Authors: Ira Katznelson
Whereas the chamber’s Republicans voted with perfect cohesion to support the Brooks amendment, the nonsouthern Democrats divided 12–7. Coming from Arizona, Idaho, Indiana, Iowa, Pennsylvania, and Rhode Island, none of the negative voters either supported or represented states with a poll tax. What motivated them was a concern that approval of this amendment might ultimately doom the legislation and split the party. The degree of likeness between southern and nonsouthern Democrats
79
was considerably higher than Republican and nonsouthern Democratic likeness,
80
and surprisingly closer to the more expected low likeness of southern Democrats and Republicans.
81
In short, this was a vote that pitted a unanimous Republican Party against Democrats from all regions who found themselves torn between principle, both for and against, and instrumental calculation.
For more than half a century, civil rights voting in the House had provided a form of theater. Those casting votes about lynching or the poll tax understood that southerners in the Senate would filibuster to block the legislation. For the first time in decades, the subject of soldier voting detached this veto instrument. Like their House compatriots, southern senators in the main understood that they could not afford to be seen to block soldier voting. Moreover, most southern representatives found explicit talk promoting white supremacy to be discomfiting and instrumentally counterproductive. They appreciated that such language, as distinct from constitutional argument, risked appearing unpatriotic, insufficiently committed to the war against the dictatorships. Expressing this ambivalence, Nat Patton, a former judge and a member of the House from East Texas, remarked in late August, “I don’t want to disturb the poll tax, but I don’t want to deprive the boys of their chance to vote.”
82
Cross-pressured this way, and also aware that “their objections were dealing a death blow to their party’s campaign in the North,” where “the Negro vote in northern cities which swung away from Republicans in 1933 was going back to the party of Abraham Lincoln as a result of the poll tax discussion,”
83
the representatives from the eight poll tax states agreed not to block the House’s consideration of soldier voting by procedural objection, and thus permitted the chamber to send the bill to a congressional conference by unanimous consent.
84
Nonsouthern Democrats were keenly aware that the era’s Great Migration was bringing new black voters to their constituencies who might decide close elections.
85
The poll tax issue was especially attractive to these Democrats because it offered a chance to endorse and secure black sentiments without challenging or offending their white constituents. Most southern representatives, in turn, knew the poll tax was the least important of the barriers to black voting, and the one most difficult to defend at a time of war.
86
So they principally looked to other, less emblematic but more essential, means to preserve the region’s electoral exclusions.
Furthermore, southern members in the main understood that if they went too far and were too insistent in opposing soldier voting because of its poll tax feature, they might endanger the period’s mostly tacit but sometimes explicit agreement to leave well enough alone south of the Mason-Dixon Line. Especially during the war, there was little interest on the part of Congress to press the South to transform its exclusionary franchise, for much the same reason the armed forces had justified the decision to maintain military segregation. John J. McCloy, then assistant secretary of war, who headed the Advisory Committee on Negro Troop Policies, wrote in July 1942 that this was no time to confront racial prejudice and discriminatory acts “irrespective of whether the White or the Colored man is responsible for starting them,” adding, with respect to segregation, “I doubt that you can convince the people of the United States that the basic issues of freedom are involved in such a question.” This position restated the policy of racial separation that President Roosevelt had endorsed in 1940, which argued that “changes now would produce situations destructive to morale and detrimental to the preparation for national defense. . . . It is the opinion of the War Department that no experiments should be tried with the organizational set-up of these units at this critical time.”
87
Writing in the
Atlanta World
, the South’s principal black newspaper, the African-American columnist Charles Howard Sr. wondered at the start of the soldier-vote debate in May 1942 whether “the law makers in Congress are going to disfranchise a couple of million white boys to keep a couple of thousand Negroes from exercising the prerogative of American citizenship.” Observing that “it is going to be pretty hard to grant the ballot to one group and deny it to the other,” he quite presciently predicted that “the anti-Negro group in Congress is pretty ingenious.”
