The Pitfalls of Writing the First Draft of History


As an American political historian, I’m occasionally asked to offer thoughts about the current state of our politics or, more dangerously, to make predictions about its future. At times like those, I feel obligated to remind people that my professional training is in the field of hindsight.

I certainly appreciate the instinct behind such inquiries. Our understanding of the present is, of course, shaped by our understanding of the past. Indeed, I’d suspect that most scholars who work in modern American history were drawn to their topics largely by a desire to find answers to contemporary questions. I know that’s true for me.

That said, it’s an incredibly difficult task for scholars to think historically, in real time, about events unfolding before us.

If we are empiricists, then we have to ground our conclusions in a firm foundation of hard evidence. Obviously, when it comes to drawing conclusions about the recent past or the present, so much of the material to which we would normally turn just isn’t there.

Archival materials, the sources that serve as the backbone of most historical inquiry, typically take decades to come into view. For political history, congressmen and senators don’t turn over their papers for public consumption until the end of their careers, if not the end of their lives; many then place restrictions on the materials that delay their use for several decades more. Presidential libraries open with greater speed, but even then there’s a considerable lag and a number of restrictions. The George W. Bush Presidential Library and Museum, for instance, opened in April 2013, more than four years after Bush left office, but many records will remain off-limits to researchers for a full twelve years. We simply don’t have the key materials we would normally use.

There are other sources that can help us fill the void left by archival absences: more quickly available records of congressional hearings and court cases, oral histories and press interviews with key participants, election results, polling data, the deep dives of investigative journalists and studies by scholars in fields more attuned to the present, including sociology and political science. While some historians are skilled at using such tools of social science, for many it can prove to be unfamiliar territory. We need to tread carefully.

All these notes of caution are especially true when it comes to placing a recent presidential election in historical context. I’m writing this post ten days after the 2016 election, and while we now know Donald Trump will win, we don’t yet have much sense of the precise contours of his presidential victory. The popular vote has yet to be confirmed, and the electoral vote hasn’t been finalized either. Exit polling that we might be tempted to use to break down the electorate demographically is still imperfect and incomplete. Indeed, it won’t likely be until January before we can even start to speak with any confidence about how the 2016 electorate broke down according to lines of race, gender, class, age, etc.

As difficult as it is to determine what precisely happened in an election so soon after it, it’s even more difficult to draw any larger conclusions from it. Presidential elections naturally loom large in the national imagination, but we have to remember they’re simply one data point in a larger stream, and an imperfect one at that. The initial reactions of the press and general public about the meaning of any one election result are invariably overblown. Lyndon Johnson’s landslide win in 1964 didn’t mark the death of movement conservatism, just as Ronald Reagan’s victory in 1980 didn’t demonstrate the end of the New Deal order. The success of Barack Obama in 2008 didn’t herald the country’s sudden transformation into a “post-racial society” as many excitedly claimed at the time. In the same vein, Donald Trump’s election in 2016 likely doesn’t signal a complete return to racial reaction.

Even in the largest of landslide results, there are still significant crosscurrents pushing in the opposite direction. In closer elections, it’s even more important not to lose sight of the losing side. In this immediate instance, we’d be especially unwise to disregard the loser. Hillary Clinton has a significant edge in the overall popular vote; even within the Electoral College, a shift in three states of roughly 50,000 votes (out of more than 123 million total) would have resulted in a Clinton victory and, of course, an entirely different set of instant conclusions about what 2016 supposedly meant.

All these caveats aside, we can still draw out some initial conclusions from the election, but only if we resist the temptations to teleology.

For my own part, I’ve been busy over the past year co-writing a textbook with my colleague Julian Zelizer, chronicling the course of American history from 1974 to the present day. When the election began, we had a full draft in place, with just a few pages set aside to cover the campaign and bring the book to a close. We’ll now be expanding the section on the election considerably to cover a year’s worth of tumultuous events and highlighting some earlier themes that helped lead to it. We won’t, however, let one day, still dimly understood, overwrite our earlier interpretation of four decades, an interpretation based on a deep reading of primary sources and secondary literature.

Key themes of the book will remain, if perhaps in slightly altered forms. Increasingly pronounced levels of political polarization, for instance, served as an important through line in the book, and this bitter presidential campaign has provided ample evidence of that. But if our electoral maps are once again starkly divided between red and blue, a closer look shows some fluidity in the movement of key counties and constituencies from one party to another. Political analysts had confidently assumed, for instance, that the so-called “Reagan Democrats” of the industrial Midwest were a thing of the past, but these early results suggest they’re still alive and well.

While we’ll call attention to the persistence of white working-class voters, we’ll be sure to note that other constituencies were just as mobilized, though in the opposite direction. The Bernie Sanders movement, for its part, revealed a powerful strain of economic populism that at times overlapped the Trump campaign on issues of trade, but also demonstrated stark differences on matters of race and social justice. Comparisons and contrasts between these two camps will, I think, offer more fuel for those scholars who smartly encourage us not simply to rethink the red-and-blue maps but to move “beyond red and blue” altogether. Polarization persists, but in fluid and less fixed camps than we might have thought.

A second major through line of the book has been the steadily growing influence of new forms of media on American life across the last several decades. Sections we already had in place detailing the impact of innovations like CNN, Facebook, and Twitter now seem prescient in light of the role they played in the rise of Trump. Early on, CNN aired his full speeches without much comment and then placed several of his key aides on staff to spin events his way; Facebook promoted false news stories that seemingly swayed some voters to his campaign; and Twitter served as his main means of engaging voters, a form he mastered as much as FDR did with radio or JFK did with television.

While such examples deepen our understanding of the importance of these forms of media, they’ll merely sit alongside earlier examples that point in other directions. CNN promoted Trump’s arguments against free trade and free migration, but this came after decades of the network promoting globalization. Facebook worked to silo some Americans in their political views this year, but it’s also served to pull individuals out of their isolation. Twitter mobilized Trump’s message, but it just as importantly helped launch and lengthen the Black Lives Matter movement.

In such ways, the 2016 campaign has seen a continuation of such earlier trends and others. There will, of course, be new developments born in this moment but historians would be wise to wait before issuing any grand pronouncements.

If political history teaches us anything, it’s that a first draft that’s written hastily is often wrong.

Kevin M. Kruse is a professor of history at Princeton University. He specializes in the political, social, and urban/suburban history of twentieth-century America, with a particular interest in conflicts over race, rights, and religion and the making of modern conservatism. He is the author most recently of One Nation Under God: How Corporate America Invented Christian America (2015). He is also an OAH Distinguished Lecturer.


Share your thoughts