Question of the Month: June

Simon Purdue's picture

H-Nationalism’s Question of the Month series offers a forum for discussing the big questions surrounding research, pedagogy, and practice in the field of nationalism studies and the history of nationalism. Use the reply feature to join the conversation! Email Simon Purdue (purdue.s@husky.neu.edu) of Northeastern University if you’d like to propose a question of you own. If you need technical assistance with logging in and posting comments, please contact H-Net’s Help Desk (help@mail.h-net.org).  

 

Dear Subscribers,

I hope you are all keeping safe and well in these times of unprecendented crisis. We are seeing events unfold that are shaking America to its core, and as scholars of nationalism, race, and related fields it is vital that we pay attention, listen and acknowledge what is going on in every city in the nation. 

Although at times such as these we may lose track of time, another month is upon us, and so too is another question as part of our Question of the Month series. This month we are asking:

Which digital research methods are having or should be having the biggest impact in nationalism studies? Have approaches like GIS, topic modeling, or network analysis changed how we think about our topics? Do these methods bring methodological risks with them?  

As always, we look forward to recieving your comments and ideas on this question. Please don't hesitate to reach out directly to me if you have any questions!

Best wishes,

Simon Purdue -- Series Editor

It is getting late in the month and I would be sorry to leave Simon’s fine question unaddressed. It concerns me and my work closely (and here I should declare an interest).

For the past 10 years I have relied increasingly on digital humanities to pursue my research on the emergence and spread of cultural nationalism in 19th-century Europe. My main research output is in the form of a multimedia interactive website including many forms of GIS positioning, geographical and social network visualizations etc. its is called ERNiE (short for "Encyclopedia of Romantic Nationalism in Europe", companion portal of the book of that same title) and can be accesssed at http://ernie.uva.nl. A manual is on-site.

In my view the spread of nationalism, at least in 19th-century Europe, works through an alternation of moments of diffusion and condensation. The latter is what we are familiar with as activism on the ground, located in society and usually studied within its proper societal parameters. To restrict ourselves to analysing those manifestations of nationalism entails, as we know, the risk of methodological nationalism. To counterbalance that it is useful to factor in the diffusion elements: how the "message" of nationalism spreads (also between and across "nations" and countries), mainly carried in the form of human mobility and cultural communication. This is a very complex process, involving cultural expressions in various fields of knowledge production and artistic production, from folklore, history-writing and philology to historical novels and choral festivals. These fields mutually interact, cultural actors play a role in politics, and things spread between countries. This means we need to trace a great multitude of cultural gestures in many directions. Digital humanities are an indispensible aid to help us see the forest through the trees, or indeed through the swirling clouds of pollen in the air, and to maintain chronological anchorage in the process.

Given that complexity, the advantage of computer-assisted research is obvious, but for me is mainly twofold. [1] Managing and spatially locating (GIS!) large datasets of cultural interventions and repercussions breaks through the methodological nationalism that our archive-based, nationally-compartmentalized documentary sources tend to impose. [2] It demonstrates that culture, for all its diffuse (indeed unnoticeable and, yes, "banal") presence, predicts rather than follows political developments and has definite agency in the history of nationalism.

Thirdly, we learn from our tools. To capture something as multi-meaningful as "culture" into a study of national movements, is itself a very useful exercise in topic modelling.

Dear All,

My own experiences with the digital realm are more modest than Joep’s but hopefully still afford some insight. As some subscribers may remember, a while back I did a series of interviews (via email) with various scholars about digital projects. Those are here, here, and here. I would be interested in seeing these renewed and, if anyone wants to take up the torch, you can feel welcomed to reach out to me at (editorial-nationalism@mail.h-net.org).

