Best Guess or Worst Doubt: What's the Role of Conjecture in Writing Biography and History?

Lois Leveen's picture

In this post for the H-CivWar Author's Blog, Lois Leveen seeks input on how to balance a desire for accuracy in our scholarship with the inevitable unknowability of the much of the past.


At a plenary session of the 2021 Society of Civil War Historians conference, Poet Laureate and acclaimed memoirist Natasha Trethewey asked Nina Silber and David Blight to reflect on the pleasures of doubt in their (and all of our) scholarly work. Trethewey's query resounded like a cross between a Final Jeopardy question and an especially challenging round of Stump the Band.


Nina Silber replied that doubt "opens up a space where we historians have to do things like use our imagination, which we are maybe not always good at." But, she added, doubt is always part of the historian's process, raising "all the kind of 'what if' questions that we have to work our way through." In his response, David Blight equated doubt with a fear that is inevitable when writing history: "fear of what we don't find and doubting ourselves, maybe there's doubt on every page in a great work of history" – a proclamation about doubt he then immediately began to doubt, adding, "Can I say that?  Well, maybe not, you know, or do I really know?" Then, to bring the point home, he concluded, "Especially in biography, I found that you know there are some things you really know and then there are some things you don't quite know."


(Confession:  all the quotations in the above paragraph are from the notes I hastily typed during the virtual conference session, which might make any of us doubt their absolute accuracy – indubitable proof of the larger point about doubt, accuracy, and attempts to piece together past events.)


Back in 2013, Liz Varon and I were invited to speak at the former White House of the Confederacy, in a program in which we reflected on the use of fact and conjecture by biographers and by historical novelists. The work of historians and biographers, Liz explained, "demands that we walk a fine line between scientific detachment and humanistic engagement.  We are storytellers, and to make our stories compelling we have to sometimes pull in close to our subjects, to walk in their shoes, to use our powers of empathy and our skill as writers.  Our interpretations and stories are rooted in documentary evidence but they also require flights of imagination." Liz was drawing most specifically on her experience writing the biography of the white Civil War spy Elizabeth Van Lew, but also reflecting more broadly about the field of history. Particularly for our nonacademic audience, she was making an important distinction about how to write about the past: "We historians are permitted to speculate, and required to mark those speculations for our readers."


In that 2013 program, I was speaking as a historical novelist, describing the challenging process of balancing deep historical research with the need to craft a compelling story for fiction readers. Although I had already authored a number of historical articles, I did not anticipate then that I would now be crossing over to join Liz in writing a biography, one about the real woman whose life had inspired my first novel:  one of the many free and enslaved Black people who participated in the same Civil War espionage ring as Van Lew. Given my training in academic research and writing, teaching myself to write novels had been a challenge. But after years of getting to freely use my imagination to create characters, dialogue, incidents, and plots (albeit historically plausible ones), I am once again dwelling in that challenging place all the historians I've quoted above describe.


Thus my questions for readers of this blog: 

  • How much conjecture does one allow one's self? 
  • When do you make an intuitive, speculative leap based on limited evidence?
  • Conversely, when do you decide to disregard what seems dubious in the historical record?

These questions feel especially pressing when writing the biography of an African American woman, given limitations in the historical record regarding what information about lives like hers was gathered and preserved. Piecing together this particular woman's biography is further complicated by her own lifelong predilection for avoiding detection – she was, after all, a Black undercover agent in the Confederate capital – and for artfully altering elements of her life story as she recounted it, a rhetorical and political strategy that my project demonstrates was fundamental to her postbellum activism.


Here are a few specific examples of how such questions are playing out across this project:


Biographies tend to begin with the subject's birth, but like many people born into slavery, no evidence seems to survive of when, where, or to what parents this woman was born. I have located five primary sources, created across twenty years, that each indicate her age. The various ages they provide do not add up to a consistent birth year: she is 8 in 1850; 14 in 1855; 18 in 1860 but also 24 in 1860; and 30 in 1870. I've had all these primary sources at hand for several years, but only recently did I realize that if I discount the extreme outlier (24 in 1860) and look at the exact month and day for all the other documents, it's likely she was born in November or December of 1841. But with what certainty can I assert that? As confident as I might feel that one 1860 source is inaccurate, I'm considerably less confident that the rest are perfectly accurate.


Evaluating contradictory sources presents one kind of challenge; even more daunting are the many instances when there seem to be no sources at all. All the direct sources created by her that I have thus far located consist of about a dozen letters she wrote after the war.  One was written to Van Lew (her former enslaver as well as her fellow spy) and the rest to various white male government officials. None of her correspondence with other Blacks from any point in her life has survived. How do I reconstruct the friendships and alliances that must have been important to her, in the absence of any direct evidence?


For example, based on where she traveled and what she did immediately after the war, I have some hunches about African American religious and community leaders across the eastern U.S. with whom she likely interacted. Despite how significant these leaders were in their day, none of their papers have been preserved. This raises a host of what Nina Silber referred to as "'what if' questions":  What if her trip North in the immediate aftermath of the war included, or perhaps was entirely prompted by, a desire to visit an antebellum mentor? What if, while on that trip, she met and perhaps even stayed with so-and-so? Was it a chance encounter with such-and-such while in Virginia that led to her postbellum speaking engagement in Brooklyn? Was it at yet another postbellum Northern speaking engagement that she met I-think-I-know-who, which somehow led to her brief residence in Florida? In all of these instances, I feel I have good reason to surmise about how things might have happened, but not one iota of evidence. Despite Silber's reflection, what I lack is not imagination; it's any possibility of ever finding proof.


While these are only a few specific examples from my project, I suspect many of you have grappled with similar challenges in your own work. So returning to the questions above, I invite you to share how you grapple with conjecture, intuitive leaps, dubious sources, or the elusiveness of proof, in your research.

In 1999, Ron Chernow delivered a public lecture on biography-writing for the New York Council for the Humanities (where I worked at the time), which is archived at He raised some of the same questions that Lois Leveen poses here.

He began with the challenge of incomplete records, contradictory accounts, and the unknowability of historical actors' states of mind: "[If the dead subjects of biographies could awake,] perhaps ... they would bitterly refute many of the assertions confidently made about them. They might reveal errors, supply missing circumstances, disclose the secret biases of witnesses. They might expose the entire exercise as pure balderdash."

He ended with a call for humility: "If due humility is one prerequisite for a good biographer, I think that respect for the subject is another. ... Fortified with the secrets of people’s lives, fingering through their private letters and diaries, you can easily be tricked into believing that you’re as grand as your subjects and can treat them with flippancy or familiarity. It’s not good for a biographer to coddle people, but neither should he manhandle them. In short, if you have decided to make your living by waking the dead, at least ease them gently out of their graves and treat them with the dignity they deserve."

-- Phil Katz (Council of Independent Colleges)

As a trained historian and archivist, my feeling on this topic was changed in graduate school reading biographies on Progressive Era U.S. Presidents written in the early-mid 20th century, in the absence of access to their personal papers or archival collections. Having worked with collections of original archival materials connected with those presidencies, I often found the narrative prose and summations of the writers contradicted with the original records found in archives long after the books' being written. What bothered me was the declarative statements that those historians knew the answers or the motivations behind choices and decisions. As an archivist, I live in the incomplete. We always know there are records or perspectives missing, which makes describing the materials challenging.

I am the Military Collection Archivist for the State Archives of North Carolina. I deal with writing biographies (to date) for finding aids and oral histories for over 1,700 veterans since I began. My rule of thumb is to leave narrative out of the piece where it is not needed, or at least to not allow narrative styling to dominate the information. I teach my archives interns to be as accurate as possible without being wrong. It may mean you're not 100% precise on a job title, but you have the general job field correct. For military veterans where we have few records or no correspondence, we are left missing a lot of service details that can't be found anywhere apart from unit records deep in the National Archives, and which fall outside of my ability to access or need to for the work level I perform. But, I work with so many families who review biographies and correct information, yet that information is not widely different from my own or wrong. It just doesn't have all of the story. I use language like "believed to be" or "may be connected with" to alert readers that the information isn't precise. I realize this is different from biographical writing for book publication, but I feel alerting readers to uncertainty in my presentation of the information in the text (not in the footnotes) is vital to giving readers the ability to seek out the answers on their own if interested.

A Civil War book I wrote while an undergraduate in a very undocumented area of Kentucky discussed that there were no known military battles or skirmishes in the county. In 2007, that was the case, but I stated more research needed to be conducted as more records were being digitized in order to locate that type of information. In 2014, my old history professor found an obscure newspaper article in an Ohio newspaper that described a skirmish of Confederate and Union local raiding parties in the town where my college was, but no one was killed there. Another obscure reference was found in 2017 by a former history student I went to college with, thanks to records being put online and searchable. The local county history written in the 1990s discussed its Civil War information as being 100% correct, but didn't have either of these incidents in it. So, by 2014, that book was irrelevant and unreliable. Mine could still be used with the knowledge of the acknowledgement. I feel that when people have a certainty about history, they get lax in their research or statements for the sake of effect and reputation. I was publishing my book for information, not a publication deal. I think that really matters, and in my experience as a public historian the public appreciates that approach as well.

