Rewriting history

By John Hooper

In the centre of Stanford University’s sprawling campus stands an impressive clock tower. Through glass at the base you can admire its complex, early-20th-century mechanism and read the time from a little clock face that replicates the bigger one above. Next to it, someone has thought to place an even smaller, digital clock. So people know the real time.

Stanford, the university that gave birth to Silicon Valley, is probably the world’s least analogue seat of learning. Hewlett-Packard, Google, Cisco Systems, Yahoo, PayPal, Sun Microsystems and Instagram were all founded by its alumni. So it is not surprising that Stanford should also be at the heart of a movement that has swept across campuses in recent years with the speed—and, some would say, the destructiveness—of the forest fires that break out in the hills above the Santa Clara valley.

Digital humanities, still relatively little-known beyond academia, involves using information technology to study the arts and social sciences. Depending on how you define it, the term can encompass anything from building a computer-generated, three-dimensional representation of an ancient Sumerian city to feeding thousands of 19th-century novels through software to count the number of words for moral qualities and then sort them by author, date or whatever criteria you choose to put in.

Visualising the results of the analysis on, say, an interactive chart or map, is crucial to most digital-humanities projects: it can enable researchers to draw at a glance conclusions that would otherwise lie unnoticed in a morass of facts and figures.

“It’s wonderful to have the data at your fingertips, to bounce the facts around in a way that would be impossible without computers,” says Giovanna Ceserani. Over lunch in the Stanford Faculty Club, she produces a laptop to explain what she means. Ceserani is studying the 18th-century Grand Tour. Her database is John Ingamells’s “Dictionary of British and Irish Travellers to Italy, 1701-1800”, a storehouse of information on more than 5,000 men and women who journeyed through Italy in search of education, inspiration or, sometimes, just entertainment. Ceserani has narrowed her gaze to the 69 who became architects. “Even so,” she says, “it is impossible to remember all that is known about all of them.”

Tapping on the keyboard, Ceserani summons up a bar chart. Beside the vertical axis are the names of her subjects. The horizontal axis measures the number of months they spent in Italy. Multi-coloured bars stretching from left to right show where they stayed, each separate colour corresponding to a town or city. One colour predominates: red, for Rome.

That of itself is no surprise. The ruins of the old imperial city were the high point of the Grand Tour. But when Ceserani ran the same exercise for the aristocrats with an interest in architecture, she discovered something that would never have struck her just reading through the entries in Ingamells’s monumental directory. “The amateurs, like Lord Burlington,” she says, “went as much to Venice or Naples as they did to Rome.” The data, when visualised, showed that a long stint in Rome was a kind of litmus test: “It almost defined what being an architect in the 18th century was all about.”

Prompted by what she had seen in the bar chart, Ceserani began looking again at the architects’ rela­tionship with Rome. What she found using conventional research techniques was that “it was not just about looking at great old buildings. Rome was a unique place where you could interact with people from other countries and social classes, sometimes gaining patronage in the process.” And not only that: Ceserani’s non-digital studies showed that the aspiring British and Irish architects were far more engaged with their Italian contemporaries than had been realised. They studied under masters like the great engraver Giovanni Battista Piranesi; they tried to join Italian architectural societies and win Italian architectural competitions. “When the Architects’ Club, England’s first architectural association, was founded in 1791,” Ceserani says, “one of the four requirements was to belong to an Italian academy.”

Her project is an example of the way digital human­ities should work. Prompted by an original view of data that had long since been available, she formed a hypothesis that inspired her to look at the literature in a new light. In this case, the hypothesis is one that challenges the view of the Grand Tour as a jaunt for dilettantes. For architects at least, says Ceserani, “it was where professionalisation took place.”

With its palm trees, courtyards and artificially weathered sandstone masonry topped with red tiles, much of Stanford looks the way a Moorish palace might do in a video game. On the top floor of a building in this distinctive, Beau-Geste-on-the-West-Coast style is the Centre for Spatial and Textual Analysis, or CESTA, the inter-departmental unit that is digital humanities’ most important bridgehead on campus.

