The archaeology of reading seeks insight from what's in the margins

History as conventionally practiced shouldn't care about Gabriel Harvey. He knew important political and literary figures during the reign of Queen Elizabeth I, but he never broke into their esteemed ranks. The son of a rope-maker, Harvey attended Christ's College Cambridge and became a fellow at Pembroke Hall, where he began a lifelong friendship with Edmund Spenser, poet of The Faerie Queene. Harvey's fellowship was in part owing to the patronage of Sir Thomas Smith, a member of Parliament and diplomat who came from the same Essex hometown. After earning a doctorate in law from Oxford, Harvey moved to London to work as a professional reader, a kind of scholarly adviser, for prominent people. He'd read important books to confer on with his client, and he frequently wrote notes in those books while doing so—marginal asides that could provide historical context to a passage that his client could consult when he read the book, or brief asides that came into his mind while he was reading. Harvey wanted to be seen as the literary and intellectual equal of Elizabethan courtiers, but after losing a late 1590s pamphlet feud with Thomas Nashe, who satirized him as a common opportunist, he never regained his reputation. Gabriel Harvey died in 1631, in the town where he was born.

French historians Lucien Febvre and Henri-Jean Martin, in their influential 1958 study The Coming of the Book: The Impact of Printing, 1450–1800, estimate that nearly 20 million books were printed in Europe prior to 1500, and there were "fewer than 100 million inhabitants in the countries where printing developed, and of them only a minority could read." Harvey was but one reader in that small minority, and he was quite smitten with Olaus Magnus' Historia de gentibus septentrionalibus, which was first published in 1555. Harvey praised the Historia in the margins of his copy (this translated from Harvey's original Latin):

"This book is useful, as it is replete with the inquiry of the greatest and largest variety of things. It is illustrated not only with external examples but also with distinct pictures of domestic matters. Thus, it is full of delight, pleasantness, and incredible things, easily imbuing the mind of the reader with enjoyment." [1]

"He's just talking about how wonderful this book is while he's reading it," says Earle Havens, scrolling through a digital version of Historia on his laptop. Havens, the Nancy H. Hall Curator of Rare Books and Manuscripts for the Sheridan Libraries, explains that Magnus' work is a massive book on the natural history, ethnology, physical world, and mineralogical universe of Sweden, adding that demand for it was so high that it was reprinted 15 times in four languages by the early 1600s. The owner of this copy calligraphed his name and date in ornate script on the first page: Gabriel Harvey, 1578.

Harvey's copy of Historia de gentibus is one of 13 books once owned by him that have been digitized by the Archaeology of Reading in Early Modern Europe, a collaborative project of the Sheridan Libraries, University College London's Centre for Editing Lives and Letters, and the Princeton University Library. Launched in 2014 with a $488,000 grant from the Andrew W. Mellon Foundation, the Archaeology of Reading is developing an open-source, searchable viewer to improve scholars' ability to study the reading practices of engaged writers during the printing revolution. [2] The viewer created by the project went live online in September. Digitized versions of rare books tend to look the same: high-resolution page scans/images that can be navigated with page-turning clicks or vertical scrolling. Sometimes the text is searchable. Sometimes foreign languages are translated. Sometimes the text can be copied and pasted for quotation. The AOR viewer has more options. An individual digital surrogate of a volume's page occupies most of the browser screen, and a navigational tool in the lower right-hand corner controls zooming in and out and rotating the page. At the browser's bottom is a horizontal crawl of the book's pagination, with pages labeled for recto and verso views. A button toward the top of the page opens up a sidebar, where transcriptions and translations of marginal text can be read and searched.

