 Welcome. Thanks so much for coming. This panel is entitled Engaging the Complexity in Peacebuilding, a Systems Perspective. So, no one here probably would argue that peacebuilding and conflict resolution processes are complex. They are. They're complex. They're messy. They're wicked. They're wicked and messy and complex problems and processes. And undoubtedly we all experience the pain of that reality at some level every day in our organizations or in the communities that we work within. Sometimes it shows up as a confusion about what we're doing and why. Maybe it shows up as a difficulty communicating with our funders or our boards or our leaders that we're working with. Sometimes it shows up as not being on the same page with the people you're working with or outright conflict. So, you know, it's not that we don't think of things holistically we do, but the problem I think sometimes we have is being able to juggle all the complexities and put that together to kind of think about what we're doing and what kinds of effects we expect. So, in setting up this panel, what I did is we have a really esteemed group for you. And each of them has a bio that just warrants me applauding them and telling you a lot about them. But I'm going to refer you, since we're on such short time, to the packets that actually have people's biosketches. But what I did in asking, pulling these folks together and asking them to speak to you is I asked them to think about what questions would you want folks to be asking about the context in which they work? And once you're asking those questions, what kinds of tools or ways of practicing would you be encouraging them to use to better answer or engage those questions? And sometimes that engaging those questions just means being able to ask better questions. So, I don't want to take any of their time in speaking to you. So, I'm going to go ahead and turn it over right away to Mr. Robert Sigliano, who's going to get us started. And we'll go from there. And hopefully at the end, we'll have a little time for questions. But if they haven't helped you see by the end of this how it is that you might be asking better questions of the context you work within and the work that you're doing and what kinds of tools you might be using, be sure to tackle them at one of the breaks or in another session because that's what we're here to give you. So, we'll get started. Thanks, Robert. Thanks, Karen. Good morning, everybody. Happy to talk about my favorite subject. And first, I want to say that I could explain complexity and peacebuilding to you, but it's just too complex so you wouldn't get it. I won't go there. So, I want to also say that there is a certain difficulty in talking about complexity and systems approaches to peacebuilding in that it's hard to do that without implicitly criticizing all the good work that's happened in the past. I mean, granted, there are good, bad, and ugly outcomes that we've had in this field. I do a lot of work with family of organizations or initiatives under the umbrella of the Amityar Group. And they did a, this is funded primarily by the PR and Pam Amityar, and he's the founder of eBay. And so, it's the proceeds of that endeavor that they're pledging back. And they did a report called the first decade, the first billion. So, they had given a billion dollars in philanthropic resources to these various initiatives, peacebuilding and social change. Hope Lab that Ari mentioned is one of their organizations that they support. And what's been great about that is that they said, look, we did a great work. All these organizations did great work in the last 10 years, had great impacts, and we are wanting more. And that's hard. I work a lot with the people that they've, that work in these organizations and there's a bit, they feel that. Like, so we didn't, we did something wrong. I mean, what was, and it's not that, it's just that we're saying, let's look at a different way of being, different level of impact, a different way of operating. So, I want to jump in on this. I want to, these are my answers to Karen's questions that she gave us. So, if we're that short on time, I could stop now, Karen, because this is the quick presentation. So, the key question I want people to ask, or I wish they would ask is, is what they're doing? Is the issue they're working on, is the cause that they're championing, championing something that's fixable or something that's complex? And more directly or more, more, more elaborately, is it, to what extent is what you're working on a fixable issue? Where if it took some resources, it took some technology, it took some information, and you would then get a sustainable and effective outcome. Is it, to what degree is it that? And to what degree is it much more complex? It's much even harder to understand, let alone engage with and have impact on. I'll talk a little bit about this concept of a systems practice, which is really not a specific tool, but it's a set of tools for how we engage that complexity. And I want to leave with, I think, the one fundamental, there's lots of changes I would see in our practice, but the fundamental one is to help us collectively and individually break out of what I would call the vicious virtuous cycle. Which basically says that we, it's the very good intentions and practices that allow us to be effective to the degree we are effective, that actually undermines our ability to really engage systemically and holistically over the long term with many of these most difficult problems. So again, I think a lot of you know this list of issues that I won't call a lot of attention to these, but there is a general feeling that there is something wrong, we want something more. I hear donor fatigue a lot, and I want to call attention to the fatigue that is felt by so-called beneficiaries, which I think is actually much larger than those that donors feel. All of these are symptoms, I think, of this basic problem. So the question that I talked about is, it's almost like we live on two different planets. There are two different ways of being effective in the world. There's the planet fixable world where you actually can directly fix things. You can build a well and have access to fresh water in an impoverished community. And then there's planet complex where our best habits, our linear logic models, our engineering principles are much less helpful to us than they are in the planet fixable context. And so I want to just elaborate a little bit on what I see as the distinction around that question. So unlike Ari, I'm not a graphic artist or game designer, so these are not very sophisticated graphics. So if that black blob represents the context within your work, that's the Congo, that's Nigeria or Pakistan, and the yellow dot is the thing you care about, that's say poverty alleviation, gender empowerment, access to fresh water, whatever it is. What distinguishes planet fixable from planet complex is the degree of interrelationship, interconnection between the thing you care about and the surrounding context. So if it's a relatively straightforward connection, that's fixable. So the simple example is like a spark plug in an engine. It does interact with the rest of the engine, but it is not so interdependent on the rest of the engine that if the air filter breaks, the spark plug breaks. So it is something you can actually fix by replacing the spark plug or repairing it in some way. Complexity says there is a much different interrelationship between the thing you care about and its surrounding context. And if that's the case, then this system, that sort of messy set of gray dotted arrows there, that is the point at which you need to engage. Because engaging directly in the problem, like going directly after the poverty or the fresh water, may not be the way to long term sustainably address that issue. And so the challenge is how do we address the system? And so fundamentally that's the system's perspective to me. So that's the reason why you ask that question, is if you should be addressing the system, don't try to treat it like it's a fixable problem. So in a sort of simple example that if you take a village in Malawi and say needs access to fresh water, is the reason why they don't have fresh water