 Robert Daunton is the director of the university library and the Carl H. Forsheimer University professor at Harvard. He's a former Rhodes scholar and MacArthur Fellow. And his daunting list of publications includes such influential books as The Business of the Enlightenment, A Publishing History of the Encyclopedia, The Great Cat Massacre, and other episodes in French cultural history, and The Forbidden Bestsellers of Pre-Revolutionary France. As these titles imply, Bob Daunton is one of the founding figures, in some respects, in the discipline that's come to be called The History of the Book. And I hope to talk to him, encourage him to talk about what we might think of as the trajectory or the arc of his own career as a historian in a few minutes. He served as a president of the International Society for 18th Century Studies, president of the American Historical Association. He's a trustee of the New York Public Library and the Oxford University Press. He's received honors from the French government in recognition of his work. And he's also, unlike many academics, such an accessible writer that his books actually sell widely. And I hope at some point in our conversations to talk to him a little bit about the difference between academic prose and good writing. I also should add one other thing about Bob's recent work. It's implicit in his work has always been an interest in the circulation of information in societies. And perhaps it was natural for that reason for him to become more and more concerned with the role of digital technologies and with libraries. And I remember when he took the job at Harvard as a librarian, he had been, of course, a professor at Princeton for many years, a distinguished and widely admired as a teacher and a scholar. And when he made the move to Harvard, some colleagues of mine at MIT, when they learned about it, expressed surprise as if what a strange thing for a professor, a teacher to do to become a librarian. Well, it didn't strike me at all as a strange move, either for a professor in general. But it was certainly not a strange move for Robert Darden. In a certain sense, it's a kind of culminating expression of intellectual and scholarly and teacherly interests that have defined his career from the very beginning. And I hope in our conversation, both in the conversation he and I have in the beginning and then in the question and answer segment, which we will open to you in the audience after we conclude our conversation, that we can take up some of these matters. But let me begin, Bob, by asking you to talk a little bit about your first career as a journalist. Because in some sense, your profile is unusual, although as you know, I half share it, so I'm very sympathetic to it. Well, I should apologize maybe for lapsing into autobiography. Maybe it's not all that interesting for people out here. But newspapers are interesting. They're interesting in an almost sentimental way. I mean, time was when newspapers were all over the place, and all of you would have read probably two or three newspapers over your morning cup of coffee. And probably none of you read in print a newspaper today, or very few of you. Actually, how many of you did read, if you say, a print newspaper? I would say, well, a good third. But you think a third, David, or something? That's encouraging, yeah. But most of the students. Maybe it just means we're old. It's a very young looking group. There's some seed from here. So having said that, and having made my apologies, I should explain that I have, in a way, a very boring conventional life as a university professor. On the other hand, I was born in a family that was really rather swashbuckling. My father was almost a textbook example of the lost generation. He was born in 1897 in a small Midwestern town. He volunteered to fight in World War I. He was seven months in the trenches, which was a very long time for an American. He went in as a good Irish Catholic. And when he came out, he tore off his St. Christopher's medal and threw it away as a kind of skeptic, I guess. He went to the University of Michigan for the better part of a year, drank a lot of beer. Apparently, he was a great skirt chaser, I'm told. I never knew him, as I'll explain in a minute. And went into newspaper work. So he worked on several small, two small papers in Michigan. Somehow he got a job in the Baltimore Sun with H. L. Minkin. So he got to know Minkin. He actually wrote a little bit for the smart set. And then he moved up to the Philadelphia Inquirer, and from there, this is all in the 1920s, onto the New York Post, which was then a great newspaper, a really fine, the best newspaper in New York. And he kept moving. But at this time, he'd been through two marriages. And he was living sort of not down and out, but he was living a quasi-Bohemian life in Greenwich Village during the Prohibition era. He became desk editor and then managing editor of the AP in New York, and finally joined the New York Times, where he was one of the top reporters and timed his life perfectly to be assigned to cover the Pacific when World War II broke out and then was one of the first foreign correspondents killed in the war. So I was then three years old. I never knew him at all. That's why I'm sort of trying to get up his history in a way. And my little brother, John, was only 11 months old. So we've been fascinated, my brother and I, ever since as to who this person was. But part of the fascination is with journalism as experienced in that generation. I was then predestined to join the New York Times. And the publisher of the Times told my mother that my brother and I would always have a job. So from a very early age, I began scribbling newspaper articles for a period of time. Worked for the Times for a long time. He spent his whole career there. My brother won a Pulitzer Prize. He's a terrific correspondent, writer, reporter, editor. I'm very fond of my little brother. But he really is a wonderful writer, I think. In any case, I was the black sheep of the family because I became a college professor. And it happened sort of by mistake because I was intending, of course, to be a newspaper reporter. I mean, I had this notion of guys in trench coats with hat those days, they all wore hats. And a turned up collar covering crime in the streets. It's a romantic vision that's sort of tough guy, shoe leather reporter type. But I trained to do it and it worked a lot in school newspapers. And then I did boot camp on the Newark Star Ledger in 19. We were going back over this. I think it was almost the time that you were on the Newark evening news. So we could have crossed paths. But we've decided we didn't. In any case, covering police headquarters in Newark, New Jersey was a great way to understand the basics of how to write news stories. And the basics are not all that complicated. But I would argue they involve an apprenticeship in this mysterious thing we call news. And now maybe we're getting into the heart of the sort of subject here before us in a media center, because what is news? It seems to me that there are two ways to answer the question. One is news is what you read in newspapers. And so that's a kind of good working definition. You could go back over newspapers and try to sift out what in them constitutes news value or news. But another view of it, which is a contradictory to the first, is that news is a cultural construct that newspaper people learn by doing. And that's what I think I picked up on the Newark Star ledger, how to write a lead sentence that would get past your editor, how to write fast, how to be accurate. I never realized that you always had to know the name and the age and the address of the victim. And if you didn't have that and the perpetrator, the suspect, if you didn't have all that information, the Night City editor would look at you as if you were an idiot. Well, graduate students in history sometimes don't get things like that, because they can often be so enamored of theory and steeped in Foucault. I'm an admirer of Foucault, so I'm not putting him down, but so steeped in him that they don't worry about the age of the person and where that person was born, and so on. I mean, I think graduate students in a subject like history, but maybe subjects that some of you are in should all do time on a daily newspaper. It's very good for the soul, but it's probably good for your perception of what goes into this cultural artifact that is communication. So I'm rambling on and on. Let me pursue this, Bobby, taking it further into your own career as a historian. Do you think that that work, you went to Harvard as an undergraduate, you were a Rhodes scholar. One of the reasons I, the newspaper experience helps to explain the clarity and precision of Bob's prose, I'm sure. But I think another explanation is that his PhD is not American, that he got his PhD when he was at Oxford. At a time, I suppose, when the English PhD was much less Germanic than it became. Yeah, well, people advised me. I first did a kind of a souped up MA degree, and then finally a PhD, which they call a D fill, an Oxford. And my tutors there said, don't get a D fill. That's for Americans. Well, I wasn't American. And they said, well, no, that's not the way into higher education. What you should do is get a research fellowship in an Oxford college, and then be elected as a fellow. And not write a book until you're 45 or 50. Some review articles, none of this professionalization. They were very scornful of the American PhD. And also, indeed, as you said, of the way Americans write. For me, it was maybe newspaper experience. But what was really impressive was having to write an essay and read it to your tutor every week so that you heard your prose hitting and offending the ear of your tutor. And you could watch him sink deeper and deeper into his chair. And you knew something was wrong, but what wasn't exactly? After a while, I think Oxford brought out in its students the importance of clear, unaffected prose. And most, at least in history, most English historians write much better, I would say, than most American historians who are often enamored of these very conky, Germanic sentences with lots of sort of buzzwords sometimes scattered through them. Now, maybe I'm not being fair, but certainly that was a kind of ideology, an implicit ideology in the way history, and I would say all subjects, were taught at Oxford. But when I finished my, I sort of did a defil for the fun of it. I had a scholarship and I had plenty of time and I spent all my summers working for the New York Times in its London Bureau. So I was continuing to prepare for a career as a journalist and then was given a regular job in the city room of the Times. And I lasted three months. I walked into Abe Rosenthal's office and I said, as gently as I could, I quit. And then I took a long walk behind the New York Public Library in Bryant Park. And I felt this weight depart from my shoulders. It was very hard to be the son of a famous correspondent who was a kind of war hero. I had to walk past a plaque to him that was on the wall every day. And really, I had to succeed him. But that was difficult to do if I wanted to be myself. So without going into psychoanalysis, I think it was actually a very liberating thing. But the main point is I love doing history. And the kind of history that I absorbed at Oxford was archival history. The idea was to get your hands dirty. It was very easy to do that. I had a little motor