Let me offer two comments on the original question. The first is that digital innovations are so deeply rooted in how we work that it is easy to overlook their influence. These digital innovations don’t all rise to the level of a methodology, but their transformative impact is nonetheless there. For example, when I assign older monographs in graduate reading seminars, especially works from before the mid-1980s, I often explain to students that they have to approach them slightly differently because they were composed and edited by hand and by type writer. The process of editing and slimming text is so much easier than it used to be that the standards for concision and interpretive front-loading are much higher than they once were. There are of course other factors, including the economics of publishing and the powerlessness of early-career scholars, that contribute to this, and it would be silly to argue that scholars writing before the PC were incapable of brevity and wit (who among us can match Imagined Communities), but I think there has nonetheless been a shift in expectations. Anyone who has slogged through E. P. Thompson’s classic The Making of the English Working Class knows the feeling of understanding that a work is important but struggling to find where its core argument lies (outside of its first few pages). Indeed, many canonical works in U.S. historiography—I think of David Montgomery’s Beyond Equality among others—meander from primary source to primary source before a series of concluding paragraphs that lay out a chapter’s key insights. This is much less common today. Because the writing-and-revising process is so integral to the generation of insight, at least for historians (I can't speak for other disciplines), I’d suggest that the rise of the PC has led to more than just cosmetic changes.  

Another subtle but transformative change has been the rise of digitized, word-searchable databases. Here again the issue of convenience may obscure a deeper change. In a sense, the evolution of historiography has tended to be path-dependent, at least if one thinks in terms of the rapidly evolving bodies of regional scholarship that emerged in the middle of the 20th century. Early discoveries in the archive and in published primary sources pushed subsequent scholarship in specific directions, especially as scholars engaged in interpretive debates and fashioned synthetic narratives of subjects like the coming of the U.S. Civil War. Of course, new scholars always sought to define their work through additional archival discoveries, so I wouldn’t want to make the process sound too deterministic. But I’d suggest that a good part of the value of digitized databases of texts is how they have brought to the fore primary sources otherwise lurking in the background and likely to be neglected because they were not manifestly relevant to key figures or events addressed earlier in the field’s evolution.

This has definitely been relevant to my own work. When I started my dissertation research on the United States’ Reconstruction, I did so largely with hard-copies of mid-19th-century primary sources, especially political newspapers, and microfilm.  By the time I was finishing and then revising for book publication, the volume of sources that were online and word-searchable was truly stunning. Indeed, there may be methodological nationalism cooked into some of these databases since, at least for U.S. history, they reflect in part a sense that the Library of Congress and other organizations have a mission to make American history available to Americans. Then there is the tremendous market for research tools in the U.S. university system, which no doubt biases the digitization process towards American sources. At any rate, sites like the Library of Congress’s Chronicling America are astounding in their volume of content and ease of use. Perhaps the best example of this from my own work is my recovering of the post-Civil War career of Paul Du Chaillu, a controversial explorer of Africa who made himself famous by speaking across the United States about gorillas, cannibalism, and the fate of “the negro.” Despite the fact that Du Chaillu did this exactly as the Republican Party was overseeing the expansion of African American civil and political rights, and exactly as Americans were having an acrimonious debate over the relationship between race and national identity, the scholarship on America’s Reconstruction more or less ignored Du Chaillu. Browsing bound volumes of newspapers led me to be interested in his career, but I could not have rendered the textured assessment of it that appears in my Between Freedom and Progress without the ability to search for his (fortunately) unique name as well as key terms like “gorilla” and “Ashango-Land.” As I turn to another project—a history of the word “reconstruction” centered on the mid-19th-century United States (and by they way, the term gets borrowed from discussions of the “Reconstruction of Poland”)—I couldn’t imagine doing the project without these resources.

My second comment is that there seems to be real hesitancy on the part of historians of the United States to embrace advanced digital methodologies like topic modeling. My impression, from knowing a few scholars who work closely with these methods, is that they experience outright hostility towards their work in some cases. As someone who in an earlier life worked in statistics, I find this hostility to be somewhat perplexing. If some specific methodological innovations are recent, the underlying capacity for elaborate computational work has been around for some time, and has been driving fields like economics for well over a generation. It is perhaps only among historians, or perhaps only historians of the United States and a few adjacent fields, that such methods are often deemed to belong to other disciplines. I have sometimes wondered whether, for historians of the United States, digital methods aren’t the new cliometrics—a sophisticated set of tools that promise to be transformative but end up compartmentalized in a narrow subfield. I hope not.

Best wishes,

Dave Prior