a function of the fact that they just need a well? If it is, dig the well. If it's much more interconnected, so the reason why they don't have access to fresh water is because they're entrenched economic interests. Someone gets paid to tote water from 10 kilometers away into the village three times a day. Is it about religious and clan conflicts? Is it about poor governance? Is the community taking its scarce resources and using it to bring in health practitioners rather than access to spending it on fresh water? If that's the case, then you're in a complex situation. So just digging the well doesn't work. And actually, I use this example because I build off a TED talk by someone from engineers that borders in Canada who gave a really illuminating talk about their project that dug 103 wells in villages in Malawi after about a year and a half, 81 of them broke down. And it was largely because they were treating it like it was fixable and not because it was like it was complex. And that was really what the source of both the breakdown and the lack of repair came from that mess of gray hours, came from the system. And so what does that look like in terms of how we engage? This is some way, skips down, actually, Karen to your third question. So if we're trying to do is engage the system, by engaging the system over time, we can improve whether that's a level of peace or a level of poverty, whatever you want to put on the vertical axis, is that engaging the surrounding system over time indirectly affects the issue we care about. It's just a different model of how we engage and how we be effective. So that's the fundamental sort of, for me, reason why you ask that question and what it does to your practice. The second question, so what are the tools? What do you do do this? I break it down. This is my organization of my toolbox. So there are ways, there are tools we can use to listen to the systems. So systems are very difficult. There's not like a label that says, here's the system of Congo. There are mental constructs that can be very difficult. We run into huge problems of information management, which I'm sure, Nate and Scott, you're going to settle for us, right? Right, okay. So how do we do that, whether it's through a visualization technique, it's a data management technique, but it's still difficult, poses challenges, but there are ways to deal with this. Engaging the system, as opposed to intervening in it, we've got to get off of the addiction to projects, which again, is really maybe structured by a lot of donors, that we want to think about things that can be done in six months a year, maybe two years at the most. Engagement doesn't have a beginning and an end. It doesn't have a before and an after. It hasn't always. And that's, again, a different way that we think about engaging, but there are also tools for engaging effectively. And same on learning and learning systems, so also I think a lot of what Scott and Nate may talk about, the ways in which we do learn over time, we do learn from multiple and big data sources. Just a quick, I won't spend a lot of time talking about the tools, but things like systems mapping, more holistic analysis of systems, key things about engaging, finding leverage points in the system, working in networks, but really networks that are effective, that really lead to effective action. Monitoring, outcome harvesting type approaches, building these larger learning infrastructures that are multi-party and pull information and data out of lots of places that work across organizations. These are all just, again, some of the particulars in that toolkit. But I wanted to finish up with talking about the harder issue for me, which is this idea of the vicious-virtuous cycle. And I think that, again, we all play a part in, and I play a part in, and we also play a part in both executing it and being stuck in it, but also disrupting it. So this is the system I think we are trapped in as peace builders in a lot of ways. And so it basically starts with this idea that there's a strong need and desire for assistance. There are any number of problems out there that we could assist with, which leads to those of us, like in this room, in government and non-governmental organizations and academic institutions feeling a need and a desire to help. And it drives us to this idea of fix. Can we find a fix? Can we alleviate suffering now? Can we open up the access to the fresh water? Can we help conduct an election? And that does lead to some positive impacts, which, again, increases the desire, hey, that worked. Let's do something else. But unfortunately, in the complex space, so in those areas of the world where we work, there is true complexity at play, oftentimes those fixes tend to fail. So either they're unsustainable, they last like the 18 months of the well project I mentioned in Malawi, or they actually have negative impacts. So it would look like a good idea to begin with, ended up having negative consequences because the system sort of was the empire strikes back. The system finds a way to counteract and undermine whatever that change was. Also, I think our attempts to sort of fix in the short term undermines our ability and our desire to engage for the long term. So it was like the question I asked Senator Feingold about the fact that we tend to engage around crises and dramatic moments and not deep long term underlying root dynamics and patterns that we see over and over, but we just keep falling into again and again. So that for me is the vicious part of this. Is the two sort of red parts of that diagram where we tend not to have incentives, we don't get rewarded, we don't get funded to engage deeper systems. People in those bureaucracies know that there is a larger system at play, but say I and I in my program at U.S. aid or wherever, I don't have the ability to structure funding long term flexible adaptive mechanisms that will allow you to use that systems practice that I mentioned. And so I guess what I want to leave you with is where are we in this and how can we break out of it. And I'll come back to the work I do with the mid-yard folks. It's on a much smaller scale, but working with all layers within that organization, both the people who are writing the checks, the people who are leading the organizations and the people who are executing, it's a microcosm of our peace-building universe, but working at all those levels and having that conscious effort where we sort of, what is it, the 12-step program start with awareness, right? It's just even being aware of this, I think is a huge step forward for us. And we have a role to play in that. And I know people have to meet budgets and write grant proposals and all that. And I think we do have the potential to address this and to not fall victim to it year in and year out. That's me. Thank you. Perfect. Perfect timing. Nate, you want to come up? Get us... I don't know how... I'm a Mac. Does anybody know how to do that? Yeah, where is it? Oh, is it right there? Okay. Here we go. Forget how to use a PC. Hello, everyone. So, as we all know, it used to be when we talked about conflict assessment and early warning, we were implicitly or explicitly talking about conflict between unitary and monolithic states. So the tools for assessing such things and for building peace around such a framework were very different than what they are now. Now we're concerned as much or more with intrastate conflict or cross-border conflict. And intrastate conflict could telescope down to every level of granularity. You can be concerned about inter-communal conflict or intracommunal conflict, sectarian conflict, political conflict or criminal conflict. So it's become a much more complex and dynamic problem that we're trying to deal with. So because of that, we've all accepted, you know, as gospel truth that we need to be dealing with complexity. We have... there's a million models and tools and things, and we all recognize this. We know that we need to be thinking about second and third-order effects. We know we need to be looking at cross-cutting issues. We know we need to be looking at the different variables and understanding the relationships between them, whether they're antecedent, independent, interdependent, whether