It’s not enough to just have information — intelligence requires context

Structure Data 2013 Samantha Ravich National Commission for the Review of R&D in the Intelligence Community

Session Name: Motion, Context And Integrating Intelligence.


Announcer Samantha Ravich


Our next speaker said she’s going to be talking about some big ideas, so this should be fun. She is Samantha Ravich, she’s the co-chair for the National Commission for R&D in the Intelligence Community and she’s going to be talking about motion, context and integrating intelligence. Please welcome Samantha Ravich to the stage.



Thank you. Here’s the story: 825 days ago Tarek al-Tayeb Mohamed Bouazizi set himself on fire. Who was he, you might wonder, and what does he have to do with motion, context, and integrating intelligence? Bouazizi was the young Tunisian fruit vendor who set in motion what has become known as the Arab Spring. For years, Bouazizi, like many others in the world who live in a shadow economy, was harassed by police for lacking a permit to operate his fruit stand, a permit that was virtually impossible to get in Tunisia. The only way his street vending could continue was to bribe the police. So on that day in December 825 days ago, Bouazizi didn’t have the bribe money. Roughed up and humiliated by the police and municipal officials, Bouazizi grabbed a canister of gasoline, doused himself, lit himself on fire and shouted How do you expect me to make a living?”. His act galvanized a town, then a country, then a region, and then the world. Events had been set in motion. The story was evolving. Liberal secularists around the world saw it as their moment to throw out corrupt dictators and promote economic and civil liberties. The Muslim Brotherhood moved to enshrine Islam as the guiding force in social, political and personal life. And Al-Qaeda and their affiliates saw a pathway to further their own goals of radical Islam and attacking the west.


Each had their own internal and external backers that were seeking ways to move money, people, weapons, and other forms of support to these hotspots. On January 14 of 2011, less than a month after Bouazizi’s act of defiance, Tunisia’s president Ben-Ali was forced to flee the country after 23 years in power. Protest then began in Lebanon, in Oman, in Egypt, in Syria, and in Morocco. In February, Hosni Mubarak, Egypt’s president for 30 years, resigned. A few months after that Gaddafi, ruler of Libya for 42 years, was dragged from his hiding hole and shot.


The Middle East and North Africa have not stopped churning as a result of these events. Syria is in the middle of a bloody war. Nearly a hundred thousand and perhaps more are dead or missing. Syria’s very large stockpiles of chemical weapons may now be in motion, still controlled by an increasingly desperate regime, but eyed on by a fractious opposition that is more and more in the hands of Al-Qaeda elements.


If we could go back to the day before Bouazizi struck that match, would we have been able to predict the trajectory of that evolving story? Would we be able to communicate that story so other would also understand what we were seeing, and more importantly, would we be able to help direct it toward the outcome we would like, by undermining malevolent forces and strengthening salutary ones? Probably not, or not very well.


Why not? we have the best intelligence service in the entire world. We have the most sophisticated signals intelligence and imagery intelligence, so how could we not have seen what was coming? Because we failed to see the motion. I’m a global security strategist, and as such I try not only to understand the risks and opportunities in the world, but think through ways to minimize those risks and seize the opportunities. When I examine what we could have done to propel or curtail the Arab Spring, I think the answer lies in recognizing, really recognizing that the world is in motion as never before, and that may seem obvious, but our intelligence is typically still more in the format of a still image, or a standalone conversation. What we need are new ways to see and measure movement of people, of ideas, of technology, of ideology, of money, of weapons. And that’s not enough. We have to put that motion into context so we understand when things such as local sentiments are actually ripe for revolution. To do this, we need to integrate technology and analysis in non-traditional ways.


Let’s start for a moment with the idea that the world is in motion. We all know the statistics on the world-wide use of the internet connecting people, but it’s not just on the web that people and ideas are in motion like never before. International air travel has more than doubled in the last 15 years. The number of people living outside their country of birth has increased by 60 million over that same time. It’s like all the people in the United Kingdom have gotten up and moved somewhere else. And money? That stuff that actually makes the world go round? The amount of money being converted from one currency to another grew by 20 percent from 2007 to 2011, to reach almost 4 trillion dollars a day. Now more people around the world have the ability to transact business in other parts of the world, which is good for legitimate businesses, but it is also good for those who conduct illicit affairs. Proliferating dangerous technologies, or drugs, or weapons, or fomenting revolution far from their country of origin. Things are on the move.


