the riddle of the library

(originally written for Wired US in 1997, this article was never published; a heavily edited version was eventually printed in the Daily Telegraph's "Connected" section, 04.98)

It was, I thought, a straightforward assignment. Exciting, too, if you like that kind of thing (and I do). The Vatican Library, possibly the greatest ancient manuscript library in the world, had teamed up with IBM in a bid to make its archives available on the Internet. The story had tremendous potential. Of all the criticisms made of the net, the one which really stings, the one which comes up again and again, is that it has no depth, that its information is too random, too contingent. But if an institution as venerable as the Vatican Library can dare to make its contents available online, surely this is the kind of of seed from which a new stratum of the net could grow, a library stratum, a realm of properly catalogued, in-depth resources, where serious research of traditional kinds could be done in a traditional manner using the most modern technologies. Surely this is the as yet unrealised dream of many a scholar?

Go to Rome, said my editor, meet the folks involved, do the story. Classic Wired. Open and Shut. Or so we both thought. But we had reckoned without the labyrinthine politics of the Vatican - and, for that matter, of IBM. One of the things the net is often praised for, is that it breaks down hierarchies. But as I was to find out, venerable institutions, whether Big Blue or Holy See, are very proud of their hierarchies and the structures and strictures that are implicit within in them.

To begin with, everything seemed straightforward enough. There were two key men involved in the project: Father Leonard Boyle, an Irish priest of the Dominican order who had been the prefect of the library since 1984, and Fabio Schiatarella, the link man at IBM Italy. In early May I telephoned both of them and arranged to meet them both at the library in June. As soon as it was all confirmed I arranged a hotel and booked my flights. I was off for a week in Rome. Nice.

As a link between libraries of the past and philosophies of the present I took the opportunity to read The Name of the Rose, Umberto Eco's famous novel about dark goings on in a fourteenth century monastery. It was a good choice - or perhaps a bad one. I was soon embroiled in tales of Machiavellian monks being killed by a poisoned copy of Aristotle's Poetics, which is cunningly concealed at the heart of a library the structure of which mirrors that of the world itself. My imagination was stirred yet further as I began to discover a little about the Vatican library itself.

One of the most valuable text archives on the planet, the Vatican library houses 150,000 ancient manuscripts and 1,500,000 printed books, 8,000 of which are incunabula, or books printed before 1500. This place has its own museum of 19,000 objects; it also has a coin collection of 330,000 coins. It has a 100,000 prints and drawings. It has documents from virtually every civilisation that ever produced written records. It contains the oldest extant Bible manuscript, written in AD350 in Constantinople, and some of the earliest surviving examples of copies of works by Aristotle, Dante, Euclid, Homer, Virgil and Ptolomy. Its printed books include two copies of the Gutenburg bible and Platina's De honesta voluptate, Europe's first printed cookbook, acquired in 1475. Poetry, music, art, medicine, history, science, law, literature, geography - the library has it all.

But most important are the manuscripts - all of them are unique in themselves and therefore unique to the library. When I arrived on the scene, IBM and the Vatican Library had just completed a pilot project exploring the feasibility of scanning these and making them available online. Of course, this was not the first time that IBM had been involved in such a venture - their digital library project is now a vast operation that spans the globe. Operations are ongoing in Seville, Spain, to scan an archive of 8 million images which include Christopher Columbus's personal records (for example the letters he wrote while travelling to his son, Diego); in St. Petersberg, Russia, where a team is beginning to put the Hermitage museum's enormous archive of 3 million objects online; in Osaka, Japan, at the the Museum of Ethnology; and at the Library of Congress in Washington D.C., where IBM is beginning to attack what is probably the world's largest collection of objects, some 115 million all told (James Billington, the library's director, wants 5 million of these online by the year 2000!).

Add to that the digitisation of the audio files of the BBC in London, of EMI digital publishing, of several German radio stations (where many archive recordings have deteriorated to such an extent that they can only be played one more time) and the libraries of many universities and other institutions, and the entire IBM digital library initiative adds up to an incredible 50 exobytes of data (that's 1018 bytes).

It's fair to say though that in terms of scholarly value, the Vatican Library project is right up there with any of these. According to Willy Chiu, the director of the IBM Digital Library Initiative, "some of the most treasured pieces have not really been touched or catalogued." When Chiu first visited the Library's vaults, Michelangelo's drafts for the design of the Sistine Chapel were "lying on a shelf." And now, with one eye on the preservation of such documents, many of which are deteriorating with age, and one eye on making its incredible contents more widely available, the Vatican Library is going online.

