“Let our ballots secure what our bullets have won”: Union Veterans and the Making of Radical Reconstruction?

Editor’s Note: This post is part of the Detailed: A Semi-Occasional series within Muster. Read the introductory post here. 

The passage and enforcement of the Reconstruction amendments is one of the most remarkable expansions of political rights in world history. Political scientists studying the expansion and contraction of political rights treat Reconstruction as a classic example that illustrates how these changes are driven by the strategic interests of political elites. They argue that electoral competition between rival elites or political parties drives them to expand suffrage when they expect new voters will give them decisive support.[1] From this perspective, recognizing that restored representation to white Southerners would empower the Democrats, the Republican party strategically extended suffrage rights and other protections to African Americans to win elections in the South and retain control of the federal government.[2] Thus, Republican commitments to equal civil and political rights were only as deep as the electoral value of African American ballots.

While true in part, this isn’t the whole story. Rather than rush to give and protect rights for African Americans, Republicans pursued expansions of civil and political rights fitfully, and were deeply concerned that “Radical” Reconstruction policies would alienate Northern voters.[3] Where did popular support for this revolutionary agenda come from? In newly published research, I show that the war turned Union veterans into a pivotal constituency that voted and mobilized for Republicans and their Radical reforms in the critical elections that shaped Reconstruction.

Scholars have underappreciated the important role of veterans in shaping post-war politics .[4] Returning Union veterans comprised at least 20% of the post-war Northern electorate. Veterans’ organizations, including the Grand Army of the Republic, were active participants in political conventions and campaigning—particularly for Republicans—in 1866 and 1868 elections. And the intense experiences of military service may have primed soldiers for political change. Through their sacrifices to defeat the Confederacy and their experiences with slavery and African Americans, Union soldiers may have developed commitments about the meaning of the war, what had been achieved, and how victory should be preserved: what Barbara Gannon calls “The Won Cause.”

Yet, figuring out whether Union soldiers’ views on slavery and racial equality were changed by the war and what the political consequences of that were is bedeviled by two thorny problems:

1. Measurement: What counts as evidence of soldiers’ political views, as a whole? The diaries and letters of individual soldiers, the commonly used sources, have had differing interpretations. And, since we can’t collect these for all soldiers, it is reasonable to ask: whose diaries and letters are we reading? How representative are they of the Union Army?
2. Causality: How did Union veterans acquire their political commitments? Even if we accept evidence that soldiers disproportionately supported Republicans (such as, the soldiers’ votes in the 1864 Presidential Election), one might reasonably ask: did veterans already possess these views when they enlisted, or were they transformed in the crucible of war?

I bring some quantitative tools of social science to bear against both of these problems. First, I work to clearly and systematically measure the post-war political support veterans gave to Republicans and Reconstruction policies in elections. Second, I employ rigorous statistical analyses that, with transparent and plausible assumptions, enable us to evaluate whether military service caused a change in soldiers’ political commitments.

Union Veterans Increase Post-War Republican Votes

Casting the widest net, I examine whether higher levels of Union enlistment caused counties to experience greater support for Republicans in elections after the war. I calculate the fraction of votes cast for Republicans using county election returns for federal elections collected by the Inter-university Consortium for Political and Social Research. I link data compiled by the American Civil War Research Database on the military records of virtually all Union Soldiers with the 1860 US Census to calculate the enlistment rate among military-aged males for counties in eight states where soldiers’ residence was recorded.[5]

A skeptical reader might raise the objection that counties with higher enlistment may have had more votes for Republicans even before the war, and that enlistment did not cause any changes. To address this concern and provide evidence of the causal impact of wartime enlistment, I employ a method called differences-in-differences. In essence, this method compares changes in Republican voting from before to after the war in counties within the same states with higher versus lower enlistment.

After the war, counties with greater enlistment saw much larger increases: compared to 1860, counties in the top 25% of enlistment saw Republicans gain 10 percentage points in 1866, while counties in the bottom 25% gained only 2.

By looking at changes within counties, it cannot be that unchanging attributes of counties (e.g., a history of abolitionist mobilization or pre-war support for Republicans) explain this pattern. To interpret these results as the causal effect of enlistment rates on Republican voting, we need only assume that counties with different levels of enlistment would have had similar trends in Republican votes, had the war not happened. While this is impossible to know, we can see in the provided figure that, prior to the war, counties with different enlistment rates had the same trends in Republican voting. Thus, we can plausibly say that wartime enlistment caused substantial gains for Republicans in key post-war elections.

Yet, we cannot say for sure that these aggregate effects are the result of soldiers themselves changing their votes; it could be that non-veterans in places with greater enlistment became more supportive of Republicans, through, say, the canvassing efforts of veterans.

