1 This note offers general observations on picture and photograph collections in certain Canadian archives and libraries, along with a few thoughts on the subject of indexing historical photographs and other illustrations. It has resulted from a number of sporadic field trips over a two-year period to provincial archives and other major repositories of photographs, prints, and drawings1 in order to find historical illustrations of domestic life in Canada, ca. 1840-1920. Forming a research collection, these are intended to support curatorial, interpretive, and restoration needs at National Historic Sites and to serve other scholarly purposes. Most of this collection consists of historical photographs, the vast majority taken after 1870, of Canadian homes and home life.2 (It includes such ancillary views as the interiors of commercial establishments containing domestic goods.) So far the collection amounts to about 5,000 reproductions — prints, photocopies, a few slides, and two reels of microfilm.

2 This note is based on a cross-section of archival research in every province and is a very general guide to major collections of historical photographs, prints, and drawings.3 It is written for researchers new to such collections who are interested in broad themes in Canadian social and material history.

3 Like most textual collections, most picture collections are not organized or indexed with these research interests in mind. This is not really surprising, nor does it mean that these are the only interests that matter. I am not implying that Canadian picture collections are alike. They are enormously varied, not just in size, scope, and budget, but in cataloguing and storage methods and facilities, policies toward researchers' access to originals, making of reproductions, the level of subject indexing, and degree of captions attempted. Another important variable for the researcher is the quality (more significant, sometimes, than the number) of the staff in each institution.

4 If one is engaged in material and social history research, one must abandon thoughts of a trouble- and tedium-free search for a representative sample of the visual document being sought. Such a sample is likely to require an item-by-item search of collections, drawers of index cards, catalogues, inventories, and lists of subject headings. The reason for this is that not only do most picture collections lack systematic or satisfactory subject indexes, but a significant number possess a preponderance of uncatalogued material.

5 Researcher access to uncatalogued material varies by institution. To give some examples: in the Provincial Archives of British Columbia there is (apparently) no access. At the Public Archives of Canada the Guide to Canadian Photographic Archives, now in a new edition, offers brief descriptions, item counts, and outside dates of uncatalogued collections. This can be supplemented by looking up the Division's "guide files," some of which contain more detailed finding aids, but there is no item-level access to uncatalogued photographs. The researcher can request cartons of some of the Division's (roughly) 9 million uncatalogued items, representing 95 per cent of its holdings.4 By contrast, researchers at the Vancouver Public Library's Historical Photographs Division have item-level access by subject to some 40,000 "uncatalogued" photographs. Its index and approach are seemingly unique and inexpensive and have application to uncatalogued collections of a manageable size.

6 In many collections, for uncatalogued as well as catalogued pictures, the researcher must rely on the archivist as a major finding aid. It is then crucial to determine in advance of a field trip the names of those most familiar with the picture collection — it might be a cataloguer as well as an archivist — and to give detailed notice of the research topic and the projected visit. Even so, item-by-item searches of collections, inventories, and catalogues might be unavoidable. Of course, item-by-item searches of collections that number in the tens of thousands (or in the millions) can hardly be expected or recommended.

7 Item-by-item searches may be more productive than would at first appear. This is especially so when catalogue cards are not image-bearing, or when a subject index, in the end, refers one to the bulk of a particular collection. When catalogue cards are not image-bearing, the research time required will be at least doubled, and time will be wasted copying references to pictures discovered to be of no special interest. It is less mind-numbing, in my experience, to flip through often fascinating historical images than to wade through and transcribe such captions and indexes. This all depends on how large, and how accessible and physically easy to flip through, the picture collections are. If one is required to handle glass negatives, a great deal more care and time will be required.

8 Item-by-item searches may be the only recourse for material or social history researchers in certain picture libraries such as that of the Art Gallery of Ontario, the National Gallery of Canada, or the Metropolitan Museum in New York where images are filed by name of artist only. One cannot complain, as this system serves their own research purposes. This is not the case in public research institutions, such as the Picture Division of the Public Archives of Canada, which maintains a reasonably impressive subject index in its public service catalogue.