Think of these early marginalia as the Printing Revolution's comments, status updates, and tweets. Harvey's comments, which scholars have sifted through for more than a century, are the project's first case study for exploring marginalia in texts. Five transcribers went through each of the 13 books page by page, documenting underlined passages and marks Harvey made (he developed a catalog of symbols that he used again and again) and translating his comments into English. All of this information went into an XML file, a format used for storing and sharing information on the web. A team led by Sayeed Choudhury, the Sheridan Libraries' Hodson Director of the Digital Research and Curation Center, built the database and visual interface to process and interact with that data and view the documents. The viewer allows anybody to peruse Harvey's marginalia without traveling to a rare book collection. A reader can also search his comments within individual books or across any of the transcribed volumes. The most oft-cited location in Harvey's marginalia? Rome, mentioned 41 times. Julius Caesar appears 153 times.

Harvey mentions gout three times, twice in Magnus' Historia, which has an entire chapter on the ailment, including a treatment, and it's curious how Harvey refers to it: "Can it be that the keeper of the great seal, Nicholas Bacon, the father of Francisco Bacon, and William Cecil, the English treasurer, are without this remedy? Two most prudent and delightful men." Now, Lord Nicholas Bacon, philosopher Francis Bacon's father, and William Cecil, Queen Elizabeth's chief adviser, were two of the most powerful figures in England. Harvey wondering whether they know about this book's gout treatment would be like one of Chelsea Clinton's Stanford classmates tweeting at Secretary of State John Kerry and Attorney General Loretta Lynch, wondering whether they've read this one New Yorker article. "How weird is that?" Havens, AOR's principal investigator, says of Harvey's gout comment. "How revealing is that of how somebody is thinking about people in their own historical moment? On the one hand, he's writing about how great this book is that he's reading. On another, he's writing about something utterly obscure in the context of the greatest living people in his world.

"The question is, What is all this stuff?" Havens continues, adding that, yes, such comments offer biographical detail, such as whom Harvey knew and what he enjoyed reading—but the AOR scholars believe marginalia also offer an insight into the history of reading. "And for our research purpose, we want to understand how people read books."

Contemplating how people read sounds like a trivial or esoteric pursuit. We have evidence of human written communication going back roughly 5,000 years; the book as we know it has been around for more than 500 years. Surely we've gained some command of this whole reading thing. Reading is, as far as prevailing attitudes are concerned, good for mental health, offers an armchair education, and can improve empathy, and it can do all these things for us anytime we curl up, alone, with a good book.

But this notion of reading as an individual pursuit—see: Harold Bloom's solitary reader—might not be how people always read. Reading has long been a key component of knowledge creation, which is always in flux. What we know about today—who we are, where we come from, how to deal with each other—is informed by what we have been able to extract from the historical record. Laws, science, history, the story of humans from the ancient past to this moment are entirely shaped by what we collectively read about it. The contemporary application of U.S. legal statutes can often come down to how a group of nine people read and interpret the Constitution. And understanding how people read at different points in time can help unlock the study of history itself. If everything we know is pulled from the historical record, it's worth considering who put that record together and how, and what their contemporaries thought of it. Marginalia [3] offer the closest insight we're going to get to knowing what somebody thought about a text at the moment he was reading it.

The story of Christianity and its many splits, schisms, reformations, and offshoots is a history explicitly born out of different readings of the Bible. It isn't a coincidence that the Protestant Reformation takes place shortly after the invention of the printing press. "In the case of the Bible, we've known for a long time that people read the same text in very different ways," says Arnoud Visser, a professor in the Department of Languages, Literature and Communication at Utrecht University and director of the Annotated Books Online project, a crowdsourced digital archive of early modern books with marginal notes (early modern here meaning roughly 1450 to 1750). Visser also serves on the AOR's advisory board. He tells of an edition of the Bible including commentary by the 16th-century humanist scholar Erasmus that once belonged to the Protestant reformer Martin Luther. Erasmus' explications of the Scriptures prompted Luther's animosity, evidence of which can be found in the notes Luther wrote in the margins of this Bible. At times Luther "completely bursts out in angry German comments, really shouting at Erasmus on the page, calling him an atheist, a skeptic, a scoundrel, and things like that," Visser says. "In one angry note you can see that he immediately slammed the book closed—the ink stains are on the other side of the page.

