives that people tell themselves and the emotions that these stories generate. Narratives shape identity and the social construct of reality—we interpret our lives through stories. These have the power to transform relationships and communities. The purpose of PeaceTXT is to leverage mobile messaging (SMS) to market peace in strategic ways and thereby generate alternative narratives. SMS reminders have been particularly effective in catalyzing behavior change in several important public health projects. In addition, marketing to the “Bottom of the Pyramid” is increasingly big business and getting more sophisticated. We believe that lessons learned from these sectors can be combined and applied to catalyze behavior  change vis-a-vis peace and conflict issues by amplifying new narratives using timely and strategically targeted SMS campaigns.

Last year, Sisi ni Amani sent the following SMS to 10,000 subscribers across Kenya: A good leader initiates and encourages peace and development among all people and is not tribal. “In a nation divided along ethnic lines, where a winner-takes-all mindset fuels rampant corruption and political violence, changing perceptions of good leadership is a daunting endeavor. And yet, according to post-campaign data, 90 percent of respondents said they changed their understanding of ‘what makes a good leader’ in response to the organization’s messaging. As one respondent commented: ‘I used to think a good leader is one who has the most votes, but now I know a good leader is one who thinks of the people who voted for him, not himself'” (NextBillion Blog Post).

PeaceTXT is about marketing peace using mobile advertising by leveraging user-generated content for said text messages. We’re in the business of selling peace for free by countering other narratives that tend to incite violent behavior. Preparedness is core to the PeaceTXT model. To be sure, local mobile-based advertising is hardly reactive or random. Indeed, billions of dollars go into marketing campaigns for a reason. To this end, we’re busy developing an agile SMS protocol that will allow us to send pre-determined customized text messages to specific groups (demographics) in targeted locations within minutes of an incident occurring. The content for said text messages will come from local communities themselves.

The next step is for Rachel and her team to organize and hold several local focus groups in July to begin generating appropriate content for text messages to de-escalate and/or counter police-community tensions, rumors and insecurity. I’ll be back in Kenya in August to review this user-generated content so we can add the text messages to our SMS protocol and customized SMS platform. I’m thrilled and can’t wait to work on this next phase.

The Quest for a Disaster Early Warning System (1988)

I assign this short report (PDF) by Kumar Rupesinghe as required reading in my courses and professional seminars on conflict early warning and crisis mapping. This paper could be published today—20 years later—and still be considered forward looking. I highly recommend reading the report along with Rupesinghe’s new edited book on Third Generation Early Warning Systems, which I reviewed in detail here.

I stumbled across this report while working for Norway’s Former Secretary of State at the Peace Research Institute, Oslo (PRIO) in 2006. Rupesinghe was also a research fellow at PRIO and I found the worn-down report hidden away in the Institute’s library. I carefully scanned the report and have been disseminating it as widely as possible ever since.

The excerpts below are the main highlights of the report. While some of these may seem obvious, keep in mind that Rupesinghe was writing this in 1988—well before the field of conflict early warning became formalized.

I realize there are numerous excerpts below but this just reflects how important I think this piece is.

  • It is crucial for the viability and credibility of developing information and communication systems to discuss ways in which the information can be used. Information is useful if acted upon, and when the information so produced provides choices of action to policy makers as well as to the victims to the victims of the impending disasters.
  • Discussions on ‘early warning’ systems would remain academic if information systems are developed which bear little relationship to social policy or social action.
  • Important to the discussion of early warnings are some of the issues related to the demand for a ‘new information order’. Here we have to raise the entire question of who controls information. Generally, discussions relating to early warning systems emanate from the North, and particularly in environments, which can handle large amounts of information. Little attention is paid, however, to the victims of disasters, or to the competence of local NGOs to strengthen their own capacity to handle information, to evaluate and control their own environment.
  • Akira Onishi suggests a highly sophisticated information system, since in the fields of present day technology, particularly in the astounding developments of computers in the 1980s, extraordinary sophisticated handling of information has become possible both in software and hardware systems. Onishi suggests this field of research as been stimulated by the progress made in global modeling.
  • The distinction between a ‘natural’ and a ‘social’ disaster has also been challenged, particularly by those who have witnessed the recent famines in Africa:

The causes of the African crises are increasingly perceived as man made, or they are at least attributed to human activities more than to natural phenomena. Such terms as ‘man made’ and ‘natural disasters’ widely used to distinguish two categories in the past disaster jargon, are of little relevance or may even be misleading. It is now understood that some of the major interlinked factors behind the growing disaster problem in Africa are man made.”

  • Within the discussion on restraining military technology and redeeming modern science and technology for the good of humanity, Marek Thee has proposed that research and development be subjected to national and international scrutiny. Further, he writes:

National and international technological assessment bodies can be established to serve as a kind of ‘watch and early warning system’ against military excesses. Such concurrent international measures may not be easily achieved. But, if we have the political will informed by human rationality and a comprehension of the scientific-technological global interdependence, reinforced by an awakened public opinion, the barriers for change are not insurmountable.

