When James C. Bonner published his seminal essay “Profile of a Late Antebellum Community” in 1944, an examination of Hancock County, Georgia in the 1850s, antebellum southern historiography was in the throes of an “agrarian revolution” instigated by Frank L. Owsley and the Young Turks at Vanderbilt University. The Owsley school was contesting the traditional interpretation of a planter‑dominated antebellum southern society, using manuscript census examinations of real property ownership to prove the existence of a ubiquitous middle‑class yeoman southern society. Although indebted to the path‑breaking work of Owsley and his students for opening up manuscript census records as primary source material, Bonner’s representative community “test case” nevertheless quite directly rejected Owsley’s “yeoman” society thesis.
Instead of a simple society of small landholders, Bonner found a complex society of planters and farmers, professionals, merchants, tradesmen, overseers, farm laborers, and factory workers. Instead of a yeoman democracy, he found rich planters forcing tenant farmers off the land and into either agricultural wage labor, factory work, or migration. Clearly the society that Bonner had uncovered was neither U. B. Phillips’s two class planter‑poor white society nor Owsley’s middle-class yeoman society, but a society composed of many socioeconomic groups, each deserving additional attention.
Unfortunately, later southern historians, while forever citing Bonner’s contribution, have failed to follow up the complexity he uncovered. Instead, for the next thirty years, analyses of southern society continued to revolve almost entirely around planters and slaves. In the late 1970s when the “new social history” finally started penetrating the antebellum South, historians rediscovered Owsley’s yeoman and tenant farmers. But of the non‑farm groups identified by Bonner we know next to nothing. Much of our ignorance about these “invisible people” derives from our lack of knowledge of southern towns, villages, and hamlets in which many of them lived and worked. This essay takes a second look at the non‑farm town sector of Hancock county in the 1850s and examines how inclusion of towns in the picture of antebellum southern society necessarily redirects many of the questions currently being asked about the antebellum South.
Most social historians of the South totally ignore the non‑farm and town‑based population of the South. Indeed Bonner himself paid short shrift to the non‑farm sector of the community and makes no mention of Hancock’s villages. J. William Harris’s recent study of the plain folk and gentry of Augusta’s hinterlands, in which he includes Hancock County, ignores any towns and dismisses “the opinions of mechanics [i.e., tradesmen]…in an overwhelmingly rural and agricultural society” (77), and writes off factories and factory workers as contrary to republican ideology (33). In his study of the Georgia upcountry, Steven Hahn found that “the towns were little more than glorified hamlets having populations numbering in the low hundreds and distinguished solely by the presence of small trading establishments and the county courthouse. And their impact on local affairs was not extensive” (97).[ref]Hahn does pay some attention to non-farm peoples, acknowledging that “some laborers found jobs in the tiny Upcountry towns and villages” (22) and that artisans, many of whom owned land and combined the trade with farming, were important in the “grid of local trade” (21‑22,71). But Hahn otherwise finds little room for non-farm peoples in the social structure of his yeoman society.[/ref] Robert Kenzer’s self‑sufficient Orange County, North Carolina farmers little needed towns. “Although the few towns had a more diverse occupational composition than the countryside, they failed to provide an alternative source of employment for the county’s farmers” (29). Harry Watson, who places Fayetteville at the center of his sociopolitical study of Cumberland County, North Carolina, nevertheless writes off southern towns as “little more than the banks, stores, shipping facilities, and factorage houses that were necessary to market the staples of the region” (287).
On this point, the “new economic historians” are in complete agreement with their antebellum bedfellows. For Douglass North, the South’s lack of towns and locally oriented industries and services resulted as a natural outcome of the cotton and slave economy (130‑133). Anderson and Gallman state that, in the South, non‑agricultural activity was insignificant and “the centers of population concentration in which such entities normally locate were few, small, and widely separated. There was no rich history of farm and village or town interaction” (42).
The work of southern colonial urban historians and historical geographers has gone generally unincorporated in the general social histories of the antebellum period. Ernst and Merrens stress the functional role towns like Camden, South Carolina played in the southern colonial economy. Earle and Hoffman, based on evidence from one gazetteer, find distributed from Maryland to South Carolina 262 identifiable cities, towns, and villages containing altogether 16,000 houses and over 100,000 persons (59‑61). On the other hand, antebellum urban historians, enamored with southern “cities,” do little more than acknowledge the existence of such “rural” towns and villages (Goldfield; Huffman).
Not all historians of the antebellum South ignore the role of towns and non‑farm employment. Political historians, at least since Abernethy in 1926, have stressed the importance of towns in antebellum politics and the influence of politicians, attorneys, merchants, bankers, and professionals who lived there (Abernethy 1926, 521‑522; 1930, 530; Sellers, 341‑342). More recent political historians have identified political divisions between town and country (Thornton, 41‑42; Watson; Jeffrey 148; Folsom, 377‑378).
More directly, a few social historians have noted the importance of the town in the social and economic life of the South. According to James Oakes, “Most of the slaveholders’ social, legal, political, and religious activity took place in the small town. But with its merchants, shippers, and country stores, the town served primarily as the nucleus of a vast and decentralized marketing system that tied the products of slave labor into the international economy” (92). John Hebron Moore devotes over a third of his book on antebellum Mississippi to towns, townspeople, and non‑farm activities. John T. Schlotterbeck stresses the importance of the town of Orange Court House to the “social economy” of Orange County, Virginia and the “growing service economy” which “created a surprisingly large non‑farm population for an area that had neither large towns nor factories” (15‑16). William L. Barney’s examination of Dallas County in Alabama’s Black Belt stresses the importance of the towns of Selma and Cahaba and the rapid socioeconomic changes occurring in an antebellum society even more complex than Hancock County.