"This is the heart of the whole issue: that the margins can actually tell us very exciting things about how people transmitted ideas, and all sorts of unexpected, surprising things with their books," Visser continues, adding that historians have not always regarded marginalia as meaningful. Paying attention to reader interactions with their books is "a new insight from the history of reading that has been learned over the past 20 years. We can see evidence of the plurality of reading and unexpected sides of reading, and that has prompted the idea that looking for readers and traces of readers in books is exciting."

Gabriel Harvey, Visser says, is a "particularly suitable example to study this process because he makes clear, in his annotations, that he is constantly reading books with other books in mind, and other books on his desk as well. The annotations in some of his works are clearly meant to help his clients make sense of the text. He gives advice and pointers to other useful literature very often. This situates the book in a context of other books that are also relevant."

Understanding how Harvey read isn't going to reveal how everybody read, but understanding how scholars have studied Harvey over the years illuminates how the Archaeology of Reading project hopes to spark a more rigorous examination of annotated texts. Harvey's annotations were first noted in the late 19th century. English historian G.C. Moore-Smith compiled an overview in Gabriel Harvey's Marginalia, published in 1913. According to Virginia F. Stern's definitive Gabriel Harvey: A Study of His Life, Marginalia, and Library (1979), Harvey owned 180 printed books, a broadsheet, a folio sheet, and 10 manuscripts. [4] That may sound like a modest library for a professional scholar—a 2014 report conducted by BookTrust, a British literacy charity, determined that the average adult in the U.K. owns about 200 books—but his library is the window to what he knew and what he thought about this pool of knowledge.

"When you think of the term 'Renaissance man,' someone with vast learning, it's hard to get a handle on what that means," says Chris Geekie, A&S '09, a PhD candidate in the Krieger School's Department of German and Romance Languages and Literatures who transcribed Harvey's notes from two books for the AOR. He says spending time reading Harvey's marginalia has given him a sense of the torrent of information to which early book readers were exposed. The transcribers kept a list, which now runs to more than 1,000 names [5], of every unique individual that Harvey referenced in his notes. "He might not know who that person is, but it's a historical figure that he read in some Roman history and he's referencing them as an example or a negative model," Geekie says. "He's got this huge universe of references that he's relying on and expecting his reader to know. That breadth of knowledge is really startling."

Geekie came by this impression through an updated version of old-fashioned scholarship. He spent two years in Rome doing research on his dissertation and transcribing a pair of heavily annotated Italian joke books, bound as a single volume, from Harvey's library. He worked from digital scans, and he shows me a sample page. It almost looks like an abstract black-and-white print. Harvey filled in nearly every square inch of available white space with annotations of some kind. "I could easily spend hours on a single page," Geekie says, going through what the transcription process entailed. He'd note any marks Harvey made—underscoring, drawn lines, bracketing, wavy lines, symbols placed in the margin (e.g., Harvey frequently drew the symbol for Mars, a circle with a trident-shaped barb jutting upward to the right, in the margins of text concerned with war). Then he'd copy the notes Harvey made in their original language, chiefly Latin and Italian, with a smattering of French, Spanish, and ancient Greek, and translate them into English. Each page's transcriptions and translations resulted in an individual XML file, which would be uploaded to GitHub, an open-source software repository. Page by page, book by book, the AOR project created a data set focusing on Harvey's interactions with texts that can be searched and studied from anywhere with web access.

"I think this project is part of a larger general trend toward consulting vast amounts of information," Geekie says. "It's very difficult to process and understand all of this without some kind of computer that just holds it in one place. You could read all of these books, but you're going to miss something. You're going to forget something. Or you're not going to know something." As an example, Geekie brings up a symbol that Harvey uses twice in the books thus far transcribed. It looks like the Greek letter pi with a tilde on top. [6] It occurs once in Thomas Freigius' Paratitla seu synopsis pandectarum iuris civilis (printed 1583), an overview of Roman law that was used as the basis for legal systems then emerging around Europe. It occurs again in Magnus' Historia, in a chapter dealing with Swedish law. And when Geekie was working on the texts nobody knew what it meant. "I'm sure if there was some sort of modern canonical legal scholar, they might have known that symbol, but I don't know anyone who does that stuff," Geekie says, adding that he turned to a digital search engine to help him search and sift through large amounts of text.

In other words, he turned to Google Books.

Digital search engines weren't an option when Anthony Grafton and Lisa Jardine researched and wrote the seminal article for contemporary history of reading scholars. "Studied for Action: How Gabriel Harvey Read His Livy" was published in the history journal Past & Present in 1990, and ever since it's been the model for contextualizing not only Harvey's marginalia but reader annotations in general. For the article, Grafton and Jardine read Harvey's heavily annotated copy of Livy's history of Rome, Romanae historiae principis (printed in 1555), in the contexts suggested by Harvey's notes. They argue that Harvey read the Livy three times. Twice he read it for its discussions of Roman military and political history, to offer scenarios for discussion of Elizabethan political issues. Later in life, he read Livy in concert with Augustine's The City of God, as if comparing moral lessons of ancient pagan and Christian ideas about what it means to be a good man. Reading for Harvey was an active pursuit. Or, as Grafton says, "he read a book to do some particular thing that the book was the right tool for."

Grafton, a professor of history at Princeton, says he and Jardine didn't know what they were going to find when they began transcribing and translating Harvey's marginalia in the 1980s. "In the 1970s and '80s, scholars were working in many areas of what's now called the history of books and readers," he says. Both Grafton and Jardine are listed as principal investigators on phase one of the the AOR project, though Jardine died in October 2015. "People were working on how books are produced, how books are sold and distributed, how you can analyze the contents of bookstores and libraries to get a sense of what people were reading. What nobody had a way to grab is, How do you know how anybody read a book in a different time and place? Lisa and I had both worked on these learned, Latin-writing humanists, and that's really what Harvey is. We knew that they annotated their books."

Harvard professor emeritus and librarian emeritus Robert Darnton, a distinguished books scholar, was at Princeton at the time, and Grafton says they talked about marginalia in early books. Grafton says that Darnton doubted that elaborate marginal notes in Latin could offer insight to a reading process.

"I was totally unsatisfied with the answer that it wasn't reading," Grafton says. Since both he and Jardine were curious about these issues, and Princeton's library had Harvey's copy of Livy, "we decided to go to the library and have a look at this Harvey book. It was like looking at an old stone monument, overgrown with ivy and trying to figure out the ivy, the notes. We had no idea [what we were looking for]. We just knew that we wanted to find out what Gabriel Harvey did with this book that made him so passionate about it that he filled it with these beautifully written notes."

The research was "done the old, slow-food way," Grafton says. "When one of us had time, Lisa or I would go to the rare book room with a notebook and a pencil and copy as much as we could and then Xerox it so each of us had [it]."

While they were copying Harvey's marginalia, they were thinking about what he was saying, making notes of the books he alluded to, and then reading those volumes. They were doing what historians do, sifting through the connections between historical moments to see what conclusions they could draw. After a few weeks of reading and research, they started noticing how, as outlined above, Harvey read Livy in specific contexts and settings for specific reasons. "It took a couple of months before we had figured out that reading for him was something that could be very political, very goal-oriented," Grafton says. "It was something you did with other people as a professional, not something you only did on your own in a study."