  • Although billions of dollars have been invested in developing sophisticated data banks and early warnings, we have to note that even the most expensive systems have shown a striking inability to forecast political events.

‘Quite simply, the record has been terrible, despite all the technological improvements of the last twenty years. The Soviet intervention in Czechoslovakia, the Yom Kippur war, and the Argentine take over of the Falklands all caught the American government by surprise.’

  • A specific proposal for [a United Nations] early warning system was made by the Speial Rapporteur, Prince Sadruddin Aga Khan in his report on Massive Exodus and Human Rights to the UN Human Rights Commission on December 1981. In his study, Sadruddin Aga Khan drew attention to the increase in the phenomenon of mass exodus:

‘which is becoming a tragically permanent feature of our times, … The problem is to become more serious with time unless imaginative and concrete measures are urgently taken to contain, if not avert, situations of mass exodus. … It is however my considered opinion that if we are to succeed in any measure to spare future generations the spectre of millions of uprooted peoples, more is required than reports and resolutions, however pertinent and useful they may be.’

  • A recent report of the United Nations that the UN’s own capacity to deliver is its existing commitments is seriously at stake. The report suggests that ‘today’s structure is both too top heavy and too complex.’ Further, ‘its present organizational structure is too fragmented’, and ‘without enlarging secretariat functions’, a ‘leaner Secretariat will enhance productivity and improve efficiency.’
  • In many cases, the problem has not been one of insufficient information about probable crises, but that the specialized [UN] agencies or the secretariat lack the mandate or the authority to act on the information.
  • The United Nations alone would not be in a position to assist in the building of adequate information systems, without the participation of a range of NGOs. This is particularly relevant for developing countries, who themselves can be involved in an active partnership in the exchange of information.
  • What is increasingly realized is that the NGO community in general has a profound role to play in early warnings, monitoring, providing immediate relief, and finding creative ways of resolving conflicts. […] NGOs, particularly in the Third World, and the media, can play a crucial role in early warnings, sounding the alarm in cases of emergency.
  • Most of the discussions with regard to early warning systems have emanated from a concern with the early prediction and reporting of events which could lead to social disasters. […] However, these systems have been developed in advanced environments where the intention is to gather data so as to predict events in distant places. This leads to a division of labor between those who predict and those predicted upon. And this in turn tends to draw attention only to those efforts which continue to reinforce existing unequal distribution of information.
  • A democratic flow of information is the first condition for a democratic and open system of warnings and resolution. However, ‘information’ is a highly explosive and political issues, especially in the Third World. Many countries have elaborate laws to prevent people from gaining access to information, or censorship laws which prevent people from reporting on what actually happens in a society.
  • Thus, efforts must be made to strengthen the capacity and competence of Third World NGOs to communicate locally and internationally so as to create a democratic global communication system. An information system of the monolithic type developed by the superpowers cannot be encouraged as far as the NGOs are concerned.
  • [There is a] wealth of knowledge available within the local societies [hence the] importance of finding ways of tapping this wealth of information [and] of involving the local societies and integrating their work, so as to build local competence in monitoring and evaluating, their own experiences.
  • […] NGOs, both international and national, will have a vital role to play in the development of a global, decentralized early warning system. They now need the capacity to build information systems, and to provide the basis for rapid information exchange. In general, NGOs will have to confront the monopolization of information with a demand for the democratic access to information technology. Further, the working conditions of most NGOs and NGO networks, especially in developing countries, remain difficult.
  • Increasingly, bureaucracies are exhibiting their incapacity to manage the complexities of our global village. And today the alternative structures most likely to succeed these bureaucracies are rapidly emerging. The most common term for these structures is ‘networks’. They tend to be decentralized, where policies tend to be flexible and fluid, where staff relations are not monolithic and hierarchical, where the structure tends to be polycentric rather than monocentric.
  • NGOs lack systematization and standardization of information. Each small NGO uses its own methods often based around their previous normal systems. Thus, NGOs have to build competence in standardization of information and cataloguing procedures, to facilitate exchange of data and easy retrieval.  HURIDOCS, Human Rights Information and Documentation System, ahs been established on this basis as a network to assist information sharing and usage.
  • Low cost computers provide opportunities for information recording and retrieval and for the development of data bases in highly specialized areas. Once electronic mail becomes cheaper than the conventional post, every NGO will be able to do cheap mass mailings in a fraction of the present time.
  • With regard to the use of satellite technology, there is a strong case for involving grass roots movements and international NGOs alike in the learning process about satellites as well as radio communication. The aim would be to ensure adequate access to the airwaves for the non-governmental sector as a whole and guarantee the democratization of satellite communication.
  • In this discussion of communications and information for early warnings, we have stressed the need to strengthen the capacity and the competence in the South to store, retrieve, and analyze their own information. […] Here Northern NGOs and donors have an important role to play.