scooter and I could scooter off to Paris, take the ferry, go to Paris, and just order up boxes from the archives. The sensation of opening a box full of manuscripts is there's nothing like it. It's much more exciting than covering a holdup in the middle of Newark. They come in boxes about like this. There's usually a kind of a ribbon-like attachment on the side. You undo the ribbon. You fold back the top of the box. And inside are what the French call chemise. They're dossiers, folders. Piled up, you take one out, you put it down, open it up, and start reading. And out of this, one letter or memo or whatever it is after another, somehow you've got to produce history. Well, I think one of the things you're simplifying here is the difficulty of reading what you see in those archives. I mean, I remember my son is an influence by Dorton. He is a historian of the French Enlightenment in part, and works in territories that were pioneered by Dorton. And I remember when he won his fellowship to go to France to do his research, he had to take a special, he took a special course in how to read 18th century French handwriting. And apparently it's a very difficult task. No, it's easy. I mean, it's 18th century, not to put down your son. But 18th century French handwriting is easier to read than modern French handwriting. It's true, whenever I teach a seminar, I usually have photocopies of one of these dossiers, usually from the papers of the Bastille. And they're wonderful documents because they're interrogations in which you can follow a police interrogator laying traps for the prisoner. So it's question, answer, question, answer. And at the bottom, the prisoner signs every page to testify to its accuracy. So it's a drama, you know, and you see into the lives of these usually rather obscure people, but that's not history. I mean, history is somehow putting together all of this raw material into a story that is going to be, one hopes, important for the attempt to understand the human condition as it was lived 200 years ago or 300 years ago. And that's what I mean is, that's the drama of opening up the box because you come into contact with vanished humanity, you know, with people who've died and disappeared 200 years ago and 99% of the cases known has ever heard of them, but you have, and you have the opportunity to try to bring them back to life. So one way to put it is the historian plays God. You can bring this person back to life. I'm reluctant to interrupt God, but let's turn now, on the basis of what you're saying, Bob, to the actual sort of substance of the kind of history you did, because let's leave aside the question of books for a moment and book history, but turn to the question of the way in which so much of your work, and there are some other historians, of course, Natalie Davis comes to mind, but and of course some continental historians who work on materials that seem to be what, I guess, one of the phrases is history from below. What is it that helps to explain why you went to sources that and wrote books about materials that an earlier generation of historians would have thought were irrelevant for the writing of history? What lies behind that? And maybe define more clearly than I could do what the substance of this kind of history involves. Sure, well I can make a stab at it. I don't know if all of you have heard of this phrase history from below, but it was a kind of rallying cry for historians and students in general in the 1950s, 1960s, and it had a kind of left wing tinge to it. The idea was that history involves everyone, ordinary people, people at street level, not just aristocrats, generals, kings, queens, and the rest of them. The French version of this was histoire total to do a total history to try to understand the entire society as it held together in a systemic way 200, 300, 500 years ago. So it was another whole vision of the human condition and some of the historians I found most inspiring managed to get that across. I mean you can't do it perfectly, but people like Edward Thompson or my favorite French historian Pierre Goubert really gave you the feel for life as it was experienced by ordinary people long ago. And that I think provides another dimension to your own view of the world because you see the world as steeped in the past. It seems to me most historians have a very shallow view of the world because it's so present oriented. And so history from below, in my case, I was studying the enlightenment, the history of ideas and so on. And it seemed to me it was very much a matter of famous writers having exchanges with one another and producing famous books or great men and great books. There were no women included at all at that time. And people pretty much read famous philosophic works of the Enlightenment. That's how I was trained to do it. But I got attracted by the second rank, third rank and then 10th rank writers in order to understand what writing itself was, what the, how the literary world was put together, what the rules of the game were in this thing we call literature. So just to give you one example, going through the archives, I found police reports on writers. So you're in the police archives. I mean, I had been in police headquarters in Newark, New Jersey, but now I found myself in the police archives where it turned out that the police did a census of every writer they could find in Paris between 1748 and 1752. They found 501 writers. That's a lot. And they actually produced printed forms with rubrics that were then filled in by hand. I mean, this is sounding like Foucault, you see, because you had in fact a kind of grid, a conceptual grid that was then filled in with information by police spies and police officers. Well, this census was simply astonishing. There you found Voltaire, Rousseau, Diderot, Montesquieu, the most famous writers, and surrounded by about 490 writers that no one had ever heard of. And it was possible to look at literature from below, to understand the life of hack writers, of scribblers and garrots, to see also how they fit into society, because these rubrics were really interesting. I mean, one, for example, is age. So I had a kind of demographic chart about the age structure of writers at that time. In other words, origins. Do they come from the provinces and flood into Paris? It's one of the main themes of French literature. Then they had a rubric called Signalement, which is the way they looked. And it turned out that the police were reading physiognomies, and they were very tough in the way they described these people. And atrocious physiognomy, a hideous meme, just one thing after another. But the idea was to read faces for characters, which I found fascinating. And then finally a rubric called Istoire. And there they told the life story, as much as they could piece it together of these writers. Well, these life stories were simply, in many cases, extraordinary. And I could go on and on about it. But you see what I mean. The point is, here's the Enlightenment. In the year 1748, you've got the first great works of the Enlightenment. Not exactly the first, but 1748 is a year when French literature explodes. You've got de l'esprit des lois. You've got the announcement and the prospectus of the encyclopédie. You've got important works of Voltaire coming out by 1751. Rousseau has published his first discourse. I mean, the whole Enlightenment seems to appear in print in a very few years. Now it's not as if the police thought used the category Enlightenment in their work. That's something that was foisted on these writers in retrospect. But they understood that literature was a power and that the means of communication, the kind of thing you study here in a media center, these means were important and needed to be followed by the police. So in trying to put together the pattern of lives lived in Grubstreet, Paris, I thought that I was getting at literature from below. And that was the idea in any case. Well, Bob, this encourages us to make a nice transition more explicitly now to the sort of whole discipline of what now is called a discipline of the history of the book. And I wonder if you would address it. You could talk a bit about your own contribution to it, but how you define what you think about how the field emerged and why it's significant. Yeah. Well, when we say the book, we're not really being accurate. We really mean the means of communication. So historians of the book frequently study film. They study dance, notation. They do lots of things. It's not at all limited to history of the book. That phrase somehow took. It began in French as histoire du livre. And then it really passed into German, geschichte des bouques phasens, which is a real mouthful. In English, it still makes you feel a little awkward. At least when I say history of the book, people look at me as scant and they say, well, which book? So then I say, well, book history. And they say, well, what's that? Don't historians always use books? The idea is to understand how the printed word, or often the manuscript word, or the sung word, which is my theme here at MIT Wednesday after next, how words shape consciousness. How do they become a force in history? The nice thing about books is that you can measure them. You can follow their path as they spread through society. You can, to a certain extent, trace their effects, although that's very, very tricky, of course. So the history of the book, as we now call it, or the history of books, or book history, has, I think, come together as a new field of study. Now, it's true that people have studied books as long as there have been books. And there's a great tradition, notably in the English-speaking world of bibliography, that analytical bibliography that has been taught in library schools and English departments for quite a while. It was really invented at the very end of the 19th century and developed into a rigorous way of studying the physical aspects of books. But that isn't what we mean by book history now. It's one aspect of it. While the English were developing analytical bibliography, largely to figure out what Shakespeare probably originally wrote, the Germans were studying the catalogs of book fairs in Leipzig and Frankfurt, and were developing a more economic view of the history of books. The French studied it largely through state archives. I mean, everything seems to be filtered through the state in France. And so they were very interested in censorship, state control, that kind of thing. We began trying to put all the pieces together, and the lead was taken not at all by me, but by people in the previous generation, especially in Paris. And the greatest of them was Henri-Jean Martin, who wrote a thesis, two volumes like this. It's about 1,000 pages. But once you've worked your way through it, you have a new view of how rich the study of books can be. This was linked with the so-called Annales School of History. That is a school that tried to see history from below and to do what I refer to as total history, histoire totale. So how could you do a total history of books? And those of us in the, if not the second generation, maybe a half a generation behind Martin, tried to use Annales techniques in order to understand whole communication systems. So in my case, it often took the form of drawing diagrams, which may seem pretentious, but the idea was to trace a communications circuit from author and publisher and printer to, it turns out, the wagon drivers were very