Learning from History
"If history repeats itself, and the unexpected always happens, how incapable must Man be of learning from experience." Was George Bernard Shaw right about the lessons of the past?
Special Guest: Dr Kristian Gustafson https://www.brunel.ac.uk/people/kristian-gustafson
For more information on Aleph Insights visit our website https://alephinsights.com or to get in touch about our podcast email podcast@alephinsights.com
This podcast uses the following third-party services for analysis:
Podtrac - https://analytics.podtrac.com/privacy-policy-gdrp
Transcript
Hello and welcome to the Cognitive Engineering Podcast produced by me, Fraser McGruer for Aleph Insights. In this series of podcasts we take a look at interesting topics and discuss what we think they tell us about analysis and decision making. I'm here with Nick Hare and Peter Coghill of Aleph Insights and also this week with our special guest Kristian Gustafson and this week we're discussing learning from history. Nick could you introduce us to our, well I was going to say our new friend, my new friend Christian. Tell us about Christian. Yeah I've known Christian
Speaker B:for many years. I guess we probably met about 10 years ago when I started running Futures and Analytical Methods in Defence Intelligence and Christian is of interest for all sorts of reasons. Firstly because he's an excellent all-rounder. He knows a lot about all kinds of different things. But secondly because he's an academic who happens to work in an intelligence field. So it was a useful person to talk to on the academic side and we did a few sort of workshops and talks and things together. But then he also happens to be a practicing intelligence analyst himself and I'm sure he will tell you a bit more about that and has worked in various bits of the government sort of as you know sometimes interleaving being an intelligence analyst and being an academic. So yeah and we've stayed in touch all that time and whenever we get together something interesting happens. And one of the things that he knows a hell of a lot about that I know virtually nothing about is military history and he's always got some interesting examples of things that are like a situation today that happened in you know the time of Alexander the Great. And I think one of the things that I'm interested in is you know actually how useful is that information for policymakers should they be paying more or less attention to him. So
Speaker A:that's why we got him on the podcast. Nice introduction. I was wondering where the history bit was but you've explained. But Christian perhaps you can tell us fill in the gaps a bit tell us a bit more about yourself, your background, where you're from, what you do etc. Yeah not the
Speaker C: IA operations in Chile in the: Speaker A:period of history the 60s and 70s or what they could have learned? Well what we can learn now
Speaker C:about intelligence operations and intelligence analysis based on that period. So what you know can you take a specific case and learn general lessons from it? Can you generate enduring lessons from very specific cases in history? And that's really the fundamental question that underlies any after-action reporting, any lessons learned process and history in general is what can I generate general information from specific cases? Got it. And we'll come on to that in a second. So what do you do at the moment? What's your main thing right now? The day job is Senior Lecturer at Brunel University and then I have a sideline in uniform for the British government.
Speaker A:Okay okay. I mean you guys are probably burning to ask some questions but let's just go straight to the question that you've asked yourself there is can we learn something general from something
Speaker C:specific? Can we? Yes but with difficulty and not that well I think is the overarching take on it. There are you know there are a whole, well there's a couple different branches you can go down here. One is the problem of constructing history. So there's the problems that exist in what you generate and call history. What that's derived from the problems of determining causation, the problems of perspective and the perspective of the history writer, the problems of evidence. There's also then problems in the usage of history. How public figures especially government officers in this instance is what I would focus on make use of history in decision making. And then you can also go down the branch of how you can you know can institutions and governments actually learn from history. Can which is a more fundamental question can government institutions actually learn using air quotes. So defaulting back to that first one which is the problems of constructing there is the the problem which is a general one which guys like Thucydides and Herodotus even acknowledged which is that history is has to be by necessity an abstraction a simplification. The analogy I use a lot with my students is history has to be at scale because a one-to-one scale map of the world is useless. It doesn't help you negotiate the world. History ultimately is presented as narrative. We pick up a book, we read it from the front to the back. That gives an illusion of cause-effect cause-effect of linearity which doesn't exist in real life. Real life moving forward is chaotic. We happen to go left instead of right. We happen to go up instead of down. But when we write history we turn that into a single linear pathway and that eclipses a lot of complexity, nuance, happenstance, chance, and seems to give them an interconnected meaning which they may not possess in the first instance.
Speaker A:Which then sort of leads us to a false conclusion that we could learn from something then right?
