What happened to the u.s. marriage rate during the fifties?

J South Hist. Author manuscript; available in PMC 2010 Dec 15.

Published in final edited form as:

J South Hist. 2010 Feb; 76(1): 39–70.

PMCID: PMC3002115

NIHMSID: NIHMS191544

In 1864, the editor of the Southern Literary Messenger received a letter from H. R., who identified herself as an eighteen-year-old, unmarried woman from Buckingham County, Virginia. Hattie, as the editor called the anonymous letter writer, admitted suffering from a “chill feeling of despair” brought on by the “execrable war.” She wrote that

the reflection has been brought to my mind with great force that after this war is closed, how vast a difference there will be in the numbers of males and females.

Having made up my mind not to be an old maid, and having only a moderate fortune and less beauty. I fear I shall find it rather difficult to accomplish my wishes.

She asked the editor, “[D]o you think that I will be overlooked ‘amidst this wreck of matter and crush of men and horses’[?]”1

Social historians of the Civil War have generally agreed that fears like Hattie’s were well grounded in demographic realities. Nearly 620,000 men were killed in the war, a number approximately equal to the deaths in all other American wars from the Revolution to the Korean War combined. The deaths of huge numbers of men, Nancy Cott has argued, rendered “the assumption that every woman would be a wife … questionable, perhaps untenable.”2 The death rate was especially great in the Confederacy, which lost approximately one in five white men of military age in the conflict.3 Catherine Clinton has stated that the reduced population of young men “demographically deprived” southern women of husbands.4 Drew Gilpin Faust, in her study of elite southern white women during the war, has argued that the loss of such a large proportion of the South’s male population undermined the region’s established pattern of family formation and threatened the identity of white women as wives and mothers. A generation of southern women faced the prospect of becoming spinsters reliant on their families for support.5 Similarly, in a recent study of white southern womanhood in the late nineteenth century, Jane Turner Censer has expressed the notion that the Civil War “constituted a watershed” in the likelihood of marriage for southern white women.6

If it is true that the war condemned a generation of southern women to spinsterhood, then demography, specifically the imbalance in the number of men and women, contributed to what some historians have described as a war-provoked “crisis in gender.”7 Unfortunately, despite repeated calls for more research, the demography of the United States during the Civil War and its aftermath remains a neglected subject.8 A few scholars have attempted to reconstruct the population history of selected counties, but the geographic diversity of the nation’s population, economy, and wartime experiences is too great to permit an inference of general patterns based on these local studies.9 Consequently, there are no studies of the war’s effect on postwar marriage in the South at large.

This article attempts to fill that gap. It relies on samples of the federal decennial censuses from 1850 through 1880 to compare white marriage patterns before and after the war.10 Although this study presents marriage estimates for all regions, the discussion focuses on the South, which suffered three times the rate of military deaths of the North.11 The results suggest that a modest version of what demographers call a marriage squeeze affected southern white women who reached marriage age during the war. Faced with a shortage of potential spouses in the postwar period, some women postponed marriage or chose less appropriate husbands. Diaries, letters, and memoirs of southern women supplement the quantitative analysis and document women’s wartime fears of spinsterhood. However, the results of this study demonstrate that women’s feared spinsterhood failed to materialize over the long term. The vast majority (approximately 92 percent) of southern white women who came of marriage age during the war married at some point in their lives. Indeed, the marriage squeeze on southern women apparent in the 1870 census is no longer evident in the 1880 census.

Before examining the war’s effect on marriage patterns, it is important to understand the economic, demographic, and cultural context of marriage in the nineteenth-century United States. First of all, before they married, young couples were expected to acquire the economic resources to establish an independent household. The age of marriage, therefore, depended on real wages, inheritance, and the relative cost of purchasing land, farms, farm machinery, and homes. Although studies are few and subject to possible biases, most scholars agree that the ready availability of inexpensive land in colonial America made marriage feasible at an early age. As a result, marriages occurred several years earlier, on average, in colonial America than in Europe, and much higher proportions of the population eventually married. Community-based studies suggest an average age at marriage of about 20 years for women in the early colonial period and about 26 for men. As population densities increased and land prices rose in the eighteenth and nineteenth centuries, American couples delayed marriage, and a higher proportion remained permanently unmarried. The published census figures for 1890, which are the earliest that permit estimates of age at marriage, reveal that the mean age at marriage was 23.8 for white women and 27.8 for white men—little different from those ages in England.12 This discussion demonstrates that the Civil War occurred in the midst of a long, gradual increase in the average age at marriage.

Whether and when young men and women married also depended on the presence of desirable alternatives to marriage. Twentieth-century demographers have observed that women are more likely to postpone marriage, remain permanently unmarried, or disrupt current marriages when they have greater opportunities for education and for entering the paid labor force.13 In the antebellum South, negligible opportunity for young white women to participate in the paid labor force may have increased the cultural incentive for early marriage.14 The author of a recent article examining marriage patterns in 1860 has concluded that women’s participation in the paid labor force was the most important factor in determining regional differences in marriage timing. All else being equal, white men and women married later in life in areas with high proportions of single, wage-earning women and earlier in areas with low proportions.15

Finally, marriage timing and the proportion of men and women who eventually married are strongly correlated with the availability of potential marriage partners. An imbalance in the sex ratio—perhaps as a result of sex differentials in migration, rapid changes in the size of birth cohorts, or sex differentials in mortality—reduces the likelihood of marriage. Severe imbalances in the number of men and women can cause a marriage squeeze.16 The more prevalent sex faces greater competition for spouses, typically causing them to delay marriage, marry individuals of less appropriate age or condition, or remain unmarried. Sex differentials in migration in the eighteenth and nineteenth centuries, for example, resulted in men greatly outnumbering women in the West and women significantly outnumbering men in the East. The imbalance created a marriage squeeze on men in western census regions and on women in eastern census regions. Far more women remained permanently unmarried in the East than on the frontier, where they continued to marry at the young ages and high proportions characteristic of the early colonial period.17

Differences in cohort size and cultural expectations about the appropriate age differentials between husbands and wives put additional pressure on unmarried women, who typically married men four or five years older than themselves. The number of children born in the United States increased each year.18 As a result, younger women searched for marriage partners among smaller cohorts of older men. In 1860, for example, white women from 20 through 24 years old outnumbered white men from 25 to 29 by 13 percent.19 Although this difference suggests that many young women would be forced to delay marriage or remain permanently single, the greater tendency for men to remarry after widowhood ensured that most women in the nineteenth century eventually married.

