painting of christine de pizan
Historian Christine de Pizan (1364 – c. 1430) from the Harley MS, “The Book of the Queen,” at the British Library.

How do we think about our responsibility as researchers and writers to communicate the significance of our work to our professional peers, as well as to engage the public?

Last week a conversation erupted on Twitter, prompted by a keynote address at a conference on “High Stakes History” at Columbia. The event was part of Columbia’s “History in Action,” a pilot program funded by the American Historical Association-Andrew W. Mellon Foundation Career Diversity Initiative. A lot of social media focused on Jill Lepore’s keynote. A Professor at Harvard and a staff writer for the New Yorker, and the author of an enviable corpus of thoughtful, engaging and briskly-selling books on a wide range of topics, Lepore put her hand right on the high voltage rail of a humanist’s nervous system. Who do we write for, and why do we write the way we do?

All academic disciplines struggle with how to engage the public. We worry about the role of public investment and access to research, about the public’s understanding of key concepts in the humanities and the sciences, about the applicability of our work for modern democracy. History has a unique public audience for a variety of reasons. In writing last month about the wild success of the musical Hamilton, I noted that history seems accessible, or at least like it ought to be. There is the stuff that happened in the past, and then you just write about it. Right? And people want to read about it—Barnes & Noble has the equivalent of Kellogg’s cereal aisle shelf space set aside for American history, seemingly lots of it about the founding generation and the Civil War.

Of course it’s not really that simple, and actually many readers don’t expect it to be. Open access advocates are sure there are important audiences for academic research, and at least at the journal I know best, the William and Mary Quarterly, we do have a robust non-academic readership. The question is what constitutes purposely “writing for the public,” and how that differs significantly from the writing academics do for one another — and which might also be of interest to the public. For historians this often boils down to “narrative” versus “argument.” “People care about stories, not arguments,” was one tweeted paraphrase of Lepore’s talk. Storytelling is one of the oldest human forms of communication. It is not a simple thing to tell a story well and with meaning. One of the masters of the genre (and Lepore’s teacher), John Demos, teaches a course on narrative history that pushes students to think about form and expression as well as evidence and argument. These debates about narrative versus argument have been happening for eons; I imagine Thucydides saying “look, guys, narrative is the only way to write history of the Peloponnesian War.”

The question ought not be, however, one versus the other. Academic writing is expository. For academic writing, argument is essential, and narrative is optional. Academic research is the accumulation of new information by many different means. The significance of this information is articulated through evidence-based argument, the heart of historical disciplinary practice. Argument doesn’t preclude narrative — a very fine writer can craft a narrative that conveys a variety of important arguments, but pure narrative can never substitute for argument in professional exchange.

Why not? Don’t professional historians appreciate a good story? Every historian I know loves a good story. Academic writing, however, is the formulation of research into new knowledge. That might be in the form of genuinely new information, or it might be an importantly fresh perspective or interpretation. Using new methods and tools as well as the regular revelation of new materials means that historians are generating new knowledge at a rapid clip.

So how do we know what’s new? A fundamental responsibility of academic writing is to explain the relationship of new scholarship to its forebears. Knowledge doesn’t accrete in a linear or progressive fashion, of course, but explaining how research and interpretation is related to the literature that’s come before it is fundamental to our evaluation of the work. After all, historians have been writing about the American Revolution since shortly after the American Revolution. As a professional historian, how would I know whether the next book I see on either an oft-studied topic or an entirely fresh subject is important to read and digest, to inform or incorporate into my own research perspective or plans, and to integrate into my teaching? I just watched an exchange between an experienced former journal editor and a manuscript reviewer who asked “if I think I’ve seen something like this argument before but I can’t quite place it, what should I do?” And of course the former editor encouraged the reviewer to try to address that issue as fully as possible, noting that expert peer reviewers play a key role in signaling to editors how a submission relates to the existing scholarship. In other words, historians are particularly attuned to the history of history.

