Saturday, March 19, 2011

Jeff Thompson: The Poetics of Data


In this series of posts I will chart possible ways to re-contextualize data sets as poetic texts and the ways that writers, visual artists, and musicians have and are using data in their work. Through this shift of focus, “hard” data can be read, analyzed, critiqued, manipulated, and be inspiration or source.

This conversation flows from my own studio art practice. Originally trained as a painter and sculptor, over the past seven years or so my work has transitioned into new media and performance. Along with this shift, I have felt the interest in image and object draining away and an interest in science, language, and technology rise to the front. I spend much less time in museums and galleries than I once did. No more hours in front of Bonnards or Turrells; instead I have found myself spending much more time thinking about things like the phenomenology of earthquakes or how to build microcontrollers.

[ My former obsession: a painting by Pierre Bonnard in the collection of the Minneapolis Institute of Art; next to it an image of the Rayleigh-Taylor Instability, a fluid dynamics model. ]

Vanessa Place sums it up well in the book Notes on Conceptualisms: “… we should consider that from [Barnett] Newman’s nugatory zip to Gerhard Richter’s squeezed woods to last season’s gallery show of airbrushed portraits of the cast of Hogan’s Heroes, visual images are being systematically drained of image, leaving behind the image referent – language”1. If instead the creation of images is increasingly from language – and that language can take the form of a sitcom script or a WikiLeaks spreadsheet or a raw text file of numbers – countless new languages and texts, and as a result images, become available.

Before examining contemporary practices as I will in later posts, it will be useful to re-contextualize historical examples. I say re-contextualize because the history of data and the arts, in the strictest sense of visualization and sonification, is likely no more than 30 years old. By looking at historical works, in this case Lucretius’ long-form didactic poem On the Nature of the Universe, the river drawings of Leonardo DaVinci, composer John Cage’s use of the I Ching, and the visual artworks of the group Art & Language, we can trace a path from scientific observation to spreadsheets and our contemporary networked culture.

[ A portrait of Lucretius ]

Written in the 1st century BC, Lucretius’ On the Nature of the Universe (also referred to as On the Nature of Things) presents an interesting starting point for this discussion: the use of poetry to argue scientific and philosophical knowledge. As an example from Sir Ronald Melville’s translation2, Lucretius explains how lightning occurs:

For indeed I have shown above the hollow clouds
Must contain very many seeds of fire
And must receive many from the sun’s hot rays.
Therefore, when the same wind that has driven them
Into one place together, has squeezed out
Many seeds of fires, and in so doing itself
Has intermingled with the fire, the whirlwind
Finds its way in, whirls round in the narrow space
And in the hot furnace sharpens the thunderbolt.
For the wind is kindled in two ways: by the heat
Of its own motion, and by contact with the fire.
Next when the wind has reached a mighty heat
And the strong impulse of the fire has entered,
The thunderbolt, now as it were ripe, cleaves through
The cloud by a sudden blow, and the heat, shot out,
Lights all the place beneath with flashing flames.
(lines 271-284)
While the exact details of his account are certainly suspect, this accumulation of careful observations is science in its purest form. The chapter to which this excerpt belongs is introduced in a synopsis at the beginning of the book as dealing with:

Thunder, lightning, and thunderbolts. Waterspouts, clouds, rain, etc. Earthquakes. Why the sea is always the same size. Volcanoes (Digression: difficulty of assigning the true cause to all phenomena). Nile floods. Why some places are fatal to birds. Peculiar properties of springs. Magnets. Epidemics3.
A strange yet beautiful selection, to be sure: it is important not to forget that in the case of a data set (as with a poem) it is both the content and structure that are important and hold meaning4.