Perhaps Hancock County is not a representative county for a study of either Georgia or the South. Indeed many southern historians have found much smaller percentages of nonfarm peoples than Bonner. The percentage of nonagricultural occupations among all occupations reported in recent studies for Georgia counties in 1860 ranged from 7% in Glasscock (bordering on Hancock), 10% in Hart, 15.4% in Jackson, 16.7% in Carroll, 17% in Taliaferro, and 17% in Edgefield, S.C. compared to Hancock’s 29.2%. (Hahn 21; Harris 21). However, as Table I indicates, Hancock is not at all that exceptional. The percentage for the entire state of Georgia was 31.4%, reflecting an occupational structure more like that identified by Schlotterbeck in 1850 Orange County, Virginia (28.0%), than those reported in recent studies of the Lower South.[ref]Table I lists for comparison both Bonner’s older calculations and new calculations computed from a data base of the 1860 Hancock County population census developed for this study. The differences between the two are not very significant, hopefully lending greater credence to both studies. Unless otherwise noted, all computations presented in this paper are derived solely from the new data base. For Hancock County, the following occupations were included under each occupational group:
| Occupation | Hancock (Bonner) (%) | Hancock (Database) (%) | Georgia (%) |
| Farmers | 36.6 | 33.6 | 45.6 |
| Farm Laborers | 20.1 | 21.9 | 19.9 |
| Professionals | 4.9 | 5.3 | 5.1 |
| Merchants | 2.9 | 3.2 | 5.9 |
| Overseers | 14.1 | 14.0 | 3.2 |
| Factory Hands | 9.7 | 10.9 | 2.0 |
| Tradesmen | 11.7 | 11.1 | 18.4 |
Overall, Hancock has more overseers and factory workers, and fewer planters, farmers, and tradesmen than the state as a whole but the overall match seems quite good.
Such analysis of occupational categories should be entered into with extreme caution. Census takers had a great deal of leverage in assigning occupations, making occupational comparisons quite hazardous. The census taker for Hancock identified every farmer who farmed his own land and most tenants as “planters” as long as they had enough agricultural produce to qualify for the agricultural census. No distinctions were made according to either personal or real property. However, some farmers who combined other occupations‑‑professionals, merchants, tradesmen‑‑were often listed by the non‑farm occupation.
Census takers in many Georgia counties in 1860 used the occupational category “laborer” although the census instructions explicitly stated that census takers were to clearly define “farm laborers” from other types of laborers. In Hancock county the term “laborer” was used for only seven individuals, three of whom were probably factory hands and the other four probably farm laborers. For the state calculations listed above, the 11,272 laborers (7.2% of total occupations) in Georgia were assumed to be “farm laborers” but perhaps many of these worked in factories, trades, services, transportation, or a combination of these.
Also problematic are those individuals for whom no occupation is given. According to the census instructions, all males over the age of fifteen should be given an occupation yet 217 males over age 15 have none listed. Although the Hancock census taker did not use the occupational category “student,” many of these individuals are obviously at least part‑time students, as indicated in the census column noting school attendance. (For most cases, the total number in the household attending school was listed next to the head of household making exact identification of students difficult.) Eliminating likely students still leaves 126 individuals whose occupation is unaccounted for including 56 heads of household. Perhaps many of these individuals truly have no occupation as did Frank Barbrough, a thirty year old, married, father of two living in Linton who had listed “Nothing” for his occupation. Others like J. M. Stanford, a prosperous merchant, were simply oversights.
Of the 76 non‑heads with no occupation listed, a full 48.7% are living in the households of “planters” (two‑thirds of this group share the same surname as the head and are thus likely sons) and another 10.5% are in the households of overseers or farm laborers. Undoubtedly the bulk of this group should have been classified as either overseers or farm laborers. The other non‑heads are in households headed by merchants (5.3%), professionals (13.2%), and tradesmen (10.5%), half of whom share the same surname as the head. This group could represent clerks, apprentices, laborers, or boarders. Most likely, more thorough classification of the occupation of all males over 15 would have the net effect of significantly increasing the percentage of farm laborers and overseers and slightly reducing the percentage of all other groups. If all heads of household, and non‑heads living in houses headed by planters, overseers, or farm laborers were classified as “farm laborers,” this would raise the percentage of farm laborers in the community from 21.9% listed in Table I to 30.0% and reducing the percentage of farmers and planters from 33.6% to 30.5%.
Occupations for women and children create additional problems. Although women and children can be eliminated from analysis, this undoubtedly distorts the occupational structure of antebellum southern communities. In Hancock County, among other occupations, 46 female heads of households were listed as “Planter” while 98 other female heads had no occupation. In some families, children as young as 8 years old listed occupation as “Farm Laborer” and most of the teenagers living in Montour Village, the factory town, listed a factory occupation. Older students, while living at home, undoubtedly performed many of the activities of farm laborers.[ref]In the analysis which follows, women with no occupation listed and all children under age 16 were excluded.[/ref]
Although one could forever argue definitions of hamlets, villages, towns, and cities, what constitutes sufficient population and diversity, and whether rural villages are really that different from the surrounding countryside, such definitions will always be arbitrary. The least limiting defintion for a social scientist, but perhaps most frustrating for an historian, assumes a continuum of places. Hope Tisdale Eldridge sees urbanization as “a process of population concentration” without worrying about cutoff points (338). Darrett Rutman envisions “rural and urban as simply two extremes on a continuum, with the Chicagos, New Yorks, Birminghams, and Londons at one end, farm life at the other, and a range of places in between” (4). This study will also assume such a continuum.
Nevertheless, the terms hamlet, village, town, and city have historical value and arbitrary definitions are sometimes useful for organizing research. The definition might arise out of theory, out of the historical data, or out of a combination of the two. The most critical definition for the study of urbanization is that for “hamlet,” the smallest “urban,” non‑farm unit deemed worthy of study. In a most thorough analysis of these smallest units, Glenn Trewartha defined a hamlet as at mimimum four active residential units (at least two non‑farm) with two additional active functional units (residential, business, social, etc.) with a total of five actively used buildings and maximum linear distance between the outermost buildings not to to exceed one‑quarter mile. For general purposes, he set a population minimum of 20 and, to separate hamlets from villages, a more arbitrary population maximum of 150 or 38 residences (37). Although Trewartha based his definition on 20th century Wisconsin, his understanding of the minimum requirements for a hamlet make a good starting point from which to examine the antebellum South with the expectation of possible downward revisions.
