Bureau of Anomalous Regulation
DOC-000-17 / Anomaly Regulation Case File
CASE-1240 - "SMOKER'S CURSE"
Control Code-Modifiers
HYP-L
Hazard Types
T
Protocol
PALE HORSE
Lead Investigator
Dr. Eleanor Pask / BAR-OPM-40718
CONTAINMENT PROTOCOL
CASE-1240 is not subject to physical containment. The effect attaches to an act performed under the conditions set out in the Phenomenological Summary, not to any object, location, or person, and it occurs wherever those conditions are met. Bureau measures are confined to observation, reduction of qualifying instances, and restriction of Bureau personnel. CASE-1240 is held under Control Code HYP-L. The effect is known only through its outcomes, and the alteration of prior events that it is understood to involve has never been directly recorded. See Incident CASE-1240-1.
OPM maintains the instance record opened in the Foundation Era and continued without interruption since reclassification. The record pairs observer logs of ignition times, taken to the second against the posted scheduled time, with arrival data from transit authorities. Analysis is performed quarterly and reported to the Lead Investigator. Controlled trials follow the randomized-ignition design adopted in 1961, with every ignition time drawn in advance. No trial is to involve an event whose delay is deliberately maintained. Designs that attempt to establish the length of a delay before ignition are discontinued following Incident CASE-1240-1.
To reduce qualifying instances, DAI supports civilian restrictions on smoking at stops, shelters, platforms, and transit waiting rooms where such restrictions are already under civilian consideration. Support is advanced on public-health grounds alone, and no Civilian Communication is issued. DAI does not originate such proposals, and no other reduction measure is authorized. Civilian belief in the effect is classified as self-normalizing under REG-800-12. No amnestic procedure applies, and personnel who encounter the belief are neither to confirm nor to dispute it.
Bureau personnel are prohibited from lighting a cigarette while awaiting any overdue event, on or off duty, except as assigned participants in an approved trial. Violations are to be self-reported to OBP, and an unintentional violation is recorded as Level I.
PHENOMENOLOGICAL SUMMARY
CASE-1240 is a temporal effect associated with the lighting of a cigarette by a person awaiting an overdue event. The effect has no identified carrier. No property of the smoker, the cigarette, or the location has been found to bear on its occurrence; it has been recorded wherever its conditions have been observed.
Three conditions qualify an instance. A person is awaiting an event that has an expected time of occurrence; that time has passed by at least one second; and the person lights a cigarette. Under these conditions the awaited event occurs shortly after ignition, at a median interval of 57 seconds and an observed range of 9 to 143 seconds. Where no cigarette is lit, the event occurs at whatever time it otherwise would. The boundary falls at the expected time itself, to within the one-second resolution of available records. A cigarette lit one second before the expected time has no observed effect; one lit a second after it qualifies.
Scheduled road, rail, and water transit account for the large majority of recorded instances, because transit routinely records both expected and actual times. Instances have also been recorded involving postal and courier deliveries and the arrival of persons expected at an appointed time.
In every instance examined, vehicle location data, operators' logs, and the statements of drivers, passengers, and awaited persons place the event, at the moment of ignition, exactly as far from occurring as the interval that followed: a median of approximately one minute. None of these sources indicates the delay that would otherwise have been expected. Because ignition times in controlled trials are set at random, anticipation of the arrival by the smoker is excluded as an explanation. OPM's working model holds that ignition under qualifying conditions determines which of the event's possible recent histories took place, and that the history in which the event is delayed leaves no trace. Whether that history occurred in any sense has not been established. How far the determination extends beyond the awaited event is unknown.
In the typical observed instance, the smoker attends to the arriving event and discards the cigarette. Cigarettes recovered at observed stops following qualifying instances show a median burned length of 11 mm.
The effect has been known in civilian folklore since at least the Foundation Era, in the form of the saying that lighting a cigarette will bring a late bus or train. Civilians generally attribute it to coincidence and selective recollection. OPM estimates qualifying instances at ████████ per day. No cumulative consequence of the effect has been identified beyond the arrivals themselves. OPM has recorded that it has no method by which such a consequence could be identified, since it would reside in histories that are not retained.
NOTES & OBSERVATIONS
18██-11-02 — Disposition, Foundation-era notation retained — Investigator H. ██████
The porters' saying holds. Of 212 late arrivals entered in the book since March, 71 were attended by a passenger lighting a cigarette on the platform after the train was due, and every one of the 71 came in within two minutes of the lighting. The remainder came in as late as they pleased. I can find no way the smokers could know. No harm is done that I can see, save to the cigarettes. Recommend the book be kept.
1913-02-17 — Reclassification — Documentation Specialist ████████ / BAR-OPM-00088
Foundation-era file translated into Modern Era format and registered as CASE-1240. The original arrivals-book annotations and the investigator's disposition are retained. Hazard assigned T on the evidence of the tally. Control Code HYP-L assigned, the alteration of prior events having been inferred and not observed. Protocol computes to PALE HORSE. No override entered.
1931-09-30 — Tally Summary — Investigator C. ██████ / BAR-OPM-01944
Successive investigators have carried the tally at railway platforms and, as those services came into use, at tram and motor omnibus stops. It stands at 4,006 qualifying instances, and in each the awaited arrival followed ignition within 150 seconds. Timing is by observer's watch and is not reliable below ten seconds.
1961-06-19 — Randomized Ignition Trials — Dr. M. ████████ / BAR-OPM-07215
To exclude the possibility that smokers light up in unconscious anticipation of arrival, ignition times were drawn from published tables of random digits and assigned to observers in advance. Observers lit at the assigned second regardless of any sign of approach.