The formal arrangement of this work, like Homer’s Illiad and Virgil’s Aeneid. On the Nature of the Universe is in dactylic hexameter, a poetic meter consisting of six parts (or feet), each built out of a specific set of short and long syllables. This structure is necessarily lost in translation to English, but can be represented graphically as:

– uu – uu – uu – uu – uu – x
In dactylic hexameter, “ – ” is long, “u” is short, and “x” is either long or short. The term itself is derived from the finger (dactyl) by looking at the knuckled sections from hand to fingertip, as is seen in the drawing above5. A very good example of dactylic hexameter being read aloud can be heard at:
This strict structure being applied to scientific findings mirrors contemporary data storage such as the universal CSV (comma-separated values)6 format. An example of lightning strike data7 written in CSV format would read as follows:
The first line is a header, explaining what each column stands for; the following lines are the actual data (telling us here that the strike took place on the first of February in Winters, Texas). When opened in Excel or a similar spreadsheet program, the data gets cleaned up into more readable columns. This format will be examined in more detail in future posts. As a nice coincidence, the National Climatic Data Center’s lightning reports, like Lucretius, are divided into six sections per line…

[ River investigations by Leonardo DaVinci and some possible combinations of the I Ching symbols ]

By way of segue from ancient Rome to the latter-half of the 20th century, we look to Leonardo DaVinci’s drawings of rivers and John Cage’s compositional works made using chance operations and the I Ching.

DaVinci’s research occurred many centuries before inexpensive sensors and multi-channel data-logging, so instead his investigation into the flow of rivers and waterways was carried out with pen and paper in the form of drawings. Similarly, in the 1950’s composer John Cage turned to the I Ching, an ancient Chinese book used for divination or, in Cage’s case, for inserting chance operations into his compositional practice. The I Ching is essentially a visual database, with different combinations of long and short lines representing different meanings depending on the context8. Like Lucretius, both DaVinci and Cage used analog methods, each more systematic than the next to gather concrete and discreet pieces of information about the world and while these are not the data formats we are used to thinking of, they presage a shift to an information culture that would gather steam in the later 20th century with the emergence of the computer.

Building on the intellectual history of Duchamp and Cage, varieties of what is now broadly called Conceptual Art arose in the late 1960’s in the United States and Europe. The group Art & Language emerged from this multivalent practice. Based primarily in the UK, artists in the United States (most notably Joseph Kosuth) were also considered part of Art & Language’s collaborative and often intentionally elusive practice.

[ Images of the Index project installed at Documenta ]

In the collection White Heat Cold Logic: British Computer Art 1960-1980, Graham Howard makes a case for viewing the work of Art & Language, especially their Index projects, as related to the rise of cybernetics and the database9. The first incarnations of the Indexes, made in early 1970’s, consisted of collected writings of the group, while later versions shifted to massive stockpiles of words and symbols. Shown in filing cabinets and drawers with cross-references or maps pinned to the wall, the viewer could navigate the texts in many different ways; this might be completely random or by following a system of their own devising. The structure of the Indexes was intentionally non-linear and unlike a book, and can be seen as a predecessor to the hyperlinked structure of the web.

[ Printout from Index 03 ]

Later Indexes pushed this structure further into what we recognize as databases and algorithmic writing, taking the form of dot-matrix computer printouts. Index 03 included 64,000 possible combinations of its own text, represented in a shorthand code created for the project. An example line included in Howard’s essay:

C(EX)AB1(X) & N(EX)BC2(X) & N(EX)CD3(X) & N(EX)EF5(X) & N(EX)FG6(X) & C(EX)GH7(X) & N(EX)HI8(X) & N(EX)IJ9(X) & N(EX)JK10(X) & C(EX)KL11(X) & N(EX)LM12(X) & C(EX)MN13(X) & C(EX)NO14(X) & C(EX)OP15(X) & C(EX)PQ16(X).

This interest in merging logical operations and invented linguistic systems with the formal, visual/verbal, and associative – what we can loosely refer to as “poetic” concerns – presents us another possible Rosetta Stone to begin reading data as text.

