DH vs. DH, and Moretti’s War

With “The Differences between Digital History and Digital Humanities,” historian Stephen Robertson, who became the director of the Center for History and New Media last year, added to the slew of responses to the recent differences special issue “In the Shadows of the Digital Humanities,” which included deep criticisms as well as staunch defenses of the digital humanities, and Adam Kirsch’s troll-piece, “Technology is Taking Over English Departments: The False Promise of the Digital Humanities.” In similar terms to suggestions by Ryan Cordell (“On Ignoring Encoding“) and others that Kirsch (and Fish and Marche before him) unduly conflates digital humanities with digital applications in literary studies, especially those in English literature, Robertson suggests that we separate out what makes digital history distinct from digital literature, further suggesting that “DH” is (and ought to be) digital literature, and digital history is something separate from DH altogether. Should we have two different DH-es? One for history, and one for a humanities that is defined as primarily literature?

n.b. I intend this post to be a broad response to the recent controversies, and not just a critique of Robertson’s piece. I wholeheartedly endorse his basic goal of enhancing our understanding of what is digital history. That said, I think it is a bad idea to isolate digital history from the rest of DH. Here’s why…

1. DH vs. DH, or Why History should not be an Isolationist Discipline

Robertson writes:

My first reaction was a desire to highlight how the absence of digital history produces an impoverished vision of what digital humanities is. On reflection, in the US at this particular moment, fighting to get inside the big tent of digital humanities does not seem the most productive response. Instead, more can be gained from taking the debate as an opportunity to emphasize what makes digital history different from digital literary studies and ‘dh.’

Personally, I wish he had followed his instinct on this one, but instead we have the hint of a controversy: digital history is not digital humanities. My reactions to the main objective of the piece, and of the conference it advertises, are as follows.

Historians don’t have to fight to get into anything; Robertson’s predecessors at the RRCHNM, Roy Rosenzweig and Dan Cohen, along with other historians such as Peter Bol and Timothy Hitchcock, have ensured that DH includes history. They have not been shunned by the humanities–far from it. In Spring of 2013, Cohen held an Avenali Departmental Residency Fellowship at the University of California, Berkeley, a prestigious award granted by that school’s Townsend Center for the Humanities. That fall, Hitchcock was invited as the Lansdowne Lecturer in Digital Humanities at the University of Victoria. These are not isolated examples. Historians are in the DH tent already. If there is any dissonance between DH and history, the problem is instead that the digital has taken longer to get into the tent of history. But that’s an issue for another time.

If the recent media controversies about the meaning and value of DH have excluded history from anywhere, it is the battlefield over distant reading. Interestingly, distant reading easily appeals to historians, so why have digital historians not come under fire? The answer is that the fight is a turf war over methods in the study of literature. That historians should want no part in this is true. However, this is no reason to cordon ourselves off from DH at large. For my explanation of this, see part two, below. But first, let’s take a look at Robertson’s argument. Why does he think we should separate history from the rest?

Robertson quotes Tom Scheinfeldt:

Calling our work “digital humanities” has made it more difficult for us to make it understandable and creditable in disciplinary context: the unified interdisciplinary message may be useful with funding agencies or the Dean of Arts and Sciences, but it may be less so with one’s departmental colleagues.

Be that as it may, we all know that we should describe our projects to different audiences in different terms. Moreover, if historians are less interested in hearing about digital humanities, perhaps it is the ‘digital’ that has turned them away. When you are explaining your GIS query to a historian who doesn’t understand or appreciate digital methods, it really isn’t going to matter what H you put in DH. The “D” is the disconnect.

In any case, the disconnect between historians and DH is given as a reason to explore the differences between digital history and “‘dh’/digital literary studies” (as Robertson divides things). Two main differences are suggested.

First:

The collection, presentation, and dissemination of material online is a more central part of digital history.

A quick look at the list of projects which encoding standards maintained by the Text Encoding Initiative (TEI) shows that the collection, preservation, and dissemination of material is not endemic to digital history alone. The Perseus Digital Library and the Women Writers Project are typical examples, and in terms of the digital skills and tools that such projects require — XML, SQL, Saxon, html, css, etc. — they are no different from the Old Bailey Online, directed by a group of historians.

Second:

In regards to digital analysis, digital history has seen more work in the area of digital mapping than has digital literary studies, where text mining and topic modeling are the predominant practices.

To me, this seems like splitting hairs. Different disciplines in history could be divided up by those who have more work in digital mapping and those who don’t. More importantly, this supposed difference obscures the fact that beneath the surface of the technologies, under the hood of digital mapping, text mining, and topic modeling, are a collection of computing languages and competencies shared by all: XML can be used for for mapping (KML) and text mining (TEI); understanding databases will help you on all counts; query languages such as SQL are used in all cases. Beneath superficial differences, a big set of underlying technologies are shared by all.

Robertson explains why historians haven’t been using text mining and topic modeling as much as their counterparts in other fields:

Part of the explanation for why more historians have not undertaken text mining and topic modeling projects lies in the limited availability of machine readable texts: historians more often rely on unpublished sources than literary scholars, and on handwritten records that it is not yet possible to effectively use OCR to transform into machine readable text.

Some historians spend their lives rooting around in the archives for handwritten notes. Others, especially those whose periods are further removed from our own, rely heavily on published sources, many of which are available online. The Perseus Digital Library is helpful to ancient historians as well as to classicists of a more literary persuasion; the same is true in my own field, early Chinese history, where we rely heavily on electronic sources such as the CHinese ANcient Texts CHANT database (paywall).

If fields like nineteenth-century English literature enjoy the lion’s share of digital editions, this is not an accident or a gift of the gods. It is because, at least in part, scholars saw the value in building these resources, and proceeded to do so, creating institutions like NINES. In the process, they developed technologies and standards that are useful in all disciplines.

Moreover, and in any case, few DH projects like the ones I’ve cited here would present themselves as belonging to any single discipline. Linguists and legal theorists might find use in the Old Bailey Archives; Sinologists of all kinds use CHANT; cultural historians and anthropologists can benefit from the Women Writers Project; these are just a few examples.

I’ve gone to great lengths to respond to Robertson’s piece because I think it would be a mistake to try to divide up the digital humanities or for historians to withdraw from the DH accord. At best, the whole exercise misses the point that the digital technologies being used are often the same, whether one is making maps or mining texts. At worst, it threatens to evoke the animosity of scholars in other humanities fields, especially in languages and literatures, who already feel that they are being attacked by a host of disciplinary, political, and economic forces. The last thing we should to do as historians would be to toss our colleagues from literary studies to the sharks.

That said, Robertson makes some good points too, and I hope these will be the springboard for the conference this November.

As Tom Scheinfeldt has pointed out, these practices do not originate in humanities computing, the dominant origin story for digital humanities, but in oral history, folklore studies, radical history and public history. I’m not saying that the presentation of material online is not part of digital literary studies: electronic scholarly editions and manuscript collections such as The Shelley-Godwin Archive are longstanding parts of that field, but, as the current debate indicates, at present they are not its predominant focus.

This is an excellent point, and historians would be right to remember that DH tools and techniques don’t spring uni-lineally from humanities computing. But why should this necessitate drawing a firm line between history and DH? In another context, I was once told by a historian skeptical of DH that “historians are not librarians.” Dan Cohen notwithstanding, this is basically true. But for us who use digital resources, tools, and skills, we should remember that many are developed in libraries and maintained by librarians. It is through interdisciplinary engagement that they come to us. Putting up disciplinary walls would stymie potentially fruitful interactions.

To return to the issue of sources:

Historians more often rely on unpublished sources than literary scholars, and on handwritten records that it is not yet possible to effectively use OCR to transform into machine readable text. And a wealth of printed, published texts have been digitized, only to be gated, available only to the small number of institutions that can afford the subscriptions, and even then without the API needed for computational analysis.

These are problems, but the solution is not to jettison history from the humanities. On the one hand, literary scholars also deal with manuscripts, and all such researchers would benefit from working with computer scientists to take advantage of developments in OCR, crowd-sourcing, and related technologies. On the other hand, if our sources aren’t digitized, the solution would be to get them digitized. In no small amount of cases, I imagine that this could be done more easily through cooperation with people in other fields of DH. As for paywalls and APIs, these are most certainly problems that would be better addressed by a broad coalition of scholars from many disciplines, as the economic forces behind the paywalls are strong, to say the least.

In all cases, I think pulling history from DH is a big mistake, counterproductive at best. Now, what about all of the recent attention to DH? Why does it seem as if history were absent?

2. Moretti’s War

Robertson’s piece comes in the wake of another blow-up about DH which views it as a threat to the health of literary studies. Let’s step back a moment and ask why this criticism is so persistent in the first place. The most corrosive vitriol splashed at the digital humanities by the likes of Fish, Marsh, and Kirsch is — I believe — a distillation of sour grapes left-over from the early 1980s, not from humanities computing circles, but from theoretical controversies in literature departments. This is Moretti’s War. Before Graphs, Maps, and Trees made a splash in the discussion of Digital Humanities, Franco Moretti lamented in Signs Taken for Wonders (1983) what he saw as the abuses of certain theoretical trends, the “lexicogrammatical euphoria of the last few years” (24), and he was looking for a way to put literary studies on a quantitative footing, to make it more scientific and less new-agey. His answer at the time was to suggest reviving genre studies. But the battle didn’t stop there.

In “Conjectures on World Literature” (2000), Moretti set his famous concept ‘Distant Reading’ in unabashedly polemical terms, attacking the method most cherished in literature departments–not just excessive theory, but close reading itself was now a problem:

The United States is the country of close reading, so I don’t expect this idea to be particularly popular. But the trouble with close reading (in all of its incarnations, from the new criticism to deconstruction) is that it necessarily depends on an extremely small canon. This may have become an unconscious and invisible premise by now, but it is an iron one nonetheless: you invest so much in individual texts only if you think that very few of them really matter. Otherwise, it doesn’t make sense. And if you want to look beyond the canon (and of course, world literature will do so: it would be absurd if it didn’t!) close reading will not do it. It’s not designed to do it, it’s designed to do the opposite. At bottom, it’s a theological exercise—very solemn treatment of very few texts taken very seriously—whereas what we really need is a little pact with the devil: we know how to read texts, now let’s learn how not to read them.