Speaker C:Yeah we identify a cause and we say that caused this but we don't necessarily know that that caused this. We may view that particular event as a number of one of a number of causes which led to an outcome. So for instance the great the great historical question is what caused the first world war? Well it's very difficult to begin to say what caused the first world war. There's an almost infinite number of things you could start bringing up. You could talk about industrialization, you could talk about you could talk about increase in population, you could talk
Speaker B:about the scramble for Africa. And ultimately you might you might just want to say what caused the first world war was just you know the day before the first world war and everything that was going on that day. And that probably is that might be the most complete answer but it's but it's not useful like you were saying like what we want is an abstraction. Yeah there's a nice example of this
Speaker C:which hopefully I remember correctly which is General Lewis Mackenzie who's the Canadian commander in Sarajevo during the Bosnian war, the siege of Sarajevo, and a French politician I think it might have been Chirac flew in and there was some sniper fire and they jumped under an armored vehicle and Chirac turned to Lewis Mackenzie and said who started this? Who started the shooting? But Mackenzie's answer I think is the best which he says some asshole 500 years ago. So proximate versus ultimate causes of events. Yeah a foreign leader getting killed in Sarajevo wouldn't have been good either for all sorts of reasons. So yeah so that's that's a you know a good point to say you know in terms of satisfying our desire to understand what unleashed a particular cause of events we might say yeah it was that assassination but that assassination also has precursor events which also have significance. So that is a decision of the historian. What do I determine is sufficiently far back to ascribe as a cause? Or do I need to go back further? How far how many antecedents do I do I give? The other problem in that of course is is selection of evidence. So historians tend although that has changed over the last few years tend to default to documentary history especially when writing about government events. You have a problem because while bureaucracies are ultimately record-keeping machines that's a key function of bureaucracy not everything is kept. So there's imagine a series of concentric circles getting you know from big to small that there's everything that's possible to know about an event. There's everything that's remembered wrong everything that's observed of all that everything that's remembered of all that everything that is recorded of all that everything that is archived of all that everything then the historian has access to later everything the historian actually accesses and then everything the historian actually uses. And that last bit everything the historian actually uses is a very small proportion of what's known.
Speaker B:And the filtering process is actually correlated with what ends up in that filter, you know, so it's not just, you don't just get a random sample of things. There's a specific process which means that you're going to end up with only a certain kind of information which is probably going to be misleading in summary.
Speaker C: be they wanted to hide in the: Speaker D:So in a world where we're recording more and more, arguably we record nearly everything. What are, what do you see are the future problems for historians? I mean, I can see that there's potentially way too much data to process and to trawl through for people like you in the future. And the problem historians will face is not that there's no data, but there's just way too much.
Speaker C:Yeah, that one-to-one scale map of the world when you have all available information, what do you use and how do you select it? Which would shift the bias from whereas before it might have rested with archivists and it might have rested with secretaries and that which was recorded and that which wasn't. If everything is recorded then the bias comes into how the analyst, how the historian selects evidence, how he selects it, what he...
Speaker B:Yeah, I think we actually did a podcast about digital history and what was being recorded and I think that one thing I remember from doing the numbers was that a much higher percentage of data than ever before is being lost, but that was being more than offset by the enormous amount that we're creating. So there'll still be a lot more data, but it's certainly not like we're capturing everything. If you look at what's on the internet archive, it's a tiny percentage of what's actually being created. But anyway, I think I'm kind of interested in just getting back to the issue of... Let's just assume for a minute that we do have an idea of what happened, of what the facts were, even though we know there's lots of problems about actually reconstructing what events actually occurred. But let's assume that we've got that solved. We have a good understanding of a certain period, like the French Revolution and all the events that led up to it and so on. And it's a good enough understanding to be able to pull together some lessons from it and to understand what chains of events were related to other chains of events and so on. So the question is, under what circumstance is knowing all that going to be useful now, right? So if we know all that, what's it going to tell us and how do we exploit that knowledge so we can make better decisions now?
Speaker C:Are you specifically asking about old history or history in general?
Speaker B:Yeah. When does old start? Surely it's a continuum. And how do we exploit that knowledge so that we can make better decisions now?