The American Civil War increased the difficulty of family formation caused by sex differentials in migration and differences in cohort size. Following the bombardment of Fort Sumter on April 12, 1861, the nation underwent a massive mobilization that removed large numbers of single men at peak marriage age from their communities to various fronts in the war’s western and eastern theaters. Although all states had high participation rates, the institution of slavery allowed the South to mobilize a greater proportion of its men of military age. Gary Gallagher estimates that the South had mobilized between 75 and 85 percent of its white male population of military age by the end of the war. In contrast, approximately 50 percent of the white male population of military age enlisted in the North.20

Despite the obvious hindrance that military service posed to courtship and marriage, observers frequently noted that the war acted as a catalyst for marriage. Bell Irvin Wiley’s early social histories of Confederate and Union soldiers document the obsession of unmarried men with the possibility of losing a fiancée or not finding a wife after the war. Letters to relatives were replete with inquiries about who was marrying whom and exhortations to local women not to marry other suitors, especially slackers and men exempt from the draft.21 Green Berry Samuels, for example, wrote to his future wife Kathleen Boone in April 1861, begging her, “Dont be so cruel as to fall in love with some of the nice young men about F. Royal whilst I am gone away to fight the battles of Va.” In a subsequent letter, Samuels had harsh words for men who stayed home. “Should Mr. Lehew tease you about my being at Harpers Ferry, tell him you would not have a sweetheart unless he was willing to risk his life in defense of his country and also that you would never marry any man who staid at home and had nothing better to do than teaze the ladies.”22

A flurry of marriages occurred early in the war, whenever men went on furlough, and then again at the end of the war. Richmond, the Confederate capital, hosted hundreds of wartime marriages, leading observers to marvel at the “marriage frenzy.”23 In 1863, after receiving a visit from her engaged nephew, who had lost a leg during the war, Judith McGuire of Virginia wrote, “I believe that neither war, pestilence, nor famine could put an end to the marrying and giving in marriage which is constantly going on. Strange that these sons of Mars can so assiduously devote themselves to Cupid and Hymen; but every respite, every furlough, must be thus employed.” In early 1865, McGuire again commented on “a perfect mania on the subject of matrimony. Some of the churches may be seen open and lighted almost every night for bridals, and wherever I turn I hear of marriages in prospect.”24 As she traveled home with a group of other southerners at the war’s end, Kate Cumming heard a soldier declare that “the first thing he intended doing, after he arrived home, was to get married. I heard many of the soldiers say the same.”25

During the war southern women, too, aspired to marry. Drew Faust has noted an “intense absorption of unmarried southern women of all ages with courtship and matrimony” and a corresponding dread of “spinsterhood.”26 Based on overwhelming evidence from diaries and letters, women favored men who fought. When Eliza Andrews met a friend from Boston, apparently a Confederate sympathizer, she “felt uncomfortably conspicuous” walking with him because of his “broad-cloth coat and a stovepipe hat.” “I am almost ashamed, nowadays, to be seen with any man not in uniform,” she continued.27 Unmarried women expressed growing alarm as the draft removed the few men left at home to the front, a concern parodied by Vanity Fair (see Figure 1). “There is but few men at home,” complained a southern woman to an unmarried cousin, “and what there is I reckon has declined the idea of ever marrying.”28

“Awful Possibility.” Illustration by Howard Del, from Vinify Fair, March 29, 1862, p. 154. Reproduced with permission of Special Collections, Musselman Library, Gettysburg College, Gettysburg, Pa.

As time passed and casualties mounted, some women became resigned to life without a husband. Others were willing to compromise on acceptable partners. In 1862 Ada Bacot complained of “two fashions which have crept into society … [t]hat of marrieng for money, & that of a woman marrieng a man younger than herself.”29 Military service conferred cachet upon the soldier, often regardless of his class. After the war, wealth became less important in the economically devastated South when contracting marriages, and many women married below their social class. Susan Bradford Eppes met her “Soldier in Gray” following the battle of Gettysburg, and they married after the war. “I hope we will not have too much trouble with my trousseau,” she remarked. “I wish they were willing for me to have only simple clothes for I am marrying a poor man and I do not ever intend to live beyond his means. Father would be willing but Mother and the sisters think, because they had these clothes I must have them, too.”30 Some southern women in areas occupied by the enemy risked social ostracism by courting and marrying Union soldiers. Historians of the occupied South have written, “Letters and diaries of Union men in every occupied community reveal considerable social intercourse between Federals and ‘secesh’ girls which in a good many instances led to romances and marriages.”31

The shortage of suitable men after the war gave those remaining many choices of women to marry, allowing widowers to remarry and others to try to escape their former obligations. Though more evidence is needed to draw concrete conclusions, a few northern and southern men may have attempted to remarry without divorce. Southerner Anna Bragg related to her husband news of a widower with three children remarrying and also described the wedding of Captain Paine to Miss Mary Frincks. “Some say he has a wife and child living,” Anna Bragg noted. A Union chaplain turned down the request of a woman who “had the hardihood to ask me to marry her to a man who confesses that he has a wife in Reading Pa. and who says his wife has had a ‘nigger baby’ since he came to the army,”32

After the war, white southerners responded to interracial marriage with violence. In 1870 Frances Harper, who had been an abolitionist, described a conversation with a black man whose son had “married a white woman, or girl, and was shot down, and there was, as I understand, no investigation by the jury; and a number of cases have occurred of murders, for which the punishment has been very lax, or not at all … .”33 Widespread fears that emancipation would increase the incidence of interracial sexual encounters led states to pass more laws prohibiting interracial marriage “during the Civil War and Reconstruction than in any comparably short period.”34 The deaths of so many young men during the war probably contributed to such fears. John Blassingame, for example, has argued that the death of white men in the war led to a postwar increase in sexual contacts between white women and black men in New Orleans.35 The number of interracial unions no doubt remained quite small. Although instances of interracial marriage and cohabitation occurred during Reconstruction in numbers large enough to suggest some initial level of toleration from white neighbors, the vast majority of white women—confronted with the possibility of violence, rigid enforcement of miscegenation laws, and the vast social distance between themselves and black men—married white men.36

Not only the deaths of white men but also their wounds affected the prospects for marriage in the aftermath of the war. One of the most important roles of nurses, official matrons, and volunteer hospital visitors was to help wounded men cope with the psychological impact of their injuries. “I constantly hear the unmarried ones,” wrote Kate Cumming, a nurse describing her amputee patients, “wondering if the girls will marry them now.” Years after the war, another southern nurse, Fannie Beers, had “never forgiven” a “heartless girl” who rejected her betrothed. The young man had suffered a facial wound and lost a leg. He told Beers about his engagement to “one of the prettiest … girls in ‘Massissip’” and asked her to write a letter telling the young woman about his wounds. While they awaited his fiancée’s reply, Beers eased the wounded man’s worries that he would have to “let her off” by relating “instances of women who only loved more because the object of their affection had been unfortunate.” She later regretted nurturing his hopes, for it was her “misfortune to read to him a very cold letter from his lady-love, who declined to marry ‘a cripple.’” Though “inconsolable” for a short time, he soon decided that she would not have been a good wife.37 As for southern women, faced with the choice of marrying amputees or cripples, men from lower social classes, or no one at all, some of these women ultimately married disabled veterans.

Female hospital workers and visitors treated disfigured patients as heroes instead of shrinking from them in horror or pity. Judith McGuire, a volunteer nurse, eased the anxiety of a man eager to travel home to marry his sweetheart. “Ah,” she said, “but you must show her your scars, and if she is a girl worth having she will love you all the better for having bled for your country ….”38 Nurses described how their patients cheerfully dealt with amputations and crippling injuries. “We have a room with seven men in it, who have lost a limb each. It is a perfect treat to go into it, as the men seem to do little else but laugh,” Kate Cumming remarked. “They are young men, and say to me, I must tell all the young ladies to come and see them, and that they will make excellent husbands, as they will be sure never to run away.”39

Letter writers and diary keepers commented frequently on wartime marriage, but after the war many of them stopped writing; the resulting silence created a gap in evidence about postwar marriage patterns. During the war, many Americans sensed that they were living through exciting, unique times. In order to record their experiences and reactions, they started keeping personal diaries, only to stop writing when the conflict ended. Many southerners stopped confiding to diaries because the humiliation and pain of defeat left them unable or unwilling to express themselves in writing. Furthermore, letter writing decreased from wartime levels as soldiers and refugees returned home. Women, especially, avoided recording events and sentiments that could be perceived as dishonoring Confederate veterans and their military service, and imbalanced sex ratios and the marriage squeeze may have served to remind southerners of their loss.40

