Slavery or Sheep? The Antebellum Realignment in Vermont 1840-1860

1^ I am grateful to the staff of the Vermont State Archives in Montpelier, especially Scott Reilly, for their assistance in finding and copying a great many pages of handwritten antebellum election reports. Gregory Sanford, the Vermont state archivist, gave me detailed comments on an early draft, saved me from some errors, and directed me to important sources. I also thank Larry Bartels, John Geer, Michael Donnelly, Karen Jusko, and participants in a seminar at Princeton’s Center for the Study of Democratic Politics for their helpful suggestions. Masha Krupenkin carefully turned the handwritten township voting returns and population figures into machine-readable data, and I am deeply grateful to her for her hard work and professionalism.

2^ Of course, some disagree. For example, Egnal (2009) defends the Beards’ economic interpretation of the Republican surge.

3^ For the critical 1848, 1856, and 1860 presidential elections, see Luebke (1971); Gienapp (1986, especially 482-551); Huston (1987, chap. 6); and Holt (1999, 377 and passim).

4^ On the Anti-Masonic party in Vermont, see Ratcliff (1995). That party in New England generally is ably discussed in Goodman (1988). On the Liberty Party in Vermont and nation-wide, see Johnson (1979; 2009). Free Soil is treated in Rayback (1970), Mayfield (1980), and Alexander (1990). Vermont elections without the required majority winner are discussed at: http://vermont-archives.org/govhistory/governance/Majority/summary.htm..

5^ Vermont’s population grew by just 12% from 1830 to 1860, while the number of voters rose by 45% from 1836 to 1856 (Arnold 1980, 14; Burnham 1955, 814).

6^ References to county-level vote returns in presidential years are taken from ICPSR file 08611, provided by the Inter-University Consortium for Political and Social Research at the University of Michigan, accessed September 17, 2010. Aggregate gubernatorial returns are from Carter (1989). County and town returns for presidential and gubernatorial elections and for referendums are reported by the author from the original records in the Vermont state archives in Montpelier. The Vermont State Archive lists statewide returns at http://vermont-archives.org/govhistory/elect/index.htm. Gubernatorial results are at http://vermont-archives.org/govhistory/elect/results1/pdf/stoff1gov.pdf.

7^ The Know-Nothings did capture the Vermont Council of Censors in 1855, the body responsible for amendments to the state constitution, but this one-time victory was never repeated for any statewide office. The Council was elected at town meetings, where the Know-Nothings’ secretive and well organized party was at an advantage, not in a regular election (Hand 2002, 7-9).

8^ All correlations in this paper are Pearson r’s, with county observations weighted by the total votes cast in the election being explained. Thus the correlation between the 1860 vote and the 1853 referendum is weighted by the votes in 1860.

9^ The successive temperance votes in each town were more volatile and less tied to party than might be expected (Bigelow 1970, chap. 4), perhaps because pragmatic Vermonters were learning from experience what worked and what did not, and because the initial four votes were taken at annual town meetings, whose attendance was variable. For further discussion, see http://vermont-archives.org/govhistory/governance/Referendum/ref.html.

10^ Some interpretations of this period lay emphasis on Protestant evangelical revivalism and moral fervor as the sources for temperance and abolitionism (Cawardine 1993; Strong 1999). But as those sources note, local conditions mattered. Since the two movements are unrelated in county- or town-level voting, it is difficult to see how they could have a common cause in Vermont. Similarly, efforts to tie either one to denominational differences at the town level find no substantial relationships. Quakers seem to have been disproportionately for abolition and Episcopalians disproportionately not, but the great majority of Vermonters belonged to neither denomination (Bigelow 1970, 19-44, chap. 4). The key distinction may have been “Old Light” versus “New Light” theologies, and those divisions were intra-denominational, and thus are not easily measured (Roth 1987; Bassett 2000).

11^ Gregory Sanford reports that some years ago, the Georgia archivist was unable to find a copy of the motion in the state legislative records, which raises the sad possibility that the story is apocryphal. The original source in Vermont historiography appears to be Crockett (1921, 444), which quotes other colorful remarks about Vermont by antebellum Southerners, but without giving formal citations.

12^ Sheep counts are taken from the U.S. Agricultural Census of 1840. See also Benton and Barry (1837) for 1836 figures. The two county-level counts correlate very strongly (r = .99).

13^ Of course, not everyone in the North favored tariffs, and not everyone in the South opposed them. Merchants engaged in north-south trade were particularly resistant to sectional feelings. See Freehling (1965, 274) on Charleston, and O’Connor (1968, 166-167) on Boston, New York, and Philadelphia.

14^ Wool prices are taken from Wright (1910, 347-349). Connor (1921, 193-194) extends the series further into the twentieth century. I have followed other authors in quoting medium grade prices to illustrate changing price levels, as in Taussig (1889, 152) and in the ditty on the following page. U.S. production was confined mostly to middle and lower grades because it was not competitive with the British on higher grades (Taussig 1889, 144-152; Cole 1926, vol. 1, 300-302). Adams (1944, 140-144) gives wool prices in index numbers and in purchasing power indices, but they do not seem to track nominal prices closely, and I have chosen to set them aside. He also gives tariff rates on wool at various time periods (p. 155). To measure general economic conditions, antebellum wholesale prices are used. They are much more volatile than in the current era, and they track the Panics of 1837 and 1857, as well as newspaper reports of other hard times. A weighted index of wholesale price levels in this period appears in Smith and Cole (1935, 158, 167).

15^ Of course, some towns were exceptions. For example, Bristol is located near the center of sheep-raising, fervently pro-Whig Addison County, but against the mountains, giving it a waterfall. It had mills and forges from its early days (Potash 1991, 104; Munsill 2009, 103-115), and it clung to the Democrats through most of the antebellum period.

16^ Lamoille was the boyhood home and frequent adult residence of Joseph Poland, the editor of the state’s leading abolitionist newspaper from 1840-1844, which was published in Lamoille (Ullery 1894, 321-322). In consequence, Lamoille was the only county where the Liberty Party ran well in 1844 (25%), taking substantial votes from the Whigs and a few percentage points from the Democrats as well. No other county gave the Liberty Party as much as 11%.

17^ In a brief cross-sectional comparison of a few counties in 1844, Benson (1961, 158-159) doubts this association for New York State. I hope to take a closer look at those counties over time in a subsequent draft. Freehling (1965) argues that the tariff issue mattered for Southerners, but that (to oversimplify his argument) the debate was really about slavery. Among other evidence, he cites a private letter from Calhoun in 1830 making that point (p. 257). However, without denying that farsighted people saw a connection between the tariff and slavery, nor that it formed a useful rhetorical point (“You wait: The tariff is just the beginning.”), I find it difficult to read and dismiss as window-dressing the very lengthy discussions and elaborate calculations about the tariff in Southern writers at this period, for example, Turnbull (1827). Calhoun himself in his well known and lengthy Senate “speech on the force bill” on 15 February 1833 makes no mention of slavery (Cheek 2003, 411-441). Nor is that surprising: Slavery faced no serious threats in the 1830s. Parallel skepticisms apply to Anbinder’s (1992) argument that the Know-Nothings were often attractive due to their subsidiary stands on temperance or slavery rather than to what they spent nearly all their energy on--their shrill anti-Catholicism. However, all these claims deserve careful consideration, and their impact on mass electorates seems researchable using election returns.

18^ The same is true for the Whigs. The Democrats lost between 10 and 20 percentage points almost everywhere; the Whigs lost between 3 and 10 percentage points almost everywhere. The pattern is closer to a uniform swing against both parties than to the conventional ecological regression assumption, in which each old party’s defection rate to a new party is constant across geographic units.

19^ Texas annexation and the question of whether it would be slave or free also figured in the 1844 election. But tariffs mattered in a way that slavery did not. Vermonters had no difficulty choosing Henry Clay, a slaveholder and author of “the American system” of high tariffs (55%) in place of the single-issue abolitionist candidate of the Liberty Party, James Birney (8%).

20^ In the five presidential elections from 1856 to 1872, GOP votes shares in Vermont were 78%, 76%, 76%, 79%., and 78%.

21^ From 1856 to 1860, Democratic vote shares gained proportionately by more than 25% in just two Vermont counties—Rutland and Addison, the two counties with the most sheep. Wool prices were up in 1860 compared to the Panic of 1857 period and its aftermath, and the Democrats were in office. By contrast, the economy as a whole was somewhat down from the two preceding years.

22^ One need not soft-pedal the horrors of slavery, as Owsley and Craven are sometimes inclined to do, to recognize that Northern abolitionists and Southern “fire-eaters” were locked into a shrill war of words.

23^ Some historians dispute this point, of course. Freehling (1965) argues that the tariff issue mattered for Southerners, but that (to oversimplify his argument) the debate was really about slavery. He cites a private letter from Calhoun in 1830 making that point (p. 257). Without denying that farsighted people saw a connection and that it formed a useful rhetorical point (“You wait: The tariff is just the beginning.”), I find it difficult to read and dismiss as window-dressing the very lengthy discussions and elaborate calculations about the tariff in Southern writers at this period, for example, Turnbull (1827). Calhoun himself in his well known and lengthy Senate “speech on the force bill” on 15 February 1833 makes no mention of slavery (Cheek 2003, 411-441). Nor is that surprising: Slavery faced no serious threats in the 1830s. For similar reasons, I am skeptical of Anbinder’s (1992) argument that the Know-Nothings were more attractive for their subsidiary stands on temperance or slavery rather than for what they spent nearly all their energy on--their shrill anti-Catholicism. However, all these claims deserve careful consideration, and their relevance to mass electorates seems researchable using election returns.