«Volume Title: Strategic Factors in Nineteenth Century American Economic History: A Volume to Honor Robert W. Fogel Volume Author/Editor: Claudia ...»
This PDF is a selection from an out-of-print volume from the National Bureau
of Economic Research
Volume Title: Strategic Factors in Nineteenth Century American Economic
History: A Volume to Honor Robert W. Fogel
Volume Author/Editor: Claudia Goldin and Hugh Rockoff, editors
Volume Publisher: University of Chicago Press
Volume ISBN: 0-226-30112-5
Volume URL: http://www.nber.org/books/gold92-1
Conference Date: March 1-3, 1991
Publication Date: January 1992
Chapter Title: The Slave Family: A View from the Slave Narratives Chapter Author: Stephen Crawford Chapter URL: http://www.nber.org/chapters/c6967 Chapter pages in book: (p. 331 - 350) 11 The Slave Family A View from the Slave Narratives Stephen Crawford The slave narrative collections of the Work Projects Administration and Fisk University contain over 2,200 interviews with aged ex-slaves taken primarily in the late 1930s. Although the interviews were largely unstructured and vary greatly in quality, they have rightly come to be accepted as important sources for the study of slavery. I There has been a growing realization during the past fifteen years that the narratives can provide quantitative evidence for research on slave historiography.* These quantitative uses must take into account the considerable biases inherent in a source that consists entirely of the memories of aged ex-slaves. But, if these biases are reported and corrected for, the resulting quantitative measures are extremely useful summaries of this vast body of information.
One of the most important issues that can be quantitatively studied with the narrative source is slave family structure. The vast majority of the ex-slaves talked about their family experience under slavery, providing a snapshot of slave family structure in the immediate antebellum period. By examining the data and cross-tabulating family type with relevant variables, such as plantation size and location, it is possible to observe how slave family structure was altered by both slaves and masters.
The nature of the ex-slave interviews makes the narrative sample a unique data source. Because few of the ex-slaves reached adulthood before emancipation, the sample consists of children reared under various family types.
More important, each ex-slave did not provide information for just a single
1. See Eugene D. Genovese, Roll, Jordan, Roll: The World the Slaves Made (New York, 1974).
Herbert G. Gutman, The Black Family in Slavery and Freedom (New York, 1976), and George P.
Rawick, From Sundown to Sunup: TheMaking of the Black Community (Westport, 1972).
2. Paul D. Escott, Slavery Remembered: A Record of Twentieth Century Slave Narratives (Chapel Hill, 1979).
332 Stephen Crawford cross section but, rather, a chronological history of his or her family. To illustrate, consider the case of a slave child born into a family containing both parents but whose father was subsequently sold away. The informant was then sold from his mother and on the eve of emancipation was living without family, alone in the slave quarters. To quantify this information requires two family distributions. The first is given here by the family at the time the informant was taken away from the family (or at the time of emancipation, if the slave were not separated from the family of origin). It is a history of the marriage, in all but legal terms, of the slave’s parents. The example given would be coded as a female-headed family created through sale of the father. The second distribution focuses on the informant at the time of emancipation. The example would be coded as a child living alone in the quarters without family owing to sale or transfer.
I term the first distribution the “Family of Origin.” It summarizes all the information in the narratives about the marriages of the informant’s parents.
The second distribution I term the “Slave Household Type.” This distribution is a snapshot of slave households that contained children just prior to emancipation.
Developing the two distributions requires accurate coding of often sketchy qualitative information. The narratives did not come from structured question-and-answer sessions designed to generate easily quantifiable data.
Interviewers were encouraged to cover important topics, such as family life under slavery, but, more often than not, the aged informants simply reminisced about their experience as slaves. Thus, the family-type categories have been created out of numerous individual coding decisions, the possible biases of which will be discussed.
Table 11.1 presents the Family of Origin distribution.
This distribution focuses on slave marriages by categorizing the ex-slave’s family type either when he or she was separated from the family of origin or at the time of emancipation. Roughly two-thirds of the ex-slaves grew up in families defined as “two-parent consolidated,” meaning the family lived together on the same plantation, or “two-parent divided residence,” meaning the father lived on a different plantation from his wife and ~ h i l d r e n. ~ remaining third were The raised for at least part of their childhood in a single-parent family, almost exclusively female headed. To understand this distribution requires an examination of the individual family types.
Half of the ex-slaves who provided information belonged to two-parent consolidated families. These families were easy to categorize because exslaves often reminisced about mothers and fathers and their relationships during slavery. In general, these were enduring relationships that began under slavery and extended into the post-emancipation period. Less than 2 percent
3. I will use words like “wife,” “husband,” and “marriage” throughout to describe relationships in the terms the ex-slaves did.
333 The Slave Family
Noret Family of Origin is given by the structure at the time the slave was sold from the family or at emancipation.
of these two-parent consolidated families were voluntarily broken when slavery ended.4 A surprising feature of the two-parent consolidated households is the virtual absence of stepparents, especially stepfathers. Only 2 percent of the category identified the fathers as a step, as opposed to a biological, parent. Apparently, once a slave marriage was broken by death or sale it was rarely reformed, at least not in the eyes of the slave child. Thus one explanation for the high percentage of female-headed families is the lack of remarriage, rather than an absence of marriage altogether. For some reason, be it the availability of potential new spouses or reluctance on the part of the slaves, few women re-established two-parent households through remarriage. * The second category in Table 11.1 comprises families defined as twoparent, divided residence. Such families are inferred when the slave said that his or her father lived on a nearby plantation but visited often enough to maintain the family bond. More often than not, the father spent Saturday afternoon through Sunday night with his family. In a minority of cases, a weekday visit was allowed. A small minority of less fortunate fathers were allowed only irregular, infrequent, or merely seasonal visits.