Finally, people, particularly southern women, may have stopped writing about marriage after the war simply because the fears of spinsterhood expressed during the war were not realized. Risk assessment experts and cultural demographers have noted that perceptions of demographic patterns often do not match reality and are biased in predictable ways. Extraordinary, catastrophic, and uncontrollable risks, such as those posed by epidemics, natural disasters, and wars, are perceived to have a much greater impact than more ordinary and controllable risks.41 It is also possible that fears of spinsterhood spurred women to accept unconventional suitors—men who were younger, much older, of a lower social class, from a distant community, or of a different ethnic group—and thus reduced what would otherwise have been a larger cohort of unmarried women.42

Given the dearth of qualitative evidence on postwar marriage patterns and the possible biases of observers, quantitative data are required to document postwar marriage patterns. The Integrated Public Use Microdata Series (IPUMS) samples of the 1850, 1860, 1870, and 1880 federal censuses, recently constructed at the Minnesota Population Center, permit scholars to test, for the first time, qualitative assessments of the war’s effects on marriage patterns.43 Several estimates of marriage—including the mean age at marriage, proportions never married, and ratios of men to women—can be constructed for each census year. However, it is impossible to determine year-to-year variations in marriage during the war itself (1861–1865) or in its immediate aftermath (1865–1869); there may well have been some short-term effects on marriage that were resolved in the five years between the end of the war and the 1870 enumeration.44 Furthermore, the federal decennial censuses of 1850 through 1880 do not record the year or duration of current marriages, the number of times each individual had been married, and the state or locality where current marriages were contracted. Therefore, it is possible to examine only the long-term impact of the war on first marriages. The short-term effect of the war on the timing and incidence of marriages and its short- and long-term effects on remarriage are difficult or impossible to discern from census records.

Table 1 shows various measures of the timing and incidence of first marriage for white men and women in the census years 1850, 1860, 1870, and 1880 arranged by age and census region.45 The Singulate Mean Age at Marriage (SMAM) is a convenient way of summarizing the timing of marriage for the population that eventually married.46 At the national level, the mean age at marriage was about 27 years for men and 23 years for women, with little apparent trend between 1850 and 1880. However, the sectional averages provide modest support for the hypothesis that the war created a marriage squeeze among southern white women. The mean age at first marriage for southern white women rose 0.7 years between 1860 and 1870 (from 22.3 years to 23.0 years). At the same time, the mean age at first marriage for southern white men fell 0.7 years (from 26.6 years to 25.9). In contrast, SMAMs for both men and women in the North rose together modestly (about 0.4 years for men and 0.3 years for women).

TABLE 1

Nuptiality Measures for the White Population of the United States, 1850–1880a

The greater death rate of young men in the Confederate forces and the resulting imbalance in the number of men and women seeking to marry is the most likely reason for the distinct southern pattern. The imbalance is reflected in the sex ratio (the number of white men from 20 through 29 years old per 100 white women in the same age range). In the North, the sex ratio declined a modest 1.0 percent after the war, from 98.3 in 1860 to 97.3 in 1870. In the South, the decline was a much steeper 15.8 percent, from 104.0 in 1860 to just 87.5 in 1870. In essence, for every 100 white women from 20 to 29 years old, there were 4.0 more white men the same age in 1860 and 12.5 fewer in 1870. The sex ratio, however, is a crude measure of potential marriage partners. It includes many men and women currently married and thus unavailable as potential spouses. An estimate of the marital sex ratio—the ratio of the expected number of first marriages among white men to the expected number among white women—is therefore also included in Table 1.47 The marital sex ratio, plotted by section in Figure 2, highlights the war’s differential impact on the availability of potential spouses. In the North, the marital sex ratio declined a modest 4.3 percent between 1860 and 1870. In southern census regions, however, the ratio declined a sharp 25.5 percent from the slightly female-biased, prewar value of 0.94 in 1860 to a very large female bias of 0.70 in 1870. This means that for every 100 southern white women expected to be entering marriage in 1870 there would be just 70 southern white men.

Marital Sex Ratio, Northern and Southern Census Regions, 1850–1880.

The lower age at first marriage for southern white men after the war, as compared with before the war, may have been related to the decrease in competition for property and in the purchase price of farms and businesses and to the greater probability of their inheriting property. After the war, there were fewer people in a position to make large purchases. Furthermore, land prices collapsed, which made it cheaper to purchase a new farm. A sample of farms in eight counties of the South suggests that the average cost of an improved acre of land was almost $30 in 1860; ten years later it had fallen to less than $15.48 Moreover, men who lost brothers in the war probably received larger inheritances. In Orange County, North Carolina, for instance, the average number of sons mentioned in wills declined 28.2 percent after the war.49 War widows with property also may have made attractive wives for men who would not have otherwise had the resources to marry.50

Lower farm prices, less competition, and greater shares of inheritances, however, were probably offset by reduced wealth, reduced savings, reduced incomes, increased taxes, and the bleak economic conditions of the postwar South. Most white families—especially slave-holding families—lost substantial real and personal wealth because of the war, which probably lowered the value of intergenerational transfers to young couples considering marriage.51 Some families, moreover, incurred substantial legal costs to regain property confiscated during the war. Although young white men could expect less competition from other southern white men for farms, they faced increased competition from a small but growing number of blacks and northern whites. Reduced incomes, higher taxes, property fallen into disrepair, and greatly reduced numbers of livestock added to southern white men’s financial hardships. Without more research it is impossible to say whether and to what degree the decline in the average age of marriage among southern white men after the war resulted from economic change.

The percentage of the population who had not married by the time they reached the ages of 45 through 54 (labeled “single” in the table) in each census year is also shown in Table 1. Because of the very small number of individuals who marry for the first time after age 45, the percentage single in this age group is typically interpreted by demographers as the percentage of the population remaining permanently unmarried. The results suggest that the vast majority of white men and women in the late nineteenth century eventually married. Overall, only about 7 percent of white men and women 45 through 54 years old in the four IPUMS samples analyzed here never married. There was little apparent trend at the national level in the percentage of those who never married. (It should be remembered, however, that, in the 1870 and 1880 censuses, the majority of men and women 45 through 54 years old had married before the war.)

A sharp East/West gradient in the timing of first marriage and the percentage remaining permanently unmarried is also evident in Table 1, especially among women. In the North, New England women had the highest average age at first marriage and the highest percentage never married. The average age at first marriage and percentage never married decline steadily as the census moves west. One reason for the East/West differential appears to be the availability of potential marriage partners. Sex ratios were lowest in the East and highest in the West. Other factors include regional differences in land availability and in the availability of paid work for women.52 In southern census regions the same East/West relationship holds, with women in the South Atlantic region having the highest average age at marriage and the highest percentage never married, while women in the West–South Central region had the lowest. Although there are too few cases in the census samples to be confident with the results, the combined Mountain and Pacific census region appears to be a clear exception to the East/West gradient for men: sex ratios are so high that many men delayed marriage and a very large percentage (about 30 percent) remained single at the age of 45 through 54 years.53 A modest sectional differential is also evident in Table 1. Southern men and women married slightly earlier than northern women. The earlier marriage age did not translate into a lower percentage of southern women who never married, however. With the exception of the 1850 census year, northern women were a little more likely to marry eventually than southern women.

