Buttons sit on opposite sides and nobody can prove why

The convention itself

Pick up a man's shirt and the buttons are on the right-hand side as worn, with the buttonholes on the left, so the left front laps over the right. Pick up a woman's and it is reversed. The convention is remarkably consistent across garment types, countries and centuries, which is what makes the absence of an explanation interesting.

It is not written into any regulation. No trade body imposed it, no law requires it, and manufacturers who ignore it are not penalised beyond the mild confusion of customers.

It persists anyway, and it has persisted through enormous changes in how clothing is made and who makes it. That is a long time for a habit to survive with no stated reason.

The dressing-servants explanation

The most repeated account holds that women of the classes who bought elaborate fastened garments were dressed by servants, and that a garment buttoned by somebody standing opposite is easier to fasten if the buttons face them. Men dressed themselves, so their garments suited a right-handed wearer.

It is a tidy story and it has an obvious weakness. The great majority of women in the periods concerned did not have anyone to dress them, yet their clothing followed the same convention, and mass-produced clothing adopted it long after the practice would have mattered.

The counter-argument is that fashion has always been set by the classes that could afford it and copied downwards, so a convention originating in wealthy dressing rooms could easily have spread to garments nobody was helped into. That is reasonable and it is not evidence.

Swords, horses and infant feeding

Several other explanations circulate. One says a man drawing a sword with his right hand needed his coat to lap in a direction that did not catch the hilt. Another says women riding side-saddle wanted the overlap facing away from the prevailing wind on the side they sat.

A third holds that a woman carrying and feeding an infant on her left arm benefits from an opening that a right hand can manage. Each of these has an internal logic, and each explains part of the pattern.

What none of them has is a contemporary source stating the reason at the time it was established. The explanations are all reconstructions offered afterwards, which is precisely the pattern one sees whenever a custom becomes fixed before anyone thought to write down why.

Why a convention outlives its reason

Conventions survive because breaking them costs something and keeping them costs nothing. Once buttonholes are punched on a particular side across an entire industry, patterns, machinery, training and customer expectation all point the same way.

There is also a signalling function that developed after the fact. The side a garment buttons on became a marker of which department it came from, which matters commercially even when it means nothing physically. A garment that buttons the other way reads as borrowed, whether or not the wearer intends it to.

That secondary meaning is now the main reason the convention holds. The original cause, whatever it was, has been irrelevant for a long time.

What buttons themselves record

The buttons are worth a look on their own account, because they date and place a garment more reliably than the fastening side does. Buttons made from horn, shell, wood, glass and metal each belong to particular periods and price points, and the shift to moulded plastic is a twentieth-century marker.

The number and placement carry information too. A jacket's buttoning point — where the front fastens relative to the waist — has moved up and down over the decades and is one of the clearest ways to date tailoring by eye.

Working cuff buttons, sometimes called surgeon's cuffs, are a survival from a time when a sleeve that could actually be unbuttoned had a practical use, and they now function purely as an indicator of how the sleeve was constructed.

Living with an arbitrary rule

None of this makes the convention worth defending or worth attacking. It is a piece of inherited arbitrariness, like which side of the road traffic drives on, that solves a coordination problem without needing to be correct.

It is occasionally useful. Anyone sorting through unlabelled secondhand clothing can use the buttoning side as a quick signal, and anyone having a garment made can simply specify.

And it is a good example of a general truth about clothing history. A great many details that look like design decisions are the fossilised remains of practical arrangements, and a fair number are the remains of arrangements nobody bothered to record.