When a Promise Has No Author
What to do with an arrangement everybody honours and nobody agreed, and how to convert it into something the rota can rely on.
Five arrangements on one team, traced back to what agreed them
Two of the five have documents behind them. Two are honoured by everybody and sourced by nobody, and one survived the pattern it belonged to. This is one employer's own trace, not a statement of what any rule requires about varying arrangements.
Some arrangements have no author at all. Not a manager who left, not a document nobody can find — nothing. Everybody honours it, everybody believes somebody agreed it, and the belief is all there is.
The constraint in “When a Promise Has No Author” becomes useful only when its source, cost and effect on coverage are visible. A team considering explore the provider’s current overview for employee time clock records can add timely work and attendance context, with transparent settings, limited access and a clear correction route.
These are different from rules nobody can source, because somebody is relying on them. That makes deleting one a decision about a person rather than a tidy-up.
Teams documenting “When a Promise Has No Author” may use PubMed shift-scheduling research as an independent reference. Recheck current requirements before changing a live rota.
How they form
Usually by accumulation. A pattern accommodates something once, then twice, then as a matter of course. Nobody granted anything, and after three years a person has reasonably concluded that something was granted.
The second route is inheritance: an arrangement attached to a role, honoured for successive holders of that role, long after whatever prompted it.
Why they are worth taking seriously
Because the reliance is real. Somebody has arranged childcare, a second job, a course or a commute around this, and the fact that no manager signed anything does not change what withdrawing it would do.
Whether any status has attached to it is a question for somebody qualified in the place concerned, and it is usually not the question that determines what to do.
The wrong first move
Announcing that the arrangement never existed. It is often technically accurate, it is always heard as a trick, and it converts a scheduling question into a trust problem that outlasts the pattern by years.
It also tends to be wrong in a specific way: the arrangement usually did begin with something, and the something is findable if anybody asks the longest-serving person in the department.
The right first move
Ask, neutrally, what the person understands the arrangement to be and where it came from. Most people answer precisely, and a reasonable number volunteer that they are not sure it was ever formally agreed.
That conversation converts an unspoken expectation into a stated one, which is the only state from which anything can be decided.
Three honest endings
Confirm it, with a record and a review date, because it works and costs little. Vary it, by agreement, into something that works better for both. Or end it, with notice proportionate to how long it has been relied on.
- Confirm it, with a record and a review date, because it works and costs little.
- Vary it by agreement into something that works better for both sides.
- End it, with notice proportionate to how long it has been relied on.
All three are defensible. What is not defensible is the fourth ending, where nothing is said and the arrangement is quietly breached one week when the rota is short.
The group version
Where a whole group honours an unsourced arrangement — a shift nobody is asked to cover, a category of work one team never does — the same method applies and the conversation is harder, because there is no individual to agree with.
Here the useful move is to state what the organisation now intends, explain the reason, and give notice long enough that people can plan. Consultation requirements differ by jurisdiction and are a question for somebody qualified where you are.
Why they should not simply be formalised
A reflex to write everything down converts a flexible local practice into a rule that is harder to vary than the practice was. Some unsourced arrangements are better left as judgement with a note.
The test is the same one that applies to any constraint: would somebody be harmed if this were silently dropped? If yes, record it. If no, note it as practice and leave it alone.
What the register should hold
The arrangement, who relies on it, the best account of its origin including "unknown", its current status, and the date of the last conversation about it.
Status is the field that matters: confirmed, under review, or unexamined. Most of these sit in the third category for years, and moving one to the first takes about ten minutes.
The pattern this reveals
Teams with many unsourced arrangements almost always had a long period with no stable management. The arrangements are not a discipline problem; they are a record of people organising around an absence.
Reading them that way makes the conversations easier and usually more productive, because it is accurate.
What it costs to leave them unexamined
Two things, both compounding. The arrangements spread, because what one person has visibly been allowed becomes the baseline for the next request. And they become harder to vary each year, because reliance accumulates with time.
Neither cost is visible in any report. They surface as a pattern that cannot be changed and nobody can explain, which is the state most teams with a long gap in management eventually reach.
What good looks like
A team where every arrangement is either recorded with a review date or honestly marked as a practice nobody has sourced, and where nobody is surprised by the contents of that list.
That state is reachable in a few weeks of short conversations, and it survives the next three managers, which no amount of informal understanding does.