• Project Timeline

• Introduction to the LOC collection (PDF)

• 1941 introduction to the LOC collection (PDF)

• Introduction to the Mississippi Narratives

• Instructions to the interviewers (PDF)

• Sample narratives

• Introduction to the Index

• Acknowledgements

• Foreword by Dr. Charles Joyner

Mississippi

by Ken Lawrence

Search, Discovery, and Editing

Even as the first volumes of The American Slave appeared, George Rawick began to search for more unpublished slave narratives and to urge others to join the search. During a trip to Mississippi in the spring of 1973, he challenged audiences at Tougaloo College and several other campuses to help locate more Mississippi narratives. He pointed out that the collection in the Rare Book Room of the Library of Congress contained only twenty-six of them, 174 pages all told (contained in volume 7 of The American Slave), and that the WPA must have collected a great deal more in a state whose people had once mostly been slaves.

Scholars elsewhere had the same idea, and previously unpublished WPA narratives were found in Virginia, Georgia, Louisiana, Texas, Oklahoma, Ohio, Colorado, Minnesota, and possibly other states. Some will be published in this series; others will appear under different imprints. The finest body of new material that has been published to date is the Virginia collection edited by Charles L. Perdue, Jr., Thomas E. Barden, and Robert K. Phillips, published by the University Press of Virginia under the title Weevils in the Wheat.

Other sources, too, are being mined for records left by former slaves, especially newspapers and magazines that published their recollections in the late nineteenth and early twentieth centuries; many of these will be collected and reprinted. Post-emancipation records of ex-slaves, including those of the Freedmen's Bureau, Black newspapers, and state and federal legislatures from the Reconstruction era, are finally attracting long-overdue treatment by historians.

For the time being, it may be that nonblack historians will have more use for these materials than black historians will, for, as one black woman commented after Rawick's lecture at Millsaps College, “I learned all that from my grandmother.” But as the experience of slavery continues to recede, it is likely that more and more black scholars will turn to these materials; perhaps as that happens the methodologies for using them will be revolutionized. In the meantime, we hope that by publishing these additional materials now we will not only draw attention to previously unused materials but will also help broaden and deepen the understanding of the entire body of material.

George Rawick's challenge prompted an immediate quest for the missing Mississippi narratives. At first, the research staff at the Mississippi Department of Archives and History (the main depository for Mississippi's WPA materials) was unable to locate any. But an article in the Winter 1968 issue of Mississippi Folklore Register, “The WPA-Collected Archives in Jackson” by George W. Boswell, informed us that box number 83 contained “Reminiscences of former slaves.”

As it turned out, this was a lucky report. Boswell said there were “twelve boxes filled with folders of typed folklore manuscripts.” Actually there were 232 boxes, several file cabinets, and dozens of microfilms in the WPA collections, with folk materials, including narratives of former slaves, scattered throughout. Though the numbering code had been changed since Boswell examined the collection, the archivists were able to locate a box labeled “slave stories.” Eureka! The box contained several stenographer's notebooks filled with shorthand, a couple of WPA books and pamphlets, and eight folders containing 459 pages of slave narratives.

But what a mess! Many of the narratives were written in almost illegible handwriting; others were faded carbons on yellow paper that crumbled at a touch, fastened with rusting staples and paper clips. Much of what was originally written was rendered unintelligible by pencilled-over editing. Often there were three or four different editing stages of the same narrative; but just as often a highly edited narrative appeared alone, without the various stages of its evolution. A few were clipped from old newspapers, pasted to sheets of yellow paper and sent in to WPA editors.

Jan Hillegas catalogued the findings and discovered that the box contained ten narratives identical or similar to the ones in the Library of Congress collection and seventy-three new ones. Since more than half of the known narratives were still unaccounted, she reasoned that there must be more. (As it turned out, the box we were looking at wasn't even the same one that Boswell had noted in 1968.) Two and a half years later, after plodding through most of the 232 boxes; microfilms of WPA “Source Material for Mississippi History” for eighty of the state's eighty-two counties; and twenty-two bound manuscript county histories, she believes she has uncovered all of the ex-slave documents in this collection (as well as a few from other sources accumulated incidentally along the way). During the same period, she has also retyped every page because of the poor quality of almost every original. This, in turn, placed the largest editing burden on her shoulders.

