Seriality and the Computational Sublime
Presented at "Infinite Possibilities? Seriality: A Symposium"
Davis Musesum and Cultural Center
Wellesley College
Saturday, March 6, 2004
Seriality and the Computational Sublime
The computer is a metamedium. It can represent almost any other medium. In this sense, there are many new media projects that look like some of the works in Infinite Possibilities: Serial Imagery in 20th-Century Drawings or address similar issues of the multiple, the variation, the sequence. What I would like to address today are some of the ways that seriality is fundamental to computational media, even if some of the resulting works might not at first glance - or first experience - seem like serial works. What I would like to suggest is that the underlying seriality of computationally based art and how this so-called "language of new media"1 may lead to new forms of narrative seriality in art.
"Given a diophantine equation with any number of unknown quantities and with rational integral numerical coefficients: . . . devise a process according to which it can be determined by a finite number of operations whether the equation is solvable in rational integers."2
In 1900 David Hilbert first posed what came to be called the Entscheidungsproblem or decision problem. Fundamentally, the question was whether a mechanical or algorithmic procedure could, as Roger Penrose paraphrases, "in principle solve all the problems of mathematics (belonging to some suitably well-defined class) one after the other?"3
In 1936 a 24-year old mathematician named Alan Turing published the paper, "On Computable Numbers, with an Application to the Entscheidungsproblem." In it he proved that there are, in fact, noncomputable functions. Turing was not the first to disprove Hilbert's underlying assumption - Kurt Goedel's incompleteness theorems of 1931 accomplished that - or even the first to prove that certain problems were not "effectively calculable," in Alonzo Church's earlier phrase.4 Turing's novel methodology, however, laid the groundwork for what came to be known as a Universal Turing Machine and is otherwise known as a computer.
In order to prove his hypothesis, Turing had to establish, according to George Dyson, the nature of computability itself. To do this he invented the idea of the following machine:
"We may compare a man in the process of computing a real number to a machine which is only capable of a finite number of conditions . . . which will be called "m-configurations". The machine is supplied with a "tape", (the analogue of paper) running through it, and divided into sections (called "squares") each capable of bearing a "symbol". At any moment there is just one square . . . which is "in the machine". We may call this square the "scanned square". The symbol on the scanned square may be called the "scanned symbol". The "scanned symbol" is the only one of which the machine is, so to speak, "directly aware". However, by altering its m-configuration the machine can effectively remember some of the symbols which it has "seen" (scanned) previously . . .. In some of the configurations in which the scanned square is blank (i.e. bears no symbol) the machine writes down a new symbol on the scanned square: in other configurations it erases the scanned symbol. The machine may also change the square which is being scanned, but only by shifting it one place to right or left."5
The point is that this conceptual "machine," which can only make one mark at a time on a single, infinitely long piece of paper and only move the paper left or right one square at a time, i.e. serially, can solve for any computable problem. And while it is called a machine, conceptually it is software - a set of procedural instructions.
According to Sol LeWitt, one idea of seriality is "a way of creating art that did not rely on the whim of the moment but on consistently thought out processes that gave results that were interesting and exciting."6
The Turing Machine is nothing if not a serial process. But are the results interesting and exciting? They are, I would argue - or at least they can be - sublime.