Historians can and do play a vital role in the public humanities. As journalists and media critics, and in podcasts, blogs, and more traditional outlets — including books — historians are and need to be voices for perspective. There are vital reasons not just why but how we write for one another, too.



Karin Wulf

Karin Wulf

Karin Wulf is the Beatrice and Julio Mario Santo Domingo Director and Librarian at the John Carter Brown Library and Professor of History, Brown University. She is a historian with a research specialty in family, gender and politics in eighteenth-century British America and has experience in non-profit humanities publishing.


38 Thoughts on "The Importance of Academic (History) Writing"

I think there’s potentially a lot to say about the concept of “new knowledge” and how that’s formulated through institutional structures and traditions, which may or may not be of service to different sorts of goals held by historians.

But that’s not why I wanted to comment. I really just wanted to include the link to the tweet which Karin quoted above without citing, so that she can incorporate the link into the text and thereby give appropriate credit to its author. Obviously blogging citation practices are fluid, informal, and open to discussion, but I think there should be a consensus that if possible, we credit anything that we directly quote.

Thanks, Tom and yes, absolutely– I think that one’s embedded now though I was really responding to the full exchange (of which you were a part!)– if I were clever enough I’d have storified it.

There was a period of time, especially in the 1980s, when quantitative history, or “cliometrics,” was all the rage, and being very technical and mathematical, it was mostly not accessible to the general public. A few, like Fogel and Engerman, with their trade book “Time on the Cross” made an effort to reach a wider audience (by stuffing all the most technical material into appendices, as I recall), but most quantitative historians didn’t even try. The return to social, cultural, and political history brought narrative back to the fore and once again made scholarly history available to nonexperts.

Thanks, Sandy. I can’t share your categorization of quant research = not accessible and social-cultural history = narrative thus accessible, but I appreciate your reading the post and sharing your perspective.

Well, perhaps that is too stark a contrast. No doubt there are many social/cultural histories that are not written very accessibly either. But here is my question back to you: can you think of ANY other book in quantitative history that succeeded as a trade book besides Fogel/Engerman?

Yes, please keep writing books. Made up of those flat paper things (what do you call them?–printed pages!) set inside cloth-covered boards. One thing, one constant thing, that the public(s) could be made aware of is the axiomatic understanding that history is not experience. History is a record of experience. As such, come into play perspectives, as cited above, and agendas of the recorders–and scholars as the re-recorders. Which isn’t bad; it’s simply real. (A good helpful contribution from post-modernism.) Thank you for this contribution, a fine–refined–and needful assessment, analysis, and argument.

Thanks– I think historical method, that is, the selection, analysis and interpretation of historical source materials is as critical (maybe more) to teach than events of the past.

I have had the opportunity to study historical reasoning and writing, along the lines that Karin describes. There are indeed two kinds of new knowledge. First, and perhaps simplest, is when an historian reads, say, a diary and reports for the first time the events described therein. More compex is when a new explanation for well known events is offered. In this case history is, or at least borrows from, the social sciences. New concepts or theories in these sciences, and even the physical sciences, can be brought to bear for the purpose of explaining past events. These may range from economics and psychology to climate change. I recently encountered an historical explanation based on memeplexes, something I had not heard of before. See

As to the audience, one cannot expect the lay reader to know the literature on a given event, so it is very difficult to present the technical arguments found there. I do not think this is a matter of the lay readers not wanting to hear these arguments, that is wanting just a story, but rather of the arguments’ necessarily technical nature. The same is true of popular science writing. But then this is an empirical question. Then too there are many degrees of lay knowledge.

Thanks, David. Like all disciplines, history has methods and tools of analysis that change over time and with the practitioner. They aren’t necessarily unavailable to anyone else but are part of the training. And the literature reflects that.

I have a somewhat different view on all of this. There are several definitions of academic/scholar, most of them somewhat unflattering.
* Of or relating to institutionalized education and scholarship, especially at a college or university.
* Of or relating to studies that rely on reading and involve abstract thought rather than being primarily practical or technical.
* Relating to scholarly performance: a student’s academic average.
* Of or relating to the conservative style of art promoted by an official academy.
* Having little practical use or value, as by being overly detailed, unengaging, or theoretical: a dry, academic exercise.
* Having no important consequence or relevancy.