9 Most public archives and libraries offer some level of subject indexing and content analyses of their picture collections. The sophistication of these offerings varies widely from one institution to another. As an Archivaria article noted:

Because there is no consistency or predictability from one picture archive to another, it is often worthwhile to peruse all the subject headings indexed.

10 Furthermore, within each institution, the nomenclature selected may not be rigorously applied in its index. For example, pictures of domestic edifices might be listed simultaneously under several categories, such as "residence," "house," "farm house," "homestead," "buildings," "architecture," "pioneer life," and soon. (Adequate cross-references cannot be taken for granted.)

11 Where subject entries are retained exclusively in drawers and on cards, reviewing all these terms used is very fatiguing. This is especially so where subject entries are the same size and colour as the references and cross-reference cards in the drawers. Researchers should encourage these archives and libraries to provide bound listings of the subject entries they employ.

12 Such a listing is the National Photography Collection's spiral-bound notes produced in 1983, "Descriptive and Subject Cataloguing at the National Photography Collection, Public Archives of Canada." It is a roster of all headings and "see also" references used, but does not represent an "official" publication or policy.6

13 Other tools for picture archivists devising a subject list include Elizabeth W. Betz, Subject Headings Used in the Library of Congress Print and Photographs Division (preliminary edition, 1980). Many references are also listed in John Maounis, compiler, "Cataloguing Historical Photographs: A Select Bibliography," Picturescope, 28, no. 4 (Winter 1980), p. 21.7

14 Should the New York Public Library's Picture Collection publish the list of subjects it indexes, it would probably be a boon to picture archivists and librarians interested in a comprehensive system. It encompasses an astonishing wealth of detail and cross-references, and a very wide scope of historical data. The predictability of the system and ease of access to the pictures are all a researcher's dream. The heavy demands on it testify to this. It accommodates an average of 150 to 200 researchers a day, and about two-thirds of its collection of 2.5 million images, gleaned from some 20,000 books and about 1,500 periodicals, is out on loan at any one time.

15 Of course, this feature cannot be applied to collections of original material. A small percentage of the New York Public Library's picture collection is non-circulating because the images are rare or expensive. In the circulating portion, a researcher has no way of knowing whether pictures that would be of vital interest are out on loan. In any event, in no other institution does a researcher have free and easy access to such a quantity of material so thoroughly subject-indexed.

16 Another comprehensively indexed collection of historical images is contained at the Bettmann Archive in New York. Unlike the New York Public Library, researchers have access to it by correspondence, and for a research and publication fee. In 1966 Otto Bettmann published the Bettmann Portable Archive, an "invitation to come in and see." It offers a glimpse of the Archive's visual index, which attempts "to extend cross references beyond the obvious, to search out associative values."8 In spite of this welcoming and inspirational tone, a visit to this commercial institution demonstrated that its subject heading list was not readily available as a model, and that the institution has not extended the frontiers of subject indexing since 1966.

17 The subject indexing of visual material is recognized as a difficult and demanding task. Notwithstanding copious entries, there is always a chance that the minutiae of a researcher's special area of interest will be overlooked, evoking consternation and disbelief. This is compounded when the archives or library has obviously spent most of its time indexing details of no real consequence to the researcher or anyone he or she knows.

18 Whether at all justified, this is how I felt about the special attention now being paid to the indexing of photographs by the name of the photographer or studio in archives with no clear mandate to collect visual material relating to the fine and decorative arts. Presumably this development has been influenced by the organization of collections in fine art galleries and museums, and by the International Museum of Photography at George Eastman House in Rochester, New York. This concern is relatively new. As the former chief curator of the National Photography Collection wrote:

19 Across the country, this archival interest in the oeuvre of Canadian photographers and the technical development of photography seems quite pervasive. There is no harm in this situation unless the content of historical photographs comes to be considered, in acquisition and indexing policies of archives and libraries, to be entirely secondary to the technical expertise and artistic vision exhibited by the photographer in question. I recognize that there can be compensations, such as in the happy circumstance where a particular photographer made a point of recording the type of information being sought. Thus, I was pleased to find some indexed listings and concerted acquisitions of the work of a few persons who were passionate about photographing domestic life in Canada.

