Sundown versus consistent GOP by county

Studying US politics, one striking factor I noticed is the extremely consistent Republican voting of certain counties in the rural Midwest and nonplantation South. The constant partisanship of these counties contrasts strongly with the reversal frequently noticed by psephologists since the 1964 election between Lyndon Johnson and Barry Goldwater.

The extreme Republican loyalty of these counties is due to the fact that they were strongly supportive of the Union during the Civil War. Most lie in antebellum slave states, yet virtually nobody in these counties actually owned slaves. These counties’ almost entirely white populations refused vehemently to fight a war for lowland plantation owners. Those that lie outside antebellum slave states share many similar characteristics, but possessed even greater hostility towards the spread of slavery into the territories. These features led residents into extremely strong loyalty towards the unionist Republican Party, although, critically, they were opposed to any civil rights for blacks.

At its root, the American Civil War was, critically, a war over how to rid America of free blacks. In fact, as noted by the late Domenico Losurdo, the words “pariah” and “rogue”, used today for anti-American states, originated as words for a free black. Losurdo likewise notes that:

“Blacks not subject to slavery began [after the American Revolution] to be perceived as an anomaly that would sooner or later have to be rectified.”

and that:

“[Blacks in areas without slaves] were treated like ”a race legally and socially excommunicated, as the Helots of Sparta as the Pariahs of India disfranchised outcasts; a separate and degraded caste” [Leon F. Litwack, North of Slavery, University of Chicago Press, 1961, pages 67 and 97]”

Thus, these anti-slavery communities sought a means of deporting blacks to Africa, and indeed Abraham Lincoln was close to doing so after the Civil War, although there clearly existed no federal revenue source adequate to fund the cost of so doing. As the late James Löwen showed in his book Sundown Towns: A Hidden Dimension of American Racism, after the Compromise of 1877, these counties — unlike the plantation South lacking any need for black labour — attempted in many cases to expel black population who had moved out of the plantation South during Reconstruction. Before reading Löwen, I had assumed that these counties never had any black residents and the blacks had never wanted to move there, and it was something of a shock that blacks were de facto forbidden from living in almost all rural and suburban areas outside the plantation South.

Consequently, I am attempting to discover to what extent the consistently Republican counties have excluded blacks. Before Löwen died, I talked to him about this issue. Both Löwen and I noted that partisanship during the early Second Party System may provide clues as to which counties were most likely to exclude blacks a century and a quarter later. However, I did not do a systematic study of this, and it is cumbersome or expensive to get the necessary 1830s and 1840s election data. For this table I have included all counties meeting the following criteria:

1. having either:
1. voted Democratic no more than once since the Civil War, or
2. consistently voted Republican since 1920
2. and never had a population of 75,000 or larger at any census
• studying the figures tabulated below, my instinct tells me that this 75,000 cutoff point, although smaller than that usually used to define “rural” counties, is likely too large
• it is of course true that blacks could not live in many parts of counties with total populations much larger than 75,000
• nevertheless, the largest county with zero black residents in the 1930 census — Portage County, Wisconsin, ironically one of only 130 counties to back George McGovern in 1972 and indeed among his 25 best counties nationwide — had only about 33,000 total residents

Counties are shaded according to how many black residents (in censuses up to 1960) and black households (in censuses since 1970) they are shown as having. Non-household blacks such as prison inmates and live-in servants obscure non-existence of free blacks in many rural counties outside the plantation South. This was corrected in the 1960 and 1970 censuses. Nevertheless published figures in Wikipedia and other encyclopedias still state racial composition by total population rather than by householders.