This collection presented a staggering number of editorial choices. Under the circumstances, we are bound to have made mistakes and to have failed to anticipate some scholarly needs. Our general approach when in doubt was to err in the direction of inclusiveness. With that guiding principle, we included every item we found in the WPA collection which provided any information about a black person's experience as a slave or during Reconstruction. We left out roughly sixty more items (about 100 pages total) that reveal only the name of an ex-slave and owner and sometimes the ex-slave's birthplace. This gave an elementary definition of “narrative” consistent with the primary purpose of the collection as seen by its overall director.

Federal Writers’ Project Folklore Editor Benjamin A. Botkin wrote:

The chief value of the slave narratives is that of source material for the social historian and writer. From the point of view of the historian, the narratives are not so much documents as evidence—the only pure or unconscious evidence we have of the actual conditions and circumstances of slavery. From the lips of former slaves has been recorded (sometimes crudely, sometimes skillfully, but nearly always significantly) a folk history of slavery.

Botkin thought that “we have here the Negro's own story and folklore of slavery and not a white version and myth of slavery.” Though his enthusiasm exaggerated the “purity” of the narratives, our editing attempts to fit his basic framework.

In practice this means, for example, that a WPA “narrative” was included if it contained anything more than just the name of an ex-slave, his or her place of birth, and the approximate year he or she came to Mississippi. We will undoubtedly be criticized by some historians for being so inclusive, even though few of the documents are that skimpy. Genealogists, on the other hand, will probably deplore the omissions, no matter how slight.

Very few items strained our minimum criteria. Our real problems were created by the fact that so many different versions of a large number of narratives have survived. If we had included each variant, we would have doubled the size of the publication (and its cost) without materially improving it as a collection of source materials. In cases where there was no difference in content, we typically chose the earliest available version of a narrative. On the other hand, if two different versions each contained unique material, we used both, including the redundant parts as well; in these cases, we normally accepted whatever editing had been imposed along the way.

If a statement was clearly added by an editor or rewriter other than the original interviewer which seriously altered the ex- slave's account, we took it out. But if we could not be certain of the source of a doubtful addition, we left it in. This will result in a very slight tendency to include some of the distortion we sought most to remove, but we felt that all doubt should be resolved by readers, so we are publishing the material. In no case do doubtful statements we have included differ essentially from similar testimony of other narrators that is undeniably genuine. In any event, there are probably fewer than a dozen close calls in about 2,400 pages of text, not enough to impose much of a bias.

We included the WPA heading codes even though we don't understand all of them. (“241” means ex- slave interview; we don't know the meanings of “FC,” “FEC,” or “SWC,” although they probably refer to stages of editing and rewriting.) We corrected obvious typographical errors, but whenever an odd spelling or other error conceivably might have been intentional we left it alone. Because many of the most complete versions of narratives also suffered from the most editing, it is important to stress once again that these materials cannot be used for studies of language, grammar, or dialect, and it is likely that almost every version of any narrative is counterfeit in those regards.

We found clippings about ex-slaves in the collection and included them if they met our criteria for narratives. The only aberration they add is that some were collected earlier than the interviews, a fact which can only enhance their usefulness. Having decided to include those, we also included three others that we found elsewhere—those of George Washington Albright, James Hill, and Hiram Rhoades Revels.

Despite our diligent efforts to locate and include as many Mississippi ex-slave narratives as possible, there are certain to be others that haven't yet turned up. The only part of the state that appears to be suspiciously underrepresented in this collection is the Greenville-Greenwood area of the Delta. Even if there are more to be unearthed and published, the essential task has been completed and new priorities—distillation, interpretation, analysis, and popularization—are now on the agenda.

The potential for error when working with the WPA slave narratives is enormous, but there is no need to apologize for it. Our failings of whatever nature are trivial compared to those of prior generations and can be rectified as this material becomes generally available.

The Federal Writers in Mississippi -->