To see if I could get some background information and maybe hook up with some other people in the library I decided to phone the Vatican press officer, an American woman by the name of Majorie Weekes who was not, as it turned out, the world's most forthcoming of PRs. "That's 'Weekes' with an 'e'", she told me - witheringly - down the phone, after insisting that I express mail some samples of Wired to her and a formal request for interviews with Boyle and Schiatarella. "B-But I've already set up the interviews," I protested, "all I'm asking you for is some more information on the library." "We'll just have to see if it's possible," she muttered, ominously.

I called Marjorie Weekes back the next day and she said a decision hadn't yet been reached, but her manner was so curt that I decided not to speak to her again in case she took a real dislike to me and tried to jeopardise the project. But events seemed to have a momentum of their own, and that afternoon I got an email from Father Boyle. "I regret that I shall not be able to meet you," it read. "My resignation after 13 years took effect on Saturday last, 24 May. There is now a new prefect. I have nothing more to to with the place or the job."

Er, hello? Suddenly I thought I understood why Weekes had been so unhelpful - there was obviously some political storm raging inside the Vatican, and one so violent that the prefect himself had been ousted. The story was suddenly hot. I tried to call another priest I'd spoken to randomly a day or two earlier - a Father Allan Dustin, the man in charge of a digitisation project going on in the Vatican museum - but he had left for a conference in Florida. His assistant directed me to a Father Sheehan in the library - perhaps he could help. But I couldn't reach Sheehan and enquiries at IBM Yorktown, where the research labs which built the scanners being used in the library project were located, were also proving fruitless. I emailed Father Boyle back and told him I'd already booked my flight and would he please reconsider. He said he appreciated my predicament and told me to call him at his new church, San Clemente, situated a couple of miles from the Vatican and near the Colosseum, on my arrival in Rome. For the time being at least it looked as if the story was saved.

Meanwhile, I was finding out a bit more about the library itself. It has quite a history. The initial collection of what was to become the Biblioteca Apostolica Vaticana was begun by Pope Nicholas V in around about 1450, fifteen years before the invention of print. An accomplished scholar himself, Nicholas had a humanist as opposed to an ecclesiastical vision of preserving the learning of what we refer to today as "classical antiquity", and to this end he sent envoys off all over Europe to hunt down ancient manuscripts and rare texts and either acquire them outright or have copies made. His efforts produced an archive of 1,100 manuscripts which lay relatively undisturbed until Sixtus IV became Pope in 1471. Sixtus restructured and rejuvenated the library, encouraging the then librarian Cardinal Giovanni Bussi, a great scholar in his own right, to expand it, and over the next few centuries the library grew and grew, swelled by collections purchased outright, personal libraries made over to Vatican in the wills of nobles, gifts of manuscripts and books to the pope, and the often less than subtle acquisitions made during the Crusades against the Turks in the fifteenth century and through the operations of the Inquisiton, then and later.

Although Sixtus IV had planned a separate building to house the library, it stayed where it was until Sixtus V managed to get one built nearly a century later. In 1587 the library moved across the courtyard into its new home where it has remained, in one configuration or another, ever since. In 1605 a fire damaged part of the collection, but the library continued to grow throughout the 17th and 18th centuries thanks to an endless series of acquisitions and gifts, including a particularly notable collection donated by Queen Christina of Sweden in 1690. During his sack of Rome in 1797, Napoleon walked off with much of the library, although most of it was returned some twenty years later, in the aftermath of Waterloo and the Treaty of Paris. The thought of these manuscripts - some of which were already hundreds of years old - being ferried to and fro across Europe by creaking coach or leaking ship is less than pleasant.

But while the collection got through it looked as though my story was at the very least going to be seriously diverted. Ten days before I was due to leave, I'd decided to brave Marjorie Weekes once more. Big mistake. This is how the conversation went:
"Hello, Ms. Weekes? It's James Flint here from Wired."
"We're not going to be able to do anything."
"But I've already spoken to Mr. Schiatarella…"
"I've spoken to Mr. Schiatarella. We're not going to do anything. Mr Schiatarella won't be able to meet you I'm afraid."
"Is there anyone else I can speak with?"
"We don't deal with that."
"What do you deal with?"
"We deal with visuals."
"So I'm talking to the wrong person then. You have nothing to do with interviews?"
"Well, yes, you have to come through us."
"But I though you were just visuals."
"We're audio-visuals."
"So do I talk to you to get an interview or not?"
"It may be something to do with us, I don't know."
"So you won't help me out."
"Thankyou Mr. Flint. Goodbye."
And that was it. She put the phone down on me. Immediately I called Schiatarella, but he stone-walled me too. Weekes must have got to him.

