Politifact and Historical Thinking: Disenfranchisement, Then and Now

By James Grossman and Allen Mikaelian

Is it appropriate to reference “Jim Crow” when discussing the rash of new laws and measures further regulating who can vote and under what conditions? Do the rhetorical analogies fit the historical facts? Politifact, the Pulitzer Prize-winning watchdog best known for mercilessly wielding its trademarked Truth-o-Meter™, is somewhere between skeptical and incredulous. Last year, the editors of Politifact counted one Democrat’s statement evoking Jim Crow as a finalist for “lie of the year.” Politifact Georgia took to task a mayoral candidate for making an analogous comparison between a crackdown on undocumented workers and the infamous “black codes” (enacted a generation earlier than the Jim Crow era and quickly erased by Reconstruction). More recently, Politifact Florida called a Democratic U.S. Representative’s reference to poll taxes “half true.”

Each time, Politifact editors called on historians to help them judge. Each time, their analysis and resulting judgments raised important questions about how historians, journalists, and politicians evaluate the nature of truth and how the past can best be mined for constructive analogy. Amid this useful conversation between heated political rhetoric and cool journalistic analysis, historical thinking finds its place at the table as uncomfortable as it is essential.

The Politifact editors do their research. On this issue, they’ve consulted prominent historians, legal scholars, and political scientists. Our colleagues on the list include such historians as Eric Foner, Robert Korstad, Glenda Gilmore, Jane Dailey, Alex Keyssar, William H. Chafe, Thomas Adams Upchurch, Leslie V. Tischauser, James C. Cobb, Randall Miller, Paul A. Cimbala, David Colburn, Morgan Kousser, and Michael Fitzgerald. The Politifact editors clearly have made an effort to get the best thinking on the subject.

The range of perspectives in Politifact’s analyses has been striking, especially across disciplines: the legal scholars and the historians often see this issue differently. Take, for example, the statement by Democratic National Committee Chair and U.S. Rep. Debbie Wasserman Schultz (D-Fla.) that the intent of voter ID legislation was “to literally drag us all the way back to Jim Crow laws” (a statement she quickly walked back). Asked about the comparison, historian Glenda Gilmore (Yale Univ.) pointed out, “Historians of the South use the term ‘Jim Crow laws’ to mean discouraging voting in a way that impacts minority voters. So, I think that she was exactly right in her use of the phrase.” Jane Dailey (Univ. of Chicago) similarly took a broad view, noting that “Groups interested in limiting political competition have always justified suffrage restrictions as necessary checks on election fraud. Like today’s GOP, turn-of-the-century southern Democrats argued that poll taxes and literacy tests would reduce fraud at the polls.”

The legal scholars took a different tack. Michael Klarman (Harvard Univ.) offered a results-oriented perspective: “Black voter participation, which in some places had been 90 percent of eligible male voters in the late 1860s and early 1870s, was cut to essentially zero by early decades of 20th century. What Republicans are doing now … is trivial by comparison. They are making it marginally harder for poor people, disproportionately minority, to vote.” Other legal scholars were inclined to parse the difference between intent to harm the election chances of one party (as has been argued about the current legislation), and the intent to disenfranchise African Americans for the sake of it (Jim Crow era). A few historians sided with Klarman, pointing to the importance of the larger context of Jim Crow racial policies. Others might argue that, like current efforts, late 19th-century disenfranchisement was indeed motivated primarily by partisanship; had southern Democrats been able to reliably control black votes, disenfranchisement would not have been necessary. The suitability of the comparison moves easily back and forth as the subtleties accumulate.

This does not mean that a historian could not offer concise judgment. We offer one possibility: Contemporary measures that manipulate exaggerated references to voter fraud in order to limit voting draw on traditions of blatant disenfranchisement that took place in the Jim Crow South, and hence merit attention and concern. But by equating the current legislation with the massive disenfranchisement of the turn of the 20th century, politicians leave little room for thinking or talking about far more egregious attacks on voting rights.

Politifact took a different approach, and rated Wasserman Schultz’s statement “False.” “Although there are some similarities in the two sets of laws, they are outweighed both by the differences between them and by the inflammatory nature of the phrase, which all but calls the laws’ supporters racists.” It is probably more accurate to observe that Wasserman Schultz has committed the misdemeanor of exaggeration rather than the more egregious offense of inaccuracy. But one might follow that with an emphasis on why the comparison matters, on why Gilmore considers the analogy meaningful. This is not “literally” Jim Crow, perhaps even not figuratively Jim Crow. But it is like Jim Crow. Perhaps “akin” would be a good term here: the legislation (as Dailey insightfully suggests) uses the same logic as late 19th-century disenfranchisement, mobilizing unproven allegations of voter fraud to blatantly disenfranchise particular voters, even if not at the same magnitude as the earlier period. Current efforts are not the resurrection of Jim Crow; they are the descendants of Jim Crow. The DNA can be traced, its family traditions apparent.

More recently, U.S. Rep. Alcee Hastings (D-Fla.) called Florida’s requirement that some residents prove citizenship (and pay for their own postage) a “backdoor poll tax.” Politifact editors went further into the matter than ever before, gathering comments from 14 scholars and even consulting parts of a book and an article.

They garnered and relayed a great deal of history about the poll tax in the United States. Alex Keyssar (Harvard Univ.) took the broad comparative view: “Any effort to introduce an election procedure that requires some voters to incur financial costs could be thought of as a metaphoric or perhaps real poll tax.” T. Adams Upchurch (East Georgia Univ.) disagreed, and the Politifact editors ended up largely convinced by his emphasis on a significant difference between the new Florida requirements and the 19th-century poll tax—affordability. The 19th-century poor who were faced with prohibitively high poll taxes, he noted, truly could not afford to vote, and “could absolutely and rightly claim discrimination … . But a stamp? Please.”

In the end, Politifact rated Hastings’ statement as “half true.” This is probably just about right, even if the rating fails to capture all of the issues involved. Keyssar argues that as a matter of principle, this is indeed a poll tax, and the actual amount can be considered a slippery slope. But a stamp is, after all, only a stamp. The question could also be addressed with intent: once again, was the intent to make it harder to vote? Would some people predictably find it harder than others? Sometimes historians are better at identifying the right question than at providing a short answer fit for the “true/false” section of a test.

And perhaps this is the issue. Politifact admirably works to educate the public on the accuracy of politicians’ references to the past. Sometimes this is a straightforward task; often it is not. Politifact generally seeks to confirm or disprove one-for-one correspondences between the present and the past. The historians cited by Politifact appear more willing to allow for comprehensive thinking; recognize that categories like “Jim Crow” aren’t cut-and-dried; and accept the idea that intent matters. Historians, less attached to the tyranny of the Truth-o-Meter™, are more willing to engage questions by explaining issues of continuity and change, and greatly enlarging the context. Though Politifact has made a concerted effort to include historians in its analysis, the Truth-o-Meter™ might not be readily calibrated to measure their responses.

Share on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Digg thisShare on StumbleUponEmail this to someonePrint this pagePin on PinterestShare on RedditShare on Tumblr

Back to Top

Leave a Reply

Comment

* Required field

  1. Kathryn Kemp

    Why would anyone who systematically targets African Americans and persons with Hispanic surnames for voter suppression be called a “racist.” Goodness me.

    Reply