Kennedy was right about Oklahoma. But not about the rest of the evenly divided country.
Oklahoma went for Nixon early and decisively on November 8. With an advantage for the Republican nominee of more than 160,000 votes, or fully 18 percent of the 900,000 votes cast, it turned out to be Nixon’s third largest margin (after Nebraska and Kansas) in the entire election. Indeed Oklahoma was one of very few states where Kennedy’s vote count in 1960 was actually lower than Stevenson’s in the Eisenhower landslide four years earlier. But contrary to Kennedy’s dark election-eve analysis, his religion was not the only factor in his Oklahoma loss. By 1960, its prairie populism more fading memory than active force, Oklahoma was becoming a conservative state where Nixon’s smaller government and stern, cold war–fighting image could resonate.
Elsewhere Kennedy was more fortunate. In the end his five-year quest for national office against enormous odds just barely succeeded. In political terms, all that traveling, speaking, maneuvering, and writing enabled his “positives” (fresh, vigorous, new thinking, Catholic, activist) to keep his “negatives” (inexperienced, unknown risk, bigger government, Catholic) at bay.
The popular vote was the closest since 1888, when the country actually elected the man who got fewer votes, Benjamin Harrison. The Electoral College margin was the smallest since Woodrow Wilson’s second squeaker in 1916. Kennedy got fewer votes than Eisenhower had in 1956 (34.2 to 35.6 million), despite the presence of some seven million additional, eligible voters. In twenty of the states the margin was below 5 percentage points; Nixon took six of them, and one state (Mississippi) went to a slate of Jim Crow–supporting, “independent” electors. Of the thirteen cliffhangers that went to Kennedy, only his advantage in North Carolina was above 4 percent and in five it was below 1 percent.
This was not what Kennedy and his campaign leaders had expected. There had been no preelection boasts of a landslide, but between Harris’s polls and O’Brien’s organizational high command, there had at least been an expectation of a clear if not solid victory. The price of the optimistic mood after the debates was a rough election day and night.
“We started getting down to the nuts and bolts of it and it was just terrible,” recalled Donahue, whose desk was inside O’Brien’s lair. “We were leaking and there was just a little bit of leaking all the time and the leak was just enough to keep you behind. . . . Although I never really thought we could lose—mere stupidity more than anything else—I just started to lose confidence that we could make it big and [then] confidence that we were going to be good and then you just want to win.”
After voting in Boston, once again using his family’s Beacon Hill apartment as his legal address, Kennedy and his wife arrived on Cape Cod in midafternoon. He was on the telephone frequently but not constantly. On this day he was not talking economics with Walt Rostow or domestic policy with Meyer Feldman or rhetoric with Ted Sorensen. His attention was focused solely on Donahue’s nuts and bolts. And his major henchmen were focused on precincts and counties, not Kennedy himself.
Recalled John Bailey, who had traveled with Kennedy from the start, “On election day, Abe Ribicoff and I were sitting in the headquarters in Hartford. I was busy. I had a private telephone on my desk and we were trying to do some things, figure out how much we were going to win by and the rest. The telephone rang. I impatiently picked it up and said Hello. A voice said, ‘Hello, John.’ And I said, ‘Who is this?’ He said, ‘This is Jack.’ ‘Jack who?’ Then he said, ‘It’s the candidate, who do you think it is.’ That day my mind wasn’t on him. It was on Jack Somebody who was running the second or third ward someplace in the city.”
Bailey told Kennedy he would take Connecticut by 90,000 votes. Bailey was not called a master of his craft for nothing; the margin was actually 91,000.
Much, much more was at stake than 8 electoral votes, or even the fact that not counting the four FDR victories Kennedy would be gaining Connecticut for the Democrats for the first time since 1912.
Ever since the advent of national radio networks in the 1920s, election nights in the United States had been paced by the time zones, with results reported east to west. That created the possibility that results in the first states to report, from the eastern third of the country, might influence the voting farther west, where the polls were still open.
The evidence supporting this possibility has always been skimpy, but the Kennedy campaign was nevertheless paying close attention to the incoming numbers. Even a month earlier O’Brien’s weekly national report focused on the reason he was so concerned about Connecticut: “The need for an organizational effort in Connecticut is becoming increasingly more apparent. Polls in Connecticut close at 7 p.m. EST and, because Connecticut is an all [voting]-machine state, results will be known by 8 p.m. (5 p.m. on the West Coast, 6 p.m. in the Mountain Time zone, and 7 p.m. in the Central Time zone). A smashing triumph in Connecticut probably can influence 100,000 to 200,000 votes.”