I want to share a part of a poem by British poet Stevie Smith, because what’s a big data conference if you’re not going to put in a bit of poetry? But I think it really captures my point. She wrote, ” I was much further out than you thought/ and not waving/ but drowning.” So if there was a photo of that man, his raised arm could be interpreted as waving, as if he was enjoying a swim in the water. But capturing his movement may have told a different story. His thrashing about may have shown his distress, and that’s the danger of not seeing things in motion. So it would be good to have a moving picture of that man, maybe a video. Of course one example of where the US Government and the intelligence community is really putting a lot of effort is in integrating intelligence in the use of video, and the amount of video coming online, as you know, is just overwhelming. YouTube alone uploads 72 hours of video every minute, in 43 countries and 60 languages.


We’re in an exciting time, when new source of information generated by commercial devices, such as smart sensors, will be everywhere. Finding better and faster ways to aggregate and process data from all those sensors will help provide better warning signals of where the story is evolving. But right now the technology for getting the most from video really is still in its infancy. the tools to catalog and organize and search within video, but in some ways they still mimic taking information from each frame, rather than gleaning information from the movement across those frames.


So back to that drowning man. We really need to know actually more than is in the video to understand what’s going on and if we should call for help. We need to know more about the man himself. Can he swim? Was he pushed from a boat? How did he get out so far? How deep is the water? Is there an undertow? We need context. Who would know to ask and answer those types of questions, to direct resources to collect this data, to ultimately determine if we need to send out a lifeguard or simply wave back?


But let’s bring this back to real life. In the 825 days from Bouazizi’s self-immolation we’re still not able to track and predict where that story is headed. We’re still getting caught off guard. The importance of understanding motion and context can be seen in what is transpiring right now in northern Mali, in North Africa, which has had parts of it taken over by radical Islamists. One of these groups is an affiliate of Al-Qaeda, known as MUJAO, and it was able to surprise even the best national security analysts by how fast and competently it was able to move from a small, clandestine terrorist organization to a significant overt one in a matter of months. The organization’s success rested with a number of factors.


One, the surge of weapons and forces flowing from Libya after the fall of Gaddafi. In the aftermath of Gaddafi’s ouster, hundreds of radical Islamists were released from Libya’s prisons. Many quickly took up arms in Libya, or moved eastwards towards Afghanistan and Pakistan. But others moved westward, towards Algeria, finding safe harbor in refugee camps or within ungoverned territories in northern Mali.


But there was a second key factor that led to MUJAO’s success. They were able to use social media to expand and move his message. They announced their recruitment drives. They Tweeted their victories. They broadcast the group’s talking points. In one recruiting drive, over two days, 200 Africans joined the jihad and moved to Mali to start their training. Separately, hundreds of fighters reportedly came from the Nigerian Islamist group Boko Haram.


Moving weapons. Moving messages. Moving people. We are missing the boat if we see these events like individual still images. But again, even if we see things in motion, we’re missing the story unless we place it in context. There are two researchers, Lambertson and Page, of the Santa Fe institute, who has done some interesting work that I think relates a lot to this. They write that there’s too much attention trying to find that direct tip, that struck match from the Arab Spring, and what we need to be doing is developing the instrumentation and analysis we need to monitor and understand contextual tips.


What’s a contextual tip? It basically occurs when a gradual change in one thing leads to a jump in some other variable of interest. So in other words, in the Arab Spring, it’s these contextual tips that are like a warning bell that something big is going to happen. So there was a lot happening in Tunisia before Bouazizi struck that match that we should have, could have been watching. Blog postings on the growing frustration at corruption for instance, or the trending of words such as ” hogra”, which is an Algerian term used throughout North Africa to denote the contempt of rulers towards their people. Likewise, the ground game that Al-Qaeda and its affiliates were already laying in the region made it ripe for them to seize northern Mali. Following the people and weapons moving out of Libya would have given us a better understanding of what was building in Mali, and likewise, the leader of MUJAO made little attempt to hide his recruitment drive. He was trying to mobilize resources to help him in his recruitment drive, and he sent a very public signal, if we had known how to look for it and how to see it in context.


