The 1948 presidential campaign is endlessly fascinating, and it’s addressed at chapter length in Lost in a Gallup, my forthcoming book about polling failure in U.S. presidential elections. Lately, I’ve been rereading David McCullough’s account of President Harry S. Truman’s come-from-behind victory in ’48.
That chapter is a highlight of McCullough’s hefty, Pulitzer Prize-winning biography, Truman. In engaging detail, McCullough described how Truman — behind in the polls and a sure bet to lose — ran a focused, spirited, and strenuous campaign that kept him on the attack against an exceedingly confident foe, Republican Thomas E. Dewey, the governor of New York.
In broad terms, aspects of Truman’s 1948 campaign are relevant to the unfolding campaign narrative this year, notably in the public opinion polls suggesting a bleak electoral landscape for President Donald Trump.
In addition to polls that favor the challenger, other elements of the unfolding 2020 campaign are redolent of Truman-Dewey race, notably that the challenger appears reluctant to run many risks.
Dewey was reserved, not very likable, and ran a platitudinous campaign. “When you’re leading,” McCullough quoted him as saying, “don’t talk.”
Aspects of the unfolding 2020 campaign are redolent of Truman-Dewey race in 1948, notably that the challenger appears reluctant to run many risks.
Dewey’s caution brings to mind the below-the-fray, basement-based strategy that presumptive Democratic nominee Joe Biden has pursued for weeks, due partly to the coronavirus contagion. Dewey in 1948 did get out and travel the country by train.
But it was barnstorming in a genteel way, not the grueling, whistle-stop tour that kept Truman away from the White House for extended periods in the summer and fall 1948.
Another broad parallel lies in the approaches taken by the underdogs. As McCullough wrote, Truman “had just one strategy — attack, attack, attack, carry the fight to the enemy’s camp. He hammered the Republicans relentlessly ….”
“Joe Biden is not the leader of his party,” Trump said at a rally last week in Tulsa that drew a smaller-than-expected turnout. “Joe Biden is a helpless puppet of the radical left, and he’s not radical left. I don’t think he knows what he is anymore, but he was never radical left, but he’s controlled by the radical left, and now he’s really controlled.”
Truman took aim in 1948 at what he called the do-nothing, Republican-controlled 80th Congress. Trump has echoed the do-noting critique: “Joe Biden’s record can be summed up as four decades of betrayal, calamity, and failure. He never did anything.”
No two election campaigns are quite the same and there are of course significant differences between 1948 and 2020. Truman wasn’t inclined to belittle Dewey, whereas Trump relishes making fun of his foes. Truman “never criticized or ridiculed Dewey in a personal way,” McCullough wrote. Trump has a variety of unflattering nicknames for Biden, including “sleepy Joe” and even “sleepy creepy Joe.”
Truman, an obscure figure who became president on President Franklin Roosevelt’s death in 1944, faced political obstacles steeper than those confronting Trump this year. By mid-summer 1948, the Democratic party had split three ways — the breakaway Progressive party nominated former Vice President Henry Wallace and the Dixiecrat segregationists chose Strom Thurmond as their standard-bearer. Truman was the party’s mainstream candidate, heir to Roosevelt’s New Deal policies.
Republicans had won both houses of Congress in 1946 and, according to McCullough, no party ever lost the presidency after having taken control of Congress two years before.
And in June 1948, Truman faced the prospect of war over the Soviet Union’s blockade of Berlin, which the West broke in 1949 after a 11-month airlift of food and supplies to the city’s western sectors.
A prominent line of thinking in 1948 was that campaigning did little to change voters’ minds. By Labor Day, this thinking went, the race was essentially decided and all else was pageantry.
An advocate of the theory was Elmo Roper, one of the best-known pollsters of the day. Roper, who began polling in 1936, was so sure Dewey would win that he said in early September 1948 that he would release no further poll results.
On his 15-minute spot on CBS radio on September 12, 1948, Roper declared:
“So right here and now, I’d like to finish off my share in this whole business by predicting the election of Thomas E. Dewey by a heavy margin. I’d like to say that’s that, and turn my attention to other and more valuable considerations.”
In 1948, as they do nowadays, pre-election polls helped to fix the news media’s narrative about the campaign. The polls by Roper and George Gallup and Archibald Crossley — three founding figures of modern survey research — all signaled Truman’s certain defeat.
Truman didn’t much trust polls, and neither does Trump.
Late in the 1948 campaign, Truman declared, “You can throw the Galluping polls right into the ash can.” He dismissed them as “sleeping polls,” intended to lull Americans into believing the election had been decided weeks before they voted.
Not surprisingly, Trump mostly dislikes polls, especially those showing him losing. Trump assailed a recent Fox News poll indicating he trailed Biden by 12 percentage points as “another of their phony polls, done by the same group of haters that got it even more wrong in 2016.
“Watch what happens in November.”
So what do the broad themes and parallels in campaigns 72 years apart tell us?
They do not necessarily tell us that this year’s election will be 1948 redux. Times are different. The country is different. Campaigns and their dynamics are different. Whistlestop rail tours are of the past. And no two elections are quite alike.
But the parallels do point to the hazards of running a passive, take-few-risks campaign. Remaining above the fray probably cost Dewey the election. Keeping to the basement may not be a winning strategy for Biden.
The 1948-2020 parallels also signal that political fortunes are not static, that campaigns matter, and that focused, aggressive, imaginative, and tireless campaigning can impress and motivate voters, while helping to define the opposition.
More important, the 1948-2020 parallels offer cautionary reminders about a too-excited embrace of election polls, especially polls conducted months before anyone has voted.
Summertime polls seldom have significant predictive value. Besides, presidential races have sometimes turned on late-in-the-campaign developments — as may have happened in 2016, when the FBI announced in late October it had reopened its investigation into Hillary Clinton’s private email server.
Finally, the advice offered years ago by Warren Mitofsky, one of opinion polling’s most notable innovators, is pertinent here. “There’s a lot of room for humility in polling,” Mitofsky said in 1998, at the 50th anniversary of the Truman-Dewey race. “Everytime you get cocky, you lose.”