they're leading indicators or lagging indicators. We recognize this. We know this. We know that we need to be looking at multi-level analysis, because what happens at the state level is different perhaps than what's happening at the local level. So we've accepted this as truth. However, in practice, it's much easier said than done, perhaps. In Nigeria, where I work mostly these days, I think Nigeria is a... sort of the poster child for complexity. You've got issues in the Niger Delta. You've got that trending in a positive direction since the Amnesty program in 2009. Meanwhile, you have... you have the issues that we've been seeing on the news lately in the Northeast. And then you have issues in the Middle Belt, which are very different, although some of the stuff in the Northeast is now affecting stuff in the Middle Belt. And likewise, as the election approaches in 2015, what's happening in... both of those areas might start affecting what's happening in the Niger Delta. So it's extremely complex, and it's much harder to address in practice than it is to simply accept the fact that, yes, it's complex, and we need to take a systems approach to these things. So just quickly, looking at specifically what happened in the Northeast recently, in 2009, there was an insurgent uprising in Bauchi, which then spread to Borno and a few other states. After the leader of the uprising, Mohamed Yusuf was killed. Violence dropped down almost to baseline. And then in 2011, it began to spike increasingly, becoming worse and worse over the last couple of years. In 2013, in May, there was an emergency, a state of emergency declared. And then in April 2014, there was a second incident that we've all heard about. So this is Borno State. This is the data we've been collecting through our systems and our networks. And the red line is Maiduguri. So that's the capital of Borno State. And you see at the very apex of that line, that's May 2013, when the state of emergency was declared. You can see that it had a huge effect on the trend of violence, right? It dropped. However, what you see there is all the other local government areas in the state. And that had an extremely adverse trend since the state of emergency was declared in May of 2013. So over here, you see these heat maps. Before the state of emergency, you see almost all the violence was concentrated around Maiduguri. And then during the state of emergency, this was about six months later, you can see it's almost like there was a fist that stomp down and spread all the violence across the state. So to me, this shows how hard it is to actually deal with complexity or a systems approach. Too often, you hear in the public sphere, oh, they were doing too much, right? Before the Chiba incident, they're being too aggressive. They're being way too hard in their approach. And then after the Chiba incident, they haven't been doing anything. It's obviously much more complicated or complex than that. We need to be taking a systems approach if we're going to be successful in dealing with all of this. So back to fundamentals now. We talk about second and third order effects. Clearly, we need to do a better job of dealing with this factor. Whenever you have any kind of an intervention, it's not going to be enough just to defeat the enemy. You're going to have blowback of civilians that are then exposed to insecurity as a result. So that has to be a part of your calculus. You've got to look at cross-cutting issues, social, economic factors, political factors, security factors. You've got to look at gender as a cross-cutting issue and youth as a cross-cutting issue. Right now, one of my colleagues is in Conno, and she's doing a workshop to try to create some data around violence against women and girls. Antiseed and independent variables, multi-level analysis. We talked about how it's very important to understand that it's not enough just to look at Borno State. You've got to be able to look at the local government area levels as well. And most importantly, the framework has to be iterative, flexible, and relevant. And in order to be relevant, it has to be locally defined. You've got to have stakeholders on the ground who are defining what you're going to look at and how you're going to look at it. And finally, don't forget about the back burner. You've got to look at Delta as another dynamic in Nigeria. You can see, based on the data that we've been collecting, that I said that it's been trending in a positive direction since the Amnesty program, but in the last couple of years, as a measure of the number of incidences, it's been steadily climbing as we move towards an election in 2015. So we're all looking at Borno right now, but in a couple of, you know, six, eight, 12 months, we might have to be starting to look at other parts of the country as well. So don't forget about the back burner. So how can we do this? How can we begin dealing with issues of complexity when we look at a country like Nigeria or any country for that matter? What we try to do is we try to bring together data sets, whether it's the unlock data set, which is one that I've been working on in Nigeria now for a while, where we work with Civil Society to gather information on pressures on peace, whether it's Nigeria Watch data set at the University of Ibadan, whether it's the Nigeria Security Tracker at the Council on Foreign Relations, has a great data set, whether it's Accoled. As of now, we have about seven or eight different data sets, and we're uploading those all the time, encoding those, and we look at multiple sectors. So we look at demographic pressures, refugees, economic pressures, group grievance, and then you can drill down to about 70 different sub-indicators underneath that. And the purpose of this isn't just so that we can make pretty pictures and say, boy, isn't it complicated, but so that stakeholders can begin to engage with the data in a practical way, whether it's the traditional leaders, whether it's the USAID Conflict Advisor in Abuja or anybody else. And finally, we try to map stakeholders. So we go out and have workshops all over the country, and people identify their organizations and initiatives as peace agents. And then we map that so that folks who want to work on stakeholder mapping, networking, scoping can be more effective in that, and you can make that link between the hotspots and who's doing what, where to address it. Delta State, just to illustrate another way this data can be looked at, you can drill down to the sub-indicator level, and you can see this is a picture I took in Northern Delta State, which illustrates conflict between pastoralists and farmers, but you can see how the issues have changed over the years, from militancy in 2009 to now it's more about ethnic and communal tensions and criminality. So multi-stakeholder collaboration is essential. Without that, it's not going to work both on the data side and both on the implementation side, because people know different things. You need to have buy-in by everybody. They need to have collaboration. But at the end of the day, you've got to take this data back to the communities themselves, which is something that we're doing throughout the Niger Delta and the Middle Belt now. You've got to, because all of this will tell you a lot about patterns and trends, it will tell you nothing about causes and effects. So to get to the causes and the effects, you need to have people on the ground who are working with the nitty-gritty to actually analyze the patterns and trends and begin to strategize around solutions. So that's the website, where it's an open tool for anybody who's interested in Nigeria, the p4p-nigradelta.org, and I've got a bunch of cards that people would like to have that website and take a look at it. All right. Very good. Thanks very much, Karen. Thanks to Nate and Rob for setting up the conversation so well. What I'd like to talk about today is, and what I'd like to do is show you a series of examples of how to navigate through complex systems and complex problems using a mapping platform that has been developed by colleagues of mine in San Francisco, a new tech company called Vibrant Data that has just started embracing all kinds of challenges in this space. And the basic message