20 The indexing of such proper names is, of course, standard library and archives practice, and the simplest kind of indexing to do. But researchers should be aware of the possibility that pictures may be indexed under the proper name, rather than the subject, even where a reasonably substantial subject index exists. Thus, for example, even though the subject category "interiors — domestic" appears in the index, a view of one might be entered only under its geographical location or the name of its owners.

21 Similarly, captions provided for historical pictures often focus only on biographical and geographical details. This is especially frustrating when captions and images are separated. In my research, information on the picture's setting and content is crucial. A caption might read, for example, "Mr. and Mrs. J.S. Robertson, Peterborough, Ont., 1885." Are they inside or outside their home, are they in a photographer's studio, or elsewhere? Is it a head-and-shoulders portrait, or are they engaged in any activity or posed with or near any household objects? Is there anything special about the way they are dressed? Can any information be included on their ages, occupations, or status as settlers?

22 Certainly, providing such information is a terrible chore for archivists and librarians, most of whom face huge backlogs of uncatalogued material, some of which has been unceremoniously dumped on them. But in the long run such scrupulous attention will increase the usefulness and prestige of picture collections. They will be more widely understood to contain historical documents which do not rank below any other kind. Such documents are not mere illustrations to be looked up after all other research is done.

23 Among the picture collections visited, the Glenbow-Alberta Institute best exemplifies this conscientious and scrupulous archival treatment. It has produced the most rigorous and reliable subject index and system of cross-references. This painstaking task of cataloguing and indexing visual images has been accomplished in a manner wholly sympathetic to actual research concerns, and without a large staff or budget. The captions provided, nearly always seen with the image, are extensive. In many cases they go beyond the call of duty in drawing attention to elements in a picture that might otherwise be overlooked.

24 There is easy physical access to the Glenbow Collection numbering well over 300,000 items, about 70 per cent of which have been individually indexed. The tile numbers assigned to each picture provide information to the researcher and archivist. These codes serve to designate the dimensions of the negative available, and can be used further to discern the name and size of the collection from which the item derives (fig. 1).

25 The Glenbow-Alberta Institute provides photocopies (including photo captions) at very low cost and at short order, and the staff is even receptive to research inquiries by telephone or letter.

26 Still, patrons should not expect picture collection staff to do all the research on their behalf. Though an archivist, librarian, or cataloguer can either provide a map or a compass or can nonchalantly watch researchers lose their way in an uncharted morass, research success in picture collections will still be commensurate with one's own effort.

27 With this in mind, the following is a series of questions that might productively be asked, by telephone or letter, before embarking on expensive research trips to picture collections.

1. Does the picture collection observe the same hours as the rest of the archives or library? Who is responsible for the pictures or knows them best? Is the person there five days a week? Will he or she be on vacation during your visit? (Forgetting to ask these obvious questions may lead to immense research frustration.)
2. How are the collections organized? What is the physical arrangement with which you will be dealing and in what format is catalogued material? Is there access to originals and is special permission required to obtain such access?
3. What percentage of holdings are uncatalogued? If accessible, how long will it take to retrieve uncatalogued material? Might it be worth retrieving, or is it in a disorganized state?
4. What level of subject indexing or finding aids has been attempted? Are captions and pictures separated or are the caption cards image-bearing?
5. Will captions be provided with reproductions you order? What are other policies and costs regarding reproductions? Can photocopies be provided? What is the minimum length of time required to process the order?

28 It may still be difficult to determine how much time is adequate for the task at hand for those engaged in picture research on broad themes and in material history. It may be as well to allow more than seems necessary or is advised by an archivist or librarian, given the quantity of uncatalogued material, and the state of subject indexes, cross-references, and captions in so many picture collections.

29 This situation might be improved if patrons of picture collections expressed clearly (and in advance) their research interests to the archivist or librarian, and later offered a few constructive suggestions on indexing, captions, and other matters which seem to be of general concern. Thereby, they would be helping one another in a scholarly community and contributing to the increasing importance of pictures in the historical record.