For identifying hamlets, villages, and towns in the antebellum United States, the best starting place is the federal register of post office agents which lists name of post office, county, name of postmaster, compensation (in dollars), and net proceeds (in dollars) in every odd year starting in 1817. As Richard Helbock has shown, compensation correlates significantly positive with town population (9‑14). Listed in order of increasing compensation, the post offices for Hancock County in 1859 were: Shoals of Ogeechee ($20.36), Island Creek ($26.23), Powelton ($43.50), Culverton ($69.05), Mount Zion ($83.28), and Sparta ($436.47). A post office at Linton, an academy boom town built between 1857 and 1860, had a post office by early 1860 (Central Georgian 7 Mar 1860).
A most useful source for descriptions of some of these small “urban” units are the series of gazeteers published by Adiel Sherwood.[ref]One potentially powerful untapped source for antebellum urban research is archaeological investigation, starting with simply walking through and mapping the present town site. These antebellum villages did not exist solely in archival sources and the physical layout and spacing of many of these smaller towns remains basically unchanged, as Trewartha noted in Wisconsin (38‑40). Many of the antebellum buildings that these historical people lived and worked in still exist in Hancock (Moore 16‑21).[/ref] For many of the smaller places, Sherwood often mentions the number of houses or families. In 1860, he identifies possible hamlets at Shoals of Ogeechee (“some houses and some shops” as well as a factory), Mayfield (“a post village of four houses”), and Culverton (“a post‑office, five houses, and a good school”). Clustered at a distinctly higher level are three other places, here arbitrarily called “villages”: Linton (25 families and two‑story brick school building), Powelton (30 houses, two academies, Baptist and Methodist churches), and Mount Zion (“another eduational town” with no additional imformation although White in 1849 reports a population of 200). At another hierarchical level lies the “town” of Sparta, the county seat, which according to the 1854 state census had a population of 1024 (586 whites, 16 free blacks, and 422 slaves).[ref]Sherwood misses Island Creek, a post‑village with an academy. Both Mayfield and Shoals of Ogeechee spread on both sides of the Ogeechee River, divided between Warren, Glascock, and Hancock counties. The Mayfield post office was actually located in Warren county and the Hancock side of the river was more usually known as Shiver’s Mills or Rock Mills, and later would be renamed Jewell. The post office compensation analysis generally supports Sherwood’s description except for Culverton whose post office business indicates a much larger village.[/ref]
What kind of people lived in these hamlets, villages, and towns? Unfortunately, manuscript census records in the antebellum South rarely delineate towns explicitly. But by combining information from manuscript census records, tax digests, local newspapers, and county histories, usually clusters of non‑farm employed households and other townspeople can be fairly well identified for all but some of the smallest hamlets.[ref]For descriptions of similar efforts, see Edward Muller and Robert Kenzer.[/ref] Like many of the older Georgia counties, Hancock is organized by county subdivisions called militia districts. (Unfortunately the 1860 Hancock manuscript census was not organized by militia districts, complicating the following procedure a bit.) Using contemporary maps, each town, village, or hamlet from the post office records and gazeteers were first located in the proper militia district. Each militia district contained at most only one urban unit, simplifying the analysis (althouggh Sparta shared two districts).
Next individuals from the 1860 Hancock tax digest were matched with individuals of the same name on the manuscript census, particularly noting those individuals listed as owning town property on the tax digest. The newspaper and county history were searched for any mention of a link of an individual with a particular place, whether town or country. Postmasters taken from postal records were also located.
All of the individual information was recorded on a copy of the manuscript census and a most probable route of the census taker determined by identifying serial patterns of places and militia districts. The towns were distinguished by identifying a distinct serial cluster of at least four households of which over 50% of the households contain individual(s) with either nonagricultural occupations, town property, or links to a particular town from other records, with almost all (greater than 90%) of those matched from the tax digest recorded as taxed in the same militia district. The clusters were kept to a minimum by beginning and ending the series with a household containing individual(s) taxed in the same district (if matched) with either nonagricultural occupations, town property, or links to a particular town from other records. The town was then identified by linking the militia district with the proper town, village, or hamlet.[ref]This analysis assumes that the census taker tended to record most of the people living in a town, village, or hamlet in one or two continuous sessions so that townspeople appeared in distinct serial clusters on the pages of the manuscript census. The towns, villages, and hamlets identified by 1860 Hancock County manuscript census household numbers are: Culverton (258-265), Linton (750-773), Montour Village (82-130), Mount Zion (133-148), Powelton (166-193), and Sparta (1-81,131-132,646-663).[/ref] In this study, unfortunately, neither Shoals of Ogeechee, Mayfield, nor Island Creek could be positively identified as a cluster in the manuscript census records and, though these places most likely qualify as hamlets, they were dropped from further analysis. However, added as a separate unit is Montour Village, a separately incorporated factory town on the outskirts of Sparta.[ref]Following the described procedure, Montour Village would be included with Sparta for both share the same militia district are the two towns represent an unbroken series in the manuscript census. However, the factory town could be clearly delineated by the occupations peculiarly associated with a cotton factory and other characteristics such as no personal or real property.[/ref] Table II lists the number of households, dwellings, and total population for the year 1860. Of all individuals with listed occupations, 33.0% resided in towns. Towns accounted for approximately one quarter of the white population (26.7%) and white households (25.8%).
| Town | Households | Dwellings | Population |
| Culverton | 8 | 6 | 30 |
| Linton | 24 | 23 | 120 |
| Montour Factory | 49 | 45 | 245 |
| M0unt Zion | 16 | 10 | 102 |
| Powelton | 28 | 26 | 111 |
| Sparta | 102 | 91 | 431 |
Analyses of the people who lived in these “points of concentration,” the reasons for the creation of such “points,” and the activities that went on there, indicate that the southern town, village, or hamlet was a phenomenon integrally linked and yet distinct from the surrounding countryside. All of the identified urban units in Hancock county listed in Table II (apart from Montour Village) had its merchant, postmaster, teacher, physician, and at least a couple of tradesmen and wealthy planters. Table III shows that the occupational structure of the towns (combining all hamlets, villages, and towns) differed significantly from that of the countryside with a much higher percentage of professionals, merchants, factory hands, and tradesmen in the towns. Table IV shows the percentage of individuals of each occupational category living in towns. Whereas only about a tenth of the county’s planters, farmers, overseers, and farm laborers lived in towns, nine‑tenths of the merchants and factory hands lived there. The towns also contained over half of the county’s students (including only males over 15), tradesmen, and professionals. Finally, the one‑quarter of individuals of unknown occupation living in towns reflects the general county population distribution between town and country. Table IV also lists a few other specific occupations which shows that all attorneys and most carpenters, physicians, Methodist ministers, and teachers lived in towns, while Baptist ministers were evenly divided between town and country, and most shoemakers resided in the country.
