Skip to content

Sources, Style, and Subjectivity In History Writing

January 29, 2007

A Reflection On Erick Larson And The Health Of The History Profession


One of the great quests in becoming a good historian is learning to write well. Finding success in that endeavor involves a myriad of ingredients: considering one’s audience, using research material well, training, developing a style, learning to have a voice (a style corollary), rewriting, and negotiating with editors and/or professors. One of the mysteries of this quest is subjectivity: what pleases one authority or audience will not necessarily please another.

In school, one learns to adjust his or her own writing to what each professor likes. Factors that come into play include: more or less narrative, more or less analysis, more or less voices from your sources, more or less first-person interjection, and more or less literary flair. In a classroom one has at leat three or four opportunities to adjust, such that by the final assignment you have a fair understanding of what your instructor wants.

In the history publishing world, especially with regard to books, it seems that little to no adjustment period exists. With journals the revise and resubmit process resembles fairly closely what happens in the classroom. With books, however, it seems that publishers and editors make even bigger decisions on first-time submissions with even smaller samples and less interest in developing a working relationship. As best as I can see it, publishers accept proposals and writing samples, but discourage an author from sending a full manuscript without an invitation. This means that the potential success of one of the biggest quests of a junior scholar’s career – getting a book published – is based on fewer factors than a first-year student’s survey course ‘A.’

With the problems of book publishing in mind, as well as my desire to be the best history writer I can be, this Sunday I read with great interest a Chicago Tribune story by Patrick T. Reardon. Titled “The Details in The Devil,” the article meditated on Erik Larson’s The Devil In The White City (2003).

As a Chicagoan, one of my – and my wife’s – favorite activities is observing what people read on the train. Larson’s book has been a hit with CTA riders. But the book’s popularity seems to extend far beyond Chicago. Reardon reported that Devil in the White City has sold over one million copies.

But Reardon’s article was not some kind of fete for the book. He focused on Larson’s research and writing, and found that the book’s popularity as a work of history is built on dubious grounds. Here are some excerpts from Reardon’s piece:

– “Larson’s dual story of Daniel Burnham overseeing the World’s Columbian Exposition of 1893 and serial killer H.H. Holmes preying on young women in the Chicago neighborhood near the fair site has sold more than 1 million copies. Now, it’s being cited . . . by academic scholars such as Carl Smith, who lists The Devil in the White City in the bibliographical essay for his top-notch monograph The Plan of Chicago. . . . Librarians and other experts . . . are [also] recommending Larson’s book as a fine example of non-fiction writing.”
– “The problem is that key scenes in the book, as well as other descriptions, characterizations and details, aren’t based on historical sources or eyewitness accounts, but on Larson’s speculations, on what he seems to believe must have happened or could have happened. Yet he presents them as fact.”
– “This is important because the bedrock of good history books and other non-fiction works is truth. Any interesting fact in a non-fiction book . . . is all the more interesting because it’s true.”
– “Contacted for this story, Larson replied in an e-mail: ‘There are 857 footnotes in The Devil in the White City, citing material from 139 different books, articles and archives, and from hundreds of newspaper articles in 1890s editions of the Chicago Tribune, the Philadelphia Public Ledger, and The New York Times, all of which I consulted in the course of approximately three years of research. Yet these endnotes are by no means exhaustive. A truly comprehensive list, noting the source of every detail, would require a companion volume. Nonetheless, The Devil in the White City is, figuratively speaking, an open book, its logic and sourcing transparent to all.'”
– “Larson writes in the endnotes that at important points in his narrative, he ‘re-creates’ scenes, using the bare-bones facts available from a variety of accounts and weaving through them his speculations about what was said and done, how things looked and how people felt. As a result, the book reads like a novel.”
– “Unless readers[, however] are constantly interrupting the story to go to the notes in the back–and, I suspect, few do–they have no indication when Larson is recounting facts and when he’s speculating. This difficulty could have been avoided easily if Larson had written in the text, ‘The murder might have happened in this way …’ Or ‘This is how the killing may have taken place …’.”
– “The Devil in the White City belongs to a literary genre known as creative non-fiction, in which the writer uses a novelist’s techniques–scene-setting, rich descriptions, dialogue–to tell a vivid, engrossing story that’s true and accurate. One practitioner is Kurt Eichenwald, a New York Times investigative reporter and the author of three books on corporate skulduggery. . . . ‘When writing such books,’ Eichenwald noted in an essay posted at, ‘every fact–from the weather conditions, to the color of the wallpaper, to the types of meals eaten by the characters–has to come from somewhere.'”
– “That wasn’t the standard that Truman Capote followed when he produced . . . In Cold Blood. Although Capote’s book has been a model for Larson and other writers of creative non-fiction, [Capote] acknowledged the role his imagination played in the final text by terming the book ‘a non-fiction novel.'”
– “Since the mid-1990s, sales of non-fiction books have been steadily on the rise, particularly memoirs, such as Frank McCourt’s Angela’s Ashes. Readers have gobbled up true-life stories about winning thoroughbreds and landmark lawsuits, as well as personal accounts of addiction, oppression, struggle and survival, while showing much less interest in novels on the same topics. The reason: Truth sells.”
– “Of course, The Devil in the White City isn’t a memoir; it’s a work of history, complete with endnotes and a bibliography. That makes Larson’s inventions, to my mind, all the more egregious.”
– “Larson . . . deserve[s] some credit for explaining his methods in the endnotes, at least in relation to his re-creation of the murders and the visits to the fair. Even so, the notes in other cases prove inadequate. For example, he puts thoughts into Daniel Burnham’s head at various points in the book without providing any indication of what his source was. And he writes that at the fair’s opening ceremonies, socialite leader Bertha Palmer’s ‘diamonds radiated an almost palpable heat.’ As sources for this section, Larson lists six books and the Tribune for the next day, May 2, 1893. But none of the books mentions that Palmer was wearing diamonds. The copy of the May 2 Tribune, in this newspaper’s archives, includes a story giving a detailed 100-word description of Palmer’s gown and appearance, but it makes no mention of diamonds. In Larson’s defense, it’s possible he saw a different edition of the paper for May 2.”
– “That’s how historical errors are perpetuated. And that’s why the inventions that Larson engages in are dangerous. Once they’re embedded in the historical record, it’s hard to pry them out.”