Dear All, 

I very much appreciate Lois's post and will suggest that it's relevant to more or less any exercise in historical interpretation.  Let me add a few disconnected thoughts that are hopefully of some service. 

For what it's worth, I often urge my undergraduate students to think of historical interpreation as resting on a disciplined form of creativity that cycles back to the sources to check for corrobaration and contradiction.  I worry, like many of my peers, that they have been conditioned to believe that "knowing" the past means memorizing "facts," when of course which pieces of information matter and what they really mean hinges on acts of interpretation. That said, I myself don't really know what expectations my students bring to their history classes, and my healthy doubt on that topic might serve as a pivot to ask Lois what she thinks her readers will expect of a biography. My impression is that professional historians, for all their familiarity with post-modernism and related intellectual currents, have a cultivated hostility to acknowledging the existence of uncertainty. I might ask my colleagues here whether they think it is fair to say that in our own narratives we tend to try to paper together coherent and linear stories instead of helping our readers stare into the abyss of the unknowability of the past. Of course, Lois has a paricularly rich range of writing and research experiences, so she likely has her own well-developed perspective on these issues. But could it be that a conventional emphasis on rendering those coherent and linear stories is really about the author and not the reader? How can we be certain that readers wouldn't prefer--or choose some other metric--works that explore just how little we can know? There are always at least a few within professional history devoted to explicitly creative writing from the archives, but this strikes me as forever a niche field. 

Back when I was working on what became the second chapter of my Between Freedom and Progress, which focused on the once-famous but now obscure explorer of equatorial West Africa, Paul Du Chaillu, I found myself asking some of the same kinds of questions that Lois raised (his birthdate was also hard to pin down, but for different reasons).  He lived a prominent public life as a lecturer and had a robust persona and stage presence (making the case slightly different from Lois's), but revealed little about his inner life. Ultimately, I did find compelling one contemporary account of him as smooth, ambitious, and focused on worldly success, but always wondered what this globe-trotter really thought and felt. In the case of that chapter, I was more interested in Du Chaillu's imprint on the racial politics of the Civil War and Reconstruction--in public reception and debate--, so didn't dwell on it too much in the chapter, perhaps to its detriment. While I personally only love history for the historiography and intertextual lacunae, I can't help but wonder how many people who read history in their free time and take history courses do it for essentially  human, social reasons--they want to know people, living and dead, and are drawn to them.  

Finally, the paraphrase from Nina Silber (about historians perhaps not always being good with their imaginations) reminded me of the following quotation from Peter Novick's That Noble Dream, which should probably be padded with several caveats but which I'll instead just leave here by way of being cheeky: 

In most respects history graduate students seemed to remeble those in other disciplines, but in two paritculars they displayed notable characteristics. In a survey of the college class of 1961, those going on to graduate study in history were almost at the bottom of a list of several dozen disciplines in the extent to which they saw in their academic work an "opportunity to be original and creative." Only those going into geology, microbiology, and economics listed this ambition less often. And when asked to rate themselves on a scale running from "very conventional" to "very unconventional," prospective graduate students in history were almost the most "conventional," surpassed only be chemists, biochemists, and microbiologists. (378)


Dave Prior


To all,
Two quick thoughts.

1) C. Vann Woodward told some of his graduate students to read Robert Penn Warren to learn how to use words to depict characters who would catch readers' interest.

2) In THE POETICS Aristotle linked the two rhetorical figures "energia" and "enargia." "Energia," the root meaning of energy, is defined as a kind of movement that sets things "before the eyes." "Enargia" is vividness of presentation. (It was especially needed in courts where witnesses and litigants needed to reproduce events through language with the vividness of ocular proof.) "Enargia," Aristotle argued, was needed to produce "energia" in an audience by setting things "before the eyes, for the hearer should see the action as present."

Thanks to all for responding, as I navigated my '99 Saturn across the U.S. to start my research fellowship at Library of Virginia. A few replies to specific comments in this thread -- and then one more plea/demand for input:

1. Phil Katz quoted Ron Chernow on a biographer's need of "respect for the subject." Scholars of African American history (and Black Diasporic history more broadly) have explored this topic in particular when we feel compelled to "expose" details of our subjects' lives that they strove to keep unexposed. As Greg Childs thoughtfully phrased it: "When as a historian, in other words, is it okay to let fugitives remain at large, to not follow the dictum of the discipline and rigorously comb the archive for more evidence and traces of a subject who was clearly trying to avoid being captured and documented and disciplined?" (quoted from )