Speaker C:Well, that's a good question. There is already a bias amongst decision makers for lived or experienced history. So decision makers will default to the things that they remember. And there are a few sort of biases in that or problems in that usage, which I'd like to get to later. But specifically, people tend to look at lived experience, which is problematic because they presume that lived experience gives you an edge or advantage over read or recorded history. That's not necessarily true. So you have famous historians or historiographers like Gaddis who talk about the landscape of history. So the historian has a particular advantage over the observer of actual events. So the observer of actual events is limited to his own sensory input. So what he hears, sees and recalls. And there are already filters in that. The historian has the advantage of sitting almost godlike over time, can move back and forth, can change perspective, can look at different points of view and therefore can construct interesting things. And all of those things help that what Machiavelli described as the kind of vicarious learning, the thing that you might not be able to, you know, this is the whole point of The Prince is here's some mistakes people made before. Pay attention to those. So he's done history. Here are the lessons. Here's the lessons. Yeah, here are some lessons. And, you know, guys like Machiavelli, you know, he states that if you're not going to get it quite right, you can't generate rules that work in every situation, but you can, he called it, get the odor of it, which is interesting. So the example, I think we've discussed this before in other cases, it's if you're looking at, for instance, what Russia is going to do next or what is Russia thinking about doing, we in the West have a very particular perspective, which is very new. We, we, we think we are on, this is one of the faults. I, we think we're the new men. We've trans, we've, we've gone past, uh, past things. We're in this new space where we're creating our own story. Well, the Russians have a, uh, a view of life, which is much more rooted in their deep history. They have a, and whether that is constructed in a modern era or whether that's real, it's a case that they pay more attention to their, to their older history. So when we were talking about, for instance, um, uh, when people were wondering what's Russia going to do vis-a-vis the color revolution in Ukraine, the Maidan, there were people saying, well, nothing, because if you look at Russia today, it's ruled by the oligarchs. The oligarchs are all important. They make money and the financial consequences of, of them making a move in the Ukraine are, are catastrophic. Another view would say, well, actually Russia cares an awful lot about its deep history. Russia as a modern state came out of the Kievan state a thousand years ago. It was Christianized in the Crimea. There is a history going back from that thousand year period to today where Russia values the Crimea. That's important insofar as it constructs a strategic meaning for a place for a state like Russia. So knowing the ancient history, well, it's not going to tell you how certain things are going to happen. It's going to tell you that this is something that the Russians care about.
Speaker B:You may have put something down which is, uh, I think it's, it's almost like what you're saying is that in order to have a model of Russia now, we need to know what historical beliefs the Russians have now. And it's a contingent fact that let's say the Russians have a certain understanding of history, which affects their decisions. Now that's, that's true of Russia now. It may not be the case that I personally need, I need to go and understand the history of the Kievan Rus or whatever to help me understand what the Russians are going to do. I just need to find out what the Russians believe were the case. And I think that's a bit different to me saying, you know, well, we're in a Brexit type situation. We've got to make some rules about Brexit. We've got to think what's going to happen. Well, it's a bit like, you know, the American war of independence, or it's a bit like when the Irish free state came about or, you know, looking for some other historical example that gives us some guidance as to what the good and bad things would be to do. And that's what I've got in mind, I think is sort of not, not so much understanding what do others believe about history, but what historical examples are actually useful to us now. And how do you know if you pick the right one?
Speaker C:Right. So, I mean, first of all, I'd say that you're an analyst. You've been an analyst. You understand that there's a difference between generating a causal understanding of a model or an event and giving someone else a causal explanation of it. How you come to an understanding is not necessarily how you deliver that understanding to somebody else. So you may not, I may not want to walk into a room to brief a general and start with my first event on the list being the Christianization of the Rus in 858. Um, but that's something that I have brought into the, to my model of, of understanding. Now, where that becomes useful is that the habit of policymakers, especially, and those who have fairly non-professional understandings of history tend to work by generating analogy. Yeah. And analogy is a simplification of a simplification. That's generally what we, we construct. So, uh, uh, the, uh, something that's often used as an example in this is the analogies used during the Cuban missile crisis by JFK's government. And we talk about that model because we have all the recordings of, of XCOM. So we, we heard everything they said in their decision-making. And one of the things that was said very early on is, uh, this is like Pearl Harbor.
Speaker B: at, should we invade Iraq in: Speaker C:Analogy is the critical tool that, that most people use because that's how, how we think. It's rather more determining what the right analogy is and getting the best fit analogy. And then not using that beyond its utility. The, um, Neustadt and May who are two American guys at Kennedy School of Government talked about always working out. Uh, the first question is what's the story until now? So the policymaker needs to, uh, not just cherry pick, uh, analogies, which seem to suit, which is, is what people will tend to do, but to situate yourself within a story. Um, so you're looking at what's the story up until now, where do I sit and what will this look like as we go forward? So what's my part in this unfolding, uh, in this unfolding story, not viewing yourself as the ultimate result of history, which is a bit, it was just an egoist fallacy, which, which creeps in a lot, but as part of, so what's the story until now. And therefore I can use that if I construct that story, I can better find, uh, and, and stretch it into the future rather than, than just into the past. I can help pick the right, um, models for particular bits of understanding. I can help generate, um, a, um, framework in which I can hang, uh, the right analogies in the right order to help get through different parts of a problem. But to simplify and say, this problem is just like that problem, uh, is, is not, or is rarely a safe bet in history. And I think it's, you know, the, the paraphrase of the Santayana quote, now history doesn't repeat itself, but it sure does rhyme.