So how can we learn to do this? I think it requires people like you, technologists, engineers, scientists, and people like me, working together to develop new capabilities. Because important as the new technology is to understanding this world in motion, I really do believe that that technology will only be truly groundbreaking if it is created hand-in-hand with the analysts and the strategists, so we can see things in context.


Robert Gallucci, the president of the MacArthur Foundation, recently published a piece in the Chronicle of Higher Education where he wrote that something is seriously wrong in the relationship between Universities and the policy community in the field of international relations. He blamed the universities for taking too much of a theoretical turn, paying little more than lip service to the importance of interdisciplinary work. He underscored that nothing can replace the value of insights that emerge from the integration of knowledge and research with history, and economics, and politics, and culture, and religion and geography of a region. I wholeheartedly agree. Interdisciplinary insights will provide the context. They will help make sense of the story. I want to see those insights integrated into the tools and capabilities that are produced in the collection and exploitation of intelligence. I want tools created with the insights of those who understand context, which can help humans ask the right questions, technology that can help us all tell the story, and provide guidance for those who are entrusted to shape its outcome.


Afghanistan provides an unfortunate example of where we could have most benefited from such integration. For over a decade now we have had the most amazing information on Afghanistan. We can literally map every square inch of that country. We can count nearly every poppy being grown. We can follow the movements of a single suspected terrorist from high up in the sky. But we miss the context of how to interpret and use this data. In the words of General Michael Flynn, who is now the head of the Defense Intelligence Agency, but at the time he was directing intelligence in Afghanistan, he wrote ” The vast intelligence apparatus is unable to answer fundamental questions about the environment in which the US and allied forces operate, and the people they seek to persuade.”


Why was it unable to do what General Flynn needed? Lots of reasons; and some we may not be able to change. But one stands out, and I think it is something we have the power to fix. Because it requires thinking anew of how we seek to learn about the world. This requires bringing three communities that don’t always talk to each other to collaborate in new ways and to build the tools and capabilities that we will need to succeed in this world in motion. We need the subject matter experts and the analysts to explain their understanding of the elements within the system. We need to listen to the strategists as they explain how we want the system to move or not move to best serve our interests, to get the outcomes that we most want. And we need to empower the technologists to build the tools to capture different types of data. We must also empower the technologists again, you, to encourage people like me and my cohort to think about the art of what could be possible.


So at its most elemental, what we need to do is to better understand the evolving story of the world, be able to tell that evolving story to others, and be able to understand how to effect change within that evolving story. We must recognize also that this challenges the established way of doing intelligence. To get it right, we’re going to have to take risks and experiment, and challenge that often siloed thinking within the established community, the technologists over here, the scientists over there, the analysts there, no-one talks to the strategists, the policy makers. It can’t work like that anymore.


So this is the challenge that I present to you today. Let’s evolve from measuring still images to measuring motion, and let’s evolve from seeing isolated events to seeing context, and let’s evolve from building siloed capabilities to really building integrated intelligence, and these are tough challenges. But I look forward to figuring them out together. Thank you.



Well, that leads us into our first break of the morning. We hope you’ve had a good morning so far. We’re going to go into a one hour break, so please network amongst yourselves, talk to each other, see what everyone else is up to, and then be sure to visit the sponsor workshops. Remember, these things fill up early, so get there as soon as you can. Hadapt is having a sponsor workshop in Oceanic, Alterics is having a sponsor workshop in Aquitania West, and Juniper Networks is having a sponsor workshop in Aquitania East. Stop by the GigaOM research table and find out what’s happening there, refreshments are located in Olympic and in the main hallway, and the general session will resume at 11:45. Thanks everybody.

firstpage of 2

Comments have been disabled for this post