Reading rare books by candlelight

On a Thursday evening in winter quarter, after a shared dinner of Thai takeout, 30 rare book enthusiasts with freshly washed hands entered Green Library’s Hohbach Hall and found themselves in inky semi-darkness.

Twenty one-of-a-kind medieval books lay open on stands; the only light in the room came from clusters of battery-operated candles. “There was this moment of awe,” says senior comparative literature major Eren Yurek. “It was kind of sublime. We thought we knew what to expect, but we couldn’t have envisioned that moment.”

The hosts of the “Manuscripts by Candlelight” event were Benjamin Albritton, Stanford Libraries’ rare books curator, and Elaine Treharne, the Roberta Bowman Denning Professor in the School of Humanities and Sciences, who had invited the cross-disciplinary group of faculty and students to experience how these centuries-old books would have appeared to the medieval people who prayed and sang from them in darkness unrelieved by artificial light.

After a brief introduction to the materials, the members of the newly launched collaboration Manuscript Sciences at Stanford were invited to explore ornate Latin chronicles, prayer books, music books, and Arabic manuscripts, the earliest dating from the 13th century.

Stanford Libraries has no white glove policy; touching the rare books (with the exception of the script and the gold and silver illumination) is encouraged. For two hours, faculty and students pored over the manuscripts with careful enthusiasm, observing how pigments ground from metals and minerals gleamed, how pages that were opaque by day were translucent when turned by candlelight, how illustrations appeared more divine, or more sinister.

“The darkness made the images much more aggressive, much more monstrous. We were kind of amazed by them,” says Yurek of the Paris Apocalypse, a facsimile of a 13th-century manuscript he had previously examined in both natural and artificial light.

It wasn’t a perfect simulation – the flames of beeswax and tallow candles flicker as the fuel they produce burns, and when a draft or exhaled breath stirs the air. But even a battery-powered candle passed over the gold leaf of a 13th-century Quran can shift the viewer’s perspective enough to inspire new insights. “There are quite a few of us on campus working in object studies who recognize how helpful it can be to think through medieval daily experiences when approaching the materials that survive to us,” Treharne says.

Some students pointed the candles like flashlights, illuminating slivers of the page while casting the rest in shadow. Treharne noted that wax spatters on the manuscripts indicate medieval readers likely did the same.

“In fact, what the students were doing with those candles was not at all unlike how medieval, monastic, and devotional or secular readers would have engaged in close proximity to these smaller books,” she says.

*  *  *

Of the hundreds of thousands of medieval manuscripts known to exist today, we lack the most basic context for around 70% of them, says Treharne. Not only do we not know who wrote them, we also don’t know where they were written, why, and, in a smaller number of cases, when.

“Uncertainty is second nature to a medievalist for whom so much foundational information simply doesn’t survive,” Treharne said. “How do we account for that uncertainty in a computational world, where you need data in order for the computer to answer the questions that you need it to answer?”

Treharne, an expert in the study of the medieval book as whole object, is driven by the question, how do we catalog one-of-a-kind objects about which so little is known? And how can we harness the power of technology to enhance accessibility and further the study of these materials?

Many of the processes that shed light on these works are not automated, at least not yet – examining the shapes of thousands of individual letters, for example, as Treharne did recently in order to estimate the date and provenance of a pair of manuscripts. “As scholars, we’re trying to think about ways in which we can make more discoverable and intelligible the materials that survived from the Middle Ages, to create tools and platforms for interested parties, not just to see the materials but to be able to understand and study them,” she says.

Stanford has been applying computational methods to humanistic inquiry for decades, most recently under the umbrella of the Digital Humanities. But current leaps in technology make answering these questions more important than ever for Treharne. “It’s really the right time,” she says.

Earlier this year, an international conference organized by Treharne and Albritton and funded by Stanford HAI brought manuscript scholars, librarians, book artists, and digital experts to Stanford, where, for four days, they considered the question of how to automate discovery about handmade, one-of-a-kind objects in an age of artificial intelligence and machine learning.

*  *  *

Manuscript Sciences at Stanford is an initiative Albritton describes as a virtuous circle – getting scholars across disciplines interested in taking advantage of Stanford’s remarkable collection of primary source materials, and then leveraging their expertise to add to the existing body of knowledge about them.

“What we’re trying to do is increase the opportunities for students to encounter these materials in a variety of contexts,” he says. “We want to encourage engagement. And then at the same time make it clear how important it is that the information that comes out of that engagement goes back into the record so that it enhances discovery.”

Imaging technologies applied to ink can tell us what elements make up different pigments, for example, while DNA testing on pages made from animal skin may offer clues about where a manuscript was created. Multispectral imaging can uncover text that has been scraped off and written over. Most famously, scientists at the SLAC National Accelerator Laboratory used X-ray fluorescence imaging to reveal the Archimedes Palimpsest, two works by the ancient Greek inventor and mathematician hidden in a 13th-century prayer book.

“The beauty of working with books is that it gives us not only the content but the materiality of it,” says musicology PhD student Christina Kim. “They’re bound with woods and metals and vellum, so already that’s three different classes of things to study.” She says the potential is there to approach these manuscripts from many angles. “Take political science. The church was a spiritual center and also a political center. These books were fancy for a reason.”

For now, it’s about building community through events like the Manuscripts by Candlelight evening, where Kim says a relaxed collaboration emerged as scholars of different disciplines exchanged observations.

“It allowed us to just naturally look at the books in an inter- and multidisciplinary way,” she says. “It felt amazing, this moment where three or four of us just gathered around and were able to get to know the book really well.”

Says Albritton, “I can imagine that as we get more engagement across the campus, questions will arise that we’ve never thought of, or ways of interrogating this material may come up that we just can’t conceive because of our disciplinary boundaries. And that would be really exciting.”