The separation of divided-residence families undoubtedly placed strains on the marriage and the family. For instance, Jane Sutton’s parents lived on farms that were geographically near, but the father’s absence seriously endangered the relationship between the father and his children.
My pappy’s name was Steve Hutchins. He b’lon to de Hutchins what live down near Silver Creek. He jus’ come on Satu’day night an’ us don’ see
4. The low level of voluntary disruption suggests that slave marriages were largely by choice.
If slaveowners forced slaves together, I would expect higher levels of separation when there was no longer a slaveowner to require that parents stay together.
5. An alternative explanation is that the reforming o families by stepparentage was so easy that f the new parent was completely accepted. This seems unlikely given both the explicit mention of stepparents by ex-slaves and the lack of any discussion of the loss of parents by death or sale among those in two-parent families. Even if the new parent were completely accepted, I would expect that there would be some discussion of the loss of the natural parent.
334 Stephen Crawford much of ’im. Us call him ’dat man’. Mammy tol’ us to be more ’spectful to ’im ’cause he was us daddy, but us aint care nothin’ ’bout ’im. He aint never bring us no candy or nothin’.6 More often, however, the ex-slaves’ stories demonstrate the strength of the family bond. Charly Davis’s parents lived on adjoining plantations and Charly’s father avoided patrollers during extra visits with his wife and family.
My mammy and pappy got married after freedom, ’cause they didn’t git de time for a weddin’ befo’. They called deirselves man and wife a long time befo’ they was really married, and dat is de reason dat 1’s as old as I is now.
I reackon they was right, in de fust place ’cause they never did want nobody else ’cept each other, no how.’ That fathers and husbands clandestinely, and under threat of punishment, made extra family visits demonstrates the cohesiveness of these families.
Samuel Boulvware remembered that his father came to see the family even though he faced a whipping if caught.
My daddy was a slave on Reuban Bouwave’s plantation, ’bout two miles from Marster Hunter’s place. He would git a pass to come to see mammy once every week. If he come more than dat he would have to skeedaddle through de woods and fields from de patrollers. If they ketched him widout a pass, he was sho’ in for a skin cracklin’ whippin. He knowed all dat but he would slip to see mammy anyhow, whippin’ or not.* The most important quantitative measure of the strength of dividedresidence families is the extent to which they voluntarily reunited after freedom. Information on the post-emancipation history is available for half of the families. In 80 percent of this subsample, the family reunited. The importance of the father’s role in these divided families is revealed by where the family reunited after emancipation. The sample is very small, but, in the twelve cases with information, eleven reported the family reuniting on the father’s plantation and, in the other, on neither the mother’s nor the father’s plantation.
Given all the indications of cohesiveness in the two-parent dividedresidence families, it seems correct to group these families together with the two-parent consolidated families to obtain a measure of families with strong, unbroken bonds between the parents. Fully 62 percent of the ex-slaves were raised by parents who had an unbroken marriage. What of the remaining exslaves?
The single-parent-family category is more difficult to code than either of the two-parent categories. When the family was never formed because the father was unknown or when it was broken due to parental death or sale,
6. Rawick, vol. 7, Mississippi Narratives, p. 151. Note that all references are cited by abbreviated reference to George P. Rawick, editor, The American Slave: A Composite Autobiography, 19 vols. (Westport, 1972).
7. Rawick, vol. 2 ( l ), South CarulinaNarratives, p. 252.
8. Ibid., p. 68.
The Slave Family coding was straightforward. It was also straightforward when the ex-slave talked about the father not being present. In roughly 20 percent of the cases, however, the father was mentioned as having a different owner from the rest of the family, but the father did not visit. Although it might be argued that some of these families were really two-parent divided residence, I have chosen to code them as single parent. These families clearly exhibit a weaker or possibly nonexistent bond between the mother and children and the absent father. To include them in the divided-residence category could call into question the aggregation of two-parent consolidated and divided residence into an overall two-parent category.
Single-parent families were created by four factors, two of which were unique to slavery. The general causes were parental death and bastardy. The causes unique to slavery were the sale of a parent and the existence of a white father.9 Nine percent of the female-headed families resulted from death of the father. If we add orphans to this figure, the total percentage of the ex-slaves providing family information who reported their fathers as dead is 4.5 percent. Most of the father-headed families were created by the death of the mother. Taking these together with the orphans leaves a total percentage of dead mothers of 3 percent. Parental death was not a major cause of singleparent families in the slave narrative collection.
The number of children with unknown fathers is difficult to investigate using the narratives. A two-parent family could well have been composed of children with different fathers who were simply accepted into the family. If “illegitimacy” is defined as ex-slaves who said that they did not know the identity of their father, 9 percent of the children in the mother-headed category could be defined as illegitimate. This group can be further divided in two segments: one in which the child was clearly illegitimate and another in which the parents were separated before the child knew his or her father. Henderson Perkins fell into the former category: “In dem days, ‘twarnt so particular ‘bout gettin married, and my mammy wam’t before I’se born, so I’se don’ know my father.” lo John Finely told a somewhat different story but one that also indicates his mother and father were never “married”: “My pappy an on dat plantation but I don’t know him ‘cause mammy never talks ‘bout him ‘cept to say, He am here.”ll In the other group were children such as Easter Wells who was very young when she was separated from her father.
I never saw my father; in fact, I never heard my mammy say anything about him and I don’t guess I ever asked her anything about him for I never thought anything about not having a father. I guess he belonged to another
9. White parentage is treated as a separate category from bastardy because of legal issues regarding slavery. Whites could not form a legal family with blacks even if they chose to.
10. Rawick, vol. 5 (3), Texas Narrarives, p. 180.
11. Rawick, vol. 4 (2), Texas Narratives, p. 35.