The small changes in nuptiality depicted in Table 1 do not indicate a dramatic impact from the Civil War.54 Even when the results are limited to the South, the mean age at marriage changed only slightly and then returned to its long-term trend. The vast majority of men and women in both sections continued to marry. It is unclear, however, whether the modest, short-term marriage squeeze evident in the period data translated into reduced lifetime chances of marriage. Table 2 shifts the perspective from the estimates centered on census years to actual birth cohorts. IPUMS samples for census years 1900, 1910, 1920, and 1930 were added to the analysis to follow the various birth cohorts into old age. Because of the focus on birth cohorts, Table 2 is limited to the native-born population, and the sectional analysis is based on section of birth.55

TABLE 2

Nuptiality Measures for the Native-Born White Population by Birth Cohort and Section of Birth

Cohorts experience strong period effects, such as wars, at different ages. Individuals born between 1821 and 1830, for example, were between the ages of 31 and 40 at the outbreak of the war in 1861 and between 35 and 44 years old at its conclusion. Given lower military participation rates by men in their thirties and forties relative to men in their twenties and the large proportion of men and women in the cohort who married prior to the war, the expectation is that the war’s effect on the timing and incidence of first marriage among the 1821–1830 cohort was minimal.56 There may have been some impact, of course, at older ages. The small proportion of women still single in their thirties at the outbreak of the war may have seen their chances of eventually marrying diminish, especially if unmarried men in their age cohort married younger women, thereby taking advantage of the reduced number of young men in the postwar South. Likewise, women widowed in their thirties may also have faced more competition for husbands of the appropriate age.57 Overall, however, the effect of the war on first marriage for the 1821–1830 birth cohort was probably quite small. In contrast, the 1831–1840 and especially the 1841–1850 birth cohorts were at peak marriage ages during the war, and a higher proportion of men in those cohorts fought and died. The expectation is that women in these birth cohorts, especially in the South, would have experienced a dramatic marriage squeeze after the war and that men in these cohorts would have had much less competition in their search for wives.

The cohort results in Table 2 provide little support for the hypothesis that the war had long-term consequences for women’s chances of ever marrying or that the war’s impact fell more heavily on specific birth cohorts. At the national level, the female mean age at first marriage rose slowly and steadily with each cohort, from 22.6 years for the 1821–1830 birth cohort to 23.4 years for the 1871–1880 cohort, suggesting the importance of long-run economic factors such as rising farm prices and increasing participation of single women in the paid labor force. Likewise, the percentage of women remaining single at older ages rose steadily with each birth cohort from 7.5 percent among women in the 1821–1830 cohort to 10.7 percent among women in the 1871–1880 cohort. In this context, the percentages of women who never married in the 1831–1840 (7.8 percent) and 1841–1850 (8.0 percent) birth cohorts do not stand out as anomalously high. There is evidence for a modest, short-lived marriage squeeze when results are confined to men and women born in southern census regions, however. There, the mean age at first marriage rose about 0.6 years for white women in the 1831–1840, 1841–1850, and 1851–1860 birth cohorts before declining back toward the 1821–1830 cohort level for women born after 1860. There is no indication that women in these cohorts faced a diminished lifetime prospect for marriage. Approximately 92 percent of the southern-born white women in the 1831–1860 cohorts eventually married, roughly equal to the percentage for adjacent cohorts. Men born in the South are unique among the groups shown in Table 2 in having no long-term upward trend toward higher mean age at first marriage. In contrast to women born in the South between 1841 and 1860, men in the 1841–1850 and 1851–1860 southern birth cohorts had about a 0.4 year lower mean age at first marriage relative to men in adjacent birth cohorts.

Overall, however, the war had a modest effect on the timing and incidence of first marriage. Only the deaths of single men would have affected the timing of first marriages. As far as the effect of the deaths of married men is concerned, no informed estimate exists of the number of widows produced by the war.58 Possibly, a rough estimate could be made using age-specific marriage rates and age-specific service records. Unfortunately, there are no comprehensive enrollment statistics for the Confederate army that include the necessary detail. However, information on the marriage status of a sample of Union soldiers was collected by the U.S. Sanitary Commission (the “Gould” sample); this sample may substitute for statistics on Confederate soldiers, although the Confederate army probably drew from a wider population base than did the Union army and thus included more married men.59 If one regards the estimate of the proportion of married men in the Gould sample (28 percent) as a lower bound, the war widowed more than 70,000 southern white women.60 A fall in remarriage rates for women widowed by the war would, all else being equal, reduce the impact of the war on single women’s marriage prospects.

Although the number of war widows cannot be measured with the census, the 1880–1930 IPUMS samples—which recorded current marital status—permit a rough estimate of the regional distribution of widowhood in the period after 1880. Table 2 includes estimates of the proportion of native-born white men and women ages 40–49 and 50–59 currently widowed by birth cohort. Women widowed by the war, of course, could have remarried in the fifteeen or more years following the end of the war, and many women enumerated as widowed in 1880 and later censuses lost their husbands before or after the war, but it is nonetheless clear that the Civil War had a major impact on the incidence of widowhood and possibly the rate of remarriage.61 Approximately one in five southern-born women aged 40–49 in the South were currently widowed in 1880, compared with just one in nine among northern-born women. Figure 3, a county-level map of widowhood in 1880, highlights quite clearly the sectional impact of the war on subsequent widowhood. The prevalence of widows in the South appears to be concentrated around urban areas and along the Mississippi River, suggesting a degree of geographic mobility among widows to areas offering greater access to wage labor and social support networks and higher overall mortality rates in counties adjacent to the river.

Percentage of White Women from 40 to 59 Years Old Currently Widowed, 1880 Census.

Source: North Atlantic Population Project and Minnesota Population Center, NAPP: Complete Count Microdata. Version 2.0 (Minneapolis, 2008).

The modest impact of the war on long-term marriage patterns may also be explained in part by adjustments in the age differential between spouses and an increased tendency of men and women to marry individuals outside their customary group. If women relaxed their standards of acceptable partners, the impact of the war on first marriage may have been lessened. Table 3 indicates more variability among young women’s spouses in 1870 than in earlier census years, suggesting less discrimination on the part of women in the immediate postwar years. Southern women aged 20–24 in 1870 were more likely to be married to a younger man (the percentage who did so increased from 6.4 percent in 1860 to 8.6 percent in 1870) or a much older man (up from 6.4 to 8.0 percent).62 Southern women aged 20–24 were also more likely to be married to a man born in a northern census region (up from 4.8 percent in 1860 to 5.6 percent in 1870), a foreign-born man (up from 2.4 to 3.1 percent), or a man with little or no real estate wealth (up from 58.4 to 67.2 percent). A similar tendency of women to select marriage partners from a somewhat wider pool of eligible spouses is also discernible in the North, but to a much lesser extent than in the South. One notable exception is the greater tendency of native-born, northern women to many foreign-born men. Between 1850 and 1880 the percentage of northern, native-born, white women marrying foreign men increased from 5.6 to 12.5 percent, approximately four times the rate of native-born women in the South. The greater percentage of northern women married to foreign-bom men reflects immigrants’ preference for settling in the free labor North before the war, their continued avoidance of the economically stagnant South after the war, and high numbers of male immigrants relative to female immigrants.