| Occupation | Town (%) | County (%) |
| Farmers | 13.5 | 42.7 |
| Farm Laborers | 6.8 | 28.7 |
| Professional | 11.1 | 2.6 |
| Merchants | 9.2 | 0.6 |
| Overseers | 5.5 | 17.9 |
| Factory Hands | 32.9 | 1.0 |
| Tradesmen | 20.9 | 6.6 |
| Occupation | N | Percent Town |
| Farmers | 352 | 12.5 |
| Farm Laborers | 229 | 9.6 |
| Professionals | 55 | 65.5 |
| Merchants | 34 | 88.2 |
| Overseers | 147 | 12.2 |
| Factory Hands | 114 | 93.9 |
| Tradesmen | 116 | 58.6 |
| Students | 91 | 54.9 |
| Unknown | 126 | 25.4 |
A breakdown by place of birth for individuals with occupations in Table V shows that, although Georgia‑born individuals filled the bulk of both urban and rural occupations of Hancock County, the urban population was much more diversified. Table VI shows more directly that only Georgians and Virginians had a relatively higher propensity for living in the country, while migrants from North and South Carolina were much more likely to live in towns. Most foreigners (62.2%) and almost all northerners (89.5%) were townspeople.
| Place of Birth | Town (%) | Country (%) | Total (%) |
| Georgia | 78.9 | 90.7 | 86.9 |
| Virginia | 2.5 | 2.8 | 2.7 |
| North Carolina | 5.2 | 2.5 | 3.3 |
| South Carolina | 2.9 | 1.4 | 1.9 |
| Northern States | 4.2 | 0.2 | 1.5 |
| Foreign | 5.7 | 1.4 | 2.9 |
| Place of Birth | N | Town (%) |
| Georgia | 1098 | 29.2 |
| Virginia | 34 | 29.4 |
| North Carolina | 42 | 50.0 |
| South Carolina | 24 | 50.0 |
| Northern States | 19 | 89.5 |
| Foreign | 37 | 62.2 |
Although most economic and social historians stress the economic function of towns, evidence from Hancock county indicates southern towns came into existence mostly for social reasons. Almost every town in antebellum Hancock was founded in conjunction with an academy.[ref]Although Sparta was founded in 1795 as the county seated for Hancock County and did not have its own academy until 1818, academies, particularly the renowned Male and Female Institute started in 1832, played an important role in the social life of the town in the mid-nineteenth century (Boogher 368). The Sparta Lyceum organized in 1856 also acted as an educational magnet for a while drawing large audiences in the late 1850s (Central Georgian 29 May 1856).[/ref] Powelton (1815), Mount Zion (1823), Farmersville (1832), Island Creek (1856), Linton (1857), and Culverton (1859) were all founded around academies (Boogher 368-69; Smith 23-26,101-113).[ref]Some other small hamlets, Rock Mills and Shoals of the Ogeechee, were founded around mills on the Ogeechee River, Rock Mills was immediately across the Ogeechee River from Mayfield in Warren which started an academy in 1843 with the support of many Hancock families (Boogher 376; Smith 101). Two other academies, the Hancock County Academy (1794) and County Line Academy (1838) apparently were not associated with any town development (Boogher 368-369).[/ref]
Sherwood, writing in 1827, noted that in Mount Zion, “the [male and female] academies were first erected, and the inhabitants crowded around them to enjoy the literary advantages which they proferred to their children” (79). Similarly, in Powelton, the Male and Female Academies, that had just recently “attached a Library and now a new Chemical Apparatus,” “were the attactives which drew the people around them that their children might enjoy the advantages of education” (Sherwood 1827,88).[ref]The first five principals of Powelton School, opened in 1811, came direct from Bowdoin College in Maine (Smith 111). Richard Malcolm Johnston, a long‑time resident of Hancock County, principal of Sparta Male and Female Academy in the early 1850s (Southern Recorder 18 Nov 1851), and later a famous educator and novelist, as a lad attended the school in Powelton near his birthplace and “preserved his memoirs” of his schoolboy days in the novel Dukesborough Tales (Preface, 1892 edition).[/ref]
Although Hancock County educational success may not represent the experience of other Georgia counties, evidence in Sherwood’s 1827 Gazetteer suggests otherwise. Hermon, a village in Oglethorpe County, “like many others in the state, was built for the purpose of supporting the academy, and enjoying the advantages of education” (62). In Salem in Clarke County, the male and female academy “may be said to have created the village,” even though “an extensive tannery” supplying “a large section of country with leather” was later established (Sherwood 1827,94). Boogher acknowledges that the establishment of a school often attracted “sufficient settlers to form a village” (100).
Altogether Boogher identifies 583 academies chartered in 97 of Georgia’s 132 counties between 1783 and 1860, although he does not identify whether these were “town” or “country” academies. In the county description of 1860, Sherwood identifies 95 academies, institutes, high schools, or colleges in 55 Georgia counties, almost all of which he identified with one town or another.[ref]Sherwood also summarizes school statistics from the “Governor’s Message” which shows, of 102 counties reporting, a total of 108 secondary institutions.[/ref] Fourteen counties‑‑Bibb, Cass, Chattooga, Coweta, Elbert, Floyd, Greene, Houston, Monroe, Newton, Putnam, Spalding, Talbot, and Troup‑‑were distiguished like Hancock with the presence of three or more secondary institutions and might all be termed “educational” counties. Analysis of the 1860 manuscript social statistics census schedule of Georgia indicates that Sherwood’s figures are significant underestimates for the census actually lists a minimum of 235 colleges, institutes, academies, high schools, etc. in 79 counties, with 35 counties containing three or more secondary institutions.[ref]This analysis includes all educational institutions apart from common schools, eliminating those counties like Putnam where every educational institution was called an “academy.”[/ref]
Academies were also businesses. Hancock leaders stressed the positive economic as well as social benefits that the academies brought to the county. As Sherwood notes for Fayetteville in Fayette County:
Since the people have patronized and built up good schools here of high grade, the town has been growing. No internal town will increase much without the aid of schools, and no people need to wonder or complain that their place is small, if they fail to nurture and encourage schools. (1860,65)
The absence of a railroad perhaps forced Hancock’s citizens to seek alternative investment outlets like academies even more. Certain academies, like the Sparta Male and Female Academy, had the privilege of operating lotteries which attracted attention from as far away as New York City (Boogher 106-108). State school appropriations at times brought in additional revenue for some academies (Boogher 70-78). Planters from all over Georgia and the rest of the South sent their children to Hancock’s highly respected academies. In 1856, of the students at the Villa School, operated in conjunction with the Mount Zion Academy, nearly three quarters of the 75 students were from out of county (Smith 26‑27). An academy meant boarding houses which brought significant money into the county at a normal rate of ten dollars per month per student, along with income from tuition averaging around forty dollars per year per student (Smith 69‑69; Central Georgian 5 Dec 1860).
