The Costume Institute at the Metropolitan Museum of Art holds more than 33,000 garments and accessories spanning five centuries, per the museum's own collection figures, and its long project of exhibiting dress as art history — institutionalized with the 1995 arrival of Harold Koda and Richard Martin's scholarly program and, from 2005, Andrew Bolton's exhibition-making — is the strongest single argument that a garment can be studied the way a painting is. The annual Costume Institute exhibition, mounted each May since 1948 under the museum's auspices, has become the reference point for what fashion-in-the-museum means. This essay reads that project from its documented record.
The argument the museum has quietly made is not that fashion belongs beside art because it is beautiful, but because it carries authorship, technique, and period meaning in the same demonstrable way.
What did the Costume Institute actually change?
Before the 1970s, dress in encyclopedic museums lived mostly in decorative-arts departments, studied as textile or craft. The Met's Costume Institute — formed from the Museum of Costume Art, absorbed by the Met in 1946, per the museum's published history — professionalized the field: dating garments by cut and construction, attributing houses and hands, building the scholarly apparatus a discipline requires. Diana Vreeland's consultant exhibitions of the 1970s and early 1980s brought spectacle and audience; the Koda-Martin era brought catalogue rigor; Bolton's exhibitions — 2011's Alexander McQueen retrospective among the most visited in the museum's history, per the Met's own attendance figures — fused the two. The sequence matters: audience first would have been circus; scholarship first would have kept the audience away. The museum ran both.
How does an exhibition argue authorship?
Through the object and its label. The 2011 McQueen exhibition, "Savage Beauty," presented some 100 garments and accessories from the designer's 19-year house archive, per the Met's exhibition records, organized thematically around Romanticism, tailoring, and the exotic — an art-historical reading of a fashion oeuvre, complete with an illustrated scholarly catalogue. The 2015 "China: Through the Looking Glass" and 2019 "Camp" exhibitions extended the method to aesthetics and theory. What the label does in each case is what a wall text does for a canvas: name the maker, date the work, and place it in a line of influence drawn from documents, not from the house's own storytelling. When the museum attributes a silhouette to a specific atelier hand, it is performing the same authorship criticism the painting department performs on a workshop canvas.
Why did the houses come to the museums?
Because the museum confers canonicity, and the houses recognized it. Designers now negotiate archive access with curators, donate bodies of work, and treat a museum exhibition as a career milestone in a way the industry's own awards never quite managed — a shift documented across the flood of fashion exhibitions at major institutions since the Met's McQueen moment, from the Victoria and Albert Museum's fashion program to dedicated museums like Palais Galliera in Paris, per each institution's exhibition records. The flow runs both ways: the 2018 Met exhibition on Comme des Garçons's Rei Kawakubo drew on loans the house made specifically for scholarship, per the museum's exhibition materials.
What does the open-access turn change?
The Met's Open Access program, launched in 2017, released images of public-domain works — including costume holdings — for unrestricted use, per the museum's program announcements. A student can now study the construction of an 18th-century robe à la française in the same interface that serves Rembrandt. The Costume Institute's physical galleries rotate, and much of the collection rests in study storage; the digital turn is what makes the 33,000-object argument public rather than professional. Other institutions followed — the Art Institute of Chicago and the Rijksmuseum's textile holdings among them, per their open-access programs.
Where does the argument stop?
Two honest limits. Conservation: garments are fragile in ways canvases are not, and the costume exhibition's short runs and low light levels — museum conservation standards restrict exposure — mean the discipline's public face is intermittent by necessity. And theory: whether a runway collection is art in the same sense a painting is remains a live argument among philosophers of art, with no museum settlement; the Costume Institute's success demonstrates that dress rewards the methods of art history, which is a different claim. What the record establishes: a century-long institutional project that made the garment legible as authored work. The stronger claim — settled canonicity — the record leaves open, and this essay leaves it there too.