88
So it proved to be. Faced with these various conflicting currents of language and value, southern members crafted strategies, made choices, and offered justifications that could somehow reconcile, or appear to reconcile, democratic norms with the desire shared by committed racists and moderates alike to maintain segregation and determine the character and pace of change without external intervention.
Even with the poll tax suspension, the 1942 act met this test. With the compliant assistance of fellow party members from other parts of the country, the Democratic majority crafted a bill that through its timing, burdensome procedures, and states’ rights protections that minimized the federal role offered only the appearance of remedies for the severe logistical problems posed by soldier voting.
This had been an issue the Democrats had initially preferred to avoid. The first bill to clarify how that might actually happen had been introduced in April 1942 by Joseph Martin, the Republican minority leader and chairman of the Republican National Committee, who declared that the three million individuals scheduled to be serving by November “ought not to be deprived” of the chance to vote.
89
President Roosevelt, by contrast, “not at all certain that much can be done about it,” thought the task of soldier voting to be so imposing that, in May, he simply advised the Department of War and the Department of the Navy to “remind the boys by posting notices . . . summarizing laws in each state.” He also considered issuing an executive order commanding the armed services to work with existing state regulations to facilitate voting by absent soldiers.
90
As Martin had insisted, soldier voting was an issue Congress was mandated to consider. Even before U.S. entry into the war, the Selective Service Act of 1940 had stipulated that conscripts be permitted to vote in all elections under the laws of their state, either in person, without being required to take a leave of absence of more than one day to do so, or by absentee ballot (in turn, they were not eligible to vote in other states in which they were stationed). The diversity and patchiness of existing state laws in 1942 made necessary at least some degree of federal inducement and oversight along the lines Martin proposed to advance cooperation between state authorities and the military if soldiers were to have any realistic chance to vote. Moreover, a high premium was placed on the successful functioning of American democracy in the dire circumstances of that spring, marked by the surrender of American forces to the Japanese on the Bataan Peninsula and on Corregidor, the drive led by Germany’s Field Marshal Erwin Rommel from Libya toward Alexandria, the visible ban in France of Jews from public facilities, including restaurants, libraries, and public gardens, and the less visible start to mass killings at Auschwitz.
91
With only Rankin objecting (“It seems to me that we have enough on our hands to whip Germany, Italy, and Japan without pandering to those vicious elements who are constantly waging war on private enterprise and on the white people of the South”
92
), the House conferees accepted the Senate version, including the poll tax suspension and the inclusion of primary elections. The bill also suspended the requirement that prospective voters had to register in person. Passed in mid-September by overwhelming votes (47–5 in the Senate; 248–53 in the House) that were characterized by uncommonly unified voting by Republicans and nonsouthern Democrats, the statute, as we have seen, rendered these features moot in practice because only 1 percent of the armed forces succeeded in utilizing its procedures. It also did not disturb in any way the capacity of each state to establish qualifications for voting and judge whether a citizen had met them. To Maryland’s Millard Tydings, who had asked, “Is it the Senator’s contention that any proper limitation upon the qualification of a voter is not wiped out by the measure now proposed?” Senator Green replied, “That is correct . . . the bill has nothing to do with the qualification of electors. They all remain as they were, whether they are right or wrong, constitutional or unconstitutional.”
93
IV.
W
HEN
C
ONGRESS
again debated soldier voting in 1944, voting patterns had changed dramatically. The two crucial House votes—the 328–69 passage of the Rankin states-rights version, and a narrower 273–111 vote to agree to the conference report, which tilted ever so slightly in the direction preferred by President Roosevelt—were passed by a coalition of Republicans and southern Democrats. Only nonsouthern Democrats supported the type of federal ballot that he had proposed.
