As someone who has frequently felt like a failure (because there are always plenty of things I’m failing at), there’s something that I’ve known for a few years now:
No system can truly be trusted if it has not yet failed. And the more times it has failed without dissolving, the more it can be trusted.
Here’s a typical American example: You decide that, in order to keep your mental health in order, you’re going to go to the gym for one hour, five days a week, every week. Six weeks in, you have succeeded perfectly. One hour a day, five days a week, every week.
Six weeks is a long time. There is cause for optimism. But this is an unstable system. Because we have no idea what failure mode looks like.
Something goes wrong. You get sick the following week. You lose a little momentum, say you’re just going to take this week off but next week comes and your engine is stalling out. You can’t wake up early enough. A big project at work takes over your consciousness. Four weeks later you’re thinking “I feel terrible. Remember when I was taking care of my mental health by going to the gym?”
You can extrapolate this out for any length of time and it’s still true. You’ve exercised five days a week for 146 weeks, no exceptions. Your system is probably pretty good, but you can’t be certain, because you haven’t seen what happens yet when it fails. You’d much rather be in a position where you’d exercised 5 days a week for 118 of the previous 146 weeks and had two really unsuccessful months and a whole bunch of weeks that got blown up by a variety of excellent failure modes.
In some systems, failure means death or worse, so it isn’t an option. My friend Dustin runs a nuclear reactor and he follows what we all hope is a good system when running that reactor. He has never had the reactor melt down on him. I’m a drug addict who hasn’t taken a drug or a drink of alcohol in 18 years. When she crosses the street, I make sure Louisa is always holding my hand and we both look to see if there are any cars or trucks or buses coming before we step out. Neither she nor I have yet been crushed by a motor vehicle. These types of systems need to be as airtight as possible, but they can never fully be trusted, which is actually a good feature because perpetual fear and caution are desirable in cases of life and death.
But for the vast majority of systems, not only is failure not a matter of life and death, failure is critically important to verify the strength of the system.
To be clear, I’m not talking about “learning from it” or “there there little one, we all fail and that’s okay” (true as that is). I’m talking about failure mode as the most important test of any system because it is inevitable.
So the post-failure message of “Go easy on yourself” misses the mark completely. Failure is in fact the only reassurance we can truly have that our system works. “Two steps forward, one step back” isn’t some compromised reality we all just have to live with… In fact, it is the surest path to victory.
How many systems do we set up for ourselves that don’t account for (or softly welcome) failure? For me, it’s too many. I need to stop that shit.
Much love, y’all,
P. S. While I haven’t yet failed to send this newsletter weekly, I’ve definitely sent out some lame ones that I threw together last minute or were just plain not good (okay, under the circumstances, most of them have been that). Failing to send out an email that I’m super proud of (but sending it anyway) has been perhaps the most critically important failure mode I’ve introduced into any creative system I’ve ever had.
P. P. S. Not sure exactly when, but I’m probably going to start making some blog posts on https://zachphillips.blog to accompany this and maybe even some YouTube videos. I mean, I’ve been set up to do it for a while now. I need to come up with a system for that stuff similar to this newsletter… In other words, get ready for some lame videos.
P. P. P. S. Here are a couple of pictures of life right now.