that I want to get across is that when we confront the huge complexity that we often do in the conflict and peace-building world, we don't have to despair or bury our heads in the sand and push it to one side. We can actually embrace it, dive head first into it and come out the other side with some simple but hopefully robust guidelines for creating priorities about how to take action and how to intervene, or as Rob put it, much better to engage with the system in question. So I guess in terms of the questions that Karen set up for us, what kinds of questions should we be asking in the peace-building world and what sort of tools should we be using, I just wanted to start with a question that I was asked when I started my previous job at the Skoll Global Threats Fund. I was told on day one that our mission was to find a transformative or game-changing intervention that we could make in Middle East conflict. In a Israeli-Palestinian conflict in particular. Yes, laughter. So if you've been following the peace process lately, you'll know that I've failed completely and utterly and miserably in that, but I feel like I'm probably in good company there. John Kerry and Martin Indic couldn't fix it either, so I don't feel so bad. But what I did at least take out of that was to begin to grapple with what that question actually meant. What would it mean to create a transformative intervention? And in the process of grappling with that question, it led me to this collaboration with my colleagues at Vibrant Data and some of the work that I'm going to talk about today. So just perhaps to frame that question in a different way. How, when we have a multitude of options for engaging with a conflict system, how do we choose one over the other and what basis do we make that decision? How do we justify that decision as having greater or less a likelihood of success? Also, how do we avoid what Daniel Goldman called in a recent book, systems blindness. That is the possibility that we'll make an intervention at some point in the system and then get totally blindsided when that intervention unleashes a cascade of effects that circles around and hits us from behind in some sort of unforeseen way. We've all heard examples of that and Robin Nate just spoke quite eloquently on that point as well. So, in trying to avoid that, it occurred to me that the first thing you would need to do, at the very least, was to create some kind of map of the system. So you had some kind of idea of the basics of the architecture of the system. So you wouldn't be entirely groping in the dark exactly where that lettuce in the Middle East I'll circle back to. But for now, I just want to take a sideways leap and here is a map of a complex system and we all know this beautiful example that was hit the front page in the New York Times and was on the Daily Show and everybody took a shot at it and had a bit of a laugh about it because it was how could you possibly understand something like this, right? So I think there's a serious point to be made here which is that whenever we see something like this I think we're paralysed by two extreme urges, one of which is to try and understand the system as a whole all at once and say something meaningful about it which causes our head to explode very rapidly and the other is to try and zoom down and zero down on one particular issue, screen out the complexity and just hope to God that you can intervene at this point and everything else is equal and you don't have to worry about the system effects that we know might be lurking in the background. What I want to talk about today I guess is kind of a middle way where a technique for stripping down systems like this to something manageable, something that can give you a strategic roadmap that's simple enough to be comprehensible and to be actionable but at the same time which still takes into account those system properties and hopefully doesn't lead to you getting blindsided. So around the same time I was grappling with my challenge my colleague Eric Burlow who's the founder of Vibrant Data was decided to use this opportunity to give a TED Talk about this issue and to make a simple point about the kind of methods that you could use to reduce these complex problems to something more manageable and he started by converting this into a network diagram and starting with the point that the whole objective of this was to increase populist support for the Afghan government and if you then turn this map into an ordered network this is what it looks like and if you look one, two, three degrees of separation away from that ultimate goal then actually the network reduces quite quickly in size to about a quarter of its original size Eric's actually published some work in the Proceedings in National Academy of Science suggesting that even going two or three degrees of separation away is perhaps more than you need to that often effects in these networks are quite localized to the first and second connections but then you can look more closely at these 25 connections and you can categorize them into different categories if for example you're a nonviolent civil society actor trying to get involved and engage with this conflict somehow you might screen out those that are not actionable things like the terrain you might exclude actions that require military capability and you might come down to actions that as a civil society organization you could actually take about seven of them coloured in orange there and you might notice that they can also be grouped into two main categories so after starting with this incredible hairball of a complex map all of a sudden you're down to a couple of sort of basic strategic directions that you could potentially take back to your organization you could take to your community of NGOs and try and move in the same direction to engage with the conflict so to be fair you know that was a bit of a throwaway example everyone was using that as a throwaway example Eric's not an Afghanistan expert he didn't develop this analysis in consultation with Afghanistan experts as ideally you should but I want to switch gears now and talk about an example where he has just done that on the issue of personal data on an issue that's increasingly pervasive in our society and increasingly pervasive in peace building and conflict so the challenge is to do with the personal data economy and Intel Labs actually came to vibrant data with a question which was what are the core challenges at the root of the personal data economy and how can we influence them so if data is the new oil as is often said how can personal data be made to work to the benefit of the public at large rather than having people be exploited as a result of it so they gathered 50 experts on this issue and asked them to name the key challenges in this arena they came out with about 100 different issues so this is clearly a complex system things like the growth of a market for data services the development of tools to anonymise personal data the ability to see personal data exhaust increasing the amount of data voluntarily shared and so on and then they asked how each of these 100 or so issues are connected to each other are they connected and if so how so for example if you make an improvement or you solve one of these challenges does it make every other challenge in the system easier or harder so for example if you develop tools to anonymise personal data you might increase the amount of data that's voluntarily shared so an improvement in one node leads to an improvement in another that's not necessarily the case some of them might go the other way so they looked right across the system and lo and behold once you go through every single pair of issues they came up with about 6,500 different links and so you know we're back where we started again with this horrible hairball that's very intimidating and makes you want to run away so what do you do now so this is the key point that I want to make today is that as awful and impenetrable as this looks to begin with it does have structure there is structure within it and you can look for it you can find it you can tease it out and you can use it to your advantage so for example you can look at the asymmetry between links so for every