The Once and Future Web

Complexity can feel daunting, but I think the real problem is how we struggle to talk about. It's possible to think about complexity in more general terms and unlock tools to understand it on a deeper level.

Photo: Frank Zappa, photographed by Norman Seeff in 1976.

Why Do Things Have To Be So Complicated?

Complexity is everywhere — emphasis on everywhere. We see it in all the systems we can observe, from things as big as the large-scale structure of the universe's visible mass, all the way down to things as small as proteins. That ubiquity is not a coincidence.

The people who formally study complex systems describe a phenomenon called emergence. It's an observation that a system has a capacity greater than the sum of its parts would suggest. That might sound a bit cliche, but be assured: the consequences are profound, and we're still trying to understand them. Even straightforward, deterministic systems with only a few rules can really surprise you.

Conway's Game of Life gives us a clear example. It has only four rules and a grid for the automatons to live on, but Conway showed it's Turing complete! So from a computability standpoint, it's as much a Turing machine as the device you're using to read this post.

We can try to limit the number of system components. The three-body problem is one of the simplest systems, with just three masses orbiting each other due to gravitational forces. But even here, we find the emergence of chaotic behavior that requires careful simulation to accurately predict. So the next time you benefit from a high-resolution weather forecast, spare a thought for the people who had to simulate Earth's entire atmosphere.

Now, I've focused on natural systems so far, but human systems definitely feature complexity, as Avril Lavigne laments in her 2002 single, “Complicated.”

Georgia Warr: Why do things have to be so complicated? Sunil Jha: Ah, the eternally wise words of Avril Lavigne.

— Loveless by Alice Oseman

We've established that complexity is ubiquitous, so while most people don't formally study it for a living, everyone interacts with it. Humans have developed highly specialized skills and knowledge to deal with it. And within those communities, we're comfortable talking about specific kinds of complexity. That's where jargon comes from.

But that's a shattered view of the whole.

The problem is we don't have a robust, universal way to talk about complexity. And siloed off from each other, how can we expect to systemically engage with complexity?

I think people have some sense of the scope of failure. That's where the negative connotation of complexity comes from. People describe how it overwhelms them, produces opportunity costs, and unleashes serious harm. And that's if they're even paying attention. It's even worse when people willfully ignore complexity or treat it as a shallow aesthetic.

That's why I mention Lavigne; I think the way we use “It's complicated” is noteworthy. The phrase means something is difficult to describe because of its complexity. And because it's not specific to anything about that something. It's so flexible. It's a rare, universal statement about complexity.

So that's one universal tool, even if it's just a warning. Caution is a good start, and it's certainly better than denial. But I think we can go further and engage with complexity on generalized terms.

The Frame

I've had this intuition about a generalized approach to complexity, but I wasn't sure how to articulate it. Thankfully, I recently came across this incredible quote from Frank Zappa's 1989 autobiography:

The most important thing in art is The Frame. For painting: literally; for other arts: figuratively — because, without this humble appliance, you can't know where The Art stops and The Real World begins. You have to put a “box” around it because otherwise, what is that shit on the wall?

Objective reality — what Zappa calls the Real World — has endless, ubiquitous complexity. We can model only parts of it at a time. That means we must explicitly choose what information to include, and everything else is left out. The complexity of the Real World still exists. But within The Frame, we focus our attention and glean meaningful insights that we can use back in the Real World.

These “Zappa frames” are just what I was looking for, and to make this more clear, let me tell you the story of epidemiologist John Snow.

Snow studied the 1854 London cholera outbreak. Doctors knew that people were contracting cholera, getting sick, and dying. But there wasn't agreement on how cholera spread or how to mitigate future transmission.

It was a complex situation. People are already complicated. But these Londoners lived closely together, so there was a lot of interaction complexity. Snow needed a way to clearly explain the transmission. He needed to piece together a meaningful picture of these Londoners' lives.

Snow's solution was a map.

Look at that. It turns out John Snow did, in fact, know something after all.

This famous map of cholera cases shows geographic clustering around the Broad Street water pump. It was evidence for Snow's hypothesis that people were drinking contaminated water (and the germ theory more broadly).

Snow was trying to analyze a complex system, this section of London and all the people living there. With a carefully constructed Zappa frame, he could highlight the most helpful information and provide tremendous clarity. And it changed how people think about the problem. Snow's contributions and popularization of data visualization are recognized as foundational pieces of modern Western epidemiology.

