Critical Visualization: Rethinking the Representation of Data 9781350077270, 9781350077249

Citation preview

Acknowledgements The idea for this book began to take shape during a conversation at the Urban Ecologies 2013 conference in Toronto, where we recognised that our research interests intersected, along with a shared sense of a missing critical discourse. A number of events, discussions, and trips helped the project development, for which we would like to thank some very supportive people at our respective institutions past and present: York University and ocad University (Patricio); Griffith University and University of the Arts London (Peter). A huge thanks goes to our contributors, mentors, mediators and supporters, including: Danah Abdulla, Janet Abrams, Kim Albrecht, Heath Bunting, Tim ­Cadman, Lizzie ­Coles-Kemp, Tom Corby, Tony Fry, Lorraine Gamman, Donna Haraway, Alfredo Jaar, Laura Kurgan, Simon Maidment, Hem Patel, Margaret Pearce, Philippe Rekacewicz, Damon Rich and Jae Shin, Anna Ridler, rndr, Bronwen Robertson, Rebecca Ross, Jeremy Till, and ­Rebecca Wright. Special thanks go to our indefatigable research assistant, Aaron Tucker, illustrator Marc Ngui, and amazing designer Chris Lee. We would also like to thank our editorial assistant at Bloomsbury, Libby Davies, and the editors at Bloomsbury, Louise Baird-Smith, Rebecca Barden and Lee Ripley, who sowed the seed back in 2014. Finally, Peter would like to thank Lana Lê. P.H. & P.D.

220221_CritViz_Book_Full_revisions.indd 6

7/20/22 9:45 AM

220221_CritViz_Book_Full_revisions.indd 7

7/20/22 9:45 AM

I An Introduction to

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 8

7/20/22 9:45 AM

I

n 2020, as the impact of the novel coronavirus pandemic ­began to make news, the phrase ‘flattening the curve’ became a familiar trope: data visualization as a shared cognitive image. For those fortunate enough not to contract the virus or witness its effects on a friend or family member, data was the face of covid-19: as Bruno Latour wrote in Le Monde newspaper in March 2020, this was ‘a virus whose trace is known only to scientists and whose effects can only be understood by collecting statistics’ (Latour, 2020). Embedded in the phrase and the simple graph that accompanied the article — showing the number of cases since the outbreak (y-axis) against the number of days (x-axis) — was a statistical argument for mass behaviour change. A steep epidemic curve showed the unchecked spread of the virus; a flatter curve showed how the reproduction rate (R-rate) of the virus could be reduced, and the burden on healthcare systems lessened, with mitigation measures such as physical distancing, face masks, and lockdowns. But whether the ‘flatten the curve’ visualization contributed to the mass behaviour change it required is a more complicated question, as evidenced in the wildly differing cultural and demographic responses to quarantine, distancing and mask-wearing orders across the world.

9

Figure 1.01  —  Graphic showing ‘flattening the curve’ (by slowing the spread of pandemics) and ‘raising the line’ (increasing healthcare capacity), so that healthcare demands hopefully stay within capacity. RCraig09, licensed with CC BY-SA 4.0.

The paradox of our data-heavy era is that mass data literacy (or ­numeracy) has not followed the masses of data. The kind of scepticism toward scientific evidence that Latour decried in a seminal 2004 essay critiquing critique has become part of an infodemic, fuelled by conspiracy theories and fake news, and disseminated through seemingly un-poppable filter bubbles. Our ability to collect data has far surpassed our ability to read data, ­construct and evaluate evidence and discern facts. Likewise, our ability to critique data visualizations, evaluating their integrity, their sources and the political-economic contexts of their production have not kept up with the pace of output. The veil of neutrality drawn over data v­ isualizations becomes more opaque and impenetrable when the socio-technical infrastructure that produces them is obscured behind

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 9

7/20/22 9:45 AM

10

­so-called neutral technologies: human decisions sealed behind computa­ tional logic (i.e. algorithms). This is one reason why this book makes a case for a critical approach to visuali­zation, to confront a familiar problem exacerbated by the filter bubble effect (Pariser, 2012). On the one hand, there is the blunt deployment of data visualization to support powerful agendas. On the other, there is a blind ­distrust of visualizations that do not conform to a given prejudice. To counter this effect, we advocate both the rigorous critique and unfolding of data visualization and the practice of creating visualizations that are counter to, or critical of, dominant narratives. To paraphrase the geographer Denis Cosgrove’s remarks on maps: always make visualizations, always question visualizations (Ross, 2006). This book is written for those whose interest, business or practice is the ­visualization of data, as well as for those who study data and its impact in fields such as human-computer interaction, science and t­ echnology studies, criti­cal data studies and digital humanities. As authors, ­researchers, designers and educators in communication design and media p ­ ractice, we are interested in how data and its collection, m ­ anipulation and presenta­tion both empowers and disempowers the people and the interests that are represented (or misrepresented) by the data. Our premise is that data visualization has flourished in terms of technique and ­creativity, but that this has come at the cost of critique. As we shall discuss, the idea of what it means to be critical has itself been critiqued, but we ­remain convinced that a robust field of data visualization warrants a ­robust c­ ritical discourse. In the field of communication design and graphic design, the deficit of robust critical discourse becomes particularly apparent when contrasted with related fields such as critical data studies and critical cartography (to which we frequently turn for cues on method). In graphic design circles, as in mainstream media, the burgeoning of data visualization across disciplines has brought with it a form of information fetish or aestheticizing of information that tends to stunt the growth of critical discourse. Entangled with this fetish is the position, influentially demonstrated by Edward Tufte, for example, that the ‘efficient’, ‘clear’ and ‘excellent’ visual display of quantitative information will reveal the truth and help us solve all manner of social and technical problems in a fair, unbiased way (Tufte, 2001, pp. 13–15). The achievements of dataviz proselyte Hans Rosling, for example, included showing how data visualization can counter popular misconceptions about global health, but did not emphasize critical thinking around the circumstances of its gathering, selecting and presenting.

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 10

7/20/22 9:45 AM

11

Figure 1.02  —  Hans Rosling presents The Joy of Stats. © bbc

A CR I T ICA L F RA M E WOR K To establish a framework for critical visualization, we might begin with a series of statements, which we explore in the subsequent chapters: 1. Data visualizations are part of larger data assemblages — complex socio-technical systems, composed of many apparatuses and elements that are thoroughly entwined, whose central concern is the ­production of data (Kitchin, 2014, p. 24). As we discuss in Chapter 3, a good example of a data assemblage is the census, a large amalgam of apparatuses and elements that shape how a census is formulated, administered, processed, communicated and how its findings are employed. 2. Data assemblages enhance and maintain the exercise of p ­ ower within society, particularly where they assemble institutions, regulations, laws, administrative measures, scientific statements and moral propositions (paraphrasing Michel Foucault’s notion of the ­dispositif ). One way in which a data assemblage maintains power is in the ­deployment of categories to organize what, or who, gets counted and how. As discussed in Chapter 3, once a person or group of people is classified, for example, as autistic, affluent, bi-racial or healthy, the classifi­cation has an effect on that person or group, which philosopher Ian Hacking terms a ‘looping effect’. In this way, data classifications ‘make up people’ (Hacking, 2006). 3. Data visualization critique reveals the assemblage, or explicates the assumptions, apparatuses and elements that are entwined in the form: critique does not simply pass judgement on a visualization’s formal clarity or excellence (although this certainly helps us make better v­ isualizations). Here we lean on Foucault’s account of critique: ‘A ­critique does not consist in saying things aren’t good the way they are. It consists in seeing on what type of assumptions, of familiar notions, of established, unexamined ways of thinking the accepted practices are based’ (­Foucault, 1981/2002, p. 456). In revealing the assumptions and unexamined ways of thinking, critique also reveals what or whose interests are being excluded.

Chapter I

220221_CritViz_Book_Full_revisions.indd 11

Critical Visualization

7/20/22 9:45 AM

12

4. Data visualizations need to be situated in time and space to ­counter the visual rhetoric that connotes omniscience, or the ‘god trick’ (­Haraway, 1988, p. 581), that suggests data has been gathered, and is ­being viewed, from nowhere — a position often confused with objectivity. One way to situate a visualization is to draw from the expertise of those familiar with the data site and processes of its generation. We can also borrow a framework developed in Actor-Network Theory to analyse how things take shape through identifying, enrolling and mobilizing other interests. For example, in Chapter 2, we explore how Florence Nightingale’s famous polar area graphs developed as part of her process of accruing support from politicians, military chiefs and newspapers, making herself indispensable to the network. 5. Critical visualization practice uses assemblages, e ­ xplication and an analysis of power to create counter-narratives or ­criti­cal v­ isuali­za­tions. This is vital because — as is evident from point 2 of the ­frame­work — data visualizations and data assemblages function to e­ nact power and, as such, tend to serve the interest of the powerful. As ­Catherine ­D’Ignazio has eloquently argued, ‘there is profound ­inequality between those who are benefitting from the storage, collection, and analysis of data and those who are not’ (D’Ignazio, 2017, p. 6). Three of the terms deployed here — assemblage, situated knowledge, Actor-­ Network Theory (ant) — indicate the discourses we are adapting to the task of critical visualization. The concept of an assemblage goes back to Gilles Deleuze’s and Felix Guattari’s (1983) account of the heterogeneous entities — humans, non-humans, organizations, laws, things — that come together to produce power or agency through their relations. The situated data visualization refers to Donna Haraway’s (1988) account of the ‘situated knowledges’ produced by subjects who acknowledge the contingency of their own position in the world, and, in doing so, produce greater ­objectivity than those who claim to be neutral. Actor-Network ­Theory ­describes (confusingly) not a theory, but a set of ‘material semiotic’ tools and methods of analysis for explaining how heterogeneous relations between human and non-human actors are continuously forming and reforming to enact power (Law, 2007). The tools of ant help us explore how a data visualization translates the disorder of the world into patches of ­orderliness — by ­imposing categories, defining, counting and sorting entities into those categories, and then deploying

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 12

7/20/22 9:45 AM

13

Figure 1.03  —  Assemblage. Marc Ngui

Chapter I

220221_CritViz_Book_Full_revisions.indd 13

Critical Visualization

7/20/22 9:45 AM

14

that order to influence behaviour and shape policy. ant emerged in the early 1980s and, unlike prevailing literary and linguistic approaches to meaning-making, focused on empirical case studies in science and technology to tease out how social and material things (‘actors’) — such as electric ­vehicles, door-closing mechanisms, scallops and Louis Pasteur’s ­laboratory — enacted power by identifying, enrolling and mobilizing social-­material relations (Callon, 1986; Latour, 1992; Law, 2007). Because it characterizes power as networks of associations, ant does not tend to account for factors outside of those associations (e.g. a pervasive ethno-­ centric bias) that are nonetheless influential (Müller, 2015). Assemblage theory, however, accounts for such a bias, and helps explain how data visualization enacts power. In short, we are applying to data visualization some of the ideas and methods from the last three decades of scholarship in science and techno­ logy studies (sts) and geography (Müller, 2015). Our aim is to show how visualization both represents and enacts interests, and how it structures and imposes power relations. Rather than evaluating a visualization’s stand-alone excellence, clarity or integrity, then, we can use the framework detailed earlier to situate and contextualize a visualization’s claims to authority and to remind us that every visualization assembles an alliance — or ragtag army — of interests, each with their own history. Krause (2019) sets out a case for writing ‘data biographies’ with questions probing the authorship, purpose, m ­ ethod of collection, intended audience, impact and limitations of every data set. In a similar vein we can introduce a visualization to our framework with questions loosely derived from the basic investigative method taught to trainee journalists: the who, what, when, why and how? The sequence that follows leans rather more heavily on the what, because, as we shall see, unravelling the conditions of a visualization’s development nudges us to consider the ant-ian ‘moments’ of identifying, enrolling and mobilizing that accompanied its production and pre-production. So we arrive at who, when, why, what, what, what and what: Who made the visualization and for whom? When did they make it? Why did they make it? What social/cultural conditions was it made under? What belief systems is it reinforcing or challenging? What processes, or translations, preceded its production? What has been excluded?

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 14

7/20/22 9:45 AM

To apply the questions to an example, let’s turn to a popular source for the aestheticizing of information as a form of cultural production: David ­McCandless’ much-cited book Information is Beautiful (McCandless, 2012) and its cover image, the ‘Colours and Culture’ visualization.1

15

Who made it? David McCandless, a British data journalist. When? In 2009, just as ‘infosthetics’ and dazzling graphics were g­ aining prominence, fuelled by widely available data mining, scraping and ­mapping technologies. W hy? The full title, ‘Colours and Culture: The Meanings of Colours around the World’ indicates that the intention was to demonstrate how colours mean different things in different cultures (e.g. red means good fortune in China, but debt in the West). In what social/cultural conditions? The reassuring idea that data visual­ ization can help make sense of complex, bewildering, interconnected problems becomes more appealing when an expert guide offers his or her services. As a data journalist, McCandless has developed a practice of data interpretation using spreadsheets, and subsequently design tools, to ‘tell stories’ with data (McCandless, 2015). Any entrepreneur requires a level of showmanship, and we can speculate that the ‘Colours and Culture’ visuali­zation was produced more for its form than its function, as we shall explain shortly. W hat belief systems is it reinforcing/challenging? On close inspection, the visualization reveals its dependence on tertiary sources, namely Wikipedia and ‘general web’ (presumably Google). This might indicate the strangely skewed sense of world geography evident in the infographic: there are two continents represented (Africa, Asia), three subcontinents (America, South America and Eastern Europe) two religions (­Hindu and Muslim), one Indigenous group (Native American) and two ­countries (Japan and China). What has been excluded? There is no Western European ring and the ­designation ‘Asian’ is unclear. In fact, the logic behind the categories in the key is unclear, if not entirely ad hoc. In the graphic, ‘America’ on the outer ring, has 53 entries. ‘South America’ on the innermost ring, has just five entries (red for danger and success, green for death, purple for mourning, blue for trouble).

1 F  or a viewable and zoom-able visualization, see p.7 of the book sampler: http://infobeautiful2.s3.amazonaws.com/IIB_Sampler_ Dec2016.pdf (accessed 21 May 2021).