Now, a charitable interpretation of Moretti’s article might say that he laments those who practice close reading at the expense of all other methods. Would that it were merely that, but I understand it to be more. His target is the predominance of national narratives in comparative literature (presumably with implications for other fields), a situation for which he blames close reading. He goes on to conclude:

The point is that there is no other justification for the study of world literature (and for the existence of departments of comparative literature) but this: to be a thorn in the side, a permanent intellectual challenge to national literatures—especially the local literature. If comparative literature is not this, it’s nothing.

This is achieved, apparently, through distant reading.

So Moretti suggests that close reading as a method of comparative literature reinforces the national and local literatures and undermines the very basis of the field’s existence. I think he is barking up the wrong tree. If the problem is the predominance of national canons at the expense of world literature, the cause is not close reading, which as a research method cannot alone result in the selection of some part of ‘world literature’ at the expense of another. Rather, I think he should be looking at the social, political, and cultural forces which result in the existence of a canon in the first place. I am reminded of a Comp Lit conference at NYU in 2008 entitled “Age of Comparison?” It was hosted by Zhang Xudong, a prolific translator of western literary and social theory into Chinese, and many of the participants were studying non-Western subjects. Some participants were surprised (and noted their relief) at the relative absence of French and German literature from the panels, and yet G.W.F. Hegel still loomed large in the discussions, such that one participant wondered, exasperated, “why is it that when we look back, we always see Hegel staring back at us?” The problem here was very much the sort of thing that Moretti would like to challenge, but it had nothing to do with the method of close reading.

It’s tempting to go into a deeper critique of the article on other points. But it order to stay on topic, I will say only that I think Dipesh Chakrabarty’s Provincializing Europe (2000; see especially chapters two and three) is more powerful than “distant reading” as a critique of those Eurocentric comparisons or analyses which take as normative the western European historical experience.

In the end, by promoting distant reading as a superior alternative to close reading, Moretti distracts from his goal here (i.e. to critique nationalism in Comp Lit). Meanwhile, he manages to alienate some scholars of literature who might otherwise be sympathetic to the new methods he advocates elsewhere. And this article also fails here to acknowledge the contributions to digital methods by literary scholars interested in canonical texts and close reading, from at least as early as the 1980s. For example, studies of authorship (in the canon) and meter (the closest of readings) have been common in the journal Literary and Linguistic Computing (established in 1986), and scholars doing this kind of work, such as David Hoover at NYU, have been seeking to put the study of literature on more quantitative grounds–similar to what Moretti was advocating in 1983. But he misconstrues the problem by posing it in needlessly divisive terms of distant versus close reading. The result is that he (but not only he) manages to make DH into something that many see as the dragon threatening to destroy the humanities. Thus Stanley Fish sullenly concludes that “whatever vision of the digital humanities is proclaimed, it will have little place for the likes of me and for the kind of criticism I practice” (NYT Jan. 23, 2012). For those whose scholarly authority (not to mention livelihood) rests on close reading, the misleading dichotomy that Moretti has created implies to them that DH something to be slain, lest they be destroyed by it.

So, in short, Moretti’s distaste with the excesses of literary theory and national literatures saturated his advocacy of new methods. Because he had phrased this advocacy in polemic terms, a Morettian take on the digital humanities is one where DH is in direct conflict with the approaches to literature that dominated the late 20th century literary departments, linguistic theories and close reading. As a result, so many criticisms on DH are in essence, or at least in part, attacks on distant reading. They conflate a part of DH for the whole, and they see DH as a distinct threat to now traditional methods of literary study, but they didn’t fire the first shot. This is Moretti’s War, and DH got dragged into it.

Conclusions

I am convinced that there is no conflict between close reading and digital tools (even when used for distant reading). In general, the best approach would be a combination of all available methods of research and discovery; in particular cases, the approach should be selected based upon the nature of the questions and sources involved. I think Moretti’s polemic stance against close reading is a problem because it drives a wedge between two methods that should work very well together, and for no good reason.

What does all this have to do with Robertson? Perhaps it is natural that historians find themselves out of place on the the battlefields of Moretti’s War. But putting forth the suggestion that digital history is not digital humanities just expands the polemic war to a new front. But I think, given that the tools and skills of DH are still quite new, and that many of the technologies in all branches of digital humanities and social science research are shared, to divide up the territories of DH would serve none of us well. What’s worse, we might end up squandering the electricity of the moment, when institutions like THATCamp and DHSI are showing just how much we have to gain from cooperation and collaboration.

Having said all of this, I agree that we should ask what all of these technologies and tools have to offer history as a discipline. But let’s not define digital history negatively, as ‘not DH’ or ‘not literary studies.’ It should suffice to ask: what do historians do, and how will these digital things help us do it better?

What should we do about Moretti’s War? Let’s just agree that there is no inherent conflict between DH methods and other methods and that the needs of individual research projects should determine what methods ought and ought not be used, and then we can get on with the work of scholarship.