The saleswoman at Avenue Montaigne turned the jacket inside out without asking

The saleswoman at Avenue Montaigne turned the jacket inside out without asking. Seams bound in silk grosgrain, the lining hand-stitched at the armscye, and — there, just above the left interior pocket — a tiny label bearing not the logo but a single word: Tailleur. She said it as though I should already know. I did, barely. It's the house code for tailored pieces, a classification system Dior used in the fifties to organise the atelier's output. Robes. Cocktail. Tailleur. It still appears, occasionally, inside jackets that will never be turned inside out by anyone but their owner and a very patient cleaner.

That label is the kind of house signature we've stopped looking for. We know the Saddle, the Book Tote, the cannage quilting. We recognise Maria Grazia Chiuri's tulle from a hundred metres. But Dior's actual codes — the structural ones, the ones that shaped how clothes were built rather than how they photographed — have receded behind the logo. Some of them still appear. Most people don't notice.

The founding vocabulary

Christian Dior opened in 1947 with a collection the press called the New Look, a term he never used himself. What he did use, obsessively, was a technical vocabulary borrowed from architecture and horticulture. He named his silhouettes after letters and flowers: the A-line, the H-line, Tulipe, Muguet, Flèche. Each one was a structural proposition. The Tulipe dress, from spring 1953, tapered from a rounded shoulder to a stem-narrow hem. It required fourteen metres of taffeta and a system of interior boning that turned the torso into something between a corset and a scaffold.

The atelier worked in what Dior called lignes — not trends, not themes, but literally lines. A collection was a series of geometric problems solved in fabric. He would sketch a silhouette as a single fluid stroke, then hand it to the premières to translate into pattern pieces. The gap between the sketch and the garment was where the house's actual technique lived. A 1955 invoice I once saw at the archive listed a day dress as 'robe ligne A, taille cintrée, jupe 14 lés'. Fourteen panels in the skirt. The line was the concept; the construction was the signature.

Very little of this language survives in the way we talk about Dior now. We talk about the bar jacket, which is fair — it's the most famous thing the house ever made. But we don't talk about the way that jacket's peplum was engineered to cant forward by three degrees, or the fact that its shoulders were built over a pad stuffed with horsehair and shaped to push the clavicle back. The jacket made the New Look, but the interior construction made the jacket.

What lasted, what didn't

Some of Dior's codes embedded themselves so thoroughly into fashion that they stopped reading as Dior. The nipped waist, the full skirt, the emphasis on the hip — by 1955, every couture house in Paris was working some version of that silhouette. It became the fifties, not the maison. What stayed specific to Dior were smaller things. The grosgrain ribbon used to finish interior seams, always in a colour slightly darker than the shell fabric. The practice of lining evening gowns in silk taffeta even when the outer layer was wool, because taffeta held the shape and made a particular sound when the wearer moved.

Yves Saint Laurent, who took over in 1957 at twenty-one, kept the structural rigour but softened the references. His Trapèze collection in 1958 was still built on a ligne — the name gave it away — but the silhouette released the waist and swung free from the shoulder. It was a Dior code deployed against Dior's original premise. The house didn't know what to do with it. The press loved it. Clients were confused. Saint Laurent lasted six years.

Marc Bohan, who followed, spent twenty-nine years at Dior and is now mostly remembered for not being someone else. He was technically conservative and commercially consistent. What he understood, and what kept him employed through the seventies and eighties, was that the house's real clients didn't want revolution. They wanted the grosgrain seams and the taffeta linings. They wanted the Tailleur label. Bohan gave them that, and the house stayed solvent.

Gianfranco Ferré, in the nineties, came closest to retrieving the architectural language Dior had used at the founding. He built dresses the way Dior had built dresses: as structures first, as garments second. A white poplin shirt from his 1996 collection required a sleeve head so complex the atelier had to draft it in muslin four times before cutting the final cloth. It was commercially insane and technically flawless. The house fired him after seven years.

The current proposition

Maria Grazia Chiuri has now been at Dior longer than Saint Laurent and Ferré combined. Eight years. She arrived with a clear mandate — make the house speak to a younger client, make it feel political, make it move product — and she has done all three. The Book Tote is now more recognisable than half the bags in the archive. The slogan T-shirts sold. The tulle skirts photograph beautifully and ship consistently.

What she hasn't done, or hasn't tried to do, is retrieve the old codes. The lignes are gone. The technical vocabulary is gone. Chiuri works in references — she'll pull from the archive, from Bohan or Ferré, but she pulls images, not methods. A recent collection referenced the fifties bar jacket, but the jacket itself was unstructured, the peplum a loose ruffle, the whole thing soft where the original had been engineered. It looked like the archive. It didn't behave like it.

This isn't a failure. It's a choice, and probably the correct commercial one. The client who wants a sculptural jacket built over horsehair padding is not the client buying into Dior in 2025. That client exists, but she's likely at Alaïa or The Row, houses that still treat construction as content. Dior's current client wants the logo, the recognisability, the tote bag, the discourse. She wants the house to have a position on feminism. She does not, as a rule, want fourteen panels in her skirt.

The risk is that the house begins to forget its own technique. Not the imagery — the imagery is everywhere, exhaustively referenced, easy to mine. The technique. The way a sleeve was set, the way a seam was bound, the specific angle of a peplum. Those things live in the atelier, and ateliers have institutional memory until they don't. Skills leave when people leave. A première retires, and the way she used to pad a shoulder leaves with her, unless someone thought to write it down or film it or teach it to the next generation.

What remains

There's a room at 30 Avenue Montaigne — not the public-facing boutique, the atelier upstairs — where they still keep the wooden dress forms used in the fifties. Each one is labelled with a client's name and measurements, carved into the wood at the base. Madame S., Madame D., Madame R. The forms are no longer used, but they haven't been thrown out. Someone dusts them.

I asked once whether the current atelier ever references them, whether a première might pull an old form to check a proportion or a fit. The answer was no, not really, but they're there if needed. It felt like an honest answer. The house knows what it has. It just isn't sure, right now, what to do with it.

The grosgrain seams are still there, inside jackets that cost five figures and will be worn a dozen times. The Tailleur label still appears, though I've met clients who own the jackets and have never turned them inside out. These aren't the codes that sell bags or generate runway coverage. They're the codes that survive in the gap between what a house says it is and what it actually knows how to do. Dior still knows. It just doesn't lead with it anymore.