Narrative made sense for cultures based on tradition and a small amount of information circulating in a culture -- it was a way to make sense of this information and to tie it together (for instance, Greek mythology). Database can be thought of as a new cultural form in a society where a subject deals with huge amounts of information, which constantly keep changing.This quote has been rattling around my skull ever since. I've tracked down the original interview in which Manovich made the comment, which, in its entirety, reads (new parts bold):
The shift to the database can be understood as part of the larger shift from a traditional “information-poor” society to our own “information-rich” society. Narrative made sense for cultures based on tradition and a small amount of information circulating in a culture – it was a way to make sense of this information and to tie it together (for instance, Greek mythology). Database can be thought of as a new cultural form in a society where a subject deals with huge amounts of information, which constantly keep changing. It may be impossible to tie all together in a single (or a set of) narrative but you can put it in a database and use a search engine to find what you are looking for, to find information which you are not aware of but which matches your interests and finally to even discover new categories. In short, a narrative is replaced by a directory/index.(Manovich builds out this argument more fully in The Language of New Media around page 228, where he argues that "database and narrative do not have the same status in computer culture.")
Now, if the database -- which Manovich himself defines as "a structured collection of data" -- is "new," what about these weird contraptions?:
The first image is from Georg Philip Harsdörffer's Fünffacher Denckring der Teutschen Sprache, or the Five-fold Thought-ring of the German Language, described in his Philosophische und Mathematische Erquickstunden (1651). It stores different units of language, similar to morphemes, on five different nesting wheels, which then can be spun to quickly generate words, many of which didn't actually exist but, because they were morphologically correct, could exist in German (or so Harsdörffer thought). It was also used to generate poetry, since the user could easily find rhymes by keeping the last syllables of a word stable while moving the first few rings. If my calculations are correct (and I admit to not having taken math since high school), the Denckring allows 97, 209, 600 permutations -- 300 times as many words as the Oxford English Dictionary, which took over seventy-five years to produce the first edition.
The second image is from Ramon Llull's Ars magna. Llull is a pretty famous guy in the history of computing, and justifiably so. In the thirteenth century, he developed a system of nesting wheels that 1) stores the nine divine dignities in different variables (shown in the image above), 2) could quickly combine those dignities to construct premises, then 3) combines those premises to construct logical syllogisms that argue for the existence and indeed greatness of the one and only Christian God. In The Search for the Perfect Language, Umberto Eco counts 1,680 arguments in Llull's system -- that's 1,680 theodicies, 1,680 proofs, 1,680 treatises on God.
The third image
It seems databases -- structured collections of data used to manipulate, analyze and generate huge amounts of information -- have been around for a long time. And I didn't even go back to Porphyry's tree, from the third century, or taxonomic systems developed by Aristotle, since taxonomies and hierarchies often involve classifying any amount of information, rather than organizing large chunks of data.
Of course, the fact that humans have been structuring data for a long time doesn't refute Manovich's main point: that the database has taken on a new significance in contemporary culture, in a way that Lullian circles never did for late medieval or early modern Europe . My issue isn't with the comment as much as it is the attitude, the tone of the argument, which seeks to identify the present against the past. That is: we're information-rich, "they" were information-poor; we've got database, "they've" got narrative. Not only do these relativist claims not hold up to scrutiny, they ironically end up obscuring Manovich's (important and interesting) point by locking media into time periods narrowly defined by one over-arching principle.
This approach tries to make media history safe, simple, explainable. One the one hand, it seeks evidence of already-formed beliefs about the past, merely confirming a history that's already been written; on the other, it seeks obvious differences, drawing a line in the sand between the past and the present. In both cases, what is truly significant about the database, or any media practice -- the failed experiments, the crazy plans, artifacts like Lullian circles that don't quite fit into the narrative we want to find -- all that gets lost. And what we're left with is a pretty boring picture of both the past and the present.
I'm reading a great little book right now by Murray Cohen, Sensible Words. Published in 1977, it digs up a ton of interesting universal language projects and grammar books from the 17th through 19th centuries, advocating an approach that honors the diversity of thought that existed at the time. Cohen writes:
Our methods of historical analysis, our selection of terms, figures, and sights (or cites), speak who we are by showing how we think and, even, what we do not know we believe. If we do not resist the blinders that the conventions of our own discourse put on us, we succumb to histories that put a premium on a continuous track (which may have some hurdles or some well-marked, gracefully curving detours) that finally includes us along the Way.Yes, Cohen is very clearly influenced by Hayden White, Thomas Kuhn, Michel Foucault, and other thinkers challenging the dominant narratives of history. What's been so surprising, as I read this book published over thirty years ago, is how relevant his critique of historical methods remains today, especially in media and literary scholarship. I'm staring at a stack of books on 17th-century poetry that don't cite Harsdörffer once. It took me a week, and several random emails, to track down a copy of Leibniz's dissertation (untranslated, a scanned copy of an 18th-century edition). What little I have found on proteic poetry contains blatant contradictions of fact (does Kuhlmann's sonnet have 152, or 169 permutable parts? didn't anyone count them before writing their essay?).
So when, after spending hours researching Lull and Harsdörffer and Kuhlmann, I read this distinction between database and narrative -- an admittedly small point, made seven years ago by a thinker whose work I deeply respect, and who's more historically sensitive than most -- I really start to question every narrative I've absorbed, every survey course, every canon. If Kuhlmann's proteic poetry didn't make it into the historical record, what else are we missing because of a still-prevalent Whiggish historiography? What are we losing right now because Google Books doesn't deem it significant enough to digitize? And what insights have we missed about our contemporary media culture because we haven't taken the time to significantly understand the past?