Union Veterans’ Combat Experiences Affected their Partisanship

To find out whether the war changed veterans, I restrict my focus to soldiers and examine whether different wartime experiences affected their political views after the war: did increased exposure to combat casualties make men more Republican? A simple comparison might lead us astray: Republican and Democratic soldiers may have chosen to join different units and at different times, and Army commanders may have taken unit partisanship into account when issuing battle orders.

To isolate the effect of combat casualties on partisanship, I take advantage of a “natural experiment”: men joining the same regiment enlisted, moved, and fought together during the war. Yet, in battle, different companies were arrayed at different points along a short line, and as a result of the contingencies of battle, some arbitrarily experienced more casualties than others. If we believe these differences were effectively random, like the assignment to treatment and control in a clinical trial, we can attribute differences in partisanship to the causal impact of company casualties.

Biographical directories compiled in nine Indiana counties in 1874 record the partisanship for nearly thirty thousand people. Digitizing and linking this data to the military records of soldiers from those counties, I find that within the same regiments, soldiers serving in companies with more combat deaths become more likely to label themselves as Republican versus Democrat. And consistent with the effectively “random” exposure to casualties, there were no such differences in pre-war demographic traits. This is compelling evidence that one key feature of military service, combat, turned individual soldiers into Republicans.

Union Veterans Backed Radical Policies

Did veterans merely support Republicans without actually backing Radical policies? To answer this, I combine data on enlistment with town-level returns in referenda over Black suffrage in Iowa and Wisconsin.[6] Importantly, these states held referenda beforeand after the war.  Looking at changes in support for suffrage after the war, towns with higher enlistment saw greater increases in support for Black suffrage.

These are aggregate patterns but mathematically it must be that, after the war, support for suffrage increased more among veterans than among those who stayed home. This is another difference-in-differences analysis: if we believe that, absent the war, soldiers and non-soldiers would have had similar trends in support for Black suffrage, this is compelling evidence that military service caused substantial increases in support for suffrage.

Critically, the Wisconsin referendum took place in November 1865. It occurred well before national Republican leaders publicly embraced any of the key policies of Radical Reconstruction. Thus, these results suggest that returning soldiers may have pushed the party toward these positions from below.  This would turn the electoral competition argument on its head: it was veterans and their Radical allies who, as pivotal voters for the Republican party, pushed party leaders to embrace a stronger Reconstruction agenda.

Conclusions

In Congressional elections of 1866, Republicans won a supermajority that empowered them to legislate Radical Reconstruction. They benefited from increased support in places with more enlistment: a ten point increase in enlistment was enough to secure the seats needed to override Johnson’s veto. I show that these effects were the result of war experiences changing both soldiers’ partisanship and their direct support for Black suffrage.

My conclusions about the role of veterans in the Republican party during Reconstruction parallels recent work in political science showing that it was the local-level incorporation of African American voters and mobilization by labor unions that pushed Northern Democrats to embrace civil rights in the 20th century.[7]  More work remains to be done to understand the nature of veterans’ support for Reconstruction and its consequences.  Nevertheless, we can draw important lessons from these cases. Rather than exemplify the strategic logic of elite-led rights expansions, Reconstruction shows that expanding and protecting political rights depends on buy-in from broad grassroots coalitions among those already enfranchised.

[1] Ben Ansell and David Samuels, Inequality and Democratization: An Elite-Competition Approach. (New York: Cambridge University Press 2015); Humberto Llavador and Robert Oxoby, 2005. “Partisan Competition, Growth, and the Franchise.’’ Quarterly Journal of Economics 120(3):1155–1189;  Dawn Teele, 2018. “How the West Was Won: Competition, Mobilization, and Women’s Enfranchisement in the United States.’’ Journal of Politics 80(2):442–461; Giovanni Capoccia and Daniel Ziblatt, 2010. “The Historical Turn in Democratization Studies: A New Research Agenda for Europe and Beyond.’’ Comparative Political Studies 43(8–9):931–968.

[2] Richard Valelly, The Two Reconstructions: The Struggle for Black Enfranchisement. (Chicago: University of Chicago Press, 2004).

[3] Xi Wang, Trial of Democracy: Black Suffrage and Northern Republicans. (Athens: University of Georgia Press, 1997).

[4] Though, Cf. Mary Dearing, Veterans in Politics: The Story of the G.A.R. (Baton Rouge: Louisiana State University Press, 1952); Brian Jordan, Marching Home: Union Veterans and Their Unending Civil War (New York and London: W. W. Norton and Company, 2014)

[5] Connecticut, Massachusetts, Maine, Illinois, Indiana, Iowa, Vermont, and Wisconsin.

[6] Data from Iowa referenda are available from https://www.icpsr.umich.edu/web/ICPSR/studies/4284. Data from the Wisconsin referenda were provided by Michael McManus.

[7] Eric Schickler, Racial Realignment: The Transformation of American Liberalism, 1932–1965 (Princeton: Princeton University Press, 2016).