This comes with the territory when the dichotomy of academic vs. non-academic is created and used. In simple terms construction work by actors (in history of course) has consequences. In this case to separate academics (i.e., professional historians) from non-academic butchers, bakers, and electricians. But we should not forget that most of the actors in the ongoing story we call history are just this – non-academic butchers, bakers, and electricians. The historian’s role (as a professional) is to describe and communicate as clearly as possible the history mostly created by non-academics. Too many historians – but still much fewer than social scientists – insist on elevating their theories of history above the history that’s actually constructed. I say all this recognizing, of course that historians’ theories are also a part of history’s construction. So in studying history, as professionals and communicating what we uncover, we need to always keep an eye on the work of the historian (all historians) in constructing the history that’s communicated. To see the accuracy of these claims we need look no further than the several “schools” of US history that have described and explained the nation and its developments over the last 250 years. From Adams to Beard to Turner the descriptions and explanations change because the people who were making the nation changed and thus changed the nation and its history. In other words the explanations and theories of history invented by non-historians are almost always more complex and original than those historians can and do invent. Sometimes it’s difficult in fact for historians to keep up.

Ken, this seems rather confused, or at least I have trouble understanding it. You seem to think that the fact that writing history is itself an event in history, is somehow important when it comes to the act of writing. I do not see that, except in the sense that one can only use the concepts and tools available at the time.

I am particularly concerned by this statement –“Too many historians – but still much fewer than social scientists – insist on elevating their theories of history above the history that’s actually constructed.”

On the contrary, explanation (as opposed to mere description) is central to historical writing, just as it obviously is to social science.

But which explanations and descriptions should historians (and even more so social scientists) focus on? Since historians and these scientists supposedly focus on studying some object, just like physicists’ object of study is physical events. So as physicists base their descriptions and explanations on what they are able to piece together from the objects they observe, shouldn’t descriptions and explanations of historians and social scientists be based on what they are able to piece together from observing and studying such events as elections, stock markets, war, fashions, etc. Why do historians and social scientists need to add to these? Physical scientists don’t — look closely and often enough and a star will reveal all of itself. Sound advice in my view for all historians and social scientists.

And the fact that doing history (studying, writing, etc.) is itself a part of history is not something for debate. It just is. How that effects the work of historians and social scientists varies from situation to situation. But it always in some ways effects that work.

Looks good, Sandy, albeit pricey. I no longer study the logic of historical explanation. When I was at Carnegie Mellon the nascent philosophy department was joint with the history department, so I spent a lot of time with historians, including reading their stuff and team teaching. Since my research was on the logic of complex issues, I found myself studying their reasoning.

Roberts just rehashes the old debate about “covering” laws in the explanation of historical events. His solution to the the validity of their use — that, though useless in explaining macro-events, covering laws are indispensable in connecting the steps in an explanatory narrative — is no resolution at all. The fundamental problem with covering laws is that to work as proposed they would have to exist and originate outside of history. From some higher power or a realm beyond the human. Such power and realms may exist. They are just beyond human perception and understanding. So what do we gain by attempting to explain history through their use? Nothing. Plus the downsides of using them are multiple and serious. So I suggest the entire notion be assigned to the dustbin of history.

Covering laws arise outside of the discipline of history when their methodology originates in the social or physical sciences. Examples range from game theory to climate change. There is no mystery here. You seem to be confusing the academic discipline of doing history with history itself.

Covering laws are so called “general propositions” that supposedly “correlate” events of a type (i.e., wars) that we want to explain. Such as all wars involve combat. In simple terms the statement follows that when a war begins combat will follow. First, such successions of events sometimes turn out to be wrong. For example, the “cold war” involved no direct combat between the US and USSR. But more than that such supposed laws can make us lazy in observing events. And for scientists that’s a bad thing. Using observed regularities from past observations to help with future observations can be useful, so long as we don’t let them distract our attention from current observations.

There is much more to explanation than careful observation, usually some sort of causal mechanism. Stars do notvreveal all of themselves, by any means. There are mechanisms like fusion, gravity, electromagnetism, etc. Same with history.

David, explanations and theories of what events mean come with the observations of the events. They are created by the same interactions by which the events are created. Why would an historian or social scientist omit these as part of the research. The US Civil War was fought, according to those who created the war, fought in it, ended it was: 1) to free the slaves; 2) to protect the union; 3) to protect states’ rights; 4) to defend the south; 5) to protect slavery. Does the Civil War historian need to add other explanations? It seems examining these will occupy the efforts of any historian for an entire career. The historian might bring some thoughtful questions and a plan to examine these more closely. But this is the civil war as it existed. And yes there is more involved than just observation, but not much more. In every instance historian and social scientist must return to observation of the events under study, or a surrogate for those events (i.e., model, experiment). Or risk becoming disengaged from those events as economists have with economic events and sociologists have with social events.

Ken I do not understand your first sentence. Your “come with” seems ambiguous at best. Our understanding of fusion did not come with the observation of stars. But the explanation of stars as fusion reactors obviously required that stars first be observed.

Your five observations of the Civil War call cry out for explanation. Why did things go the way they did at this particular time? Note for example that southern slavery was not under direct attack. It was merely prohibited in the new territories. In fact the south attacked the north, not vice versa. Why?

Your last sentence seems simply preposterous to me. You seem to be questioning the validity of entire scientific fields.

When events like stars are observed scientists take empirical events of varying sorts and attempt to put them together to form a summary of what the observations are. This is done again and again. Eventually scientists begin to agree on a summary of what’s happening. And then keep observing, which may of course change the summary. In simple terms the star explains itself. Scientists just try to get the summary right.

Why do the explanations of the civil war need to be explained? Yes, investigated for origins and pathways. Yet, investigated as to who/what proposed them and how they did so Yes, investigate as to forms and disputes. But they don’t need to be explained. They are explanations! And along the way in observing we may likely run across other explanations we missed.

Science works because it is so simple. Just observe and report. (I fully acknowledge here that observation can and often does become quite difficult and complicated. So the simple formula is often not simple at all. But the basic formula remains.) Look for how things are built, by whom, in what ways, and over what time period. That holds whether studying stars or the civil war. Stephen Hawking is a great theoretical thinker but as he theorizes he is not really a scientist. He may inspire scientists, as did Sagan and Einstein. That’s not a bad thing. Just a different thing.

Scientific explanation involves far more than “summaries of observations.” Scientists do not observe as though their minds were blank slates simply absorbing empirical data. They come to observe in order to test hypotheses, which are themselves derived from theories they develop, which are only partly the result of prior observation. Even the instruments scientists use to observe are theory-laden. And observations are themselves not always determinative in confirming or disproving theories. As Pierre Duhem long ago taught us, there are multiple layers involved in science in any one of which changes can be made to bring theories into alignment with observations. Ir is a much more complex business than you make it out to be.

I agree, Sandy. In fact I have long argued that science (including scientific communication) is one of the most complex of all human activities. It involves millions of people, around the world, building systematically on one another’s findings and ideas, over hundreds of years. It is also poorly understood.

But I think we have wandered far off topic.

Sandy and David, if science were a religion or an eschatological system I would agree with your assertions. But it is not. It’s just a bunch of folks working hard to figure out how things work. That includes things like stars and things like wars and things like love. And the only, I repeat only source for such “figuring out” is the things themselves. We may “guess” about such things and talk about their control by gods or natural processes, but really the only guide we have to reach any understanding is the things themselves. I respect scientists’ guesses and for experienced and mature scientists such guesses (theories) are probably more useful in advancing understanding than just asking the “person on the street.” As Curie observed many times science is mostly just hard and continuous work. Don’t glorify it. Just keep working. In my experience historians have a similar set of arrangements. Lots and lots of hard work to figure out something from the past. And like scientists the only guide historians have for this work is the thing(s) themselves. Hopefully as historians we can listen intently and closely enough to hear what these things say.

If you read your Kant, you’d know that the “things themselves” are not directly knowable. Categories of the mind filter all perceptual information. And much in natural science is not directly perceivable anyway. Are light waves, electrons, musons, bosons, etc. “things”? What about the “strings” in multidimensional string theory?

The best and most kind thing I can say about Kant’s philosophical system is “Thank God it’s now dead.” Kant summarized all the developments of the Enlightenment up to about 1790 and that’s generally where the world of philosophy in Europe has remained since then. His system excluded any way to assess or critique the assumptions on which these developments are based. His system institutionalized the separation of necessity and freedom. Forcing morality, objectivity and art into separate pathways, separate realities if you will. For over two centuries those who came after Kant have worked, most unsuccessfully to figure a way out of the devilish trap his system creates. American philosophy was spared this devilish trap. (We got them indirectly mostly through the European economic theorists we imported.) In my view mostly because of Peirce, William James, and Pragmatism. But until the 20th century American philosophy had little influence in Europe.

Some of the things for which we have Kant to thank The bifurcation between subject and object, law, which your comments clearly employ. But more than that Kant’s system made it impossible to focus the needed attention on the advent and impacts of the preconceptions (ecological and political) upon which industrial revolutions were based. That our issues with ecological and economic impacts of these revolutions remain unresolved today we owe largely to Kant.

Ken, it is certainly true that science is about figuring out how things work and the only way to do that is to study the things. This is trivially true, but it in no way follows that science is simple. On the contrary, figuring out how things work means explaining how they work, which is a very complex process that has taken centuries of building to date. Science is also very creative, constantly creating new explanatory concepts, which turn out to be discoveries, like the sequence molecule, atom and quark, none of which were directly observed. You seem to have missed the creative part.

Ken: On an historical note, consider the following sequence. Planck “guessed” (your term for it) that heat came in irreducible lumps, which he called “quanta,” Einstein extended this guess to light. Bohr then extended it to electrons, thereby making Rutherford’s guess (that the atom was like the solar system, with electrons circling a nucleus) stable. This created atomic science, where others guessed fission, then fusion. The final guess explained that stars shine because they are fusion reactors, finally answering a question that was thousands of years old. The whole thing took about 60 years, these “guesses” are now fundamental knowledge and the guessers are regarded as among the greatest scientists in history. These quanta have never been directly observed.

Science is certainly a collective enterprise. A community. And the creation of communities is always a creative process. Artistic if you will. So I agree that scientists are indeed creative in the construction of the scientific communities of which they are members. Observation in science is also a creative process. Witness the different approaches (methods) for observing and the varied meanings given to those observations (models). But there is nothing unusual about this process. As I said making communities is always a creative process. Take a look at how scientific communities describe themselves. As communities of observation and testing. Observation is the central element of scientific method or process. The core skill of the scientist is to make observations and then test these with further observations. On atomic observations billions of dollars have and continue to be spent on observing, especially things like quarks and photons that are difficult to observe. Creativity at times also involves guessing. For scientists for example, guessing about the fuel source of stars. My concern about scientists guessing is that the guesses may interfere with the observations and their compiling. Baring that scientists have my permission to guess as much as they like. On a personal note. Just returned from a meeting of physicists on string theory. One of the concerns voiced there is that string theory is indeed beginning to interfere with observing. But in my view physics ceased being a science over 50 years ago. Second personal note. Imaginary constructs are quite common among scientists. Gravity is probably the most famous one. They are partly a short-hand to aid conversation and partly a wish that something like this exists. Or otherwise many observations make no sense. And that’s frightening, especially for scientists. For historians the most common imaginary construct is “history.”

Guess you haven’t read Rawls, Ken–the influence of Kant on modern philosophy remains strong. As for what you describe as the process of science, Kuhn would call that “normal” science. He wrote a whole book, you may recall, about “The Structure of Scientific Revolutions.”