Chapter I

220221_CritViz_Book_Full_revisions.indd 15

Critical Visualization

7/20/22 9:45 AM

16

The troubling impact of a closer inspection of ‘Colours and Culture’ is that the visualization’s claims to authority, objectivity and even credibility fall apart amid the ruins of its incoherent logic. At risk of lingering too long amid the rubble, we might, prior to departing, consider the apparatuses on which it depended for its appeal to authority: Wikipedia, ­specifically the world’s largest online site of encyclopaedic knowledge production and — presuming that ‘general web’ means Google — the world’s largest search engine. The national, corporate and gender biases inherent in Wikipedia’s declared ‘neutral point of view’ (npov) approach to entry-writing have been the subject of extensive debate (e.g. Menking and Rosenberg, 2020). Similarly, the notion that search engines like Google provide a level playing field for ‘all forms of ideas, identities and activities’ is confronted by Safiya Noble, for example, who exposes the social problem of data discrimination and the oppressive effects of search algorithms (Noble, 2018). One fierce criticism of McCandless’ Information is Beautiful platform is levelled by designer Neville Brody, who, in a debate on the bbc, argued that ‘the idea of making beautiful information misses the point: trying to make information pretty can hide the core message. Information is political’ (bbc, 2010). Brody’s critique supports our premise that the collection, sorting, arrangement and presentation of data empowers some and disempowers others (point 2 of our framework) — to create a beautiful visualization that purports to show how colours mean different things to different cultures, but to populate it with data gleaned from one dominant region using apparatuses whose effect is to perpetuate a particular epistemology or particular worldview, is to effectively undermine the claim to pluralism. In the McCandless example, the who-when-why-what questions served to poke behind the glossy facade of a visualization to reveal the larger assemblage, or actor-network, to which it relates. Law (2007) has argued that there is little difference between the assemblage and actor-network, but while the actor-network approach insists on symmetry in exploring the strategic, relational and productive character of heterogeneous entities, assemblage points to the imbalance of power that results. The imbalance of power also provides a rationale and motive for m ­ aking new, critical visualizations (framework point 5). Those equipped to deal in data generally have power and resources on their side. D’Ignazio identifies the imbalance between the data-haves and data-have-nots in her case for engendering a ‘creative data literacy’ that benefits the ­traditionally disempowered. Because it is state and corporate actors who possess the resources to collect, store and analyze data, individuals (e.g., citizens, community members, professionals) are more likely to be the subjects of data than to use data for civic purposes. There is a strong case to be made for cultivating data literacy for people in non-technical fields as one way of bridging this gap. (D’Ignazio, 2017, p. 6)

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 16

7/20/22 9:45 AM

As we demonstrate in Chapter 3, one rhetorical function of a data visuali­ zation is to erase or obscure the decisions and machinations that accompanied its development. For the statistical display of quantitative information to be clear and ‘excellent’, it must eliminate all uncertainty, ambiguity and preparatory work including consultation and participation. Tufte’s oeuvre demonstrates this effect, and the process of concealment in visualizing data is strikingly analogous to the way human input is often concealed in algorithmic decision-making. Cathy O’Neil’s aptly named book Weapons of Math Destruction provides a series of case studies where data and algorithms were deployed, often blithely, to discriminate against their subjects, such as a Washington school district fifth-grade teacher who was fired because her performance was red-flagged by a new algorithm that failed to take into account key contextual information; the teacher went on to a successful career in a private school (O’Neil, 2017, p. 7). Our argument is not to make all data visualization messy and inconclusive, but to draw attention to the rhetorical effect of concealing decisions behind decisive graphical user interfaces.

17

This book, then, makes a case both for critiquing visualizations and for making critical visualizations. Both activities draw from the framework above, and in some cases subvert the rhetorical appeals to authority implicit in rules and guidelines on user-friendly or efficient visualization. In both aspects of critical visualization, we take cues from critical carto­ graphy, which, as we explore in Chapter 2, emerged in the post-SecondWorld-War years, out of a discipline that had followed a similar discursive framework as data visualization: a ‘cartography of progress’ (J. Crampton, 1994, p. 2) governed by internal technical interests and disengaged with ideological concerns (Kitchin and Dodge, 2007). DE F I N I NG T H E F I E L D The field of data visualization is itself subject to different disciplinary per­spectives and conceptions of its scope. Disciplines of relevance here are media studies, statistics, computer science, geographic information science and design — all of which have differences as well as commonali­ ties in their approaches to visualization and its critical discourse. In his recent cultural history of data graphics in news media, Murray Dick uses the term ‘infographic’ interchangeably with ‘data visualization’, arguing (after Alberto Cairo) that the two concepts exist on a spectrum from

Chapter I

220221_CritViz_Book_Full_revisions.indd 17

Critical Visualization

7/20/22 9:45 AM

18

description to ­ex­planation. Outside of the news context, however, the term ‘infographics’ is quickly disparaged for its association with news­ papers’ tendency toward distorted graphics that exaggerate the recent rate of change (Dick, 2020). In statistical approaches to data visualization, the field is characterized by the pursuit of efficient methods for displaying quantitative data: Fienberg categorized graphical methods for illustrative, analytical, computational and decorative purposes (Fienberg, 1979, as cited in Dick, 2020) — the ­latter category sitting uncomfortably close to Tufte’s disparagement of ‘chartjunk’. In human-computer interaction (hci) and geographic information science (gis), the emphasis is on how the human brain best processes and analyzes information, supported by empirical measurement and user testing. Johanna Drucker’s critique of this, the dominant a­ pproach to interaction design, is that it is ‘reductively mechanistic’: Its goal is to design an environment to maximize efficient ­accomplishment of tasks — whether these are instrumental, analytic, or research oriented —  by individuals who are imagined as autonomous agents whose behaviours can be constrained in a mechanical feedback loop (Drucker, 2014, pp. 151–152).

In our own design-centred work, we have argued that dominant data visualization discourses neglect critical contexts. We have proposed new cate­ gories for contexts of use that elevate artistic visualization — that which reflects on cultural conditions, challenges dominant assumptions, and offers new modes of representation — alongside journalistic and scientific contexts of visualization (Hall, 2011). We have also argued for the use of the term ‘diagrams of power’ to capture the ways in which a diagram can be said to be both visual and non-visual, representing data and communicating ideas on the one hand, and arranging bodies and things on the other (Dávila, 2019). This sense in which design is ontological — enacting a way of being in the world — has been argued by Willis (2006) and Fry (2012), who have critiqued ‘instrumentalist’ tendencies that evaluate ideas, concepts and discourses in terms of their functional (and economic) success. The dominance of instrumentalism, or what Murray Dick calls a ‘function­ alist-idealist’ approach to data visualization, is exemplified in the central figure of Jacques Bertin, whose ‘efficiency principle’, introduced in the book Semiology of Graphics (Bertin, 1967/2011), has sustained a fifty-year ­influence on data visualization since its publication (Harvey, 2019). Bertin drew from his experience as a cartographer to establish a system using seven variables for the graphic organization of information: size, position, colour, orientation, shape, tone and texture. The reduction of complexity, the use of a grid for establishing scale and measurement of space, and the flattening of volumes (e.g. the spherical planet) are long-established premises of cartography, but Bertin’s system facilitated the further abstrac­tion of information into a typology and syntax with its own internal logic, antici­ pating the arrival of geovisualization. For Bertin, visual logic deter­mines

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 18

7/20/22 9:45 AM

the communication of signs and symbols; propositions can be created by the rigorous application of the signs in a system. But his system is ‘monosemic’ (meanings are singular), which leaves no room for symbolic or cultural contexts or audience (Dick, 2020).

19

Where cartography had tended to conceive of maps as tools for storing and communicating spatial information about the environment, Bertin’s system focused on their role as methods for analysing and processing ­information (MacEachren, 2004). This laid the groundwork for the codification of graphic efficiency, simplicity, and consistency in Tufte’s works on statistical graphics, and a computer science-based approach to data visualization, evident in, for example, Ben Shneiderman’s (1996) golden rules of inter­face design, and Per Mollerup’s (2015) guidelines on data design. A contrast­ing approach to Bertin’s positivist semiotics would be a social semiotic perspective on how signs establish and change their meaning in different times and cultures (Hodge and Kress, 1988). At the heart of cartographic principles, and underlying Bertin’s approach, is the idea of orientation: that the person reading the map can locate themselves, and the person reading the visualization can effectively make comparisons and inferences without feeling that the rules of scale and consistency — or the data represented — are changing. But abstracted from the task of representing measurable space, Bertin’s system and its successors in data visualization imply a point of view without actually revealing one. The authority of a standardized system for representing data, like the standardized cartographic approach of flattening the earth and overlaying a longitude and latitude grid, has the effect of making the subject disappear. When satellite imaging technology is combined with Bertin-ian s­ ystems of visual logic, as in Geographic Information Systems (gis), where data (e.g. traffic or weather patterns, or census data) are overlaid with spatial information, the effect is compounded, since the ­visual syntax has no capacity to show the origins, location or authorship of the data ­being visualized, yet it delivers the view from nowhere with absolute authority. Laura Kurgan has argued that, ‘For e­ very image, we should be able to inquire about its technology, its location data, its ownership, its legibility, and its source. To facilitate that inquiry, an image and its associated data should remain closely linked’ (Kurgan, 2013, p. 26). The widespread public use of satellite imaging and the Global Positioning System (gps) on mobile phones and in-car navigation has exacerbated the loss of a sense of scale. Whereas the scale of a printed map is fixed, on a screen it slides continuously with the simple zoom action of pinching one’s finger and thumb. In her many projects that problematize the authority and singular gaze of the satellite view, Kurgan has highlighted the work involved in rendering the view from nowhere, which includes extensive post-production work, to ensure the satellite view is flattened, seamless

Chapter I

220221_CritViz_Book_Full_revisions.indd 19

Critical Visualization

7/20/22 9:45 AM

20

and lit uniformly. gps itself embodies an ‘unhinged’ sense of stable and fixed location, since it comprises an orbiting network of 24 military satellites whose data must be reconciled with adjustments informed by Einstein’s theories of relativity. As Kurgan puts it (citing Peter Galison), ‘gps and a whole new set of technologies linked to it have introduced, or hyperbolized, a profound decentering or disorientation (Kurgan, 2013, p. 16). Defining the field, then, demands a trans-disciplinary line of inquiry that never loses sight of the socio-technical nature of data design conventions, their histories and interests. Crossing disciplines is easier when we trace genealogies. Geographer Jeremy Crampton points out that the origins of gis lie in the thematic maps and statistics — the ‘technologies of management’ — that emerged in the nineteenth century (Crampton, 2010, p. 10). Tufte and cartographic historian Arthur Robinson both concur that the origins of thematic maps are tightly entwined with the history of carto­ graphy, extending back at least to Edmund Halley’s map of trade winds (Tufte, 2001, p. 26; Robinson, 1982, p. 46). Denis Wood extensively tracks the post-Second-World-War invention of thematic mapping as an attempt to professionalize and securitize an academic interpretive mode of analysis distinct from the more instrumental ‘base mapping’ (D. Wood, 2010, pp. 121–126). As we have argued previously (Hall, 2008, p. 122), maps, thematic maps and visualizations are all projections of particular cultures at particular times and places, from Medieval mappae mundi, which divided the earth into three continents with the Holy Land at the centre (Robinson, 1982, pp. 9–11), to nineteenth-century disease and morbidity maps, which reflected concern with the social conditions of crowded cities — and speculated dubious correlations between geography and developmental disorders (e.g. Robinson, 1982, pp. 174–176). By the same token, network visualizations of the late twentieth century privilege a popular notion of the network as abstracted from physical space, fetishizing the spectacular complexity of technicoloured links and nodes in empty 3D space (Hall, 2008, pp. ­120–124).

Figure 1.04   —  Edmund Halley’s pioneering map of trade winds, 1686. Edmond Halley

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 20

7/20/22 9:45 AM

21

Figure 1.05  —  Barrett Lyon’s Opte map of the internet, 2003. The Opte Project

The terms ‘data’ and ‘visualization’ also warrant some unpacking. As ­Daniel Rosenberg writes, in recent histories of science, data are commonly charac­ terized as incontrovertible facts, reflecting the etymology of the word, which we consider in Chapter 3. The earliest use of the term data is in a 1646 theological text, where the word refers to a list of theological pro­ positions ‘accepted as true for the sake of argument — that priests should be called to prayer, that liturgy should be rigorously followed, and so forth’ (D. Rosenberg, 2013, p. 20). In seventeenth-century scientific tracts, the term data was used to identify facts that were, by agreement, ‘beyond ­argument’. The legacy of this connotation is present in the rhetorical function of the word today: data means that which is arrived at through experiment and observation, but it is also the premise from which we start—‘that which is given prior to argument’ (D. Rosenberg, 2013, p. 36). This can be contrasted with the word fact, derived from the Latin facere, ‘to do’. Hence, a fact is something that has been done, or exists. Facts exist, but data is given.

Chapter I

220221_CritViz_Book_Full_revisions.indd 21

Critical Visualization

7/20/22 9:45 AM

22 The basis for facts and data is evidence, a word that has a Latin root that suggests more active human witness, coming from the Latin verb ­videre, ‘to see’. The connotations of the three words can thus be understood in terms of their contexts of contemporary use: facts exist, evidence is based on what we see (and know) and data, based on facts and evidence, are assembled to make a case or an argument. Or as Rosenberg puts it: ‘facts are ontological, evidence is epistemological, data is rhetorical’ (D. ­Rosenberg, 2013, p. 18). We might like to believe that the visualization of that which is given is the less political practice of making data clear, accessible, and even beautiful, but the term ‘visualization’ comes with a baggage of its own. The term, again deriving from videre, emerged in the nineteenth century to denote mental images of things that could not be seen (D. Rosenberg, 2013). The contemporary meaning of the word, according to Orit Halpern’s extensive study of the development of vision and reason, straddles the actual practices of depicting and modeling the world, the images that are used, and the forms of attention by which users are trained to use interfaces and engage with screens. (Halpern, 2014, p. 23)

Halpern argues that visualization has shaped thought by training our perception, defining reason, and — by extension — transforming notions of government­ality. Governmentality, a term conceived by Foucault to denote mechanisms for regulating populations, in the twentieth century became tied to data, calculation and neoliberal economics. At the same time, visualization came to describe the processes of bringing into sight that which is not already present: Visualization slowly mutated from the description of human psychological processes to the larger terrain of rendering practices by machines, scientific instrumentation, and numeric measures. (Halpern, 2014, p. 23)

In short, the visualization of data can be understood as a technique or technology for disciplining and managing populations and bodies, training them to see and understand the world through particular — economic, statistical, cognitive — lenses. One focus of this book is how visualization can also be used to reveal the world through different lenses, but typical­ ly, design’s role in the data visualization assemblage tends to be one of complicity with power, sealing unexamined ways of thinking behind ­powerful visual performances. Halpern shows, for example, how visualization environments such as the ibm pavilion at the 1964 New York World’s Fair played the role of influencing public acceptance of computing, and contemporary ideas about psychology, attention and space (­Halpern, 2014, p. 143). ibm and designers Charles and Ray Eames, whose firm designed the t­ wenty-two-channel installation ‘Think’, which projected a ­simultaneous flow of apparently disconnected film and animation clips,

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 22

7/20/22 9:45 AM

sought to get viewers to make non-linear connections between images at the speed of thought, to ‘think like a machine’. The effect and affect, argues Halpern (2014, p. 83), was to create what György Kepes called an ‘experiential’ form of vision.

23

Figure 1.06  —  ibm Pavilion 1964 World’s Fair. Courtesy of International Business Machines Corporation, © International Business Machines Corporation

Figure 1.07  —  ‘Think’ multi-screen film by Charles and Ray Eames at the ibm Pavilion, 1964 World’s Fair. Courtesy of International Business Machines Corporation, © International Business Machines Corporation

Data visualization, then, is much more than the technical pursuit of clarity through the excellent display of objective information. Visualization structures a world and ways of looking at it, disciplines forms of attention and, as we argue in Chapter 6, engages viewers to see the world differently. As we discuss in Chapter 3, data is neither objective or raw — Lisa Gitelman amply demonstrates this (Gitelman, 2013). Data is always situated. Data is always gathered at a certain time in certain places, by certain organizations with a certain purpose.

