Data Visualization as Documentary Form: The Murmur of Digital Magnitude

Abstract

Data Visualization as Documentary Form:The Murmur of Digital Magnitude Tess Takahashi (bio) In the last decade, data visualization has emerged as a form of documentary. While data visualization is not typically thought of as being in the same category as documentary film, these visually pleasing and often amusing depictions of data culled from the natural and human world pop up repeatedly on our web feeds, often over geographic maps of the world that are detached from their larger context. In other words, data visualization is a rhetorical form that visualizes a relationship between the material world and its particular form of representation. At a time when both documentary forms in the art world and popular feature-length documentaries have flourished, data visualizations have become increasingly popular and powerful forms of documentary. It goes without saying that data visualizations are primarily visual, not aural, representations. Rather than documentary cinema's authoritative voice-over or the voicings of authentic witness, data visualization is typically silent. Yet I argue that data visualization not only has a metaphorical "voice," like the voice of documentary film, but also claims to represent the literal material "voices" of others, even if those voices are seen as inarticulate, unrecognizable, or impossible—the voice of Earth, for example. This essay [End Page 376] asks how we might figure the documentary voice of data visualization as a metaphorical form that "speaks" to the eye. Further, it asks what thinking sonically about data visualization might reveal about voice and documentary. At stake here is what becomes of material embodied voices as they are absorbed into the authoritative voice of what is now our most ubiquitous documentary form: the data visualization. Today, data visualizations often draw on a wealth of "Big Data," which Steve Lohr has described as "the rising flood of digital data from many sources, including the Web, biological and industrial sensors, video, e-mail and social network communications."1 In the face of the incomprehensible scale of Big Data, data visualizations present an image that squeezes data from the big world into the instant of a glance—decades of rising sea levels, one hundred years of increasing temperatures, the extinction of species after species—by drawing on a combination of numerical measurement and photographic evidence. The most common contemporary forms of data visualization are static maps or looping, bite-size gifs, but data visualization has a longer history of attempts to make enormous amounts of data graspable in a glance. This tradition includes the work of Scottish engineer William Playfair, who in the late 1700s created the bar chart, the pie chart, and the line graph—forms that, in the words of nineteenth-century civil engineer Charles Minard, can be used to "speak to the eyes."2 At that time, the supposedly immediate form of communication offered by charts and graphs constituted "visual breakthroughs, innovations that allowed people to see patterns in data that they would otherwise have missed if they just stared at long tables of numbers."3 In more recent history, from the 1980s through the early 2000s, MIT professor Edward Tufte's three influential books on data visualization have chronicled the numerous ways that we have envisioned information over past centuries, ranging from hand-drawn hatch marks over sailors' maps to suggest the turbulence of the wind to clunky, artifact-ridden digital clouds generated by 1990s computer software.4 Today in the age of Big Data, the sophistication of data visualization tools has grown by orders of magnitude, like the sheer volume of data in the world. If the amount of data generated from the beginning of recorded time to 2003 rings in at a measly five billion gigabytes, we now generate that amount every two days.5 How might what William Davies describes as the now "default" collection of Big Data change what we think of as a "voice," and what counts as "speech" for data visualization as a documentary form?6 While data visualization allows the eye to take in large amounts of data in the space of a glance, the embodied human voice cannot [End Page 377] be taken into the ear "at a glance." Steven Connor has described the space of...