Paul Fyfe responds to Kate Holterhoff

Holterhoff underscores the challenges for books like Reductive Reading which seek to make contributions in each domain of “digital humanities and literary scholarship.” Holterhoff notes the lack of quantitative results for the stylistic signatures of Dickens, Eliot, and Barrett Browning. Indeed, Reductive Reading downplays its own statistical methods. Its most reaching computational claim—that speech tags mostly occupy the middle of sentences at the start of the nineteenth century and but get entirely mixed by the century’s turn—sees only the light of a few pages at the end of the book (124). Nor does the book publish data for follow-on analyses or replication.[1] But, as Holterhoff concludes, these are not its primary goals. While admitting some early “doubt,” Holterhoff “come[s] to believe” this research will matter—as if the book, true to its central argument, persuades over its sustained experience rather than by its moralizing / most direct claims. Modeled in its arguments about how stylistic units aggregate into theories of ethics and fictional personhood, Reductive Reading subordinates computational literary studies to arguments shared broadly across literary studies. Indeed, Daniel Wright on Twitter—a reductive medium if there ever was one—forecasted his review of the book for Victorian Studies in praise of the “gorgeous, eye-opening close reading in the service of a brilliant argument about morality, style, and scale.”[2] If well received in that venue, what would a review look like in the journal Digital Scholarship in the Humanities? Can or should such a book possibly do both? And what does that suggest about what Holterhoff calls “the field of digital humanities and literary scholarship?”

[1] Sarah Allison, “Other People’s Data: Humanities Edition.” Journal of Cultural Analytics (Dec 8, 2016).

[2] Daniel Wright (@dannybwright), “Just finished @sarahallison’s incredible book.” Twitter April 6, 2019.

%d bloggers like this: