BACKGROUND, RESOURCES, AND COMPARATIVE APPROACHES
The vast collection of slave narratives, housed in the Library of Congress, was compiled from interviews conducted between 1936-1938 by Federal Writers' Project field workers. The typescript consists of over 10,000 pages from more than 2,000 interviews with former slaves throughout the United States. Nearly three percent of these interviews were conducted in Florida (Mormino 1988:405). George P. Rawick has published the entire collection in an initial series of eighteen volumes and two supplement series. Seventy-two interviews are published in Florida's volume (1972a). Donald M. Jacobs has indexed the entire collection to make it possible for scholars to conduct research on particular issues or geographic regions (1981), and Howard Potts has published a comprehensive name index (1997). Further audio-visual and online resources have made the narratives still more accessible in recent years for classroom and scholarly use. Ira Berlin, Steven F. Miller, and Marc Favreau have published an anthology of the narratives together with an audio-cassette that allows students to hear excerpts of the original recordings (1998).
An online version of Rawick's compilation can be searched by key word functions. In addition, the site offers discussion forums and other resources, including private "classrooms" with discussion boards and posBible lesson plans that help teachers and students navigate the collection more effectively (http://www.slavenarratives.com). Sites maintained by the Library of Congress (http://lcweb.loc.gov/exhibits/african/intro.html) and the University of Virginia (http://xroads.virginia.edu/~HYPER/wpa/ wpahome.html) also offer background information, photographs taken at the time of the interviews, as well as text and sound files of selected narratives.
Though the narratives potentially have tremendous value in teaching and research, they also pose a number of challenges that must be taken into account before the collections may be effectively used. For example, Rawick explains that certain "systematic biases" exist in the narratives, which in many cases may give a slanted view of the treatment of slaves. On "matters concerning ... the sexual exploitation of women, whipping and punishment," the surviving versions of the interviews may be "heavily biased in the direction of grossly exaggerating the humaneness of the institution" (1997:xxxii-xxxiii).
The slave narratives were often censored at the state level before they were sent to Washington, and controversial topics may have been avoided during the interviews or omitted in transcription.' The former slaves may also have felt inhibited discussing race relations because the majority of the interviewers were white while the former slaves were "blacks, almost invariably very poor and totally destitute, and often dependent upon public charity and assistance from white-dominated charities and public officials" (Rawick 1997:xxxii). Jacobs explains further that the economic climate of the 1930s when the interviews were conducted contributed to an almost nostalgic attitude towards slavery on the part of many former slaves who could "look back on their youth under slavery as a time when they at least managed to have something to eat" (quoted in Rawick 1997:xxxii).
All of these complications are compounded by discrepancies among the interviewers in the transcriptions of dialect, which the interviewers were encouraged to preserve in their notes despite most field-workers' lack of linguistic training and, in many cases, lack of tape recorders. In the introduction to the supplementary volumes of his collection, Rawick cautions that "the slave narratives do not generally provide a reliable source for those seeking to study black speech patterns and black English" (1975:xxix). The vastly different methods interviewers employed in recording direct speech can be misleading because …