TABLE 3

Characteristics of Men Currently Married to Native-born, White Women from 20 through 24 Years of age

Finally, the intrinsic dynamics of the postwar American population mitigated the long-term effects of wartime mortality on marriage. High fertility before the war resulted in large differences in cohort size and rapid elimination of short-term gender imbalances. Table 1 indicates that despite the short-term shock evident in the 1870 census, the sex ratio among white men and women aged 20–29 and the marital sex ratio had returned to near their prewar levels by 1880. Theoretically, therefore, there were still potential marriage partners for the vast majority of postwar women, though unequal geographic distribution of men and women still exerted strong influence locally.63 Indeed, after the initial report from the 1880 census was published, a number of observers commented on the nation’s growing surplus of men. In February 1881 the Washington Post commented, “It appears that the disparity between the number of males and the number of females is increasing, and unless we shall have a very destructive war there is no chance for a restoration of the equilibrium.”64 Three months later the same paper waxed eloquent about women’s prospects for marriage:

The new census … shows that in the United States the men outnumber women to the extent of nearly a million. This reverses the popular belief regarding the number of superfluous women, and the ladies are to be congratulated on the new condition of things. This is one of those paradoxical cases in which the minority is more powerful than the majority, and the fewer there are of women the better off they will be. They need no longer be haunted by the fear that there will not be men enough to go around. They will command the matrimonial situation, and may dictate their own terms. There will be a corner in women. Nothing can prevent it. And it will last. This misproportion exists not only between adults of the two sexes, but extends down into the cradle, so that the coming woman, as well as the woman who is already here, is likely to have things pretty much her own way.

The old maid has no longer any reason for being. If she remain single, it may, without any special effort of gallantry, be taken for granted that she does so of her own accord. It cannot be because there are not men enough, and to spare. She will probably marry, and have an establishment of her own … .65

Thus, despite its enormous death toll, the war had a modest, short-lived effect on the timing and incidence of first marriage. On one hand, for a brief period after the war, southern men who had survived the conflict enjoyed demographic advantages in the search for a wife. Relative to southern men born a generation earlier or later, white men in the postwar South had more potential spouses to choose from and married at a slightly younger age. On the other hand, unmarried southern white women in their twenties at the outbreak of the war faced an acute shortage of available men after the war. Unsurprisingly, a small number of women in this cohort delayed marriage or compromised on marriage partners. The vast majority eventually married, however, and the war did not create a large cohort of lifelong spinsters or so-called maiden aunts. Although available census data limit the analysis of the timing and incidence of first marriage, an analysis of widowhood in the 1880 and later censuses suggests that many women widowed during or after the war were unable to remarry. High levels of widowhood in the postwar South among relatively young women probably reflects both high death rates of southern men during the war and low remarriage rates of southern widows afterward.

The results of this study are remarkably consistent with the results of Louis Henry’s classic study of the effect of World War I on marriage patterns in France. Like the Confederate South, France lost between 15 and 20 percent of its young men in a few years of war. But, despite the severe marriage squeeze experienced by French women after the war, the vast majority of them eventually married. Indeed, among French women who came of age during the war, there was only a modest increase in the percentage remaining unmarried at age 50. In examining this paradox, Henry concluded that relatively small changes in age preferences compensated for the large imbalance in potential marriage partners. The French marriage market, like the postbellum South’s marriage market, proved to be surprisingly flexible.66

Although the war did not create a crisis in white southern women’s ability to marry and fulfill their expected roles as wives and mothers, it did nonetheless have some effect on the overall character and meaning of southern marriage. Most important, the war utterly transformed former slaves’ ability to choose partners, marry, and form families. In addition, as many historians have argued, the war challenged traditional gender roles and individual marriages in diverse and interesting ways. Husbands seriously wounded in the war or suffering psychologically from their experiences no doubt affected their households and their wives’ and children’s lives, as did the loss of property and wealth.

The demographic evidence highlights several promising areas for future research. The degree of community diversity in demographic behavior is an interesting and largely unanswered question, especially given known patterns of local enlistments and the clustering of war mortality in specific companies and regiments. Although low sample densities in existing IPUMS census samples currently prevent a detailed investigation of variations by community, higher sample densities in future samples and detailed community studies by individual researchers may show that marriage patterns in specific communities varied widely from the sectional patterns observed in this article. The results on sectional marriage patterns discussed here will provide a standard by which other scholars can compare other communities or individual marriages.

A more promising area for research is the social demography of widowhood. There is a stark, surprising contrast between the extremely high levels of widowhood in the postwar South and the paucity of historical research. Nearly one in three southern white women over the age of 40 were currently widowed in 1880, representing about 300,000 women and many more dependent children. Although most of these women were not war widows, the war clearly contributed to marked sectional differences in the rate of widowhood. Despite these staggering numbers and the growing interest in the social history of the war and its aftermath, there are almost no studies of postwar southern widowhood. How these women and their families coped with the dire economic environment of the postbellum South deserves the attention of social historians.67

References

1. Editor’s Table. Southern Literary Messenger. 1864 February;38:124–25. [Google Scholar] (first, second, and fourth quotations on 125; third quotation on 124). This work was supported in part by NIHCD grant numbers 1 K01-HD052617-01 and 1 KO1-HD051494-01. The authors would like to thank Catherine Fitch. Steven Ruggles, Douglas Bradburn. Diane Sommerville, Orville Vemon Burton. Stephanie McCurry, and participants in the University of Minnesota Early American Workshop for helpful comments on earlier versions of this paper.

2. Vinovskis Maris A. Have Social Historians Lost the Civil War? Some Preliminary Demographic Speculations. Journal of American History. 1989 June;76:34–58. [Google Scholar]Cott Nancy F. Public Vows: A History of Marriage and the Nation. Cambridge, Mass.: 2000. pp. 78–79. [Google Scholar]

3. Vinovskis Have Social Historians Lost the Civil War? pp. 38–39. Vinovskis considered men of military age to be those listed as 13 through 43 years old in the 1860 census. The estimate of the ratio of southern men killed in the war is based in part on the number of deaths among Union troops.

4. Clinton Catherine. Maiden Aunt. In: Wilson Charles Reagan, Ferris William., editors. Encyclopedia of Southern Culture. Chapel Hill: 1989. pp. 1553–54. (quotation on 1553) [Google Scholar]

5. Faust Drew Gilpin. Mothers of Invention: Women of the Slaveholding South in the American Civil War. Chapel Hill: 1996. pp. 139–52. [Google Scholar]

6. Censer Jane Turner. The Reconstruction of White Southern Womanhood, 1865–1895. Baton Rouge: 2003. p. 32. [Google Scholar]

7. The Southern Lady: From Pedestal to Politics, 1830–1930. Chicago: 1970. Anne Firor Scott has described the war’s impact on southern women. [Google Scholar] In Scott’s interpretation, the accelerated social change war and southern women’s transition from the private patriarchal southern household to the public world of political activism. More recent studies, while agreeing that the war challenged traditional gender roles, suggest that southern women did not eagerly embrace change. After the war, many women sought to restore southern men’s traditional masculine self-image. See Faust Mothers of Invention.; and Rable George C. Civil Wars: Women and the Crisis of Southern Nationalism. Urbana: 1989. [Google Scholar] The Civil War as a Crisis in Gender: Augusta, Georgia, 1860–1890. Athens, Ga.: 1995. LeeAnn Whites has made the most forceful case for interpreting the war as a crisis in gender. [Google Scholar] The notion of a so-called crisis in gender encompasses more than women’s ability to marry and fulfill their expected roles as wives and mothers. Excellent studies of the war’s impact on gender can be found in Clinton Catherine, Silber Nina., editors. Divided Houses: Gender and the Civil War. New York: 1992. [Google Scholar]; and Clinton, Silber, editors. Battle Scars: Gender and Sexuality in the American Civil War. New York: 2006. [Google Scholar]