Speaking with Grafton about how he researched Harvey was vital for Sayeed Choudhury and his team of developers when building the AOR interface. He knew the scholars wanted a more sophisticated way to view pages. "At first we heard, 'If you could just get the images on the screen, and if I can just scroll through them, page through them, and zoom and rotate, that would be better than anything I have now,'" Choudhury says. But as engineers, the team didn't merely want to develop a better viewer. They wanted to create a research tool that might catalyze new ways of thinking about the subject matter. "I sat next to Grafton at one of our meetings, and I think in that two hours I received a fire hose of information," Choudhury says. "The tacit knowledge he has, we will never fully capture it. But the pathways that he might take through something like this, or what he thinks are the richest, deepest kinds of questions that we can ask of this material—I was trying to get a snapshot sense of what an expert sees when he looks at these pages and what might be interesting to be able to do."

Listening to a scholar like Grafton discuss Harvey's notes helped Choudhury and his team shape the reader's functionality. They began to understand why it was important to be able to search terms, such as the names of historical figures, and why the context of the terms matters. [7] "We think one of the biggest breakthroughs of our approach will be the ability to link across not only different data in this context but to medieval manuscripts, to classics documents," Choudhury says. "So something mentioned in the documents in the classical period is referred to in the medieval manuscripts and interpreted, annotated, and morphed in early printed books, and then again in modern books, and so on. Maybe individuals know these references because they've had access to those documents, but why can't we ask those questions in a broader sense? What was the concept of war over time, from the early classics all the way to the modern time?"

Geekie demonstrated this research power while he was investigating Harvey's curious ~/π symbol. Geekie knew Harvey often quoted phrases from other books in his notes as a way of organizing his thoughts. So whenever the symbol appeared, he started searching in Google Books for phrases that surrounded it, using Harvey's exact wording. "I searched different variants, and I found that he was quoting a book on law," Geekie says, adding that through using Google Books a great deal, especially its 16th-century holdings, he has become adept with that era's alternate spellings and lettering. The search engine pointed Geekie toward what areas of 16th-century legal writing he should consult, and eventually allowed him to locate what he argues is the clarifying source of Harvey's notes in this instance.

In a post titled "Swedes, Lawyers, and Pi" on the AOR's Bookwheel blog, Geekie explains his theory of what the ~/π symbol means: It's a shorthand way to allude to the classical Pandects that outlined Roman law. "I need to stress that this kind of research tends to rely on moving across the surface quickly," Geekie says. "You really need a familiarity with the period to get the most out of this. That's why a legal scholar would have been able to understand that passage much more quickly and better than I have. But this started me wondering, 'What the hell is this symbol?' Someone might come along and say, 'Your pi theory is completely wrong,' but first we had to get something out there about it."

This translation process has echoes throughout written history: the role transcribers played in translating classical texts into medieval texts, medieval texts into modern texts, and now texts in general into digital texts. Geekie simply wrote the history of the ~/π symbol's first draft. "Transcribers on these sorts of projects are first-level interpreters of the text," he says. "Since we're trying to make this data and information presentable to a larger public, we have to organize it and understand it, even superficially. We have to decide how we're going to interpret this at least for now until someone who knows more can come along and fix it. Not all the time, but for more obscure elements, we're operating on a series of assumptions and hypotheses that we hope future scholars are going to figure out."

Before future scholars can do that, the AOR team needed to know what today's researchers thought of their creation. The project's first phase [8], from 2014 to 2016, focused on Gabriel Harvey and creating the AOR viewer that was made public in September. Over the summer, Havens invited more than 100 scholars, librarians, and curators who work on early modern texts at Johns Hopkins and around the country and world to peruse and explore a beta version of the viewer, seeking their feedback.

The response has been promising. "I'm teaching this class on the intellectual life of the Italian Renaissance, and we're doing a week or two on Petrarch," says Christopher Celenza, Johns Hopkins vice dean for humanities and social sciences. "One of the things I wanted the students to understand was what it was like to live as an intellectual in the pre-modern world with pre-modern books, and precisely how working directly on texts could inflect the way you were thinking about intellectual life, your version of history, and so on."

