What Has the Digital Meant to American Periodicals Scholarship?

Note: this is a peer reviewed, uncopyedited post-print of an article that appears in American Periodicals 26.1 (2016), which is now available at Project Muse. It is part of a phenomenal forum on Digital Approaches to Periodical Studies that includes essential pieces (in order of appearance) by Elizabeth Hopwood, Benjamin Fagan, Kim Gallon, Jeffrey Drouin, and Amanda Gailey.

What has digitization meant for periodical studies, and what might it mean in the future? We should first consider how the digital archive changes notions of access, both political and practical. James Mussell notes that “the conditions that permitted newspapers and periodicals” to become the central medium of discourse in the nineteenth century—“their seriality, abundance, ephemerality, diversity, heterogeneity—posed problems for those who wanted to access their contents” in print forms.[1] The periodicals archive is vast and largely unindexed. In ways so basic and fully transformative that we easily overlook them, digitization and its attendant technology, keyword search, have already changed periodicals scholarship entirely, allowing researchers to easily identify topics of interest across swathes of newspapers, magazines, and related materials, and to just as easily incorporate those media as evidence for historical, literary, or other claims. As Ted Underwood reminds us, “[a]lgorithmic mining of large electronic databases has been quietly central to the humanities for two decades. We call this practice ‘search,’ but ‘search’ is a deceptively modest name for a complex technology that has come to play an evidentiary role in scholarship.”[2] Though other forms of computational analysis will certainly influence periodicals research in the near future, the most dramatic methodological shift has already happened.

Patrick Leary wrote more than a decade ago in “Googling the Victorians” that “[f]ortuitous electronic connections, and the information that circulates through them, are emerging as hallmarks of humanities scholarship in the digital age.”[3] Given the diversity of material available in mass digital archives such fortuitous connections are more likely to be drawn across media. Even Google Books, for instance, includes a huge array of eighteenth- and nineteenth-century periodicals that were bound by libraries, bringing periodicals into the purview of all its users, whether they go there seeking periodicals or not. Such access, coupled with more academically-oriented databases of historical newspapers and magazines, have meant that more scholars can and do cite periodicals today than did a few decades ago. More recently, Leary noted that “[s]earching the newspaper collections for references to the people, ideas, publications, institutions, or events one is writing about is quickly becoming not merely an option but a necessity” even for more “casual” users of digital archives.[4] We might claim digitization has rendered periodicals more tractable to humanistic argumentation, broadly conceived, than they once were.

We might also contend that the ubiquity of digital periodicals research has not been unequivocally positive. As Leary, Bob Nicholson, and many others have pointed out, access to digital periodicals archives is unevenly distributed, with the largest collections sold by commercial providers beyond the means of smaller institutions and independent scholars—and often closed to computational analysis beyond search even for scholars at subscribing institutions. Large-scale public collections, such as the Library of Congress’ Chronicling America newspaper archive, have brought welcome attention to certain varieties of periodicals. For example, Chronicling America’s guidelines for state digitization programs ensure that the digitized papers are spread geographically across each participating state. Such policies bring rural and small-city papers more often into scholars’ fields of vision, mitigating the dominance of urban papers in earlier periodicals scholarship.

On the other hand, such collections risk tipping the scales of our attention in equally misleading ways. In his analysis of English-language Canadian dissertations, for instance, Ian Milligan thoroughly demonstrates first that dissertations written in the “database age” draw more on newspaper content than those written before, but secondly—and perhaps more importantly—that they draw ever more disproportionately on particular newspapers: “[n]ewspapers readily accessible online are being used more frequently. They are also being used in a more sustained manner.”[5] While it’s perhaps unsurprising that digitized newspapers are cited more frequently than undigitized papers, Milligan demonstrates in precise, quantitative detail how quickly and dramatically the newspapers used in dissertations have skewed toward what is available online. Even as digitization has amplified the practical scope of historical periodicals research, allowing researchers to draw materials from across a variety of periodicals sources, it has paradoxically narrowed the potential scope for such periodicals research for most scholars.

This narrowing effect is most troublesome when we consider that public digital collections include far fewer publications from immigrant and minority communities than from middle- and upper-class white communities, and fewer even than digital archives offered by commercial publishers such as Gale, Readex, and EBSCO. What Lauren Klein writes of the digitized archive more broadly certainly pertains to periodicals: “in spite of what Ed Folsom…has extolled as the ‘epic transformation’ of the archive, characterized not only by increased access to content, but also by the proliferation of paths through that content that are facilitated by the digital archive’s underlying database structure, the issue of archival silence—or gaps in the archival record—remains difficult to address.[6] Most digital archives were built almost exclusively from microfilm collections, which can be scanned much more quickly than paper, allowing a given project to generate significantly more content from a given grant or library budget item. While eminently understandable from a practical point of view, such decisions reinscribe acts of valuation, selection, and preservation made by previous generations of scholars, who indirectly dictate what is digitized and thus what is studied. The archive—whether manuscript, print, or digital—has never been objective or neutral, but we must attend to the unique ways in which database technologies insinuate false ideas of completeness. These limitations remind us that we cannot consider the digital archive sufficient; we must both advocate for increased digitization of those periodicals which are not represented in current collections and make the best use possible of physical periodicals collections in our scholarship and teaching.

Perhaps, too, we need to more carefully distinguish our objects of study, theorizing the digital periodicals archive in media-specific terms. Ellen Gruber Garvey traces the central metaphor of modern newspaper databases to nineteenth-century “‘clipping services’ idea that newspapers were a vehicle for data that could be isolated and re-sorted according to multiple criteria.” Likewise digital “databases disassemble the newspaper via keyword searches and offer a pileup of hits to articles on those topics…rather than slicing through what others were reading in the newspaper at that date.”[7] Garvey shows that such disambiguation is neither ideal for periodicals research nor, paradoxically, particularly novel to the digital age, except perhaps in the scale of information available to any individual researcher. Certainly, drawing significant connections across databases of “superabundant information” poses an under-examined epistemological and methodological challenge for periodicals scholars. Maurice Lee has cautioned scholars about New Historicist readings drawn from large-scale digital archives, as “what seems like a meaningful correlation can result from sheer quantity and chance, particularly under conditions of mass information in which seemingly unlikely coincidences are highly probable and easily targeted by bias-confirming searches.” Lee allows that “[b]lunt quantitative analysis may help to contextualize New Historicist anecdotal evidence,” and we might balance worries about the coincidences inherent to digital search with opposite worries about the coincidences of physical archival research.[8] Can we draw a significant difference between those connections we dismiss as “chance” and what we laud as “serendipity?” Stephen Ramsay advocates a “Screwmeneutical Imperative” within the digital archive, an embrace of “browsing” as “one of the most venerable techniques in the life of the mind”: “There are so many books. There is so little time. Your ethical obligation is neither to read them all nor to pretend that you have read them all, but to understand each path through the vast archive as an important moment in the world’s duration—as an invitation to community, relationship, and play.”[9] If there are “so many books,” there are even more periodicals—and perhaps even more fruitful paths through them for scholars who can exploit the algorithmic affordances of the digitized archive.

Considered only as a surrogate for the physical newspaper or magazine, the digitized periodical can only disappoint. The digitized periodical, like the digitized book, constrains its original to the size of the computer screen, smooths its textures, even plasticizes its aroma. The most compelling digital projects are those that do not attempt to replicate the the reading room, but instead take remediation seriously, drawing on the unique affordances of the digitized text to trace patterns across the archive and even highlight its limitations. In the past decade we have seen a surge in projects that employ advanced computational techniques to track topics across Richmond’s Daily Dispatch during the Civil War,[10], assess and map language patterns in historical Texas newspapers,[11] trace the evolution and circulation of themes in abolitionist newspapers (and other digital archives),[12] detect poetic content in historical periodicals using image analysis,[13] or track reprinting across nineteenth-century US newspapers.[14] In a recent talk, Klein referred to topic modeling as “a technique that stirs the archive,” rearranging its contents in ways that defamiliarize them to offer new interpretive purchase. Klein’s formulation captures the experimental and exploratory aims of many computational periodicals projects, which employ computers to model ideas and questions across texts that would be difficult to an individual human reader to test.[15] For that reason, much computational work has grappled with systemic questions around information exchange, geography, networks, seriality, and thematics.

To offer my own brief polemic in closing: when we conceive of the digitized periodical primarily as a surrogate for its analog original, we foster misguided debates about replacement and foreshorten our imagining about what the digital might offer periodicals scholarship. As Alan Liu has shown, narratives of “new media”—whether millennial or apocalyptic—too often stage an “exaggerated encounter between old and new” despite the historical reality, which is more often a “thick, unpredictable zone of contact—more borderland than border line” between media. “The trick” for scholars, Liu argues, “is to play the ‘old’ and ‘new,’ ‘codex’ and ‘digital,’ and ‘literary’ and ‘informational’ off each other in ways that thwart any facile modernization narrative and foster surprising recognitions about the scholarly and cultural potential of new media.”[16] The digitized newspaper or magazine is a new edition of its source text—its type newly “set” in text or XML files—with new limitations, yes, but also new affordances. The best digital periodicals scholarship moves between scales and media, demonstrating how observations at the macro level, drawn from digitized corpora, align (or do not) with observations at the micro level, drawn from archival texts. Such work pairs the unique capacities of the computer with the interpretive talents of the human reader, to the benefit of both.

While computational work in the humanities is hardly new, to date we have seen comparatively few projects that take the digital remediation of periodicals seriously. This should not be surprising; the sheer scale of epistemological and methodological shift induced by digitization has required an extensive period of preparation and examination. Ever-growing digital archives coupled with increased training in digital research methods have placed us at an important horizon. The way forward is neither to eschew the archive in favor of “distant reading” nor to spurn computation as categorically (or necessarily) positivistic. Even digitized periodicals remain voluminous, messy, provocatively difficult. Indeed, it is precisely these qualities which make periodicals compelling to both humanities scholars and computer scientists—who, we should remember, largely are not programmers but scholars seeking complex theoretical and engineering problems amenable to computational exploration. What questions could we ask of the digitized text that we could not ask of the analog, and how might such investigations turn back productively toward the archive? Who outside of our departments or disciplines might likewise be interested in our wonderfully chaotic data? Can we foster computational work in active dialog with other American periodicals research? There is a reason why fields such as book history and archival studies flourish in the digital age. The act of digitization brings the analog more sharply into focus; remediation makes its media perceptible. As arguably the first mass medium, American periodicals are uniquely suited to helping us understand our moment of informational abundance—its opportunities and elisions, its connections and ruptures. Already a technology of aggregation and exchange, historical periodicals can speak forward to the digital even as the digital helps us grapple anew with periodicals’ production, form, and content.