2. Matthew Peek and Dave Prior raise questions of what role narrative can/should/ought not play in our work. This is always tricky territory, and especially so with my subject. Her story has been "mistold" very often, by journalists and historians. One example: There is a very awful book by a woman who credits herself with "inventing" the field of "sizzle history." Published by a very major US publisher and selling zillions of copies, this book purports to be a nonfiction account of four white women spies in the Civil War. It is full of invented scenes, dialogue, and plot twists Why did the publisher allow this? Because it sells (the first word of the title is "Liar" so maybe someone at the publisher has a sense of irony???). One of the white spies is Bet Van Lew, and some incredible new untruths about "Mary Bowser" are thrown in to add a little color, as it were, to the book. One claim is that throughout the war, Mary Bowser sewed secret messages into Varina Davis's dresses and then smuggled them to a seamstress to who undid the messages and passed them to Van Lew. No one with the least understanding of the lives of the enslaved, or even the availability of needles and thread in the Confederacy, let alone the operations of the Richmond underground, could believe it (the author must have known this was a lie she was inventing, because Van Lew's journal, which she cites, makes it clear this never happened), but now it's printed as fact. Another invention in this "historical" account is even more disturbing: it involves Bet Van Lew "saving" Mary Bowser from capture by loading her into a cart, covering her with manure, and sending her to Philadelphia. How demeaning a twist on the white savior myth is that? African Americans who challenge white supremacy must expect to be literally covered in crap, not by white supremacists but by their closest allies. These untruths are now taken as historical fact, cited by everyone from the *Washington Post* on (not in a book review, mind you, but in a stand alone article about Mary Bowser, in which Liz Varon and I are "quoted" although the "reporter" never contacted either one of us).

The popularity of accounts like this underscores growing public interest in women's history and African American history. They also underscore how little historical understanding people have, if they are swallowing tales like these. So more than just playing whack-a-mole with attempts to disprove false claims (although I spend a lot of time doing that, though it seems often to little avail), my goal with my project is to engage that public interest in women's history/African American history to teach a more accurate and complex understanding of the American past. Notably, my working title for this project includes the phrase "a biography of race in nineteenth-century America," because my interest is not in creating a new lone-hero tale but in using this woman's life to explain larger issues/movements, from Liberian colonization to postbellum black women's activism. Certainly, there are examples of biographies that disprove the mythologizing of their subjects (e.g. Nell Painter's *Sojourner Truth: A Life, A Symbol*) or that piece together snippets of evidence not just to recuperate the experiences of a single person but to demonstrate larger forces of which that person's experiences are representative (e.g. Caleb McDaniel's *Sweet Taste of Liberty*). One of the challenges of my project -- likely the topic for one or more future posts in this series -- will be figuring out how much to include of the fascinating things I'm learning that are not directly about my subject but about those with whom she interacted -- or those whom she never met but whose actions somehow touch on her story. All toward the goal of "narrativizing" the past in a compelling way for readers, but not of fudging the truth to make the story work.

But now for that plea/demand for input: I can't help but notice that most of the responses in this thread have been fairly general, referring to what so-and-so says about the nature of writing biography, or overarching advice one gives to one's students, etc. Matthew does readily admit having written a work that now, because of more recently uncovered evidence, he categorizes as unreliable and even irrelevant. A cautionary tale regarding certainty indeed (and this is something I will raise openly in my book, by alerting readers not only to what I've gotten wrong in previous articles about this woman, but also underscoring the distinct likelihood that new evidence will challenge my interpretations of what evidence I have assembled). But none of the thoughtful responses to the original post have actually grappled with what I was hoping they would:

how do you decide what evidence is reliable, and what is not?
when and how do you allow yourself to make an intuitive leap?
what is a comfortable relationship between interpretation and speculation for you as a historian?

It's safe space, colleagues -- can we have a bit more honest disclosure about our individual processes?

A follow-up that, I hope, is more responsive to the questions Lois posed. (Obviously, she does not need my pontifications about THE POETICS.)

When I mentioned Robert Penn Warren, I should have added that his narrator in ALL THE KING'S MEN might be a helpful guide in dealing with some of the questions she raised. Jack Burden is a history grad student drop out, who faces, not always successfully, the burden of recovering the truth of a past--including the racism of the South--that is not always easy to access. "All of us historical researchers," at one point he ironically proclaims in a work of fiction, "love truth."

Because Lois asks about dilemmas we face in our research, let me share a recent one I have faced and am doubtful that it is worth trying to solve. It is: what should we do when biographers of major figures use their sympathetic imaginations to enhance the story of their subjects by slandering the character of a minor, but once privileged, figure more or less forgotten by today's Reconstruction scholars.