Hancock leaders went about promoting their schools with the same gusto that they tried to get a railroad. An advertisement in the Savannah Republican addressed the anxieties of parents to protect their children from the “vice and immorality so common in large towns and cities,” recommended the Sparta Male and Female Academy, “located in a quiet, but pleasant and healthy village, and among a people marked for their intelligence , morality, and high christian character”. The ad also noted the “cheapness and accessibility‑‑only three and half hours ride from Milledgeville…by stage” (Central Georgian 7 Jan 1858).[ref]Many examples of advertisements by academies are listed in Boogher (281-320).[/ref]
Like railroads and factories, however, academies could go belly‑up and often did, even such a renowned institution as the Male and Female Academy which sold off the Male Academy building to the highest bidder in 1859 (Southern Recorder 4 Jan 1859; Boogher 99). Earlier “a model school” for ladies in Sparta had “attracted considerable attention,” “but it had its day and faded away like all flowers” (Sherwood 1860, 77). Unless the town diversified, an academy in trouble meant a town in trouble. Powelton declined as the school fell out of favor. Farmersville (associated with hamlet of Devereux’s Store) died in the 1850s with the financial failure of its academy but in the same decade the villages of Island Creek, Linton, and Culverton sprang to life around their schools.
Linton provides an excellent example of an educational boomtown, growing from a wooded hill in 1857 to a full‑fledged village in 1860. In October 1857, the Washington [Baptist] Association of Baldwin, Hancock, and Washington counties “voted to build ‘a permanent school of high grade'” and associated village on land donated by Dr. John Stone near Darien Baptist Church (Smith 105; Central Georgian 15 Oct 1857). Many in the county doubted that such a “dubious enterprise” would succeed, remembering that the “Farmer’s Academy was projected and failed in this county, under somewhat similar circumstances.” But supporters believed the Farmer’s Academy failed because “that school was simply for the benefit of a few families, while the Washington Institute (the official name of the school) “embraces several counties” and had the official backing of the Baptist Association (Central Georgian 14 Jan 1858). Originally called Buffalo, the name of the town was soon changed to Linton in honor of Judge Linton Stephens (half‑brother of Alexander Stephens), a resident of Hancock County and a strong supporter of the Insititute. A total of twenty four residential lots of 3 5/8 acres, twelve on either side of a mile‑long main street centered around the school, were laid out and sold off on 23 November 1857 (Smith 105; Central Georgian 3 Dec 1857). The school opened in February 1858.
The town attracted people of some means who rapidly built large two‑story residences that could accommodate large families and boarding students (Smith 105; Central Georgian 5 Dec 1860).[ref]Incorporated as a “dry” town, reflecting the town’s Baptist origins, no whiskey was “to be made or sold within a radius of three miles” (Smith,105).[/ref] By 1860 Linton was “very nearly completed” with its Institute, 19 houses, 3 stores, and a wagon repair shop (Smith,110‑11) and was granted a post office in early 1860 (Central Georgian 7 Mar 1860). A visitor in April 1860 noted the “beautiful village” of Linton with its “splendid brick” Institute building soon to be completed. “Several handsome and commodious dwellings have been completed within the past few months, and others are being built. These buildings are not put up as if they were only designed for temporary abodes, but at a cost, some of them, of four or five thousand dollars.” The Institute had about one hundred pupils (Central Georgian 18 Apr 1860).
The Institute building was “the center of Linton’s social life. Not only was the great auditorium the stage for commencement days and special school events…but it was the place for erecting the community Christmas tree and serving the community oyster supper” (Smith 110). And the school was not just for townspeople for many children living in the surrounding countryside attended, walking anywhere from one to four miles to get there (Smith,107). The school bell symbolized the town-country symbiosis for it “‘called children to their lessons, told of deaths and weddings, signaled fires, called the doctor for emergencies, and rang out for state and national events,'” and farmers and housewives for miles around set their clocks by its regular peal (Smith 109).
End‑of‑term exercises in such academy towns were a big cultural event for the entire county. At the Sparta Male and Female Academy, Hancock citzens were treated to speeches, musical entertainments, compostion readings, and “various Literary performances” (Central Georgian 18 Jun 1857). At the close of the fall term in 1860, an all‑day festive celebration at the Washington Institute included “fire‑balloons,” the Linton Brass Band, the Mendelssohn Singing Society, a concert performance of “The Grand Battle” “with all instruments taking part,” and finished off with a “large display of fireworks” and “a grand supper” (Smith 69).
Montour Village, the factory town, provides an interesting contrast to Linton. Compared to the State of Georgia, Hancock County had an unusually high percentage of factory workers. Almost all of the factory workers in Hancock County in 1860 worked for the Montour Company (originally named the Hancock Manufacturing Company and known in Hancock County as “the Factory”), which had a planter‑financed steam‑driven cotton mill and mill town on the south side of Sparta. The mill town contained approximately 245 total white population in 49 households and 45 dwellings.[ref]A local newspaper reports that the mill town embraced “about 300 inhabitants” in 1858 (Central Georgian 11 Feb 1858).[/ref] The 121 individuals with listed occupations included 100 factory workers, 3 farm laborers, 3 professionals (superintendent, physician, engineer), and 15 tradesmen (carpenters, shoemakers, machinists, and a carriagemaker). (Most of these tradesmen were probably employed at the mill.)