Speaker A:Um, so I want to ask is, um, especially given that we talk about analysis and decision making, that's what we're here for. We've talked a lot about analysis at the moment. Um, there's something you talked about at the beginning as well as about institutions is I was just wondering, I'm sure there are mechanisms for how this kind of analysis is brought into decision making. Um, but I was just wondering, um, how, what trends there might be in that or how that might be changing at the moment of, of, of, yeah, how, how you bring in historical intelligence analysis into policymaking to decision making now. Um, is that, is there a trend? Is that changing? How do governments, people, you, how do governments, uh, use people like you?
Speaker C:Now that's a good, uh, question. There's, there's two sort of branches you could go down there. One is, is, um, after action reporting or lessons learning. So for instance, something like the Butler inquiry or the Chilcot inquiry, where you look at a disaster and say, right, what can we learn from that disaster? What can we change? How can we improve? And then the other side is, is how you use history in, in current decision making. I would tend to argue that the role of history in current decision making would tend to be in the analytical space rather than in the, necessarily in the policy determining space. Um, that the, the people doing the analytical groundwork are the ones who, who need to know the history and need to draw on the models. And that should then be presented depending on the character, because how you present information to policymakers depends ultimately on the character and the background and the intelligence of the, of the decision maker. But in, in terms of policy advice, rather than in, in raw history, I think giving history to some raw history to some people when they don't have time to read it or they won't have time to digest it, but it would be like giving unprocessed, uh, collection material.
Speaker B: ade and it takes into account: Speaker A:Um, before we continue, we're close to be needing to wrap up there. I've got one question I want to ask Christian, but I just wanted to ask to see if you've got anything you want to ask or if indeed Christian, if there's anything you want to add at the moment, uh, we'll go with Peter first.
Speaker D:So in, in very general terms, thinking about the progress human race has made, and I mean progress in a Steven Pinker sense that we have much less violence than we've ever had.
Speaker C:Yeah, I would, I dispute Pinker, but go on.
Speaker D:We're, we're better off now than we were a hundred years ago. And a hundred people, a hundred years ago were better off than the people 200 years ago. It, you know, everything now is rosy compared to how it's been, uh, how much of this general progress would you ascribe to be us being thinking, planning beings who are able to learn from what we've got wrong in the past, uh, and, uh, corrected our behavior as a result. And how much would you, uh, ascribe to the just general, general technological process that we've made, which do you think is the stronger driver for change?
Speaker C:Yeah. I mean, we human societies learn in, in, after a fashion. So as, as, as Nick said, we try to bake in to the design of our societies and our cultures, memory, uh, and, and, and knowledge, but for everything we fix, we often break another thing. So it's, and here we have a problem of historical perspective. You're viewing the world as, as, as a pleasant place to be, um, because you live in a relatively very pleasant part of it. Um, if you were in the middle of fact, in many places in the Middle East or Sub-Saharan Africa right now, you, you may not have such a rosy view, um, of the world. Um, the other problem with that view of, of tending to say that, that we are inevitably progressing as, as a species and, and viewing that progression, that upward progression as inevitable is that it excludes, um, the, the, the very frequent, um, you know, if you look at history, civilizations are like a tetherball and they, they keep going up as long as someone keeps hitting them up. As soon as you stop hitting them up, they tend to spin down. So we have over the last 5,000 years of recorded history, lots of cases of very well-developed civilizations. Yeah.
Speaker B:This time we've got it right. Yeah. I'm sure they all say we've learned from those disasters.
Speaker C:And that's exactly what, you know, so you talk about the economics, this is when we have, um, uh, the, the prime minister saying we've banished boom and bust. We've, we, and this is the, the, the big problem that I, I mentioned earlier, I call new man theorem. So yes, there's all this history that came before, but we've surpassed that now. We are the new men. None of the lessons of the past. This time is different. This time is different. And so every time I hear someone say, oh, but in our ever increasingly complex world, I reach for my revolver because that is code for history doesn't matter now. Right. We've, we've gone past that. And that tends to, especially we tend to, especially as, as our society, perhaps focus on technological advancement and technology as overweighing other aspects of, of our society. So we, we say, well, there's the internet and there's this connectivity and there's big data. So all these past lessons don't matter because those existed. Those are derived from times where we didn't have all those things. Therefore that past history doesn't matter. So in military history, we have all sorts of people saying, no war is no longer Clausewitzian. It's no longer determined by chance, violence and, uh, and reason. It's something different now because insert technological. Yeah. Um, and, and that, and that really gets me worried because that's, that sets you up for that fall off that cliff.