- 612 trials with ignition assigned after the scheduled time: arrival followed within 150 seconds in all 612.
- 300 trials with ignition assigned before the scheduled time: no association with arrival.
Anticipation is excluded. The randomized design is adopted as the standard for all further trials.
1988-01-12 — Memorandum: Reduction of Qualifying Instances — Supervisor A. ████████ / BAR-DAI-12054
TO: OPM, CASE-1240 file
FROM: DAI, Supervisor A. ████████
RE: Reduction of qualifying instances
In view of OPM's frequency estimate, and of OPM's statement that it has no means of assessing cumulative effect, DAI will support restrictions on smoking in transit waiting areas where such restrictions are already under civilian consideration. Support will be advanced on public-health grounds alone. The public-health case requires no fabrication, and no cover story will be issued. DAI will not originate proposals and does not expect to reduce instances at unsheltered stops.
2026-08-14 — Current Analysis — Dr. Eleanor Pask / BAR-OPM-40718
Arrival data from ██ transit authorities for 2019 to 2025 were set against 3,418 observer-logged qualifying ignitions. Arrival followed every qualifying ignition within 143 seconds, at a median of 57 seconds. Ignitions logged before the scheduled time show no association. The boundary falls at the scheduled time to within the one-second resolution of the vehicle location records, and those records agree in every instance that, at ignition, the vehicle was exactly as far from the stop as its arrival time implies, a median of approximately one minute away. Retention of HYP-L is recommended.
Aggregate punctuality figures for the authorities in the dataset run marginally better than their operating conditions predict, by a margin of the order the frequency estimate would produce. I have found no way to separate it from ordinary variance, and record it only because it is the first measurement in this file that may show the effect at scale.
DISCOVERY LOG
The saying that a late train can be brought in by lighting a cigarette was current among the porters at ██████ Junction in the Foundation Era. It reached the Bureau through Investigator H. ██████, who was at the junction on an unrelated inquiry when his own train ran late and a porter offered him the saying as practical advice. His notes do not record whether he took it.
At the investigator's request, the stationmaster had the junction's arrivals book annotated from the following March. A clerk marked each late arrival at which a waiting passenger was seen to light a cigarette after the train was due, and entered the minute of the lighting beside the minute of arrival. The clerk kept the annotations for eight months. They are retained in the file with the investigator's disposition, dated that November.
Foundation-era practice rested on guidance, and no containment was attempted. Successive investigators kept the tally as a matter of practice, and it continued after the file was reclassified into the Modern Era framework in 1913. It has grown by accretion since, from a clerk's marks in the margin of an arrivals book into the longest continuous observational record held by OPM.
CASE-1240 remains classified HYP. The Bureau has recorded the effect continuously for more than a century, at a rate it can only estimate, and has never once observed it take place.
INCIDENT LOG
1974-10-14 – Incident CASE-1240-1: Upstream Observation
The upstream observation trial was designed to record the delay that ignition removes. An observer was stationed at a point on the route approximately twenty minutes' running time before the trial stop, in telephone contact with a trial officer at the stop. The ignition signal was to be given only on a report from the upstream observer that the vehicle had not yet passed his position, and only if that report was received at least ten minutes after the vehicle's scheduled arrival at the trial stop.
The vehicle was scheduled at the trial stop at 16:02. The trial officer's log records the ignition signal at 16:14:54 and the vehicle's arrival at 16:15:52. The upstream observer's log records the vehicle passing his position at 15:55, consistent with a normal running time for an arrival at 16:15. It also records a telephone report to that effect, which the trial officer's log shows as received at 15:56. No report meeting the condition for the signal appears in either log.
The trial officer, Investigator J. Mercer / BAR-OPM-19230, gave the following statement on the day of the trial.
I gave the signal. I don't know why. The log's right, he rang through just before four and said it had gone past, and I read it back to him. Then at quarter past I gave the signal.
The trial was run twelve times between 1974-10-14 and 1975-03-21, with three trial officers in rotation. In every run, the upstream log recorded a passing time consistent with arrival within a minute of ignition, and the trial officer had given the signal without the qualifying report on record. None of the three officers could account for the decision.
Upstream observation trials were discontinued on 1975-03-21. OPM concluded that any report establishing the delay belongs to the history that ignition removes, and that no design of this kind can succeed. The prohibition on Bureau personnel lighting a cigarette while awaiting an overdue event was issued on 1975-04-02, on the grounds that the series had indicated that qualifying ignition can alter Bureau records without trace.
2019-11-26 – Incident CASE-1240-2: Self-Reported Violation
Observer J. ████████ / BAR-OPM-33871, off duty and waiting at a bus stop in rain, lit a cigarette at 18:31:07. The bus was scheduled at 18:27 and arrived at 18:32:02. She reported the violation to OBP the following morning.
It was raining. It was four minutes late and I lit up without thinking. I don't know how late it was going to be. Nobody does now.
OBP was unable to verify the report. Vehicle location data place the bus approximately one minute from the stop at 18:31:07, and nothing in the record distinguishes the instance from ordinary lateness. A Level I violation was recorded on the Observer's statement alone.
It is the only violation of the personnel prohibition on record. OBP has noted that the prohibition can be enforced only through self-report, and has made no change to it.
EVIDENTIARY ATTACHMENTS
Upstream observation logs — trial 1240, 1974-10-14
Post-instance recovery — evidence 1240-E7
Platform still — CAM 3, Rivermere Transit, 2024-11-14
End of file.