Technological change encourages ideas of supersession of the sort exemplified by Victor Hugo’s Archdeacon waving a book at the cathedral of Notre-Dame and lamenting “ceci tuera cela.”^{Footnote 1} The gun-designer Samuel Colt refined the notion when he argued that the telegraph would “supersede […] commercial newspapers” by separating paper from the news, thereby suggesting that as part of supersession you could target the bad yet retain the good in a simple, binary division.^{Footnote 2} From here we have confidently stepped to notions of books without binding and libraries without walls. Digital devices, it is assumed, allow us to remove problematic material constraints while holding on to the essential informational resources.

From its quasi-oxymoronic title onward, Digital Paper, Andrew Abbott’s “manual for research and writing with library and Internet materials,” resists such separations. Research may have “gone digital,” yet “paper” can still provide valuable conceptual as well as physical structuring. Hence, rather than making academic tasks lighter, the “new tools,” Abbott claims, “make it harder than ever for students to learn the disciplines of research” because, while dramatically increasing problems of information “overload,” such tools undermine the structures that previously bore much of that weight.^{Footnote 3}

Abbott comes to this discussion with some authority. His essay “Reflections on the Future of Sociology” remains an early but useful caution against current enthusiasms for “big data” in sociological research, while The System of Professions used as a case study the relationship between information technologists and librarians.^{Footnote 4} Abbott might appear likely to embrace the latter and shun the former, yet Digital Paper challenges technologists less for shunning librarians’ material practices than for naively embracing their idealist view of information. Scholars, he argues, abandoned libraries long before modern information technology [IT] entered because librarians increasingly failed to provide apt support for academic research. Despite interjurisdictional fights, Abbott holds, IT professionals and librarians have long shared a blindness to the realities of scholarship. Hence in Abbot’s view “there is nothing very new about the digital library” [16] and the “Google project […] is very old news” [35]. While “[s]cholars desert[ing] libraries” [24] has a long history, now, Abbott maintains, they should desert, or at least demand much more of, digital tools.

These information professions, Abbott suggests, share democratic, universal, and nonhierarchical ideals. It takes as much bravery to speak out against democracy and universalism as it does to denounce digital tools, but Abbott unashamedly does both. Invoking another cathedral image may help explain why. To emphasize the egalitarian character of Open Source software, the software developer Eric Raymond invoked the notion of a bazaar, in contrast to the hierarchy of the cathedral.^{Footnote 5} Abbott, by contrast, favors cathedral-like structures for scholarship, arguing that the hierarchical “jurisdictions” of the disciplines provide distinct, structuring resources essential to producing their “view from somewhere” and that the world of print taps into “longstanding systems” [61] to support such structuration. New tools, by contrast, offer an idea of “knowledge from nowhere,” [83] as Abbott puts it, giving “[n]one of us […] a solid way to judge authority or quality online” [61]. Here Abbott reveals robust faith in the ranking of academic print publishers (modestly putting his own second from the top) and of peer review (perhaps unsurprising for the editor of the American Journal of Sociology).

Surveying digital resources, Abbott appears more discouraged by prior work absorbed into the digital world than by work born digital. In his view, in the process of digitization, much indicative substructure is superseded by such things as inept scanning, fallible character-recognition, and obtuse keyword searches all presented to us by capricious user-interfaces and supported by problematic business models. The Net is never slow to support the last point. Abbott cautions that “[t]o understand the problems with newer and online reference tools, it’s helpful to examine the new economics of libraries” [73] while elsewhere he cautiously applauds the digital version of the American Library Association’s venerable Guide to Reference Books [72]. A visit to the Guide’s website since Abbott wrote, however, finds that “the current business climate makes the continuation of the Guide impractical.”^{Footnote 6}

Abbott does not limit his criticism to libraries and digital tools. As it attempts to highlight the tortuous, nonlinear practices of research, Digital Paper indicts scholars for leaving these hidden behind the linear products we make public, arguing that “[t]he finished logic of our articles and books is a façade, put on after the fact” [xii]. To counter this deception, he uses examples from his own work, bravely laying bare the enduring but usually unacknowledged inner reaches of academic practice. In this way, Digital Paper is a reflexive work, though occasionally it can feel more recursive than reflexive, as when, for example Abbott exemplifies the challenge of research in libraries with a account of his own research on libraries. The detailed explications of implicit practices, furthermore, sometimes give the feeling of incautious reification, classifying what I suspect are relatively seamless shifts in research practices into distinct and dauntingly separate categories.

This reflexive writing in turn encourages reflexive reading (I should confess that I learned and inwardly blushed as I read), and the book closes by invoking Kant, who has influenced much of Abbott’s sociology, and the “categorical imperative,” with which Abbott hopes to encourage researchers to “Do your research and write your text in the frame of mind in which you yourself would want to be researched and discussed” [246]. Given how easy it is to train students as attack dogs, this is an important and generous caveat that endorses Abbott’s evident faith in both the practices and the discursive products of scholarship, however much in his eyes the latter misleadingly hide the former.

As digitally driven ideas of supersession challenge conventional notions of scholarship, we may need to cling to that faith. I recently attended an enthusiastic discussion of moocs (Massive Open Online Courses) which celebrated the ease of machine-grading hundreds of thousands of student assignments. “But what,” one person asked, “about those disciplines whose discursive products are not amenable to machine grading?” “Oh,” a confident voice replied, “we won’t need those any more.” Abbott’s work persuades us that we will and offers valuable ideas about how to produce them. Yet focused on the bizarre bazaars readily found in the digital world, Abbott leaves us trapped by another binary as sharp as the Archdeacon’s or Colt’s, for he provides little sense of whether there might be alternatives for surveying the material practice of scholarship to standing on the gothic parapets of academic cathedrals.