issue in this network does it have a lot of links going out of it affecting other issues and not very many coming in which is something you might like so you don't have much competition to influence that particular node but if you do influence it it impacts a lot of other things you can also look at how strong are the effects that that particular node has on other issues in the system and you can end up with this subset of issues that have a disproportionately strong effect on a disproportionately large number of other issues in the system and you might argue that that is a subset of issues that maybe fit the definition of potentially transformative or potentially catalytic they then ask the experts to look more closely and zoom in on that group of issues and see if they could be categorized in any way and they manage to categorize them into four different groups and then came out with these sort of what they call four grand challenges of the new personal data economy platform openness data literacy digital access and digital trust which Intel Labs are now creating incubators around to try and engage with the space so an example of how you can get from something terribly complex to something relatively simple so I mentioned we were looking at the Israeli-Palestinian conflict we did a similar kind of process at least a pilot process with a small group of experts to ask the question what issues could impact the ability of a nonviolent movement to get to a successful resolution of the conflict defined in terms of creating security, mutual recognition and reconciliation and civil and political rights for all parties concerned we identified 73 issues, 1874 links and identified six issues with that characteristic of high link asymmetry that is not many issues affected them but they affected many others we also looked at positive and negative feedback loops to go to Rob's point about vicious and virtuous cycles if an issue is embedded in a positive feedback loop its effects could potentially be amplified so if you improve that issue if it then links to others that feedback and amplify it again then you can get into a vicious or virtuous cycle depending on whether you want that thing to happen or not so there were a number of this is very much work in progress still but there were a number of promising signs there that we could actually whittle that very complex system down as well we've also been dabbling with a visualization of the regional dynamics in Syria which we think also has some potential but I want to finish with on a slightly different note it doesn't have to be issues that are the nodes in the network it can also be organizations you can look at the ecosystem of organizations and ask the question does the ecosystem of organizations that you work with have a structure and can we use that structure to identify potential synergies, inefficiencies overlap and new ways that we could collaborate together for example so this is data from the water and sanitation water and sanitation funding space this is from data from Brad Smith and his team at the foundation center and it's a cluster network diagram of funding organizations the funding node is a funding organization and they're clustered together if they fund a lot of the same grantees and they're far apart if they don't fund many of the same grantees and you can see that using this clustering algorithm it sort of sorts out into about eight different funding communities which are colored different here but you can also zoom in and have a look at the funding profiles of individual organizations so Google.org is sort of funding in a small space with relatively similar funders the Gates Foundation has a more diverse profile Hilton Foundation a more diverse profile again and even smaller organizations like the Peninsula Community Foundation is really sort of punching above its weight in terms of funding a real high diversity of different organizations with a relatively few grants so I'll leave it there but I just want to finish with you know just reinforcing a couple of simple messages that we don't have to be paralyzed by complexity that we do have tools now that can help us navigate through complex problems and extract some simple guidelines for prioritizing action we even the most messy complex problems have structure to them and if you can extract that structure you can find it and put it to your advantage and make it work for you and lastly nothing that I've talked about today would be a substitute for granular expert knowledge in fact all these analysis are built on top of granular expert knowledge and once they're done they should be returned to experts to decide whether they make any sense or not and how exactly they should be implemented at the end of the day these analyses are not about providing answers really so much as they are about hopefully enabling people to ask better questions about the system that they're engaging in I'll leave it there, thanks Thank you for framing so clearly what the questions are about systems and systems thinking and the path that goes from seeing patterns to working with them so there are two main distinctions that we see in the systems thinking one is using these models and methods to look backwards so how can we use the data, the perspectives the relationships, the linking models, the networks to see what happened in the past and there are lots of wonderful models and methods to do that and that was where I began my work in systems but I'm also an actor, I'm a decider, I'm a leader I am a practitioner and so that wasn't enough for me I needed a way to think about systems that would take me forward and lead into action as well as sight now it's important to be able to see the patterns it's very important to see the patterns you can learn lots from them and in a moment of action you need patterns that will also lead into action the other piece is a distinction between system models that come from the top deductive models where you have a conception, a picture an understanding of the pieces that would be important and then you build a systems map from your beliefs down into reality the other is an inductive model where you take data, as Nate does and you build up from the detailed data so as you're thinking about doing systems work those are two very important questions you need to ask are you looking back with your systems work are you looking forward and are you going from ideas to a system model are you going from data up into a systems model no, not even nice, they're all useful but they're important questions to ask now what I want to share with you today is a theory of change that is based on one particular systemic perspective so I love the question this morning about what would be the theory of change for the gamers in piece and I think this is a theory of change that works with gaming though I loved his solution that the theory of change builds the game and not the other way around is lovely so what I'd like to do is to define for you the way we think about systems now this is a complex adaptive system individual agents they have the freedom to act in unpredictable ways but when they interact patterns emerge that are systemic patterns agents, interactions and patterns so you can imagine that there would be multiple ways that the agents might interact and as they do over time the pattern appears now this pattern can be a peaceful pattern or a conflictual pattern it can be a pattern for an individual in terms of fear or courage or wisdom or it can be a community pattern of ethnic or cultural perspectives or rituals or choices it can be a political pattern of governance at any level patterns emerge over time now those patterns then constrain the actions of the agents in the next cycle of coming to be so once a pattern begins to form the degrees of freedom for the agents in that system are constrained in the future and so we have agents, interactions and patterns and then perpetuating the pattern in the action of the agents now this is happening at many scales in many ways and if you begin to think about the peace and justice context in which you're looking and you think about this kind of causality not A causes B or even non-linear causality A causes B causes A but if you think that this is the way that patterns of peace or conflict emerge then there are three really important questions you want to ask what are the agents that are engaged and that's the kind of work that you're doing Scott is finding what those agents are how do they interact and interactions can be physical, social, informational with money what are the interactions that are there and finally what are the patterns that are showing up what are the agents what are the connections and how are they engaged what are the boundaries that lie among sets of agents how can you change the conditions at the local level so that the pattern shifts itself so an intervention does not become A to B an intervention becomes how do we look at the existing system understand the conditions that are shaping the pattern that exists and then influence those conditions so that different patterns might emerge now the trick is you can't predict what's going to emerge because you're not the only one setting conditions for that system so these are questions that you can ask if you think possibly this is the way things are happening now the next set of questions is if you see a pattern in this way how do you deal with the fact that you can't predict or control what's going to happen in response to your intervention and so all of you talked about the fact that action had to be flexible iterative, local, sensitive multiple scale and we call that adaptive action and it's three questions what what do you see, what do you think what are you thinking so what and obviously the last question we could do it in unison right now what so what's the current pattern that's coming out of this complex system so what does it mean in terms of the potential for violence or nonviolence for support, health coherence or dissonance now what can I given the resources and access that I have in this moment what can I or my institution do to shift those conditions but because the system is complex because we at least part of the system is living in Rob's cloud we have to think well now what's going to happen so that total iterative engagement with the system what so what now what so as each of you was talking I was thinking about the ways in which this is an operationalization of the complexity that you were looking at one other important distinction and set of questions that I'd like to ask adaptive action can be very short in fact we find that practitioners in the field do this spontaneously and very well and the distinction between someone who is good in a moment and not good in a moment is the quality in which they do this process it can also be very long term one of our associates is working with the World Bank and looking at agricultural policy from an adaptive action perspective of multiple scales so it can be microscopic macroscopic same process one other distinction I want to share with you and this is also one that everyone else has pointed toward we like to talk about the distinction between a finite game and an infinite game and a finite game now what the boundaries are you know what the measures are you know what the goal is you can get better and worse there is such a thing as prediction and control and expectation and the purpose of such a game is to win these were the ones that you were talking about Rob in terms of let's solve this problem is this a fixable problem if the problem is fixable if it's a finite game there's a fix and there's an expert who knows it and you can get them in to fix it unfortunately those aren't our real sticky issues we get stuck on the others the others are infinite games and an infinite game has no boundaries has no set score card has many many different ways to keep score you don't know who the players are and who the observers are and they switch all the time anyway and someone can be your opponent is this sounding familiar in many of the context that you're working in infinite game and the purpose of these games is to keep playing not to win or to lose so that what you're doing perpetually and this I think is what Rob was talking about is engagement is engaging with that system to play an infinite game now you can play a finite game inside an infinite one so you can zoom in on an individual project you can put in a well even though you know the real issue is a cultural transformation and climate change in the community you can go in and do a finite game and put in a well and know still that that's in the context of an infinite game so you zoom in and do a fix that is possible zoom out and realize that you're continuing to play the game and so it isn't that either one of these is the only way to play to have capacity in both and to be able to choose wisely what are you seeing finite or infinite so what does it mean for your options for action now what can you do and then yet again so the questions that you'd ask would be what game are you playing so what game is most fit to purpose in this moment should we be playing a finite game or an infinite game and how do we go about doing that and now what can we do to frame that game in the deeper complex system where there are the parts becoming coherent as individuals or families all becoming coherent as larger governance spaces political spaces, cultural groups and the greater whole which is the massive interdependency of the multiple scales that you're looking at what so what and now what and there are resources here which we'll share and I'm hoping that we'll have a good opportunity to speak more thank you we do have some about minutes for discussion thank you everybody for your comments and for being so succinct amazingly I was worried that when I saw this slide Dex I'm like please be sure it's not more than 15 minutes and you all were fantastic so thank you we've got time to talk and since I'm up front and got the mic I'm going to start but I will open it up here in a minute and I want to kind of draw attention to the role of visualization in grappling with complex systems I mean all of you at some level have really talked about visualization we talked about you know the problem with levels of analysis and the connections no one really talked about time but obviously time is in there some things happen slowly some things happen fast and some things are truly dynamical where some very small input to the system creates a whole dynamic that you would never have imagined I mean the reality is is these are unbounded systems and we can drill down but we are working with infinite games and agents that have changed the rule sets for us regularly so I guess I want to see if you guys can help us a little more effectively speak to the role of visualization in as a practice and in working with our communities and our colleagues as a means to reify some section of the system so that we can align ourselves more effectively not only for that first step of action but for the learning and also if we could talk about and I don't know I think we will have a space to put more resources but some of these things are high tech and some of us are in environments where we need really low tech solutions I mean we need techniques that help us to visualize and elicit these kinds of things without the data required for a heat map or you know the complex networks you know what kinds of things are available to us to start to do this with our communities and our networks to not just have this a non holistic and a non systemic so basically like the role of visualization and I guess since we haven't heard from you for a while we'll start here so one thing that all the visualization techniques have in common and I actually want to place those on a continuum from more quantitative models and more qualitative sort of maps of a system that all of them though are I think what they have in common is that they are ways to help you see the whole even though you are stuck in a piece and those that you're working with are stuck in pieces and until you can see the whole it's just like any looking at the Grand Canyon from the top of the ridge and down in the valley I mean it's from your feet up and different possibilities will suggest themselves and I want to maybe call attention to two I think really important impacts of the visualizations one is on the more qualitative side