Understanding complexity and interdependencies will help international affairs practitioners avoid overconfidence in applying narrow lessons from the past.
In the last blog, I introduced one of the most important concepts for an international affairs professional to grasp from Nassim Nicholas Taleb’s book Antifragile. If you missed the post on convexity or the follow up on how it applies to an international legal research team, they’re here and here.
Understanding convexity and how it positions practitioners to benefit from uncertainty is a lesson that most international affairs graduate programs and on-the-job training miss. In the last post, I described how to chart a convex path. Remember, you can leverage convexity to be antifragile to uncertainty, now and in the future, by bounding your downside, removing caps to your upside, and ensuring you have an option to exit at any time. This way you don’t need to predict the average stressor because you’ll be prepared for the worst.
Now, I’ll review the second major concept: complexity, the second and third-order effects, and their interdependencies. I started this series with convexity to give you the solution before the problem. Charting a convex path is Taleb’s solution to complexity.
Here’s a Taleb quote that sums up the problem:
“Black Swan effects are necessarily increasing, as a result of complexity, interdependence between parts, globalization, and the beastly thing called ‘efficiency’ that makes people now sail too close to the wind…The world is getting less and less predictable, and we rely more and more on technologies that have errors and interactions that are harder to estimate, let alone predict.”
Every clause of that quote is doing real work. By the end of this blog, I want each concept to land for you, not as abstract philosophy, but as a description of the actual environment you work in every day.
Complexity: more states, more unknown interactions
We throw the word “complex” around in international affairs: complex environment; complex stakeholder landscape; complex geopolitics. There’s plenty of academic international relations theory on complexity, but little on the concrete applications for practitioners.
For Taleb, complexity means a system with many interacting parts whose combined behavior is opaque: not just difficult to predict, but genuinely unknowable in advance from examining the parts individually. Think global supply chains of natural resources or vaccines, or international alliance networks. The whole behaves in ways the parts don’t always predict or explain. There are positive feedback loops at frequencies impossible to notice. What counts as a strong feedback signal to one international partner may be weak to another.
As complexity increases, our models of the situation and the future lag; not because we’re underfunded, but because the number of new combinations a complex system produces vastly outpaces our ability to observe and draw accurate conclusions. In a complex system, genuinely novel combinations keep appearing and history becomes a progressively less reliable guide. Not useless—but we need to understand the limitations of not knowing the whole picture.
Why the interactions and interdependencies are opaque
Taleb identifies three specific types of interdependence in complex systems that together produce opacity:
- Temporal interdependence: A variable depends on its own past changes. In international affairs, we’re most familiar with this type of path dependency and how if we miss a key change, it impacts our understanding.
- Horizontal interdependence: Variables are linked to each other simultaneously. A currency depreciation affects food imports affects political stability affects alliance reliability. Pull one thread and several others move. We might think we can assume which systems are linked, but that is based on past experience.
- Diagonal interdependence: Variable A depends on the past history of variable B. You weren’t paying attention to B weeks ago; now it’s on your radar. Today’s crisis was partially shaped by something that happened elsewhere, in the past, in a system outside your analytical frame.
Put these together and you get a complex system where the causes of what you’re observing are genuinely opaque. You don’t see everything that’s happening, and what you do see is only partially the result of what you think.
When interdependence becomes fragile
Complexity and interdependence are not inherently catastrophic. What makes them dangerous is tight coupling and over-optimization. Tight coupling is when the links between parts of a system allow failures to propagate rapidly, with little buffering or slack to absorb the shock. There is little buffering or slack because over recent decades, the institutions and systems we rely on have been progressively optimized for performance under “expected” conditions. On standard efficiency metrics, this looks like progress. The problem is that complexity makes future conditions less likely to match our model.
Think of the difference between a loosely organized coalition where partners can absorb setbacks independently versus a tightly integrated multilateral operation where one partner’s political crisis rapidly degrades everyone’s capacity. The latter is tightly coupled. When something breaks, it breaks fast and wide.
When Taleb refers to the “beastly thing called efficiency,” he’s referring to things like minimal redundancy, minimal slack, maximum leverage, and just-in-time everything. Taleb’s line about reliance on technologies “with errors and interactions that are harder to estimate” is about optimizing with reliance on technology, especially digital technologies, that are engineered with many layers of abstraction. Technology is fast and efficient because it makes assumptions based on patterns from the past. However, each layer then has incomplete data because the future is not like the past, which creates potential failure modes that are opaque to people operating several layers above.
As complexity grows, the opacity of interactions even within the technologies we rely on creates fragility. Taleb’s metaphor: “Sailing too close to the wind” is a reference to a high-performance maneuver in sailing. It’s fast, efficient, and technically impressive. It also means you have almost no margin for error. A sudden gust doesn’t give you time to correct. The very thing that makes the posture high-performance makes it fragile.
How a practitioner can leverage interdependencies and second and third order effects
This should all sound familiar. International affairs is a tightly coupled, opaque system where practitioners can most easily anticipate temporal first order effects of an actor’s decisions. We don’t often plan for the second and third order effects, nor can we predict the horizontal or diagonal interdependencies. With hindsight and history, practitioners know that we should, but most graduate programs and on-the-job training don’t teach you how to chart a convex path. Instead, we’re taught to try to get better data and improve prediction.
How can an international affairs practitioner who wants to support decision making for domestic and international prosperity make a change? The first step is to learn research strategies that limit your role in the problem of over-relying on techniques and technologies that ignore complexity. Second, to remove barriers to your upside and ensure you can gracefully exit your strategy at any time, you need to be ready for two types of leadership questions: the ones they ask, that are often news inspired hair-on-fire urgencies; and the ones they aren’t yet asking, which requires a practitioner to ready for the long-term with dedicated time to deeper research. I teach how to do the latter with a two-hour research habit.
Shutdown
Taleb says that Black Swan effects are necessarily increasing. Not because rare events are inherently more common, but because the structure of our systems amplifies their impact rather than containing it. In other words, even if the base rate of rare events were the same, their systemic impact grows because of the way we’ve structured our institutions and technologies.
An international affairs practitioner operates in a complex and opaque environment. According to Taleb, the question isn’t how to predict severe socks better. Instead, the question is how to be positioned so that when unforeseen events occur, you’re on the right side of the convexity curve, i.e. you’ve been doing deep research your colleagues said was a waste of time, until it isn’t. Opacity isn’t a technical limitation to engineer around. It’s a structural property of complex, nonlinear, tightly coupled systems, also called the modern world. More data doesn’t eliminate fat tails. Faster models can’t capture reinforcing feedback loops nor prevent them from producing cascades no prior data predicted.
In the last blog, I introduced convexity as Taleb’s recommendation for how to perform with integrity in an uncertain world. Not more data. Not faster models. Not more optimization. Now you know the problem, go back and read about the changes required to chart a convex path. Even Daniel Kahneman didn’t think the average international affairs practitioner was up for the challenge. But I believe in you.