I have in mind recent biographies of Grant intent on rehabilitating his reputation as president by among others Ron Chernow, cited as a positive example in a previous post. By now most professional historians--if not the general public--know that Chernow selects evidence partially. Grant's positive record on the KKK gets the attention it deserves. Grant's 1873 pardon of all Klansmen in federal prison does not get mentioned. Most Reconstruction historians also know that Chernow is one-sided in his portrayal of Grant's dispute with Charles Sumner. But Sumner is a major figure and some historians will come to his defense, as will John Stauffer in his forthcoming new biography of Sumner. The distorted portrayal of the diplomatic career of the historian John Lothrop Motley is another story. Portrayed as being simply a pawn in Sumner's dispute with Grant and Secretary of State Fish, Motley was fired by Grant and Fish as minister to London.

Retelling Motley's story does not raise all of the issues Lois confronts in retelling the story about a previously unrepresented African American woman. Motley was once celebrated with George Bancroft--another historian/diplomat--as the foremost historian in the US. There is plenty of information to be found about him. But telling his story does raise questions about proof for other reasons. Because, against his will, he became a figure of controversy, information about his record as a diplomat is often colored by whether someone sided with Grant and Fish or Sumner and Motley. Here the evidence gets cloudy. From May 1, 1869, to May 19, 1877, he kept a 1,100 page journal. But his heirs respected his admonition that “After my death these journals are on no account to be published 23 Jany/75.” G, W, Curtis, the editor of his letters, not wanting to stir up old quarrels, omitted any references to Motley’s disagreement with Grant and Fish. Grant's and Fish's friends were not so reticent. They aided the two in putting the most favorable spin on Motley's firing with an eye on both contemporary politics and the historical record. That spin was picked up by Fish's friends at Columbia University--he was Chair of the Board of Trustees. These friends were colleagues and friends with William A. Dunning. It is this version of Motley that Chernow and other Grant accolades perpetuate.

Today the so-called Dunning school's portrayal of African Americans has thankfully been discredited. But its portrayal of Motley continues to dominate. Lois deals with the complications of telling a story from scratch about an African American woman with insufficient evidence. Motley raises the question of how to retell a story in which the politics of Reconstruction historiography--past and present--impacts the evidence used.

My doubt: would anyone would be interested in an account that temporarily puts at center stage a Boston Brahmin who has become a Rosencrantz or a Guildenstern in the story of Reconstruction? .

Dear All, 

I've been scratching my head about the issues Lois raised in her original post and am circling back here in hopes of offering some more helpful input. I'll admit that I think the best advice I can offer on her original questions: 

  • How much conjecture does one allow one's self? 
  • When do you make an intuitive, speculative leap based on limited evidence?
  • Conversely, when do you decide to disregard what seems dubious in the historical record

is in fact general in nature. The closest thing I can get to a broad set of principles in response to these questions is that 1) how one answers these questions depends on what the purpose of the writing is and who one is writing it for, and 2) that there are necessarily trade offs once you consider (1). Let me try to apply these points to Lois's three questions in hopes that they at least help push the conversation forward. 

How much conjecture: I don't see the norms of the historical profession or our research methodology as in and of themselves requiring any limits on conjecture, as long as you make it clear to your reader that that is what you are doing. Biographers often try to step into the private lives and internal deliberations of the people they study, even where evidence is scanty. So I don't see any reason why you could not write an entire book on Bowser just offering conjecture, as long as you walked your reader through where you were pondering the unanswerable and where you were drawing on sources. But once you opt for a specific audience, such as a scholarly one through a peer reviewed article or a book intended for a broader audience, you may have to grapple with some constraints. A peer reviewer, I'd think, might ask what the purpose of the conjecture is. What does it do for your larger "intervention" and why is it necessary for them to follow you through your own process of coming to a clear opinion based on imperfect evidence? People reading the work in their free time might ask themselves whether the process of you conjecturing is what they want to read about. (As per my previous post on this thread, I actually think they might be quite interested in that... but am just guessing.) So once you've opted for a certain audience, you may face tradeoffs in terms of how conjectural you can be and whether you can hold your readers' attention. Would you agree with me that the problem with "sizzle history" is not that its conjectural but that the conjecturing is done in bad faith? The author doesn't want to understand the past, but to sensationalize it.  

Making Intuitive, Speculative Leaps: Like above, I see nothing methodologically wrong with making such leaps as long as you are transparent with your readers and introspective about how your own intuition might lead you astray. In my previous comment I suggested that people read history in their free time in part because we're social animals--we want to know and understand other people, dead or alive. If that's the case, won't your readers want you to render a coherent picture of who "Mary Bowser" was as a human? That will require you to take some leaps, since the extant sources only tell you so much about her. (I found Brook's initial point about rendering vivid accounts quite interesting.) Again, I think it is likely the expectations of your audience and the format of your writing that will impose tradeoffs on how many leaps you can make before you start to lose their attention. There is nothing about the study of history itself that prevents you from doing this endlessly, as long as you're conscientious about what you're doing. Incidentally, I remember reading John Demos's The Unredeemed Captive in college and one observer in the class pointing out that the book's conclusion, in which Demos tries to step inside the head of the captive based on the values of her culture, was nothing more than informed speculation. That speculation did, however, provide the book with a sense of narrative completeness, or so I think.  I should note that there are cases where I find the speculative leaps unconvincing, such as with Karl Jacoby's Strange Career of William Ellis. At root, Jacoby wanted his readers to like Ellis, and I'm not convinced Ellis himself would have cared for such sentiments beyond his ability to manipulate them. Perhaps one way to put this is that, as a biographer, don't you confront readers who will be relying on their own intuitions as they read yours? 

Disregarding What Seems Dubious: Isn't this just the other side of the speculative intuition question? Instead of adding in details that you can't back with sources but you feel connects the dots, you discount (the better word?) dots that you think lead the interpretive line astray. Transparency with your reader, so they know which sources you're marginalizing in your composite portrait, remains the key. Method itself doesn't forbid it, but requires justification. 

So I'm not sure how helpful that is, but if there is a single take away perhaps its that you have to think this through in terms of whether your readers will follow you where you are going. Historians often assume an infinite resevoir of will power and energy on the part of our readers to understand the past, when in fact both are limited and contingent--even among professional historians. When I initially raised questions about who our readers are and what they want out of our works, it's because how you answer those questions may dictate how you handle the three questions you raise. No doubt others here may have more concrete advice. 

Best wishes, 






As someone who writes both fiction and non-fiction I'd like to suggest that the issues surrounding conjecture raised by Lois apply primarily to narrative history and historical fiction. Scholars who take an analytical approach to their subject, meaning that their primary goal is to evaluate data, seek corroboration, document trends, and explore gaps in the narrative, don't worry about conjecture because it is part of the process. Conjecture can be presented as solution to a mystery or conundrum based on the evidence for (or absence of evidence for) the subject being discussed.

Narrative writers face a different problem because they are trying to write a story from A-Z. This forces them into a linear model of presentation that is not flexible enough to easily allow for conjecture unless the writer transitions from a narrative to an analytical discussion and then reverts back to the narrative. Scholars writing from an analytical perspective are not constrained by linearity. Their analysis of historical issues can be presented asymmetrically, i.e., from a variety of perspectives, as long as the question being explored remains at the center of the discussion. I hope this makes sense.

Thanks to Dave and Alex for keeping this thread going. As it happens, I've already written an entire novel about "Mary Bowser," (published by William Morrow/HarperCollins and pretty well regarded, including by some historians who indulge such flights of fancy as long as they are clearly labeled as fiction). In a novel, one must keep things historically plausible while glibly disregarded certain facts in favor of narrative arc, character arc, etc. But now, I am writing a biography of the real figure -- whom I prefer to refer to NOT as Mary Bowser, as she did not use that name in any extant documents I've located. But that is a topic for another post! For now, the point is that this project rests on the distinction between fact and fiction -- and as Dave reminds us, between sensationalizing (and/or oversimplifying) the past rather than seeking to understand it in its full complexity. (Alex, I'm not sure why you think narrative accounts don't require conjecture -- understanding why, or even how, a historical/biographical subject got from A to Z, or even just from J to K, requires quite a bit of conjecture. Unless one has been smart enough to choose a subject who left very detailed journals, correspondence, etc, which is not the case for my project. But if I'm misunderstanding your point, please clarify.)

All of this semi-hypothetical-how-ought crystallized a bit for me since my original post in this thread. I strolled into a particular archive with a pretty decent hypothesis about the relationship Mary Jane Richards (as she was then calling herself) had with a particular individual. But what I discovered in that archive -- not a word from her perspective, mind you -- suggests a much more complicated dynamic. "suggests" is a key term, because most of the historical evidence/documentation that might have revealed more is likely lost forever. Of course, now I must conjecture on what I did find, even as I continue to try to find more. But I will do so knowing that if I hadn't happened into that archive and that particular source -- which no one else had ever thought to consult -- I would never have up-ended my own earlier hypothesis.

It's thrilling to come across such a find, but also sobering, because it underscores how easy it is to think we understood someone/something, when despite our best efforts to date, we did not.

It occurs to me that I ought to post this original query on the H-Biography list, because despite my commitment to using the story of this life to tell larger stories about race in the mid-nineteenth-century US, there is a real difference in doing biography. Were I simply writing about Black resistance in the Civil War, or about any of the other themes in the chapter of this life (African American communities in the antebellum North, expatriation to Liberia, postbellum Black activism, etc.), I would not have to rely so heavily on conjecture because I could work with a broader range of evidence. But when the extant evidence is not directly about this particular figure, or those with whom she interacted, I wander into extrapolation, and conjecture, and so it goes.

I note that we still aren't hearing from other folks offering up some specific examples from their own work. Are you all somehow operating with more certainty?

I spent a couple of years researching the publications and biography of a relatively obscure (to modern scholarship) popular schoolbook author.  I found that the absence of any personal papers beyond a single large manuscript notebook thoughtfully preserved at Columbia imposed severe constraints on the research and pushed me into the realms of conjecture more than I originally wanted.  This person had a very long and vigrous career of pubIishing dozens of books and inventing interesting educational contraptions and classroom aids, and was Harper Brothers’ top-selling text author in the 1860s and 70s, with copyright revenues approaching a princely $200,000.  Yet I could not hunt down any surviving personal papers; the accumulating tangential evidence from family wills, Harpers contract files, local newspapers in his home town, and other detritus led me inexorably to the local historical society, to which his family had donated his personal library upon his death at 92 early in the 20th century.  The local society claimed it had nothing on him, but it also wouldn’t let me search its unorganized collections where I am confident a good deal of pertinent information if not personal correspondence, is probably stored.  (My struggles on that end are another story of the type I’d love to see an H-Net network tackle, concerning local archives and the pros and cons of their gatekeeping of materials having scholarly significance).

All of this goes to the point made in this thread about conjecture, leaps of faith, and Chernow’s remarks about how the dead returning to life could completely reverse what we’ve said about them.  In my publications about this person, I offered biographical details found nowhere else that I had corroborated, but also with the caution that his personal papers remain elusive.  I also was forced into the difficult choice when working with incomplete materials: to withhold this entire evidence from the jury or to present it, along with my inferences and conjectures, for that jury to assess.  Like other scholars in similar straits, I took the latter course.  Readers want conjecture because they appreciate the dilemma the writer faces with incomplete or absent materials, and they expect the author to offer his or her own informed observations in those circumstances, always with appropriate cautions about the tentative nature of conjecture.  The notebook I mentioned is a trove of fascinating drafts of letters, articles, textbook plans, and occasional budgetary information that would have been much more valuable when compared to his personal papers and incoming correspondence.  But I had to work with what I had.  It’s possible that someone in the future will uncover this person’s papers and publish a dissent about my conclusions and conjectures — in which case, the record will be much improved and the story more complete.

I've begun preliminary work on another prominet educational figure of that time and, well, guess what: no surviving papers (yet found)...



To continue the dialog, I offer a few more thoughts and examples.
First, on sources and conjecture. Source material is invaluable for historians and biographers. As Taylor Branch quipped in writing PARTING THE WATERS about the Civil Rights Movement, all biographers secretly wish that the FBI wiretapped everyone, as they did Martin Luther King, Jr., for the evidence it provides. Nonetheless, even the best sources rarely eliminate the need for conjecture. All sources need to be interpreted. The same source can lead to different interpretations. Sources also have to be evaluated for reliability. Even data quantitative historians rely on for their analysis comes from sources that may not be perfectly reliable. Will the 2020 census data be reliable for historians in the future? A major controversy over the legacy of Johns Hopkins results from conjecture based on census reports. Sources can often be contradictory, especially when they involve historical controversies. Some of the most complicated sources come from unreliable narrators. Like an unreliable narrator in fiction, they are rarely completely unreliable. Most of the events they report took place, but their interpretations of them are sometimes trustworthy and other times not. Conjecture plays a large role in deciding what information is reliable and what is not.
Having made those preliminary observations, let me return to my earlier post on the dangers of sympathetic conjecture by those writing biographies of one figure that might affect readers’ understanding of another figure. My example was recent biographies intent on putting Grant on the right side of history that feel the need to take sides in his ugly dispute with Charles Sumner and how that affects their portrayal of John Lothrop Motley, a once well-known historian/diplomat largely forgotten today.
Both Charles Calhoun and Ron Chernow want to justify Grant’s demand in 1870 that Motley resign as US minister to the UK even though the treatment of Motley was at the time and for many years after highly controversial. As with any controversy, there is conflicting evidence. A major problem with Chernow’s and Calhoun’s partisan portrayals is that they cherry-pick evidence. For instance, Chernow justifies Grant’s action by writing that Motley was a “pariah in London” who was “banished from polite society” (698.) But the London Daily News responded to Motley’s “unexpected and unexplained recall” with “extreme and unfeigned regret” and praise for his “firmness and discretion.” No minister could be “more sensitive to the honor of his government, more attentive to the interests of his country, and more capable of uniting the most vigorous performance of his pubic duties with the high-bred courtesy and conciliatory tact and temper that make those duties easy and successful.”
But the portrayal of Motley also raises questions of conjecture linked to uses source that are not fully reliable AND ones that are, for unknown reasons, not available. In what follows I will provide examples of the following:

