An Overview of the Costume Documentation Project

Jan Glier Reeder
Consulting Curator for the Brooklyn Museum Costume Collection at The Metropolitan Museum of Art

Presentation Slides

I am pleased to have opportunity to outline for you the procedures and protocols that the Costume Documentation Project (CDP) staff developed to accomplish the goals set out in the Andrew W. Mellon Foundation grant awarded to the Brooklyn Museum in 2005. The goal was to more clearly define the scope, content and quality of the museum's costume collection. To this end, the project staff's mandate was to fully catalogue, photograph, assess and inventory each item. A further directive was to identify during the assessment process a corpus of four thousand objects that represented some of the best examples from each object type in the collection. These were to be dressed on forms or otherwise properly mounted for high resolution photography that at the end of the project would be made available on ARTstor, which as I am sure most of you know is an online art image library and research initiative used by many educational and nonprofit institutions globally. We referred to this select group as the "Special 4,000."

The material to be included in this project was the costume portion of the museum's costume and textile holdings. This incorporated all seventeenth- to twentieth-century Western European and American fashions and accessories; twentieth-century jewelry; costume-related lace pieces; Eastern and Western European folk costume; and ecclesiastical objects. Other regional costume material, most notably African, Asian and Islamic, and all flat textiles were not part of the project. The duration of the project was designated to be three years, ending in January 2009.

Initial Planning Stage

Upon my arrival in fall of 2005, much of the groundwork had been laid by museum staff. A 15,000-square-foot work space at 25 Elm Place in downtown Brooklyn on the Fulton Mall had been obtained and made habitable; the technological infrastructure connecting us to the museum's computer network and to its digital laboratory had been laid; and initial plans for the interior build-out had been drawn up. Additionally, minimal records that included only the accession number, credit line, the object name and sometimes a brief description for each recorded object in both the costumes and textiles holdings had been created in the museum's database system —The Museum System, or TMS, as it is known; a group of costumes that had been stored in one of the more remote areas of the museum had been moved to a more accessible area; and one intrepid art handler had been working over the summer identifying contents of boxes.

For the first three months after I arrived, I worked with museum staff to complete the processes already underway which included the final design and build-out of the Elm Place space; planning and assisting with the preparation for the transfer to the new space; and writing job descriptions and hiring Project staff. The first of the new staff to take her position was the Collections Manager, Elizabeth Fiorentino, who over the next two months directed a team of six art handlers in preparing the objects for the transfer to the downtown space and created and implemented the logistics involved. Liz is going to tell you more about this later, but suffice it to say that the collection, which had been stored in nine different areas at the museum, was physically transferred to the new work space in eleven days; and for the first time in its history it was housed in a single dedicated area. (This box, one of the last to arrive in the new space, was appropriately marked “The End” in packing tape, by relieved art handlers when they sealed it up at the museum.)

Simultaneously during the planning period preparing for the move, a professional photographer, Lea Ingold had been hired and took on the task of planning a photo studio within the space and selecting and ordering all of the photographic equipment needed for both the documentary photography and the high resolution images of the Special 4000 that she would be capturing.

We moved into the space in late December and there were four of us in addition to the photographer in place at that time. Our first task was to establish a three- year work plan, but we did not know how many objects we were dealing with. Estimates had varied considerably from 35,000 to 70,000. In order to establish the plan, then, we first conducted two object counts, one based on the number of shell records for costume and textiles that had been entered in TMS, as I described earlier, and the other a physical count, based on the actual number of containers we had moved, and estimates of the number of objects in each container. For the record count, it was necessary to sort out the textile records from the costume records and to define all of the object types and titles within them to search under. In regard to the physical count, while it was possible to estimate with some degree of accuracy the number of objects hanging in garment carts, educated guesswork was required for the contents of the many of the boxes, drawers and cabinets that held small objects such as gloves, stockings and handkerchiefs. The spreadsheet lays out our methodology and the totals from the two counts. In the first column are object types and the second column names all of the titles within those types under which the shell records were searched and in the third column are the results of that count, which was 26,799. In the fourth are results of the physical count, which was 20,034. Both of these were considerably lower than we had expected, even though we knew that many records contained more than one object. We decided on the larger number, 27,000 records, as our point of departure. To establish the three-year plan, we divided our three years into twelve three-month quarters, divided 27,000 by twelve months, and established the goal of processing 2,250 objects per quarter. We then broke that down into the number of work days per quarter and then further into the number of cataloging hours per day to come up with the goal of cataloguing forty objects per day at eight minutes per object. We decided to start with twentieth-century garments because it was the most familiar material, the easiest to physically manage, and therefore a good way to test whether our procedures were viable.

We also recognized it was going to be important to vary the type of material we were cataloguing to avoid tedium, so we intentionally built in a change of object types from one quarter to the next. As you can see in this first three-year plan, we switched to cataloguing hats after two quarters of twentieth-century garments, then went back to twentieth-century garments later.


By January 16, 2006, we were ready to begin. All staff were in place, we had a physical design that allowed for flexibility, adaptability and accessibility to each other, and the photo and other technical equipment were installed. It included two Nikon digital cameras tethered to Macintosh computers at two photo stations; a Hasselblad H2 format camera with an Imacon back, also tethered to a Mac, in the photo studio; four laptops on rolling carts; and ten PCs in offices. There were eleven of us on staff at a given time. In addition to myself in the position of curator were assistant curator Grace Hernandez in charge of overseeing the integrity, consistency and accuracy of the database records and assisting me in all ways; Liz Fiorentino, the collections manager, who oversaw the physical and technical logistics including keeping track of the objects; conservator and dresser, Glenn Petersen, who styled, dressed and prepared appropriate mounts for photographing the special 4,000, as well as overseeing the care and handling of the collection and cataloguing the shoe collection; Lea Ingold, the professional photographer, who was succeeded by Lolly Koon mid-way through the project; and six research assistants, all of whom held master's degrees in the study of costume and textiles—Jessa Krick, senior research assistant, Jennifer K. Holley, Katie Netherton, Elizabeth Randolph, Jessica G. Rall, Rachel Tu, Deborah Saville, and Katherine M. Winters. The research assistants performed daily and weekly rotating tasks which included object preparation, data entry, documentary photography, accession record review during the cataloguing process, object research, and one discreet research project on a topic related to the collection to be completed by the end. Some of these will be presented this afternoon.

The Workflow

The staff worked as ensemble team in the staging area to process each item. After a preparatory stage, where objects were located and tagged, items were brought forward one and at a time and hung on a costumer or placed on the table. The cataloguer verbally provided the data about the object to the research assistant at the computer who entered it into the database record. Another research assistant was reviewing the original accession record for the object. These records, which were bound in books, were laid out on tables at the back for easy accessibility. The research assistant read the record aloud before the verbal cataloguing began, and at the end any discrepancies were noted and any pertinent additional information was added to the database record. A time-limited attempt was made to reconcile any discrepancies between the existing and the new records by referring to the donor files which were in an adjoining room. (If discrepancies were not easily resolved, the appropriate box was checked on the front card and the process moved on to the next object.) When cataloging was completed, the item was placed in a designated holding area to be photographed on a hanger or other appropriate mount in the photo station. In the case of the "Special 4,000," the object was sent to the photo studio to be styled and dressed on a dress form or mannequin for high resolution imaging. Simultaneously, during the cataloguing process the collections manager was planning the next group of items to be reviewed and in addition updating locations, entering component counts, and resolving cataloguing or numbering conundrums. After an item was photographed, it was then placed in the appropriate storage container which was its permanent location for the duration of the project. The photographic images taken each day were stored in an onsite server overnight and transferred the next day to the digital laboratory. The digital lab was a newly established museum department created for the efficient management of digital imaging museum-wide. The lab's staff performed a crucial role in the CDP process collating and linking images generated by CDP photographers to the database system as well as uploading them into Luna, the Digital Asset Management System that was implemented in the course of the project.

The Cataloging Process

In collaboration with the museum's registrars, the CDP team developed a collection-specific data screen (front screen) that included 24 data fields. The information that the cataloguer verbally provided to the data entry person, as I described earlier, was to fill in those fields. This information included object title, medium, description, date, maker, detailed label information, a brief assessment of condition and storage needs, and one of three recommended designations: the "Special 4,000" category; museum worthy, to be maintained as part of the core collection; or candidate for deaccession. Color-coded stickers were affixed to the tags on each garment to indicate the designation: pink for "Special 4,000," green for recommended deaccession, and a single mark written on the tag for core collection.

The criteria for recommending deaccession were primarily condition, redundancy, and lack of historic or aesthetic distinction, as outlined here, compared with other similar works in the collection. The criteria for a masterwork included importance within the history of fashion and/or a given designer's oeuvre, historic significance, outstanding aesthetic appearance or technical virtuosity, provenance and rarity. These reasons were noted in curatorial remarks field on the front card.

