The Hour That Had to Be Agreed

AUntil the middle of the nineteenth century every town kept its own clock by the sun. Noon in Bristol fell about ten minutes after noon in London, and nobody minded, because nothing moved fast enough for it to matter. Railways changed that. A train leaving one town by local time and arriving in another by a different one turned timetables into puzzles and signalling into a hazard, and the companies imposed a single clock along the line, the Great Western adopting London time as early as 1840. The received account runs from that fact to a conclusion: the railway made standard time. A survey of 402 British and continental towns published in 2017, built from council minutes, newspaper notices and church-clock ledgers, complicates it. Adoption dates were scattered across five decades, and towns on main lines proved as likely to be late as early. The uniform response the railway account predicts never appeared, and the argument since has run between those who study engines and those who study workshops.

BHelena Marsh, an economic historian at the Institute of Transport History in Leeds, holds that the survey counted the right towns and the wrong dates. Between 2003 and 2019 her group worked through 1,140 printed timetables and 318 accident reports covering 1838 to 1889, dating adoption not from a council's formal vote but from the first timetable quoting a single time for the whole line. Measured that way, towns lying within twelve kilometres of a station changed their clocks a median of nine years earlier than towns beyond that distance. The discrepancy, on her account, comes from reading minutes and calling them practice: a council could ratify in 1878 what its stationmaster had enforced since 1856. She is careful about scope. The relationship held only where a town had a daily service in both directions; where trains called twice a week, nearness to the line predicted nothing at all, because the timetable never became a daily discipline.

CThe account makes a prediction, and Marsh has tested it. If a shared timetable is what teaches a town the new hour, places reached by the telegraph before the railway should have standardised early as well, since the wire also carried a time signal. Her group traced 46 such towns across England, Wales and the Rhineland. Adoption within five years of the wire arriving stood at 71 per cent where the town already maintained a public clock on a church or an exchange, and at 8 per cent where it did not, since a signal nobody could see corrected nobody's watch. Standardisation, in other words, was effectively confined to places that already possessed something to correct. Marsh therefore treats the timetable as an instruction a town may or may not be equipped to obey, and she accepts that where no public dial stood the instruction went unheard for reasons her transport model does not contain.

DAnselm Roux, who studies labour history at the Centre for Industrial Archives in Ghent, replies that a clock obeyed by nobody but a stationmaster is not a standard. Across 88 municipal minute books from the Franco-Belgian industrial belt he dated every adoption vote and set it against the opening of local mills: between 1851 and 1893, adoption followed a median of three years after the first works employing more than five hundred hands began operating in the parish, and bore no relation to the year the railway arrived. Standard time, he argues, was imposed by the shift bell rather than by the engine, since a factory paying by the hour needs one hour that everybody accepts, and much of what the survey scores as railway influence is a workshop discipline that happened to reach town along the same road. He concedes two things. His parishes lie within one industrial belt, and minutes record decisions rather than motives.

EThe dispute stopped being antiquarian in 2022, when a parliamentary committee reviewing the seasonal changing of clocks called both historians to give evidence. The committee's revised procedure requires those proposing to abolish the change to demonstrate a public benefit, whereas the earlier practice had obliged those defending the present arrangement to show that abolition would do harm. Roux points out, uncomfortably for his own position, that a country can standardise its clocks only once, so the mechanism he defends explains a single episode and cannot be stretched to cover whatever legislators do next. Marsh answers that a mechanism confined to one episode at least leaves an archive behind it, whereas the readiness she describes can be undone by a single generation that stops winding the clock. Both regret the same habit: that the literature keeps reporting the date a town adopted the new hour, a figure one council vote can move, instead of evidence that people actually kept it.

FWhat is emerging is not a verdict but a division of territory. In the manufacturing districts of northern France, Belgium and Lancashire, the timing of adoption follows the ledger of mills, and the railway explains what is left over; along the thinly settled corridors of the American plains and northern Sweden, where no large works existed to set a bell, the transport account is left in possession of the field. The awkward cases lie in between, and no apportionment has yet been proposed that both camps accept. A larger question is untouched by any of it: whether households in such towns actually abandoned the older reckoning, or kept two times in parallel for a generation, one for the train and one for the fields and the church. Diaries and dated letters could settle it, and no collection yet assembled runs long enough. What both reject is the assumption written into the earliest chronologies, namely that the date on a council minute can be read as a report on how people lived.