Agency and Fundamental Frame Boundaries

Zappa frames seem like a compelling way to think about complexity, but what fascinates me most is how they center choice. In Zappa's definition, he's clear that our agency is required to give interactions with complexity meaning. Otherwise, it just looks like noise, “What is that shit on the wall?”

We choose what makes it into the frame, what gets our attention, and what gets ignored. That emphasis on choice means Zappa frames readily expose the ethical dimensions of systemic thinking. And because it's generalized, that ethical focus doesn't get lost in the tangle of details specific to a domain or cultural bias.

For example, there's a bitter coda to Snow's story. Despite its clarity, Snow's framing was challenged by cultural squeamishness around poop. It's disappointing that it limited the effectiveness of his ideas, but it's downright embarrassing that we've only partially learned the lesson. It's still a problem today.

But having clarity is valuable. We've established how complex systems touch the lives of everyone, so we're obligated to make good choices when we build Zappa frames. And by recognizing that Zappa frames have boundaries, we acknowledge that we have an incomplete view and must be open to corrections. That openness makes it easier to fulfill our obligation to always include information that accounts for potential harm. Excluding that information puts people at risk and diminishes the value of the Zappa frame.

There's a lot more to explore here, and I'd like to write about it. For now, I'll leave you with a reminder to make good choices.

Be like Charlie. Make good choices.

Another fundamental aspect of Zappa Frames is the nature of their simplification. Simplified models offer clarity. Snow's map highlighted a geographic signal that would have been hard to spot in a stack of individual patient reports. But if you include too much information in the frame, you risk losing the quality of that contrast.

But I'm talking about something more fundamental. There are concepts called incompleteness and undefinability. They are specific to constructing mathematical models, but I think they can be generalized here. They suggest that if you tried to build an all-encompassing Zappa frame, it would collapse under its own incoherence. You can't fit all of the Real World into The Frame.

That seems really important to me. It implies that Zappa frames are a kind of fundamental structure of complexity. We can't know complexity all at once; we need to access it in measures via Zappa frames.

That leads me to rethink integrative levels. People who study complex systems often note apparent hierarchical organization. Life is a pretty good example: Organisms are made of cells made of proteins made of molecules. What's nice about these levels is how they satisfy incompleteness and undefinability. The levels' models don't need to be coherent with each other. Details from the lower levels can be safely excluded from the Zappa frame.

I've always been fond of integrative levels because of how tidy the organization looks. And, at least in my cultural experience, people often put things into hierarchies. But thinking about Zappa frames makes me question if they're “real.” Integrative levels have been helpful as explanatory tools, but are they a feature of objective reality, or are they an artifact of how we access complexity through Zappa frames?

This new frame understanding suggests they're very arbitrary. That's somewhat disorienting, but we must remember what's vital: When we build models, we should always prioritize avoiding systemic harms and increasing clarity; not satisfying some harmonic aesthetic in a tier list.

Pushing on the Deep Structure

They created us. And they knew enough of beauty to teach it to us. Maybe they could find it themselves. But only if you pick a side, Maeve.

There is ugliness in this world; disarray.

I choose to see the beauty.

— Westworld Season 3 Episode 8: Crisis Theory

Complexity is a fact of reality. It cannot be circumvented, overpowered, or ignored. Anyone who tries risks unleashing severe harm. But that's not destiny. Zappa frames, instead, show how vital our agency is. We decide the nature of the interaction. So the systems we build can and must engage complexity and use that sophistication for the greater good.

That's why Zappa frames give me hope as a tool. They satisfy an intuition I've had for a while. There's a “deep structure” within emergent complexity — something that suggests there's a kind of unity to all human endeavors, even across domains.

It's an intuition that draws me to storytelling that centers agency, framing, and metacognition. Stories like Westworld, Watchmen, Everything Everywhere All at Once, and Heartstopper — superficially very different genres — give you brief glimpses of that deep structure. It's what makes them so compelling to me and what led me to start writing about Heartstopper.

People tend to focus on production value as a proxy for quality during this era of Prestige Television, and there's nothing wrong with a fun spectacle. But storytelling is a metacognition framework, and I consider that to be its primary and most valuable purpose. Stories that draw on the deep structure connect to all human systems; when they teach us about it, we might truly see the world.

And that's something I'd like to write about more, but in the meantime, I'm hopeful. Hopeful we might learn more about the deep structure: how to see it, how to talk about it, how to push on it. Because we sorely need to discard the systems that suck and build new ones that give us what we need to thrive.

With the right frame, we can find the path.