important. We could talk about wagon drivers and smugglers, if you like, they're very interesting characters, but it was crucial, it turned out, transportation was crucial in the total history of the book. And then from there to booksellers and libraries, and finally to readers. And the readers, of course, would influence writers indirectly, but the circuit could be, in that sense, completed. What I tried to design as a model for doing book history, in other words, was a total history in which you would study all aspects of the book and not simply the words on the page or the life of the author. So I found myself drifting into the papers of publishers. Now it turns out that there is only one publisher from the 18th century whose papers have survived in a large mass. And that was the Société typographique de Neuchâtel, that's a publisher in the little Swiss town or city of Neuchâtel just across the French border. And that was not an accident because France was surrounded in the 18th century by publishing houses all the way from Amsterdam and Brussels through the Rhine country into Switzerland and down to Avignon, which was then papal territory. The publishing industry flourished around, what I call a fertile crescent of publishers. It flourished because there was a tremendous demand in France for books that were not censored. The censorship was quite active. It was fairly permissive, but still repressive at the same time. And there were book inspectors, there was a special book police, there was above all a booksellers guild which had a monopoly that was controlling the production and distribution of books from Paris. But the people outside of France in Brussels or Geneva, they published whatever would sell. They even had agents in Paris who were sending them letters about what was selling and what was likely to be published. They did serious market research and they didn't call it that, but there's no question about what it actually was. So that their commercial correspondence is full of information for what you could call the sociology of literature. And these papers are simply thrilling to read if you're interested in that kind of a question. They have letters by smugglers. Some of them, many of them of course were illiterate, but the smugglers were organized in teams and the captain of a team of smugglers could write the letters. In fact, they didn't call himself a smuggler, a contrebandier, he called himself an assureur. Now an assureur was an insurance entrepreneur. It was an insurance game. Smuggling was called assurance. It was a major industry, which they worked out very carefully so that the person who shipped the books would take out an insurance policy for them to get across the border from Switzerland into France. And if the books were cost, their value would be repaid. The cost for the insurance was usually about 16% of the commercial value of the books, but the smugglers themselves were illiterate peasants carrying books on their back and baskets that usually weighed 60 pounds, 50 pounds in winter. It's all worked out very carefully. If they got caught, they would be branded on the left shoulder with the letters G-A-L for Galerien, or galley slave, and then sent to roll in the galleys in Toulon or Marseilles for nine years. It was a tough business, nine years that the galleys is enough to kill you. So yes, there was a, you know, there's a lot of hugger mugger and razzle dazzle in this. Bob, here's a question that I've often wondered about. The, a lot of the work you're talking about took place really, I suppose, before this happened, but I'm wondering how much the, what we call today the discipline of the history of the book has been not created but encouraged, incited, strengthened by the awareness in the digital age that the book is in some sense an artifact, an endangered object. In other words, what the relation is between our recognition that the book is a historical artifact, what the relation between that and the fact that we exist in an era in which the book seems like a vulnerable or an endangered species in some way. Do you think there's anything? Well, I think there's a lot to that, although I don't agree that it's an endangered species. Neither do I, actually. I've been invited to more conferences on the death of the book. And, you know, more books are published every year than the previous year. I've got some statistics on that. It's simply amazing how the production of books expands and expands. I daresay next year will not show the curve continuing to go up. We're facing tough times, but the production of books has not declined. Publishers, I think, are a little bit like farmers. They like to complain and they say the weather's terrible and profits are down, but the book is actually thriving and it's not surprising because the codex, as opposed to the volumen, the book where you turn leaves and you can thumb through, this invention is, I think, one of the greatest inventions of all time. We don't know when it happened. Maybe around the first century of the common era. We don't know who invented it, but it spread and it took and it still is very powerful. So that's just part of the answer. If you count this as an answer, I do agree that there is more fascination with the book today than there was, say, 10 or 20 years ago, partly because so much information is available online. And so sometimes students pick up books as if they're handling some queen object. And I teach a freshman seminar at Harvard on the history of books. We begin with, it goes from Gutenberg to the internet. And we begin with the Gutenberg Bible. Well, we're privileged because I always have a librarian who coordinates the teaching with me and for this session, the head of the Rare Book Library walks in with a Gutenberg Bible under his arm. He plops it down on the table. And the great thing is he doesn't just ask the students to stand back and admire this first major printed work, but to come up and turn the pages, to feel it, to look at it. And then he asks some questions about its structure. Even if they can't read the Latin, he says, why does this first part have 44 lines to a page? And then you keep 42 lines and then you go a little further, it has 43 and then 44. What does that mean? What's going on? Or why is the- What's the answer? Well, the answer, we think, you know, this is, these are, we know nothing about Gutenberg, next to nothing. I think there are less, less is known about Gutenberg than is known about Shakespeare by far. But the conclusion that most experts have reached is that Gutenberg was trying to save money on paper. Paper was by far the most expensive element that went into the production of the book and continued to be so until mass-produced paper began in the early 19th century. So Gutenberg realized that he could squeeze more words onto a page by various tricks. He packed the type more thickly and he began sawing off ascenders and descenders on type. That is, you know, the T that sticks up, you saw off the top, you've got that much more room or the bottom of the Y. And sure enough, if you look at Gutenberg's type real closely, you will find that there are sawed off chunks of type. It's endlessly fascinating, but the argument goes that Gutenberg was not producing a luxury object. On the contrary, this is the beginning of the democratization of knowledge. It was a down-market Bible, even though now it's, you know, there were millions and millions of dollars, but probably, and this is borne out by studies of some of the bindings too, his Bibles were sold to relatively poor monasteries as opposed to great rich courts of the princes. So you can learn a lot by the physical study of the Gutenberg Bible, and this is just kind of beginner stuff, but my point is this, the students of course were just knocked out by being able to touch the pages of the Gutenberg Bible, but I sensed in them a fascination with ordinary books from various times in the past, and that is part of what I believe is making book history so attractive to people today, because most undergraduates think all information is online, and they go to their computers for, first, if they want to find out about something, but once they're drawn into the world of books, I think they discover really another fascinating universe. So your message is no allergies for Gutenberg. That's right. Right, but this also allows us to make a transition to a topic I'm sure will be a central issue for people when we turn to questions in a few minutes, and that is the relationship between, something you've begun to write about a good deal, the relationship between the emerging technologies and the culture of a book, and what the responsibilities of libraries are, and one of the, one of the most interesting sort of sentences of yours I've read recently came from that piece that you published fairly a few months ago in the New York Review of Books about libraries and the fate of the role of libraries in a digital age, and you said your sentence was, information has never been stable, and I'm wondering if you would unpack that sentence for us, why you felt it was important to make the point that information is not stable now and never has been stable. Well, I think people have the illusion that anything goes now in the world of information. You've got blogs, you've got ordinary citizens who are tossing stuff into the internet, and so you can't trust what you read, and I think one should never trust what one reads, that you should always read skeptically and critically, but I was trying to make the point that it was like that in the past as well. That texts which look very firm and solid when printed on paper, the type is embedded into the paper, it's not just sitting on top of it. There's a kind of firmness to a book produced in the era of the common press, and so you tend to look at it as if the information there was absolutely stable, but it wasn't. If you take the case of Shakespeare, the first plays of Shakespeare printed in quarto format are notoriously faulty. They're full of nonsense. I mean, you can take whole pages and you can barely make out anything at some of the so-called bad quartos of Shakespeare. The first folio, which is what we depend on more than on anything else for knowing Shakespeare's works, is also riddled with mistakes, but in the case of Hamlet, and now I hope I'm remembering my facts correctly, but there's the Hamlet I think in the first folio is twice as long as one of the quarto Hamlets. So what did Shakespeare actually write? The disparities between the two versions of King Lear are enormous, and I could go on and on, and in fact this is of course created the industry of Shakespearean scholarship really matters to us that we get as corrected text as possible, but I think anyone would agree who studied the question that Shakespeare's texts, the most important writer in the English language are faulty, untrustworthy, and we are only guessing at what was actually written in some of the most beautiful passages, some of the best poetry in Antony and Cleopatra, for example, was produced by Compositor C, and Compositor C was, he just messed up everything that he set in type, and then we've got Compositor D and A and B, I mean for Shakespearean scholars this really matters, and I tried to carry the argument further and to go into the newspapers as produced quite a while ago. The example I chose was report a news article about the battle of Brandy wine. This I stole from a PhD thesis written by