The projects outlined in this essay suggest some possible models for this re-reading, and the other posts in this series will investigate other possible methods, both historical and contemporary through the lens of artistic practice. It is important to remember, however, that the data sets used by Lucretius, DaVinci, Cage, and the members of Art & Language were different from those generated by a scientist in a lab: these sets were created or found by artists and contextualized as art products. The required leap of logic comes in finding techniques for reading “hard” data not created for artistic purposes.

As I find myself explaining to non-artists and especially those in the sciences, artists of all kinds (visual artists, poets, musicians, etc) are valuable because of the freedom of process our disciplines allow. We are able to think and work associatively, to make great leaps without worrying about always framing a coherent argument, and a willingness to leave untidy edges when we’re finished. This investigation into the poetics of data will necessarily approach the topic from that standpoint.

I leave you with another selection from Lucretius, illustrating the artist’s leap: here contrasting mechanisms of evaporation, atoms and molecules, and erosion in one breath.

And clothes hung up beside a wave-tossed shore
Grow damp, but spread out in the sun they dry.
But how the moisture first pervaded them
And how it fled the heat, we do not see.
The moisture therefore is split up into tiny parts
That eyes cannot perceive in any way.

Then too, as the sun returns through many years,
A ring on a finger wears thin underneath,
And dripping water hollows out a stone,
And in the fields the curving iron ploughshare
Thins imperceptibly, and by men’s feet
We see the highways’ pavements worn away.
(lines 305-315)

- - - - - - - - - - - - -

1. Notes on Conceptualisms. Fitterman, Robert and Place, Vanessa. Brooklyn: Ugly Duckling Presse, 2009.
A cursory look online cannot confirm whether this is a real show, or made up for the purpose of satire.

2. On the Nature of the Universe. Lucretius, translated by Sir Ronald Melville. Oxford: Claredon Press, 1997.

3. On the Nature of the Universe. Lucretius, translation by R.E. Latham. Middlesex: Penguin Books, 1951 (pg. 26).

4. Understanding the structure of the language of data is imperative. While we have done much better unpacking the language of popular media – Lev Manovich’s older but still very relevant Language of New Media is an aptly-titled example of the unpacking of the language of cinema and video art within the context of new media art practice – the critical dialog about the “hard” language of the sciences is underdeveloped.

5. Metrica:

6. As part of its mission to make data more accessible, the US Federal Government’s website mandates that all data sets be stored in “machine-readable”, standardized formats. Other common formats include TSV, KVM, and XML.

7. Lightning data set:,31.9685988
From the National Climatic Data Center’s live lightning data feed:

8. By way of example, three long lines can mean creative force, heaven/sky, northwest direction, father, the human head, strength, and the dragon.
Source: #Trigrams

9. White Heat Cold Logic: British Computer Art 1960-1980. Ed. Brown, Paul; Gere, Charlie; Lambert, Nicholas; Mason, Catherine. Cambridge: MIT Press, 2008 (pg. 334), from the essay: Conceptual Art, Language, Diagrams, and Indexes by Graham Howard.

Oct.,2010- performing a live soundtrack to the
1974 film "Zardoz" at Drift Station.
Jeff Thompson received his BFA from the Minneapolis College of Art and Design and his MFA from Rutgers University. He is currently Assistant Professor of New Genres and Digital Arts at the University of Nebraska, Lincoln.

Thompson has exhibited and performed his work internationally, most recently at SITE Santa Fe, Jersey City Museum, Weisman Art Museum, Hunter College, White Box Gallery, and Museo Arte Contemporaneo in Argentina. Thompson was awarded the Van Lier Fellowship from Harvestworks in 2008 and a commission from Dispatx, an alternative curatorial platform based in Spain and NYC, in 2007. In addition to his studio work, Thompson co-founded the Texas Firehouse, an alternative gallery space in New York City from 2007-2009 and is currently a co-founder of Drift Station Gallery and Performance Space in Lincoln, Nebraska, USA.

No comments: