What Ringing the Bell Used to Mean

A town bell had a small number of uses and a strict grammar. Hours, services, deaths, assemblies, and fire. Everyone within earshot could distinguish the patterns, which is what made the system work.

The alarm was the rarest and most consequential. It required someone to judge, in the moment, that the situation justified interrupting every other person's day. Ringing without cause was a serious matter, because it spent something that could not be replaced quickly, which was the assumption that the bell meant what it said.

That constraint is the useful part of the inheritance. A signal reserved for genuine emergencies stays legible. A signal used continuously becomes noise, and the town stops looking up.

This publication takes its name from a bell that is not in the sequence, the one that rings after the ordinary count is finished. It is the one worth paying attention to.

The grammar is the part worth dwelling on, because it is what made the system work and it is the part that does not survive translation to a feed.

A pattern meant one thing. Not approximately one thing, and not one thing depending on who was ringing. The number of strokes and their spacing carried the message, and everyone within earshot had learned the vocabulary without being taught it formally.

That shared vocabulary is expensive to build and easy to destroy. It survives only if the signals stay rare enough to remain distinguishable, which puts a hard limit on how often the alarm can be used.

The modern equivalent is more regulated than the old one and rests on the same logic. The modern rules for emergency alerting specify who may originate a message, what codes exist, and how receivers must behave, which is a written version of the grammar the bell carried informally.

The codes matter for the same reason the strokes did. A distinct code for a distinct hazard lets a receiver decide what to do, and a system with one undifferentiated alarm forces everybody to treat every message as maximal.

The statute behind them makes the underlying bargain explicit. Access to a channel that can interrupt everyone is a privilege granted on conditions, and the conditions exist because the interruption is the scarce resource.

Scarcity is the whole design. An alert that fires often is not a stronger alert, it is a weaker one, because the population learns the base rate and adjusts. Nothing in the technology prevents this and nothing in the technology can fix it.

The rules are amended continually through the notices that amend them, mostly to add capability: more codes, more languages, more delivery paths. Almost none of the amendments address frequency, because frequency is a judgement rather than a specification.

Which returns the problem to where the bell left it. Someone has to decide that this particular situation warrants waking everybody, and no rule can make that decision for them.

The old arrangement handled it with consequence. Ringing without cause was a serious matter, personally attached to the person who pulled the rope, and that attachment is the mechanism modern systems have the most trouble reproducing.

A committee that issues an unnecessary alert bears no comparable cost, and the incentive under uncertainty runs toward issuing it. Nobody is ever criticised for the warning that turned out to be unnecessary.

The record survives either way. Decisions to alert are kept as records, which means the pattern is auditable after the fact, and auditing it would answer a question that is currently argued from impression.

The name this publication took is a small joke with a serious point behind it. A clock that strikes twelve has said something ordinary. A thirteenth stroke says the count itself is wrong.

It is a useful test to carry into reporting. Most of what arrives announcing an emergency is the twelfth stroke, on time and unremarkable. The thing worth attention is usually the one that should not have been possible, and it rarely arrives loudly.