Oliver Sparrow is the Director of the Challenge Forum. He has spent the bulk of his career in Shell, chiefly in strategic planning. He left in 1996 to set up a think-tank operating out of the Royal Institute for International Affairs, aimed at generating an overview of the challenges of the next twenty years - the Challenge Forum is its consultancy arm.

We are interviewing experts in foresight and policy makers and we are asking them how they scan for wild cards and weak signals and whether they use that in their work. We have set up an online community for experts and those interested in foresight and we are hoping that they will submit their own wild cards and weak signals and maybe talk to each other about the future.
 
 I think this is a flawed methodology. What you will end up with (and we have seen this with five or six government departments in Britain and in many other experiments) is a heap of bullet points, with no connection between them. This is almost totally useless. When I was in an oil company, I had about 20 people looking at competition. They went cut out every item they could find in the newspaper and entered it into a database. A vast heap of pointless facts began to emerge. It quickly became clear that you needed a thinking model of what mattered, how individuals played within that, and essentially what the key dimensionality of the problem was. Then, rather than digging for isolated facts and incidents and wild cards, one actually then understood a flow, and the fact that the flow could have two or three contrary forces operating within it. So many outcomes at least meant you talked to policy makers in a way which brought out the key dimensions. These are the things that really matter. You can have plans – for example, the military have hundreds of thousands of people working on a really circumscribed set of issues, such as an amphibious invasion. When you have the whole world as your domain, you simply can’t write down everything that might happen – or at least if you do write it down, you can’t possibly read it. I know of someone who did a study for DEFRA where he ended up with a very large room covered in 12 point font bullet points from floor to ceiling on all four walls. It was useless. That is the weakness of an approach in which you gain an immense amount of stuff. The fact of someone having written it down does not take you any further, it’s not useful. Sorry, that is a depressing point, but I felt it was important to say it.
 
We have had these comments before, especially about connecting and elaborating on them. We responded to that and we take selected wild cards and try to elaborate on them what possible outcomes they may have and connect them to weak signals and other wild cards.
 
 I do commend the scenario methodology. If you have three or four well constructed scenarios, you will find that almost every wild card, other than martians landing and giant meteorites, falls neatly into those scenarios. Exogenous events, so an earthquake is just an exogenous event – you can’t do anything useful about it in risk management terms or preventive terms. Looking for utility: how useful is this going to be? Utility only works if it gives you some mechanism of reducing the risk or some method of closely constraining its consequences.
 
 What is your background, and how did you come to do what you do today?
 
I was in the Foreign Service, then Shell Planning and then I was at the Royal Institute for International Affairs. Currently I run lots of little companies.
 
 Do you now run scenario workshops for companies and other entities?
 
Yes. They tend to be ‘between’ organizations, rather than ‘for’ them. By that I mean we tend to work with a large organization to produce conversations with its regulator; or, as I am doing in Mexico, we get a major organization to talk with its innovation and renewal end. The object is not to produce scenarios, but one of a dozen tools.
 
 In your methodology, do you use a concept such as a wild card or weak signal?
 
No, never, because you typically have one of three situations. Either people don’t know how to talk about the problem at all, they have not evolved a language or model for thinking about it; or that model has evolved but not propagated sufficiently through the organization, you have slightly different models or wildly different models. Often there is an official model, which is all about safety and quality and being nice, and another that is all about being nasty and cost cutting. Those two often run in parallel and often people have not assessed their values correctly. They will go along making decisions with their eyes closed. They will not look at that side of the world because it is often too frightening. It is like that old analogy: ‘Why you are looking under the street light, even though you lost your key somewhere else?’ ‘Well I can see here.’ That happens with companies. They feel comfortable in an area and so look for solutions there and won’t go and look in the dark for solutions, so one can push them in that direction. The third situation where you want to use scenarios is when there is some area of deep contention in the organization, where it’s too painful to address the issue in the present, but if you put it 10 years in the future they are all happy to talk about it. For example, in human resources policy: if you say: ‘You have to start employing a more international group of people,’ or ‘You have to bring your technologies into general management,’ the instant reaction is: ‘Hey, that is my job’. But if you say: ‘Let’s think we can do this five years in the future,’ it is all quite painless. Specifically, looking for wild cards one looks for an educated mind. We put people together in a room and very dedicated minds have explored much further than the domain in which the company is happy to exist, but they have never had anywhere to put their thoughts before. As an example, I was working for a South American oil company which discovered a really large deposit. They were their essentially their own regulator – there was an ineffective Minister of Energy and the country was extremely innately corrupt. The issue of how to control this, and whether they needed an external regulator and some law to control legislators was an area of discussion that they could not handle on their own. All of them had thought it, but nobody felt they could say it. So we got them together and there was a huge row and everyone got very angry and then they settled down and eventually they sorted it out. They have all got the good ideas. If you go into a technology company, anyone interested in technology has put themselves 10 years in the future – they simply don’t have the machinery to table what they have thought. The second point is you that need machinery. It is not enough just to do it and walk away – you need a conveyer belt onto which ideas are poured, evaluated and funded. That does not happen on its own. I look for process within organizations and there are four elements one is looking for. There is what we call insight – having a thinking model about how the operating environment works; where are our sources of legitimacy; if we are an NGO, why do supporters support us; if we are a commercial organization, how are we making money? That bundle of concepts needs to be clear throughout the organization. Secondly, values need to be clear. These are not ethics, they are decisions we take, e.g. how we employ people. They are not objectively saying ‘this is how it should be’, it is just historically how we have done it, or this is where our brand sits. You have to ask yourself, does this make sense? Take, for example, a company made clothing and cloth. The values present in a company making commodity cotton cloth were about low cost and competing with China, whereas the fashion end of it could not cope with that – it did not share the same values. Values are very important. Unless you have some sort of harmony through the system, your decisions are always going to skewed, because you won’t have the best information. If you say as senior management you want to do something, and no one can understand why because it goes directly against a huge faction of the workforce, they will think you are stupid and won’t do it. The next thing is options, which broadly concerns the dimensionality of the choice open to us. In a commercial organization it might be internationalism or not; it might be stop all the selling and just make things; it might be to become virtual. There are a set of dimensions along which questions can be posed. That gives you space in which you can say ‘Where am I, where is my competition and where is all that going to be in five years and then where is a viable place to be?’ In the very broadest sense, you can say ‘We want to go that way and not this way’. There is some sense of the geography you are navigating. It’s very close to insight, but it is thrown forward and has choice in it. The fourth thing is narrative –turning all of that into language and reflective understanding in the whole of the organization. Unlike 20 years ago, when only senior management got outside an organization and saw the world, now everybody in an organization is tightly connected to your customers and external contacts. Your young people are your antennae close to the technology. If they don’t know what you want, they cannot innovate to supply it. If you have no mechanisms in place to conduct conversations with them and extract what they understand, you are going to be 10 years out of date. This capacity to have these conversations at all layers of the organization is what makes for the seeds of innovation and renewal. You also need the machinery and a reward structure that picks it up and does something with it. If you have all of that structure in place, plus the machinery to make it tick in the right order, you will get, firstly, sensitivity to operating environments and wild cards and things that might go wrong. Secondly, you will get sensitivity to innovation that might help, rather than things that might hinder. So an organization which can respond to change, anticipate change, innovate and renew itself, is essentially one in which information flows are very rich; and to be rich they need to have these tools and machinery in place. Without all of that, it is like any complex system – you pull out the heart, the liver, and it stops working. When it is all there, the most dramatic things happen and when it is not, they don’t. In many laboratories, clever people sit in utter frustration, because they don’t know what is expected of them or they come up with something really clever and someone will say ‘Why are you doing that?’ Unless there is a clear conversation going on about what we might want if it were possible, then innovation never happens. It is exactly the same about responsiveness to change and about sensitivity to unexpected events. If you don’t understand the structure, you will get surprised. Most people who get surprised don’t have a thinking model. There is a panic in the UK about cloned cows, because people don’t know about biochemistry and are ignorant and it sounds frightening. The press is exploiting this. It should not be a surprise, as this has been done before. If we try to identify a wild card, it is virtually impossible to say which are going to come into prominence, because it depends on how it is talked about and pitched.
 
 In your scenario work with organizations do you try to get a group of people from all levels of the organization to work together?
 
No. If you mix very junior with very senior staff, the junior staff will be tongue-tied and the senior staff won’t give of their best. Far more important is to get people who don’t normally meet, but are of similar level. It depends on what kind of problem you are solving, e.g. in communication or analysis you will do different things.
 
 Do you ever get them to think about options that might have unforeseen consequences or something that might potentially go wrong? Do you speculate about the future and how do you get them to talk about that?
 
That is a pay off – either do A or B, then C or D could happen. That is decision trees and those 1950s structures. They are not terribly helpful. If the problem is immensely complex and you do an analysis, no one will read it.
 
 Do you think it could be counterproductive to imagine a future where everything could go wrong and that would stifle the conversation and stop it going any further?
 