This was obviously bigger that I thought! Who was this woman who could pull the strings of the mighty IBM? How awful could the scandal be? As I called my editor I idly wondered to myself if the Pulitzer had ever been awarded to an Englishman…

The Wired machine quickly went to work, nudging IBM US for more information and trawling Italian sources for tales of rum-doings in the Holy City. We came up with some rumours and some good gossip from the past - a tale of a trusted academic who had stolen some valuable manuscript pages, and the story of the Holy See website which had been taken down due political in-fighting almost as soon as it had gone up, and remained "under construction" for an entire year as a result - but nothing concrete about the current situation in the Library. My editor told me I should just go to Rome, try to contact Father Boyle, and go ahead regardless on a story about the digitisation. Oh yeah. Almost forgotten about that in all the excitement.

So I flew to Rome. Just before I left I'd managed to reach Father Sheehan, who agreed to try and set up a tour of the Library for me - I was to phone him on arrival. The story hung in the balance. It still seemed quite likely that I would get to Rome and Boyle and Sheehan would both let me down. I'd be left with nothing to show for it but a nice tan, a paunch from too much pasta and a head full of Umberto Eco-style conspiracy theories (worse things have happened, I thought to myself).

Things looked particularly bad when, on the afternoon of my arrival, I couldn't reach any of my contacts. The next day - when I finally got hold of Father Boyle and he tried to deny me an interview for a second time, only reluctantly offering the slim possibility of a brief meeting on the forthcoming Saturday - they looked considerably worse. I read more Eco and became more convinced than ever that a grim secret lay at the heart of the Vatican Library. But I was preparing myself for a week full of nothing to do (thoughts of that Pulitzer slipping away…) when my luck changed - on the Wednesday morning (I had arrived in Rome on a Monday) I managed to get a call through to Father Sheehan and he told me he'd set it all up, could I be at the Library in an hour? I grabbed my passport - the Vatican of course being a separate country - and set off across Rome.

In some ways the Vatican is the ultimate city block, sort of like the World Trade Centre of the spiritual economy. At its public entrance, the Santa Anna gate, you have to pass two sets of guards: first, the ceremonial ones, dressed in blue hose and stockings and big ornamental berets; and then the internal police, who dress in contemporary trousers and shirts. It's the internal police who deal with visa applications, in a small hot room off to the side of the single road leading in and out. But apart from the police and a certain amount of bureaucracy there are no barriers to entry - no gates or wire grilles. After all, as countries go, the Vatican is pretty small. Inside are a total of about three short roads. If you break in, you haven't got far to run.

Once past the guards you immediately enter a complex of buildings and courtyards abutting the enormous fortified wall that runs along the back of St. Peters and encircles the city-state as a whole. Mostly constructed out of the light brown "tile bricks" so characteristic of Italy, here are the offices of the Holy See, the private apartments of the cardinals and the pope, the Vatican Museum (accessible from outside), the Vatican Archive (essentially a public records office) and of course the Vatican Library. In addition there's a supermarket, a drugstore, and a gas station. All of the latter are immensely popular with those who work here, since prices are kept artficially low. Most popular of all is the drugstore, because there staff can buy new drugs from America that have not yet been made legal in Italy proper. Of course, when I tried to go in for a poke around I was promptly seen off by the guard. I had a pass for the Library alone, and was not supposed to go anywhere else.

Sheehan turned out to be an amicable Irish American priest with short cropped white hair and a round face. He met me in one of the entrance halls and led me up a marble staircase and along quiet white corridors to the library itself, housed on various levels inside of the main square of buildings built around a dull courtyard which doubles as a carpark. Sheehan has just completed a catalogue of the Library's collection of books printed before 1500, the incunabula, a task that has taken him eight years. Cataloging has always been a problem for the library. It didn't have any kind of comprehensive catalogue until 1928, when the Library of Congress and the Carnegie Foundation [NB - maybe it's one or the other, I don't understand the relationship between these two institutions] was called in to help build a card catalogue, and even then the manuscript archive wasn't included. Today, the manuscripts are still not completely inventoried and Sheehan points out that in fact that's one of the reasons for the popularity of the library amongst scholars: "there are still exciting discoveries to be made here."

He took me through the reading rooms - one for printed books and one for manuscripts - whose long tables were busy with people perusing ancient texts on sallow pages. The ceilings - like most of those in the Vatican - were covered with frescoes, and the adjacent book stacks stretched off so far into the building beyond that I couldn't actually make out where they ended. I was surprised by how small the reading area was by contrast - both rooms were full and there couldn't have been more than 150 people here - even though I already knew that limited space for scholars was one of the main reasons that the Library had embarked on the digitisation project in the first place. Only about 2000 readers cards a year are issued, all of which go only to serious scholars. This kind of "processing power" might be adequate for an ecclesiastical library, but although the Vatican Library is traditionally the personal library of the Pope it is ultimately a humanist library, and was conceived as such from the start. There is a great demand for access to its manucripts and the library tries its best to meet that demand quite apart from any religious function it might have. It already has a photographic department which makes microfilms and ektachromes of manuscripts and books to order, but requests can take months to process.

The original inspiration to do something more came from Latin America. Obviously, the Catholic church has a huge presence there - more than half of all Catholics will be Latin American by the year 2000 [Bernstein & Politi, His Holiness, Doubleday 1996] - and one effect of this is a large scholarly community who need access to the resources held in Rome for their researches. "To look at just a couple of pages of these manuscripts can take weeks at the moment," says Lourdes Peña, PR manager for IBM software in Latin America. "Not only do scholars have to travel to Rome, but because of the Vatican Library's space restrictions they also have to be highly qualified." It's not simply a matter of religious ties, either. Many of the Vatican manuscripts date from around the time of the European discovery of the Americas, and some of them originate from pre-Columbian civilisations, making the collection invaluable for modern Latin American historians.

It should come as no surprise then that four or five years ago, faculty members at the Pontifical Catholic University of Rio de Janeiro (PUC-Rio) began to wonder if it would not be possible to put at least the most heavily demanded portions of the Library online. Their initiative coincided with corresponding dynamics within IBM and the Vatican Library and the three interests set up an informal partnership, which worked because while the Library had the manuscripts and IBM had the technical know-how, "PUC-Rio had the staff and researchers, and was the kind of institution needed to close the gap between the Vatican and IBM," according Peña. Out of this collaboration a pilot project was born, which was completed earlier this year [1997] and which involved the scanning of some 60 manuscripts, a total of around 20,000 pages.

To see the business end of the operation, Sheehan took me down in a small lift to the manuscript vault, the only area of the library which is airconditioned. Inside, the vault's about the size of a football field, and with three-foot thick walls the collection is about as secure as it could possibly be. Apparently it's atom bomb proof, though I'm not sure I believe that - it's at ground level, after all. Radiation resistant, perhaps.

By the enormous reinforced entrance door were the scanners and their operator, Irma Schuler. A German expert in handling and microfilming manuscripts, Schuler - who wears white cotton gloves at all times - is continuing to scan manuscripts even thought the pilot project is over so that the machines won't lie idle while the powers that be decide on the structure and timing of the project's second stage. She showed me how the scanners worked and the images that they produced, which are of an almost unreal quality. Parchment and vellum are made from animal skin, usually sheep, goat or calf. The skin is washed and lime is used to remove any hairs. It is then pinned out and scraped smooth with pumice stone. Umberto Eco describes this process in his book: "Nearby I saw a rubricator, Magnus of Iona, who had finished scraping his vellum with pumice stone and was now softening it with chalk, soon to smooth the surface with the ruler. Another, next to him, Rabano of Toledo, had fixed the parchment to the desk, pricking the margins with tiny holes on both sides, between which, with a metal stylus, he was now drawing very fine horizontal lines." [Umberto Eco, The Name of the Rose, Picador, 1983, p. 185] All of these marks - the pin holes, the stylus lines, are picked up by the scanner and visible on screen, as are the ripples in the parchment itself which result from stretching and drying.