one of my students, a guy called Will Slaughter who wrote I think a beautiful thesis. He showed that the report on the battle of Brandy wine, which as you know was lost by George Washington, appeared first in a loyalist newspaper in New York, so Brandy wine's near Philadelphia, already something's funny in the eyes of the readers. It's taken to Nova Scotia, to Scotland, I think it was Edinburgh, reprinted, copied because everyone just lifts passages from everyone else in the world of early journalism, transmitted down to London, recopied in various London newspapers, but in London they were betting on the American Revolution, literally on the stock market, there were bulls and bears then, but also politically in Parliament because the government was teetering and about to collapse, the Lord North Ministry was in great trouble and if it looked as though this upstart general, Washington, was winning, the government would fall and so would the stock market. Now a lot of people thought that was what was going to happen and so they disbelieved the story and they said, well it can't be true because it came out of New York, not out of Philadelphia. Furthermore, Washington uses the word arraying my troops in one of the documents, the quote that comes in the story, no general would say arraying my troops, well it turns out Washington said arranging my troops and it was a typo, but the whole thing gets misconstrued to such an extent that the Battle of Brandywine is finally reported as a victory for Washington, who however was killed in the battle. In fact, he was killed four times in the course of the American Revolution according to the English press. I mean, newspaper reporting at that time is just wildly inaccurate and you can't... So the larger point would be that the people who today claim that the universe of digital information is so unreliable that we should ignore it or we should recognize it as a special case or naive about the inconsistency and contradictoryness of all information in all times. Yes, that's right. I think that if you're an historian, you should be very skeptical about what you read in the Daily newspaper yesterday and 200 years ago. I think we should be skeptical about all texts, but that's part of the fun because you can't... I mean, the story of how the Battle of Brandywine was misconstrued is itself a subject of study, sure. And in fact, it had tremendous repercussions because it was then translated into French and Spanish and German. And the French version is great American victory, we should rally behind these colonials against our enemy, England, just when France was deciding to, in effect, join the Americans. So the significance is, so to speak, in the way information is transmitted, not just in the words that appear in print. And that's the kind of thing that we're trying to do in the history of books or communication. I think it's what people do now when they study how news is reported on the web or other forums, what you'll be talking about after the election as you just told us. So I think the focus is shifting and it's a great moment for trying to understand the process of communication and how that affects this other mysterious thing that we call public opinion. It's a real force in history, but it's very difficult to get your hands on it, to find out how it intervened when it was crucial, what influenced the shape that it took. So I'm for a total history of communication, including communication by the internet and by songs and by jokes, by graffiti, by all the media of any period in order to understand communication in a large sense of the word. I'd like to make a transition. This in a way will be my final question to you, Bob, before we turn it over to the audience. And I know we have some librarians in the audience. I'm counting on them to bring us to our real topic. But let's begin moving in that direction by asking you, let me begin, by asking you to talk a bit about what you think the responsibilities of libraries are in this incredibly complicated space we now occupy. I mean, there are obvious limits on the resources of libraries. They can't be all things to all people. Maybe one way to approach it would be to, I hope that you've often written and talked about research libraries and that should be an aspect of our topic, but I was hoping you might begin by talking about, if you think of a significant difference between the responsibilities and obligations of research libraries as against what we would call, say, public libraries, what they will be like in the digital future that is impending and what they have been like. Right, well, I'll take a stab at it. Certainly in New York City, the Neighborhood Library, so there are now 86 branches to the New York Public Library, which as you know, isn't public, it's a private library. It's a private, three foundations came together to create it. The branches are extensive, they're important for the life of the neighborhoods where they make books, but not just books, also videos, CDs, all sorts of things available. One of the things they make available actually is access to computers. I mean, everyone in MIT, I'm sure it takes it for granted that you've got a computer on your lap, if not under your pillow. That's not true. On your wrist. Right, it's not true of everyone in New York. In your left nostril. Not by any means, there are a lot of people who can't afford computers, who are intimidated by computers, but who need access to them. So the Neighborhood Library has become a place in New York City where computer training takes place, where people have access to the web, and where a lot of very small businesses, often by immigrants from Korea or from Latin America, can begin to use data sets and information in a way that will make a difference for the small businesses, groceries, or whatever they are that they're setting up. I think it's a very important service. And furthermore, as you know, a great many people in New York City don't speak English. Well, the Neighborhood Library gives courses on English and in English, but also English plus Spanish or English plus Korean, whatever it might be, to the local population. I mean, this is a vital function. And we found that when we had to, I say we, I'm one of the trustees of the library, we had to close the libraries two days a week. So we went from a schedule of seven days a week to one of five days a week. The neighborhoods protested. I mean, they really counted on this. For some neighborhoods, it's the most important building in their area. And when we finally got enough funding, we went back to a seven day week and it made a big difference. So I'm trying to say that libraries have a vital function, I think, in the life of communities, ordinary libraries. Research libraries are something else, but as you know, no one has solved the problem of how to store and preserve digital documents. So we migrate them through various formats and so on, but they're not like books. They could disappear. They could become inaccessible because of inadequate metadata. Or they could simply have one or two of these zeros and ones drop out, which happens through what I call erosion. I'm sure it's not scientifically correct, but you can lose just a few of those and the whole document disintegrates. So research libraries, one of their minimal functions. You were talking about the instability of platforms. Yes, that's right. Instability of energy. It shows you what a brilliant technology the book is. You can drop it from the top of a building and it's still a book. So I mean, this is a minimal function, but we depend on libraries to preserve our cultural heritage, research libraries. Don't depend on Google. I mean, Google may not be here in 10 years. In 20 years, it won't be here. The technology will have been replaced by something else and the product of that technology, this wonderful, huge mega library being digitized at this moment, it may not be accessible anymore. So we'll still have our great research libraries. And that's one fundamental duty, I think, of libraries, but well, I could go on and on. And we might go into this story. One implication is don't destroy what you have, including the card catalogs themselves. Yes, I'm for keeping card catalogs. Newspapers? Yes, especially newspapers. I don't know how many of you have read digitized newspapers or microfilm newspapers. The number of missing pages and thumbs over the columns and mistakes, it's just terrible. And often the microfilmers or digitizers simply skip pages. And when they come to books, well, I don't know, maybe some of you have horror stories of reading books. My favorite example is Tristram Shandy, which is almost the most favorite book from early modern England for a lot of avant-garde literary professors, people like you, David. Why? Because it's so postmodern. And Lawrence Stern, he's writing along and you read the first edition of Tristram Shandy and then you come to a blank page and Stern says, well, it's up to you, the reader, to make it up. The reader becomes drawn into the text itself, becomes a kind of collaborator or sometimes even an arguer, an enemy of the author. And at another point, he has one of the characters dies and the next page is black, just a black page and mourning for the character who died. It's very playful, it's very clever and it's using the physical qualities of the book to get across in a self, almost a parody type of way the fact that what you're looking at is an artificial medium that is trying to communicate something about life and death in an imaginary world. Well, when you look at it in a digitized version, the people doing the digitizing thought, well, there's a blank page, we don't have to digitize that and the black page, well, that's clearly a mistake and furthermore, in some cases, they begin with volume two instead of volume one and they're missing pages and so on and so on. It's, and then which edition of Tristram Shandy is going to be chosen by the digitizers? They have, I dare say, Google has not a single bibliographer among its many thousands of engineers. But it matters that they select this edition rather than that edition. Now, I speak as a fan of Google. I think Google is marvelous, but there are real limits to what they are doing and therefore the research libraries are going to keep and make available important copies of different editions of all of these works. So that sounds like an advertisement for libraries, but frankly, I'm here not to fling a commercial out at you. The reason I left Princeton, which has a great history department to come to Harvard was not really to be at Harvard, but to be part of the shaping of the new information age through the library because the library is a crucial place where it's happening. That's where we have some leverage and if we can use this leverage effectively, I believe we can make the new information age work for the benefit of the general public and not for some private company that's intent on making money out of it all. So there's a lot going on and I feel that librarians and libraries are absolutely central in what is going to emerge as what we call the information age, even though every age was an age of information, as I say. Well, we're at the stage in our event when the really intelligent things happen. The audience begins to control the discourse. Let me ask you to, there are microphones on either side. Try to keep your questions and comments concise. We begin with Professor