Yes: ‘we are all doomed’ does not take matters further. I like GANT charts, which do help: horizontal bars which say: ‘this is going to be important around then and then this kind of thing will matter a great deal a bit later on’. Then you have a flow of what the issues are. If you are dealing with top-end policy makers, say, trying to give a prime minister a sense of what may be happening over the next 10 years, then you must deal with this very generally. What you are trying to find out is where he is likely to be ignorant and what is the most useful information you can supply in a short time. If you are dealing with a highly focused company, then the issue might be the price of lithium for batteries. Then the scenario variables will be where the lithium will be sourced, price, and what to do under various different futures, e.g. Chinese cartelization of rare earth minerals. If what matters to you is lithium, and the lithium is all coming from China, Bolivia and Afghanistan, we might look at what other sources might be and what prices and what does the supply curve look like – the analysis develops nicely. We are trying to look at the very broad nature of the problem, and refine that to a series of answerable questions, and focus down on serious things. There are techniques to say, for example, how much would it be worth for us to know this or that, and put a value on the information, but that is several layers down. That is a class intelligence approach to anything. It is similar in many ways to the scientific method. What is terribly difficult is when someone comes to you and says ‘It is frightening, and can you tell us what is the matter?’ And we say ‘No, we find out what matters to you.’ You can say things about energy supply, for example, which are straightforward and concrete, so you pull it down to definable bits and pieces. How much would it be worth spending in order to avoid what kind of damage and what is the probability? That becomes a set of actuarial questions and you come up with an answer.
 
 Can you envisage any major wild cards, positive or negative, that may occur in the next 20 years?
 
Arriving at a clear understanding of what constitutes awareness and cognition and the ability to put those into a machine that would change absolutely everything, from the workforce, to how we think about eating cows, to how we think about the value of individual human beings, etc. That incidentally will happen in the next 20 years. Go on the internet and look for the word ‘singularity minus black hole’. There is a movement called the ectopians, who say that somewhere in the next 50 years we will find a much better way of being alive and aware. We will move into a much faster, more durable substrate and will do amazing things and go to amazing places. If you then ask yourself, if the ultimate quality of awareness is the same between you and me and my dog and the thing you had for lunch, and the only differences are memory and a few other minor differences, then all our value systems collapse. That will happen in the next 20 years. One the European Union should strongly consider is that global warming is perfect. By that I mean everything has an absorption curve. Imagine you have a clear glass of water and you put in a drop of ink and it becomes darker, and you keep doing that until it is completely black and then adding more in does not make any difference. Well, carbon dioxide does indeed absorb infrared and does indeed produce effects in low concentration, but it is now saturated and adding more won’t make any difference. There are very expensive schemes being set up, like the Mediterranean interconnect, that will be a waste of money. I am not saying this is true, but if it were, the science community and the policy community would be discredited for a generation. Another wild card is Chinese society collapsing under its own weight. The only thing holding China together is the People’s Republican Party and habit. China could go down and take a lot of debts and make the current financial crisis look like nothing. Maize or wheat or rice gets a major epidemic and it is all virtually monogenic –virtually all the corn grown in the world today comes from the same places. At a conference in Russia in 2010 about wheat rust there was a general sense that we could lose one of the major crops. Ireland had this in 1840 and went overseas. We have no overseas to go to. The seabed is lined with methane hydrate, and if you take the pressure off, it turns back to an extremely potent methane gas, and we would be in a pressure cooker. We know in the past there have been great belches of methane, followed by a climate spike. The American coastline has had a lot of extra erosion because of extra agricultural production and there could be a landslide. This would move at 200 or 300 miles an hour, causing a major ‘outgassing’ of methane. A physicist discovers the grand unifying theory in the next two or three years. This would have the same scale of impact as Einstein, out of whose discoveries came nuclear weapons. Somebody cracks RSA, which is a standard inscription. If this was done in a public domain way, the entire banking system and ATMs would stop working and could not be used, as they would be insecure. The whole thing is based on one inscription algorithm. A systemic error in a widely used operating system like a telephone system, such as android, which allowed a virus to pass across the telephoning networking system and brought the whole thing down. This would bring western commerce to a screeching halt.
 
 Can you see any weak signals that indicate that these wild cards might happen?
 
There is a generic weak signal, which is that the more complex something becomes, the more vulnerable it becomes. Technology is our equivalent of the Irish Potato Famine. Let us say something went wrong with logistics. Because all our shops are restocked on an eight-hour cycle, within 24 hours the urban population would be starving and there would be mass evacuation into the country, where there would be no supplies. Anything where the core elements that allow our contemporary complexity to be maintained are weakened or brought down would be systematically disastrous. The US have been thinking about this quite hard – as usual in military terms. One of the issues is that maybe things become too complicated to manage and we have to go to something more modular and simpler. It is a basic criticism of the EU, China and India. It is why India is so different to China: in Indian society, whenever it splits it tends to stick back together, whereas Chinese society does not. A lot of our technology structures are more Chinese than Indian. We may need to think of ways for things to be more local and distributed.
 
 Are there any of these wild cards or weak signals that you think should be given top priority in EU research – of the ones you have discussed?
 
Many are not researchable. It is more useful to look at what EU research should be focusing on. Take energy. There has been a lot of nonsense spoken about renewables. To give you an example, I spoke to 2,600 of the top ten chemical engineers in Europe just after the biofuels fiasco. Not one of them had been consulted on this. Nobody had asked the guys who know about this. For energy systems to work, they need to be giant, unified and homogeneous – completely the opposite to renewables, which are patchy and difficult. We need a new version of an energy currency, taking low grade energy, making it something that can be stored, and moving that stored thing around, e.g. aluminum as one idea. This will work for ocean thermal, remote wind and hundreds of things. In other words, it is a unifying system and will all pull together. Similarly, there are all sorts of silly things said about battery driven cars. Oil is a long change of carbon atoms with a fur of hydrogen stuck on it. We know where to get hydrogen from renewables, so we could use entirely green carbon or recycle the carbon. This would be using conventional fuels, which we know how to use and which won’t blow up – we know how to lubricate, we know the fine details. We are taking an intermittent difficult power supply which is durable and storable and which we know how to handle. There is no discussion going on that I know of within European circles that feeds into policy about these big structural issues about energy. Academics know an immense amount about a tiny thing, but they know remarkably little about their whole field. What is needed are enough people who know economics, sociology, chemistry, engineering and physics, who can talk to wide groups of people and extract ideas from them. These people don’t exist. When we were doing technology foresight here, it emerged that 95% of the important issues had a social or economic basis, to do with users and motivation. To radically change the European science base, we need to stop thinking in these 19th century disciplines and silos, and start thinking about how we will solve this particular issue.
 
 So it is not to do with prioritization, but with revamping the way we think and about and approach problems?
 
Exactly. How I run a workshop is to get people to produce their ideas and put them into some sort of crystallization, and then come up with a prioritization of the five or 10 things we know about and need to solve to make better decisions. If one is running state-funded research, the belief of certain scientists is that you set, deliver and review the tasks and the decision-making process sits outside any problem-solving engine, because it is about deep understanding of quantum mechanics, or whatever. There is a perfectly viable domain there, but that should not be the whole of it. Across Europe, the research element of the total R&D budget, private and public funded, is about 15%-20% and all the rest is development, i.e. the remaining 80%. That is hopelessly badly directed. Academic science does produce amazing results. But I one needs to systematize the development process. I am building a database for a Latin American company of all the people they have and all the things those people know. It is a big network of people who are motivated by the power of being in the network to make it good. That sort of thing does not cost much money and produces a self-aware community which detects the weak signals, because it has been thinking about them for the last 10 years. User groups on the internet know an amazing amount about their subject. In the artificial intelligence or economics groups on the web, there are top flight minds who have been talking to each other for five or 10 years. They are miles ahead of what is going on in the official media. So it is about encouraging the unofficial unpublished chatter and having ‘journalists’ within that. Summing it up, there is an old joke which says ‘I am sorry to write the 80-page letter, but I did not have time to write the onepage letter’. No-one reads 200-page reports. What we need are executive summaries and communications targeted at the people who need them. You need to think like sociologists about how groups communicate, how they identify problems, how they deploy information and resources on it and prioritize, and how they create closure. If you think about it in those terms, you will see instantly there is a structure missing in European knowledge management.
 
 What about methods for scanning?
 
Phillips carried out some psychometrics in organizations to see who was a good predictor. He found that 90-95% of people like orderly competition, systems and clear direction. Such people like to be told ‘The answer is cost control and your job is to sell another 10% next year and you will sell it to that province’. They get very excited and rush off and do it. He called this psychological type hedgehogs. They have one trick, which is to roll up in a ball and stick out their spines. They like clear models. Foxes, on the other hand, are sceptical of everything, and accept no one model of the world. The fox has one big secret and the hedgehog has many and they hate each other. In 2002 Philips asked people to forecast some general things and some things in their immediate area. The hedgehogs were clear and definite and virtually always wrong, because they were single minded. The foxes’ attitude was: ‘On the one hand, on the other, it is rather complicated’ and they were always right. Foxes are about five percent of the population. Find your foxes. It is often hard to extract information from them, because they are sceptical and what they have to say is complex. You don’t exactly go to people to see wild cards, you go to see the insight these foxes have put together. They don’t perfectly match each other, but they generally overlap. These are broadly the people you want to have in your scenario workshops. But if you produce something wholly ‘foxish’, the hedgehogs will reject it, so you then have a translation process, which makes it sound very definite and very clear.