What is in a name?
Back in graduate school in early 2000s I used to joke with my roommate, telling him that we should set up a political risk consultancy and call it Hindsight 20/20. After all as political scientists that’s what we mostly do. Explain events after the fact. Or at least that’s what many social scientists claim. Neither of us was satisfied with that though, having a keen interest in prediction, forecasting, and anticipation. Now almost two decades later I name my newsletter Hindsight 20/20.
It is not only nostalgia for an ironic inside joke that I decided to name this newsletter Hindsight 20/20 though…
What Matters?
We say hindsight is 20/20 when we fail to anticipate or foresee something that we should have. Now we look back and can explain what went wrong etc. but what is the use? When you look back, the vision is almost always perfect. What you need is foresight, even if it is imperfect, not perfect hindsight.
But when does this regret, this lamentation about our failure to see what was coming matter? It matters when the impact of what we fail to foresee turns out to be big and I mean really big. Life-altering big. Thinking about uncertainty thoroughly and engaging Kahneman’s System 2 before making a decision is cognitively costly and demanding. We simply cannot sustain a constant vigilance against the forces of unknown. So we need to be selective in bringing the big guns out and let habit, System 1, “fast and frugal heuristics”, and routine take care of the rest when failure is not catastrophic. Focus on the few big, important things and the rest will sort itself out (mostly through trial and error).
I will be writing on two themes in this letter; uncertainty and international affairs. What connects them would be a focus on the high-impact events (often low-probability or even “unknown unknowns”, but not always) rather than the routine, the everyday, the business-as-usual. On sorting out the signal from the noise, what really matters from what just seems so. What not to do and what to avoid more than what to do.
It is easier said than done of course. So in a way this is a public ledger for myself as I try to make sense of an increasingly uncertain (or so it seems) world. What was I thinking when I identify a particular event/development as a signal? Perhaps more importantly how I was thinking? How do my models of the world fare when faced with the reality and there is a record of them, staring in my face in the archives of this newsletter?
The hope is that I will say “hindsight is 20/20” less and less and that this will also help you do the same.
On Decision-making under Uncertainty
This is not another newsletter or blog about “mental models” which seems to be all the rage in the past several years (e.g Farnam Street, Poor Charlie’s Almanack etc.). Nor will it try to recycle the methods and frameworks of the “wisdom of the crowds”/“heuristics and biases” approaches that relies on the works of Kahneman, Tversky, Tetlock, and others. Good Judgment, Inc. and its Superforecasters are the gold standard here but there are others such as Metaculus out there. Full disclosure; I was one of the Superforecasters out of the Good Judgment Project and continue to work with Good Judgment as a Superforecaster so I might be a bit biased in here. However, I think GJ’s public record speaks for itself.
There are two aspects that are part of the approaches mentioned above that I will be talking about though.
How to ask better questions? Asking the right question is perhaps underrated but it is fundamental to reducing uncertainty (when we can). It is really a craft to ask a question (or a series of questions) that is both relevant (in making a decision) and answerable. My hunch is that by asking the right questions (and keeping track of the answers) we can tame some of Taleb’s Black Swans, perhaps even turn them gray, thus getting a handle on them. After all, Taleb himself said, most Black Swans are thus from the perspective of a particular observer.
How to reduce noise? Recent research suggests noise reduction plays a large role, perhaps larger than debiasing, in improving forecasting accuracy. Kahneman’s new project/book deals with noise. I also think it is fundamental to reduce noise thus making the signal stronger/clearer in this information-rich environment to improve judgment and subsequently decisonmaking.
In short, the idea is to ride the Taleb - Tetlock curve. Identify the domain of the problem (known unknowns vs unknown unknowns/deep uncertainty) and its pay-off/consequence structure (e.g. binary with no regard to magnitude vs variable where size/magnitude matters) and only after that apply appropriate tools/models for decision-making (e.g. uncertainty reduction and early signal detection for known unknowns vs avoidance of ruin/absorbing barrier, minimax, robust decision-making etc. for unknown unknowns).
Most interesting and consequential things relate to Rumsfeld’s unknown unknowns or deep uncertainty and happen in Taleb’s 4th quadrant but we have better tools (and better chance of success) if we can move into the domain of known unknowns (or Taleb’s 3rd quadrant) for decision-making or at least clip the left-tail if we cannot move away. What I want to do is to explore the ways to do just that.
On International Affairs and Geopolitics
Thinking, teaching, writing, and speaking on international politics is my bread and butter. I focus on rare but high-impact events like crises and wars, major changes in the distribution of power, the creation and the unraveling of international orders etc. Foreign and security policies of major powers such as the US, Russia, and China play an outsized role in all those events. So I focus on them and, since I am interested in decision-making, on their leaders.
Reducing uncertainty, asking the right questions, using the correct models (or no models at all, a subject I will talk in future issues), recognizing the signals etc matters a lot in this domain where the consequences, positive and negative, could be staggering (and even existential).
There will mainly be two types of writing on international affairs:
Short commentaries focusing on whether an event is a meaningful signal or just noise. In other words, should you pay attention to it or ignore it?
Longer analyses on the future of geopolitics that will include pre-mortems. In other words, they will include a section where I assume my analysis/forecast turns out to be completely wrong and I describe why that may be the case.
A note on terminology. Lot of this is called geopolitics in the popular parlance, although strictly speaking most of what passes as geopolitical analysis is neither classical geopolitics (in the tradition of Mahan, Spykman, and Mackinder) nor “critical geopolitics”. However, given that this is not academic writing I will yield to the popular usage and will use geopolitics and international politics (as it relates to foreign and security policies) interchangeably.
This newsletter is a work in progress and I’d love to hear your comments and suggestions. What would you like to read or hear me talk about (remember there will be podcasts)? What you cannot get elsewhere but have been looking for? Let me know in the comments or email me at balkan@devlen.com
Please feel free to forward this newsletter and share it with those you think might be interested in.
Thank you for subscribing and joining me in this journey of trying to make sense of an uncertain world!
Balkan