1. Counties with zero blacks (to 1960) or black households (1970 and after) in at least one census are shaded dark red
• Counties with at least three consecutive censuses with zero blacks or zero black households are bolded as well as shaded dark red
2. Counties that do not fulfil criterion 1) but have fewer than ten blacks (to 1960) or no more than five black households (1970 and after) in at least one census are shaded red
3. Counties that do not fulfil criteria 1) or 2) but have no more than twenty-five black households in at least one census since 1970 are shaded pink
4. Counties which had at least ten blacks in every census to 1960 and at least twenty-six black households in every census since 1970 are unshaded

County	State	Total black population									Number of black households
		1880	1890	1900	1910	1920	1930	1940	1950	1960	1970	1980	1990	2000	2010
Boone	Illinois	48	48	53	43	35	16	32	13	33	26	32	48	107	321
Edwards	Illinois	83	86	103	73	78	65	51	9	17	0	0	1	2	5
Ford	Illinois	110	138	99	87	96	47	29	21	55	22	15	0	10	11
Johnson	Illinois	133	218	259	164	116	171	156	60	23	4	9	9	8	11
Lee	Illinois	64	63	70	62	117	328	501	598	797	100	129	148	153	176
Ogle	Illinois	107	82	51	33	19	26	17	16	31	8	16	12	53	122
Stark	Illinois	4	14	6	9	7	6	10	9	10	0	3	4	2	7
Randolph	Indiana	591	606	596	235	214	136	99	123	90	23	29	12	23	28
Steuben	Indiana	8	6	8	22	10	21	8	12	25	7	16	19	42	40
Wabash	Indiana	254	141	151	167	174	128	87	53	67	18	23	8	28	29
Cass	Iowa	21	12	9	22	31	32	1	22	14	1	0	1	9	14
Page	Iowa	247	232	234	262	250	225	50	106	48	12	0	9	17	15
Brown	Kansas	263	422	488	457	450	381	291	173	138	47	36	79	72	58
Chautauqua	Kansas	144	224	184	118	143	135	87	52	55	15	8	2	3	6
Doniphan	Kansas	914	767	813	683	597	633	587	403	372	99	104	50	46	56
Osborne	Kansas	61	106	118	69	16	26	18	25	23	4	7	6	2	2
Allen	Kentucky	1,069	1,098	1,042	910	798	482	368	246	265	76	62	60	81	64
Butler	Kentucky	820	773	725	561	475	397	272	149	72	20	13	7	11	14
Casey	Kentucky	608	516	504	278	168	130	75	65	45	9	22	6	10	17
Clay	Kentucky	706	413	564	494	449	299	393	471	455	73	86	112	114	104
Clinton	Kentucky	311	188	175	94	65	55	52	47	30	9	7	0	7	12
Cumberland	Kentucky	1,567	996	922	1,024	915	798	780	574	478	124	143	150	117	98
Edmonson	Kentucky	199	452	458	439	360	187	154	125	102	25	25	44	11	14
Jackson	Kentucky	45	54	19	22	4	18	21	3	0	2	0	0	1	3
Knox	Kentucky	662	778	754	1,059	677	595	619	515	390	126	101	91	106	107
Laurel	Kentucky	267	555	654	657	333	303	344	281	254	58	32	66	119	144
Leslie	Kentucky	28	32	75	132	98	19	25	7	0	0	0	0	3	3
Lewis	Kentucky	229	177	175	141	88	62	30	23	20	5	23	13	9	11
McCreary	Kentucky					28	27	38	31	0	4	10	0	1	4
Monroe	Kentucky	661	540	682	705	597	484	450	364	351	106	113	146	123	103
Owsley	Kentucky	89	84	73	75	56	41	51	33	15	3	0	0	2	3
Pulaski	Kentucky	1,196	1,291	1,336	1,187	976	861	814	777	622	179	179	224	217	269
Rockcastle	Kentucky	437	155	157	125	70	75	29	11	2	0	0	0	2	1
Russell	Kentucky	354	266	294	207	225	224	240	153	142	24	33	44	37	37
Whitley	Kentucky	237	769	752	1,111	600	299	365	180	150	33	54	56	35	38
Garrett	Maryland	112	185	126	107	45	24	4	9	39	0	1	0	12	55
Antrim	Michigan	5	32	49	35	26	21	6	4	7	0	4	4	13	10
Hillsdale	Michigan	59	48	35	22	19	11	5	31	28	15	25	12	37	54
Missaukee	Michigan	0	0	7	15	16	11	3	2	1	0	2	2	5	8
Sanilac	Michigan	32	15	11	8	6	8	8	7	4	3	4	2	15	24
Camden	Missouri	114	97	95	46	31	13	12	5	11	3	4	27	39	67
Douglas	Missouri	26	33	27	8	0	0	0	2	2	0	0	0	2	2
Gasconade	Missouri	167	86	68	41	53	43	8	9	2	2	0	8	10	10
Ozark	Missouri	13	12	26	16	6	4	0	9	0	0	0	0	3	2
Putnam	Missouri	23	35	16	20	8	1	1	2	2	0	0	0	1	0
Stone	Missouri	28	10	4	1	0	0	0	2	3	0	0	0	7	10
Taney	Missouri	4	3	2	1	4	1	0	5	0	0	0	5	38	151
Warren	Missouri	954	725	614	478	315	327	305	259	356	116	173	136	178	226
Keya Paha	Nebraska		24	0	7	12	1	2	0	0	0	0	0	0	0
Allegany	New York	377	331	353	325	216	169	162	142	102	30	31	21	70	66
Genesee	New York	151	129	161	162	191	339	343	429	741	251	326	370	370	456
Livingston	New York	293	247	232	344	357	473	515	608	839	221	197	206	219	220
Orleans	New York	224	194	108	147	112	84	199	319	1,462	423	493	488	499	522
Tioga	New York	445	360	309	242	161	146	155	177	189	86	100	82	98	118
Wyoming	New York	98	57	47	88	94	76	322	581	807	55	26	33	34	23
Avery	North Carolina					243	277	259	204	151	20	13	20	16	26
Mitchell	North Carolina	503	553	536	348	56	56	58	48	38	9	12	13	12	14
Wilkes	North Carolina	1,924	2,034	2,437	2,591	2,322	2,696	2,825	2,707	2,694	665	928	1,060	1,098	1,183
Yadkin	North Carolina	1,544	1,368	1,174	1,187	1,102	986	1,175	1,097	1,128	316	410	449	455	455
Clinton	Ohio	1,463	1,296	1,199	939	730	633	644	633	680	226	199	261	315	332
Fulton	Ohio	55	23	21	6	13	25	6	23	36	10	5	17	32	53
Major	Oklahoma				90	45	41	30	20	12	4	2	1	3	7
Bradford	Pennsylvania	537	599	309	234	137	115	140	75	57	18	45	40	52	75
Cameron	Pennsylvania	8	84	39	31	46	21	22	6	11	2	7	0	6	5
Huntingdon	Pennsylvania	280	315	352	305	717	930	1,087	995	958	143	166	173	178	181
McKean	Pennsylvania	326	299	302	251	313	217	191	80	75	18	17	37	24	38
Potter	Pennsylvania	27	55	65	54	25	9	20	34	47	7	18	7	20	20
Snyder	Pennsylvania	19	5	3	4	3	2	14	53	156	2	7	10	50	80
Tioga	Pennsylvania	115	91	85	73	106	89	119	119	77	19	45	40	55	64
Union	Pennsylvania	133	52	65	59	28	32	214	514	649	6	28	57	73	151
Wayne	Pennsylvania	31	33	18	20	48	100	174	315	383	3	3	25	94	149
Campbell	South Dakota	0	0	0	0	1	0	0	1	0	0	0	0	0	0
Hutchinson	South Dakota	0	0	1	3	0	0	0	2	1	0	0	0	1	6
Lawrence	South Dakota	97	63	106	177	50	28	17	12	6	7	0	10	15	24
Sully	South Dakota	5	7	6	49	58	48	18	14	11	0	0	0	0	1
Turner	South Dakota	0	0	1	0	17	1	0	1	0	0	0	0	2	4
Carter	Tennessee	628	697	661	660	569	528	503	402	435	96	157	156	203	271
Cocke	Tennessee	1,447	1,307	1,261	1,051	849	726	671	635	712	175	190	196	253	263
Grainger	Tennessee	829	685	650	483	447	306	255	208	177	43	64	40	29	38
Hancock	Tennessee	482	727	273	481	256	162	234	110	99	27	10	22	12	6
Hawkins	Tennessee	2,641	2,268	2,154	1,805	1,201	1,069	1,001	1,061	1,050	285	301	329	359	321
Henderson	Tennessee	3,016	2,365	2,637	1,918	1,766	1,693	1,701	1,701	1,713	456	575	663	826	931
Jefferson	Tennessee	2,500	2,153	2,174	1,639	1,428	1,276	1,088	974	926	223	295	281	366	358
Johnson	Tennessee	470	350	368	377	306	301	193	178	138	36	24	20	24	25
Scott	Tennessee	157	366	335	97	16	15	1	6	3	0	0	16	2	2
Unicoi	Tennessee	119	219	130	131	4	13	11	11	5	3	5	3	2	6
Wayne	Tennessee	1,069	884	1,144	845	681	406	394	270	206	64	47	40	44	40
Kane	Utah	0	1	0	0	0	0	0	1	4	0	0	0	1	2
Floyd	Virginia	1,274	1,175	1,075	837	675	515	502	490	532	104	96	112	130	133
Grant	West Virginia	503	379	262	253	232	240	249	242	232	47	57	46	35	38
Morgan	West Virginia	197	275	220	177	150	178	201	147	126	28	37	31	32	38
Preston	West Virginia	206	134	162	151	147	92	58	124	100	19	25	33	30	39
Ritchie	West Virginia	64	36	26	26	13	7	9	5	1	0	0	2	7	8
Upshur	West Virginia	201	256	221	226	196	201	155	92	71	19	20	29	22	27
0 blacks/0 black households		5	4	3	3	4	5	8	1	8	20	22	19	3	3
<10 blacks/≤5 black households		11	9	13	12	13	15	21	23	22	35	32	30	22	17
≤25 black households											59	56	53	47	43
0 blacks/0 black households		5.32%	4.26%	3.19%	3.19%	4.26%	5.32%	8.51%	1.06%	8.51%	21.28%	23.40%	20.21%	3.19%	3.19%
<10 blacks/≤5 black households		11.70%	9.57%	13.83%	12.77%	13.83%	15.96%	22.34%	24.47%	23.40%	37.23%	34.04%	31.91%	23.40%	18.09%
≤25 black households											62.77%	59.57%	56.38%	50.00%	45.74%

Conclusions:

Given that the great majority of rural areas outside the plantation South excluded blacks after Reconstruction — especially in the Plains and Mountain States as can be seen from the South Dakota counties shown above — it cannot be said that the persistent Republican counties were exceptional. In fact, many consistently Republican counties in far eastern Tennessee, the Eastern Panhandle of West Virginia, the eastern Pennyroyal of Kentucky, and parts of eastern Kansas notably never entirely excluded blacks.

The documented sundown county of Scott in Tennessee (heavily discussed in Sundown Towns) and the highly probable sundown county of Ritchie in West Virginia (not discussed at all; I intend to discuss this case in a supplementary post) lie geographically isolated from non-sundown persistent GOP counties in the same states, although the probable though undocumented sundown counties of Tyler and Doddridge border Ritchie County and have historically been almost as Republican. Both Scott and Ritchie are adjacent to secessionist and Democratic sundown counties, as are other counties in the above table in Kentucky and Missouri.

Discussing with Löwen before he died, we both agreed that it was generally easier to confirm the sundown status of counties which once did have significant numbers of black residents. Counties that never had any black residents — like those in the Great Plains or Kane County, Utah — we both agreed as very difficult to research. What the smallest maximum pre-sundown black population that can be easily researched is does not seem to be easily measured: although the Plains counties were never researched, Löwen did research a great deal the sundown policies of the entire Ozark region in Missouri and Arkansas, and the above table shows above that even at the end of Reconstruction this region had very few blacks. Missaukee County in Michigan is another similar case — not discussed individually by Löwen, but as part of a wider sundown region.