It's not until you see the manuscripts themselves that you realise why you can't just slap them on a flat-bed scanner and stick them on the net. The first problem is the sheer size of many of them, often up to 24" x 36" and 3" thick when closed. The parchment pages don't lie flat like paper: they wrinkle and curl and seem to fight one another for space in the bindings, which are usually made of heavy, tooled leather or even of wood. But apart from being extremely bulky they're also extremely delicate, sensitive even to light and heat let alone manhandling.

In order to be digitised these tomes must be placed in a special supportive easel and scanned from above. Because of the thickness of the pages, every time a page is turned the camera must be manually refocussed because a millimeter or more has just been added to the focal length, making it a very time consuming process (Schuler estimates that it takes her a week on average to scan one manuscript.) The camera has to have a large depth of field so that it can keep in focus all the details of each page despite the undulations in the surface of the parchment. Even then the pages often do not lay flat enough, and a piece of perspex has to be laid across them to keep them in place. And the entire operation has to take place in a special curtained off, humidity-controlled area positioned as close to the manuscript vaults as possible, where there's only enough room here for two scanners to be set up, and be permanently overseen by a Library official who ensures that Schuler's position with regard to these priceless objects can never be compromised. As Father Boyle was later to remind me, the Vatican is still very much a patriarchy, and women are still regarded as inherently untrustworthy by the hierarchy (something to do with Eve and an apple, I think).

The scanners themselves are extremely impressive. They have a resolution of 3000 by 4000 pixels, and although the software which drives them is currently only able to register 2500 by 3000 of these even that can produce an image up to ten times as fine as one produced by traditional chemical photography. Developed in IBM's Yorktown labs, the scanner scans each line of 4000 pixels thirty-two times, an approach which allows the device to filter out noise almost completely. A single page takes about 7 minutes to scan in 16 million colours; the resulting image is about 14 megabytes in size. This means that you can blow-up a 1 inch square detail to fill a 19" monitor screen without any distortion - something which is impossible to do with a magnifying glass.

Schuler explained to me that one of the problems that palaeographers face is that over time the heavy iron content of the inks used in the scriptoriae eat into the parchment in an acid-like manner as the centuries go by. Because most manuscripts are written on both sides (although they're numbered on only one) the result is a kind of dark shadow behind the text you are trying to read, called "bleed-through". IBM have developed a technique which can process the image and remove this effect on screen. Father Boyle had originally seen this software at work in the Seville operation, where it was being used to remove blotches from those letters from Columbus to Diego, making some of them readable for the first time. The rather disconcerting result of this is that, at least as far as readibility is concerned, the digital copy is actually better than the analogue original. As Sheehan is quick to point out, "nobody's ever going to be happy if they're really interested in a manuscript and they don't see that manuscript - there's something about the feel, the smell, the aura of it that's part of the package." Later, when I finally spoke with Boyle, he was to agree: "there's one thing [the technology] will never, ever do - it will never give you a manuscript or a book in your hands. It makes available the thing but it is not the thing." Nevertheless, by making making details - such as marginalia and esoteric abbreviations - properly legible for the first time, the digitisation project promises to make a significant difference to the main problem faced by the palaeologist: reading the text.

With all of this technology and effort going into it, the pilot project was not going to be cheap. Willy Chiu estimates that IBM's costs so far are around the US$3 million mark, and although the computer corporation has provided most of the hard cash that figure doesn't take into acccount some local investment by PUC-Rio [precise figures unavailable] and the resources made available by the Vatican Library. "The cost of digitising beyond this," says Chiu, "would be prohibitive for any institution to bear." We are talking about a project here that IBM estimate will take anything up to a century to complete, and although Father Boyle later made the point to me that "the library has been existence for 550 years, so it doesn't matter to us if it takes 50 or 100 - we're not fly-by-night", it's still a pretty major investment in anybody's book. And it has to be paid for.

So why were IBM so keen to pour so much money into helping such a potentially awkward and publicity shy institution as the Vatican Library? There seem to be two answers to this. The enabling factor was obviously the personal chemistry between Father Boyle and Fabio Schiatarella, who between them had made the whole thing possible (Boyle was to confirm this later - the digitisation project has been a very interpersonal operation from the outset). The second answer is that on a corporate level, the digitisation of the Vatican would act as a great centrepiece for the digital library initiative. Everybody has heard of the Vatican; everybody is familiar with its reputation as an ultra-conservative and ultra-traditionalist organisation. For IBM to be able to say that it took its technology in here and made it work, now that would be something. This is obviously why Schiatarella was not prepared to rock the boat by talking to me. Press publicity is not so important to IBM on this one. It is only potential customers they want to impress, and Planet Library is small enough that those who they wanted to hear about it would hear anyway. The greatest danger was that the project might be trivialised.

It comes as no surprise then that in the beginning Big Blue wanted to digitise only illuminated pages of the manuscripts, or illustrated frontispieces. But Boyle knew that if the project were to be at all useful to scholars and not just a gimmick entire manuscripts had to be scanned. He got his way: the digitisation project was only to happen if the end result was a proper online resource that made the manuscript library properly available. As he was to tell me later: "the danger in the digitalisation is that it can become gimmicky. You have a virtual library, so you can zoom through, looking at the titles. Look - that's for children. This is for scholars." And because of that, he was happy to see the service designed in such a way that ultimately it would pay its own way.

The upshot of this is that library won't go on the web for free. The search engine (which is intended to eventually include a capability for queries by image content) probably will do, but the images will be offered in a variety of resolutions and a corresponding variety of prices. No levels have been set yet, but the current library charge to print up a manuscript microfiche is betweeen US$30 and US$50 per page, and the Internet service will no doubt hope to undercut this, at least at the bottom end.

Finally, to protect the images' copyright once in the public domain IBM has developed a "watermarking" technique, a program that encodes a visible design (with an ineradicable "noise" element) into every image that would be made available on the public networks (a private, unadulterated, intranet would be made accessible within the library itself as well as in various other institutions around the world). IBM and the Vatican make a great play about how this image is "large but faint", and does not interfere with the reading of the manuscripts. Personally, I did not agree - the image fills most of the page and, while it may not technically interfere with the detail of the lettering, it destroys the aesthetics of the page completely. So much for the manuscripts being great works of art - corporate paranoia seems more important. The practice wouldn't be so obnoxious if it weren't for the fact that an "invisible" watermark will also be made available - for a price. It seems that it's only the poorer end of the market that is to be mistrusted.

Once my two guides had explained all this we left the vaults and Sheehan took me back up in the lift to see the area which had been the main reading room between 1587 and 1890. On the way we looked into the courtyard into which the top of the vault protrudes and across which priests and officials wander on their day to day business. It's overlooked by the room in which Pope Gregory drew up the Gregorian calender, a tiny example of the momentous events to which these walls have over the years borne witness.

The reading room itself isn't bad. It's included now as part of one of the Vatican Museum tours, so if you're ever in Rome you can pop in and have a look. A large, highly decorated space, its frescoes took 130 painters 13 months to complete. While not the greatest I've seen in terms of technique their subject matter is fascinating, worthy of an article in itself. Down one wall are scenes from the great libraries of history: Babylon, Athens, Florence, Alexandria, some of which actually show apprentices preparing parchment for the scriptorium. Down the opposite wall are scenes from the great papal Councils. But most wonderfully, each of the four sides of the pillars which run down the centre of the room and support the roof has an image of a different inventor of or contributor to the languages of the world. The entire classical myth of the development of language is here: the word is first spoken by Adam at one end of the room, and passes via Moses, Abraham, Hercules, Cadmus, Pythagoras, Simonides, Evander, Hieronymus and the others on the pillars in a direct route to Christ, the author of celestial doctrine, at the other. Around and about various manuscripts are displayed in cases: when I was there ten mediaeval versions of Ptolomy's Geographica were on display, as were fourteenth century copies of Suetonius's De Vita Caesarum and Seneca's Tragedia. But the stars of the show were the two oldest surviving manuscripts of Virgil, dating from the fourth and fifth centuries, and being shown in their entirety for the first time.

Tourists wandered amongst the cabinets, content merely to glance at these texts which, for good or for ill, form the superstructure of knowledge upon which modern Western culture, like the arch of a cathedral, was built. Since the second world war that arch has become self-supporting: we've managed to a large extent to kick away these ancient texts from beneath us. But it's easy to forget that many of the enlightened attitudes that today we hold so dear (as well as many of the less enlightened ones that today we feel the need to re-examine) came from the culture of copying and preserving and studying and re-studying that was created and nutured through numberless years by nameless monks in countless scriptoriae across Europe. If we want to know ourselves, we cannot ignore their work.