8. Vinovskis . “Have Social Historians Lost the Civil War?” Phillip Shaw Paludan has also called attention to the lack of census-based research in “What Did the Winners Win? The Social and Economic History of the North during the Civil War,” In: McPherson James M., Cooper William J., Jr., editors. Writing the Civil War: The Quest to Understand. Columbia, S.C.: 1998. pp. 174–200. [Google Scholar]

9. Burton Orville Vernon. My Father’s House Are Many Mansions: Family and Community in Edgefield, South Carolina. Chapel Hill: 1985. pp. 263–64. [Google Scholar]; Kenzer Robert C. Kinship and Neighborhood in a Southern Community: Orange County, North Carolina, 1849–1881. Knoxville: 1987. pp. 12–17.pp. 99–100. [Google Scholar] A few studies have investigated the war’s impact on selected marriages and the general nature, character, and legal definition of nineteenth-century marriage. See, for example, Bleser Carol, Heath Frederick M. The Clays of Alabama: The Impact of the Civil War on a Southern Marriage. In: Bleser Carol., editor. Joy and in Sorrow: Women, Family, and Marriage in the Victorian South, 1830–1900. New York: 1991. pp. 135–53. [Google Scholar]; Bynum Victoria E. Unruly Women: The Politics of Social and Sexual Control in the Old South. Chapel Hill: 1992. pp. 11–14. [Google Scholar]; and McClintock Megan J. The Impact of the Civil War on Nineteenth-Century Marriages. In: Cimbala Paul A., Miller Randall M., editors. Union Soldiers and the Northern Home Front: Wartime Experiences, Postwar Adjustments. New York: 2002. pp. 395–416. [Google Scholar]

10. Only white marriage patterns can be ascertained in the prewar period; the 1850 and 1860 censuses enumerated slaves in a separate, more limited population schedule, and because the slave schedule did not group or identify family members, it is impossible to infer marital status and estimate age at marriage. These samples are part of the Integrated Public Use Microdata Series (IPUMS) created at the Minnesota Population Center (University of Minnesota, Twin Cities Campus, Minneapolis). Ruggles Steven, et al. Integrated Public Use Microdata Series: Version 3.0. Minneapolis: 2004. [Google Scholar] The samples are available for free public downloading at http://ipums.org. IPUMS samples have been used in diverse projects including studies of the onset of fertility control in the nineteenth century, long-term trends in interstate migration, long-term changes in American family structure, and the migration of Appalachian whites to the Midwest in the twentieth century. For example, see Hacker J. David. Rethinking the ‘Early’ Decline of Marital Fertility in the United States. Demography. 2003 November;40:605–20. [PubMed] [Google Scholar]; Hall Patricia Kelly, Ruggles Steven. ‘Restless in the Midst of Their Prosperity’: New Evidence on the Internal Migration of Americans, 1850–2000. Journal of American History. 2004 December;91:829–46. [Google Scholar]; Ruggles Steven. The Transformation of American Family Structure. American Historical Review. 1994 February;99:103–28. [Google Scholar]; and Alexander J. Trent. Defining the Diaspora: Appalachians in the Great Migration. Journal of Interdisciplinary History. 2006 Autumn37:219–47. [Google Scholar]

11. Vinovskis Have Social Historians Lost the Civil War? pp. 38–39.

12. Haines Michael R. Long-Term Marriage Patterns in the United States From Colonial Times to the Present. History of the Family: An International Quarterly. 1996;1(No. 1):15–39. [PubMed] [Google Scholar]Wells Robert V. The Population of England’s Colonies in America: Old English or New Americans? Population Studies. 1992 March;46:85–102. esp. 95-97. [Google Scholar]

13. Becker Gary S. A Treatise on the Family. Cambridge, Mass.: 1981. [Google Scholar]Goldscheider Frances Kobrin, Waite Linda J. Sex Differences in the Entry into Marriage. American Journal of Sociology. 1986 July;92:91–109. [Google Scholar]Ruggles Steven. The Rise of Divorce and Separation in the United Stales, 1880–1990. Demography. 1997 November;34:455–66. [PMC free article] [PubMed] [Google Scholar]Waite Linda J., Spitze Glenna D. Young Women’s Transition to Marriage. Demography. 1981 November;18:681–94. [PubMed] [Google Scholar]

14. Wyatt-Brown Bertram. Southern Honor: Ethics and Behavior in the Old South. New York: 1982. p. 229. [Google Scholar] Suzanne Lebsock has observed that southern women’s inability to earn a living outside marriage put them under enormous pressure to marry. Indeed—while marrying for money of ten brought power to men—for most southern women marriage was a “matter of survival.” Lebsock Suzanne. The Free Women of Petersburg: Status and Culture in a Southern Town, 1784–1860. New York: 1984. pp. 22–23. (quotation on 23) [Google Scholar] Although southern spinsters and maiden aunts lived meaningful lives and were valued for their important roles as family caretakers and surrogate mothers, they remained dependents and rarely earned income through market work. Carter Christine Jacobson. Indispensable Spinsters: Maiden Aunts in the Elite Families of Savannah and Charleston. In: Coryell Janet L., et al., editors. Negotiating Boundaries of Southern Womanhood: Dealing with the Powers That Be. Columbia, Mo.: 2000. pp. 110–34. [Google Scholar]

15. Hacker J. David. Economic, Demographic, and Anthropometric Correlates of First Marriage in the Mid-Nineteenth-Century United States. Social Science History. 2008 Fall32:307–45. [Google Scholar]

16. Schoen Robert. Measuring the Tightness of a Marriage Squeeze. Demography. 1983 February;20:61–78. esp. 61. [PubMed] [Google Scholar]

17. Bailey William B. Some Recent Changes in the Composition of the Population of the United States. Publications of the American Statistical Association. 1913 March;13:379–92. [Google Scholar]; Yasuba Yasukichi. Birth Rates of the White Population in the United States, 1800–1860: An Economic Study. Baltimore: 1962. pp. 54–55. [Google Scholar] For definitions of census regions, see note 45.

18. During the antebellum period the number of annual births in the white population increased by an average of 2.8 percent each year. McClelland Peter D., Zeckhauser Richard J. Demographic Dimensions of the New Republic: American Interregional Migration, Vital Statistics, and Manumissions. 1800–1860. Cambridge, Eng.: 1982. p. 74. [Google Scholar]

19. Estimate made using the 1860 IPUMS sample.

20. Vinovskis Have Social Historians Lost the Civil War? Gallagher Gary W. The Confederate War. Cambridge, Mass.: 1997. pp. 28–29. [Google Scholar]Neely Mark E., Jr. The Last Best Hope of Earth: Abraham Lincoln and the Promise of America. Cambridge, Mass.: 1993. p. 125. [Google Scholar]Escott Paul D. After Secession: Jefferson Davis and the Failure of Confederate Nationalism. Baton Rouge: 1978. pp. 63–64.pp. 116–19. [Google Scholar]

21. Wiley Bell Irvin. The Life of Johnm Reb: The Common Soldier of the Confederacy. Indianapolis: 1943. pp. 212pp. 270–85. [Google Scholar]Wiley . The Life of Billy Yank: The Common Soldier of the Union. Indianapolis: 1952. pp. 187–89. [Google Scholar]

22. Carrie Esther Samuels Spencer, editor. A Civil War Marriage in Virginia: Reminiscences and Letters. Boyce, Va.: 1956. Green Berry Samuels to Kathleen Boone Samuels, April 24, 1861, and May 6, 1861; p. 82.p. 88. [Google Scholar]

23. Barber E. Susan. ‘The White Wings of Eros’: Courtship and Marriage in Confederate Richmond. In: Clinton Catherine., editor. Southern Families at War: Loyalty and Conflict in the Civil War South. New York: 2000. pp. 119–32. (quotation on 119) [Google Scholar] Although Barber includes an analysis of median age at marriage recorded in the Richmond marriage register between 1860 and 1880, the results are probably unique to Richmond’s population, which grew rapidly during the war.

