A rifle that might trip after just a few magazines, a torpedo that might hit and not detonate, a service pistol that would attract attention in the case of broken slides: all of these have an unpleasing commonality, which is that big armies will not hesitate to take equipment which they had already known was not entirely satisfactory.
The trend is not mystery or negligence per se. It is a procurement fact that manifests itself anytime the speed, magnitude, and institutional inertia interface with engineering constraints. Arguably, weapons tend to be passed to troops before the inherent liabilities are identified, and doctrine, training and field fixes would tend to bridging the gap until the system catches up.
The early M16 in Vietnam symbolized as few cases of the human price of “good enough” as could be. It came as a lightweight, high volume of fire solution to jungle combat, though soon after, reliability issues arose. In the story of the Hill Fights by Sam Lichtman, “there were no cleaning kits issued” with the rifles, and dirty cases and plugged ones came at the most inopune moment. It was not a design sin, it was a systems problem. The rifle was developed based on a single propellant, yet the military switched to a new one without completely testing the new switch behavior in the firearm. Chalky deposits it brought, increased cyclic stress, and a spectacular increase in stoppages until such measures as chrome-lined chambers and the broader modifications made to the M16A1 renewed confidence.
The same message about small reveals that the slightest changes can result in the failure of a chain of events can be seen once again in the case of Beretta M9. The initial failures with slides caused injuries and an emergency inspection regimen, but subsequent analysis has termed a major cause as ammunition being loaded to a greater than anticipated pressure, rather than a simple “bad pistol” story. Beretta established a safety measure in his reply such that a failed slide would not leave the frame instead of falling away, which became standard in later pistols. The bigger lesson here was not that brand reputation is at stake but that interoperability objectives, e.g. NATO compatibility, could commit design decisions and the support equipment and quality control decide whether the weapon would work to scale.
In cases where bullets are not the problem, the defective systems may exist as there is no alternative that comes immediately. The initial ills of the Mark 14 torpedo depth-keeping and defective detonators compromised attacks until deep-sea trials revealed that many hundred yards of field testing were necessary. The similarity to the early M16 lies, not in category but in process, in the fact that it was only when it was put to actual use that both failures became difficult to believe.
The bruising incidences led to the development of institutional structures directly. Independent operational testing capability was established by Congress in the 1980s, and current watchdog work still focuses on early, realistic, iterative testing. An evaluation by the GAO (2026) revealed that the Department of Defense test policy lacked full compatibility with the best practices, involving testers early and conducting iterative testing, including the wider use of digital twins and digital threads.
Those suggestions are an indication of a bitter lesson: introducing weapons into service before their ecosystem is prepared: ammo, maintenance, training, documentation, and quality control build failure modes that cannot be addressed through heroics. Flawed-but-issued systems have a history of urgency being rational, but still having a debt, and that debt being paid in retrofits, constrained tactics, and even lives.
