The Absence of Muslim Americans in the Religious Bodies Census
There are certain movements and cults which claim a number of adherents, but are not so organized as to make their presentation as religious bodies [denominations] advisable… Because of the distinctive nature of these movements and for the reason that they do not have a distinctive membership, the Bureau did not consider it feasible to attempt to obtain any definite statistics.1
The Census Bureau listed as examples of the “movements and cults” it would not count some 57 “small sects,” including the Jehovah’s Witnesses, Father Divine’s Peace Mission, Frank Buchman’s Oxford Movement, and Native American Churches. It might not have been easy for the Bureau to gather congregational data about some of their movements, but in some cases the Bureau’s explanation is baffling. The Jehovah’s Witnesses, for instance, published an annual yearbook, and they claimed more than 50,000 members in the mid-1930s. They were and had a distinctive membership and were organized much like the Protestant denominations the Census Bureau took as its norm.
There were several Muslim societies and movements on this list of “cults” and “small sects”: the American Mohammedan Society, the American Moslem Brotherhood (King Solomon Temple of Religious Science), the Moslem Temple in Detroit, Michigan. The Bureau did not include these Muslim groups. In fact, it did not include any expressions of Islam in the United States in any of its published decennial religion censuses.
The legislation that created a permanent Census authorized the Bureau to collect statistics about “special classes,” including “religious bodies,” but left the task to the Bureau’s discretion. Building on efforts that stretched back to the 1850 population census, the Bureau was most comfortable surveying those religious organizations that conformed to the Protestant denominational model, what Kathleen Flake terms “the American idea of a church.”2 On that basis, the Bureau in 1926 excluded Buddhist temples on the West Coast, claiming that it was impossible to ascertain their membership.
The Bureau’s brief explanation conveys its condescension. “Certain movements and cults” were not very well organized. It was not “advisable” or “feasible” to count them. Because the Bureau had discretion over its survey of “religious bodies,” its officials did not have to justify their choices. The Bureau could have found a way to include Muslim institutions in its Religious Bodies census. It chose to not do so.
In his Muslims of the Heartland, Edward Curtis notes that “234 of the 670 self-identifying Syrians in the South Dakota state census of 1915 were Muslim.”3 It is hardly an insignificant figure. In 1926, for instance, the U.S. Census Bureau counted fewer than 200 members of several denominations in South Dakota, such as the Society of Friends (Orthodox) and the Christian and Missionary Alliance. But it included those groups. Especially after a wave of emigration from then French-controlled Syria and Lebanon, there were sizeable Muslim populations in many parts of the Midwest and Northeast. Why are they absent from the Religious Bodies census? The first explanation is that the Bureau—unlike the state of South Dakota in 1915—did not survey the religious affiliations of individuals. It counted the members of a large but still limited number of denominations and movements.
There were a number of American mosques by the mid-1930s, including a mosque founded in Ross, North Dakota, in 1929, as well as others in Iowa, Maine, Michigan, and Indiana. Perhaps the Bureau did not know what to do with individual mosques, let alone less visible Muslim associations and socities. By “religious bodies,” the Bureau meant “religious organizations,” envisioning denominations and churches with multiple congregations in different geographic locations. However, by the mid-1920s, when the Bureau counted “Jewish congregations,” it actually counted “all persons of the Jewish faith living in communities in which local congregations are situated.” The Bureau could have taken a similar approach to the Muslim residents of communities that had a mosque, and it may well have found some way to enumerate these populations had it persisted with the Religious Bodies census through and beyond 1946.
The Bureau also did not capture the marked increase in African American conversions to Islam. The Bureau did not include the Moorish Science Temple (MST), organized by Timothy Drew (who changed his name to the Noble Drew Ali) in Chicago in 1925. As Edward Curtis notes in his Muslims in America, the MST “was the first example of an independent, African American Muslim missionary group devoted to the cause of spreading Islam, however, defined.”4 The MST was a significant movement by the 1930s, with temples in a number of cities and towns. In other words, it was precisely the sort of “Religious Body” that the Census Bureau should have counted.
The absence of Muslims points to the central paradox of the early twentieth-century Religious Bodies censuses. They are a treasure trove of reliable data about a large number of religious movements in the United States. The sociologist Rodney Stark quite fairly praised their “amazing range and detail.”5 At the same time, the censuses have severe limitations, most notably in the exclusion of certain religious movements and groups.
The individual schedules from the 1926 Religious Bodies census, along with the tabulated and published data from that and other decades, will make it possible for researchers to visualize particular religious ecologies across the United States. One could do so for rural Mountrail County in North Dakota. For example, here is a Seventh-Day Adventist congregation in the town of Stanley. But there is no record of the Syrian-Muslim community in Ross. The Religious Bodies census provides considerable evidence of a large amount of religious diversity in cities such as Detroit and Chicago, but it still misses many expressions of Black religious life—Muslim and otherwise.
There are, however, many other sources of information and data about Muslim American communities in the early 1900s. There are also city directories, insurance maps, atlases, and advertisements in Black newspapers. Also, the U.S. government surveilled a number of Black religious movements. For instance, the FBI under J. Edgar Hoover compiled a list of Moorish Science temples in the 1930s.
Diligent and creative historians have documented these urban and rural worlds. Judith Weisenfeld’s New World A-Coming: Black Religion and Racial Identity during the Great Migration narrates the rise of movements the U.S. Census Bureau ignored: the Moorish Science Temple, the Nation of Islam, and Father Divine’s Peace Mission. Likewise, Edward Curtis’s Muslims of the Heartland recreates communities the Census Bureau did not count.
Religious Bodies, 1936 (Washington: Government Printing Office, 1941), 1:7. ↩︎
Kathleen Flake, The Politics of American Religious Identity: The Seating of Senator Reed Smoot, Mormon Apostle (Chapel Hill: University of North Carolina Press, 2004), chapter one. ↩︎
Edward Curtis, Muslims of the Heartland: How Syrian Immigrants Made a Home in the American Midwest (New York: New York University Press, 2022), chapter one. ↩︎
Edward Curtis, Muslims in America: A Short History (New York: Oxford University Press, 2009), 36. ↩︎
Rodney Stark, “The Reliability of Historical United States Census Data on Religion,” Sociological Analysis 53 (1992): 91. ↩︎