In 1997, John Simon created one of his signature works, Every Icon. The idea is simple enough. Taking a 32 x 32 grid - the standard size for a Mac icon according to Apple's vaunted human-computer interface guidelines - Simon created a Java applet - an algorithm coded in a particular software language or, in other words, a Turing Machine - that causes each square of the grid, serially, to be colored black or white. It progresses by counting. Hence, over time, every single, possible icon will be imaged. You will note, however, that the icon linnked here has been running since 1 minute after midnight on April 9, 1998, and it has only made it to the 36th square. There are approximately 4.3 billion combinations possible along the top row alone. The second row will take roughly 5.85 billion years to compute - your results may vary, depending on the speed of your computer. Simon has calculated that for the icon to turn completely black, its final state, will take several hundred trillion years or, as he puts it, "a very, very long time."7
An unimaginable amount of time. And that is part of the point. Every Icon evokes a kind of sublime awe, which surely cannot be far from the sentiments of the poet Percy Bysshe Shelley when he wrote in from Mont Blanc: Lines Written in the Vale of Cham
Thou art pervaded with that ceaseless motion,
Thou art the path of that unresting sound ---
Dizzy Ravine! and when I gaze on thee
I seem as in a trance sublime and strange
To muse on my own separate fantasy,
My own, my human mind, which passively
Now renders and receives fast influencings,
Holding an unremitting interchange
With the clear universe of things around;
A critical difference between Shelley's Mont Blanc and Simon's Every Icon is that Shelley is attempting to represent his experience, while Simon is dynamically generating it. To put it another way, the sublime counterpoint to Shelley's humanness is nature - the divinely created "clear universe of things around"; to Simon's it is the virtual - algorithmically, serially, procedurally created.
Here I can't help but take a brief detour and note that in the introduction to her book Frankenstein, Mary Shelley Wollstonecraft wrote about evenings during Alpine sojourns with her husband Percy:
"Many and long were the conversations between Lord Byron and Shelley to which I was a devout but nearly silent listener. During one of these, various philosophical doctrines were discussed and among others the nature of the principal of life, and whether there was any probability of its ever being discovered and communicated . . .. Perhaps a corpse would be reanimated; galvanism had given token of such things: . . .."8
In the late 19th century, speculation was whether the new medium of electricity could animate the inanimate, like a spark generating a fire.
In the mid-20th century, Turing formulated his famous test of a computer's ability to perform human-like conversation, which goes roughly as follows: if a human judge engaged in a natural language conversation with two other parties in another room, one a human and the other a machine, and the judge cannot reliably tell which is which, then the machine is said to pass the Turing test. And according to the Wikipedia, "Turing . . . estimated that by the year 2000, machines with 109 bits (about 119MB) of memory would be able to fool 30% of human judges during a 5-minute test."
In the early 21st century, my computer, in fact, has 256 MB of memory, and while I don't think the Microsoft Office paperclip character would pass the turing Test, significant progress has been made toward what Turing called "machine intelligence," whether it is a computer defeating the world chess champion or simply speech recognition built into your car's navigation system. The breathless rhetoric of "artificial intelligence" has abated, but the converging sciences of emergent systems - what Steven Johnson subtitles his book as The Connected Lives of Ants, Brains, Cities, and Software - has given new life, so to speak, to the serial marking and movement of infinite strips of paper that comprise the Universal Turing Machine. In other words, with computational seriality, simple procedures give rise to complex systems
One of the ways that this idea has become popularized is with John Conway'sGame of Life, which Martin Gardner wrote about in Scientific American in 1970 and which has since then become a kind of cult classic.9
The rules for Game of Life are very simple:
For a space that is 'populated':
Sum Each cell with one or no neighbors dies, as if by loneliness.
Sum Each cell with four or more neighbors dies, as if by overpopulation.
Sum Each cell with two or three neighbors survives.
For a space that is 'empty' or 'unpopulated'
Sum Each cell with three neighbors becomes populated.
What you begin to get a sense of is how a simple set of rules can have unexpected and unexpectedly complex results. And it is these simple procedural rules that many artists are working with.
Eric Ishi Eckhardt and Justin Bakse, for instance, took the rules for Game of Life and, what else, turned them into a competitive game, Life vs. Life.
Both Game of Life and Life vs. Life illustrate the idea of emergence, and for some these emergent systems, like Wollstonecraft's reverie about galvanism, are tokens of things to come - of the sublime possibility of procedurally generated but autonomous intelligence.
Four artworks in an exhibition called eVolution, curated by Christiane Paul for Art Interactive in the Boston area, are poetic evocations of that sublimely feared and desired goal.
Maciej Wisniewski'sInstant Places; Christa Sommerer and Laurent Mignoneau'sA-Volve, Rebecca Allen'sBush Soul, and David Rokeby'sGiver of Names each present as a possibility of a kind of intelligent responsiveness that Jack Burnham complained in his 1968 Beyond Modern Sculpture was a lack of traditional sculpture.
"An inherent defect, psychical and physical, in sculpture and automata - and even today's Kinetic Art - is that they do not respond to man in any intelligent fashion. They are dead souls and made alive through 'art' and prearranged mechanized motion."10
While I would argue along with Marshall Mcluhan that it is never a case of one medium replacing another - despite what Burnham seems to imply or at least desire; nevertheless, unlike On Kawara'sOne Thousand Days One Million or Sol LeWitt'sModular Open Cube Piece, computational media can be not only conceptually but effectively unlimited, like Simon's Every Icon, or actually responsive, like Rokeby's Giver of Names.
I am not trying to argue that all computational art is serial art, but I am suggesting that the procedural seriality exemplified by some of the artists in Ad Infinitum, is a critical and distinctive characteristic of computational media and that within this context, dynamic systems make possible a whole new instantiation of seriality in process not just as process.
There is at least one other way that computational media intersects with notions of seriality, which I would like to touch on.
In an interview with Hans Ulrich Obrist, Mel Bochner says that The Theory of Painting "incorporates language written on the wall which grammatically maps the physical relationships; the language is, in a sense, the instructions or the genetic code of the piece. It is presented as equivalent to the physical evidence."
Besides the notion of the procedure, Bochner is also foregrounding the relationship of how his language maps the physical space as a kind of reflexive narrative of it. With computational media, the narrative space is created by actual use of the work, reversing the relation between description and object. The new media object literally creates the narrative.
In his seminal book The Language of New Media,Lev Manovich argues that the database is a new "symbolic form" as significant to the contemporary world as linear perspective was to the modern age. For example, in a sense, the World Wide Web is one huge hyperdocument for which any node is accessible from any other node - which is simply a way of saying that it is one huge database of all those links.
Especially from the user perspective, the database format is the dominant way we interact with the information explosion - we search, browse, link from one object to another, whether or not it is formally part of a database. The reason this is significant is because of how it affects the idea of narrative.
To explain this change in worldview, Manovich presents the Saussurean notions of the syntagmatic and the paradigmatic.
"To use the example of natural language, the speaker produces an utterance by stringing together elements, one after another, in a linear sequence. This is the syntagmatic dimension."11
It is also the narrative dimension. It is also serial. Language is strung together in a particular order to tell a particular story. In the paradigmatic dimension, "all nouns form a set; all synonyms of a particular word form another set." He goes on:
"Elements in the syntagmatic dimension are related in praesentia, while elements in the paradigmatic dimension are related in absentia. For instance in the case of a written sentence, the words that comprise it materially exist on a piece of paper, while the paradigmatic sets to which these words belong only exist in the writer's and reader's minds."
The database, however, privileges the paradigmatic and there is a "place," a space, a cell in the database table, for every noun, adjective, part of speech or however the topic is being paradigmatically categorized. It is the narrative that is in absentia. Which means that the narrative can be "authored" on the fly by the actual use of the database or by the algorithmic procedures that the artist has used to filter the database and structure the resulting flow of information.
A perfect example of this "inverse serialization" - what Manovich calls "database cinema" - where language is the dynamic product of an interaction, not the pre-fixed object of attention, is a magnificent project by Ben Rubin and Mark Hansen, Listening Post.
Imagine that each of the 121 video screens of Listening Post represents a cell in a database table, and the reading head of the Turing Machine goes to each one in a precise order out of which the content is created. This isn't actually what happens, of course, but metaphorically it ties together the procedural seriality of the computer itself with the resulting serial narrative, which is produced algorithmically, not authored in the traditional sense. And it represents a sublime vision of the awe-full mountain vistas of data that confront us every day, like an unknowable but meaningful virtual Alps.
This is an edited version of a paper first presented at "Infinite Possibilities? Seriality: A Symposium," Davis Musesum and Cultural Center, Wellesley College, Saturday, March 6, 2004
