Exhibitions, galleries and institutions

How curators find archives, and how an archive gets found

Illuminated angel figures with trumpets line a path towards a decorated Christmas tree at night.
Illuminated seasonal display.Photograph by Robert Harper. Robert Harper Photograph Library, Archive.London. From the archive, shown for illustration.

Curators rarely stumble on an archive. They follow a name through credit lines, authority files and catalogues until it leads to a door. An owner's task is to make sure the trail does not go cold.

An archive nobody can find is not hidden, it is merely unpublished, and unpublished is a condition its owner can cure in an afternoon.

The New York Public Library keeps a database of more than 110,000 photographers and studios, most of whom you have never heard of, and that is rather the point. Curators do not find archives by wandering into them. They follow a name: from a credit line in a magazine, through an authority file that confirms who the person was, to a catalogue that says where the work now sits. The usual advice to archive owners is to "raise the profile". Ours is duller and more effective: treat discoverability as a publishing job, with four or five specific databases to feed and one document to write.

How do curators actually look for a photographer's archive?

Research on photographers runs along a small number of trails, and all of them are made of paper.

Credit lines. Magazines printed the photographer's name beside the picture, and bound volumes survive in libraries. A researcher working on, say, British fashion editorial of the 1980s reads the gutters as carefully as the spreads.

Other people's biographies. Studio photography was taught by apprenticeship, so careers interlock. A curator preparing a show on a famous name will list that person's assistants, printers, stylists and rivals, because any of them may have kept what the famous name threw away.

Obituaries and catalogues. Obituaries name employers, colleagues and clients. Exhibition catalogues name lenders in their acknowledgements. Footnotes name collections. Each is an index to the next source.

Authority files and union catalogues. Once a curator has a name, they check whether it exists as a record, and where the holdings are.

The last trail is the one an owner can do something about, and the first three explain what to publish. They also explain why a photographer who was busy and well credited can still be invisible: the trail exists in print but ends at no address.

Which databases do researchers check for photographers?

Resource Run by What it holds How an owner can help
Photographers' Identities Catalog (PIC) New York Public Library Names, nationalities, dates, locations and addresses for photographers and studios, with links to other authority files Send additions or corrections by email or the feedback link, quoting the record ID
Union List of Artist Names (ULAN) Getty Research Institute Structured records for people and corporate bodies who make art and architecture Institutions, projects and individual scholars can offer contributions by contacting the Getty vocabularies team
Discovery The National Archives Catalogue descriptions from archives across the UK, and a directory of repositories Archives with an ARCHON code can upload descriptions through Manage Your Collections
Archives Hub Jisc Collection descriptions contributed by UK institutions Contributors send exports, set up a harvest, or fill in a spreadsheet template

A few details from each are worth knowing.

PIC describes itself as an experimental interface, and says its data comes from biographical dictionaries, catalogues and databases, from Wikipedia and Wikidata, and from library staff working through census returns, trade directories and death records. Each record lists its sources and links out to the Virtual International Authority File, ULAN and Wikidata where a match exists. The site gave a figure of about 116,100 names as of May 2016.

ULAN is larger and stricter. When we checked, the Getty reported 525,300 person or corporate body records carrying 1,483,845 names, since one person may have several spellings. Its scope is "creators of the visual arts", including firms and studios, and a minimum record contains a numeric ID, a name, a role, a nationality and life dates. People who cannot be given at least those are left out.

The National Archives says Manage Your Collections lets an archive add, edit, delete and restore information about its collections in Discovery, either from its own spreadsheet mapped to Discovery's fields or from a template that follows the international ISAD(G) standard. Archives Hub lists seven cataloguing systems that have successfully supplied it with data, and offers a spreadsheet for everyone else.

What is a name authority, and why does a photographer need one?

A name authority record is a numbered entry that says: this Robert Harper, born here, active then, a photographer, is one person, and these other spellings are the same man. Without it, every catalogue has to guess. Common names are the hazard. A curator who types a name into a search box and gets a nineteenth-century cleric, a jazz bassist and a professor of pharmacology will not, on the whole, persevere.

The minimum facts are modest, and ULAN's list is a good one to adopt: name, role, nationality, dates. To those add places of activity, the names of employers and teachers, and the publications where the work appeared. These are exactly the facts that let a stranger connect your photographer to a trail they are already following.

An archive nobody can find is not hidden, it is merely unpublished, and unpublished is a condition its owner can cure in an afternoon.

What should an archive owner publish to be found?

Five things, in order of usefulness.

  1. A collection-level description. One page stating who made the material, when, in what formats, how much of it there is, how it is arranged, who owns it and how to ask for access. Archivists call this a finding aid, and we have a separate guide to writing a finding aid for a photographic collection.
  2. An inventory, even a rough one. Folder titles and counts are searchable text. A curator looking for hi-fi advertising or Scottish landscapes in 1975 will search for those words, not for the photographer. Our guide to cataloguing a photographic archive explains why folder level is the realistic first target.
  3. A sourced biography. Dates, training, employers, publications, each stated plainly. Leave out adjectives; curators distrust them, and so do search engines.
  4. Stable addresses. A page a scholar can cite in a footnote in 2027 and still reach in 2037. Do not rename things for fun.
  5. Metadata that travels. Any image you publish should carry the creator, credit line and rights holder inside the file. The free add copyright metadata tool writes those IPTC fields without recompressing the picture, and our article on IPTC photo metadata fields for archives explains which ones matter.

Notice what is not on the list: a showreel of the best fifty frames. Highlights tell a curator that the archive has taste. An inventory tells them whether it has what they need.

Does a private archive belong in public catalogues?

Often, yes, though the route depends on who you are. The National Archives ties Manage Your Collections to the ARCHON code it issues to record repositories, and says the service can be an archive's sole means of getting collection information online. Archives Hub asks would-be contributors to apply. Neither is a place to upload pictures. Both are places to state that a body of material exists, what it covers and whom to ask, which is precisely the information a curator lacks.

Authority files are more open than owners assume. PIC invites corrections from anyone and names an email address for them. The Getty says institutions, projects and individual scholars interested in contributing to ULAN may contact its vocabularies team, and publishes guidelines for doing so. In both cases the price of admission is evidence: a published source for each fact. That is another reason to put a sourced biography on your own site first.

Listings of people and venues help in the other direction too. Our own index of photographers and photography directory of galleries, museums, fairs and festivals exist partly so that names and institutions can be looked up in one place.

What does this look like for a working archive?

Take the Robert Harper Photograph Library as a case, using only what is on the record. Robert Harper was born in Scotland, studied photography in London, spent two years at the Condé Nast studios assisting David Bailey, Helmut Newton and Lester Bookbinder, and then worked as a freelance assistant to Terence Donovan, Tony McGee and Alex Chatelain before opening his own studio. He was a regular contributor to Elle and The Telegraph Magazine for more than fifteen years.

Every clause of that paragraph is a trail. A researcher on Bailey, Newton or Donovan who lists their assistants arrives at the name. A researcher paging through Elle arrives at the credit line. Our piece on the Condé Nast studios years sets out that connection and says no more than "assisted", because that is all the record supports. The fuller account is on the Robert Harper biography page.

What Archive.London has done about the far end of the trail is publish. The complete 503-folder inventory is online as a searchable table and a CSV file, with frame and shot counts for each folder. There are 330 described collections across 24 categories, and a statement of ownership: 825 Ventures Ltd is the owner and rights holder. More than half of the archive is still on film and has yet to be scanned, so for most folders the words are, for now, the only thing a curator can search. We say that openly because it is useful to know before a visit, and because an archive that admits what it has not yet catalogued is easier to trust on what it has.

What should a curator ask an archive first?

If you are the one searching, three questions save a wasted journey: is there a list at folder level; what proportion is scanned; and who can grant permission to exhibit and reproduce. A single rights holder shortens the last answer considerably.

Curators, picture researchers and exhibition organisers who want to look further than the inventory can request professional access to the Robert Harper Photograph Library.

Frequently asked questions

How do curators find photography archives?

Mostly by following names. A curator starts from a published photograph, a magazine credit, an obituary or the biography of a better-known photographer, then checks name authorities such as the Getty's Union List of Artist Names and the New York Public Library's Photographers' Identities Catalog, and collection catalogues such as The National Archives' Discovery and Archives Hub, to see where the work is held.

How do I make a private photo archive discoverable?

Publish a collection-level description with the photographer's name, dates, formats, quantities, subjects and a contact, on a stable web address. Then put the same facts where researchers already search: union catalogues, name authority files and structured data on your own site. A folder-level inventory that can be downloaded does more than a gallery of highlights.

What is a name authority record and why does it matter?

It is an agreed, numbered entry for a person, so that every catalogue means the same individual. The Getty's Union List of Artist Names and the Virtual International Authority File are examples. A photographer with a common name and no authority record is easily confused with someone else, and a curator cannot cite an archive they cannot distinguish.

Can a private owner add collections to The National Archives' Discovery?

The National Archives says Manage Your Collections is open to any archive with an ARCHON code, the identifier it issues to record repositories, and that it can be an archive's sole means of getting collection information online. Owners without a code should ask The National Archives whether they are eligible to be listed before preparing data.

Do curators use Google to find archives?

Yes, alongside specialist sources, which is why plain descriptive text matters. Search engines cannot read an unscanned transparency. They can read a folder title, a date range, a list of sitters and a credit line. Embedded IPTC metadata in any published image carries the creator and rights holder with the file when it travels.

Sources

Published by Archive.London. Corrections are welcome under our editorial policy.

For picture agencies, publishers and researchers

Fashion, beauty, jewellery, music, portraits and British life from the 1960s onwards, with 52,684 analogue originals still to be scanned. Tell us what you are researching.

See representation and researchWrite to the archive