Sundown Towns: A Hidden Dimension Of American Racism (87 page)

48
Greene, Kremer, and Holland,
Missouri’s Black Heritage,
151; Walter E. Williams,
The State Against Blacks
(New York: McGraw-Hill, 1982), Chapter 8.
49
The First Lady and Interior Secretary Harold Ickes made some important symbolic gestures, such as allowing Marian Anderson to sing at the Lincoln Memorial in 1939.
CHAPTER 3: THE GREAT RETREAT
 
1
Emma Lou Thornbrough,
The Negro in Indiana
(Indianapolis: Indiana Historical Bureau, 1957), 224.
2
I define the “traditional South” as Virginia, North Carolina, South Carolina, Georgia, Florida, Alabama, Tennessee, Mississippi, and Louisiana—states historically dominated by slavery. Note 63 in this chapter defends this definition.
3
Admittedly, some towns in the Midwest went sundown vis-à-vis African Americans earlier, during the Civil War, and some even before it, but most towns in that region and elsewhere did so later.
4
Unlike most southern Illinois towns, Cairo never tried to get rid of its black population. It simply had too many blacks. Also, some whites in and around Cairo adhered to traditional southern race relations, using African Americans as cotton pickers, maids, janitors, railroad labor, etc., and would have opposed any expulsion. Map 1 shows this area of traditional southern race relations in Illinois.
5
Cairo Bulletin
quoted in
Marion Daily Republican,
8/12/1920.
6
Historian Jeri L. Reed, e-mail, 6/2002;
Guide and Directory for the City of Rogers
(Benton County, 1907), 4; “Siloam Springs, Arkansas,” brochure in collection of Siloam Springs Museum; Michael Birdwell, e-mail, 5/2002;
Terry County Herald,
3/27/1908; Betty Dawn Hamilton, e-mail, 7/2002; Eulalia N. Wells,
Blazing the Way
(Blanket, TX: author, 1942), 77.
7
Owosso and Shiawassee County Directory, 1936
(Owosso: Owosso Chamber of Commerce, 1936); David M. Chalmers,
Hooded Americanism
(New York: Franklin Watts, 1976 [1965]), 165; David M. P. Freund, “Making It Home,” Ph.D. dissertation, University of Michigan, 1999, 225.
8
“Come see Woodson” ad from Dale Steiner.
9
There was a wave of anti-Chinese activity just before the Civil War, including expulsions from many mining camps and small towns in 1858–59. See Stanford Lyman,
Chinese Americans
(New York: Random House, 1974), 60.
10
Craig Storti,
Incident at Bitter Creek
(Ames: Iowa State University Press, 1991), 112–15, 118, 125, 159–60; Bill Bryson,
Made in America
(New York: Morrow, 1994), 127; Grant K. Anderson, “Deadwood’s Chinatown,” in Arif Dirlik, ed.,
Chinese on the American Frontier
(Lanham, MD: Rowman & Littlefield, 2001), 423; Ed Marston, “Truth-telling Needs a Home in the West,”
High Country News,
9/25/2000; “Files Found in Oregon Detail Massacre of Chinese,”
New York Times,
8/20/1995; Brigham Madsen,
Corinne: The Gentile Capital of Utah
(Salt Lake City: Utah State Historical Society, 1980), 252.
11
Florence C. Lister and Robert H. Lister, “Chinese Sojourners in Territorial Prescott,” in Dirlik, ed.,
Chinese on the American Frontier,
55; Kathy Hodges, e-mail, 6/2002; Li-hua Yu, “Chinese Immigrants in Idaho,” Ph.D. dissertation, Bowling Green State University, 1991, Chapter 7; Priscilla Wegars, e-mail, 8/2003, summarizing her “The History and Archaeology of the Chinese in Northern Idaho, 1880 Through 1910,” Ph.D. dissertation, University of Idaho, 1991.
12
Jean Pfaelzer, paper presented to the American Studies Association, Washington, DC, 2001; Pfaelzer, “A Proposal for
Driven Out: Ethnic Cleansing and Resistance,”
n.d., 4, 10; Visalia oral history via Gilbert Gia, e-mail, 9/2002; Keith Easthouse, “The Chinese Expulsion—Looking Back on a Dark Episode,”
North Coast Journal Weekly,
2/27/2003, northcoast
journal.com/022703/cover0227.html
, 2/2004; Kevin Hearle, e-mails, 9/2002, 5/2004, quoting Diann Marsh,
Santa Ana: An Illustrated History
(San Diego: Heritage, 1994), 95–96; cf. Pfaelzer,
Driven Out
(New York: Random House, forthcoming).
13
Lynwood Carranco, “Chinese Expulsion from Humboldt County,” in Roger Daniels, ed.,
Anti-Chinese Violence in North America
(New York: Arno Press, 1978), 332–39.
14
The
Humboldt Times
overstated its case: several other towns kept out Chinese Americans in 1937. Also, apparently not all Chinese Americans were expelled from Orleans, an inland mining hamlet barely in Humboldt County.
15
Easthouse, “The Chinese Expulsion”; Carranco, “Chinese Expulsion from Humboldt County.”
16
Joseph A. Dacus,
Annals of the Great Strike
(New York: Arno Press, 1969 [1877]; Online Encyclopedia of Seattle and King County History,
historylink.org
, 3/2003.
17
A few places drove out their blacks well before 1890.
18
Cf. Jack S. Blocker Jr., “Choice and Circumstance,” Organization of American Historians, Toronto, 4/1999, 2–3.
19
Ibid.; Loewen,
The Mississippi Chinese
(Prospect Heights, IL: Waveland, 1988).
20
I did not study Alaska or Hawaii and did very little research on Montana and the Dakotas.
21
I used 1930 statistics in
Table 1
to avoid contamination from the Great Depression. I didn’t want readers to imagine that African Americans may have retreated to the cities for economic reasons. Using 1930 avoids this issue, because the Crash happened on October 24, 1929, only months before enumerators fanned out to take the 1930 census, and it had prompted little population movement by that time. (For the record, the Great Depression actually prompted some migration
from
big cities back to subsistence farming in the South.)
22
Columns with < 10 African Americans include those with none.
23
I define the “traditional” South as Virginia, North Carolina, South Carolina, Georgia, Florida, Alabama, Tennessee, Mississippi, and Louisiana. A later section in this chapter explains why the table includes Arkansas and Texas. The Appendix tells why
Table 1
omits Alaska and Hawaii, describes idiosyncracies in some states, and discusses other methodological issues underlying the table. Arizona, New Mexico, and Oklahoma were not states in 1890; I substituted data from 1910, the first census after statehood.
24
Minnesota had more counties in 1890 with no blacks, but more in 1930 with just a few blacks.
25
The Appendix also comments on other states whose counties showed only slight trends. It also summarizes recent work by historian Jack Blocker Jr. that reinforces the conclusions presented here.
26
Additional causes included a farm depression in the South, harm done to cotton by the boll weevil in 1915–16, and the South’s continuing denial of civil rights to African Americans.
27
Beloit (WI)
News,
8/25/1916, quoted in C. K. Doreski,
Writing America Black: Race Rhetoric in the Public Sphere
(New York: Cambridge University Press, 1998), 31–32.
28
T.J. Woofter Jr.,
Negro Problems in Cities
(New York: Harper & Row, 1969 [1928]), 28–31, quotes from 31.
29
My web site lists confirmed counties in all states.
30
I don’t believe the detail that he literally met them at the line, but I do believe that the sheriff enforced the sundown policy.
31
Malcolm Ross,
All Manner of Men
(New York: Reynal & Hitchcock, 1948), 66; 83-year-old Mason County resident, 4/2000; Carol Speakman, 9/2002.
32
This figure is approximate. I may have overlooked a town or two that exceeded 1,000 in 1970 but not in adjoining censuses; I omitted some usually unincorporated places the census tags as CDPs, or “census designated places.”
33
All suspected and confirmed sundown towns are listed at
uvm.edu/~jloewen/sundown
. Another researcher might come up with slightly different numbers, because I included smaller towns that people brought to my attention as sundown towns and did not include a few small towns that reached 1,000 people only in recent censuses. Other reasons to exclude an occasional town were judgment calls and my own exhaustion. I also excluded CDPs unless I knew them to be considered towns by residents of the area.
34
These communities are listed at
uvm.edu/~jloewen/sundown
. The exception was Newton. Confirmed towns form a scattered subset of all suspect towns, clustered near locations where I had speaking engagements, knew someone, or otherwise had connections. In addition, I contacted all 56 suspected sundown towns on Map 1. The proportion of confirmed sundown towns among those 56 was very similar to that among my scattered subset, so I don’t believe any important error resulted from my nonrandom sampling.
35
I often use 1970 figures because most sundown towns were still all-white on purpose in that year. To be sure, some still are, but many relaxed their policy in the 1970s or thereafter, especially towns that grew rapidly. Using 1970 figures is conservative in that it underestimates the size of the cities today.
36
My process of information gathering was not perfectly random: when I invited audience members to share with me information about sundown towns, for example, people from overwhelmingly white towns that were
not
sundown towns would not be likely to speak, having nothing to impart. But I also made efforts to learn about
all
towns on Map 1, for example, and the proportion of confirmed sundown towns that I uncovered from that quest was no different than that obtained from volunteered sources.
37
Actually, we shall compute confidence limits for our estimate of 99.5% of 278 unknown towns, or 277 probable towns.
38
To find the standard error of the difference of two percentages, first we calculate each standard error separately. Beginning with the 194 towns on which we have information, the formula is s
p1
=
where n = the number of towns for which we have data (194), p = the proportion that were sundown (.995), and q = (1 - p) or .005. This standard error = .005 or 0.5%.
We also need to compute the standard error of the percentage of sundown towns among the 228 towns for which we have no information. Since we don’t know this percentage, we can assume just 90% will be, lower than the most likely estimate; such a conservative assumption provides a larger-than-likely standard error, which results in a more conservative overall estimate. Using the same formula, we substitute: n = 228, p = .9, and q = .1. This standard error = .020 or about 2%.
We then combine these two standard errors, using the formula s
(p1–p2)
=
find the standard error of the difference of two percentages, which = .0205 or 2.05%.
39
Statistical tables tell us that 99 times out of 100, a range that is ÷2.58 standard errors from our best estimate will include the actual percentage. 2.58 x .0205 = .053 or 5.3%.

Other books

A Fighting Chance by Elizabeth Warren
The Ignorance of Blood by Robert Wilson
Cryers Hill by Kitty Aldridge
The Billionaire's Desire by Ashley Blake
The Day to Remember by Jessica Wood
Her Christmas Earl by Anna Campbell