24. McGuire Judith W. Diary of a Southern Refugee, During the War. New York: 1867. pp. 243–44.pp. 329 [Google Scholar]

25. Cumming Kate. A Journal of Hospital Life in the Confederate Army of Tennessee, from the Battle of Shitoh to the End of the War. Louisville: 1866. p. 193. [Google Scholar]

26. Faust Mothers of Invention. p. 150.

27. Andrews Eliza Frances. In: The War-Time Journal of a Georgia Girl, 1864–1865. King Spencer B., Jr., editor. New York: 1908. p. 87. [Google Scholar]

28. Rable Civil Wars. p. 51.

29. Berlin Jean V., editor. A Confederate Nurse: The Diary of Ada W. Bacot, 1860–1863. Columbia, S.C.: 1994. p. 78. [Google Scholar]

30. Eppes Susan Bradford. Through Some Eventful Years. Macon, Ga.: 1926. 319 (first quotation), 331 (second and third quotations), 337. [Google Scholar]

31. Wiley Bell Irvin, Milhollen Hirst D. Embattled Confederates: An Illustrated History of Southerners at War. New York: 1964. pp. 177–78. [Google Scholar]

32. Gaughan Helen Bragg., editor. Letters of a Confederate Surgeon, 1861–65. Camden, Ark.: 1960. Anna Josephine Goddard Bragg to Junius Newport Bragg. February 19, 1865; p. 269. [Google Scholar]; Raup Hallock F., editor. Letters from a Pennsylvania Chaplain at the Siege of Petersburg. 1865. London: 1961. letter from Hallock Armstrong, June 17, 1865; p. 41. [Google Scholar] For more on the relative ease of marriage abandonment in the nineteenth-century United States, see Hartog Hendrik. Man and Wife in America: A History. Cambridge, Mass.: 2000. pp. 20–23.pp. 32pp. 87 [Google Scholar]

33. Still William., editor. The Underground Rail Road: A Record of Facts. Authentic Narratives, Letters, etc…. Philadelphia: 1872. Letter from Frances Ellen Watkins Harper, December 1870; p. 771. [Google Scholar]Hodes Martha, Women White. Black Men: Illicit Sex in the Nineteenth-Century South. New Haven: 1997. pp. 147–75. [Google Scholar]

34. Cott Public Vows. p. 99.Bardaglio Peter W. Reconstructing the Household: Families, Sex, and the Law in the Nineteenth-Century South. Chapel Hill: 1995. pp. 48–64.pp. 176–89. [Google Scholar]

35. Blassingame John W. Black New Orleans, 1860–1880. Chicago: 1973. p. 203. [Google Scholar]

36. On the initial tolerance for interracial couples see Sommerville Diane Miller. Rape and Race in the Nineteenth-Century South. Chapel Hill: 2004. pp. 185–87. [Google Scholar] See also Gullickson Aaron. Black/White Interracial Marriage Trends, 1850–2000. Journal of Family History. 2006 July;31:289–312. esp, 298. [Google Scholar] It is impossible to estimate precisely the number of white women married to or cohabitating with black men. There are records of interracial marriages in the census, but social pressures probably led many interracial couples to claim that they belonged to the same race. Of the more than 8,000 southern white women from 20 through 24 years old whose marriages are recorded in the 1850, 1860, 1870, and 1880 IPUMS census samples, only 11 are recorded as living with a black or mulatto husband. Peter Bardaglio’s examination of the manuscript returns for the 1860 census of South Carolina reveals only 61 co-residing interracial couples “in the entire state (outside the city of Charleston).” Bardaglio Reconstructing the Household. p. 60. In more cosmopolilan New Orleans, Blassingame counted 29 white women in the 1880 census married to black men (out of a total population of 84,801 white women) and 176 white population men married to black women (out of total of 74,848 white a men). Blassingame Black New Orleans. pp. 206–7.pp. 240–41. Aaron Gullickson’s examination of long-run trends in black/white intermarriage indicates that intermarriage patterns were responsive to the broader climate of race relations. The frequency of interracial unions declined after 1880 with the passage of Jim Crow laws in the South and segregation in the North, increased with the waning of these systems in the 1930s and 1940s, and increased rapidly in the post-civil rights era. Gullickson Black/White Interracial Marriage Trends. p. 298.p. 303.p. 309. [PubMed]

37. Cumming Journal of Hospital Life. :117. [Google Scholar]Beers Fannie A. Memories: A Record of Personal Experience and Adventure during Four Years of War. Philadelphia: 1889. pp. 205–6. [Google Scholar]

38. McGuire Diary of a Southern Refugee. p. 330.

39. Cumming Journal of Hospital Life. :105. [Google Scholar]

40. For a discussion of southern women’s postwar writings and politics, see Hilde Libra. Ph.D. dissertation. Harvard University; 2003. ‘Worth a Dozen Men’: Women. Nursing, and Medical Care during the American Civil War. chaps. 12–15. [Google Scholar]

41. Smith Daniel Scott, Hacker J. David. Cultural Demography: New England Deaths and the Puritan Perception of Risk. Journal of Interdisciplinary History. 1996 Winter26:367–92. [Google Scholar]Slovic Paul. Perception of Risk: Reflections on the Psychometric Paradigm. In: Krimsky Sheldon, Golding Dominic., editors. Social Theories of Risk. Westport, Conn.: 1992. pp. 117–52. [Google Scholar]

42. Censer Reconstruction of White Southern Womanhood. p. 32.

43. Published census tabulations of marital status begin in 1890, well after the end of the war. Census samples from IPUMS, however, can be used to construct estimates of nuptiality by region and section beginning in 1850, although limitations in particular censuses pose significant methodological challenges. Most obviously, the censuses from 1850 through 1870 did not record individuals’ marital status, which is required to determine the timing and incidence of marriage. Fortunately, the censuses contain enough information—surname, sex, age, and position in household—to permit an inference as to whether individuals had ever been married. Tabulating by age the percentage of men and women who ever married makes it possible to calculate the timing and incidence of individuals’ first marriage in each census year. See Hacker Economic, Demographic, and Anthropometric Correlates of First Marriage for more details the imputation procedure.

44. Marriage register data in Ohio and Alabama suggest a short-term postwar marriage boom among men who had survived the war. Logue Larry M. To Appomattox and Beyond: The Civil War Soldier in War and Peace. Chicago: 1996. pp. 88–89.pp. 108 [Google Scholar]

45. Various definitions of “region” had little effect on the results. Therefore this study uses the standard census bureau definition of region in order to increase comparability of the results with those of other studies. Current census definitions of region are as follows: New England—Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, Vermont; Middle Atlantic—New Jersey, New York, Pennsylvania; East-North Central—Illinois, Indiana, Michigan, Ohio, Wisconsin; West–North Central—Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota, South Dakota; South Atlantic—Delaware, District of Columbia, Florida, Georgia, Maryland, North Carolina, South Carolina, Virginia, West Virginia; East-South Central—Alabama, Kentucky, Mississippi, Tennessee; West-South Central—Arkansas, Louisiana, Oklahoma/Indian Territory, Texas; Mountain—Arizona, Colorado, Idaho, Montana, Nevada, New Mexico, Utah, Wyoming; Pacific—Alaska, California, Hawaii, Oregon, Washington. The category “northern census regions” is defined here as the combined New England, Middle Atlantic, East–North Central and West–North Central census regions; and the category “southern census regions,” as the combined South Atlantic. East–South Central, and West-South Central census regions. “Northern-born” individuals were born in a northern census region state; and “southem-born” individuals, in a southern census region state.