O’Brien’s analysis was only half right. It neglected the other significant state with a relatively early poll-closing time, Kentucky, which his report had listed in the “even” column. Just as the dimensions of Kennedy’s victory in Connecticut were becoming clear, it became equally clear that Nixon was winning in Kentucky by 80,000 votes, or more than 7 percentage points.
His broader point, however, was theoretically valid. Through mid-evening a torrent of solid Kennedy victories was reported in the eastern time zones, above all on television. The list was impressive: southern New England, New York, Pennsylvania, Maryland, Delaware, the South as far down as Georgia, as well as Michigan. Someone watching it all unfold on television saw a string of positive Kennedy showings and a clear lead in the popular and electoral vote totals.
The Kennedy compound in Hyannis Port at the end of a quiet street contained multiple dwellings. The main house, just inside the entrance off the street, was the redoubt of the candidate’s parents, who mostly stayed there throughout the long evening and night. It was nearby, in the house used by Robert Kennedy, that the nerve center was established: thirty special telephone lines plus four teletype machines, most ending up in a large, first-floor sun porch. There the campaign’s most important political people and more than a few useless hangers-on gathered, along with a half-dozen secretaries. In addition to following the action and making the occasional phone call himself, Kennedy punctuated his frequent visits to the command center with occasional solitary strolls on the large lawn outside.
The house he typically used in the compound was behind the command post. In an effort to provide a form of sanctuary, his wife had invited just one guest that night, William Walton. The three of them had an early meal. But Kennedy’s inquiries about the numbers were constant.
“I told him a couple of lies about how things were going,” recalled Donahue. “I made up a couple of stories which weren’t totally inaccurate about the fact that the voting turnout was heavy in a lot of the cities and it was raining in some of the farm areas. . . . He was really relaxed. He was in and out of the house while I was doing some other things there. . . . I think that I was rather surprised that he could remain quite as calm as he did.”
In addition to the dose of reality supplied by Kentucky, the first shock and keen disappointment was the result in Ohio. It had been the first major state to break Kennedy’s way in the slog to the nomination, and it had been a major focus of effort in the general election. Throughout the final month Harris’s surveys consistently showed a small Kennedy lead, and O’Brien’s weekly surveys of polls and pols just as consistently put the state in the “ahead” column.
And they were all wrong—by a lot. By late in the evening Nixon was well on his way to a near-landslide in Ohio, with a margin of more than 270,000 votes, or 6.5 percentage points. This was one result that produced a worried huddle of the Kennedy brothers and O’Donnell to share the pain and to wonder what had gone terribly wrong. Was it anti-Catholic sentiment in the smaller communities in the southern part of the state? A lighter than expected turnout in Cleveland and other cities? More important, they worried about what the drubbing portended in the other industrialized states of the Midwest as well as Pennsylvania, where the returns at that point were inconclusive.
“We just could not figure it out,” said O’Donnell. “Perhaps that was what was the most frightening. Was it the Catholic thing again? And if it was, what does that mean for the rest of the industrial states? After Ohio, Bobby and I took nothing for granted.”
The second shock came from Wisconsin, another state to which there was a sentimental attachment dating from the primaries, but at least one where the expectation had been of a very close vote. Instead Nixon’s victory margin was 65,000 votes, or more than 3.5 percentage points. O’Donnell recalled, “This meant Minnesota was in trouble; Minnesota was a seesaw right up until the very end.”
As the returns came in from the Midwest and then the West, Kennedy’s leads were shrinking steadily. He appeared to have won in Johnson’s Texas, but tiny margins in New Mexico and Nevada were more than balanced by clear defeats in Oregon and Washington. The first major break came around midnight, not on television but in the print press. The New York Times, probably imprudently, put its first edition to bed with the banner headline “Kennedy Elected President.” Coming just a dozen years after the Chicago Tribune’s historic goof, “Dewey Defeats Truman,” it was an odd decision, more hunch than conclusion, and one not shared in either the Kennedy or the Nixon camp.
