Introduction
Many art libraries are now making the diversification of their collections a priority, and in particular aiming to improve their holdings of material relevant to groups facing marginalisation either in the art world or in wider society. It is clearly good news that the number of resources available to buy in pursuit of this aim is large and growing. However budgets are finite, too often leaving the acquisitions librarian casting a wistful eye over reviews of titles that are out-of-reach. The problem is exacerbated by the threefold nature of the diversification we are likely to be interested in, as we seek to add resources about diverse artists, by diverse authors, or which embody diverse perspectives.
One obvious and indeed necessary response is to put forward the case for a larger acquisitions budget, perhaps including funds ringfenced for diversification. However even with a smaller-than-optimal budget, the prospects for diversification are by no means completely limited by financial constraints. Acquisition of paid-for material can be complemented by other significant initiatives. The curation of collections of relevant open-access resources is one we are implementing at Sotheby's Institute of Art; described here is another of our current projects: making diverse content in existing physical resources more visible. While we are interested in diversification as it relates to all under-served groups, this paper, simply to bring some focus to the discussion, concerns only resources which are about women artists. In our experience, similar, though perhaps never identical approaches can be taken when looking beyond gender to other dimensions of human experience, such as class, ethnic or cultural background, or sexuality.
Improving catalogue records
All libraries of physical books, almost by definition, contain content which is not immediately visible in either the library catalogue or, if employed, the discovery tool. This will presumably always be the case until that hypothetical time, which may or may not ever be reached, when our systems have access to the full text of every printed publication in the library. Working to improve discoverability of this content is always worth considering, and never more so than when diversification has been identified as a priority. There are at least two possible approaches: augmenting catalogue records; or creating separate lists or databases of relevant titles in key areas.
In our library, the focus has been on improving catalogue records. Much of the work has involved adding tables of contents or – and this is not necessarily the same thing – lists of artists whose work is reproduced or discussed in the book in question. At least three major types of content are involved: chapters or essays on an artist or theme; entries for artists, for instance in exhibition catalogues; and discussions of artists which do not comprise discrete parts of a publication but are embedded in more wide-ranging chapters or articles. Each of the three types needs considering separately.
Adding a list of chapters or essays is straightforward, even if sometimes time-consuming. It is an immediate help to searchers who know the title of an essay they wish to read, but not the publication(s) in which it can be found. When the title of the chapter or essay is reasonably descriptive, it also helps those conducting a free-text search for the topic in question. Adding subject headings may also be helpful, particularly in the case of chapter titles which are less informative, though this may require the cataloguer to spend longer examining the book than is sensible given the other calls on our time. There are also the potential disadvantages of large numbers of subject headings to consider. We follow two of the Library of Congress's guidelines: only allocating subject headings to topics which constitute at least 20% of the resource being catalogued; and adding no more than six to ten headings per resource.Footnote 1 By providing only a short list of key headings, we aim to enable readers to assess quickly the principal focus of a publication. Furthermore, readers choosing a book because of a subject heading we have allocated are likely to find themselves with a title containing a reasonably extensive treatment of the topic they are researching. The downside of this is that artists, topics and viewpoints which generally only receive brief mentions in the literature remain largely invisible in the subject headings. The wisdom of adding more or fewer subject headings is a question which surfaces regularly within the cataloguing community; does the adoption of the conscious practice of addressing historic imbalances in our collections imply libraries need to re-think policies for subject headings? It is perhaps too soon to say.
Adding tables of contents is less crucial for many libraries than it was a couple of decades ago, for discovery tools such as Summon often have access to metadata about our books beyond what they retrieve from our catalogue records; they often have access to a list of chapters, and sometimes more. However this is not always true, particularly for exhibition catalogues from smaller publishers and for older titles. For example, the chapter devoted to the Danish modernist Anna Ancher in the 1998 exhibition catalogue Krøyer and the artist's colony at Skagen did not appear in a search for books about her in our instance of Summon until we added a contents list to the record in our library catalogue.Footnote 2 Even when a library patron looking for a particular essay does see the relevant book in the discovery tool's results, there may be a benefit to adding to the catalogue record: they may be initially unsure if the book contains, or merely cites, the essay in question.
When adding a list of artists to a catalogue record, there are three principal options for libraries which use MARC21: including them as subject headings, in a contents note, or in a general 500 note. All are equally serviceable for the patron conducting a free-text search. For the patron looking at a bibliographic record to assess how useful a title is likely to be, there are however certain differences, which in our library inform our choices about when to use each option. The advantages and disadvantages of increasing the number of subject headings allocated to a work have been already mentioned. A contents note works well when each featured artist has a dedicated section in the book, and when in addition either the number of artists is modest, or the book has little in the way of other kinds of content, such as thematic essays. By contrast, in the case of a publication such as a sizeable exhibition catalogue that includes multiple essays and also a large number of artist entries, it may make more sense to separate out the artist entries into a 500 note. This way, it is easy for searchers to scan the list of essays, or browse the list of artists, depending on their interests.