Male heads of households and older children, beginning work in their middle to late teens, composed the bulk of the factory work force. The children of the mill town tradesmen (the tradesmen tending to be much older than the average factory worker) constituted 20% of the factory work force. Apart from a few apparent newlyweds, most of the wives and older females listed no occupation.
Bonner notes that factory workers formed “the lowest socio‑economic group,” “the genuine poor” in Hancock County (47). Even immigrants shunned factory work in favor of the more promising trades. Indeed only eight (3.2%) of the mill town inhabitants were foreigners (5 Scotland, 1 Canada, 1 England, and 1 Ireland) and this group included all three of the professionals. Over three‑quarters (78.8%) of the inhabitants were born in Georgia and a significant minority (13.5%) were from North Carolina. However most of the factory workers must have only recently migrated into Hancock County for there appears very little evidence that Hancock County farming families took up available factory work. (Only two factory families shared a surname with any non‑factory Hancock family.) Indeed the mill town and its inhabitants seemed to have played little role in general Hancock society.
Spartans held quite ambivalent attitudes about mill workers. When the stockholders of the Montour Company sought (and received) a charter of incorporation for the mill town in 1857, although there may have been purely economic reasons for incorporation, evidence from the local newspaper suggests the main reason was pure “social control.”[ref]Georgia Laws 1857, p. 186.[/ref] The town leaders apparently had not been too successful “keeping good order” and sought to isolate the mill town from the rest of Sparta in an effort to gain greater control over the factory workers. Incorporation made the Board of Directors ex officio the commissioners of the village with the power to pass whatever laws they saw fit “provided the same are not inconsistent with the laws of the State,” fine up to one hundred dollars, and imprison up to thirty days in the common jail of the county “for any violations of their said by‑laws, rules and regulations.” Immediately upon incorporation the Board instituted a street police in an effort “to enforce such law as will keep the community more quiet than formerly” and elected William Deas their Marshall “whose stalwart frame and intrepidity will make him a terror to evil doers.” (Central Georgian 11 Feb 1858).[ref]According to Richard Griffin, such town‑factory antagonism were quite common; many southerners believed that only “the paternal control of employers” could prevent crowded mill villages from turning into “‘hot‑beds of crime'” (Griffin 35). Interestingly, the commissioners, being stockholders, were not required to live within the corporate limits of the mill village.[/ref]
Cotton manufacturing, in contrast to the findings of Bateman and Weiss, was apparently a risky venture to judge by the experience of the Hancock Manufacturing Company. Initiated in 1851 as an investment alternative for local planter and merchant capital, total subscribed capitalization reached $78,000 in 1853 (1853 Hancock County Tax Digest).[ref]Harris reports, from data obtained in the Dun Ledgers, that planters invested “at least $150,000” when the company was organized (34), but the President, Thomas M. Turner swore before the Justice of the Inferior Court on 15 May 1852 that only $68,415.33 of the amount of capital stock subscribed had actually been paid in and employed by the corporation (Southern Recorder 18 May 1852).[/ref] In 1854, the steam‑driven mill contained 100 Looms and 3500 Spindles, employed 140 hands, and consumed about 1500 bales of cotton a year, producing “950,000 yards of Osnaburgs and Shirtings, and 150,000 lbs. yarn” (Campbell 261), a fairly substantial enterprise. Nevertheless, in 1855, the Factory faced a “severe crisis,” nominally due to “the present high price of cotton and low price of yarns and
homespuns” which, according to the local newspaper, was leading factories throughout the South to shut down operations (Central Georgian 14 Jun 1855). The company eventually sold under levy “by the Sheriff for the low price of $35,000” to gentlemen from Virginia and Rhode Island (Central Georgian 13 Dec 1855) and operations soon started up again.[ref]Actually the purchasers were only able to come up with part of the necessary capital and became part owners, forcing the old Hancock planter and merchant stockholders to retain part ownership (Central Georgian 14 Feb 1856).[/ref]
During this crisis, town leaders expressed concern over the two hundred people who would be thrown out of work, “with nothing to sustain them even for a few weeks,” “many of them without a cent to purchase bread or a weeks provision in their houses” and requested that local planters hire these poor indigents as farm laborers (Central Georgian 14 Jun 1855; 28 Jun 1855). But aversion of the crisis made such “charity” unnecessary.
Antebellum southern factory workers rarely protested their tenuous existence. Griffin emphasizes the docility of the southern workers to “the newness of the employment and the lack of European emigrants, who brought a more highly developed class consciousness with them to the North…If there were strikes, they received no publicity” (38). The one exception in Hancock County seems to prove the rule.
The Factory hands of Montour Company near this place made a strike last week, not for higher wages, but to be allowed to stop work at sun down. The arrangement being it seems for them to work under the lamps until 1st of April. The next morning, however the Factory was closed, as much as to say, those who will not work according to rule, shall not work at all. Necessity being a higher law with them young America had to cave. (Central Georgian 12 Mar 1857)
Ira Berlin and Herbert Gutman believe their discovery of the urban immigrant laborer, “the South’s invisible man,” “necessitates a re‑evaluation of the social order of the antebellum South” (1177,1191). They question “the culture of native Southern whites who remained in the countryside, despite opportunities in the city” and, following the work of Hahn, Genovese, and Owsley, find the best explanation in the general prosperity of the late antebellum era and the “Southern yeomanry’s deep attachment to the land to its unique culture” (1191).
While immigrant urban laborers surely deserve more attention, Berlin and Gutman’s focus necessarily distorts an understanding of the greater numbers of rural native “invisible men.” And although factory work undoubtedly extolled a high price in subordination of the individual to the mill and its owners‑‑a price that few farmers and only those in direst straits wished to pay‑‑there were many poor landless southerners to whom a steady wage looked quite inviting (Griffin 29‑30). Griffin notes that many of these factory workers were actually able to accumulate some savings out of their meager wages (31).