Chapter I

220221_CritViz_Book_Full_revisions.indd 23

Critical Visualization

7/20/22 9:45 AM

24

The mechanistic flavour of data visualization discourse is located as a ­pri­marily Western bias by Drucker, whose book Graphesis makes a well-­substantiated case for more interpretive, humanistic approaches to ­visual forms of knowledge production. Drucker notes that visual forms of know­ledge production (i.e. ‘graphesis’) have played a subordinate role to logo­centric and numero-centric forms of knowledge in the West, and a methodological foundation for graphesis has to be ‘cobbled together’ from an array of contributing intellectual traditions, including: knowledge as vision; languages of form; universal principles of design; Gestalt principles of perception; the semiotics of graphics; techniques of framing and reading; issues of computation vision and the typology of graphic forms. (Drucker, 2014, p. 20). L O OK I NG AT V ISUA LI Z AT ION BE YON D W E S T E R N ­PA R A DIGM S In his various writings on decoloniality, Walter Mignolo argues for ‘delinking’ from dominant Western paradigms of thought and practice, and ‘re-­existing’ through praxis, the idea that theory is doing and doing is thinking (Mignolo and Walsh, 2018, p. 7). Decolonial thinking and doing aim to delink from the epistemic a­ ssumptions common to all the areas of knowledge established in the Western world since the European Renaissance and through the European Enlightenment. R ­ e-­existence follows up on ­delinking: re-existence means the sustained effort to reorient our human communal praxis of living. (Mignolo and Walsh, 2018, p. 106)

To delink data visualization from its dominant Western paradigms requires a critical interrogation of the inherent assumptions: whose data? What interests are being furthered in this view, and what interests are being silenced? Here, critical visualization can learn from the work of geographer J.B. Harley, who introduced the concept of ‘cartographic silences’ — the knowledges buried by the territorial imperatives of the map-makers (John Brian Harley, 2002). Developing new approaches to visualization requires that silenced forms of knowledge are given an ­opportunity to ‘re-exist’ in new data, in new ways of seeing old data and  in new representations of that data. An example of how a twofold approach to critically thinking about and critically making visualization works is provided by two contrasting projects that responded to the urgent problem of waste generation (according to the World Bank, in 2050, waste generation will reach 3.40 billion tons annually, outpacing population growth by more than double). The Trash|Track project, launched in 2009 as part of Massachusetts Institute of Technology’s (mit) Senseable City Lab, posed the question ‘Why do we know so much about the supply chain and so little about the

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 24

7/20/22 9:45 AM

“­removal chain”?’ Researchers invited 500 participants to attach an electronic tag to an item of trash, then tracked the movement of these 3,000 objects through the United States, using gps and mobile phone data. These movements were then represented as elegant visualizations resembling maps of flight paths across the usa, with items colour-coded according to ­materials (electronics, glass, paper, textiles, etc).

25

Figure 1.08  —  mit’s Trash | Track project Composite Map. mit Senseable City Lab

Figure 1.09  —  mit’s Trash | Track plastic container. mit Senseable City Lab

Chapter I

220221_CritViz_Book_Full_revisions.indd 25

Critical Visualization

7/20/22 9:45 AM

26

In 2013, the Buenos Aires-based group Iconoclasistas held a number of mapping workshops with residents and workers in an area of the city next to one of the largest garbage dumps in Argentina. The resulting fold-out map, La República de los cirujas (trans. The Republic of Garbage Pickers), features the recycling plants, community centres, cooperatives, schools and the different materials (colour-coded for type, as in the mit visualization) that were being retrieved by the cartoneros — the city’s large informal trash collectors and sorters who account for most of the city’s recycling.

Figure 1.10  —  La República de los cirujas, a map of the informal economies around a garbage dump in Buenos Aries. Cartografía elaborada junto a vecinos y vecinas del barrio 8 de mayo, Costa Esperanza, Libertador, Independencia, 9 de julio, Carcova y Villa Hidalgo. Biblioteca Popular La Carcova, Centro Comunitario 8 de mayo, Centro Cultural Diego Duarte, Asociación La Colmena, Bachillerato La Esperanza, FM Reconquista, Parroquia Inmaculada Concepción, Puntos de encuentro; y los trabajadores y trabajadoras de las plantas sociales Ecomayo, 3 de Mayo, Todos Reciclados, Tren Blanco, Sueño y Progreso. Organizado por Verónica Gago y Lalo Paret, Lectura Mundi, Universidad de San Martín, junio a octubre de 2013.

Both the mit and Iconoclasistas projects set out to make visible a process that has been made invisible in the modern world: the infrastructures that exist to manage municipal solid waste, and — by extension — the industrial, economic and ecological after-lives of the things we throw away. The resulting visualizations, however, are strikingly different. mit Trash|Track presents a visual study of geographical distribution of types of waste across the usa: the patterns are interesting, sometimes surprising, but recalls a map of flight paths, suggesting a detached, aerial view of the very early stages of an investigation of the inefficiency of the removal chain. Unsurprisingly, of 205 electronic devices tagged in the project, most ended up in the US’s port cities, where they were shipped abroad, primarily to China (Duarte and Ratti, 2019). Many studies have shown how global waste is disproportionately produced in the urbanized Global North and exported to the Global South, where those who live and work with it suffer health repercussions. This picture is implicated but left outside of the scope of the project, which culminated primarily in research papers focused on the possibilities of pervasive computing, and exhibitions in South Korea, Switzerland, Taiwan, UK and usa.

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 26

7/20/22 9:46 AM

By contrast, the La República de los cirujas outcomes present a street-level view of waste management, through a participatory mapping with workers and residents close to the ceamse (La Coordinación Ecológica Área Metropolitana Sociedad del Estado or, translated: The Ecological Coordination of the Metropolitan Area of ​​the State Society) garbage dump. As Agnese Codebò observes, the populations subsisting from the landfill have been rendered invisible by Google maps and marginalized by the Argentine state, despite their essential role in managing the city’s waste. Historically, the cartoneros, like the cirujas (garbage pickers) who preceded them, have been stigmatized by the state and mainstream media, and have been marginalized by policies such as a 2018 Zero Waste Law, which reinstituted waste incineration in Buenos Aires (Codebò, 2019, p. 31). According to a 2019 report (Codebò, 2019, p. 31) there are approximately 15,000 cartoneros in Argentina, who sort through trash for materials that can be sold to recycling centres (the name comes from cartón, or cardboard, the most commonly collected material).

27

The data visualized in the cartoneros project is qualitative rather than quantitative, and — a key point of difference with the mit visualization — is produced in collaboration with the communities it maps. One might object that the volunteers in Trash|Track did participate in the map-making, but ultimately their agreement was to facilitate the synoptic view of satellite and mobile phone network tracking. Technological solutions to waste-tracking are entirely absent from the La República de los cirujas map, which visually recalls the aesthetics of the mid-­twentieth-century Isotype work of Marie and Otto Neurath, with a significant amount of narrative text, representing the working and living conditions of the trash-pickers, as well as their absence from official records and the lack of basic electricity, water and gas services in the neighbourhoods in which they live. In effect, a critique of both visualizations reveals their situatedness: the mit map, produced with partnership support from the US-based multinationals Qualcomm, Waste Management, and Sprint, provides a proofof-concept for tracking waste that might or might not attract further sponsorship should waste-tracking become a lucrative enterprise. The La República de los cirujas map was made in collaboration with the vulnerable populations invisibly supporting waste management in Buenos Aires for the purpose of advocating for better conditions and a better public understanding of their importance. Codebò categorizes it as a work of counter-­ mapping, which she aligns with Mignolo’s call to delink from dominant Western paradigms of thinking in order to understand the world from other perspectives.

Chapter I

220221_CritViz_Book_Full_revisions.indd 27

Critical Visualization

7/20/22 9:46 AM

28

A counter-mapping is a form of what we are calling a critical ­visualization — a visualization that presents an alternative way of seeing and being in the world: We can understand counter-maps as a result of practices that ­resist the c­ olonizing paradigm by proposing other possible ways of living. (Codebò, 2019, p. 46)

A critic might argue, echoing Audre Lorde’s famous remark, that ‘the master’s tools will never dismantle the master’s house’ (1984/2018) — that data visualization is a technology of management (Crampton, 2010), which cannot be extracted from the colonial territorial imperative. However, as Lewis Gordon and Jane Anna Gordon argue, it is possible to transcend rather than dismantle Western ideas through building new ‘houses of thought’ (Gordon and Gordon, 2006, as cited in Mignolo and Walsh, 2018, p. 7). The critical traditions of Western thought are, in this respect, tools that both identify that unexamined assumptions are situated in a Western lineage, and then are turned to new constructions. Mignolo and Walsh upend the order of this sequence so that it is possible to first make new constructions outside of the Western traditions and then use them to critique unexamined assumptions. They see this upending as a decolonial project built on cultural traditions of praxis — ‘understood as thought-­reflectionaction, and thought-reflection on this action’: By disobeying the long-held belief that you first theorize and then apply, or that you can engage in blind praxis without theoretical analysis and vision, we locate our thinking/doing in a different terrain. (Mignolo and Walsh, 2018, p. 7)

The project of decolonizing data visualization is beyond the scope of this book, but as a first step, we aim to set the stage for alternative approaches to and perspectives on visualization: ones that foreground knowledge and ways of being from outside of the dominant Western, ­rationalist paradigms. To begin, this project requires delinking and re-existing through thinking/doing. S Y NOP SIS OF T H IS BO OK With this idea of thinking/doing as a means to develop a different terrain of critical visualization, we have organized the following chapters to work through and work out the constituent parts of a critical visualization framework: its disruptive history, its account of how data is made (not found or given), its position on biopolitics, the quantification of people and city governance; its account of re-presenting and aesthetics, and its case for alternative critical perspectives.

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 28

7/20/22 9:46 AM

Chapter 2, ‘Disruptive Histories’, first explores the assumptions and cultural history of dominant approaches to visualization history: a linear, positivist and teleological ‘history of progress’ written primarily in Europe and North America, and featuring the familiar names of ‘great white men’: a dominant rhetoric of neutrality and a concept of objectivity recently unpacked by Daston and Galison (2010) as a transformation of values and methods, from truth-to-nature (eighteenth century), to mechanical objectivity (nineteenth century), to trained judgement (twentieth century). Looking to the emergence of counter cartography in the post-SecondWorld-War era for inspiration, we consider Keith Jenkins’ (2008) call for a disruptive, nonlinear history that functions as a ‘refractive discursive experiment sans foundations’. To this end, we consider six case studies toward an alternative history of critical visualization, one that:

29

→ foregrounds alternative epistemologies (through the ancient quipu); → identifies the rhetorical power of a visualization within its data ­assemblage (through the Brookes slave ship diagram); → situates visualizations within a gendered, Eurocentric, ­contextual ­history; → reveals how visualizations enact imaginaries and ­territorial ­imperatives and obliterate local knowledges (as evidenced in the Great Trigonometrical Survey of India); → identifies peoples (and places) previously suppressed by dominant ­narratives; and → transforms conventional understandings of a visualization designer and her role. Chapter 3, ‘Making Data’, explores how data is assembled and how it functions, utilizing imaginaries of ‘raw data’, their extraction and connotations of authenticity, accuracy and precision, reinforced every time we use a data set, all the while concealing the processes behind its creation. We revisit the concept of the data assemblage and ‘dynamic nominalism’ — how data enact categories and impose them on populations as facts, policies and worldviews. We conclude with critical questions for interrogating data, asking how and where the data was made, with what biases, and what is erased or lost when it is produced?

Chapter I

220221_CritViz_Book_Full_revisions.indd 29

Critical Visualization

7/20/22 9:46 AM

30

Chapter 4, ‘Data and the Self ’, considers biopolitics and how the self has been constructed by practices of counting, the creation of norms, pre-indus­trial notions of time discipline and extractivism, and industrial notions of energy and fatigue. We trace the link between measuring bodies and managing bodies, between governance, scientific management, and eugenics in the early twentieth century, and present-day concerns with surveillance and risk profiling. We also consider resistant practices of self-tracking, concluding with a case for satirical and playful practices in the arts, and Henri Lefebvre’s (1977/1991) case for the everyday and everyday life as the site of life affirmation. Chapter 5, ‘Data and the City’, focuses on smart cities as the locus of a dialectic between the rhetoric of the spectacular visionary ‘self-tuning’ city and messy, agonistic, participatory forms of mapping and governance. Both kinds of cities deploy visualization, but with dramatically different intentions: the smart city deploys visualization as a tool of persuasion, evoking its connotations of techno-bureaucratic solutions to old and new urban problems (from planning, to traffic congestion, to extreme weather events); the messy, agonistic, participatory and ‘accidentally smart’ city (Dourish, 2016, p. 37) deploys visualization as a tool for engaging the city’s different publics. We consider precedents of the smart city in twentieth-century imaginaries of the frictionless, machine-age city, and look at how, with the advent of climate change, city residents are thrown together in unexpected alliances to work out new ways of being and strategizing everyday life, adapting visualization tools and tactics in a make-do approach. Chapter 6, ‘Beyond Aesthetics and Representations’, tackles the patriarchal dismissal of beauty as ornamentation, and its reinforcement by modern minimalist aesthetics and a long-reified signal/noise communication model. We explore the possibilities of beauty as a means of engagement, perception as a means of augmenting cognition, and representation as the re-presenting of selected aspects of lived reality to create meaning — to either uphold or challenge dominant orders. Examples of visualizations that re-present to challenge dominant orders include efforts by W.E.B. Du Bois to create new narratives of African Americans in the early twentieth century, efforts by Bureau d’études to challenge totalizing views, and efforts by Iconoclasistas to subvert the stereotypes inherent in the long-­ revered ­Isotype visualizations by Otto and Marie Neurath. To harness its disruptive, revolutionary potential, visualization needs to be considered as an action rather than an object, and the agency of designers needs to be recognized as a force to shape the world they aim to represent.

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 30

7/20/22 9:46 AM

Chapter 7, ‘Beyond Critical Visualization Practice’, concludes the book with a provocation for delinking data visualization from Western paradigms, commencing with a critique of the same critical theory that renders universal a Western understanding of the world. Here, we support ­Glissant’s (1997) case for opacity over transparency, since the former acknowledges the irreducible complexity of things being visualized and refuses to rest in the conceit that transparency (the logic of letting the data ‘speak for themselves’) and simplicity (think of graphic design’s beloved ‘big idea’ approach) should be upheld as paradigms. In place of the pursuit of universalist and rationalist visualizations produced through top-down, synoptic views, we support local, situated, Indigenous perspectives, oral traditions, knowledge of place and vincularidad. In place of the saviour complex and pain-centred research (e.g. ‘I have visualized and mapped your problems, and you need this’), we support visualization produced with and for the people whose interests it represents, following Eve Tuck (2009): desire-­centred research, and visualizations that act.

31

­DIS T R I BU T E D CO GN I T ION A N D H U M A N IS T IC ­A PPROACH E S We have positioned this book as an introduction or provocation, because it targets an area we feel is in need of development, while acknowledging that as researchers embedded in the field of design — in UK and North American universities — we are, ourselves, learning what we have to learn. This includes the rich discourse that challenges the dualism at the heart of the fields of human-computer interaction and visualization. Better known as Cartesian dualism, named after the seventeenth-century French philosopher René Descartes’ famous meditations on first philosophy (1647), which set out reasons to ‘doubt about all things and especially about material things’, the dualist/Western tradition prioritizes theory over action, and mind over body. Descartes established a foundational philosophical method, but stopped short of using it to question his most fundamental assumption, that his language privileged a particular worldview. By imagining that an ‘evil genius’ had conjured up a world that appears real, for the purpose of setting traps for his credulity, Descartes then established proof of an ‘I’ with the affirmation that there must be an ‘I’ who is capable of doing all of the doubting: cogito ergo sum (I think therefore I am). Non-Western traditions, however, do not cleave to the conception of an ‘I’ that precedes action, that is auto­nomous and somehow apart from the world.

