Recently I had to change a headlamp lightbulb in our 14-year-old Volvo, something I felt comfortable doing since I’d done it successfully a couple of times in the past. Except this time, it was the left-side bulb, which hadn’t blown out before, and has a somewhat different setup (cars are not as symmetric under the hood as they are on the outside). Following the instructions in the manual, I disconnected this cable, pulled out that pin, rotated that housing, and so on, and replaced the bulb. Then I put the headlamp assembly back together and turned on the engine.
Success! The headlight came on!
But…oh no… the dreaded check engine lamp came on. Turned out I’d forgotten to reconnect some cable. And once I reconnected it, the check engine light stayed on. Eventually we had to pay the dealership $80 to use the fancy diagnostic tool (known as an OBD-II scanner, for onboard diagnostics) to read and reset the code.
There is some sort of joke here about changing how many PhDs it takes to change a lightbulb. It’s only at times like this that my wife calls me Dr. Rao.
Now the thing about changing a car lightbulb, unlike (for now) changing a regular lightbulb, is that it’s no mere intuitive procedure that you can figure out simply by examining the setup for a few seconds. It’s an unintuitive and arbitrary protocol. There is a temperamental set of precise extra steps around it, being monitored by a not-very-smart error-detection system, which requires specialized equipment and knowledge to fix if triggered. To use David Weinberger’s term (which I learned about from J. P. Rangaswami), modern computerized technological systems do not have much leeway. Lack of leeway makes a system… not fragile exactly, but unforgiving. Leeway is central to a distinction John Grant has been studying, between strong vs weak protocols. Without leeway, errors are costly. And life under the API is very low on leeway.
The OBD system is not an API in the sense of the internet-scale Uber dispatch system, but is an API nevertheless. And when you change a lightbulb on a modern car, you do have to briefly duck under it, and be governed by its unforgiving logic.
Think about error codes, the scanner required to read them, and the knowledge required to interpret and act on them. What sort of knowledge is this? Why does it need to exist? Is it elite knowledge or commoner knowledge? Esoteric or mundane? Why didn’t I, despite my possession of that marker of elite and esoteric technical knowledge, an engineering PhD, which my wife (making a common, but incorrect assumption encouraged by Hollywood) thinks is salient here, have the knowledge to interpret the error or the tools to fix it?
Am I really “elite” in the picture here? Is the “trained service professional” elite simply because he has the scanner? (which, to be fair, I could buy and use myself in this particular case, but not in all such cases). Is the real elite person in the picture the unknown (and possibly dead) person who developed the error code system embodied by the diagnostic system and tooling? Through some more obscure logic, is the real elite person in the picture the President of MIT? Or Joe Biden? Or Donald Trump? Or Elon Musk?
Are there perhaps no elites in the picture at all?
A great deal about our world currently hinges on the question of who the elites are, and where they are located. We bemoan the corruption of our current elites, accuse each other of being them, dream of better elites who might arise through better civilizational logics. When things go wrong in seemingly capricious and arbitrary top-down ways, we wish we didn’t have to have elites at all; and yet when anarchy rears its head, we immediately yearn for elites to emerge to reign in the chaos.
But despite all this anxious discourse around elites and elitism, when you encounter a very specific challenge (in this case a broken lightbulb) it is surprisingly hard to determine who the relevant governing elites are, what the basis for their elitism is, what the status affords them, and how they came to acquire it.
Let’s start with a simple question: Is the knowledge required to change a car lightbulb, complete with the knowledge required to deal with errors, necessary or some variety of bullshit?
The question is based on the dangerous assumption that elitism in the modern world has something to do with the knowledge required to run some part of the world. It is not an obviously bad assumption, despite many obvious problems with it. And it applies as much to changing lightbulbs as it does to making war and peace. So let’s start with that.
Keep reading with a 7-day free trial
Subscribe to Contraptions to keep reading this post and get 7 days of free access to the full post archives.