“When it was proclaimed that the Library contained all books, the first impression was one of extravagant happiness,” wrote Jorge Luis Borges in his classic of philosophical fiction, “The Library of Babel.” One of the most revered stories-as-thought-experiments ever committed to print, Borges’ fiction posits the Universe as a library (“composed of an indefinite and perhaps infinite number of hexagonal galleries”) that contains every possible text. This intellectual vision, at once playful and poised, has stirred authors (like Umberto Eco and Terry Pratchett) and philosophers (W.V.O. Quine and Daniel Dennett) alike for more than 75 years.
And now it exists! Recently, Jonathan Basile, a Brooklyn author and Borgesian Man of the Book, taught himself programming so that he could recreate Borges’ Universal Library as a website. The results are confounding. A true site-as-labyrinth, Basile’s creation is an attempt to write and publish every story conceivable (and inconceivable) to man. In the process, Basile encountered new philosophical conundrums, French rappers, and unheard-of porno search strings. The possibilities, after all, are endless…
I think the implications of your recreation of Borges’ “The Library of Babel” will dawn on the participant as she experiences it, but can you just quickly describe what it is that you’ve done here with this crazy project?
The site is an attempt at a virtual recreation of the world of the Borges’ short story. “The Library of Babel” describes not only a universal library — a library containing all possible combinations of letters, thus all books that ever have and ever could be written — but a universe that is a library, endless hexagonal reading rooms inhabited only by librarians. There’s not even any food in their world (though they do have bathrooms). The library creates a tantalizing promise of reason — somewhere in its pages are all the works lost in the burning of the Library of Alexandria, and every future masterpiece — but drowned out by infinite pages of nonsense. The life of the librarians is quite melancholy for that reason.
When I started building the site, I actually had in the back of my mind the idea that a searchable, virtual Library of Babel might make it possible to find a few of those rational arrangements of letters. I very quickly realized how incorrect I was. And that I think is the most important part of the project — it gives that brief glimmer of hope, that reason might win out over unreason, then crushes it. In this way the site is true to Borges’ vision — I think he wants us to see that all the creations of reason, of human language and thought, are haunted and undermined by their irrational reproducibility.
How on earth did you get the idea to do this?
I was lying in bed one night and the idea of an online Library of Babel popped into my head. My first thought was — it must exist already. It seems like such a natural extension of the capabilities of a computer that I was sure someone would have made it. The next day I looked for it, a bit excitedly, and was disappointed. From then on, it’s kind of been a reluctant destiny for me.
What problems did you have to solve along the way?
The first was learning to code. That’s a work in progress. As my understanding of programming has grown my ambitions for the library have as well. The version of the library that is online now is built by generating text documents with random letters, and then reading from those files to serve the web pages. It’s a silly method for a number of reasons. First of all, it could never produce a universal library — it takes more than a terabyte of storage to hold a million books, so there wouldn’t be enough storage in the world to hold 29^1312000 (a full library). Even if the universe were just racks of hard drives, it wouldn’t be enough to hold all that. It was also a problem figuring out how to host a site on this model — don’t believe any of the webhosts who claim to offer “unlimited” storage. I learned that the hard way.
More importantly, this model really ignores the power of the virtual. There’s no need to store or pre-generate books as long as they can be generated from an algorithm which takes their “location” in the library as an input. The simplest version would be just to create them in order — the first page of the first book would be 3199 spaces followed by the letter a, then b, c, etc. until you had 3200 periods (the library contains the 26 letters, space, comma, and period). But when I was just starting out, still learning basic concepts of programming, I couldn’t see a way to build a library on this model and stay true to the randomness which is essential to Borges’ story. It was some of the first visitors to the site who introduced me to the concept of pseudo-random number generators and the random seed. I’m rebuilding the library now so that it will contain every possible page of text (about 29^3200 possibilities) and be instantaneously searchable. That should be online in a couple of weeks.
The size of the library continues to produce interesting challenges — to be able to generate that many unique possibilities I have to work with numbers of about 6,000 digits. I’m working in C++ now, whose normal data structures only allow numbers up to about 17 digits (64 bits), so I’ve had to begin using a multiprecision library to extend its capacity.
The Borges story comes from a line of “projects” or stories that build off of one another. Do you see your site as an extension of this lineage? [From Wikipedia: “The story repeats the theme of Borges’ 1939 essay “The Total Library” (“La biblioteca total”), which in turn acknowledges the earlier development of this theme by Kurd Lasswitz in his 1901 story “The Universal Library” (“Die Universalbibliothek”).]