The second major break turned out to be a false alarm. Included in the array of machines in use on election night were direct, open telegraph lines connecting the Nixon and Kennedy campaigns with each other and with their national party headquarters in Washington. Their purpose, by agreement, was solely for the transmission of congratulatory or concession statements from the candidates and from President Eisenhower. Shortly after midnight the machine on Cape Cod came alive and printed out a message from Eisenhower congratulating “President-elect Kennedy.” Fortunately Salinger had the presence of mind to look before leaping into the air. He contacted his White House counterpart, James Hagerty, who explained that one of the statements he had prepared in advance had been transmitted through a horrid error. He asked Salinger not to say anything about it; in an example of long-ago comity, Salinger complied.
The Kennedy group was just as surprised three hours later when Nixon himself appeared before cameras and supporters from his encampment at Los Angeles’s fabled Ambassador Hotel to declare that the trends favored his opponent, but not by enough to produce a formal concession statement. Standing behind Donahue while watching Nixon on the television, Kennedy was his familiar detached self, murmuring at one point that his opponent was doing exactly what he would have done. But Kennedy could not have been serious; he had not been anywhere near a television camera at that point precisely because he knew the outcome remained in doubt.
At that tense point in the early morning hours, Kennedy was ahead, but only narrowly. Where counts had been completed he could depend for certain on 161 electoral votes to Nixon’s 135. There were, however, eight states with 141 total votes where counts were still ongoing in extremely tight situations: Pennsylvania, Missouri, Illinois, Minnesota, Michigan, California, Hawaii, and Alaska. To get to the magic number of 269, Kennedy had several possible routes available to him; by contrast, Nixon could not afford to lose more than 6 of the electoral votes at issue, meaning he had to sweep all of the larger states and could lose only the two newly enfranchised ones.
The information available to both sides suggested a Kennedy advantage at that hour; in fact the television showed Kennedy ahead, with 262 electoral votes. That helps explain Nixon’s appearance but also supports Kennedy’s decision to remain mute.
Nixon’s pessimistic statement, however, catalyzed a realization that there was nothing vital to be done before the sun came up. Almost immediately the set was temporarily struck on the sun porch as the senior people sought places for a nap. Kennedy himself walked next door, kissed his mother goodnight, and then returned to his own house and dozed for perhaps three hours.
The first clue that it might be over was the arrival around 6 a.m. of the first carloads of Secret Service agents, who in those days didn’t start protecting a candidate until his election appeared certain.
The second clue was the conclusion of the count in Illinois, with Kennedy ahead by the infinitesimal margin of 8,800 votes out of more than 4.6 million cast. The controversies that would engulf the integrity of that count then and for decades afterward had yet to erupt; for the moment, 8,800 votes seemed an unlikely total to be overturned in a recount. He had also finished narrowly ahead in Pennsylvania, Missouri, Minnesota, and Michigan.
“We were satisfied,” said Donahue. “We knew it was locked.”
At the time, Kennedy’s victory appeared more solid than it actually was. The election-night count had him ahead in California by roughly 35,000 votes out of more than 6.5 million, and he was trailing in Hawaii by well under 1,000 votes. But the California count included none of the absentee ballots cast by what was then a traditionally Republican constituency. By mid-November, when they were all counted, Nixon had come back to win the state by roughly the same margin Kennedy had enjoyed on November 9. And the inevitable recount in Hawaii produced Kennedy’s smallest margin of victory in the entire election: 115 votes.
In addition the slates of “independent” electors that were on the ballot in three southern states had decidedly mixed results. They were walloped in Louisiana, where Kennedy won overwhelmingly. They very narrowly won in Mississippi, getting 8 electoral votes in the process. And they achieved a small plurality in Alabama, which, under an anomalous state law, produced 6 electoral votes for the segregationist slates and 5 for Kennedy. The 15th electoral vote was collected when one of the people on the Nixon slate in Oklahoma decided to join their cabal. Their standard-bearer was one of the Senate’s most powerful segregationists, Harry Byrd of Virginia. The southern fourteen chose South Carolina’s senator Strom Thurmond (the presidential nominee of the Dixiecrats in 1948) as their vice president; the renegade from Oklahoma chose Senator Barry Goldwater of Arizona.