Ask Mark Algee-Hewitt, a lecturer from the English faculty, whether CESTA could fairly be described as the epicentre of the worldwide digital humanities movement, and he replies with a smile and a “we like to think so.” Behind him is an array of filing cabinets marked “Chinese Railroad Workers”. More than 10,000 Chinese toiled to build America’s first transcontinental railway, many of them in the employ of Leland Stanford, the founder of the university that bears his name. The mostly illiterate Chinese left behind little by way of diaries or memoirs. But one use of digital humanities can be to fill in the void where no written accounts exist. CESTA’s researchers are mining corporate and other records to build up a picture of the lives of the Chinese who helped to unite America.

In some cases, the work carried out at CESTA merely confirms what might have been assumed. Was it really worth running large numbers of texts through computers to show that the term “United States” was more often followed by a verb in the plural (“the United States are...”) before the civil war than after it?

In other cases, however, digital research can yield evidence that undermines, or even destroys, accepted theories and common assumptions. CESTA’s biggest project is Mapping the Republic of Letters, a catch-all title for a series of studies, including Ceserani’s, that aim to shed light on the internet of the Enlightenment: the network of correspondence that linked intellectuals in the 17th and 18th centuries. “We’re like the NSA,” says Dan Edelstein, a professor of French. “We look at who wrote to whom, when and where.”

An obvious target for this form of surveillance was Voltaire. Among his many contributions to the Enlightenment was his “Lettres Philosophiques”, published in 1734, which he claimed introduced Locke to a French readership. Voltaire had lived in Britain for three years and spoke English. So when Edelstein first looked at the visualisation of his correspondence, “what jumped out at me was that Voltaire wrote so little to Britain. There were only 140 letters out of a total of around 15,000—less than 1%.”

Edelstein went back to examining Voltaire’s correspondence and realised for the first time the relevance of his subject’s attachment to the notion of translatio studii—the idea that the centre of civilisation moves inexorably from place to place. “Essentially, Voltaire thought England had had its moment in the historical sun,” says Edelstein, who based a research paper on his findings.

Caroline Winterer, a professor of history and director of the Stanford Humanities Centre, had a similar eureka moment when she saw a map of Voltaire’s correspondence set alongside that of her main subject, Benjamin Franklin. “With Franklin what you get is a dense transatlantic network,” she says, “whereas with Voltaire there is just one letter that crosses the Atlantic.” That seemed to justify Franklin’s fame as a cosmopolite. But when Winterer dug deeper into the data there was another surprise waiting: “most of his correspondents were either British or American.”

This is a recurrent theme in the Republic of Letters project. The networks of the intellectuals of the Enlightenment were far more restricted than the academics had imagined. Paula Findlen, a professor of Italian history, says the same was true of Galileo. “We think of him today as perhaps the first scientific celebrity. But he lived in a relatively local world until he was forced not to.” It was only after this great polymath came under the menacing gaze of the Inquisition that he reached out for help to non-Italian intellectuals.

“In 1597, Kepler writes to him and sends him one of his books,” Findlen reports. “Galileo responds politely, but no great intellectual camaraderie comes of it. He does not get back to him until 1610, when he has need of him.”

“You’re not plugging in data to get answers,” says Winterer, “but to ask better questions.”And some of those questions remain tantalisingly unanswered.

A pioneer of digital humanities—and one of its most controversial figures—is to be found in a darkened study, piled high with books, not far from CESTA’s headquarters. Franco Moretti, Stanford’s professor of English and Comparative Literature, was drawn into digital humanities by way of cartography. He set out to produce an atlas of the European novel, which was published in Italian in 1997. “I realised that to make good maps you need sets of data,” he says. “That is how I started quantifying.”

One of his projects involved analysing all the passages in 19th-century novels that begin 50 words before a place name and end 50 words after it, with the aim of determining the place’s emotional associations. Analysing the results, Moretti noticed something that he had not set out to find: “The growth of fictional London did not match the growth of real London.” The population exploded in the 19th century, growing from 1m to almost 7m. Yet for the most part British authors continued stubbornly to locate the events in their novels in a London that stretched from “somewhere near the Bank of England to Mayfair”. Yet, says Moretti, the same is not true of French authors writing about Paris. Why not? “Is it perhaps that in London the chic areas have not changed?” he asks with a shrug. “I just don’t know.”

