The 1790 Connecticut Census

Background
With no precedents, census taking in America was far from uniform in 1790. According to the introduction to Heads of Families at the First Census of the United States Taken in the Year 1790, Connecticut (Baltimore: Genealogical Publishing Co., Inc., 1980), enumerators in New England presented population by towns or counties, while elsewhere in the new nation, it was done primarily by counties. Nationwide, seventeen marshals of the judicial districts supervised the undertaking. They were authorized to appoint as many assistant marshals in their districts as they deemed necessary (approximately 650 in all). Each assistant was assigned a district to enumerate that could comprise one or more counties, cities, towns, townships, or parishes.

Apparently the federal government provided little assistance to the marshals, who may actually have received instructions from governors of their states. Each enumerator did receive a copy of the Census Act of 1790. Although Massachusetts furnished forms, marshals in other states had to devise their own forms on paper they themselves supplied. Completed schedules varied in length from a few inches to several feet and all were handwritten on handmade paper of very uneven quality. Many contained blots, erasures, and misspellings, as well as errors in copying and counting, as any researcher who has looked at microfilms of the original records can attest. Enumerators made two copies of each return. They posted one locally for information and corrections and sent the other to their district marshal for forwarding to the President.

The objects of the first census were two-fold: to determine how to apportion representation to the lower house of Congress and to obtain information about the military and industrial potential of the new nation. Thus, an accurate accounting of males old enough for military service or full-time work (free white males sixteen years of age and over, in particular) was of primary importance. Indians were neither taxed nor counted. Free persons, including indentured servants, mulattos, and free blacks of both sexes, were listed separately, as were slaves.

Enumerators of 1790 often encountered suspicion and hostility. Citizens accustomed to the freedom of the frontier feared federal control and questioned motives for government inquiry into their personal affairs. Others feared an increase in taxes or that divine retribution would follow an enumeration of all the people.

Human problems were compounded by difficult working conditions. Roads were poor to nonexistent, travel was slow, inns were scarce, and town and county boundaries were not always clear. In spite of these problems, most enumerators met the deadline of nine months with a reasonably accurate counting of residents within their districts. Pay was based on the number of persons counted. The rate for the backcountry was $1.00 for each 150 persons counted, and in counties where population was more dispersed, pay could be as high as $1.00 for every 50 persons. In cities or towns with more than 5,000 persons, pay was $1.00 for every 300 persons.

Connecticut Towns with Alphabetical Listings
Considering the difficulties and limitations associated with taking the first census, how can we explain why some enumerators went to the additional work of alphabetizing their lists when it was not required? Hartford County census enumerators listed alphabetically, apparently by sections of towns, residents of Berlin, Glastonbury, and Hartford. In Litchfield County, residents of Bethlem [sic], Cornwall, Harwinton, Kent, New Milford, Southbury, Warren, Washington, Watertown, and Woodbury were also placed in alphabetical order. In Middlesex County, only residents of the town of Chatham were listed alphabetically and in Tolland County, only the residents of Stafford. Names of heads of households in Fairfield and New Haven counties were not alphabetized.

Connecticut Towns Listed Together
Under "Errata" on page 8 of Heads of Families, First Census of the United States, 1790 Connecticut, the fact that twelve towns in Litchfield County were all returned under the heading "Litchfield" is mentioned but not explained. Those towns included Barkhamstead, Canaan, Colebrook, Goshen, Hartland, Litchfield, New Hartford, Norfolk, Salisbury, Sharon, Torrington, and Winchester. No notation explains listing together residents of all the towns of New London County, making it difficult to determine exactly where in New London County any individual head of household lived. Separating towns listed together might be possible if it could be shown that the marshals completed one town before beginning the next.

Separating the Towns of Norwalk and Stamford
In Fairfield County, David Maltbie was assistant marshal for the towns of Greenwich, Norwalk, and Stamford. For reasons unknown today, he listed residents of Greenwich separately, yet listed residents of Norwalk and Stamford together. In 1790 Stamford included all of Darien, then called Middlesex Parish, and part of New Canaan, called Canaan Parish. Norwalk included the remainder of Canaan Parish as well as Wilton.

About 1980 the late Lois Bayles, then librarian at the New Canaan Historical Society, worked out the route the 1790 census enumerator took through Canaan Parish. Drawing on her extensive knowledge of the settlement pattern of the area, she found that the enumerator moved along in a very orderly manner, from house to house and street by street. At certain points the enumerator followed a road from Canaan Parish into Norwalk or Stamford and then returned to Canaan Parish.

Using Mrs. Bayles' study, I set out to determine whether additional divisions between Norwalk and Stamford towns and parishes could be established. The 1790 tax list for the town of Stamford, kept at the Stamford Historical Society, provides a contemporary listing of Stamford heads of household. In addition to taxpayers in Stamford, this list includes names of taxpayers in the parishes of Middlesex and Canaan in Stamford. Since a poll tax was included, most adult males were taxed, and hence listed. Logically, residents not included in the Bayles study of Canaan Parish or in the Stamford/Middlesex Parish tax list probably lived in Norwalk.

In the 1790 Connecticut census, the names of families who lived on the Norwalk side of Canaan Parish precede the names of those who lived on the Stamford side. For this reason and because the census listing is entitled "Norwalk and Stamford Towns," I reasoned that Maltbie began taking the census in Norwalk and completed that town, including Canaan Parish in Norwalk (plus a few families living in Middlesex Parish near Norwalk). He then enumerated those residing in Stamford and Canaan Parish in Stamford and finally listed residents of Middlesex Parish (later Darien) in Stamford. With the exception of families who lived on or near the town lines, I found fairly clear-cut divisions between towns and parishes in the census listing. It is evident that the enumerator did continue along streets that crossed town or parish boundaries.

The divisions between Norwalk, Stamford, Canaan Parish, and Middlesex Parish are as follows:

In the Stamford portion of the census, there are sixteen heads of households listed for whom I found no other indication of Stamford residency. Their names do not appear on the 1790 Stamford tax list or in any other extant town records. It is, however, likely that most or all did live in Stamford. They include Abigail Smith, Darius Peck, Martin Dissabrose, Capt. Nathanial Starr, Janus Waterbury, Widow Elizabeth Comestock, Widow Molly Smith, Lewis M. Donald, Widow Mary Wilson, Augustus Wilkes, Amos Stevens, Jr., Isaac Stevens, Sarah Howard, Zepeniah Slason, Widow Esther Seely, and Joseph Sudmore. Of the six women on the list, four were widows. Without knowing the names of their husbands, tracing them is difficult. In addition, economic conditions of aged widows often precluded taxation. Women lacked the right to vote and hence did not pay a poll tax.

Numerical results of this somewhat arbitrary division of Norwalk and Stamford residents in 1790 indicated Norwalk had 858 family heads and a population total of 4,759, while Stamford had 756 family heads and a population total of 3,980. Sixteen family heads and a population total of seventy-one persons are not placed. I may have unintentionally placed some families in the wrong town and will be happy to hear from readers who have information I lack.

An earlier version of this article, as well as photocopies of pages from the 1790 Connecticut census, appeared in Connecticut Ancestry 32 (May 1990):163-172.