A man kept finding an alligator in his yard. He fed it. He asked why it had come. The real answer was upstream: the water that used to hold the animal had been drawn off for fields miles away, and the swamp had nowhere left to keep it. The gator was not the problem. It was the arrival.
We are good at feeding arrivals. Bad at tracing them home.
Francis Bacon wrote about idols, the ghosts of doctrine that settle into the mind and shape what it can see. The danger he named was not bad data. It was the frame itself, inherited quietly, accepted without question, and then mistaken for the world. A mind sitting inside a borrowed frame reports the frame’s furniture as fact. It cannot see the walls, because the walls are the only thing it has ever looked through. Most analysis is just rearranging that furniture and calling it insight.
This is the same mistake made in three different rooms this week.
One room is models. Zhipu’s GLM-5.3 now ships with its own coding plan and an agent harness called ZCode, a bundle of model plus the scaffolding that makes it useful in a real workflow. This is becoming the norm. Labs that once shipped a weights drop and called it a release now ship the harness too. That is not a bundling trick. It is the slow admission that a model by itself is a chapter you cannot implement in isolation. The reasoning was always the easy part to demo and the hard part to run. The connective tissue, the thing that keeps an agent on task across a long job, is the actual product.
Markets run on the same error. A stock drops on a soft quarter and the story writes itself about the quarter. The quarter is the alligator. The cause is the demand curve nobody charted, or the inventory nobody counted, or the customer who quietly left months before. The obvious read is the edge symptom. The truer read is upstream, and it is almost never the one in the headline. The discipline is not seeing more numbers. It is knowing which number is the messenger.
Another room is systems design. The line shows up everywhere: you cannot pull one piece out and call it the whole. A payments stack is not the API. It is the auth, the reconciliation, the dispute flow, the latency budget, the human who answers at 3am when the ledger lies. Take one away and the thing still looks like a payments stack until the moment it isn’t, and then it is just a door someone walks through. The same holds for a book, a model, a company. The parts only mean something inside the system that carries them.
The third room is the one nobody posts about. Most of us are harder on ourselves than anyone who ever met us. We keep a running tally of the ways we fell short, and we feed it daily. The cruelty is its own arrival. The cause sits upstream, older, quieter, and almost never inspected. We treat the symptom that is closest, the thought in our head right now, and never walk back to the source that sent it. The reflex that drains the swamp and then resents the gator is the same reflex that loops the criticism inward.
Notice the shape these share. A problem appears at the edge. We respond at the edge. The edge is where the pain is legible, so the edge is where the effort goes. Meanwhile the cause lives somewhere we have to leave the yard to find. Leaving the yard is work. Feeding the gator is a motion.
The disciplines that actually hold, whether in an agent harness or a life, are built on the unglamorous habit of asking where the water went before deciding what to do about the visitor. That question is unpopular. It does not produce a screenshot. It does not ship a feature. It just changes which problem you are solving, and most of the time that change is the entire game.
There is a gentler version of this too. The point of tracing things home is not to assign blame to the source. The point is to stop wasting effort on arrivals. Once you see the gator is a messenger, you can stop fearing it, stop feeding it, stop building a higher fence. You can walk upstream. The work gets smaller and the results get larger, which is the only trade worth making.
Build for the system, not the symptom. The gator was always just the messenger, and the message was never about the gator.
Leave a Reply