Of all the categories listed in Table I, the smallest category, but the one after factory workers most closely linked to towns, are the merchants. Most studies of southern society include the requisite cite of typical goods purchased from a merchant’s day book, but no historian has pushed the analysis of southern merchants beyond Lewis Atherton’s excellent but dated The Southern Country Store, 1800‑1860 (1949).
Most needed are analyses of merchant records to determine who the customers were and how their purchases changed over time and differed as a function of various socioeconomic factors to understand the complex social and economic links between town and country. Robert Kenzer, using manuscript census records, traced customers from a rural general store account book in antebellum Orange County, North Carolina and found that the customers were mostly farmers living within about five miles of the store and none of these customers shopped at a larger general store in the town of Hillsborough about ten miles to the northeast. The Hillsborough customers were mostly townspeople or farmers living no more than one or two miles away. Gregory Rose, an historical geographer, used daybooks, contemporary plat maps, and manuscript census records to reconstruct the trade area for an 1850s Massachusetts general store. Rose likewise found distance to be a critical factor, but recognized that populations of competing towns also played a significant role.
An analysis for the years 1859 and 1860 of the account books of T. I. [Thomas Irby] Little & Co., a general store located in the county seat of Sparta, confirms the findings of Kenzer and Rose on the importance of distance for identifying market areas, but refutes the notion of clear boundaries between different market areas. A general store in the county seat was very much in competition with general stores located in the outlying villages, hamlets, and neighborhoods. Kenzer’s notion of an isolated neighborhood making contact with the outside world only through their local merchant misses the mark. Individuals from every district and every village in Hancock county had credit accounts with T. I. Little, although the closer districts and villages had far greater percentages of such accounts.[ref]Both U. B. Phillips (126‑7) and J. William Harris (5) have placed Hancock County within Augusta’s wholesale marketing area and evidence from the Little account books supports this. Little purchased segars and tobacco from J. Volger and his drugs from Haviland & Chichester, both identified as Augusta firms from advertisements in Campbell.[/ref]
Almost every head of household and every person with an occupation in Sparta had a credit account with T. I. Little. Although there were other general stores in Sparta, they each tended to specialize in different lines. From the planter buying “segars” and brandy, to the carpenter buying paints and varnishes, to the physician buying drugs and medicines, to the little girl buying nuts and candy, everyone in Sparta eventually went into T. I. Little’s. Indeed such a relationship with the general store was probably an essential component of being a Spartan. Even Sparta’s other merchants all had credit acounts at T. I. Little’s. Interestingly, however, only four of the residents of the mill town (contiguous with Sparta) had credit accounts. This indicates again how isolated factory workers were from the rest of the county.
Spartans were not the only Little customers. In the closer districts, as much as 50% of the heads of household had credit accounts; in the outermost districts the percentage dropped to as low as 7%. The great majority of these rural customers of T. I. Little & Co. were planters and farmers, although a few overseers and farm laborers had accounts. A quarter of the heads of household from the village of Mt. Zion (7 miles north of Sparta) and the hamlet of Culverton (5 miles east) and half of the heads from the new village of Linton (12 miles south) had credit accounts at T. I Little’s despite having their own local general stores.
However, none of the merchants in other parts of the county had accounts with Little, indicating that they had their own wholesaling connections. Judging by total value of merchandise (taken from the 1860 Hancock tax digest), T. I. Little & Co. ($12300) was quite a bit larger than the stores in Mt. Zion ($8000), Culverton ($5000) and Linton ($2000) and thus probably offered a greater variety at lower prices, inducing some villagers and planters living in the region of the other stores to spread their patronage.
The more remote parts of the county fit Kenzer’s picture of self‑contained “neighborhoods” much better. The village of Powelton (14 northeast of Sparta) with general merchandise valued at $2500 was exceptional in that only 3.6% (1 out of 28) of village households and 7.1% (3 out of 42) of surrounding rural households (Militia District 106) had a credit account with Little. Undoubtedly, a great number of factors acting over time, only some of which have been examined here, helped to give one set of merchant relations precedence over another.
Such analysis only begins to tap the wealth of information available in merchant records. More accurate analysis of differences in purchases and accounts between households and of changes over time in household consumption would reveal much about a whole set of market relations that is not very well understood. The great attention paid to staples versus foodstuffs production ignores a whole “consumer revolution” which drove participation in the market for a great many southern households that desired non‑local goods. Analyses which emphasize the self‑sufficiency of households, all the while excepting purchased goods like coffee, tobacco, sugar, salt, gunpowder, etc., just make clear that self‑sufficiency never was the objective, because households could certainly have “survived” without purchasing such goods from the local merchant.[ref]Carole Shammas has begun to make this distinction for early America (1982a, 1982b).[/ref]
A picture of antebellum southern society populated by many non‑agriculturally employed households living in towns demands that many traditional historiographical questions be reconsidered. For example, the many studies analyzing self‑sufficiency of large and small farms (Gallman; Ford,52‑60,81‑83,249‑254; Hahn,32‑33; Harris,31‑32). Foodstuffs production analysis should not just focus on yeoman or planter self‑sufficiency and nebulous barter arrangements but should define the formal or informal market required to put food into the mouths of non‑agricultural people. Sam Bowers Hilliard’s Hog Meat and Hoecake: Food Supply in the Old South, 1840‑1860 makes a start but a good understanding of the market in foodstuffs will come only with many local studies.
Aggregate calculations, using both Hilliard’s method of calculating pork and corn deficits (104‑109; 157‑161; 260‑261) and J. William Harris’s adaptation of Ransom and Sutch (Harris 200‑201), indicate that Hancock County was self‑sufficient in both breadstuffs and meat production, that planters large and small were producing enough food for the entire county population. A alternative calculation of value of animals slaughtered on the farm also shows that Hancock farms produced enough meat for the entire county.[ref]Based on the average market price for pork at eight cents per pound, Hancock farms produced 3.9 pounds of meat per human consuming unit, in excess of the 3.5 pounds prescribed by Hilliard. For some reason, historians have made no use of this “Value of Animals Slaughtered” category in the 1850 and 1860 agricultural census. Although such data may not help reveal whether farmers were able to sustain slaughter rates year after year and thus maintain self‑sufficiency, at least the data can a better picture of a minimum amount of meat produced in the county.[/ref]
But such calculations tell us little about the market in foodstuffs. Evidence from newspapers suggests that a year‑round market in foodstuffs existed and that Tennessee drovers brought a significant amount of pork into the county.[ref]The local newspaper was filled with complaints that pork was continually imported from Western markets (e.g., CG 29 Nov 1855).[/ref] Wheat, corn, and pork were always available at a price, a matter of supply and demand. If the demand and price went up, so did the supply. As might be expected in such a free market, the price fluctuated considerably. From 1855 to 1857, as reported in the Central Georgian, corn varied from $0.70 to $1.40 per bushel, flour from $3.50 to $6.00 per bushel, and pork (both Tennessee and home raised) from 5 to 8 cents per pound neat. Tennessee pork was not always available but home‑grown pork almost always made up the deficit (Central Georgian 10 Jan 1855; 20 Dec 1855).
Planters had several options to choose from when deciding to market their surplus foodstuffs. They could store the surplus on farm as security for future crop failures or in order to plant more heavily in cotton the next year (Central Georgian 17 May 1855). They could market the surplus through the same factor that handled their cotton. They could sell, barter, or give away the surplus from the farm or they could haul it into town and sell it there. What each planter did depended no doubt on his own economic calculations and prognostications. High transportation costs for breadstuffs and high local demand for pork and beef undoubtedly encouraged local marketing. Some planters in surrounding counties actually specialized in foodstuff production. William Warthen offered 1000‑2000 bushels for sale at his river plantation in Washington County in 1859 (Central Georgian 23 Mar 1859). James F. Smith of Emanuel County drove his cows into Sandersville in Washington County 2‑3 times a week in the late 1850s (Central Georgian 23 May 1857; 25 Jul 1860).
Although foodstuffs were available and sold at general stores like T. I. Little, these stores seemed to have specialized more in imported “luxury” foods and probably played only a minor role in the total food supply of smaller towns. Stores most likely played a bigger role in foodstuffs distribution of larger towns like Milledgeville to judge from store advertisements for corn, flour, and hams (Southern Recorder 25 May 1852). Evidence from local newspapers suggests that Hancock corn and wheat mills marketed most of the local trade in breadstuffs. In 1849, Hancock County had 5 corn‑mills and 10 corn and wheat‑mills spread throughout the county (White 314). Some of these mills always seemed to have a supply of either “old” or “new” flour or meal on hand, although the supply could be irregular (Central Georgian 12 Jul 1855; 7 Jun 1855). The Factory apparently had a small company store and milled and sold meal and flour to its employees, as well as many Spartans, although generally priced higher than rural mills (Central Georgian 7 Jun 1855).[ref]The agent for the later Montour Manufacturing Company showed merchandise valued at $500 on the 1860 Hancock Tax Digest.[/ref]
The pork and beef market in Hancock County shows even greater complexities than the breadstuffs market. Bode and Ginter show in their analysis of the 1860 agricultural census that many town dwellers with nonagricultural occupations held small numbers of livestock (18‑19). The 1850 agricultural census for Hancock likewise confirms that town folk kept livestock and small gardens. Most livestock sold in the towns came in on the hoof, whether local or from Tennessee and Kentucky, only irregularly and was slaughtered upon demand by drovers and farmers working out of the back of their wagons. In 1860, Sandersville in Washington County, a town much more prosperous than Sparta being located close to the railroad, bemoaned the lack of a “market-house”. “As it now is, persons bringing fresh meats, &c., here for sale, are driven to the necessity of selling from their carts and wagons, and to sell in such quantities as not to be convenient for most of our citizens. Whereas, if we had a market‑house where these articles could be offered for sale each morning, it would be not only a great convenience to those selling, but also the citizens generally” (Central Georgian 9 May 1860). Soon thereafter, James F. Smith, the Emanuel County planter who had been driving his cattle into town since at least 1857, did finally erect a Market House (Central Georgian 25 Jul 1860).
Violence in the antebellum South represents another aspect of southern towns which begs further investigation by social historians. If Hancock County is at all representative, then evidence suggests that most of that renowned “southern violence” occurred in towns, often in merchant and trade shops and, indeed, often involving tradesmen and merchants.[ref]Although a town newspaper might be expected to report town murders more regularly than murders in the countryside, the general county coverage of accidental deaths, suicides, and other such events belies such a bias.[/ref] Of the six deaths which resulted from bloodshed in Hancock County reported in the Central Georgian between 1855 and 1860, all sixe occurred in towns: five in Sparta and one in Powelton. In two of the deaths, retail grocers shot and killed menacing intruders, one of which was a grocery clerk (Central Georgian 30 Jan 1855; 8 Feb 1855; 7 Mar 1856). Two other deaths involved two brothers who were both killed in their carriage shop in a knife and gun fight with two other men (Central Georgian 18 Apr 1856; 9 May 1856). All four of these deaths seemed to have resulted from standing grudges. One of the men involved in the killing at the carriage shop was eventually killed himself after being “hit over the head with a glass decanter” (Central Georgian 29 Feb 1860). The sixth death took place at the Factory when the watchman, while scuffling with a drunken intruder, accidentally shot a female bystander (Central Georgian 7 Mar 1860).
The town provided the milieu for “southern violence” because in towns men congregated and grocers served liquor. Edward Ayers, in examining crime and punishment in antebellum Georgia, does not distinguish between town and country violence in the “rural counties of Greene (bordering on Hancock County) and Whitfield and hardly explores the relationship between towns, groceries, and “alcohol‑induced violence” (115).
Clearly the evidence presented in this essay, while not definitive, shows that any understanding of social structure, occupational structure, self‑sufficiency, consumerism, violence, and many other aspects of life in the antebellum South must begin with an understanding of southern towns, the people who lived and worked there, and the symbiotic relationship between the town and the people who lived in the surrounding countryside. As Berlin and Gutman point out, the “social order of the antebellum South” is quite correctly in need of reordering, but that reordering will come only with an understanding of the masses of “invisible people”–the landless white laborers, farm and non‑farm, native and immigrant, as well as the tradesmen, merchants, and professionals that Bonner identified almost fifty years ago–and the niche that these masses filled in the antebellum southern planter-yeoman-slave society that modern social historians have worked so hard to create.
Works Cited