CARGO CULT SOFTWARE DEVELOPMENT
After the Second World War, anthropologists documented a phenomenon in Melanesia. Island communities that had received military supply drops during the war continued to build runways, control towers, and wooden headphones after the war ended. They performed the rituals they had observed. The planes did not return.
They had copied the form without understanding the mechanism.
The term entered software through Richard Feynman, who used it to describe scientific practices that imitate the appearance of rigor without the substance. The concept applies directly to how software is built today.
THE PATTERN
Cargo cult software development is the adoption of practices, tools, architectures, or processes because they are associated with success — without understanding why they succeeded or whether the conditions for success apply.
A company succeeds while using a particular architecture. The architecture becomes famous. Other companies adopt it. Most fail to achieve the same results. They had the architecture. They did not have the conditions.
The architecture was not the cause of success. It was correlated with success in a specific context. The context did not transfer. The architecture was cargo.
THE MECHANISM
Cargo culting persists because causation is hard and imitation is easy.
Understanding why something works requires deep analysis: what problem was being solved, what constraints existed, what alternatives were considered, what tradeoffs were accepted. This takes time, expertise, and access to information that is often not public.
Copying what successful companies do requires only observation. They use this tool. They organize teams this way. They follow this process. The surface is visible. The reasoning is not.
So the surface gets copied. The runway gets built. The planes do not come.
THE FORMS IT TAKES
Architectural cargo cults adopt structural patterns without the problems those patterns solve. Distributed systems patterns appear in applications that run on a single server. Scaling solutions appear before there is anything to scale. The architecture handles problems the product will never have, while creating complexity the team must always pay.
Process cargo cults adopt methodologies without the dysfunction those methodologies address. Ceremonies are performed. Artifacts are produced. Metrics are tracked. The process runs, but the outcomes do not improve. The ritual is observed. The purpose is forgotten.
Tool cargo cults adopt technologies because they are popular. The tool was built to solve a specific problem at a specific scale. The adopter has a different problem at a different scale. The tool adds cost without delivering value. But it is on the résumé. It is in the job listings. It must be correct.
Vocabulary cargo cults adopt terminology without the concepts. Words that meant something precise in their original context become buzzwords. They signal membership in the profession without communicating meaning. The language is performed. Understanding is optional.
WHY IT CONTINUES
Cargo culting is locally rational even when globally wasteful.
For individuals: using popular tools and patterns is career-safe. If the project fails, you were using industry-standard approaches. If you deviate and fail, you were reckless. Conformity protects.
For organizations: adopting what successful companies do feels like due diligence. We studied the leaders. We followed their example. If it doesn't work, the method was wrong, not our judgment. Imitation protects.
For the industry: cargo cults create consensus. Everyone uses similar tools, similar patterns, similar processes. This makes hiring easier, training cheaper, and projects more predictable — even if the solutions are poorly fitted to the problems.
The cost is diffuse and hard to measure: projects that are more complex than necessary, teams that are slower than they should be, problems that go unsolved because the rituals don't address them. No single decision looks wrong. The aggregate is mediocre.
THE ALTERNATIVE
The alternative is not contrarianism. Rejecting popular approaches because they are popular is its own form of cargo cult — copying the aesthetic of independence without the substance of reasoning.
The alternative is understanding.
Before adopting a practice: what problem does it solve? Do we have that problem? What are the costs? What are the alternatives? What conditions made this work in its original context? Do those conditions hold here?
These questions are not difficult. They are merely unfashionable. The industry rewards quick adoption over careful analysis. Moving fast is visible. Understanding deeply is not.
But the teams that understand deeply build systems that work. The teams that copy quickly build runways in the jungle.
THE TEST
The test for cargo culting is simple: can you explain, without appealing to popularity or authority, why this choice is correct for your specific situation?
If the honest answer involves "because everyone uses it" or "because the conference talk recommended it" or "because we'd have trouble hiring otherwise" — the reasoning has been outsourced. The choice may still be correct. But you will not know why, and you will not know when to choose differently.
The planes come for those who understand aerodynamics. The rituals are for everyone else.
---
Lingenic LLC
2026