There was one more anomaly in the 1960 election. For the first and only time there were 537 electoral votes. The reason involved the enfranchisement of Alaska and Hawaii after they became states in 1958. From the total of 531 in 1956, each state got two votes for its new senators; in addition, as prescribed in a transitional statute, each got one new member in the House of Representatives until the 1960 Census laid the foundation for congressional reapportionment, at which point the total became 535 (Washington, DC’s citizens got three more subsequently.)
The evenly split popular vote in 1960 was thus 34,220,984, or 49.72 percent, for Kennedy; 34,108,157, or 49.55 percent, for Nixon; and 610,409, or 0.42 percent, for the “independent” slates.
In the Electoral College the result was somewhat more clear-cut: 303, or 56.4 percent, for Kennedy; 219, or 40.8 percent, for Nixon; and 15, or 2.8 percent, for Harry Byrd.
Or was it?
Before the week was over, Nixon, at the start of a postdefeat vacation, announced that he would not contest the result by political means (essentially recounts) or through legal proceedings (lawsuits or investigations of alleged irregularities). However, at almost the same moment he was visited by the Republican national chairman, Thruston Morton, who not only urged Nixon to support a national campaign of challenges but also informed him that the party would undertake efforts in eleven states, beginning at once.
Over the decades that followed, much has been made of the Nixon statements, but only recently has more attention been paid to what the party’s leaders actually did with Nixon’s full awareness. In his first memoir, Six Crises, Nixon put on the cloak of statesmanship to describe his claimed refusal to use challenges to contest the election result. He said he decided not to act, even against the recommendations of senior Republicans, President Eisenhower included.
But Nixon’s narrative is false. When Morton outlined his plans, Nixon could have stopped him on the spot with a simple, forceful request. He didn’t, making him tacitly complicit in what followed. When his book was published two years later, he also neglected to note that after a very brief, fighting urge, Eisenhower had changed his mind and never supported the Republican actions that followed. Among the first people into the field to stir up challenges were such close and longtime Nixon friends as Robert Finch, Leonard Hall, and Peter Flanagan.
Any campaign of the sort the Republican Party put together in the days following a presidential election must proceed with a large clock ticking away. Typically state laws require some form of official certification of the Election Day numbers as a prelude to the actual casting of votes in the quadrennial meeting of the Electoral College; in 1960 that was to occur on December 19.
With Nixon’s acquiescence, Morton’s challenge was to come up with 50 electoral votes at Kennedy’s expense once California’s official result had flipped to Nixon and Hawaii’s had flipped to Kennedy. That meant at least two of the larger states on his original list of excruciatingly close results, or at least one of the larger states plus nearly all of the smaller ones: Illinois, Texas, Delaware, Michigan, Minnesota, Missouri, New Jersey, New Mexico, Nevada, Pennsylvania, and South Carolina.
Almost immediately three of the states—Delaware, Michigan, and Minnesota—were taken off the list because of reassessed possibilities, legal or recounting obstacles, or both. In fact recounts in general quickly proved unworkable for the Republican protagonists.
The best example was in New Jersey, where, when the counting stopped on Election Day, Kennedy was ahead for its 16 electoral votes by the underwhelming margin of barely 0.8 percent. The Republicans quickly got recounts going in five counties, but by the end of the month they had failed to turn up any major deviations from the initial count. The process quickly came to a halt, as happened in five of the other states. That meant as a practical matter that the Republican campaign would have to concentrate on Texas and Illinois (with 51 electoral votes between them) and would have to prove fraud to prevent state certification and affect the Electoral College result in both states.
The Texas effort was unsuccessful almost immediately. State law did not have statutory provision for recounts. That meant the only recourse was the federal courts if the Kennedy margin of roughly 46,000 votes, or 2 percentage points, was to be reversed.
The hurdle the Republicans could not clear was jurisdictional. In what amounted to a civil rights lawsuit, they could not persuade a federal judge that he had the authority to intervene in what was after all a state election to select twenty-four electors. They did not bother to appeal.
That did not mean, however, that there was no factual basis for an attempt to challenge the result on grounds of fraud; there was, though it was anecdotal. For example, in East Texas a precinct was found in Angelina County (Lufkin is the seat) that had 86 registered voters but recorded 147 for Kennedy. And in another county farther north (Fannin, Sam Rayburn’s home), there were 4,800 official voters who somehow cast 6,100 votes for Kennedy. These anecdotes, however, are just that. Not then, nor in subsequent decades, has any direct evidence of fraud surfaced that would come within a country mile of overturning the official result.