I’ve often wondered about this. Borges has a wonderful essay, “The Total Library,” which he wrote a couple of years before “The Library of Babel.” His knowledge of world literature was incredible, and he traces the idea of the universal library not just to Lasswitz, but all the way back to the Ancient Greek atomist philosophers, as reported in the works of Aristotle. They described the endless complexity of the world emerging from the mechanical interactions of a few basic particles, and employed the analogy of the infinite complexity of language emerging from the permutations of its atoms — the letters. I definitely see the site as continuing in this tradition, viewing reason as an extension of the much vaster abyss of unreason, somewhat in the sense that Nietzsche said the living are only a rare species of the dead.
But still, what would Borges think? It’s always hard to tell when he’s being ironic, and many of the passages in “The Library of Babel” seem to contradict what he wrote only a few years earlier in “The Total Library.” There he says: “One of the habits of the mind is the invention of horrible imaginings. The mind has invented Hell, it has invented predestination to Hell… I have tried to rescue from oblivion a subaltern horror: the vast, contradictory Library, whose vertical wildernesses of books run the incessant risk of changing into others that affirm, deny, and confuse everything like a delirious god.” Perhaps, if we take him at his word, he would be horrified to see the library enter even the virtual world, and horrified to see people willingly enter its hexagons. But then, he was never one to take the border between reality and fiction too seriously; reading his story is already, in its own way, entering the world of the library. In a sense it’s a horror story, but it feels to me more like a black comedy. Perhaps he would just laugh.
Any weirdness you’ve experienced along the way?
I suppose any time something gets published, people end up seeing something the author hadn’t anticipated in it. I’ve found that becomes even more true when the author intended nothing, or perhaps when there is no author. It’s been a little bit easier to gauge people’s responses to the text from a spin off I created — once I had written some of the code for libraryofbabel.info, it really increased my desire to permute things, so I started a twitter account which is tweeting all possible combinations of 140 characters, Permuda Triangle.
People have a way of seeing whatever they want to in random letters. Any acronym-intensive community seems to love Permuda Triangle’s tweets. Businesses always seem to think that I’m tweeting their initials or ticker symbols — the First National Bank of South Africa followed me after I tweeted fnb, for example, as did a British ISP named EE and something called Net Promoter Score (NPS). The hip hop community frequently favorites and retweets what Permuda sends out — Fresha Got the Kush and a number of his fans responded when I tweeted fgk, a bunch of French rappers were into cpls: “certifié par la street” and some kids in Arizona went nuts when I tweeted fljl. Usually people who come across the tweets follow the account for a couple of hours, then realize what’s going on. Also, it sends 2400 tweets a day, so it kind of clogs your feed.
It’s harder to tell what responses people are having to the texts on libraryofbabel.info, but a few people have posted some reactions on the forum. One guy tried to find some meaning in the number of times he could find the names of various religious figures in the library. On the one hand, all words occur with a frequency which is based solely on their length, so 1 in every 20 million or so five-letter combinations will be jesus, while satan has the same distribution. On the other hand, who’s to say these names and letters aren’t significant in themselves? I was open to his interpretation.
And a lot of people have been coming to the site now by misspelling words in their google searches — especially searches for pornography, which I think reflects what people are most likely to be typing with one hand. Oddly, they’ve not only followed the links, but clicked through to the rest of the site, after searching for things like “sezx” or “milphs.” Maybe it’s a more exciting project than I’d imagined.
In his reduction of the story, the philosopher Quine asserts that you can recreate the universal library with just a dot and a dash. I don’t know anything about programming, but didn’t you prove this with your project?
I’d say yes and no. The question of what alphabet a universal library should base its permutations on is a tricky one. Borges said that only twenty-two letters would be necessary — starting from the modern Spanish alphabet, eliminating the double letters and ñ because they’re redundant, k and w because they’re foreign, x because it’s an abbreviation, and q because it’s superfluous (so he says). Oddly though, in the short story, which casts itself as a missive from one of the librarians which found its way to Borges, many other characters are present, even in the texts of the library (ö and x for example). Borges up to his old tricks.
libraryofbabel.info uses a different alphabet than Borges’ library, to acknowledge its implicit Englishness. In a way, this is to recognize that it’s not a universal library, and that nothing ever could be. The idea of reducing characters to an essential alphabet depends on a certain interpretation of language, and it’s clear that Borges treated it with irony. (Interestingly, it would be much closer to Quine or analytic philosophy’s notion of the essence of language — it does seem like Borges’ irony was lost on Quine). Only if language has a meaning indifferent to it’s expression and is universally, losslessly translatable, could any alphabet pretend to express the entire universe of possibilities. But if we acknowledge that something as simple as a diacritical mark has something to tell us, than any attempt at a universal language will always be haunted by what it has left out, and by all the languages still to come.
I was going to ask you if you have any other stories or books you’d recreate as a site, but actually this one pretty much takes care of it…
Absolutely. You know, I still think of myself as primarily a fiction writer, but I sometimes wonder what the point is now. I’ve already published every possible story I could write…