I assume you mean John Rawls, not Lou Rawls. Although the latter has more to offer in my view. John Rawls signature work, “A Theory of Justice.” is a good example of how Kantianism has failed. First, a damning critique of Kant’s system for a historian is that it is “ahistorical.” There is no time, no history in Kant’s system. Rawls relies on Kant for what he calls a “human flourishing.” In other words there are innate qualities in humans that will make liberal democracy workable. Humans do not become through history, through time, they simply have certain innate qualities that can be latched onto. As an historian this is nonsense. Second, Rawls work hides from us once again the preconceptions of the Enlightenment. Preconceptions that need to be revealed and examined. If for no other reason than that some of their intended consequences and certainly some of the unintended ones are threats today. How can we address them when we can’t even see them?

I’ve read “The Structure of Scientific Revolutions,” several times. It’s a remarkable work. But the social studies of sciences (SST) did not stop with Kuhn, as unfortunately western philosophy did with Kant. SST began to look at the work of scientists and to describe what science is in practice and what scientists actually do. That’s why I’ve said in this string that science and the work of scientists is relatively simple. The most complex part was stripping away logical constructs and philosophical principles that said science and scientists must be this and not that, when in practice that was actually a better description of the work of science and scientists. In defending the work of scientists and the continued funding of science projects we must not treat science as something other than it is – a community of workers trying to figure out how things work. I don’t accept the critiques that science in now treated as a religion by some of its supporters. But that tendency does exist and we need to tread carefully that it does not become reality.

Good point, Sandy. Actually Ken’s purported critique of various extraneous aspects of Kant’s work is irrelevant to the issue at hand, which is Kant’s theory of categories.

But then we have learned a great deal since Kant’s time. Philosophy, like science, is progressive, as is historical analysis. I would argue that the role of concepts in analytic philosophy (where analytic refers to concept analysis) is the direct descendent of Kant’s categories. As Wittgenstein put it, one can only think about what one has the concepts of. But then concepts can change over time, especially in science. In this regard the limits of science (at a given time) are the limits of language. (By coincidence the short title of my Ph.D. thesis was “Concept Change in Science and Philosophy.)

But I see no connection between Ken’s unusual views on science and Kuhn’s account of normal science. (Kuhn’s theory of science was the starting point for my thesis.) On the contrary the position I have been articulating is very Kuhnian.

This is from Pandora’s Hope: Essays on the Reality of Science Studies by Bruno Latour, p. 6. I share his views on Kant. I have nothing more to add on why I view the end of the Kantian system favorably.

For the banquet of realty, the mind provided the food, and the inaccessible things-in-themselves to which the world had been reduced simply dropped by to say “We are here, what you eat is not dust,” but otherwise remained mute and stoic guests. If we abandon absolute certainty, Kant said, we can at least retrieve universality as long as we remain inside the restricted sphere of science, to which the world outside contributes decisively but minimally. The rest of the quest for the absolute is to be found in morality, another a priori certainty that the mind-in-the-vat extracts from its own wiring. Under the name of a “Copernican Revolution” Kant invented this science-fiction nightmare: the outside world now turns around the mind-in-the-vat, which dictates most of that world’s laws, laws it has extracted from itself without help from anyone else. A crippled despot now ruled the world of reality. This philosophy was thought, strangely enough, to be the deepest of all, because it had at once managed to abandon the quest for absolute certainty and to retain it under the banner of “universal a prioris,” a clever sleight of hand that hid the lost path even deeper in the thickets.

As to my views on science they are generally representative of those of SST researchers and writers. If anything my views are somewhat more moderate than the mainstream. As I said I have great respect for Kuhn’s work on science. But SST has gone well beyond that work over the last 40 years.

A good book on how concepts change in science is Stephen Toulmin’s Human Understanding: The Collective Use and Evolution of Concepts, Vol. 1 (1977), which I acquired for Princeton U.P.. That was the only volume he finished, alas.

Comments are closed.