As discussion unfolded, the southern and nonsouthern wings of the Democratic Party first tried to find common ground to maintain party unity. That effort did not succeed. The maximum the South was willing to countenance and the minimum that pro-administration nonsouthern Democrats were ready to tolerate did not mesh. The Republicans, in turn, faced with the chance to build a winning coalition, performed a volte-face from their assertive rhetoric and vote pattern of 1942, shifting to a states’ rights position that meshed with that of the southern Democrats. As a result, soldier voting played a key role during the early stages of the development of ties between southern Democrats who feared for their social order and Republicans who especially disliked the New Deal’s alteration of the balance between capital and labor, who found the administration’s centralization of policy and administration objectionable, and who yearned for another chance to govern. “More than any other episode during the war, the controversy over the soldier vote cemented a Southern Democratic-Republican alliance,” a detailed study of wartime public policy concluded.
94
“I am actually getting to the point,” remarked Cotton Ed Smith, “where I turn to the Republicans when I want the real fundamental constitutional laws of this country adhered to.”
95
The collision of regional and party interests with the imperatives of national unity had been minimized and managed in 1942, albeit at the cost of adopting an ineffective statute. By 1944, striking political changes had altered the stakes of soldier voting for each of the three partisan groups in Congress. The Republicans were enjoying a remarkable resurgence, one that had been advanced, many observers believed, because millions of soldiers had not been able to vote in 1942, when the party had elected 209 members of the House, picking up 47 seats, close in number to the 220 secured by the Democrats.
96
The Republican Party also gained 10 senators, up to 38, a number sufficient to control efforts at cloture. For the first time in a decade, moreover, Democrats secured less than half the party vote. Republican control of both chambers now was within hailing distance. Even more enticing was the prospect of the 1944 election. With Roosevelt’s candidacy uncertain, the Republicans could anticipate a campaign against either a new candidate or an aging candidate running for an unprecedented fourth term. In turn, the Democratic Party became more proportionately southern than it had been at any time since 1932, as most of its losses were sustained in nonsouthern competitive races.
With wartime planning for production and the allocation of labor, price controls and rationing, corruption and war profiteering, recriminations and qualms about loyalty, and the inevitable clumsiness of wartime management and resentment at how executive and emergency powers were being deployed, suspicion of the federal government had grown sufficiently to bring the Republican Party’s smaller domestic government ethos in tune with popular disillusionments and resentments. Finding allies among southern Democrats, Republicans had begun to close a raft of New Deal agencies, including the National Resources Planning Board, the National Youth Administration, the Civilian Conservation Corps, and the Works Progress Administration, and sought to return Congress to its pre–New Deal role of “negotiating the local and regional adjustments to the national policies the president was advocating.”
97
How ballots for soldiers would be organized, it was widely understood, might matter quite a lot in 1944. The federal government clearly could not decline to facilitate political participation by the country’s far-flung military. But how to do so was a subject of intense and distinct interest to each party’s legislators. Eleven months before the election that would produce a new term for Franklin Roosevelt and place Harry Truman in the vice presidency with 25,613,916 votes to 22,017,929 for the Republican ticket of Thomas Dewey and John Bricker, George Gallup appraised the two parties as “running neck in neck in terms of voting strength among the civilian population.” He thus projected that “if the presidential election were being held at this time, the outcome would therefore be determined by the soldier vote.”
98
With many Americans exhausted by the war and skeptical about a president running for a fourth term, a June 1944 Gallup poll put the distance between Roosevelt and Dewey at only two points. That summer, two students of that election found that the age group of those twenty-one to twenty-nine was “11 points more Democratic than is the entire voting population,” and thus concluded that “the Democrats will be losers to the extent to which service men do not vote.”
99
Another, earlier analysis by Gallup had calculated the party gap in the military to be significantly higher, 61 percent Democratic to 39 percent Republican.
100
An August analysis also saw a near dead heat in electoral votes, assessing 248 for Roosevelt and 229 for Dewey.
101
Even with two million of the ten million persons under arms disqualified by their age, the military electorate promised to be pivotal.