46. Hajnal John. Age at Marriage and Proportions Marrying. Population Studies. 1953 November;7:111–36. [Google Scholar]

47. The marital sex ratio used here is that proposed by Shuzhuo Li, et al. Estimation of the Number of Missing Females in China: 1900–2000. Chinese Journal of Population Scierue. 2005;4:2–11. [Google Scholar], which is an extension of the measure suggested in Tuljapurkar Shripad, Li Nan, Feldman Marcus W. High Sex Ratios in China’s Future. Science. 1995 February;267:874–76. [PubMed] [Google Scholar]

48. Ransom Roger L. Conflict and Compromise: The Political Economy of Slavery, Emancipation, and the American Civil War. New York: 1989. pp. 227–28. [Google Scholar]

49. Kenzer Kinship and Neighborhood in a Southern Community. p. 218. From the beginning of 1850 through April 1861, an average of 1,81 sons was mentioned in their father’s will. The average fell to 1.30 in the period between the beginning of April 1865 through the end of 1880.

50. p. 98. Ibid.

51. In Roger Ransom’s sample of farms in eight southern counties, the average total wealth fell from $22,819 in 1860(all farms) to just $3.168 in 1870 (farms operated by whites). Although there are no reliable estimates for income. Ransom’s analysis of the value of agricultural output suggests that incomes of white farmers fell about 35 percent between 1857 and 1879. Ransom Conflict and Compromise. p. 228.

52. Hacker Economic, Demographic, and Anthropometric Correlates of First Marriage. p. 310.p. 319.

53. Because the census did not record individuals’ marriage location, duration of marriage, and migration history, regional estimates are based on an unknown proportion of men and women who lived some portion of their lives outside the region and neglect individuals who lived some portion of their lives in the region and subsequently left. Sectional estimates are much less likely to suffer this form of bias; as many historical demographers have noted, most internal migration in the nineteenth-century United States occurred along East-West lines. Steckel Richard H. The Economic Foundations of East-West Migration during the 19th Century. Explorations in Economic History. 1983 January;20:14–36. [Google Scholar]

54. In his community study of Orange County, North Carolina, Robert Kenzer observed that the average age reported in the county’s marriage register was about the same in the 1870s as it was before the war. Kenzer Kinship and Neighborhood in a Southern Community. p. 217. Orville Vernon Burton also found the average age at marriage for whites in Edgefield County, South Carolina, “nearly identical” before and after the Civil War. Elsewhere, however. Burton reports that the average age at marriage for white men fell by 0.47 years. Burton My Father’s House Are Many Mansions. p. 264. 409n6.

55. Unfortunately, no census sample exists for the 1890 census (the original manuscript records were destroyed). Where necessary, estimates were obtained by interpolating between the 1880 and 1900 censuses.

56. At the time of the 1860 census, 83 percent of white men from 30 through 39 years old and 88 percent of women of the same age had already married.

57. Kenzer Robert. The Uncertainty of Life: A Profile of Virginia’s Civil War Widows. In: Cashin Joan E., editor. The War Was You and Me: Civilians in the American Civil War. Princeton: 2002. pp. 112–35. [Google Scholar]

58. On Union widows see Holmes Amy E. ‘Such Is the Price We Pay’: American Widows and the Civil War Pension System. In: Vinovskis Maris A., editor. Toward a Social History of the American Civil War: Exploratory Essays. New York: 1990. pp. 171–95. [Google Scholar]; and McClintock Megan J. Civil War Pensions and the Reconstruction of Union Families. Journal of American History. 1996 September;83:456–80. [Google Scholar] For brief mentions of Confederate widows, see Faust Mothers of Invention. pp. 149–50.; and Kenzer Kinship and Neighborhood in a Southern Community. pp. 97–99.

59. For more on the “Gould” sample, see Costa Dora L. Journal of Economic History. Vol. 64. Mar, 2004. The Measure of Man and Older Age Mortality: Evidence from the Gould Sample; pp. 1–23. [Google Scholar]; and Hacker Economic, Demographic, and Anthropometric Correlates of First Marriage. pp. 314–18.pp. 331–40. If these men were representative of all men in the combined Union and Confederate forces, approximately 175,000 women ( 105,000 in the North and 70.000 in the South) were widowed by the war. Although she relied on a different set of assumptions, Amy Holmes estimated a similar number of Union widows. Holmes Such Is the Price We Pay. p. 174.

60. Robert Kenzer has estimated that 20 percent of Virginia men who fought in the war were married. It is unclear, however, where he obtained the estimate. Kenzer Uncertainty of Life. p. 113.

61. Orville Vernon Burton has pointed out that the number of women widowed by natural causes in Edgefield, South Carolina, between 1865 and 1880 dwarfed the number of women widowed by the war. Burton My Father’s House Are Many Mansions. 413n42. Although he is certainly correct that the vast majority of widows in 1880 were widowed from causes other than the war, it is likely that a significant proportion of younger widows lost husbands in the conflict. Widows from 40 through 49 years old in 1880, for example, would have had husbands near the age of 30 in 1865. Given typical nineteenth-century mortality rates, 15 percent of white men aged 30 could be expected to die before the age of 45, approximately equal to the risk of death for Confederate soldiers during the war. Further assuming that half of the women in their twenties widowed by the war remarried (and remained married until the 1880 census) suggests, very roughly, that about one in four currently widowed southern white women from 40 through 49 years old in the 1880 census were widowed during the war. For remarriage rates of women widowed by the war by age, see Kenzer Uncertainty of Life. pp. 125–27. Survival estimates for white men aged 30 were obtained from the 1870 white male life table in Haines Michael R. Estimated Life Tables for the United States, 1850–1910. Historical Methods. 1998 Fall31:159. [Google Scholar]

62. In Edgefield, women marrying in 1870 married fewer younger men than in 1860. Burton My Father’s House Are Many Mansions. p. 264.

63. Finite population effects mean that sex ratios can be imbalanced in a finite (i.e., small) population simply by stochastic sampling effects. The larger problem is of local geographic imbalances. Since the social organization of military regiments was done along geographic lines and there was tremendous heterogeneity in mortality by regiment, many local populations probably experienced acute shortages despite the global excess of men. Documenting such effects will require detailed community studies employing samples of the census that are denser than the samples currently available, together with a variety of other sources.

64. Interesting Facts of the Census. Washington Post. 1881 February 21;:2. [Google Scholar]

65. The Domestic Millennium. May 22, 1881. p. 2. ibid.

66. Henry Louis. Perturbations de la Nuptialité Resultant de la Guerre 1914–1918. Population. 1966 March-April;21:273–332. [Google Scholar]

67. This recommendation seconds a recent comment by Robert Kenzer. “Despite the heightened interest in the social history of the Civil War during the last two decades, especially the impact of the conflict on women,” Kenzer has observed, “no scholar has written even an article-length study on Confederate widows.” Kenzer Uncertainty of Life. pp. 112–13. For an excellent example of the demography of widowhood, female householding, and poverty in the late eighteenth century, see Smith Daniel Scott. Female Householding in Late Eighteenth-Century America and the Problem of Poverty. Journal of Social History. 1994 Fall28:83–107. [Google Scholar]