Adding lists of artists to a library catalogue record increases the number of results that appear in the discovery tool. For us, a Summon search for the Colombian pop artist Beatriz González did not retrieve the exhibition catalogue Pop América: 1965-1975 until we added a list of exhibited artists to the record.Footnote 3 As in the case of chapters and articles, there may be a benefit in adding a list of artists to the catalogue record even when the discovery tool does identify a book as containing content about them. A Summon search of our holdings for Beatriz González always found the catalogues The world goes pop and Radical women: Latin American art 1960-1985, but there was no way of knowing from the results page if these titles included passing mentions or more substantial material.Footnote 4, Footnote 5 Once we had added a list of artists to the record for each catalogue, it was relatively easy for readers to gauge the likely amount of content after clicking through to the catalogue record. The world goes pop, for instance has 266 pages but a relatively long list of artists. From this it should be clear to any reasonably observant searcher that the book is likely to contain a useful summary of each artist's life and work, but probably not an in-depth treatment.
Adding lists of artists discussed in other kinds of publication, such as book-length accounts of particular movements, is more difficult. It is often time-consuming and also has the potential to mislead, in that the cataloguer will often need to scan the book's index and make a quick and partly subjective judgement about which artists are treated at sufficient length to justify inclusion in a note. We have therefore so far rarely pursued this option. Occasionally the task seems worthwhile, for instance when a book discusses a small number of artists. In the case of David McCarthy's Pop art, from the Tate's Movements in modern art series, it was easy to create a note listing the 26 artists discussed, including some, such as the British artist Pauline Boty, for whom the literature is comparatively limited.Footnote 6 By contrast, this approach encounters serious problems when faced with a title such as Whitney Chadwick's Women, art, and society, which discusses a very large number of artists, generally briefly, over the course of nearly 600 pages.Footnote 7
Better catalogue records, or separate indexes?
A rather different approach, focussing on creating separate indexes of artists rather than adding to catalogue records, is being energetically pursued by the Thomas J. Watson Library at the Metropolitan Museum of Art In New York.Footnote 8, Footnote 9, Footnote 10 The Library has created four databases of titles in its collection about artists from particular backgrounds, beginning with the Index of African American artists. A full comparison of the merits and demerits of these two approaches is beyond the scope of this present discussion. For now, some general remarks can be made. Firstly, a library does not have to commit to one approach to the exclusion of the other. The Thomas J. Watson Library has complemented its indexes by upgrading individual records, while at Sotheby's Institute of Art we have also published lists of holdings in particular areas, such as MA dissertations by our past students about women artists.
Secondly, the two approaches are often routes to the same end. One of the major aims of the project at the Met has been to allow searchers interested in, for instance, African American art to find artists they have not heard of. This is particularly useful for searchers interested in aspects of artists’ identities which are not necessarily apparent from their names – such as being African American. In our own library, we are addressing this by adding subject headings such as ‘African American art’ or ‘Women artists – United States’ to relevant titles. In doing this we are ignoring one well-established guideline for subject headings: classifying as specifically as possible. Given the size and scope of our collection, this seems to us a reasonable step which is unlikely to mislead many of our patrons; the same would not apply to all libraries.
Thirdly, technological change may mean the two approaches are not as distinct as they might appear, especially from the patron's point of view. For example, the Cleveland Institute of Art created its Contemporary artists index in order to bring to light material about individual artists that was lying undiscovered in more general publications.Footnote 11 The Index, which was not conceived as a diversification initiative but did grow out of a concern with discoverability, began life as a separate database, but its content has since been incorporated into the library catalogue.
Discussion
The work described here makes no call on acquisitions budgets but does take time. Fortunately, it does not need to proceed at any particular pace; whatever work can be done is likely to be of value. Nevertheless, once embarked on, the logic of this approach does ask for a continuing time commitment. Having established higher standards for the description of resources, there are clear advantages to consistency in maintaining those standards. Furthermore, time will be needed for several related tasks. As argued above, the work of ethical, diversity-conscious collection development is too new for the detail of how best to go about it to be yet clear. Therefore time needs to be spent reviewing our experience and the experience of our patrons. Should we, for instance, be focussing on improving catalogue records, or creating separate indexes? Are changes to our policies with regard to subject headings called for? We will also do well to allot time to keeping abreast of the ever-evolving conversation about diversity in general: including, amongst so much else, the question of whether ‘diversity’ is even an appropriate or adequate description of the work we want to do.
The approach outlined in this article can only be one strand of wider efforts to make a library both more comprehensive in the content it offers and more relevant to the needs and interests of as wide a range of patrons as possible. It is worth repeating that making more of the content in our existing resources cannot be a substitute for new acquisitions. To return to the example of the Thomas J. Watson Library, it has complemented its compilation of indexes with a substantial programme of acquisitions in the same areas. Libraries need to make the case for budgets adequate to the task of diversification. Equally, even the largest budget does not negate the value of making best use of resources already in the collection. Furthermore, both acquisitions and enhanced description need to sit alongside other kinds of diversity work such as the creation of study guides; the review of misleading or offensive terminology; and the provision of individual support to patrons. There is much to do. At times the size and complexity of all this may seem overwhelming, aspects of it difficult, troubling or burdensome. But let us also find joy in realising how many ways are open to us for better serving our current and potential patrons.