The British Library has made Petrarch's parchment codex copy of Livy's history of Rome available online. Petrarch made notes in it, identifying mistakes and correcting them. In one passage, Livy describes a traumatic political episode in the Roman Republic. Two consuls disagree on the course of action. Livy writes that consul Valerius decides not to go into battle, and adds a sentence, written in the first person, that if he had gone into battle, it might've turned out poorly because the soldiers' morale was low. In Petrarch's copy, that sentence is inaccurately written in Latin. Petrarch consulted different versions of Livy and corrected his copy with his own hand, clarifying the narrative and spotlighting "this rather subtle psychological point that Livy was making, which is that if soldiers go into battle but their morale is low it might not be a good thing," Celenza says. "That was one little set of annotations, and if you look at that manuscript of Petrarch in the British Library, there are thousands of them. I think that's very meaningful. It's a very different and interesting way of reading.

"So, to me, when I looked at the beta version of the Archaeology of Reading, what was amazing about it was the potential, because it's explosive. You start to see what people are noticing in the margins, and once you start zeroing in on every particular passage, you realize every one of those is another potential avenue of intellectual inquiry, a bit of history and a bit of process between an intellectual and his or her reading matter."

This relationship between a pre-modern reader and text, where he—and it is almost always a he—corrects the text and comments on its arguments, feels remarkably contemporary. "Maybe we're actually a lot more like this pre-modern period now than we would have ever thought," Celenza says, and offers a comparison. Who hasn't read a New York Times article online one day, and come back to it a few days later to find a note correcting it, or an update, or the discussions that take place in the comments section. "So there's an almost infinite way of thinking about texts as never finished anymore, I think, being projected in this digital environment that makes us realize that a lot of the reading of the past was like that. I think pre-modern people were just used to that."

One major difference between now and then is the number of people taking part in that online discussion. Johns Hopkins humanities Professor William Egginton also perused the beta version of the AOR reader and believes it allows researchers to approach literary historical scholarship from a much more empirical angle, thanks to creating data sets about texts. A dictionary is merely one big data set pairing the meaning of a word to its documented usage. What happens when we group word usage and context across texts and study meaning over time? Better yet, the AOR project makes that resource universally available. "Any reading act is going to be extremely selective, just like any act of perception is," Egginton says. On one level, there's the original historical reading, what happened according to the text. The second reading, in the case of annotated books, is that scholar's comments about that text. "I think what this democratization of the second level allows for is a chance to fill in potential blanks, mainly the blinders that any particular scholar may have on."

These early modern books are like museum objects. Previously, studying them required being able to see them, which meant gaining access to private book collections, or traveling to whatever research library held them, plus having access to the collateral scholarship that shaped previous readings and interpretations. Making these texts available to a wider range of eyes means different minds asking different questions. [9] Annotations "are fascinating bits of evidence that otherwise wouldn't be available at all" to scholars without resources, Egginton says. "All we would have is the normal tools of the literary scholar, which is to do a vast amount of reading at the time and then speculate as to what are the potential meanings of a text, speculate as to what text tells us about what's important to readers at the time. Now we have one more step in the right direction, filling out a picture that is going to be necessarily always incomplete, but nevertheless you get closer and closer."

Our understanding of knowledge is both dependent on and informative of our engagement with media at any particular time in history and, for Egginton, thinking about a single reader's library of personal annotations might provide a way to understand contemporary reading practices. "We do a lot of our reading in what we call a virtual space nowadays, and we are leaving traces of ourselves there," Egginton says. "Instead of leaving traces in our personal library, we are leaving our traces as individual readers of multiple texts out there with the texts themselves."

And that subtle shift, from individual reader with many texts to many readers with single texts, could pose an interesting problem for future historians. "There's a kind of inversion of the relationship between what the potential or future researcher will have access to," Egginton says, adding that his forthcoming book, Medialogies: Reading Reality in the Age of Inflationary Media, co-written with David R. Castillo, director of the Humanities Institute at the University at Buffalo, explores this area in more depth. "I find that a fascinating question, that the very lens through which we're going to be looking at things historically is changing by virtue of the way that we're collecting and disseminating knowledge."

