On the futility of debating the US presidential election

Johanna Drott on 2020-01-19

Credit: Tim Evanson

A control system needs to be at least as complex as the system it controls. If the control system is not complex enough, then the system it nominally controls will inevitably start to act in ways that are incomprehensible to it. This is both an axiom of cybernetics, and a good metaphor for understanding contemporary debates about the US presidential election.

To say that a control system needs to keep pace with the system it controls is rather abstract. The implications of this statement does not immediately leap off the page and embroil the minds of readers with endless thoughts of what is and what should be. We’re gonna have to simplify, just a bit.

We can simplify it by using the example of the humble light switch. It is a control system which has a complexity of exactly two bits: either the light is on, or it is off. If the light is off, we can flick the switch to turn it on. Conversely, if the light is on, we can flick the very same switch to turn it off. These are the two things we can do with a light switch. We have exhausted the complexity of this control system.

If all we need to do is to have a convenient way to turn the light on or off, then this system will do fine. The control system corresponds exactly to the complexity of the system it controls. We have achieved a very local perfection.

If we have a more complex lighting situation, however, this will not cut the mustard. Let’s say we have two light bulbs on either side of the room, and we want to be able to control them individually. The single light switch we have operated with so far will not be able to handle the complexity of the situation. It can only do two things: either on or off. By adding the second light bulb, we have added more potential states the system can be in. There are now three possible states: both lights on, both lights off, or a single light on. Flicking the single switch will not enable us to move between these three states: it has a maximum complexity of two, while the light bulbs have a complexity of three.

(Anecdotal side note: upon moving to a new place, I discovered that the light switch in the kitchen controlled both light sockets in such a way that flipping it turned off the one and turned on the other. This was, shall we say, not optimal.)

The way to solve this is to add more complexity to the control system. In this case, to add a second light switch, so that each light has one. In this way, we can control where the light is on and where it’s off. The same principle can be expanded across the rest of our imaginary house, in such a way that every room gets an appropriate number of light switches. Each addition increases the complexity of the control system, hopefully keeping pace with our desire to turn lights on or off.

We could make this illustrative example even more complex by adding dimmers to the equation, so that each switch now has three modes: off, intermediary and maximum. But I think you get the general gist of what it means to say that a control system needs to be as complex as the system it controls. A single light switch turning every light in the house on or off is not sufficiently complex, and thus needs to be complexified until we arrive at sufficiency.

The debates surrounding the US presidential election find themselves in an analogous situation. The political situation in the US is a highly complex situation, encompassing hundreds of millions of people and a great variety of geographies. Grasping the sheer scope of this complexity is something that requires lifelong dedication to even approximate. Needless to say, in terms of complexity, the US political situation is a big hulking monstrosity outside the scope of any single control system available to mere mortals. It’s just too big.

What the debates do, however, is to reduce this monumental complexity to a choice between a handful of options. Trump or Bernie? Bernie or Warren? Warren or that Yang fellow? Every aspect of US politics - no matter how big, small or complex - is reduced to whether or not they fit onto the personalities or political legacies of these handful of individuals. It is a very narrow lens through which to try to understand the relentlessly complex world we find ourselves in, and the results of these analyses (no matter how meticulously done) will inevitably end up being equally narrow.

It is, to extend the metaphor, like trying to control a battleship by repeatedly flicking the light switch on and off. The debates do not allow for sufficient complexity to grasp the situation, even if you extend your analysis to a year-long series of breathless podcasts or articles analyzing every new electoral or polling event in extensive detail. The world is bigger than these individuals.

This has implications for how we approach the coming months. There will be a lot of punditry about the US election, and most of it will inevitably be of little analytical value once everything is said and done. Most of it will also be inevitable, unfortunately; there is a well-honed tradition of political punditry that will keep going no matter what. The choice that we all face — whether or not we intend or are allowed to vote — is this: do we reduce our analyses of the situation to the narrow lens offered by the debates surrounding the US presidential election, or do we do something more complex? Something, I dare say, more interesting?

[Shameless self-promotion: you too can complexify your life. It’s as easy as flicking a switch. Or a patreon dollar.]