I left Sheehan and the Library with my head reeling and spent the afternoon wandering through Rome, that strangest of urban spaces, where buildings of the ancient city punctuate the stratum of the modern town like outcrops of bedrock, harder and more stern, echoing the combination of ancient manuscript and digital scanner in the library itself, and making time itself seem strangely folded. Everywhere you go here the past maps onto the present and the two continue together, and time becomes anything but a straight line.

My story was meandering too. Although I'd managed to visit the library, without Boyle and Schiatarella I didn't have the full account I needed. I continued to pester the Italian until he finally agreed to meet me in Big Blue's picturesque offices around the corner from Rome's famous Trevi fountain, but he refused point blank to discuss any aspects of the digitisation project with me and since I didn't have too much interest in his family life or the state of the weather (which was unchanging and unbelievably hot) the interview was a complete waste of time. The date, ironically enough, was Friday 13th, so I figured that I shouldn't be too surprised that it hadn't gone well. Expecting a further refusal I went back to my hotel to telephone Boyle for the last time, as he'd instructed. But it looked like my run of bad luck was over - to my great surprise he agreed to meet me at his new church the following morning.

Boyle's new church, San Clemente, is one of the most important churches in Rome. Although the church dates from around 1100 AD, it stands on top of an even older church dating from the fourth century, which in turn was built on top of a Mithraic temple when Mithraism - an early Christian sect - was outlawed in 395 AD. It seems that in Italy, laying one technology on top of another is nothing new. Boyle himself, I was to discover, helped to dig out these buildings from beneath the present structure, and indeed wrote the tourist guidebook to the place back in 1960. A short, slender man, it's easy to imagine him toiling away on an archeological dig - although in his seventies he retains the wiry look and open face of someone who has known both the tight focus of the intellectual pursuit and the expansive energy of the physical.

After all the nonsense I'd been through I expected to find Father Boyle hostile, but he was relaxed and friendly. We sat outside under a sun umbrella in a courtyard probably eight hundred years old and chatted about the Boyle's life and the project. Originally from Tralee in County Kerry, Ireland, Boyle studied at Oxford for eight years. He moved to Rome in 1955 to take up a teaching position and spent a lot of time in the Vatican Library, pursuing researches of his own. In 1961 he relocated to Canada to take up a job teaching palaeography at the Pontifical Institute of Mediaeval Studies at St. Michael's college, which is part of the University of Toronto. Many of his summers were spent back in Rome helping with the San Clemente excavations and, of course, delving into old manuscripts back in the Vatican Library. So when he was offered the job of Prefect in 1984, Boyle already knew the library pretty well. "That is, I knew it to use it, and I knew my own little things, but I had no idea what it was in itself," he qualified. "But gradually it came through to me that this was a manuscript library and not a library of printed books." They are, he implied, quite different things.

Printed books, however old, are all, ultimately, simulacra - copies without an original. That, after all, is the nature of printing. But manuscripts are all originals: each one is an individual work of art (although rarely constructed by a single individual). The decision to digitise the manuscript collection would ultimately be an emotional one, to do with the preservation of not merely factual information but of cultural information. For these manuscripts hold within their pages not only words copied and written, but countless wordless stories about the lives and minds of the monks who wrote them and the mutating webs of language through which they moved, all hidden in the way the pages have been prepared, the stylistic devices used in the text, the different styles of handwriting and script employed, and even in the way the pages are stitched together. If there was to be any point at all in digitising the manuscripts, most of this information had to be captured.

Father Boyle was no techno evangelist. "I knew nothing about automation when I first came [to the Library]," he assures me. "I had seen it at Toronto, I think I'd used a monitor once or twice looking for a book, but I didn't have my own computer or even access to one." It was when he saw the results IBM was getting at the Spanish archives in Seville that he realised that the technology was available to achieve just this. Not only were the manuscripts and documents (such as the aforementioned Columbus letters) being recorded with extraordinary quality, but the use of software to eliminate bleed-through and blotches meant that some of them could now be read for the first time. This meant that the new tools were no gimmick; rather, their application was actively enhancing the archive. And of course, they would make the archive vastly more accessible: scholars who may have been able to journey to Rome once or twice in their careers to consult a particular manuscript would be able to pull down low images from the net in a matter of minutes from almost anywhere in the world. But even now he sees parallels between digital technology and the ancient technology of the scriptorium. "Broadly speaking it's the same thing in a different medium," he says. "The scriptorium was copying - the monks earned their living that way - and certain monasteries became very distinguished places for copying, and they got plenty of work to do, and they spread texts that way. Then printing came along and of course killed in many ways the scriptorium as such."

Gradually then a pilot project took shape as an informal partnership between the Library itself, PUC-Rio and IBM. A small number of scholars were contacted and asked to act as a sounding board for the new system, and the manuscripts to be scanned were chosen either because these scholars specialised in them, or because they were generally popular in the library. They included a Hebrew manuscript Bible, a fifteenth century Latin manuscript discussing medical herbs, and a sixteenth century Aztec manuscript - one of the very few written works remaining today from the pre-Colombian civilizations of Central America (some of the results can be seen at http://www.software.ibm.com/is/dig-lib/fmanuscrpt.htm). Occasionally though the manuscripts were chosen on personal whim. "I think the very first one we did "was a Persian manuscript," says Boyle, because I happened to like it. It's an illustrated romance, probably done in Cordoba in the middle of the 11th century, and it includes images of people playing chess and so on."

But while I finally had the story about the digitisation in the bag, I still hadn't solved the big mystery. Why had it been so difficult for me to get this far? As my interview with Boyle drew to a close, I asked him to explain the actions of the press office. Surely they were covering something up. "Oh no," he said, laughing. "[They] simply didn't like the magazine; they thought it was cheap. And inclined to go into flip stuff. You know, 'Vatican flies high' - that kind of thing." Apparently, the Holy See had been upset with some of the attitudes generated in the press by the digitisation project. Many of the articles about it had been given impious titles by over-enthusiastic copy editors - "Holy Chip" was another example Boyle gave me - and as a result the Vatican had decided to take a firmer hand in the publicity process. "Do you do that for the Metropolitan Museum, for the Library of Congress?" Boyle asked me. "No. You have to jazz up the Vatican. And this is why the communications office put its foot down when they saw Wired. They sent me a fax saying 'Nothing doing. This is one of these "hippy" magazines, which treats everything as a great lark.'"

Boyle himself was clearly more annoyed with journalists' continued questioning about the financial angle: "They all ask where's the money side, it's the Vatican, there must be a money side. But in this there is no money side. I doubt if the Vatican will ever make a penny out of this. People do do things from an idealistic point-of-view, and in this case the opportunity came, through IBM, to do something of which frankly I had never thought in this way."

It's a shame that the other partners don't always seem to share his idealism; or, if they do, are too shy to want to express it. Despite assertions from Boyle and Sheehan that the original impetus for the project came from PUC-Rio, the university is incredibly reticent about the project. Milton Kelmanson, head of the project in Rio, would not answer my questions, claiming that he had nothing of interest to say, and none of his staff responded to my enquiries. What makes this PR vaccuum so extraordinary is that 1997 is a crucial year not only for the Library project but for all links between Rio and Rome. This October the pope is coming to Rio, an event of immense significance for Brazil and, naturally enough, for PUC-Rio as well. Indeed, as part their contributions to the celebrations they have organised an exhibition, "Treasures from the Vatican: From Parchments to Bits" which has for the first time brought six manuscripts across the Atlantic and put them on public display. The originals are accompanied with an electronic installation showcasing the digitisation project, and there is an accompanying CD-ROM.

But despite this the pontifical university would not even reveal the number of terminals it has made available to scholars, or discuss whether or not they regarded the pilot phase of the project as a success (though I found out through IBM that five Brazilian scholars have access to the material scanned so far). This kind of attitude, coupled with the Vatican library's insistence that all the servers must be held in the Vatican and in the Vatican alone (ignoring IBMs advice that it would be much more sensible - and perhaps more secure - to distribute them around the globe), suggests that despite the high moral tone surrounding the project it is going to be a very long time before altruism tips the scales against paranoia - the same paranoia which has insisted that a massive and obtrusive watermark be slapped across every image before it can go out on the net, and which has also ensured that, at the moment that it has embraced the most incredible communications technology that the world has ever known, the Vatican has simultaneously failed to communicate.

Phase two of the Vatican Library digitisation project is now in the hands of the new Prefect, Father Farina. Father Farina was previously the rector of a university and was involved in the computerisation the card catalogue for printed books. Father Boyle assures me that the project is safe in his hands.

Note: I am indebted to Father William J. Sheehan's introduction to his catalogue of the Incunabula for the information concerning the history of the library.