The article gives many other examples of Larson’s subtle errors and exaggerations. Do read Reardon the piece.

With Reardon’s critique of Larson in mind, I can see why editors have a great deal of trouble in determining which works of history are going to be a success. I expect the problem is especially acute in deciding on junior scholars. And, as William Germano reminds the aspiring junior scholar in From Dissertation to Book, publishers do want to make money. They want to pick winners. But works like Larson’s unacknowledged “non-fiction novel” put an extraordinary amount of pressure on young scholars to be something cannot possibly be: novelists. In my experience there are few instances, due to the lack of resources, where an historian can work with Clifford Geertz’s admonition for “thick description.”

Without these descriptive resources, the tools at the historian’s disposal for writing an interesting story are limited. All the history writer can do is use lively prose and develop a strong “why” – an argument or thesis. And then, because opinions about style and lively prose differ widely among historians and publishers, all a junior scholar can reasonably do to develop some level of interest is develop a strong argument based on reliable resources.

So, until publishers of history – on the whole – reject works like Larson’s, then the work of professional historians dedicated to limiting their analysis to real resources will be stymied. Junior scholars in particular feel wedded to their primary and archival resources, and hence take few imaginative risks in writing their dissertations. Works like Larson’s harm not only the pursuit of truth, as Reardon argued, but also the development of the field of history all together. Larson’s success without in-text caveats pressures historians everywhere to put more and more distance between their writing and their sources. – TL


From → Uncategorized

  1. Anonymous permalink

    I’m trying to make a habit of letting print news writers know when I comment on their stories. With that, I sent this reflection to the Tribune's Patrick Reardon. Here was his reply:

    “Tim — Thanks for your note, and the link to your blog. You make a key point in the blog — that readers of Larson's book and others like it begin to expect history books to read like novels. When those colorful descriptions (based on imagination) aren't present in a real history book, they're likely to discard the book as boring. As you indicate, it's possible to write good, readable, captivating history using the facts. A lot comes down to the thought that the historian brings to the task. But, even the best historian is at a disadvantage in comparison to a novel-like psuedo-history. The danger is that a book like Larson's will make it harder for well-written history books to get published and find a readership. Good luck. Pat”

    I thank Mr. Reardon for the note. Again, please check out his full story at the Tribune for more on Larson's imaginary additions to history. – TL


  2. TL–

    This particular issue has been vexing me all day as I write. I feel compelled to stick to what my sources say, and yet there are times when a flourish here and an addition there make a much more readable manuscript. For example, I have a situation where two men went to view a piece of property that one eventually bought to house an immense new factory several years later. This viewing took place on a Sunday in November. that's about all my sources say. It seems likely that they were cold, at the very least, but I don't know that from the sources. It's also likely that the meeting was a source of anxiety for one of the parties, as this sale seems to be a crowning moment to his career as a land salesman to that point. But I don't know that. I'd love to be able to write about the hot cocoa that they enjoyed, and how one man offered some Irish Cream to the other as they braved bitter cold winds and imagined what the property would look like several years hence. But I just can't bring myself to do it, not as a serious historian.


  3. Tiff permalink

    This explains, in part, why, after publishing 3 nonfiction books (one a scholarly monograph and two reference works) I have begun work on a novel. Historical fiction.


  4. CM & Tiff: Thanks for the comments. I'll be struggling with these issues in the next month or so as I make a serious decision on how to rewrite my dissertation as a book manuscript. As hard as it is to complete a dissertation, I can totally see why even more folks are sort of weeded out at the next step. It's not that I pine for the days of merely average, barely readable first monographs (even some from the 1960s fit in this category), but publishers need to give writers room for error, room to grow. – TL


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: