INTRODUCTION
Mankind's first attempt to communicate beyond his life was with a picture - not a word. The use of pictograms was common practice eons before written language. "In short, the first statements to survive the sound of a voice were pictures, not words" (Taylor 1979, 418). It is not surprising that our oldest surviving human records are pictures: the cave art Lascaux, France and Altamira, Spain. These depictions of horses, bulls, and other animals precede the invention of the first alphabet, cuneiform, by at least ten thousand years (Encarta 1996), and speak volumes about their creators.
No other kind of relic or text from the past can offer such a direct testimony about the world which surrounded other people at other times. In this respect, images are more precise and richer than literature. (Berger, 1976, 10)The synthesis of written language and pictorial representation provides the most complete understanding of an age or an idea. Pictures convey information in different ways than text, and they convey an entirely different type of information (Layne, 1994). The combination of texts and images appeals to "all of the human senses through both sides of the brain" (Taylor, 1979, 419). The synthesis of these two modes of human communication yields an understanding that is greater than the sum of both. "Only when photographic records are used in conjunction with textual documents is it possible to understand the entire story of an event or scene" (Tolppanen, 1996, 3).
The archival value of image collections is separate but equal to manuscripts, journals, letters and other primary source material, which are the traditional records preserved in archives. Essentially, archival materials must, "contain evidential and informational value; and that these materials are kept because of their enduring research value" (Greenberg, 1993, 86). Images are valuable tools because they contain unique information that is not accessible in any other form (Tolppanen, 1996, 3). Historically, our library classification systems have concentrated on textual materials, while regarding visual materials as "works of art", most likely to be classified by art historians for their aesthetic, interpretive, or symbolic attributes. Consequently, systems of classification for visual materials as archival records have developed slowly. This attitude has changed radically over the last decade, as librarians, archivists, and information scientists began to focus more on providing intellectual and physical access to visual documents as archival records (Hahn, 1994, 3). The various tools for establishing bibliographic control of images will be the focus of this paper.
Over the past twenty-five years, the area of documentation of holdings has been identified as a critical area of concern for archives and historical societies like the Ohio Historical Society (OHS). Quite simply, this documentation should provide intellectual and physical access to any item in the collection. It should also be able to "shed further light on any object as required" (Reibel 1978, 24). Unfortunately, very few institutions can provide this type of access to even a fraction of their collection. This lack of available documentation, "is a handicap to adequate care and maintenance of these collections and to scholarly progress in general" (American Association for Museums, 1984, 53).
Similarly, access to the OHS photograph collections is less than adequate, and as a result, several schemes are being considered to greatly improve access for staff and public. Traditionally, bibliographical control is provided in the form of indexes, which serve as pointers to desired individual items or groupings of items. The OHS photograph collections, totaling more than one million images, are categorized in four separately-located groups. Each group is represented by its own discrete index, none of which contain adequate cross-relational information to the other indexes. These four groups can be summarized in the following manner:
The major purpose of this study was to investigate the most appropriate method of providing intellectual and physical access to the composite Civil War image holdings of the Ohio Historical Society. All four of the above-mentioned collections contain Civil War photographs, which may be regarded as a sub-set of the entire photograph collection. The non inter-related nature of the four collection indexes presents an access problem in need of a single source collational index. The index type and construction was considered from the standpoint of both the OHS staff and researchers from the general public. The research skills and collection use of these two user groups was investigated in order to recommend the most effective instrument of bibliographical control in the areas of index type, bibliographical record construction, index terms, and controlled vocabulary.
Because the special photograph collection being analyzed is more likely to be found in a large state historical society similar to the Ohio Historical Society, articles, model indexes, and model bibliographic records were selected with this type of large collection in mind. The reference requests that were charted are from the logs of the Ohio Historical Society, and as such will not necessarily be applicable to all photographic collections; for example, those typically found in public libraries, smaller archives, corporate libraries, or other special libraries.