Speaker B:I just want to add that from my, this is my experience in the MOD is that, um, well, it's similar to what Chris was talking about. Everyone, everyone thinks that people are fighting the last war and I don't think that's quite right. Um, I think the problem is that everyone sort of almost overcompensates for the tendency to fight the last war. Everyone thinks they're taking into account the historical important lessons. That's what everyone in the MOD thinks. Okay. Well, we know that war's changed now. We're not going to fight, you know, we're not going to fight as if it's the invasion of Iraq. Now it's going to be more like Afghanistan and it's all war among the people now. And then of course the next one is going to be different again. And if you look at the data on the nature of conflicts, it is more or less random. There's, there's very little trend in terms of the, you know, things like the asymmetry between the forces, the number of casualties, the length of the particular conflicts and all of these things, the things that the UK have got involved in. There actually isn't much of a long-term trend and yet people can't help looking for one. And you know, if you're a retired general, you've got to write a book about that trend and then that gets on the reading list at Sandhurst and then everyone's saying, well, this is what it's all about now. And none of that is people fighting the last war. It's people trying to get better, but you know, I don't feel like we'll never get it right.
Speaker C:Yeah. Christian? There's, there's problems in there too, is it's easier to learn from defeats and failures than it is from successes. The example I like to trot out from this is the end of the first world war. So at the end of the first world war afterward and everyone sort of licking their wounds, the Germans looked at the last hundred days of the first world war and said, we were beaten because of tanks and combined arms. And that's what we need to do. So the next war is about tanks and combined arms. And that's the war we want to fight. A war of movement, a war of what we ended up calling blitzkrieg, but to the Germans never really used that word. And if you look at the way the West learned from the first world war, we looked at it and we said the Somme. Nailed it. We said, you know, the Somme, that was a disaster. We never want to do that again. And so they started focusing on defensive structures. They started focusing on, and that's what led to the mentality, the Maginot line mentality. Yeah.
Speaker B:We were getting better at what the Germans were doing well, and they were getting better at what we do well.
Speaker C:Precisely. And so, and, and, but that feeds into how, you know, this is the lessons learned thing is first of all, can institutions like governments learn? Well, they're not, they're not single things. They're not a single living being. They're collections of human beings with each of their own. But also we tend to find in government when we try and learn is governments or bureaucracies. Bureaucracies tend to find things that they can fix bureaucratically. So we, we may identify a lesson, presume that we've got the perspective right. And we, we have learned the right lesson. What do you actually change in government? Well, governments, you know, it's the hammer nail analogy. We always look to improve. We're the hammer. We, we want to look to improve the nail and, and, and the face of the hammer, but there may be all sorts of other things that we're not dealing with.
Speaker D:Well, Nick and I have actually got direct experience of this. And there's a bit of a running joke between us that when, whenever government's got a problem or they want to fix something, what they do is stand up a new team because it's more or less the only thing that they can do. It's the only lever they've got to pull in is their own internal structure.
Speaker C:And so when we, when we form history, we often do it with that purpose of finding the thing that the institution does and doing that more. So that's the perspective problem and, and, and, and, and, you know, how we actually do anything. So can governments learn and, and use that in decision-making? Yeah, but that's really hard to do right.
Speaker B:I think the answer is there probably isn't a formula and, you know, we'll never fix this problem. Fraser, you're trying to wind up.
Speaker A:Well, I'm not trying to wind up because actually, because that's almost, that's, it was almost a nice summing up rounding off. However, there's something I want to ask quickly. Okay. Now I know we're talking about learning from history. And, and so I've got a question for you. It doesn't have to be about learning, but as a military, military historian you must have vast knowledge of all sorts of military history. What's your favorite military historical event or favorite battle? It doesn't have to be one that's where we can learn something or not, but you must have a favorite there. Yeah, if, if, if, if that's not bad taste talking about battles and having a favorite one, but anyway.
Speaker C: I say, because here you have: Speaker A:That was nice. That was very educational. Actually, I liked that. That was great. Well, look chaps, we're going to stop there. I really, really enjoyed that. So we'll wind up there. Um, so thank you as always to listen to the Cognitive Engineering Podcast. I'm Fraser McGruer. We've been here with Peter Coghill and Nick Hare of Aleph Insights, but most of all, um, a special thank you to our guests this week, to Kristian Gustafson. Thank you so much for being here. Thank you. Really enjoyed it. Hope you enjoyed it. Um, thanks one and all. Goodbye.
