The Barleycorn Standard
A short history of the international shoe size, which does not exist
There is no international shoe size. There is no shoe size at all, in the sense that a kilogram is a kilogram in Lyon and in Lagos. There are instead five or six regional conventions, none of them compatible, two of them derived from grain, one of them named after a European capital, and one — the correct one — used almost nowhere.
A single pair of adult feet, unchanged in length for twenty years, may be served by shoes marked 9, 8.5, 42, 27, and 275. All five shoes fit. All five numbers are accurate within their own systems. No two of them can be converted into each other without a table, and the tables disagree.
What follows is how that happened.
I. The unit
The English system is built on the barleycorn: one-third of an inch.
The origin story, repeated in nearly every account of the subject, holds that Edward II decreed in 1324 that three barleycorns laid end to end constituted an inch, and that shoes would thereafter be sized accordingly. The first half of this is roughly true and concerns inches generally. The second half is not supported by the record. The connection to footwear appears to have been added later by parties unknown, which means the founding document of Anglo-American shoe sizing is a story about a document.
The unit survived regardless. Each whole size is a third of an inch longer than the one before it. Each half size is a sixth of an inch, being half a barleycorn, a quantity the original decree does not address.
More consequential than the unit is what it measures. British sizing measures the last — the wooden form the shoe is built around — and not the foot that will go into it. The allowance between the two is left to the manufacturer, who is under no obligation to disclose it. The customer’s foot does not enter into the calculation at any point. It is not a variable in the system. It is the thing the system is eventually pointed at.
II. The offset
American men’s sizing sits one full size above British sizing.
It is not a conversion; the underlying unit is identical, and a barleycorn is a barleycorn on both shores. It is not a correction; nothing was wrong. The offset appears in the record already in place, without preamble, and no surviving document accounts for it. It has been maintained without interruption for more than a century, across two world wars and the entire history of American manufacturing.
American women’s sizing then departed from American men’s sizing by approximately one and a half sizes, so that a women’s 8.5 and a men’s 7 describe the same object. This offset is also undocumented, and is not consistent across all manufacturers, which means the discrepancy itself has a margin of error.
Against this, the United States produced the finest instrument in the entire history of the field. Charles Brannock of Syracuse, New York, the son of a shoe retailer, spent two years developing a device that measures heel-to-toe length, heel-to-ball length, and width simultaneously. He patented it in 1927. It is a genuine precision instrument, it is beautiful, and it remains the standard measuring tool in American shoe stores a century later.
Its readings are then entered into the barleycorn system, where they are used as the basis for a suggestion.
III. The Paris Point
Continental Europe declined the barleycorn and adopted the Paris Point: two-thirds of a centimeter, approximately 6.7 millimeters.
Get ganpy’s stories in your inbox
Join Medium for free to get updates from this writer.
The Continental system does one thing that no other system does. It runs a single continuous scale from infants to adults, with no separation by gender. A 42 is a 42. There is no hidden second scale, no offset, no translation required between the men’s rack and the women’s rack. Measured purely as design, it is the most coherent system in use anywhere.
Its increment, however, is coarse — larger than an American half size — and the gap between consecutive European sizes is therefore wide enough to accommodate a foot that fits neither. The standard makes no provision for half sizes. The industry, requiring them, introduced the 42.5, which has no basis in the system and is now sold in every shoe shop on the continent.
The Paris Point also measures the last. A German 42, an Italian 42, and a Spanish 42 are three different shoes bearing the same designation.
IV. The correct answer
Japan sizes shoes by the length of the foot in centimeters. A 26.5 fits a foot 26.5 centimeters long. Korea uses millimeters. China uses millimeters. Mexico uses centimeters. Roughly half the world’s population has been measuring the foot, rather than the box the foot goes in, for as long as anyone has bothered to check.
There is also a formal international standard. It is called Mondopoint. It records foot length in millimeters together with a width measurement, it is unambiguous, it is published by the ISO, and it is the correct solution to the problem in every respect.
It has been adopted by NATO militaries and by manufacturers of ski boots.
That is the complete list. The two constituencies operating on the rational standard are armed forces and people descending mountains at speed. In both cases a sizing error carries a physical consequence within a short and predictable timeframe. Everywhere else the barleycorn holds.
V. Two further complications
Australia inherited men’s sizing from Britain and women’s sizing from the United States. Both systems remain in force simultaneously, in the same shops, in the same country, on the same population.
India adopted the British system, and sells shoes in UK sizes to a market that is not, in general, measured. A size is arrived at once, in adolescence, by consensus, and then held for life.
India also operates the only meaningful control group. The chappal is nominally sized. The nominal size is not consulted. The purchaser puts the foot on, looks down, and reaches a determination in under four seconds. Some overhang at the heel is expected and is not treated as a defect. The unmeasured shoe, fitted by direct observation, achieves a success rate that the measured shoe does not approach.
VI. Present state
The systems were tolerable for as long as the customer and the shoe were in the same room. The number never had to work; it only had to narrow the search before the foot went in and settled the matter.
Online retail removed the foot from the transaction and left the number to carry it alone, which it cannot do. The industry’s response has been to construct an informal apparatus underneath the formal one. Buyers now consult written testimony from strangers regarding those strangers’ own feet: runs small, runs narrow, size up half, true to size but narrow through the toe box. This body of evidence is unverified, uncontrolled, contributed voluntarily, and more reliable than the number printed on the box.
Global footwear is a market of some hundreds of billions of dollars annually. Its sizing information is now transmitted primarily by oral tradition.
The clerk who is alleged to have laid three grains of barley end to end on a table has been dead for seven hundred years. The arrangement attributed to him remains in force.
Postscript: A note on the device
Charles Brannock built the first working prototype out of his childhood Erector set, as a student at Syracuse University. He began manufacturing in 1925; the patent was granted in August 1928. He never married, never left Syracuse, visited the factory daily until six months before his death in 1992, and still held the patent at the end. Salvatore Leonardi bought the company from the estate in 1993. Till date, more than a million devices have been sold, and they are still made in a small factory in Liverpool, New York, a few miles from where he was born.
The readings are still converted into barleycorns.