And that meant as a practical matter that the Republican effort could not have succeeded even if it had changed the result in the state about which experts and laymen have argued the most for more than fifty-five years. Tiny as it was, Kennedy’s 8,800-vote margin in Illinois was actually only the second smallest of the election in percentage terms. (The margin in Hawaii was barely measurable.) But it was and is by far the most controversial.
As in Texas, there is anecdotal evidence supporting suspicions of fraud. It was found in Chicago wards as far apart as the Sixth, on the largely African American South Side, and the 28th, encompassing some of the city’s white ethnic communities. In 1962 three people involved in the 28th ward (two precinct workers and a captain in Mayor Daley’s powerful Democratic organization) pled guilty and served brief jail terms for their roles. In addition a Nixon friend and national correspondent for the New York Herald Tribune, Earl Mazo, found scores of matches between names on tombstones in one cemetery and those on voter registration rolls. He also visited a vacant, boarded-up house in the city that nonetheless recorded 56 Kennedy votes.
The Republican effort in Chicago and surrounding Cook County was connected to the efforts of the county’s top prosecutor, Republican and vociferous Daley foe Benjamin Adamowski, to challenge the official result of his own reelection campaign, which left him trailing Democrat Daniel Ward by 26,000 votes. A recount began at the end of November and covered nearly a thousand precincts. Two weeks later the effort had produced less than one vote per precinct more for Nixon, and in 40 percent of precincts the Nixon vote turned out to have been overcounted.
With these meager results the Republicans had no choice but to head to court, where the result was the same as in Texas. Though investigations would continue well past Kennedy’s inauguration, the results were always the same. Through the ensuing years opinions, usually loudly argued, have been all over the place about Illinois in general and Chicago in particular.
Some analysts have simply assumed the very ground in contention. In addition to partisans on both sides, a more prudent approach has been to cite the very real anecdotal evidence with the awareness that, while no negative can ever be proved, what evidence there is remains far from conclusive.
After a month of accusations and defenses, as well as the accumulation of anecdotal evidence, Nixon finally appeared to have had enough. Not long before his death Mazo told the Washington Post that Nixon summoned him in mid-December, shortly before the Electoral College met and after the Illinois numbers had been certified by a Democrat-dominated, official board. Nixon asked him to stop reporting on the election in order to help avoid “a constitutional crisis.”
“I thought he was kidding but he was serious,” Mazo remembered. “I looked at him and thought he’s a goddam fool.”
Nixon then called Mazo’s editors at the staunchly Republican Herald Tribune, whereupon he was pulled from the story, which at that point included four articles of what was to be twelve. Exactly to whom at the paper Nixon spoke has never been made known.
But that was that. On December 19 the Electoral College met and Kennedy was awarded his 303 votes. He had just barely won the election, but he had won legitimately.
The question remains, however, how the result can responsibly be interpreted and explained.
It is axiomatic in a very close election that theoretically almost anything can be cited as an event or situation that is dispositive. For example, many liberals and older voters were attracted to Kennedy because he proposed universal hospital insurance for the elderly, financed through the Social Security system; they probably numbered more than Kennedy’s tiny margin in the popular vote. That, however, begs the question of how many voted for him only because of this proposal and how many voted for Nixon only because they disliked it. Similarly Kennedy’s proposal to increase the minimum wage to $1.25 an hour from $1.00 was popular, but there is no evidence it swung the election his way.
A more productive exercise could be conducted with the two nominees’ decisions on running mates to gauge the extent to which Johnson helped in the South (quite a bit in Texas and the South generally). By contrast, there is no evidence Johnson hurt the ticket even marginally in certain of the northern states; Lodge was better liked in general, but it is speculative to argue that a more liberal alternative would have delivered a single state Kennedy lost to make up for the opportunity Johnson helped provide in the South.
The list of specifics could be extended endlessly with no hope of producing a Eureka moment. It’s probably more prudent to focus on the bottom lines, the basic architecture of the results.
Kennedy’s election represented a shift of 8 percentage points from Eisenhower’s large total of 57.8 percent to Nixon’s losing share of 49.8 percent. In popular vote terms, Nixon and Kennedy had modestly fewer supporters than the personally popular Eisenhower did in 1956.