Uricchio. Hi, could I just pick up on that last point and hear more about that linkage between the tradition of the printed library and this digital era? For example, some text, textual systems, things like Wikipedia, which are pretty interesting sources and have the problem of being highly dynamic and, of course, digital with all the kinds of problems you mentioned. So could you just talk more about your vision of where you see this readjustment of the library's mission? I mean, not the conceptual mission of keeping things, but rather how are you gonna cope with an environment where not everything is in print and I guess the pressure of Elsevier, as it's said, or there's a transitional moment in the publishing industry itself. Some folks are trying other modes of publication. Old print companies are starting to have killer rates. So how are you gonna cope with that? What's your vision there? Well, I'm gonna do my best, but I don't think personally I have any solutions up my sleeve, but I'm a great believer in electronic books, in digitizing, and even in the creation of a new kind of book that can communicate things that weren't possible through the good old codex. Where to begin? First of all, it seems to me we have a responsibility as librarians in research libraries to preserve the seemingly fleeting character of messages in the new communication system. So at Harvard, we're now beginning to try to systematically preserve emails. It isn't easy. Of course, the scale is enormous and it's very sensitive because a lot of email is private and so on, but we want, see, I'm responsible for all the records generated out of Massachusetts Hall, the president's office and the corporation and the rest of it, along with the archives that go back to 1636. And most of the communication that comes out of the president's office is in the form of email or a lot of it is just spoken and some on telephone, but I would say that email more than anything else is what is escaping us. So we have a pilot study to try to see if we can preserve all of the email. Similarly, we are trying systematically to harvest websites that are generated throughout Harvard University at MIT. I think, you've got a lot to teach us because MIT has been so advanced through DSPACE and other things in trying to integrate the electronic modes of communication with the traditional modes. So that's another responsibility. Whether we can solve it all, I don't know, but it's an enormous problem. You know, the State Department began electronic communication in 1970 through magnetic tapes and other sorts of things. Almost all of that has become inaccessible. We have a 30 year rule in this country, so in the year 2000, all of this was supposed to fall in the public domain, but most of it was lost. I say most of it. I haven't done my research on this. I could be wrong. I do know that the really important exchanges, of course, were preserved, but the ordinary business of diplomacy wasn't, I think, in large part. So we are losing a vast amount of our record, and I forget how many millions of email messages were exchanged in the Clinton presidency. I read the figure of something like 500 million or something like that, and they're all just simply gone. And more millions, of course, are being produced every day in the current White House. So that's one problem. I could go on and on. Maybe I should stop at this point, but... I think my family generated 100 million last month. What I think libraries need to do is to be the nerve center of universities. It's the library, after all, that transmits the databases and the electronic communication among laboratories in very large part. So the library really is the place where it's happening, but it's not as if we've got it all solved, and the responsibility to develop things on the digital frontier, while at the same time maintaining acquisitions while more and more books are being produced each year, that responsibility is pretty heavy. How we can do it all, I don't know. We've got real problems, and a lot of it is going to involve fundraise. My name's Whitney Choutine. I'm a graduate student in CMS here, and also an aspiring book historian, so I have a lot of questions, but I'm actually gonna restrict it to one, which is kind of narrowly related to my own research. I'm looking at moving parts in early modern books right now as part of my thesis, and I work in the Hyper-Studio Digital Humanities Lab here at MIT, I'm a big proponent of digital archives and very excited about Google Books, even with all its problems, I'm very excited about all this stuff, so I'm not a pessimist, so I wanna preface it, my comment with this, but at the same time, I've been concerned that there's not enough metadata on moving parts in books attached to images or to searches, library searches, and also I've noticed that even the fact that there are moving parts in books has been kind of alighted in digital images, and I'm wondering if you have any comment on what is lost even in high resolution, perfectly scanned digital facsimiles, what is lost about the book, and also I'm not indicating the turning pages technology of the British Library, things like that. I wonder if that's almost a distraction from the whole issue. Well, I take your point because the point about moving parts in books is they move. I mean, you have to touch them and move them. How can you do that with a digital image? Now, there might be some way around that if you produce an effect of a movie, motion pictures, but as far as I know, that's not been done. The tactile quality of books in general is very important. Here, there's a danger of sounding romantic. I mean, every time I pick up a book printed in the 18th century, I hold the leaves up to the light and I try to see if someone's petticoat has not been ground down and all the rest of it. And frequently, you can pick bits of thread out of the paper itself, and I find that rather exciting because you have this tactile contact with somebody's petticoat that had been worn 250 years, 300 years ago. Not that I'm a petticoat chaser or anything like that. So that's a loss. That's a loss. There are gains, of course, and I'm sure you probably know about the copy of Beowulf in the British Library where the digital version of things made it possible to see things that had not been visible earlier, and apparently we've got one or two words from that ancient language that were discoveries. So there are things you can do digitally that you couldn't do otherwise. What other losses? Well, one of my favorites, but this also sounds romantic, is smell. A lot of people get a kick out of the smell of books, and it does, in a way, waft you into another imaginary realm. There is a French online publisher who did a survey of students at the University of Paris to find out what put them off about electronic publishing, and the main answer was, believe it or not, smell. So this online publisher developed a kind of sticker that you could put on your computer and scratch and get a sort of rare book smell. Well, that's not a serious answer to your question, but in general, I subscribe to the thesis that one medium of communication doesn't displace another. I think they coexist. They change the balance in the way communication happens, but we've learned fairly recently, actually, that manuscript publishing continued to flourish long after Gutenberg. I mean, of course, for the rest of the 15th century, the 16th century, but now we know that it was continuing throughout the 17th century, especially in England and in France, and into the 19th century. You could publish a book, according to Donald Mackenzie, who is one of my heroes among the founders of book history, that you could publish a book more cheaply if you hired scribes to copy it up to 100 copies. So there's a lot to be said for scribal publishing, which continued to flourish. And, of course, a scribal book is quite fascinating to read. I mean, the experience of reading a scribal book is, again, quite different. It's especially important in Iceland. It turns out that the Icelanders, they had enormous libraries of scribal books, and they didn't really have printing. I mean, printing existed attached to the, there were two bishops in Iceland in the 16th century, so printing existed in Iceland before it existed in what is today the United States, but they just turned out a few missiles and such works for the bishops. Everyone is just copying books, and they did it for about three centuries. They didn't have bookstores. They didn't have printing shops, really, and I think they had almost no schools, and yet it was a country that by 1700 was almost entirely literate. So, you know, you've got to think differently about digital, about scribal books, as well as digital books. It's not an adequate answer to your question, but thank you for answering. Bob, one of the implications of what you're saying is that the job is so daunting. You want us to preserve every possible form of printed culture and every possible form of digital culture. No library could possibly do this. Isn't part of the answer much more extensive collaborative work across libraries in which some libraries take on some duties and other libraries take on other duties and there's a real partnership? Yes, I think that we need to share the wealth and share the work. Libraries, of course, are very touchy about their treasures, and some libraries and archives don't want things digitized because that's their patrimony as they describe it. So, it's a ticklish thing, and it involves, if you like, political questions. But it seems to me that the Library of Congress, which is our only deposit library, there's seven in Britain, there's only one in this country, that the Library of Congress ought to preserve much more than it does preserve. The physical things themselves, including all sorts of ephemeral literature. But the interest in ephemera isn't what it might be. I mean, now we study the ephemera of Shakespeare's time if you can get a playbill from Elizabethan England. I mean, that's a fabulous thing, whereas people at the time, of course, were tearing them off walls and stepping on them. We need to collect ephemera. We do need to collect everything. And I think we need new tax payer support. It seems to me that the Library of Congress, in its larger mission, ought to, maybe it could be paid for by renting or selling bandwidth with. After all, the air belongs to the people of the United States, and why can't it be, if you're gonna be renting it out or selling it, sell it for the benefit of the people of the United States in this mission to keep and preserve everything produced in the form of communication. Alluded to the relationship between Harvard and Google, which I think preceded your coming to your job. What's the status now? Is it like one third or one half done? And secondly, what's the assumption about the end point with that mountain of data? How will Harvard use it and how will Google use it? And what do the two entities agree on? Yeah, well, it's a... Can you hear me all right? It's a delicate question. We were one of the first to cooperate with Google for books in the public domain. It turns out that the books in the Harvard University Library are, the whole nature of the library is quite different from many other research libraries. For example, roughly half of the volumes are periodicals or serials. And of the other half, roughly three quarters of them were produced outside the United States. So