visualizations become these more complex systems maps that I work with are are really just narratives they're rich stories that capture a wealth of information but tell you the interconnections and talk about the patterns that talk about the so help you get to the so what that you were talking about Glenda and so you can think of that the narrative and it still goes to the low tech question so these are the most ancient form of like human technology is the story and if you change the story if you get people to tell the story from the perspective of the whole they see what they should do they see their agency they see opportunities as you were saying Glenda they see the what to do differently if looking from that point and I've had a couple of experiences presenting systems maps to an audience that tends to have a very short attention management systems after ambassadors in various missions and in both cases in the one case the this one ambassador went from being highly skeptical we could show her anything she didn't already know about this particular country to having viewed the systems map and I'll credit Chip Walker who's in the audience to be the master who presented it at the end of it she said one of the systems maps went up my eyes kind of glazed over but it was the most useful presentation I had ever heard in this case Cambodia and that's where this it was just the power of the story you're able to tell from the map the map becomes a table of contents for telling this rich story and holding truth that you really can't hold by knowing any one piece and looking at any one events the other one is I want to share a person I met a couple months ago who is a particle physicist and he's his name is Garrett Lee see he's got a TED talk so LIS I at least see and in this Ted talk he talks about more of like the pictures that Nate and Steve were I mean we're talking about which is looking at he took all the subatomic particles that are known so a couple dozen of them right and I can't tell you what they are and he looked at eight dimensions of relationships we plotted eight dimensions of relationship which I can totally begin to explain to you but what was really fascinating about this is that they produced these diagrams they produced these visualizations that suggested where new particles might be found and he said by looking at the visualization you could see patterns and in those patterns you could see gaps and in those gaps you could theorize what should be there and then physicists could start looking for those things and that's how they go through the process of theorizing and then experimenting and uncovering these various particles and I think there's an analogy it's not quite as neat on the peace building side but we can look at those maps and now that we see key leverage points or we see gaps or opportunities that we wouldn't see if we didn't see the pattern and Nate can you tell us anything I mean are these kinds of heat maps are they easily accessible or is it something also we could do in a low-tech way with communities well what we do is we recognize that our local partners might not have good internet access for example so we we try to produce bulletins and publications and things and then also recognizing that they might not have email we then go and take it to them and we have a discussion around it so we found that people who are living in a particular local government area in a particular state you would think that they might know everything about the conflict dynamics that they could just reel it off but when they look at a map new insights definitely emerge and so we do find that a hard copy is essential for actually stimulating that kind of discussion and then at the national level we work with a working group with the larger NGOs and the donors and things and then we have a more high-tech engagement with the platform but yes a low-tech option is absolutely essential for our particular tool if it's going to be successful at engaging the local stakeholders in the process that's awesome and I guess similarly Scott like if you some of these tools that you're doing you know do you imagine could people draw simple maps connections I mean it wouldn't be quite the same but do you imagine that there are ways that you could start to map your networks or your stakeholders or your organizations and how close they are to an issue and do that kind of thing and then how would you use those with policy makers or communities yeah well I think the tool that we've produced is I think it's of relevance to the peace-building community here in terms of figuring out where to engage with communities on the ground you obviously if you want to use the platform you need an internet connection at a minimum and you know technology is hard to build and it's complicated but once the platform is designed to be as simple to navigate as possible one point I would make and I just want to go back to what you said about the time aspect of this that what we're doing generally when you create a systems map and it's a snapshot of a continuous process so you know you create this map you put it on the wall and six months later in a sense it's out of date and you know one of the hidden assumptions of what we were doing I guess is that all of these connections exist these 1900 connections amongst different issues in your Israeli-Palestinian conflict the default assumption is they're all acting similarly at the same time whereas in fact you know a few of them will be on the front burner today some different ones will be on the front burner so there's not an easy way to get around that but one point I wanted to make which may not have been immediately obvious is that this is not intended as a snapshot this is intended you know so in the personal data economy example we've got 50 experts in the room and the idea is to create engaged communities of experts who will continuously iteratively use this platform to improve their understanding of the system as it changes over time it's not a sort of once for all time thing there's actually embedded in that platform a spreadsheet that can drop out that has all the data in it that anyone in the community can edit and update at any given time so it's an iterative process and it can be done if there's an interconnection or if there's a possibility for collaboration with communities in the conflict zones that we're talking about and the one other thing I mentioned is that when what participants in these workshops universally said was the just the richness of the conversation as they tried to decide what are the issues how should we bound them how should we define them and how should we decide whether they're connected or not was in itself an enormous virtue that in the sort of narrow day to day they never took a step back and considered the system like that so everyone found that very beneficial I think if that's done in collaboration with communities on the ground it can be a really rich source of engagement as I said which can be iterative over a long period of time I'm going to jump on that and I know Glenda will have some thoughts because I think one of the ideas too is that these reifications these little maps or these diagrams are really meant to be continuous and so it really does involve developing a set of practices and communities of people who want to work in this way and Glenda I know you work with some organizations that are trying to do that I mean is there some things you can tell us about how you get people engaged in that way you know why does it take this someone really convening for people to realize wow we should be having this conversation more often and more regularly and then everybody runs back into their hole I mean I'd love to see where those things are taking place and after this I'll open it up open up here well there's so many points and what all of you have brought up a really really key and I think one of them is that the model itself is an occasion for conversation so it's an occasion to collect relevant data it's an occasion to make sense of that relevant data it's an occasion to build action together and so that's I think it's really important sometimes that complexity folks think that the complexity is an end in itself but it's really a means for setting conditions for a better conversation so keeping that in mind is very helpful one of the things about that that makes it much more tractable in a group is to realize that you're not modeling the world so often when people start to do a model they think they have to model everything but if you say instead as a French you're problematique so you're modeling an issue modeling a challenge modeling a question then it helps you simplify the models from the get go a smaller set of agents to interact with and the fact that there are multiple methods so no one of these methods is going to be right sometimes it'll be one sometimes it'll be another so the broader capacity you have for systems approaches the better off you'll be then in terms of the way that we visualize in just grassroots fundamental is that when we talk about a pattern coming out of a complex system we say there are three things that mark the pattern how are things the same how are they different and how are they connected and if we can just get a group to talk about in a problem what's the same what's different and what's connected then that's a model that they can begin to work from so even that very simple kind of conversation helps people not just get a better picture of the whole but a shared picture of the whole everybody has put into the same picture so we work with this idea of collective sense-making and also Peter Woodrow's in the audience is back there from CDA but we were working on presenting a map once and someone who's a mathematician said oh I get what you did you did performance art and what he meant was it wasn't so much the map that was important it was the act of creating it and interpreting it and living it and talking about it that was what was important and Rob dances on the stage and he talks about his math so interpretive yeah I'm gonna go ahead and start taking if people don't mind like speaking loudly I can just we don't have to get to mics or whatever and oh they asked the people on the webcast if you could wait to ask your question until you have a mic that would be great for all those watching online let's get a mic to Dan Rothbart because I know his name and then the gentleman let's say oh sorry we had how about if we come down here and get a mic there and then go ahead Dan Rothbart School for Conflict Analysis and Resolution I'd like to thank the panelists for really providing an exciting perspective here that's very valuable in many contexts my concern or question is the involves the use of this for interventions it seems to me I'm reading between the lines here that to determine whether an intervention is positive or negative that itself raises a whole series of questions about predictability and since the systems are in a complex system and that complexity is not only at one level but it's at multi levels as was implied even at a microscopic even complexity of people in a room in a dialogue session in a mediation could be considered a complex system so my question is how do you deal with that the layers of complexity with prediction that seems to be necessary to determine whether it's positive or negative I'm going to go ahead and take another one before we start and maybe even one more go ahead sir I'm Matt Levincher from George Washington University I was very taken by a couple of remarks in this panel one was this notion of the transformative intervention which seems it's a lovely ideal and then on the more sobering side Karen's comment about we should have this conversation more often and then everyone goes back into their hole and I think we have conversations about transformative interventions and then our respective holes and that is what I applaud about this panel is that you're really encouraging us to think and act more systemically so the question is that I have and that I'm grappling with in my own work is how do you transform networks of information sharing into networks of action and it seems to me the transformation we're looking for is to build communities that act in concert rather than simply understand the situation in concert and how do you build that how do you sustain that okay and unfortunately I've been I'm getting the we're going to have to start a video in four minutes so we have the idea of the causality question and how we can judge our interventions and the need for an understanding what are the problem with causality and then Matt's question about the transformative aspects of working system so can I get who can I get to start I can just say on the positive negative thing if I understand the question right I think there's never any substitute for having a clear objective in engaging with any problem I think in the Afghanistan example is populist support for the government in the Israeli-Palestinian example there's sort of three benchmarks for getting towards a resolution of the conflict and I think you judge your engagement or intervention based on whether they're bringing you closer to or further away from those objectives so it's not a matter of just engaging at any random point with the system but keeping in mind how it's ultimately reverberating through the system to have an impact on the ultimate objectives that you're interested in anybody else or go ahead Rob so I just maybe to Dan's question a little bit is that so one is I'd say prediction is much less important than learning and so even though it will be difficult to predict how to affect a certain part of a system or dynamic within a system or even a factor within a dynamic within a system you want to be in a situation where you're set up to learn whether in fact you had it right either the understanding of the system or how to engage it right and then to the degree that you're not you're set up to be able to adapt and so I use the phrase fail smart you want to be set up at the least if you're going to fail fail smart so you minimize the damage and you maximize your potential to learn from what you do and the other one is I work a lot with people on goal setting and they want to set goals in terms of substantive deliverables one maybe two three year frame you can't do in a five year frame when you're talking that far out you've got to talk about changes to underlying dynamics and patterns not just to events or conditions and that I think hooks the first question to the second question because things are uncertain in infinite games there's nothing that's going to make it certain so the question is do you have a way to deal with that certainty or not and if you do then how do you move into action and so the only way that you become more certain about the system behavior that you learn is to do something and so it leads into this adaptive action cycle so when someone or a group is talking talking talking talking it's an information sharing space they're stuck in what we call the so what space and if they'll just do something and see how it affects the system they'll get unstuck they'll begin to move again so this whole question about the adaptive action doesn't always work but when it doesn't work you can have some idea what to do and it doesn't really matter what they do to some extent just to do something so that the system begins to move okay I'm getting the wrap up can you do it in 30 seconds? yes sure, just a very quick response to the problem of everyone going back into their holes after having these wonderful rich conversations I think that was the purpose of creating a platform that can be a visual representation of a consensus expert consensus about the shape of a problem and promising and less promising points for intervention and to keep that platform alive and keep that conversation keep people returning to that same conversation over and over the reason I think people go back into their holes is it's been too hard up till now to keep a complex system there and I think we've got the tools to do it now so please at breaks I'm going to have to wrap it up here but please at breaks and across the next couple of days engage these folks engage each other I think there are places on the website where we can put up resources I think there's low tech resources there's high tech resources that are now in our video. Thank you very very much for coming I hope it was useful