As cataloguing progressed, we developed various documents that set out protocols, procedures and guidelines related to such cataloguing issues as object titles, front card fields, constituents, editorial style, numbering, and technical definitions. These were works in progress which were kept updated by what we termed the Daily Roundup. When a new protocol was determined during the cataloging process, the person doing data entry quickly noted it in the Daily Roundup, and at the end of each quarter the senior research assistant added each protocol from the Daily Roundup to the corpus of the appropriate document. In this way a fair amount of consistency in the way the information for each type of object was entered was maintained. In the example shown here, the discussion is about the order in which to catalogue lace pieces.

"Special 4,000" Photography

The goal of the "Special 4,000" photography was to make the object understandable and attractive without being overly editorial by accessorizing or using colored backgrounds or dramatic lighting. Initially it was unclear as to how much curatorial input would be needed for placing and styling the objects and positioning the mannequins, but Glenn, the conservator and dresser, who had had years of experience doing this, soon assumed the major responsibility. I would check in periodically and he would consult with me when there were certain decisions to be made. This easy communication was facilitated by the close proximity of all the work spaces. Also printouts of the reasons why objects had been chosen for the "Special 4,000" accompanied the garments to the dressing area. Glenn would note what had been indicated as important about the object, such as seams, ornamentation, or some aspect of silhouette, and plan with the photographer to take shots to include or emphasize those features. I reviewed the images each week with our photographer as well to assure they fulfilled our needs.

There was a selection of essentially five different types of forms to work with—including mannequins, display forms, and dressmaker forms—each with variations. Almost all of the forms were existing property of the museum and fortunately were adequate to most needs. The selection of form depended on the period of the garment, size or fit, construction, and aesthetic considerations. The quantity and variety of forms was such that a constant stream of objects to the photo set could be maintained. Although there was general consistency in the types and dates of objects being shot at one time, judicious selection of the order of objects was sometimes necessary to avoid jumping between different periods on any given day.

The most challenging aspect of the "Special 4,000" photography for both the dresser and photographer was maintaining the pace over the duration of the project, which required a minimum of seven objects to be prepared, dressed, styled, photographed and undressed each day. While it was fairly easy to keep up with the twentieth-century garments, where the occasional difficult object could be dressed between several easy pieces, it was impossible to keep up the same pace with the nineteenth-century garments and certain of the twentieth-century ball gowns and wedding dresses, which were often cumbersome and heavy and much more time-consuming to prepare and dress. Slow-downs in the photography were however, compensated by periods of accessory photography, which could progress at a faster pace. In addition, as the actual cataloging was completed three months before the end of the project to allow for record review and research, there was ample time for the backlogged photography to be completed.

For flat pieces and garments which could not be dressed, there were various sizes of fabric-covered slant boards. In addition, various tabletop setups accommodated photography for accessories such as shoes and hats.

The Results

In reality the forty-object-per-day/eight-minute per object time frame varied wildly. But a close eye was kept on the numbers during the quarter and the plan was adjusted when necessary to balance the goal requirement with the varying times needed to catalogue each category. This sometimes required logistical high jinks such as having two teams simultaneously cataloguing or inserting material not originally in the quarterly plan that would move more quickly through the process. But establishing the goal to catalogue 2,250 objects each quarter as top priority was the driving force behind the successful completion of the task. The goals were met each time. (Despite moments when we lost our focus.)

If you remember at the beginning of the talk, I said we estimated we would be processing 27,000 records. In the end, the exact number was 23,834 records and approximately 30,000 objects (counting for instance a pair of shoes as two objects and a dress and belt as two objects.)

We compiled numerous categorical statistics, including approximate holdings of works by Charles James, Elsa Schiaparelli, and the House of Worth as well as items from Millicent Rogers' wardrobe. We additionally identified through the process a core group of costumes and accessories designed by American women in the 1930s, '40s, and '50s definitively establishing their consummate creativity and importance. Jessie Franklin Turner, Vera Maxwell, Bonnie Cashin, Eta Hentz, Carolyn Schnurer, Valentina, and Elizabeth Hawes are the most significant among those represented in the group.

The photographs and information of the "Special 4,000" group are now available on ARTstor as well as the Met's website and 250 objects selected from the four thousand are published in High Style: Masterworks from the Brooklyn Museum Collection at the Metropolitan Museum of Art.

Over thirty thousand photographs, such as these, were captured. Our collections manager was jumping for joy. . . and we were all jazzed up.


Explore the Met's publications.


Now at the Met offers in-depth articles and multimedia features about the Museum's current exhibitions, events, research, announcements, behind-the-scenes activities, and more.

Thomas J. Watson Library

Thomas J. Watson Library is the central research library of The Metropolitan Museum of Art—one of the world's most comprehensive collections of books, periodicals, auction catalogs, and online resources relating to the history of art. Open to all visitors college age and older. Search the collections.