Google is not clear that Google is going to be venturing far into the world of non-U.S. books or serials and therefore Harvard's collection isn't as attractive maybe for Google. I can't speak for Google, but it may not be as attractive as it might be so that we might have only a million volumes or so to contribute. I mean, I'm using very round figures, but we have 16 million volumes in the Harvard University Library. It's the largest university library by far in the world. And that's a disappointment. I would like all 16 million of those volumes to be digitized, well digitized, to be available free to the whole world. But that is naive. I mean, it's utopian. I know that it costs money to digitize. And not just to do the scanning, I mean to produce the metadata to maintain it and so on, it's a very costly business. And Google has very deep pockets. So I'm grateful to Google for being willing to invest the money to make this happen. I don't see how anyone else could make it happen. Will it happen correctly and for the benefit of the whole public? Well, so far so good because despite the errors of things like Tristram Shandy that I was mentioning, the fact is that this great mass of books now from about 28 libraries is becoming available. But there are books in the public domain. So you can consult them today. You can do searches. You can do mining. You can do lots of things thanks to Google Book Search, which I salute. The question is, what will happen when we get beyond books in the public domain? And I don't know the answer to that. I'm a librarian here at the Humanities Library. I'll come closer to the mic, Marlene. Can you hear me now? I'm a librarian here, and I've been writing for quite a few years about the impact or the instability of the digital environment on libraries. And I've become increasingly unhappy with the instability terminology because I think we're talking about a number of different things and that if we don't distinguish them, that we're actually losing something. And having just read or reread your article in the New York Review of Books, it seems as if you're using the notion of instability to make an argument for continuity, historical continuity. But it seems like you're talking about two different kinds of instability, and I would like to add a third to it. So the first kind of instability I think you're talking about is information instability. And by that, I think what you're saying is that all we have are varying accounts. And none of them necessarily correspond to any particular reality. So whether it's newspaper articles or history books or whatever, all we've got is a variety of accounts. And that seems to be the way you describe the instability of information. But then you talk about the textual instability. And textual instability, my sense is that what you're describing is the fact that there may be many, many versions of any particular text, abridgments, translations, definitive additions, whatever. And you make that argument that, in fact, textual instability was greater prior to the advent of copyright because of piracy. So there could be scores of additions of any particular text in each one could be different. So that's the textual instability. But my sense is that digital instability is of another order. And that it has an impact on libraries unlike or a new impact on libraries that we have never seen before. And the way I would describe that is that if you catalog or create a bibliographic description for a printed object, no matter how difficult it is to do that, no matter how hard it is to describe that particular object, once you have described it, you don't have to worry about it changing. It will remain that object forever. Whereas digital objects are in, well, often, in a state of constant flux. And the e-journals and the databases that libraries provide links to are from one day to the next different objects. And so when we create links to these things, when we create bibliographic descriptions or whatever, we are describing the object as we know it on the day that we created that record or something like that. And so the situation for libraries right now is that they are providing hundreds of thousands of links to objects that they really cannot keep track of. And the only way, in fact, we know that we don't really have what we think we have is when students call us up and say to us, hey, it says here, you've got this journal from 1997 to the present. But in fact, you only have it back to 2001. Or you say you have this journal in the ProQuest database. But in fact, the only thing that's really there is an abstract. And so it just seems to me that the scale of instability when you're talking about libraries and the digital, it's just a whole other thing. And to talk about those other kinds of instability, in a sense, might mislead you into thinking that this change isn't quite as radical as it perhaps is. I don't know. Well, I like that argument. I'll have to think of the third dimension of instability. I think we've all intuitively had some experience of it. But you've had hands-on experience, even though you can't touch the metadata. You've had students come up to you and demonstrate the difficulty of locating texts that are constantly moving and sometimes move beyond our range to retrieve them. So I think that's a fair comment. And whenever you're dealing with historians, beware of specious continuity. It's a typical move that we make. We try to see something in the present. And we say, well, it was just like that in the past. You've got this false consciousness of how much today is different from yesterday. So that's pleading guilty. And it is also accepting the testimony of your experience as a librarian, which I think is important and something that I should learn more about. After all, I'm not a librarian. I'm a director of a library. But I don't have a library degree. Instead, I have a tremendous respect for librarians. And when someone like you tells me that my argument really was one of these historical arguments that sins in this respect, I take that on board. I'm not sure that I see a clear difference between what you call information and texts. After all, we just have texts as a way to get at information that was traveling. And you're dealing with texts in digitized forms as well, so that you could say that the same basic problem exists. It's a change in scale. And also, I agree in the nature of the information, because what's distressing about the tendency of digitized documents to disappear is not just that the O's and the ones can unravel, as I was saying. But that if the document definition isn't adequate, even if it's there in some server, it can't be found. That's what's terrifying. I mean, all of us, I think, have had the experience, if you have a large library, of looking for a book, and maybe you've moved houses, I just did, to come here to Cambridge. And now I can't find some of my books. And you could spend a long time just trying to locate it one. Well, the metadata in my head was inadequate. You multiply that by several million, and you've got your problem. So I sympathize with the problem. I was hoping maybe that you would say, in this third dimension of instability, that you had a notion of an answer. But if you do. I think you could base it. 10 years ago, libraries basically knew what they had. They had the card catalog, and they had records of journal receipts. So they literally knew what they had. And we're in the position where we subscribe to hundreds or thousands of journals. And we basically take it on faith that every month, those new issues are showing up. But are they really, I mean, it just feels as if we are forced to take so much on faith because there is no way yet devised to track, to yes, whatever. Yeah, it's an interesting point. Because you subscribe to a digitized version of an object or something born digital. And you can't check them all, can you? To know whether the text is solid or not. Thank you for the comment. That's very helpful. Hi, my name is Sans Fish. I'm a software engineer here at the MIT Libraries. I also work on the dSpace software and one of the site administrators here. I wanted to call out that one of the things that libraries are contending with right now are we're talking a lot about texts and about books and digitizing them. But data sets are, for instance, obviously we have a lot of them generated here are a big thing that libraries are going to have to tackle. So I think that deserves a little bit more attention. Especially because just for instance, I think there's a concern that a lot of data sets will be generated, a lot of experiments will be done. But you'll get the final data set and that will be archived. But there's a whole temporal continuum there that doesn't get recorded. But I wanted to know what your thoughts were on, because I know Harvard's kind of embarking on their own dSpace instance for theses. And I believe it's possibly the policy that you're going to require digital submission of theses into the system. I'm not sure. But I wanted to know what your thoughts were on that born digital material and stuff that never actually gets printed out and gets archived before it materializes. Let's find dSpace for those members of the virtual audience, if not our living audience, that don't know what it is. You're the library. Well, I'm not a librarian, as I said, I wish I were. You're the director of libraries. Well, dSpace was invented at MIT. And it's part of your open courseware. It's a way of communicating all sorts of things that go on on your campus. And it's been a terrific success worldwide. When we at Harvard last February had a debate and then a vote in the Faculty of Arts and Sciences on the issue of making scholarship openly available free online, we thought of MIT naturally and dSpace, because it's one thing to require that faculty submit an article, a scholarly article, to a repository. It's another thing to actually make it happen. So we had to design a system. We had to build a whole system for making it happen. And we used a modified version of dSpace. So we got a lot of help from MIT. And we now have built what we think is a fairly robust system for storing and preserving and communicating all of the scholarship produced in the Faculty of Arts and Sciences today and indefinitely into the future. We're in a beta phase of it. And I've actually submitted some of my own articles to it. And it's fairly user-friendly. So I'm cautiously optimistic that we will be able to persuade some of the people in the far reaches of comparative literature and philosophy and art history who are not familiar with computers to use this system to store their articles and so that we will populate the repository with the great bulk of the knowledge that is produced year in and year out. This may not be an adequate answer to your question. Now let's come back to your question about what's lost. If there's any trepidation in your community about the theses being born digital and never actually materializing and being archived and then put in. But are you experiencing your resistance to? For theses, for dissertations? For theses or anything that you're going to put in these books? Well, there was a big debate in the faculty. And as you've probably heard, it was a unanimous vote. I think it may be the first unanimous vote that the Faculty of Arts and Sciences at Harvard ever took on anything. So I got the sense there was a lot of enthusiasm for it. There were worries at the same time. And these were talked through. There was a major attempt to study the question in all of its aspects. And one of them was the worry, I guess you're referring to, namely that it might be harmful to some journals or scholarly societies to have these articles available free when otherwise subscribers would have to pay for them. There's another worry, and that concerns graduate students who are writing dissertations. We now are planning to put all dissertations online for free. Well, you might feel quite unhappy about that if you've produced a dissertation that you want to turn into a normal printed monograph. How are we going to deal with that part of our constituency? The answer, basically, is in each case to have opt-out provisions. So for example, if you're a student who is hoping to produce a monograph on the politics of Colombia, you can, it will be digitized. It could be born digital. And it will be stored and preserved. But it will not be communicated for x years. So we will have an embargo to protect the graduate students. But at the same time, we have to work very hard to make sure that these born digital documents, whether they are scholarly articles by the faculty or dissertations by the students, that they don't get lost. And as you know, no one's solved that problem. But we have, I think, a very good Office for Information Services that has developed all kinds of techniques for migrating them from one server to another and testing them and preserving them. So I think that we are being responsible in that respect. I have the feeling that we didn't really answer your first question. Do you want to restate it? Because I think I thought it would be good. OK. Kevin. I am a Kevin Driscoll graduate student here at CMS. And I deeply appreciated that you acknowledged the role of the smuggler, the compositor, the types that are in the entities that we know as Shakespeare or Voltaire. And I wonder how that expansive notion of authorship complicates the project of creating an archive or doing history. Sometimes it's very convenient to have a single author when you're writing history. But of course, as you noted, there are many, many different human beings who were participating in the creation of a text. Yeah. Well, I might have to lapse into autobiography again, because I'm now attempting to write a very large-scale electronic book myself. And it's going to be about the world of books in the 18th century. But the information is so rich, the archives I worked in contain 50,000 letters of, as I said, smugglers and booksellers and authors and publishers. It's everyone connected with books. They're so rich that if you want to get across the richness in its fullness, you can't do it in a conventional book. And so what I'm trying to do now is to write a book about the whole thing. But as a conventional printed monograph, I hope it won't be too long. But then to have an electronic supplement in which I will have various layers. Some would, a top layer might be a sort of mini monograph series of short essays, monographs on aspects of the subject. And the subject, as I say, goes on forever. So you could write a lot of mini monographs out of 50,000 letters in an archive and other archives as well that are related. But then a layer below that could have selections of the documents in English. And then a layer below that, the complete set of documents in French transcribed and tagged. And then the bottom layer could be the actual digitized manuscripts, because when you transcribe a manuscript, you're always interpreting it. There's always something that is imperfect about it. And I would imagine a reader reading, so to speak, vertically through that material, finding his or her own path. And then being able to print out what he or she wants from it in one of these espresso like book machines. I don't know if you've run across this, but we now have the beginnings of an industry in which you can download a digitized text, print it out, and have it bound in a matter of minutes, and not very expensively. So it could be that my future readers would take this electronic aspect of the book, and each reader would develop his or her own paperback that could be produced instantly. And then the bibliographers would say, well, we've got dozens of different books. I think that's great. Maybe bibliographically it's a headache, if any bibliographer will ever be interested. But it empowers the reader. It makes a kind of collaboration between the author and the reader. It gives my readers the ability to disagree with me, to argue back with using evidence that I myself have supplied, and that there might be a kind of wiki aspect to it in which we would have running debates. I haven't quite decided that. So I think that one of the excitements about the new kind of electronic communication is the ability to get across not just more material quantitatively, but to get across material that enables the reader to select and choose and use this material in different ways. And this is only, in this case, manuscript material, but I'm dealing a lot now with songs, as we mentioned. And there, you have to hear it. So electronic books can make it possible for you to hear songs as they were sung in the streets of Paris 250 years ago, and to see to what extent the actual music communicates meaning. That's a fascinating thing. And that sort of thing has slipped through the fingers of historians because in an oral society, most oral communication just disappears into the air. But now I think we have the beginnings of a possibility to capture some of that. Again, it's like the third dimension of the digitized things. So much is going to get lost, but we can capture some. So I'm a real enthusiast about the possibilities of electronic scholarship. And I think we are going, we're in the era where we are creating new kinds of books, new kinds of reading, and that authorship itself is becoming new. Hi, I'm Sam. I'm an undergraduate at Harvard. And I have a question that begins at kind of the other end of a lot of what's been discussed. I'm imagining a future where the 25,000 email messages I already have in my Gmail account, in addition to the probably several hundred that have been generated in this room over the last hour and a half. And who knows how many are out there are all being captured that that technical problem has been solved. And I'm curious how you think that's going to change the nature of the way we use that information. Obviously, a part of libraries is selectivity or even simply popularity as determined by the survival of their materials. Only the best has come down to us, perhaps, from several thousand years ago. I mean, most of the stuff I have is junk. But presumably, in kind of the catch-all world that this new technology is making possible, everything can be stored. I mean, one way you could say it's a historian's dream, but it also very well could be a historian's nightmare. And then kind of as an added proviso to that, I wonder if that changes the way we value information. Wikipedia is great. It's also a collection of abstracts in summaries. And whether that will kind of change how deeply we go into something, say, like your multi-layered book. Like, who would want to look at the second layer when we have 10 billion of these? Right. You know, I don't think I have an answer to that question. I mean, it's a very good question. But in a way, you're saying, how in the future will this surviving information be used? I'm bad enough at being a prophet of the past, not to mention prophesying about the future. So I don't know. I mean, I think you can imagine trajectories from current usages. And there you get into things like relevance ranking. Will we have better relevance ranking than we have now? Will new developments, better algorithms be developed to empower you, the person who is going to use this totally preserved information? It seems to me that's likely. And that there will be not just Google's relevance ranking, but many, many different kinds that will be adapted for different purposes. So that instead of drowning in an ocean of information, that's a sensation that goes back to antiquity. But certainly in the 16th century, people were saying, we're drowning. We can't read everything. It's hopeless. Let's just give up. But again, there's a danger of specious continuity. We really are drowning today. And if you are preserving all this ephemera, as you said, what use will be made of it in the future? There, frankly, my answer is I don't know. But it makes me feel good to know that it is being preserved. Because I think often ephemera can give us an entry point into areas that are totally new and surprising. So I found that to be true when I studied the 18th century. I could give you another commercial. I've discovered a novel in the 18th century that I think had disappeared. And I think it's a great novel. And it was preserved in a very small provincial library in Rouen, as it turns out. And now I'm getting it translated. And it's going to be, I hope, assigned to undergraduates. Maybe you would read it. I hope it's coming out next year. But that's an example of how there's a lot out there that we can still recover that has thought to have been lost, but really is not just recoverable, but very important. Well, it's not an answer to your question. I'm sorry. But good luck with it, as you use all of this, swim around in this ocean of the future. Over here on the right, do you have a question? I'm Ellen Farron. I'm the director of the MIT Press. I'd like to start by saying that when you're ready to talk to a publisher about the book you're working on, we're ready to talk. My question is stepping back, actually, to say, can you comment on the impact of word processing on the history of the book? I mean, as someone who remembers Whiteout, I really appreciate the way the ease of revising things makes me think better, and I hope right more clearly. But we're very aware, as a scholarly publisher, that it also allows for longer and longer manuscripts, more and more easy cutting and pasting, sloppiness in some cases. Sure. Well, first I would like to pay homage to the MIT Press. I love the MIT Press, and I think that you have a great record, a great backlist. You brought Bakhtin and others to the English Reading Public long before other presses ever heard of such writers. The history of word processing is really interesting, and it's not written, but from what I know about it, partly from having experienced it, I would agree with what you're saying. It makes us verbose. It also gives us the illusion that we're writing well when we're not, because things look so beautiful on the screen. They're justified. It's as if you've got a final product, and I think it creates laziness in writing. Of course, the old form of writing, either by pen or on a typewriter, produced bits of paper with signs on it, and you had to transform them in a second draft, in a third draft, in a fourth draft, so that, of course, I go back a long time, and that's how I certainly wrote my PhD thesis. It was typing it into a typewriter, and then having a pile of paper, and then retyping the whole thing, which forced you to rethink every sentence. One way to perhaps mitigate the danger of verbosity and laziness is to compose on the computer, or use word processing, to print it all out, and then to have your printed sheets there and to rewrite the entire thing, instead of taking the lazy route and