The partisan split is revealing. According to the final preelection Gallup polls that the firm later adjusted to reflect the election results, Democrats do not appear to have voted much differently in 1960 than they did in 1956. In Eisenhower’s landslide, Stevenson actually increased his vote among Democrats over 1952, winning by 85 to 15 percent. The numbers were not statistically different for Kennedy in 1960, though there were more Democrats in the mix than there had been four years earlier.
The real difference was among Independents, already a growing force in national politics, particularly in the country’s explosively expanding suburbs. According to Gallup, the margin among Independents in 1956 for the war hero president with a nonpartisan image was an overwhelming 70 to 30 percent as he won reelection. But four years later the more traditionally Republican Nixon, who had a highly partisan image, fell to a much more competitive 57 to 43 percent.
A few days after the election Meyer Feldman penned a memorandum to his fellow senior staff members attempting to make the case that Kennedy’s victory was meaningful, that a mandate could be discerned in the numbers. “On November 8, 1960, 22 states representing 103,750,000 people gave Senator Kennedy a majority,” Feldman wrote. “These states represent 58.1 percent of the population of the United States.”
It was a lame point, and no top Kennedy person, much less Kennedy himself, attempted to make it in public. The evidence suggests the victory was much more personal than political or ideological. Yet even the personal nature of Kennedy’s victory demands qualification. It is not even clear that he won the popular vote, given the unique peculiarities of the voting in Alabama, where the votes for individual electors are maddeningly difficult to characterize.
Two years earlier, in the off-year election that occurred at a time of economic downturn and in the sixth year of a presidency, when the “in” party normally suffers losses, the Democrats had achieved a substantial victory. Some correction might have been expected in 1960, but in modern times the much higher voter turnout for a presidential election has usually favored Democrats.
However, on the day he was elected, Americans also voted in two Republican senators and twenty-one Republican representatives; Republicans also increased their standing in state legislatures as well as state and local offices. That is hardly evidence of a clear mandate to “get his country moving again.”
Much more important is the simple, overarching fact that outside of his native Massachusetts more Americans voted for Nixon than Kennedy. It is hard to claim a mandate from facts like these, and he did not. Indeed as president Kennedy sometimes had a small piece of paper in his suit jacket pocket on which he wrote his infinitesimal popular vote margin—to remind himself and others pressing him for “bold” action. (His one such move after the election was confined to his Democratic congressional majority. In an effort to make it much harder for the Republican-Dixiecrat alliance to block legislation he successfully expanded the membership of the House Rules Committee, which regulates the flow of legislation to the floor for votes. Kennedy was well aware that among the 63 freshmen elected to the 435-member House in 1960, his committee-packing proposal was opposed 44 to 19.)
At the margin or tipping point of Kennedy’s election, two elements—race and religion—stand out as having had a significant impact on both the result and the margin. There was a spike in the African American turnout in northern, urban areas, overwhelmingly in Kennedy’s favor, that was a factor in his victories in New York, Pennsylvania, Michigan, and Illinois. And the issue that never seemed to go away, religious bigotry directed at a candidate who happened to be Catholic, loomed larger than any other, costing him millions of votes but also probably helping him achieve his clear Electoral College victory even as it reduced his popular vote total.
In 1960 what came to be called “exit polls” did not yet exist. However, similar work was done at the time that provides at least some relevant information. One was a series of polls conducted just before and just after national elections beginning in 1948 under the aegis of the University of Michigan’s Survey Research Center. Another is Gallup’s historical data. Still another began inside the Democratic Party and then the Kennedy campaign, sparked by three MIT political scientists—Ithiel de Sola Pool, Robert Abelson, and Samuel Popkin—with support from a New York businessman, Edward Greenfield, and evolved into an academic project after the election. It involved combining all of the campaign year’s polling data as they were collected and projecting trends forward from this massive amount of information with the aid of computers using simulation, then a new technique. In 1959 top party officials responded favorably to a proposal by the group, and preliminary data collection work began. The involvement with the Kennedy campaign, including at least two meetings with Robert Kennedy, took place between late spring and Election Day, intensifying as strategies for the general election were being put together before Labor Day.
The very first report by the group, “Negro Voters in Northern Cities” was issued in June; it strongly urged vigorous support of the civil rights movement, a position toward which Kennedy was already moving after years of temporizing and as it became apparent that despite his best efforts Johnson would get nearly unanimous southern support at the Democratic convention. “In a close election, even a moderate shift in Negro votes could be decisive in eight [key northern states],” it read in part. “Any shifts in the Negro vote could determine the outcome.”
Kennedy’s support for the civil rights movement increased even more at the convention and subsequently and reached its zenith with the two phone calls to Georgia in late October by his brother Robert and Kennedy himself that helped gain Martin Luther King’s release from prison. The campaign was decidedly low key in its handling of the incident in the national media but worked hard to get the word out through the African American press and in the distribution of at least one million “blue bombs” at churches the weekend before the election. According to the analysis by the MIT group, “That helped to push the Negro Democratic [vote] in 1960 from the two-thirds to which it had fallen to something like three-quarters of the total Negro vote.” This analysis was confirmed by the Gallup poll numbers. In 1956 the nonwhite vote broke 61 to 39 percent for Stevenson over Eisenhower; four years later the late-moving spread was 68 to 32 percent of a larger turnout.
In addition to race, the other major undercurrent in the election (arguably the most significant in terms of both the result and the margin) was religion. In the ensuing decades some have argued that Kennedy’s Catholicism hurt and some that it helped his campaign. Upon close examination, the more complex answer is that it both hurt and helped; it helped supply his Electoral College victory, and it prevented his margin in popular and electoral votes from being greater and more solid.
The Gallup data showed that between 1956 and 1960, the Republican margin in the Protestant vote stayed basically the same—a bit more than 60 percent to a bit less than 40 percent. But among the roughly one-fourth of the electorate that was Catholic—traditionally a Democratic constituency—there was a dramatic shift, from a basically even split in 1956 to a 78–22 percent Kennedy margin in 1960.
As the MIT analysts explained, “Millions of Protestants and other non-Catholics who would otherwise have voted Democratic could not bring themselves to vote for a Catholic. In total—so our model says—roughly one out of five Protestant Democrats or Protestant Independents who would otherwise have voted Democratic bolted because of the religious issue.”
On the flip side, the MIT group also calculated the votes Kennedy gained because he was Catholic. Using its enormous database and computers, the group concluded, “Taking Congressional voting as a base for estimating normal party vote, over one-third of Catholics who would otherwise have voted Republican seemed to switch to Kennedy. The best guess is around 40 percent.”
According to the Michigan survey, 10.8 percent of all voters, or about 7.3 million people, switched votes because of religion. Kennedy’s net loss was calculated at 1.5 million. However, those anti-Catholic votes were not evenly distributed. They were bunched more heavily in the southern part of the country, relatively safer territory for the Democrats back then; and they were less of a factor in the more closely contested west and north, where Catholics were more prominent.
The MIT team used data from every state to make judgments about the Electoral College impact. The states Kennedy would have lost had it not been for Catholic voters: Connecticut, New York, New Jersey, Pennsylvania, Illinois, and New Mexico. That is, 132 electoral votes. And the states Kennedy lost because of his religion: Kentucky, Tennessee, Florida, Virginia, Oklahoma, Montana, Idaho, Utah, California, Oregon, and Washington. That is, a net gain of 10 electoral votes.
It is silly to speculate about how the election would have turned out had Kennedy not been a Catholic; had he not been, he wouldn’t have been Kennedy. But it is not silly to take Kennedy’s religion as a given and recognize that his popular vote and electoral vote margins were held down by widespread religious bigotry that was almost instantly marginalized by his historic victory.
Five years of hard work—intellectual and political—had helped him win the Democratic nomination and then the presidency. He had begun seeking national office because he discovered a way forward politically; his ambition, as he saw it, had not been about some cause or issue. It was to be in Theodore Roosevelt’s arena, to be in a position to wrestle with the biggest issues and make a difference.
And politics and the issues aside, his “intellectual blood bank”—his aide Ted Sorensen—never stopped reminding friends after Kennedy was murdered why his victory mattered. “We’re still here,” he would say. For thirteen days in October 1962 Kennedy used the same attributes he displayed during the marathon of his campaign to figure out how to keep a nuclear confrontation with the Soviet Union from becoming a nuclear war.