George Santayana is credited with the aphorism, "Those who cannot remember the past are condemned to repeat it." What's left out of that saying, however, is the fact that history changes depending on who is doing the reading and writing of it. Those who cannot remember the past may be doomed to repeat it, but just as problematic are those who think history is fixed. The Archaeology of Reading project wants to help pry open the first 250 years of the book to see what we may not know. "Nothing is ever done, not even a dictionary," Havens says. "There's no OED that you have on your shelf with a little magnifying glass that you can use that has every word in it. There's always new words—the word of the year, the new term that everyone thinks is the most important that gets added every year.

"The thing that makes people go crazy is the idea of going back 500 years and doing that, where it's supposed to be more of a hermetically sealed universe," he continues. "The point is, these things are never done. There's no encyclopedia that's ever been done where there's something you can't add to it. That's true of everything."


1 This is an English translation of Harvey's original text, written in Latin. Harvey generally wrote notes in the same language as the book was printed, so predominantly Latin, though he also made notes in French, ancient Greek, Hebrew, Italian, and Spanish.

2 This explanation is a simplification of Havens' project description: The AOR is part of a broader effort of humanities scholars to create a digital research environment that "overcomes certain limitations of the codex in its analog form." The aim is to "unlock the richest surviving evidence we have of the life-long historical reading practices" of active and engaged writers of marginalia during the Renaissance and the Printing Revolution.

3 I'm using this word to describe all readers' marginal notes, but it doesn't enter the English vocabulary until the 19th century, used to describe the voluminous thoughts Samuel Taylor Coleridge scribbled into his books, six volumes of which have been published as stand-alone texts.

4 Stern listed Harvey's copy of Thomas Tusser's Five Hundred Pointes of Good Husbandrie (printed 1580), one of his most heavily annotated books, as unknown. It was held in a private collection for decades and acquired by the Princeton University Library through Sotheby's auction in December 2015. The text was immediately digitized, and the AOR team transcribed and encoded it. "It is now available to the research public for the first time in the most accessible form imaginable," Havens says.

5 "From Achilles to Zedekiah," Havens says, also mentioning a list of 100 places and 350 books Harvey references. "We are currently running data analyses on our resulting transcription work," Havens adds, "to determine where there may be statistically significant relationships between specific 'concept groups' across the languages Harvey employed in his marginalia"— e.g., if Harvey uses the word "fortuna," does he use it in similar instances where he uses words such as chance, fortune, luck, felicity, fate, etc.?

6

7 As Havens explains, by digitizing a reader's notes, like Harvey's, into a machine-readable XML format, for the first time researchers can "retrace their steps, search efficiently through a substantial corpus of their work, and reconstruct their reading practices and strategies in ways that the analog form of these printed books could never allow before these digital technologies."

8 Per Havens: "Beginning September 2016, the Mellon Foundation awarded a second $451,000 grant to the AOR project to fund a second phase, which will digitize, transcribe, translate, and encode in viewable and searchable forms the marginal notes of 16th-century natural philosopher, mathematician, and astrologer John Dee."

9 A comparative example from recent scholarship: In his 2007 paper "Foucault and the Black Panthers," Brady Heiner, an associate philosophy professor at California State University, Fullerton, contends that the influential 1970s and early 1980s writings of French philosopher Michel Foucault owe an unacknowledged debt to the writings of Black Panthers George Jackson and Angela Davis, an argument inspired by editors' notes and a postscript in two different collections of Foucault's writings and lectures. Heiner offers a radical rethink of a towering intellectual figure suggested by research pathways found in seemingly insignificant text in two books.

Bret McCabe, A&S '94, is the magazine's senior writer.