1) Two cases of what I consider inappropriate conjecture. One by Calhoun. One that I decided not to make in my unpublished effort to correct him.
2) The dilemma of what to do with sources that, although somewhat helpful, are also unreliable.
3) What I consider legitimate conjecture about a missing source.
Early in his biography of Grant as president and long before he deals with Motley’s firing in 1870, Calhoun works to undercut Motley’s credibility by singling him out in his condemnation of self-seekers at the start of Grant’s term “who had shown not great solicitude for Grant’s candidacy” but then professed loyalty “with an eye to their own advancement” (53). Calhoun provides no evidence for singling out Motley. On the contrary, he fails to mention that Motley campaigned for Grant and had repeatedly praised him. After Lincoln’s assassination Motley wrote: “I am no great admirer of military heroes, but we needed one at this period, and we can never be too thankful that exactly such a one was vouchsafed to us – one so vast and fertile in conception, so patient in waiting, so rapid in striking, had come, and withal so destitute of personal ambition, so modest, so averse to public notoriety… So long as we can produce such a man as Grant, our Republic is safe.” Furthermore, Calhoun’s accusation that Motley used Grant to get his ministry makes no sense. Motley was already famous, and diplomacy was a financial sacrifice. A contemporary called him “loved and honored” and “the most brilliant historical writer and most accomplished gentleman of his generation.” As a minister, he received no housing and spent twice his salary on social functions. He did not need the ministry for his “own advancement.”
For his account of Motley as minster in London Calhoun relies heavily on the diary of Benjamin Moran, who was secretary of the legation. That seems to make sense. Indeed, Moran’s diary is a helpful reminder of much that transpired in the legation. BUT Moran’s judgments are often unreliable. A friend of James Buchanan, who first got him his job, Moran hatred Bostonians. As the not always reliable Henry Adams warned, “Benjamin Moran . . . had an exaggerated notion of his importance; he was sensitive to flattery, and easily offended . . . His diary must be read from the point of view of his character.” Convinced that he deserved equal social recognition as a minister, Moran resented Motley for his acceptance in English high society. Nonetheless, without providing any context, Calhoun cites Moran to label Motley a “snobbish and impetuous minister” (165).

In writing about this use of Moran, I contemplated making a conjecture of my own. There is evidence that Moran may have been gay. The only minister he admired and befriended was Buchanan, also possibly gay. To be sure, Moran had married an Englishwoman, but she was nine years older, and when she died after an unhappy marriage he never married again. His diary has numerous detailed and elaborate physical descriptions of the men he meets, but not of the women he meets. In the end, however, I decided not to make this conjecture or even to hint at it. It would not add to my argument and might undermine my ethos.

But there is another issue involving sources that did prompt some more interesting conjecture. Just as Moran kept a diary, so did Motley. He began his ministry in England spring 1869 and died July 1877. From May 1, 1869, to May 19, 1877, Motley kept a 1,100 page journal. Although obviously told from his point of view, it would be extremely valuable to have access to it. But in January 1875 he wrote at the beginning “After my death these journals are on no account to be published.” Why would a historian whose own histories were based on long hidden archival material write these words, especially when he knew that Grant and others were spinning an unfavorable account about his record as a diplomat? Acknowledging that it is conjecture, I came up with one possible answer.

Unlike Sumner, who kept his fight with Grant public, Motley said good bye to political squabbles and returned to writing history—provided a house by the Queen of the Netherlands to conduct archival research in the Hague. In late December 1874 Motley’s beloved wife, never recovering from the shame of his recall, died. He wrote his admonition not to publish his journal the very next month. One plausible reason he did so is that he wanted to have nothing more to do with a public dispute that had contributed to his wife’s death. Indeed, their two daughters honored his instructions and never published the journal. They too would have been motivated to put the dispute behind them. Both daughters married important Englishmen, the elder the grandson of playwright Richard Sheridan, the younger Sir William Harcourt, who would become Home Secretary and Chancellor of the Exchequer. The lives of the daughters and their children were in England. When Motley died in July 1877 he was honored with a funeral ceremony in Westminster Abbey. He and his wife were buried in England. Better to let them rest in peace than get involved in a dispute across the ocean that had ruined both of their lives.

Who knows if that conjecture is true? But even of it is not historically accurate, it reminds us of gaps in the sources upon which all of us rely AND, more interestingly, that private attempts to honor someone may hurt the public’s perception of that person later in history.

Brook Thomas