just adjusting it here and there. But how many of us do that? Well, I don't know. And certainly editing, maybe not at the MIT press, but in general, editing has declined as a wonderful craft, and for good reason, because it's so expensive. So that many publishers now job out editing, and the in-house editors don't do line editing. It's very important to give general comments about a work, but not the line-by-line, close editing that really can improve a text. So yeah, I think that quality has often suffered as a consequence of that particular improvement, if you want to call it, in the technology. It reminds me of a study of Henry James's novels. It turns out that Henry James switched to writing on the typewriter instead of writing things out by hand somewhere in the middle of his oeuvre, and that sure enough, those long Jamesian sentences correspond with the shift to the typewriter on the part of Henry James. I don't want to make too much of that, but I think actually the rhythm of prose and the length of sentences and that sort of thing is, to a certain extent, at least influenced by the technology. My name is Peter Walch. When Diderot was describing his project for the encyclopedia, and I'm paraphrasing what he said, but his basic description of what he was doing was taking a digest of the best knowledge available to civilization and preserving it in a compact form, in case civilization collapsed again, as it had done in the past, that the best things of knowledge would be preserved. And I think in a way that's a lot of the history of civilization is assuming that knowledge is fleeting and fragile, and it's going to be gone unless you pin down the most important points. And you see that in the New Testament and all sorts of documents past history. And what we have now in the digital age is this idea, we will save everything. We won't make judgments. We won't try to pin down the most important things. And I just wondered if you had any thoughts about how to reconcile those two different approaches to preserving knowledge. It's a tough question. It's quite related to some of the earlier questions. I mean, I've been arguing that we have a responsibility to preserve as much as we possibly can, and I believe that. But I accept the point you're making, which is, well, does that mean that we will not discriminate between some things that are more important than other things? My return remark would be, well, we may have a clear idea of what's more important. But 50 years, 100 years, 200 years from now, the ideas could be very, very different indeed. And there are many examples of that process of reevaluating things from the past actually taking place. For example, Shakespeare's courtes. I mean, Shakespeare didn't even think it, but we don't know what he thought. But we guess that Shakespeare didn't think it was worth the trouble of being involved in the publishing of his plays. His poetry was different. I mean, there, he really did care about it. And he put his name on it and got it published. But not his plays. That was ephemeral literature of the time sold in the streets at a very cheap price. It sold pretty well. We have evidence of that. But it wasn't taken terribly seriously. And only later does it turn out to be that which we take most seriously. So I actually think that, well, another example would be the study in France of something called the bibliothèque bleu. These were chapbooks. And most of them were read to pieces, thrown away, disappeared. But now French scholars are seeing chapbooks as crucial sources of information about the kind of cultural exchanges that took place among peasants and ordinary people. So what seemed ephemeral three or 400 years ago now seems actually very important. Or ballad books, for example. And Ireland are very important. And a librarian in Ireland, I think it was in Trinity College Dublin, pulled open a drawer and found five ballad books from the 17th century. I think they were from the 17th century that no one had ever known about before. They were unique editions. And this was a great event. But at the time, in the 17th century, a ballad book is trivial. So maybe it's not an adequate answer to what you're saying. But I do feel that we are not capable in absolutely deciding what is the most important and most worthy of being preserved in the vast amount of things that exist today. You might make an opposite. The history of popular cultures suggests, interest in popular cultures, suggests that the argument is opposite. The things that you think are throwaway items become the cultural artifacts that contain the culture's signature. Yeah, that's what I meant. But what I was trying to get at is that the role of libraries and depositories has been to present a certain view of the world, which has then been able to codify a certain process of information. For example, the Folger Shakespeare Library collected multiple copies of the folios. And because he had that fairly eccentric idea of what was important in his collection, a whole new realm of scholarship came up because people discovered the folios weren't all exactly the same. So you could check on the typos and so on. That's been historically true of libraries is that people will have some kind of obsession, a private obsession, an institutional obsession for a certain kind of literature. For example, the Arabs, they were interested in Aristotle. They translated a lot of Aristotle and preserved it. That that kind of process of sorting through information has preserved civilization on kind of a metadata level that you don't get when you're just trying to preserve everything, if that makes sense. I'm not sure I agree. I think that it's important to preserve as much as you can because we don't know what will turn out to have been significant. But in saying that, I guess I'm not disagreeing with the fact that there was very high selectivity in the past. And that meant the past is in that respect different from the present. And the assumption was you couldn't collect everything. So people were making conscious decisions and that what we know about the Greeks is what the Library of Alexandria collected and passed on to the Romans. No, I accept that as if you like a qualitative difference in the quantitative dimension of what a civilization is able to preserve. But what to do with it, that gets back to this earlier question. And I think that we have a responsibility to preserve as much as we possibly can. But that doesn't mean that some things aren't more important than other things. What it means is I don't feel that I should be a kind of intellectual policeman who will say, OK, we're going to preserve this, but not that. I would rather try to preserve as much as we possibly can and leave it to future generations to explore in this vast world of the preserved what they themselves find to be most significant. The kind of eccentric or specialized aggregation of information that you describe, Peter, is not precluded if you have everything. There'd just be a lot of weird aggregators who will do the job that you're talking about. It doesn't seem to me that there's a conflict, that if everything, theoretically, if everything were available, there would be more folgers who would come in and do their thing out of the vast ocean of information that was there. I think we're in a new age, and I think that's entirely possible. I just think, if you look at the history of libraries, the history of collections, people are constantly, they're not saying, we're going to collect everything. We can't collect everything. They say, we'll collect these 20 newspapers because they are important to us. And we'll collect bibles because we're a Christian institution, or we'll collect texts from ancient Greece because ancient Greece is the foundation of Western civilization. They weren't saying, we don't know what we want to collect. We don't have an idea of civilization. We'll just collect everything and hope later on someone will sort it out. They were making very, I think in every library, they were making daily decisions. Although the notion of a deposit library goes very far back in the past. That's true. And certainly in Britain, the idea was that deposit libraries should have everything. The same was true actually in the beginning of deposit libraries in France, although they were royal libraries. But in principle, any book that was accepted for publication legally had to be deposited in the King's Library, et cetera. So I think there was, I accept your point, but I do think there was this notion of trying to preserve not just the peculiar things that had a particular interest for a particular collector, but also the totality. It's just that they didn't have the means to do it. Final question. Sorry, I have a quick, easy one, which is directed to you in the context of being a steward of one of the largest storehouses of information we have now, and presumably a much larger one in the future as all this data comes pouring in, which is to what this is addressed to the technical side. How comfortable do you feel as a meta-owner of those assets with the degree of security and robustness and reliability of the digital systems you're using in terms of Harvard specifically or anywhere? I have to confess that I'm not a techie. I mean, I still use, believe it or not, index cards and write on them when I do research in the archives. And I'm so ignorant that I don't appreciate the fragility of the technical world that, in principle, I'm in charge of. So I'm not ducking the question, but I'm pleading guilty. The answer, I think, for me, because I'm the director of more than 100 libraries, and I don't actually administer any of them. I mean, I try to provide some sense of direction. It gives you a lot of deniability. So yeah, I'm denying it. I'm pleading guilty. But the point is to find the best techies you can find and to supervise them and to call in outsiders to evaluate the job being done and then to have faith in them. So that's my answer. It may not be an adequate one, but I should probably lose sleep thinking about how you could have a systems failure that would permeate the entire system. Is it something you're nervous about, even in the abstract? I mean, do you have confidence that the fine people working on these problems are solving them for you as a historian and future historians, or do you think this is really something that we should be very, very worried about? I have to admit I haven't lost any sleep over. I probably should have. When I was in grade school, we were trained to react to a nuclear blast. And what we were supposed to do was to, there was a whole technique. You jumped under your desk at school and you put your hand over the back of your neck because you saw the flash. And you knew you had maybe five seconds between the flash and the blast. And I had recurrent nightmares of this when I was throughout my childhood. I also had nightmares of Hitler coming in the window since I was born that far back. And so those are the kinds of things that interrupt my sleep at night. Nuclear disaster in Hitler. Now, probably, I should wake up and it is sweat worrying about the digital preservation. But I have to admit I don't. And I'm a great admirer of the people in the Office of Information Services that are part of the Harvard University Library. So maybe it's lack of imagination on my part. I'll take that as a vote of confidence in digital storage. Thank you very much, Bob. Thank you, audience. Thank you.