The person credited with inventing digital humanities was yet another Italian. In 1949, Father Roberto Busa, a Jesuit priest who was bent on compiling a concordance of the works of St Thomas Aquinas, persuaded IBM to let him use its punch-card technology, then at the cutting edge. Busa said later that after learning that an in-house report had concluded his project was impossible, he went into a meeting with IBM’s founder, Thomas J. Watson senior, carrying one of the firm’s own posters. It bore the slogan “The difficult we do right away; the impossible takes a little longer”. Watson was won over, but the impossible did indeed take longer. The last of the 56 volumes of Busa’s “Index Thomisticus” was published in 1980.

By then punch-cards had given way to magnetic tape, which had given way to floppy disks. But it was the advent of the world wide web in 1991 that really enabled digital humanities to take off.

For Caroline Winterer, as a university administrator, one of the benefits of digital humanities is that it breaks down barriers between disciplines. These are traditionally formidable at Stanford, where students reading science subjects—techies—refer to their non-scientific fellow-undergraduates as fuzzies. “Humanists have always worked alone, like monks in cells,” Winterer says. “Digital humanities involves publishing together and, before that, thinking together. CESTA crosses the barrier between techies and fuzzies.”

But, for its critics, all this quantifying, this computing and classifying, not only runs counter to the spirit of disciplines that deal in shades of meaning, but could give students the idea that they no longer need to read, see and listen.

Caroline Winterer says she understands the concerns. “Some of the values that we try to cultivate are taste, discrimination and aesthetic appreciation, and those are not obvious targets for digital humanities. There is no substitute for the kind of deep knowledge that only long-term immersion can bring. Digital humanities can enhance that, but not replace it. People still need to read books, look at pictures and listen to music.”

Franco Moretti, who coined the term “distant reading” to describe the computer analysis of literary works, is public enemy number one for the traditionalists. His “penchant for playing devil’s advocate”, one critic argued, “has brought him as close to notorious stardom as his discipline allows.” This affable Italian clearly takes a mischievous delight in provoking outrage, coming out with remarks such as: “There is no continuity between reading and knowing.”

Some of the experiments planned under his overall direction in CESTA’s Literary Lab are enough to prompt a shudder of apprehension in the least technophobic of bookworms. One of them aims to establish whether there is something linguistic that excites suspense in readers. “We’re talking to social psychology about it,” says Algee-Hewitt in a matter-of-fact way. The project would involve volunteers, he says. “We want to hook them up to machines like lie detectors, to see if their heart rates change in response to certain words and phrases.”

Might that not be the first step on a road that could one day lead to the formulation of an algorithm for the writing of the perfect, definitive—and last ever—suspense novel? “It’s never that easy,” Algee-Hewitt retorts. “I doubt we’ll ever—I hope we’ll never—reach that point.”

Like Moretti, he insists that, within the rapidly evolving world of digital humanities, Stanford is a bastion of moderation: that those who teach the technique on campus accept that quantitative analysis has to go hand in hand with traditional methods. “Some of the younger people are tempted by the strength and purity of numbers to think that formal methods should just be forgotten,” Moretti says. “I think that is a mistake.”

Put to him the view that his methods could deter people from reading, and he replies with a question: “Have you visited the Stanford Bookstore?”

I had, and it had been a sobering experience: the huge window display running the width of the store—packed with cards, mugs and sweatshirts—found room for only 12 books. Half of them were frivolous. Of the other six, one was “Are you Smart Enough to Work at Google?” Inside, books occupied less than half the floor space. The section on Japanese history—a subject of no small importance to a Pacific coast university, you might think—offered only seven titles.

At the campus Starbucks, 17 students were using laptops, and one was reading a book. Stanford provides a glimpse of an almost book-less future. “I am the only member of my family who still reads books,” says Moretti glumly.

From the vantage point of Silicon Valley, digital humanities can be seen, not as an attack on the arts and social sciences, but a pre-emptive rescue: a way of making them attractive and comprehensible to a generation brought up on Facebook and Twitter.

What none of those involved in digital humanities disputes is that the discipline is in its infancy. As the software for interrogating data becomes more widely available, as libraries and archives digitise more of the books and manuscripts in their possession and make the texts available to scrutiny by outsiders, the possibilities for quantitative investigation will increase exponentially. With a metaphor of which any man of letters would approve, Franco Moretti says: “This is just the first chapter.”

Images: V&A, Corbis, AKG