When Max Weber suggested in 1917 that the world had been disenchanted, he meant that modernity was best understood by the expansion of “technical means” that controlled “all things through calculation.” The real power of these technical means lay not in the techniques and technologies themselves but in the disposition of those who used them, in their unshakable confidence that there were in principle “no mysterious, incalculable forces” they could not calculate and control. Such a technical rationality had replaced the “magical means” premodern people had used to placate gods and spirits. In Weber’s account, which was both elegiac and supercilious, when the “technical” superseded the “magical,” wonder disappeared from the world. The confident, calculating scientist, the intellectual hero of the modern world, was incapable of “wonder” and inured to “revelation.” Nothing surprised him, and nothing could be revealed to him.
Having conquered everything else, the calculating machines of modernity are now coming for our books. Or at least that’s what anxious writers in the New Yorker, the Los Angeles Review of Books, and the New Republic have suggested as they warn of the cultural collapse being ushered in by the digital humanities. These critics rarely discuss what most scholars do with their digital tools––marking, annotating, visualizing, and collecting texts as our literary archive gradually moves from print to digital form. They focus, instead, on the grandiose pronouncements of Franco Moretti, a professor of literature at Stanford University and founder of Stanford’s Literary Lab. “The trouble with close reading,” Moretti claims, “is that it necessarily depends on an extremely small canon.… At bottom, it’s a theological exercise––very solemn treatment of very few texts taken very seriously.” In place of “close” reading, Moretti proposes a “distant” reading, in which gradually emergent and long-term patterns in literary history are studied through the application of computational and quantitative methods to the analysis of massive numbers of texts. To critics of the digital humanities, Moretti has come to represent all humanities scholars who use a range of computational and quantitative methods to model plot structures in novels, analyze literary periods, map metaphors, track lexical changes, and, yes, read texts.
For instance, writing in the Los Angeles Review of Books, the novelist Stephen Marche argues that these new computational ways of reading are not the incidental quirks of a few misguided English professors. Rather, they are symptoms of a larger cultural tragedy that began when the Google Book Project and the Hathi Trust started to digitize millions of printed books in the early 2000s. The data-fication of books represents a cultural shift not only in what counts as a book but in what counts as reading. Lamenting the leveling effect of digitization on literature, Marche claims that turning books into data treats all literature “as if it were the same. The algorithmic analysis of novels and of newspaper articles is necessarily at the limit of reductivism. The process of turning literature into data removes distinction itself. It removes taste. It removes all refinement from criticism.”
In their opposition to machine reading, Marche and his fellow critics join the melancholy moderns who, in similar fashion, bemoaned the loss of coherent and fully integrated forms of life. To Friedrich Nietzsche’s last man, Max Weber’s disenchantment, and Hans-Georg Gadamer’s lament for a lost Lebenswelt (“world of lived experience”) we can add the loss of “literature” and the reduction of reading to a rationalized, technically determined process bereft of meaning.
Just as Weber’s elegy for a lost, magical world presupposed a specific form of knowledge, so too does criticism of “distant” reading presuppose a particular form of reading. And just as Weber’s disenchanted modernity needed its enchanted premodernity, so does Marche’s distant reading need its close reading. But what is so sacred, so solemn about reading a few books so intensively? And if close reading is, to quote Moretti, a “theological exercise,” what kind of exercise is “distant” reading?
Judging from the jeremiads against Moretti and his colleagues, “distant reading” is a profane, disenchanted exercise, a technological intrusion into an ethical practice. When we read, our eyes should move line by beloved line, page by precious page. Such immersive, personal reading makes possible emotional and intellectual experiences of recognition that transform us. Distant reading treats books as though they were elements in the regular, law-governed order of nature––particles to be calculated and measured.
On the other side of the debate, arguing for distant reading, we have scholars such as McGill University’s Andrew Piper. To read “topologically,” as he terms it, is not to begin a personal transformation but to discover patterns and scrutinize relationships among not several but dozens, hundreds, or even thousands of books. Readers in a modern close-reading tradition read syntactically, sentence to sentence, and regard words and sentence as authoritative “keys” with the potential to transform readers themselves from a state of humble incomprehension and distanced curiosity to one of privileged clarity and critical insight.
When Piper reads topologically, by contrast, he uses computational methods to map relationships among multiple elements (such as lexemes, morphemes, and phonemes) and categories (genre, format, publication information) of multiple texts. Reading, in his account, is less an exercise in fixing meaning (x means y) than in discovering the ratios that constitute texts and bind them together. Topological reading eschews traditional reading’s focus on the sentence and embraces, instead, the lattice-like structure of language itself. Instead of attempting to provide a lexical meaning of “love” in Goethe’s The Sorrows of Young Werther, Piper says simply that “love” is “equivalent to 0.00109 (the percentage of times it appears relative to all of the words in the novel) compared to 0.00065 in Faust.” Reading in this way undoes the attachment to individual books and the expectation that they will change readers in a particular way. Piper is looking for patterns, not a better self. Reading with numbers, he writes, “privileges the latency of the manifest … all of those words that have historically resisted our attention through their over-familiarization, their presence, and their over-availability.” Computational reading reveals a “lexical unconscious,” and every new graph or diagram constitutes a distinct “totality,” a different way of seeing the whole of literature.
So for its critics, “distant” reading is a desecration because it does not treat individual books as precious objects worthy of the devotional practice that reading is. But where did this notion of reading as transformative, even sacramental, come from? And is “distant” reading such a radical departure from it?