Chapter I

220221_CritViz_Book_Full_revisions.indd 31

Critical Visualization

7/20/22 9:46 AM

32

The connection between the Cartesian tradition and visualization is apparent in the way that maps are conceived as artefacts to represent ‘things, concepts, conditions, processes, or events in the human world’ (Harley and Woodward, 1987), and diagrams are conceived as plans for action in the world, or as descriptions of states encountered in the world. Because the implied order of events is linear — we create a diagram, map, or visualization usually before we proceed to act — a priority is a­ ssumed (i.e. theory before practice), along with a non-reciprocal model of cognition: the mental image, or visualization, is formed, and then our experi­ ence or data is ­organized into its structures. Mignolo and Walsh’s impulse to ‘disobey’ this conceit is supported by anthropological and psychological perspectives on navigation. Anthropologist Tim Ingold draws from psychologist James Gibson’s ‘eco­ logical approach’ to visual perception to argue that humans do not construct representations of the world inside their heads and then proceed to act on them. Instead, humans use a tacit and sensorial practice of ‘­know­ing as you go’ (Ingold, 2000). At the time Gibson published his work on perception (1950–1979), psychology assumed that people’s cognitive systems organized the raw material of experience into an internal model that would, in turn, provide a guide for action (Ingold, 2000). Gibson took issue with the implicit separation of mind from the environment: Perception, Gibson argued, is not the achievement of a mind in a body, but of the organism as a whole in its environment, and is t­ antamount to the ­organism’s own exploratory movement throughout the world. If mind is ­anywhere, then, it is not ‘inside the head’ rather than ‘out there’ in the world. To the contrary, it is ­i mmanent in the network of s­ ensory p ­ athways ­that are set up by virtue of the perceiver’s immersion in his or her ­environment. (­Ingold, 2000, p. 2)

If the human mind is reconceived as a network of sensory pathways, rather than a central processing unit — to adopt the computational analogy — the implications for data visualization are quite profound. ­Conventional ap­proaches to data visualization (e.g. Card et al., 1999; Ware, 2004) tend to see the tools of visualization (e.g. diagrams, working out maths ­problems on a notepad) as ‘amplifying’ cognition, with the underlying assumption that cognition takes place ‘in the head’: This approach — often called ‘reductionist’ — largely ignores the environment, assuming that after we have a better understanding of the individual cognizer, we will be able to understand how cognition interacts with the complexity of the environment (Liu et al., 2008, p. 1173).

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 32

7/20/22 9:46 AM

In advocating an alternative ‘humanistic’ approach to visual knowledge production, or graphesis, Drucker (2014) characterizes the deficiencies of the reductionist approach to graphical interface design and the ‘science’ of human-computer interaction (hci):

33

The hci ‘user’ combines two ideological illusions in a single paradoxical identity: the predictability of a mechanized automaton and the myth of ­auto­nomous agency. Humanistic approaches to interface need to ­recuperate the theoretical formulation of subjectivity as a part of the enunciative a­ pparatus, of positions spoken, articulated, created by the structuring and desiring ­machines of representations. The legacy of a half century or more of theoretical discourse is available for this work, ready to be brought back into play. Who is the subject of an interface? How are we produced as subjects of the ­discourses on the screen? And in our embodied and culturally situated relations to screens and displays? (Drucker, 2014, pp. 146–147)

Drucker proposes that we can also think of interface as an ‘ecology’ or a ‘border zone between cultural systems and human subjects’ (Drucker, 2014, p. 148). A humanistic design approach, she argues, is ‘subject-­oriented’ rather than ‘user centred’. Drucker contrasts statistical graphics with ­human­istic graphics, whose forms and lines express uncertainty, and situated data that supports more complex processing models to gain meaning: Interpretation is stochastic and probabilistic, not mechanistic, and its ­uncertainties require the same mathematical and computational models as other complex systems. (Drucker, 2014, p. 131)

Rather than expect typologies and formats of graphic elements to account for the ways in which meaning is produced, Drucker builds a case for an interpretive understanding of how meaning is produced by readers or viewers, citing Erving Goffman’s frame analysis as a particularly relevant model for understanding how, in contemporary digital media, we are constantly shifting frames to understand what types of information are being offered, what tasks are made possible and which behaviours are expected. An alternative approach to hci’s dependence on the autonomous and anonymous ‘user’ as a unit of study was introduced by the anthropologist Edwin Hutchins with distributed cognition. In his 1995 study of US Navy personnel aboard large ships, Hutchins showed that cognition is a distributed system, both internal and external to the cognizer. Recounting the emergency procedures enacted aboard a ship losing steam drum pressure in San Diego harbour, Hutchins observes that ‘many kinds of thinking were required’ to bring the ship to safe harbour: ‘some quite clearly both inside and outside the heads of the participants’ (Hutchins, 1995, pp. 5–6).

Chapter I

220221_CritViz_Book_Full_revisions.indd 33

Critical Visualization

7/20/22 9:46 AM

34

Applied to data visualization, Hutchins’ ‘distributed cognition’ framework proposes that cognition is an emergent property of interaction among ­humans and artefacts, ‘embodied, enculturated, situated in local interactions’ (Liu et al., 2008, p. 1174). Cognition does not arise in an individual alone, but among a group of people interacting with external artefacts. These artefacts often play a significant role in reducing the cognitive load on people, for e­ xample: providing short-term or long-term ­memory aids, timely information, know­ledge and skills, structuring cognitive behaviour without people’s conscious awareness, and even determining decision-­making strategies through ac­curacy maximization and effort minimization. The cognitive behaviour of traders in a stock market, or of medical professionals in an emergency room, then, could not be understood by studying the response of one individual to a visual display, but by observing how cognition is d ­ istributed across the screens, moving and inert bodies, surfaces, procedures and ex­pectations of the situation. As such, the study of visualization might be better served ‘in the wild’ (to use Hutchins’ phrase) — using ethnographic methods rather than isolated lab experiments. Since cognition is an emergent property of interaction, interaction is a ‘major focus’ in the study of cognition (Liu et al., 2008, p. 1179). A related aspect of this process, familiar to designers, is in Gibson’s ­concept of affordances: cognition in humans and animals does not ­require information processing of representations; we directly perceive the ­values and meanings of things in the environment because perception is structured in terms of what things ‘afford’, i.e. what they offer, provide or furnish, for good or ill (Gibson, 1979/2011). A tree ‘affords’ a surface for a primate to climb and seek shelter or food; a steaming cup on the table ‘affords’ a particular way of holding it (e.g. by the handle) to avoid burning one’s hands; a video game controller ‘affords’ an intuitive way of ­moving, jumping, shooting, etc. — based on the angles, sizes and shapes of the ­controls. A conventional data visualization, in turn, affords a particular way of view­ ing, distancing and interacting with the world, which we shall endeavour to unpack in this book. In short, it enacts a way of reducing the world to known and measurable entities for the purposes of organization and, in all senses of the word, exploitation. In the process of visualizing data, a number of considerations are silenced: things that do not fit under the categories adopted for collecting and organizing information.

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 34

7/20/22 9:46 AM

35

IS N ’ T CR I T IQU E F I N ISH E D? Throughout the book, we have leaned rather heavily on the notion of critique and criticality, with reference to two key definitions: Michel Foucault’s account of critique, revealing the assumptions, familiar notions and unexamined ways of thinking behind accepted practices; and Paulo Freire’s account of critical thinking as the revelation of injustice, oppression and misrepresentation — and their causes, so that ‘through transforming action [people] can create a new situation’ (Freire, 1968/2017, p. 21). The link between revealing and transforming is central to both Foucault’s and Freire’s accounts. Freire’s book, Pedagogy of the Oppressed, laid the groundwork for a critical pedagogy directed at societal transformation, and Foucault made his remarks in 1981 after an interviewer suggested that political and social reform is unrelated to critique. Foucault countered that ‘reforms do not come about in empty space’ and that ‘criticism consists in uncovering that thought and trying to change it’ (Foucault, 1981/2002, p. 456). It remains to be addressed that critique itself has come under some critique, not least because it has become associated with a kind of relativism, where one person’s critique is countered with another’s to prove, as it were, that all critiques are equal, that criticality is just a matter of opinion, that one person’s critical position is another’s fake news. Latour’s i­ nfluential (2004) essay ‘Why has Critique Run Out of Steam?’ initiated a l­ ively post-­­critical discussion, with the literary scholar Rita Felski (2015), for ­example, unset­tling the normative, one-size-fits-all dominance of a style of critique with its tone of detachment and nonchalance, and making a case for interpretive approaches to criticism that include aesthetic appreciation. Latour himself suggested that a critic might be ‘not the one who debunks, but the one who assembles’ (2004, p. 246). His case against critique as debunking was illustrated by the tactics of right-wing strategists who seized upon a perceived ‘lack of scientific certainty’ in climate change science in order to cast doubt on an issue for which material evidence was amassing at a frightening pace. More recently, Latour has argued for a ‘down to earth’ politics that acknowledges the simultaneous events of ecological degradation, climate change denial, massive deregulation and the escalation of inequalities (Latour, 2018).

Chapter I

220221_CritViz_Book_Full_revisions.indd 35

Critical Visualization

7/20/22 9:46 AM

36

Building on Latour’s position, we have made a case elsewhere for a ‘gene­ rative criticism’ that assembles evidence, as Latour suggests, and acknowledges the power dynamics involved in the translation of interests into designed forms (Hall, 2019). The foregrounding of power dynamics in this book also takes on board the perennial criticism of Actor-Network ­Theory’s ‘apolitical’ position and mechanistic, actor-centric account of how actors work within networks (e.g. Drucker, 2014, p. 51). We aim to show how power is delegated and then enacted through designed forms like visualizations, and at the same time, how visualizations have the ­potential to expose these dynamics at work. The most compelling case for a generative criticism is provided by the multi­ disciplinary, multi-temporal and multicultural issue of the climate emergency, which can be understood as a designed object, in the sense of the philosopher Timothy Morton’s concept of ‘hyperobject’. Hyperobjects are, argues Morton, ‘things that are massively distributed in time and space relative to humans’ and that, A hyperobject could be the very long-lasting product of direct ­human ­manufacture, such as Styrofoam or plastic bags, or the sum of all the ­whirring machinery of capitalism. (Morton, 2013, pp. 9–10)

Visualizing climate change is a design challenge that throws into sharp contrast both the limits and potential of critique. Once the multi-­temporal, multidisciplinary complexity of climate change is acknowledged, it seems imperative to make this complexity visible and actionable. In 1991, the political and cultural theorist Fredric Jameson called for an aesthetic of cognitive mapping to make sense of lived experience and to coordinate existential data with the ‘world space of multinational capital’ (Jameson, 1991, p. 54). Yet, the maps and visualizations that have attempted to reveal the complex flows of world capital, as we note with the work of Mark ­Lombardi or Bureau d’études, have tended to elicit what Brian Holmes (2006) called a ‘subjective shock’ or, worse still, political paralysis. In the twenty-first century, when the complexity of the flows of capital and ecological degradation are rendered in data, a similar state of awestruck paralysis can ensue. Morton describes an ‘ecological ptsd’ that results from bombarding people with guilt-inducing facts on climate change (Morton, 2018, p. 15).

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 36

7/20/22 9:46 AM

37

Figure 1.11  —  Timothy Cadman and Lust: United Nations Framework Convention on Climate Change. Timothy Cadman/Lust

Chapter I

220221_CritViz_Book_Full_revisions.indd 37

Critical Visualization

7/20/22 9:46 AM

38

Since critical visualization embraces the situated over the synoptic perspective and, in mapping networks, can also enact them, one response to climate change is to initiate change through mapping networks of aligned interests. The researcher Timothy Cadman’s efforts to visualize the networks of organizations involved in the United Nations Framework Convention on Climate Change (unfccc), for example, sought to both illustrate the complexity of the UN climate regime, and better integrate the organizations and how they interact. Along with a follow-up map of the governance regime for eco-system services, the climate regime map is exploratory, aiming to facilitate analysis — by showing how i­ ndividual initiatives relate to each other and to overarching concepts. A more pol­emical approach is to visualize complexity and then target specific, implicated organizations in a ‘shaming’ campaign. This strategy, long adopted by Greenpeace, is illustrated in a heat map of Brazilian timber exports designed by Paul Hamilton, which forms part of a Greenpeace campaign aimed at highlighting companies who do not comply with legislation on imports of timber and timber products. The visualization arrays the international supply chain linked to the Brazilian company Santa Efigênia A ­ gropecuária, which has been investigated for illegal logging, and is sus­pect­ed of harvesting timber from protected Indigenous areas of the Amazon. An outer ring of companies, organized by country, highlights — in red and with connecting lines — those with previous links to contaminated supply chains connected to Santa Efigênia Agropecuária and associated sawmills. The visualization illustrates the complex global trade networks, but is contextualized in a campaign with actionable goals: to force authorities in importing countries to take enforcement action against non-compliant companies. As with the cognitive maps of capitalism, the effect of visualizing complex systems can be to thwart systemic change because of their scale and complexity. Morton suggests that an ontological shift is needed to move present-day eco-reformers out of a mindset that presents ‘nature’ or the ­biosphere as something external to humans. As we discuss in Chapter 2, this mindset is tightly entangled with an anthropocentric position ­inherited from the Enlightenment, which saw nature’s bounty as an infinite storehouse for human use (Markley and Cohen, 2012, p. 55). But as we aim to demonstrate in this book, that ontological shift is not alien to humans — it is at the heart of many Indigenous epistemologies. When knowledge of land, water, their inhabitants and seasonal variations is constructed and shared locally over generations through stories, place names and maps, the notion of an authoritative, synoptic, atemporal view that is detached from territory, starts to seem ambitious and impossibly scaled, abstracted to the point of hubris. Margaret Pearce’s mapping project Coming Home to I­ ndigenous Place Names in Canada (Pearce, 2017) confronts this issue of scale with a pluralistic and participatory mapping of territory and place names, which in Indigenous epistemology describe the ‘shapes and sounds’ of sovereign lands, including sites of danger, beauty,

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 38

7/20/22 9:46 AM

39

Figure 1.12  —  Brazilian Timber Exports. Paul Hamilton/Greenpeace

Chapter I

220221_CritViz_Book_Full_revisions.indd 39

Critical Visualization

7/20/22 9:46 AM

40

signed treaties, and how to navigate ice, eggs and berries. Place names are ancient and recent, notes Pearce, ‘both inside and outside of time’ (2017). In this sense, Coming Home is a map that foregrounds situated knowledge, that which resides at different temporal and geographical scales. As Pearce noted in a panel discussion with practitioners engaged at the intersection of Indigenous mapping and feminist visualization, Coming Home is a mapping project that confronts the colonial and extractivist presumption that seeks to fix and inventorize places, and instead foregrounds Indigenous protocols of ‘reciprocity, responsibility, respect, and ‘relevancy’: Coming Home is about scale: listening to what people are s­ aying about what’s important to them, and the scale at which those v­ alues and issues reside. Paying attention to scale resists this ­ongoing notion that maps are inventories, which is a totally c­ olonial cartographic tradition. (Dávila, 2019, p. 13)

In the exhibition catalogue Maps are Territories: Science is an Atlas, David Turnbull provides a closing example from the complex and vexed ­history of Aboriginal land rights (Turnbull and Watson, 1993). The researcher Kingsley Palmer had been interviewing aboriginal Southern Pitjantjatjara people, living in the Great Victoria Desert, about the cognitive maps they sustained through storytelling and myths. When translated by ­Palmer into a finished map, however, the knowledge was closely protected by rules of access, and only when Palmer attended a land rights meeting b ­ etween the ­Pitjantjatjara people and Parliamentarians from Adelaide, was the map brought out as a negotiating tool, resulting in the successful return of ­dis­puted lands in 1984 (Turnbull and Watson, 1993). The language and tools of the occupying entities had been, arguably, subverted and turned back on the colonizer. ­Palmer’s cartographic techniques served a useful purpose to the Pitjantjatjara people, but also remind us of the tendency in mapping and data visualization to inventorize everything, and turn everything into data. As Margaret Pearce observes, counting is a cultural value (Dávila, 2019, p. 19). We hope that the situated maps and visualizations created by, with and for Indigenous groups will be among those to inspire a new practice of c­ ritical visualization. We also hope to demonstrate in this book that a critical ap­ proach to visualization must acknowledge both the value of the ‘master’s tools’ but also the importance of foregrounding new, alternative and ancient epistemologies and ways of being, including those buried by the dominant modes of being and knowing. █

An Introduction to Critical Visualization

220221_CritViz_Book_Full_revisions.indd 40

7/20/22 9:46 AM

41

Figure 1.13 — Searching for Africa in life. Alfredo Jaar. Courtesy of the artist, New York. (see p. 150)

Chapter I

220221_CritViz_Book_Full_revisions.indd 41

Critical Visualization

7/20/22 9:46 AM

220221_CritViz_Book_Full_revisions.indd 42

7/20/22 9:46 AM

220221_CritViz_Book_Full_revisions.indd 43

7/20/22 9:46 AM

II Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 44

7/20/22 9:46 AM

I

n t h ei r cr i t ica l gu i de to gr a ph ic de sign h is tory, Drucker and McVarish argue that ‘by definition, any seamless linear story is false, concealing the multiple layers of disjunction and ten­ sion at work in a culture’ (2010, p. xxviii). And yet in visualizing history, the linear timeline has become so entrenched that it has dominated how we, in the West at least, conceive of time and the idea of historical progress. This, despite the fact that Joseph Priestley himself, the creator of the visualization that popularized the idea of time flowing from left to right in a linear progressive sequence, admitted that the linear metaphor was a ‘mechanical help’ but not an image of history itself, and had its limits, including its emphasis on overarching patterns and the ‘big story’ (D. Rosenberg and Grafton, 2010, p. 20). A subsequent, twentieth-century loss of faith in big stories (or a ‘breaking up of the grand-narratives’, as the philosopher Jean-Francois Lyotard (2005) famously characterized post-­modernity), including the Modern narrative of the linear march of humanity toward greater technological progress, has amplified the question: How are histories now to be written? If the cost of neglecting certain evidence and smoothing over tensions that do not support a dominant historical premise has been to marginalize or silence other perspectives, what stories now need critical scrutinizing and destabilizing in view of other historical traditions?

45

In histories of data visualization, linear, positivist and Eurocentric narratives still remain largely unchallenged. This may be a reflection of the status of data visualization in the sciences, where images have been regarded with suspicion, as subservient to the rigorous logic of numbers (Galison, 2002). When framed as a branch of statistics, the graphical representation of quantitative data tends to be conceived as an accessory to the main subject. The few data visualization histories that have been written tend to lean towards celebratory accounts of an unsung hero, mimicking the ‘great white men’ approaches of earlier conventions in other disciplines. An influential example was a 1937 publication by the American mathematician and historian Howard G. Funkhouser, which established a Western canon of innovators, concepts and technical progress in data ­visualization that became a standard reference. It is perhaps symptomatic of the high-modern­ist era that Funkhouser’s Historical Development of the Graphical R ­ epresentation of Statistical Data (1937) was published within a year of ­Nikolaus Pevsner’s Pioneers of Modern Design (1936/2005), which similarly canonized a few works of a few white men in architecture and product design.

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 45

7/20/22 9:46 AM

46

Funkhouser’s survey commences with a quotation from Prussian p ­ olymath Alexander von Humboldt, in support of ‘statistical projections which speak to the senses without fatiguing the mind’ and ends with the nomogram — a graphical calculator developed by French engineer Philbert Maurice ­Ocagne (widely used by engineers until the advent of the pocket ­calculator). Along the way, Funkhouser (1937) establishes a history populated with now-­ familiar Western milestones: Ptolemy’s maps of 150 AD; René D ­ escartes’ ­analytic geometry (1637); Edmund Halley’s use of mortality tables (1693) and compass lines on a sea chart (1701); Priestley’s afore­mentioned inven­ tion of the horizontal timeline (1765); William Playfair’s Commercial and Political Atlas (1786) with its pioneering examples of the bar graph, line graph, circle graph and pie chart; A.M. Guerry’s histograms showing moral statistics of France (1833); Etienne-Jules Marey’s systematic survey of graphic methods (1878); Charles Joseph Minard’s cartograms, flow maps, graphic tables and famous carte figurative (1869) of Napoleon’s ill-fated Russian campaign; Léon Lalanne’s introduction of polar diagrams (1843); Luigi ­Perozzo’s stereograms (1879); Henry Gantt’s eponymous chart (1915); and Otto Neurath’s ‘Vienna Method’ (1930s), subsequently known as Isotype (cited in Funkhouser, 1937). If this early history established a European-North American canon, subse­ quent contributions from Michael Friendly (2008) and Edward Tufte (2001) solidified the geographic focus, linear direction and positivist tone, through historical accounts that characterized the steady progression of visualiza­ tion as a series of technical accomplishments, premised on the pursuit of the ‘objective’, accurate and engaging representation of data. We shall return in the next chapter to the problematic assumption that data is neutral, raw and pre-existing ‘out there’ to be harvested and represented as transparently as possible. Here, to prepare the ground for a critical history of data visualiza­ tion, we should first note the critiques of linearity and positivism, often associated with postmodernism and the ‘science wars’ of the 1990s. P O SI T I V IS M A N D OB J EC T I V I T Y The positivist position, as inherited from Auguste Comte, presumes know­ ledge gained through the scientific method to be the only valid knowledge. Critics of this position argue that what we call ‘the scientific method’ has its own cultural history and specificity, and its claims to universality or to universal rules are ill-founded, since all observation is dependent on the

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 46

7/20/22 9:46 AM

conceptual framework of the observer (Achinstein, 2004). An ­important critical design exposition on this subject is Lorraine Daston and Peter Galison’s Objectivity (2010), which situates objectivity in scientific history, tracing through the medium of the scientific atlas the invention of objecti­ vity in three overlapping ‘epistemic virtues’: the truth-to-nature ethos of eighteenth and early-nineteenth-century naturalists and observers; the rise of mechanical objectivity and the receding importance of the human o ­ bserver in the nineteenth century; and the addition of trained judgement in the twentieth century, which scientists increasingly felt was needed in order to make and interpret the mechanical image.

47

The ghostly, fading in/fading out presence of the observer, scientist and designer of visualizations in the process of making images is, in a sense, the sub­ text of a critical history of data visualization. As discussed earlier, the semiotic convention of any diagram, map or statistical chart is that what we are viewing is supposedly universal, viewable from anywhere — ‘the god trick’ (D. Haraway, 1988, p. 581). When a visualization is seeking to persuade the viewer of something — as in, for example, Playfair’s chart on the national debt — the presence of the work’s creator is often apparent in the title and commentary: ‘during this last war, content with getting possession of the money, we have left to future generations the trouble of repaying it’ (Playfair, 1801). If the visualization is mechanically or computer-generated without the visible guiding hand of its maker, as in Arthur Worthington’s 1893 ‘instantaneous photographs’ of mercury splashes (as cited in D ­ aston and Galison, 2010, p. 16), or a data feed whose visual representation is auto­ mated by algorithms, the work’s creator is much harder to spot. The presence or absence of the visualizer in the visualization reflects the degree to which he or she is implicated in the scientific method adopted. As Peter Achinstein (2004) shows, the scientific method has considerably more var­ iety than popular understanding might suggest. Some might object that while methods evolve, the essence of the scientific method is that the gather­ ing of evidence must precede its explanation, and that gathering must be done objectively, in scientific conditions. Yet even this order of events is up for debate: Thomas Kuhn (1962/2012) argued that scientists generally have a theory in mind before designing experiments in order to make empirical observations. More radically, Feyerabend (1975) used the history of science to show that scientific progress does not result from the application of a particular method and to argue that progress is often the result of violating scientific norms.

Chapter II

220221_CritViz_Book_Full_revisions.indd 47

Critical Visualization

7/20/22 9:46 AM

48

Figure 2.01  —  Joseph Priestley’s linear timeline of history (1769). Joseph Priestley

A H IS T ORY OF PRO GR E S S The second characteristic of existing data visualization histories is their linearity, their adherence to a narrative of scientific progress. This is a position ultimately inherited from the anthropocentrism of Enlightenment thinkers like John Locke, whose philosophy invoked a ‘golden age’ when humans began the exploitation of nature’s unending bounty through the development of technology (Markley and Cohen, 2012, p. 55). In contemporary accounts, data, like nature, is often portrayed as a resource that can be ‘mined’: visualizations are successful when they communicate ‘what the data has to say’ with efficiency, that is, they give the viewer ‘the greatest number of ideas in the shortest time with the least ink in the smal­ lest space’ (Tufte, 2001, p. 51). The efficient presentation of data then becomes a yardstick with which to distinguish the milestones worthy of canonization.

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 48

7/20/22 9:46 AM

In Friendly’s ‘brief history’ in the Handbook of Data Visuali­z ation, significant ‘milestones’ of data visualization meet criteria under one of the following classifications:

49

→  contributions to the development and use of graphic forms →  graphic content →  technology and enablement →  theory and practice →  theory and data on perception of visual displays →  implementation and dissemination (Friendly, 2008, p. 8). Conventional histories of data visualization, like any history of a technology, are teleological in the sense that they are unidirectional and assume an endpoint. This position is sometimes implicit in visualizations themselves, for example in Priestley’s Chart of Biography (1765) and New Chart of History (1769) (in D. Rosenberg and Grafton, 2010, pp. 18 and 115), which continued to be printed into the nineteenth century. So ingrained is the metaphor of the timeline today that it is quite difficult to imagine time flowing in other directions, despite the contention that any ­interesting narrative moves both backwards and forwards in time and digresses into subplots and contrasts. The philosopher Henri Bergson, writing at the time of Marey’s chronophotographic gun, considered the metaphor of the timeline to be a ‘deceiving idol’, and advocated the concept of ‘duration’ (time as people subjectively experience it) in opposition to the dominance of measured (clock) time (D. Rosenberg and Grafton, 2010, p. 23). Nevertheless, Priestley was convinced by the evidence of his own selection of figures and events in his two charts to argue that the mass of entries on the right of the chart represented the ‘acceleration’ of the arts and sciences in his time (T. E. Rosenberg, 2007, p. 68). The teleological flavour of the history of data visualization is evident in its stone-skipping account of innovations, which typically occur with more frequency as we near the present. History with an emphasis on the survival of the fittest techniques is to be found, for example, in Arthur Robinson’s (1982) Early Thematic Mapping in the History of Cartography, which tracks the flurry of carto-statistical activity in the nineteenth century with the availability of statistics, and the introduction of colour printing and litho­ graphy. Michael Friendly similarly describes a ‘golden age’ of statistical graphics and thematic cartography in the late 1800s, only surpassed by innovations of the more recent post-war ‘rebirth’. Certainly, evidence is bountiful of a mid-nineteenth-century European and North American interest in visually presenting numbers and distribution of populations, occupations, religions, the movement of goods and people, and ‘moral statistics’ — along with a series of technical innovations for representing the data: bar charts, pie charts, histograms, line graphs, time series plots,

Chapter II

220221_CritViz_Book_Full_revisions.indd 49

Critical Visualization

7/20/22 9:46 AM

50

contour plots and scatter plots (Friendly, 2008, p. 4). Dot maps, for example, were introduced to represent totals with the size or number of dots, choropleth maps rendered given districts in flat tones, patterns or colours to comparable numbers (density, ratio, departure from average) and isoline maps represented common quantities with connecting lines (Robinson, 1982, p. 111). But this well-established progressivist narrative has yet to be dismantled or challenged by non-Eurocentric or non-teleological histories. To disrupt a seemingly innate sense of historical teleology requires that we look more closely at the assumption, adapted from Darwin’s survival of the fit­test, that the techniques and technologies that are successful are — by definition — the best.2 To critique this teleological position, the philosopher Michel Serres uses the example of the modern-day car, which is not the end result of a series of technical improvements on previous technical improvements, but a ‘disparate aggregate of scientific and technical solutions dating from different periods’ — a Neolithic wheel, 200-year-old thermodynamics and a global positioning system from the 1990s. Serres describes linear history as ‘narcissistic’: We conceive of time as an irreversible line, whether interrupted or ­continuous, of acquisitions and inventions. We go from generalizations to d ­ iscoveries, leaving behind us a trail of errors finally corrected — like a cloud of ink from a squid … I cannot help thinking that this idea is the equivalent of those ancient ­diagrams we laugh at today, which place the Earth at the ­center of ­everything, or our ­galaxy at the middle of the universe, to satisfy our ­narcissism. (Serres and ­Latour, 1995/2008, p. 48)

CR I T ICA L CA RT O GR A PH Y: A ‘ DE F I N I NG MOM E N T ’ Diagrams which place the present day at the end of a line of h ­ uman techno­ logical progress, evoke a sense of self-importance similar to that of world maps that place Europe at the centre, following the Mercator p ­ rojection of 1569. In this respect, the critical cartography initiatives of the mid-­twentieth century present an invaluable example of how a visualization can ­critique a hegemonic one, and can disrupt a dominant, unexamined way of ­thinking. The Mercator projection, like any attempt to render a sphere on a two-­dimensional surface, is guilty of distortion: by removing the poles and lowering the equator, it gives more prominence to the northern hemisphere. By placing Europe at the centre, the projection literally placed at the centre of the world the seats of the empires that pro2 ‘... as though today’s world was the precise goal toward which all decisions made since the beginning of history, were consciously directed’ (Ferguson 1974b, 19). For critique of this assumption, and a case for non-linear accounts of technological developments, see for example Pinch, T and Bijker, W. (1987) ‘The Social Construction of Facts and Artifacts: Or How the Sociology of Science and Technology Might Benefit Each Other’. In The Social Construction of Technological Systems. Edited by Wiebe Bijker, Trevor Pinch and Thomas Hughes. mit Press.

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 50

7/20/22 9:46 AM

ceeded to colonize the rest. Critical cartography’s ‘defining moment’ (to use geographer ­Jeremy C ­ rampton’s phrase) followed the publication in 1974 of Arno Peters’ equal-area projection, which, unlike the ­Mercator ­projection, enabled accurate comparison of continent sizes (J. ­Crampton, 1994). Although the Gall-Peters Projection (as it became known) would seem to belong in a history of thematic mapping, ­Robinson did not include it in his account (­Robinson’s own projection had been endorsed by the National ­Geographic Society as the closest of all flat maps to render the round surface of the earth ‘without distortion’ (J. ­Crampton, 1994)). In fact, Robinson launched a sustained attack on the Peters projection, taking issue with its claims to universality, while conspicu­ously ignoring its declared ideological aim of challenging the bias in the Mercator projection (J. Crampton, 1994). ­Robinson’s critique of the Peters Projection curiously included an appeal to taste, arguing that the map was ‘reminiscent of wet ragged long winter underwear, hung out to dry on the arctic circle’ (1982, p. 104). As discussed in ­Chapter 6, aesthetics can guide our reading of a visualization, and ­references to taste, standards of beauty or elegance can influence our perception of value.

51

Figure 2.02   —   Cartography’s ‘defining moment’: The world map using the Peters Projection. Daniel R. Strebe, licensed with CC BY-SA 3.0.

Prior to the Peters controversy, the discipline of cartography, like that of data visualization, was ‘overwhelmingly governed by internal technical in­ terests, and did not engage external ideological interests … it was in fact a “cartography of progress”’ (J. Crampton, 1994, p. 17). Post-Peters controversy, key journals such as Cartographica, books and conferences paid increasing attention to the relationship between cartography and society. Among the influential work was that of the geographer J.B. Harley, who began ­apply­ing post-structuralist techniques to ‘re-read’ the text of a map, deploying Foucauld­ian and Derridean approaches to deconstructing the map and the ‘progressive paradigms’ of cartographic history (J. B. Harley, 1988).

Chapter II

220221_CritViz_Book_Full_revisions.indd 51

Critical Visualization

7/20/22 9:46 AM

52

The postmodern critique of teleological or linear history also took a lin­ guistic turn in the 1990s that presented opportunities for re-reading it. Linearity was ‘an illusion derived from another illusion, namely, that language also unfolds in a linear fashion,’ according to historiographer Keith Jenkins’ account of Jean Baudrillard’s The Illusion of the End (­Jenkins, 2008, p. 163). Jenkins countered the idea of a linear history with history as a ‘reflexive, aestheticizing, figuring, grammatically promiscuous, refractive discursive experiment sans foundations’ (2008, p. 166). Jenkins characterizes as ‘absurd’ the idea that there is a historicized past existing independently of our various present-day concerns (2008, p. 55). A F E W E X A M PL E S: NO T A CA NON In light of developments in critical cartography, and in the spirit of Jenkins’ call for a ‘discursive experiment’, a series of visualization case studies from history will be considered here. The function of the case studies is not to propose an alternative canon but to begin to disrupt the linear, teleological and Eurocentric bias of current data visualization histories, much as Arno Peters’ projection in 1974 challenged the internal technical flavour of carto­ graphic discourse. h a p t ic v isua li z at ion: t h e qu i pu (1200 –1532) A quipu is an information technology that was used extensively across South America as far back as 4500 years ago, through to the Inca civilization (approximately 1200–1475 AD), and following the Spanish invasion of 1532 (Graham, 2014, p. 112). Typically made of a thick cord to which knotted strings of different colours were attached, the quipu was used to record data and transmit messages across the large distances of the Inca empire, transported by runners as part of a communications system that crossed the Andes, from Colombia to Argentina, via Ecuador, Peru, ­Bolivia and Chile. Informatics scholar Paul Beynon-Davies (2009) argues that the quipu, and associated infrastructure of purpose-built roads and rope-bridges, enabled the Incas to record and administer activities across a vast empire, which was essential to its success. Messages encoded into the system of knots and strings typically contained ‘details of resources such as items required or available in store houses, taxes owed or collected, ­census data, the output of mines or the composition

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 52

7/20/22 9:46 AM

of particular ­workforces’ (Beynon-Davies, 2009, p. 171). The handheld quipu allowed ­messages to be clear, compact, portable and encrypted: a quipucamayoc (keeper of the ­quipus) was responsible for encoding and decoding the messages by ‘running his fingers rapidly over the knots, rather like a Braille reader’ (Beynon-Davies, 2009, p. 171). Quipus were typically rolled up and transported by chains of runners over long distances — up to 250 kilometres in a day — a p ­ ractical response to the challenge of a hostile and mountainous terrain, and an absence of wheeled transport.

53

Figure 2.03  —   Sketch of a quipu and quipucamayoc from El primer nueva corónica y buen gobierno, a chronicle of Inca history by the Indigenous Inca historian Felipe Guaman Poma de Ayala (ca. 1535–1616). Felipe Guaman Poma de Ayala

Sceptics might object that a portable system of knots and strings does not constitute an addition to a history of data visualization, but here, Jenkins’ case for a ‘grammatically promiscuous … discursive experiment’ supports our case for including an information storage and transmission system that supported institutional structures such as agriculture and defence. Archaeologists and scholars have studied the system of knots and coloured strings of the quipu to calculate that as many as 1536 distinct signs could be communicated, more than double the number of Egyptian and Mayan hiero­glyphic symbols (Beynon-Davies, 2009, p. 185). A renewed twenty-­ first-century interest in embodied and ubiquitous networked computing (i.e. the internet of things) has cast the quipu and associated infrastructure in a useful light: as we move away from a print and screen-based information culture, data visualization stands to gain from pre-visual methods of data representation and transfer.

Chapter II

220221_CritViz_Book_Full_revisions.indd 53

Critical Visualization

7/20/22 9:46 AM

54

The quipu also presents a case for a non-Western visualization system that supports recent discourse on embodied data and embodied k ­ nowing. David Turnbull’s ‘book of exhibits’, Maps are Territories: Science is an Atlas (1993) — produced in collaboration with Helen Watson and the Yolngu community at Yirrkala, Australia — provides a wealth of examples that challenge the Western conception of a map, including: a native American Chippewa drawing that was presented to the US Congress in 1849 as a land claim; a dhula belonging to the Yolngu community of the L ­ aynhapuy region of Arnhem Land in Australia, otherwise referred to as an Aboriginal ‘bark painting’; and a stick chart from the Marshall Islands, used not for navigation, but to teach mariners how to navigate the currents and swells, represented by criss-crossing sticks. Navigation was a physical skill, involving the expert mariner lying in the bottom of a boat feeling the ocean swells and watching the sky (Christianson, 2012, p. 19). Elsewhere, maps are identified in the carved wood coastal maps used by Inuit to navigate the waters around Greenland, presenting a model of data visualization that is haptic and tactile (Southworth and Southworth, 1982). Such devices support a method of navigation that is quite distant from the cartographer plotting a territory or a route on a map. The anthropologist Tim Ingold has written on non-Western conceptions of space that redress a Cartesian bias that contains consciousness in the brain. Ingold confronts the idea that navigation is performed with a mental ‘cognitive map’ that we reconcile and update as we move through the world, arguing that this idea is premised on the often-unstated assumption that the map affords ‘a representation of things in space that is independent of any particular point of view’. Instead, he argues for a conception of the world that ‘continually comes into being around the inhabitant’, citing Turnbull and Alfred Gell’s respective accounts of the Micronesian mariner, who crosses hundreds of miles of open sea between tiny islands with the aid of a list of stars whose rising and setting points indicate direction. The Micronesian seafarer is an example of how we ‘know as we go, from place to place’ (Ingold, 2000, p. 239). This is navigation not with charts and compasses, but an act of feeling one’s way toward a destination by adjusting movements in relation to ocean currents, wind and stars. While there is no pretense to a synoptic, view from nowhere, the knowledge and method are still transferable: the principles upon which the Micronesian mariner’s map is constructed are ­securely embedded within the percepts and practices of traditional seafaring, and therefore … it requires a knowledge of this cultural context to be able to ‘read’ and understand the map. (Ingold, 2000, p. 225)

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 54

7/20/22 9:46 AM

‘Knowing as we go’ is aligned with praxis, the recursive integration of thought, reflection and action — as opposed to the dominant idea that ­theory comes before practice, that humans plan (or design) then implement. As we discussed in Chapter 1, theorists of decoloniality have situated the dominant prioritization of theory over practice in Eurocentric imperialist traditions. If ‘theory is doing and doing is thinking’ (Mignolo and Walsh, 2018, p. 7), then a material and situated form of organizing and ­updating data while moving through the world — such as the quipu — presents a radi­ cally different way of knowing from the dominant mode of data visuali­za­ tion. Hence, the quipu can be thought of as a project of critical visualization.

55

‘pl a n a n d sect ions of a sl av e sh i p’ (1789) The diagram showing the capacity of the Brookes slave ship is arguably one of the most impactful visualizations in the history of the genre, creating an iconographic language for the abolitionist movement and subsequent projects citing the African American diaspora. A copperplate engraving, initially derived from hand-drawn cargo diagrams provided by slave ship captains, the ‘Plan and Sections of a Slave Ship’ effectively enlisted supporters of the British abolitionist cause in the late 1700s. The identity of its creator is not certain, but the banker, scientist and keen artist William Elford is generally credited with the drawing and engraving (Finley, 2018, p. 33). The visualization appeared in at least six variations as part of pamphlets and posters published by the Plymouth and London Committees of the Society for Effecting the Abolition of the Slave Trade (seast), with an estimated print run of 8,700 (Description of a Slave Ship, 1789). Showing 482 men, women and children crammed into a ship designed for the purpose of trafficking humans from Africa to the Americas, the image depicted with cold clarity the effects of a supposedly ‘more humane’ slave trade, following the 1788 Dolben Act, which regulated the capacity of slave ships based on their tonnage. Abolitionist Thomas Clarkson, who campaigned with the i­ mage, considered the regulation to be ‘perfect barbarism’. The diagram, said Clarkson, depicted the ‘rigorous economy’ of the slaver, in which ‘no place capable of hiding a single person, from one end of the vessel to the other, is left unoccupied’ (Rediker, 2008, p. 339). This capacity was achieved with the construction of platforms between decks, reducing the height between each tier to two feet and seven inches (78 cm) — not enough room for captives to sit up — and the precise ­measurements of the allocated space for men, women, boys and girls (180 cm × 40 cm for men).

Chapter II

220221_CritViz_Book_Full_revisions.indd 55

Critical Visualization

7/20/22 9:46 AM

56

The broadside version distributed by the London Committee included a text detailing the measurements, and describing the conditions — including ‘blood and mucus’ on the floors, widespread sickness, slaves packed so tightly that a ship’s doctor had to step between their legs, how the men were ‘fastened together’ with handcuffs and irons on their legs, and a resulting average mortality rate of twenty per cent, or one in five (­Plasa, 2014, p. 297).

Figure 2.04 — Plan and sections of the Brookes slave ship (1788), which played a role in the British abolition of the slave trade. Plymouth Chapter of the Society for Effecting the Abolition of the Slave Trade

Part of the effectiveness of the slave ship visualization is in its appropriation of a technical drawing style used by the British Royal Navy for representing naval architecture: precise measurements were obtained in a Government-commissioned survey by Royal Navy Captain Parrey and coolly detached labelling lent the image a ‘graphic authority’ (M. Wood, 2000, p. 25).

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 56

7/20/22 9:46 AM

Marcus Wood has also noted that the propaganda circulated by abolitionists needed to tread a delicate line by providing detailed eyewitness accounts that made the horror of the slave trade vivid and stimulated guilt and culpability among an educated English audience, without frightening off such an audience through fear or disgust (2000, p. 23). Contemporary pro-slavery imagery often included idealized representations of the African woman as a black Venus, shown in a Thomas Stothard engraving, for example, with a ‘cloud of white cherubs hovering about her [as she is] magically wafted across the Atlantic’ (M. Wood, 2000, p. 21). Comparable eighteenth-century abolitionist imagery depicted the separation of slave families in Africa, or the landing of slaves in the Americas after the Atlantic voyage. The abolitionists’ seal showed an African man in chains, kneeling beneath the question, ‘Am I not a man and a brother?’ Wood argues that the diagram of the slave ship functions as a parody of Noah’s ark, which was the subject of a theological debate: Was it big enough to hold all the animals in the world? A wooden model of the Brookes diagram, commissioned by the early leader of the French Revolution, the Count of Mirabeau, included small wooden figures, painted black to represent the slaves.

57

The slave ship as anti-ark embodies a series of balanced antitheses to the ark: greed is substituted for love, exploitation for trust, repetition for ­variation and sameness for difference. The cargoes of the two boats are perfectly a­ ntithetical. The ark emphasized the unique value of each created thing, and each life form aboard is simply unique; the slave ship emphasizes the homogeneity of the slave cargo, and each life form abroad is, in its legal s­ tatus, the same. (M. Wood, 2000, p. 32)

Formal abolition of the slave trade was declared in 1807, and as part of a calculated seast communications strategy, the ‘Plan and Sections of a Slave Ship’ played a significant role in the process. The abolitionists purposely focused on ending the slave trade, rather than abolishing slavery itself, because they recognized the economic hurdles facing the larger goal, which included negotiation with colonial legislatures and plantation owners. The diagram was widely disseminated in France and North America in the 1790s, and along with the abolitionists’ seal (‘Am I not a man and a brother?’), maintained currency in the US and North America through to the American Civil War. The seal was reproduced as a ceramic medallion by Josiah Wedgwood and became so fashionable in the late 1780s, it was being worn as a broach or hairpin by society ladies, and was appearing on the lids of snuff boxes.

Chapter II

220221_CritViz_Book_Full_revisions.indd 57

Critical Visualization

7/20/22 9:46 AM

58

Britain had profited significantly from the trade, transporting 2.5 million of the 6 million enslaved Africans in the eighteenth century, but managed to obscure this history and establish a reputation as the leading anti-­ slavery campaigner — a myth still perpetuated today (Boddy-Evans, 2018). A second visualization produced by Clarkson supported this revisionist history — a map of the history of abolition, visualized as a series of rivers and streams, each bearing the name of a (heroic) Western abolitionist. The absence of slaves in this visualization, and the presence of a genealogy of abolitionists that includes the poets Milton and Pope, reveals an almost instantly rewritten history of slavery, supported in Clarkson’s two-volume book, H ­ istory of the Rise, Progress and Accomplishment of the Abolition of the ­African Slave-Trade by the British Parliament, which he published immediately after the abolition bill was passed in parliament (M. Wood, 2000, p. 1). The place of the Brookes diagram in a critical visualization history is significant for a number of reasons: as part of an abolitionist campaign, the diagram played an important role in translating the case against human trafficking to an educated and influential public. It did this with the ‘­superb semiotic shock tactic’ (M. Wood, 2000, p. 27) of combining a technical drawing with the depiction of dehumanized people. Its potency as a visualization that gives voice to enslaved Africans is evidenced in its long and still-continuing afterlife, as a signifier of Black survival on Bob Marley’s 1979 Survival album cover, and as an image of a ‘living, micro-­ cultural, micro-political system in motion’ as Paul Gilroy writes, that has been revisited in a number of artistic works in recent years, including paintings, theatre, performance, installation, printmaking, photography and film (Finley, 2018, p. 5). In the field of surveillance studies, the Brookes diagram has become a key figure in the argument for centralizing race and Blackness in the history and practice of surveillance, prefiguring and surpassing the position of Jeremy Bentham’s 1791 diagram of the panopticon. Bentham’s design for a prison in which all inmates can be observed by a single guard in a central inspection tower was taken by Foucault, in Discipline and Punish, as a meta­ phor of the archetypal disciplinary gaze of modernity, since it replaced the spectacle of public executions with a more subtle form of control, in which inmates regulate their own behaviour based on the knowledge that they might be being watched. But the floating prison of the Brookes diagram presents a mechanism of power and a diagram of political technology that simultaneously disciplines a people through a ‘policy of terror’, and consti­ tutes the category of Blackness as a ‘saleable commodity in the Western Hemisphere’ (Browne, 2015, p. 42). Noting that the tiny black figures in the diagram are not, in fact, replicas, but have different gestures, variously crossed arms, and even seem to stare back at the ‘gaze from nowhere’, the African American scholar Simone Browne argues that the diagram also depicts a form of resistance to imperial power (2015, p. 50).

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 58

7/20/22 9:46 AM

Unlike Bentham’s diagram, the ‘Plan and Sections of a Slave Ship’ was not conceived as a design to be replicated, but as a design so abhorrent it demanded its own elimination. Its significance and longevity are due to its potency in surfacing the ongoing legacy of slavery, and its careful visuali­ zation of data that others sought to bury.

59

pol a r a r e a di agr a m (1859) Florence Nightingale’s visualization of the causes of death in the ­Crimean War is often included in canonical histories of the genre for its formal in­ novation. It is included here as a project that, like the Brookes diagram, contributed to a critique and reform of a dominant practice, and — as such — illustrates how visualizations function by assembling and translating inter­ ests into persuasive form. Nightingale, who came from an affluent and well-­ connected English family, eschewed the conventions of marriage to become a driven campaigner for the use of statistical evidence to counter dominant patriarchal narratives. Her evidence specifically challenged the unspoken mythos around the supposed glory of the battlefield death. In this sense, Nightingale’s visualization can be considered a feminist project, as an es­sential ingredient for the ‘innumerate men in power’ (Magnello, 2012), despite being firmly entrenched in Eurocentric values.

Figure 2.05 — Polar area diagram of causes of death in the Crimean war (1858). Florence Nightingale

The visualization was developed to support Nightingale’s 1859 report on battlefront conditions, to persuade the British government that an improvement of sanitary measures and administration on the battlefront would have a dramatic impact on the war. She distributed 1000 copies of the report among those in power, incorporating the graphs to support the evidence her team had gathered on the front, to show how the major­ ity of British army deaths in the Crimean campaign were the result of

Chapter II

220221_CritViz_Book_Full_revisions.indd 59

Critical Visualization

7/20/22 9:46 AM

60

­preventable diseases. Nightingale did not invent the polar area graph, which was first introduced in 1829 to show crime statistics, but she imaginatively adapted it to make her points more persuasive. The graph is derived by plotting values from the centre of a circle — the farther the point is from the centre, the greater its value — with the circle broken into twelve wedges representing the months of a year. By extending the radius beyond the circumference of the circle, the graph becomes more dramatic and persuasive, aiming toward visual, rather than verbal, persuasion. Contrary to the popular image of Nightingale as the gentle ‘Lady with the Lamp’, a critical history reveals a tough campaigner with an administrative and managerial flair, who recognized that achieving change required corralling both political and statistical support. After demonstrating her abilities as manager of the Harley Street Hospital, Nightingale volunteered her services for the war, arguing that women should be allowed on the front (as was the case with the French military, but not until that point with the British). Having gained the backing of home secretary and former prime minister Sir Robert Peel, who set up a fund in the Crimea in 1854, Nightingale travelled to the front in Turkey in the winter of that year. She and the team of twenty-eight nurses she was supervising collected statistical data on the health of the soldiers in the first seven months of the war, encountering widespread cases of frostbite, dysentery, extreme weakness and emaciation, cholera and typhus. In addition, they learned on ­several occasions that bureaucratic processes prevented cargo ships carrying supplies (e.g. warm clothes, cauliflower, coffee, rice, flour and coal) from being unloaded or delivered to the troops who needed them most. The statistical data collected by Nightingale’s team was taken back to London and analysed with the statistician William Farr.3 This was a project driven by an emerging Christian idea in the nineteenth century that the ‘desperate wickedness’ of the world (Nightingale (1859), quoted in McDonald, 2006, p. 40) could be challenged and overcome with statistics. In one sense, it reflected a shift in thinking in the n ­ ineteenth century, as industrialization and the mass mobilization of peoples called for a better understanding of economies of scale. As Ian Hacking puts it, ‘­society became statistical’ (Hacking, 1990, p. 1). In another sense, the ­statistical approach evident in the project supported the ethno-­centric position of an Empire premised on its supposed religious and ­racial ­superiority, and along with the Christian entanglement of hygiene, whiteness and godliness, appealed to the dominant belief system of the time (Stake-Doucet, 2020).  

3 Farr is credited with the systematic quantification of public health problems in Victorian England, which led to the implementation of the Public Health Acts (Magnello, 2012).

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 60

7/20/22 9:46 AM

Rather than view Nightingale’s polar area graphs as a formal innovation, then, they can be viewed as the visual product of a process analogous to the ‘sociology of translation’ in Actor-Network Theory (Callon, 1986): framing a problem, identifying relevant interests (i.e. actors), enrolling and mobilizing them (see Figure 6.10, p. 197). The four ‘moments’ in Michel Callon’s sociology of translation begin with problematization, where the researcher makes him or herself indispensable in a network of relationships. Already, it can be seen that Nightingale was a skilled manager and a well-connected negotiator, accruing support from politicians, military chiefs and newspapers, thus making herself indispensable to the network. The problem, as defined by Nightingale, was ‘the British Army’s failure, through bureaucratic inertia, to protect the soldiers’ health or to assist in their recovery’ (Nightingale, 1859). Initial moves were administrative: she implemented a reform so that battlefront medical records were standardized to support comparison. By supplying the managerial means, medical know-how and statistical evidence to support this position, Nightingale established her approach as an ‘obligatory passage point’ for addressing the problem. The second of Callon’s moments, interessement, describes how an actor in a network attempts to impose and stabilize the identity of the other actors it has defined through the problematization. In this case, Nightingale had implicated a network of interests with her identification of the problem of battlefield care, health and sanitation: the British Army, its administration and care of the wounded, Government and — upon Nightingale’s return to the UK — Queen Victoria and Prince Albert, himself a supporter of statistics and the sciences. The third moment, enrolment, describes how the researcher’s success depends on the relevant actors forming alliances. Following her Royal visit, Nightingale procured support for a Royal Commission on the health of the army, and also sought statistical advice from Edwin Chadwick on publishing her findings. The final moment, mobilization, describes the process by which it is ensured that ‘actors speak for collectivities’. It is notable that Nightingale ­responded to questions from the Royal Commission in writing and — to return to the ­visualization itself — supported her account with the iconic graphs: the polar area graph was designed, she wrote, to ‘affect thro the Eyes what we fail to convey to the public through their word-proof ears’ (Nightingale (1859), as cited in Magnello, 2012, p. 31). Callon’s four moments thus help demonstrate that Nightingale’s approach successfully objectified her cause, such that the network of actors enrolled in the problematization — graphs, the Royal Commission report, statistical medical evidence, a­ n administrative overview — became a powerful collective. In short, Nightingale’s ­project is an unusually clear account of how visualizations function in the ­assembly of evidence, and even challenge the dominant, patriarchal order. A ­critical visualization history, however, should also take into account the historical, economic and cultural nature of the assemblage supporting the visualization, and Nightingale’s unquestioning subscription to the racial ­imperatives of colonialism (Stake-Doucet, 2020) situates this visualization

Chapter II

220221_CritViz_Book_Full_revisions.indd 61

61

Critical Visualization

7/20/22 9:46 AM

62

as an object both foundational to the nursing profession and to a statistical project that privileged colonial power. This privileging is evident in our next case study. gr e at t r ig onom et r ica l su rv ey of i n di a (1802–1875) The imposition of an architectural masterplan on populations in order to impart a ruling hierarchical order and ‘civic values’ is a practice that extends back through the history of city planning and mapping, but with the British colonization of India, a particular approach to information architecture was deployed to support the subjugation and categorization of ­local populations. The Great Trigonometrical Survey (or gts, 1800– 1871), undertaken by the British East India Company in the nineteenth century, imposed a conceptual image of British rule in the visualization and implementation of ‘imperial space’ (Edney, 1993, p. 61).

Figure 2.06 — T he Great Trigonometrical Survey of India (1802–1875). Survey of India

Prior to the establishment of British rule in India after the Indian Rebellion of 1857, British interests in the region were represented by the East India Company, which led a surreptitious colonization of the continent based on trade, brutality and the establishment of bridgeheads in the coastal cities, from which point it administered an increasing number of territories of India and wielded a private army larger than that of the British Army (Dalrymple, 2015; Kulke and Rothermund, 2016). The gts was a means of mapping the vast and diverse territory of India with a uniform, geodetic

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 62

7/20/22 9:46 AM

triangulation survey, whereby a surveyor establishes a baseline between two known points and determines the location of a third point by measuring the angles to it from the baseline points. By overlaying on the territory a mesh of interlocking large-scale ‘primary’ triangles, the surveyors could establish a skeletal framework from which to create smaller-scale triangulations. This, in theory, imposed a coherent system for mapping the vast, culturally and politically diverse area as a single entity: subordinates were directed to survey resources (forests, mines, soil), commerce, populations, religions, languages, local histories, animals and medicines (Edney, 1993, p. 62). In reality, the expense and complexity of triangulation meant that the gts was considerably more ad hoc than the British rhetoric suggested; its completion and supposed coherence became a matter of principle, as a symbolic representation of the supposedly rational and advanced state of European science, compared with the perceived ‘­indefi­nite ... ­imprecise or completely false’ knowledge and conceptions of space held by Indians (Edney, 1993, p. 63).

63

As a visualization, the gts presents an image of India that appears control­ lable, reflecting the proliferation of systematic territorial and statistical surveys of the era, which, as Edney notes, homogenized or obliterated local understandings: The application of the same techniques and scales of enquiry to each and every district meant that the resultant maps and statistical tables all contained the same sorts of information and were constructed and tabulated in the same manner. They therefore obscured or denied local nuances and particular ­circumstances. (Edney, 1999, p. 36)

A parallel project was the population census of India, which the British commenced in 1865. While appearing to be a passive instrument of data gathering, the census created, by its practical logic and form, a new sense of category identity, becoming an instrument of knowledge and a form of social control. As Arjun Appadurai writes, ‘numbers g­ radually ­became more importantly part of the illusion of bureaucratic control and a key to a colonial imaginary in which countable abstractions, of people and ­resources at every imaginable level and for every conceivable p ­ urpose, created the sense of a controllable indigenous reality’ (1996a, p. 117). Over time, Appadurai notes, the system was co-opted and turned against the colonial power. Counting Indian bodies, as Appadurai puts it, unyoked social groups from the complex structures and agrarian practices in which they’d been embedded, and ultimately was turned into the idea of self-rule,

Chapter II

220221_CritViz_Book_Full_revisions.indd 63

Critical Visualization

7/20/22 9:46 AM

64

as nationalism became a mass movement. If you have been counted, you count. As discussed in Chapter 4, counting and categorizing people came to depend on a notion of the ‘average’ or ‘normal’ inherited from the desire of Quetelet, among others, to establish universal statistical laws. With the colonization of distant lands, entire populations were perceived as not average, i.e. savage and unruly, raising the stakes for the Empire’s need to govern by imposing a system. data v isua li z at ion at t h e pa r is e x posi t ion, w.e.b. du bois (1900) The sixty hand-drawn charts, graphs and maps produced under the direction of African American sociologist and activist W.E.B. Du Bois for the ‘American Negro exhibit’ in the Palace of Social Economy at the 1900 Exposition Universelle in Paris are unusually prescient, in form and content. Formally, the visualizations suggest a creative re-imagining of the bar charts, pie charts and timelines that had begun to proliferate in Europe and the US after their introduction in the late eighteenth century. Among the works were: colour-coded maps showing the distribution of African ­A merican populations and land ownership; tables with images and percentage bars showing the breakdown of income and expenditure among 150 Black families in Atlanta, Georgia; and inventive spiral graphics showing property ownership and rural/urban population distributions of ­African Americans in Georgia and the US as a whole. Created by a team of students working under Du Bois’s direction, the visualizations suggest an awareness of precedents by Playfair, Minard and Nightingale, combined with a reinvention of the forms that creatively anticipated the work of Russian Constructivists nearly two decades later. The context of this unusual and — until recently — neglected event in the ­history of critical visualization begs elaboration. Du Bois, a professor of sociology at Atlanta University and the first African American to earn a doctorate, was tasked with putting together an exhibit for the Paris Expo by the African American lawyer and educator Thomas Calloway, who had successfully belatedly raised $15,000 from the United States Congress to fund an exhibit of the ‘educational and industrial progress of the negro race in the US’. With just four months to produce the exhibit, Du Bois and ­Calloway commissioned and curated a set of maps, charts, 500 photographs, reports, and a collection of magazines, books and ephemera to combat a number of prevalent myths about Black people in America.

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 64

7/20/22 9:46 AM

Such myths included the misappropriation of Darwinism to apply it to race, to argue that Black inferiority would inevitably lead to that population’s extinction (Battle-Baptiste and Rusert, 2018, p. 33). Data sourced from the census and Atlanta University sociology lab in fact showed that the Black population of the usa was increasing, and illiteracy was ‘less than that of Russia, and only equal to that of Hungary’ (Levering Lewis and Willis, 2010, p. 29). Calloway characterized the prevalent prejudice: ‘Everyone who knows about public opinion will tell you that the Europeans think of us a mass of rapists, ready to attack every white woman exposed, and a drug in civilized society’ (Levering Lewis and Willis, 2010, p. 16). Rather than giving lectures on Black Americans that Europeans wouldn’t attend, or publishing in newspapers or books that Europeans wouldn’t buy, argued Calloway, there would be no more effective response to such slander than to showcase the best in African-American life at the Exposition Universelle, where thousands of Europeans would be in attendance. He also enlisted the support of Daniel A.P. Murray, the assistant librarian of congress (and only the second Black American to work for the library)4 to assemble a bibliography of 1400 titles by African-American ­authors, including Du Bois’s The Suppression of the African Slave Trade to the United States.

65

The exhibit set out to subvert conventional perceptions of African Ameri­ cans by presenting a ‘mirror image’ (Levering Lewis and Willis, 2010) of the white world with photographs of Black Americans, mostly taken by Thomas Askew, Harry Shepherd, and Frances Benjamin Johnston, of middle-class ministers, teachers, lawyers, skilled labourers and students. But with the visualizations, Du Bois also sought to ‘blur the colour line’, by showing how perceived differences between races were due to factors ‘­other than inherent racial deficiencies’ (Bini, 2014, p. 58). With the exhibition organized in four parts: history, contemporary social and economic conditions, educational progress and production of literature, Du Bois f­ocused the visualizations on the state of Georgia, which had the largest population of men, women and children of colour (860,000) and showed population sizes, proportions of slaves and freemen prior to 1870, proportion of African American children in school, distribution of Black ­Georgians in the state, net worth of households, acres of land owned and their value, and taxes paid on property. Such progress was set against the backdrop of socially imposed hardship, represented in the exhibit by a map of the slave trade, timelines including the emergence of the white

4 https://www.loc.gov/resource/rbc0001. 2013gen06011/?st=gallery (accessed 1 June 2022).

Chapter II

220221_CritViz_Book_Full_revisions.indd 65

Critical Visualization

7/20/22 9:46 AM

66

­supremacist Ku Klux Klan and lynchings, and an excerpt from the discriminatory Black Code of Georgia (a symbol of disenfranchisement of African Americans after the Reconstruction period). In effect, Du Bois subtly used the data visualizations at the Paris Exposition to situate historically the progress of Black Americans, framing the advancement of education, professions and populations in the context of centuries of slavery and oppression. Evaluated through the conventional lens of a statistical history of data visualization, the unconventional forms of the hand-rendered gouache and ink charts and diagrams present a challenge: the choices of shapes seem whimsical and some of the colour schemes counter-intuitive. An assessed valuation of taxable property, for example (Figure 2.07), attempts to visualize an increase in the value of property ‘owned by Georgia Negroes’ from $5.4m in 1875 to $13.4m in 1900. The chart takes the form of a circle, with the lowest dollar amount assigned to a prominent centre circle painted black and the highest amounts assigned to the thinner surrounding circles. Visual logic suggests the width of each circle represents the proportion of gain on the p ­ revious year, but the functional effect of narrower width rings is to make it impos­ sible to clearly render the assigned dollar amounts. Du Bois’s team chose to place the valuation figures on arrows protruding from the outer rings that pierce the centre circle. The visual effect suggests an invasion or encroachment of the centre circle. Similar design decisions hamper the readability of one of the spiral diagrams on city and rural populations,5 which draws attention to a red centre representing 734,952 ‘Negroes ­living in the country and villages’, but visually deprioritizes the total 123,000 or so who have moved to cities, by representing them as a single-­ coloured line, moving away from the centre at sharp angles.

5 See Library of Congress Prints and Photographs division, https://www.loc.gov/resource/ ppmsca.33873/

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 66

7/20/22 9:46 AM

67

Figure 2.07 — Du Bois’s distinctive chart of taxable property owned by African Americans in Georgia (1900). W. E. B Du Bois

In the broader context of a World’s Fair, however, where colonial exploits were traditionally presented as European triumphs, and Indigenous people were othered or exoticized as being ‘of nature’ with no history of their own, the visualizations suggested an authoritative and rigorous perspective on the living, education and economic conditions of African Americans. The exhibit also presented a unique perspective of the African-American experience, one that must have struck a chord with jurors, who awarded the installation several prizes. In contrast with contemporaneous visualizations used to classify populations, economic trade, poverty, and disease for the purpose of government and management, the hand-rendered forms commissioned by Du Bois created an identity for its subjects. In Chapter 6, we explore how these ‘data paintings’ re-presented Black Americans.

Chapter II

220221_CritViz_Book_Full_revisions.indd 67

Critical Visualization

7/20/22 9:46 AM

68

In the larger context of the fair, the visualizations hinted at Du Bois’s famous thesis of ‘double consciousness’, to be published three years later, which described the experience of being Black in America, whereby social isolation and psychic fragmentation accompanied the effect of ‘always looking at one’s self through the eyes of others’. In the case of white mainstream America, those eyes projected a disparaging stereotype at odds with an individual’s sense of self-worth; hence there were, as Du Bois put it, ‘two unreconciled strivings; two warring ideas in one dark body, whose dogged strength alone keeps it from being torn asunder’ (Du Bois, 1897). At the Paris Exposition, the struggle against racism was writ large in the setting: a small show of strength emerging out of a sea of white supremacy. com m u n i t y-bu i ldi ng w i t h isot y pe: ot to a n d m a r i e n eu r at h While Otto and Marie Neurath are commonly credited for developing a formal language of information graphics whose legacy persists in news media today, the larger pedagogical and socially responsive goals of their work tend to be neglected. It is this aspect that is of interest here, ­particularly because the most common critique levelled at the Neuraths’ ‘universal language’ of Isotype (International System of Typographic Picture Education) is that it perpetuated cultural and gender stereotypes (Bessa, 2008; Cohen, 2014). An account of the Neuraths’ philosophy, ­interests and method can help situate the claims to universalism, and better c­ onnect their work to an approach to activist urban planning and ­interventions that is highly relevant today. Otto Neurath, a sociologist, initiated Isotype in the context of post-First-­ World-War Austria, a country so devastated by war its government could not provide the support needed for the nation to recover. Inflation, food and fuel scarcity were rife, and many families set up huts and enclosures in city parks, meadows and forests to grow vegetables and avoid ­starvation. These ‘gypsy settlers’ were recognized by many observers — i­ ncluding Neurath  — a­ s the seeds of a ‘self-help’ approach to urban reform and c­ itizen planning. Neurath led a research centre to promote grassroots a­ ctivism, social economy and self-help housing. The cultivation of ‘­ordered disorder’, as Neurath put it, was necessary, since the actions of citizens driven to fending for themselves were both positive and negative — the emergence of informal economies presented a realistic alternative to market-led capital­ ism, but needed some coordination and planning. The housing crisis in Vienna, for example, had driven the emergence of squatter settlements, but also t­ hreatened anarchy. The shortage of coal had driven families to cut down trees to fuel fires, but this had dangerous consequences for the sustain­ability of the forests.

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 68

7/20/22 9:46 AM

The social transformation project underlying Neurath’s embrace of picto­ graphic communication is explicated by Nader Vossoughian (2011):

69

Neurath was concerned that the industrialized city of the early twentieth century relied on dehumanizing workers, exploiting natural resources, and objectifying human relationships. In its place, he sought to galvanize a global ‘polis’ premised on the synthesis of community and modernity. (­Vossoughian, ­2 011, p. 12)

Isotype, initially known as the Vienna Method, came out of a belief that the problems of the modern city were social and called for reprogramming and reorganization along the lines of a rationalized in-kind economy that used barter and other ‘natural’ forms of exchange. After the failure of his initial efforts to support semi-autonomous settlement co-operatives, Neurath turned to museum education and exhibition design as a means to educate and enlist the working classes in the process of transforming the social and economic hierarchies of the modern city. A pre-­Isotype example was a series of charts illustrating density levels in cities across Europe and the usa, and the impact of the pace of housing production on the living conditions in a city. A particular target was the ‘rental ­palaces’ in Vienna’s Ring, which featured ornate and oversized entranceways, but cramped living quarters. ‘The study of plans is a beginning,’ wrote Neurath. ‘They teach us better than any facade whether an apartment is light and airy’ (Vossoughian, 2011, p. 67). Neurath conceived of visual displays that could empower and educate the public on the social and economic processes behind the challenges of the modern city, from housing to production, emigration, unemployment, health and mortality. As founder of the Museum of Society and Economy in Vienna in 1925, Neurath established a ‘Department of Transformation’, whose role was to analyse statistical data, distil and organize them in an accessible way. The mathematician Marie Reidemeister (who later married Neurath and continued Isotype after his death) took on this job, and the artist Gerd Arntz was tasked with creating the graphic forms. While the linocut pictorial signs of Arntz are well-known today for their influence on information graphics, the role of the transformer developed by Marie Neurath is a less-noted but significant aspect of the design process, since it involves the framing, selecting and organizing of data — analogous to James Corner’s account of the ‘doubly operative’ processes of mapping: ‘digging, finding and exposing on the one hand, and relating, connecting and structuring on the other’ (Corner, 1999, p. 225).

Chapter II

220221_CritViz_Book_Full_revisions.indd 69

Critical Visualization

7/20/22 9:46 AM

70

Today, the role of the transformer is often assigned to a data journalist, but as Robin Kinross (2009) observes, the Neuraths’ small team established the vital and interdisciplinary aspects of the role through a systematic research process. An early visual diagram of police interventions in ­Vienna from February 1925, for example, depicts seven rows of unruly-­ looking individuals divided by a central axis, those on the left representing the number apprehended under the influence of alcohol, those on the right not under the influence. This is a visually engaging advance on a typical bar chart display of data. The orientation around a central axis allows comparisons to be made between the two categories, and the organization of data into rows, with each row representing a day of the week, reveals a pattern more meaningful than, say, organization in orders of m ­ agnitude: a quick glance reveals that the longest row — the largest number of police interventions — is for Saturday, the day after payday. Yet, Otto Neurath was later critical of how the visual ‘livening up’ of the data with illustrations of disorderly people had the effect of inviting speculation on the details of individual cases rather than the central theme of the chart, the statistical relations. He wrote: ‘If one knows nothing except “arrests under the influence of alcohol” one must make just one type for it and repeat that, as often as the statistical information demands’ (Neurath and Kinross, 2009, p. 79). This methodological emphasis on iterative ­development in the ordering, filtering and arranging of data — the task of the ‘transformer’ — had a significant impact on the development of data visualization and graphic communication in the subsequent decades. Kinross makes a convincing case for the influence of Neurath’s approach on the London Underground diagram (first produced in 1931 by the ­electrical draughtsman Harry Beck) which forms the basis of most mass transit diagrams today: by showing connective relations rather than geographical relations, Beck’s diagram — like Isotype — eliminated a large amount of detail, to prioritize what was ‘really necessary to know’ (Neurath and Kinross, 2009, p. 108). Like Isotype, the London Underground diagram employs a strict set of graphic conventions, including the use of colour, to efficiently convey information and connections. An Isotype chart contemporary to the Underground diagram showing the size of armies and casualties of the Great War (Figure 2.08) similarly experiments with forms to facilitate the communicative function of the diagram. Each figure of a soldier represents a million soldiers, with wounded figures colour-coded red and deaths colour-coded black. Symbols are organized

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 70

7/20/22 9:46 AM

into two groups of parallel rows — one group for each side in the War — which makes it possible to easily compare the scale of the armies and the proportion of casualties and mortalities. Most significantly, quantities are not represented by the size of the graphic figure, as was the case in Michael Mulhall’s earlier Dictionary of Statistics (1883, cited in Neurath and Kinross, 2009, p. 98), but by repetition of the standard unit. As with the Underground diagram, the unconventional configuration of units is determined by the meaning and communicative function (Neurath and Kinross, 2009).

71

Figure 2.08 — Casualties of the Great War. Gesellschafts- und Wirtschaftsmuseum in Wien, ‘The Great War 1914-18’, 1933. Otto and Marie Neurath Isotype Collection, University of Reading.

Other significant successors to Neurath’s systematic approach cited by Kinross are the ‘new typography’ expositions of Jan Tschichold (1928, cited in Neurath and Kinross, 2009, p. 115) — who worked briefly with Neurath at the Museum of Society and Economy — and Jock Kinneir and Margaret Calvert’s system for the UK road signs developed in the 1960s. Tschichold advocated a strict standardized system with given formats, such as letterheadings on A4 with a second colour, red, to underscore information. On Kinneir and Calvert’s highway signs, the system is revealed in a single diagram: A unit is defined as the width of a character stem, which is then used to establish the proportions of the directional lines and the ­positions of names. This has the effect of prioritizing legibility, rather than, say, the size of sign, since the whole ensemble is ‘as big as it needs to be: it has no set dimensions’ (Neurath and Kinross, 2009, p. 119).

Chapter II

220221_CritViz_Book_Full_revisions.indd 71

Critical Visualization

7/20/22 9:46 AM

72

As with many Modernist-era projects that aspired to create a standardized universal language, Isotype has, with age, come to appear tainted by the specific cultural context of its creation, ‘forged in the optimism of the first Austrian Republic’ (Burke, 2009). In the twenty-first century, the ambition of disseminating a universally understood intuitive pictorial language retains an air of colonialism. The Neuraths’ optimism remains ‘at the core’ of current pictographic signage, for example, in the widespread assumption that a picture of a figure with and without a skirt can be understood universally to denote gender-specific public facilities. As design writer ­Pedro Bessa observes, the gender bias in the assumption that the masculine (non-skirted) figure also functions as a universal signifier is evidence of Maria Isabel Barreno’s expression ‘false neuter’: ‘it embraces a ­wider ­discursive practice, in a patriarchal culture that excludes or d ­ evalues ­women depicting them as deviant in relation to the masculine norm.’ (­Bessa, 2008). Arntz’s figures, for all their elegant charm and ­efficiency, also ­perpetuate cultural stereotypes: most notoriously, the ­Isotype cate­gorization of the ‘five groups of men’ in Neurath’s ­International ­Picture Language (1936) depicts a South East Asian figure in a conical hat, an ­Indian figure in a turban, an African figure with tight curls and a South American figure in a sombrero. Of the five groups, only the ­Western fi ­ gure appears to wear a shirt. The well-intentioned but ultimately derogatory effect of the Isotype racial stereotypes is re-deployed by German artist Harun Farocki in the video In-Formation, which debuted in 2005 at ‘Project Migration’ in Cologne, an exhibition concerning the movement and displacement of peoples, and how Germany and Europe have been ‘co-shaped’ by migration since the 1950s. Farocki’s 16-minute video, commissioned by Kölnischer Kunstverein, depicts a history of migration in former West Germany through a slide show of pictographs, drawings and maps taken from bureaucratic archives. A segment on the guest worker programme and Turkish immigration in the 1960s and 1970s includes Isotype-style images of men wearing black moustaches and a fez hat, drawing out the caricature inherent in the ‘universal’ symbols. In discussing the work, the critic Brianne Cohen notes that Isotype perpetuated a ‘culturally essentializing mode of representation, albeit with positive intentions’. Cohen adds: With this artwork Farocki displays what essentially amounts to the ­official ­affirmation and repetition of offensive stereotypes in German spheres of ­government, education and reportage … these images are more akin to ­caricatures than ­cartoons … they accentuate and exaggerate ­physiognomic ­features to comment on the supposedly fundamental nature of a thing or ­person. (Cohen, 2014)

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 72

7/20/22 9:46 AM

It would be a misrepresentation of the Isotype project, however, to assign it to the backwaters of a history of well-intentioned, but ultimately flawed Modernist utopic idealism, complicit with colonialism. Otto and Marie Neurath’s ambition to educate, agitate and transform societal relations through pictorial communication retains a relevancy in a present-day era, when bureaucratic procedures and ‘big data’ rhetoric often exist to disempower the disadvantaged people in a city, town or settlement, particularly migrant workers.

73

Figure 2.09 — Sequence of stills from Harun Farocki’s video installation In-Formation. Harun Farocki GbR

A more recent example of a design project that takes on the transformer role and Isotype approach, is the ‘Making Policy Public’ programme of the Center for Urban Pedagogy (cup), founded in New York in 1997. For this programme, cup commissioned a series of posters from graphic designers working with community groups to tackle a specific social justice issue or educational need. Using simple, colourful graphics and distilled texts contained in speech bubbles, the posters set out to make the complexities of a specific law affecting a group accessible and actionable (Goetz, 2010).

Chapter II

220221_CritViz_Book_Full_revisions.indd 73

Critical Visualization

7/20/22 9:46 AM

74

In ‘Vendor Power’, designed by Candy Chang for cup’s Street Vendor project, Isotype-influenced images help depict street vendors’ rights. In the wake of then-Mayor Giuliani’s draconian crackdown on street vending in New York through a maze of red tape, the posters explained — in ­several languages — to vendors how they could navigate the law and sell their wares with a legitimate permit. A key part of the ‘Making Policy Public’ project was that designers worked in partnership with community and advocate organizations. Another poster helped inform union members about labour negotiations, and another cup project took a graphic novel format to explain the process of the juvenile justice system to young offenders. As with Neurath’s transformer role, the challenge of navigating the complexities of an issue, organizing and presenting it, was seen as a transdisciplinary project. Marie Neurath summarized the role as follows: It is the responsibility of the ‘transformer’ to understand the data, to get all necessary information from the expert, to decide what is worth transmitting to the public, how to make it understandable, how to link it with general knowledge or with information already given in other charts. In this sense, the transformer is the trustee of the public. (Neurath and Kinross, 2009, p. 72)

While a philanthropic tone pervades the project, it is this dynamic process that is arguably the lasting legacy of Isotype, more than an essentialized pictographic language that betrays its culturally specific roots. Much as Isotype sought to empower the people of Vienna to rebuild their city through grassroots, collective action, so too have graphic communication projects like ‘Making Policy Public’ sought to make complexity navigable. The legacy of Isotype is also to be seen in more recent progressive efforts to amplify the voice of non-‘universal’ populations, as in the Iconoclasistas mappings discussed in Chapter 6. Here, the racial caricatures in Isotype are re-appropriated and subverted to represent new populations. CONCLUSION The six case studies presented here each aimed to touch on different ­dimensions of critical visualization: → Epistemological: the first example, the ancient quipu, challenges the dominant theory of knowledge on which visualization rests (­epistemology). → Rhetorical: the second example, the Brookes slave ship diagram, ­exemplifies the role of a visualization as part of a larger rhetorical assemblage in challenging dominant views and legislation.

Disruptive Histories

220221_CritViz_Book_Full_revisions.indd 74

7/20/22 9:46 AM

→ Situated: the third case study, the Polar area diagram of Crimean war casualties, is positioned as a situated, gendered critique of ­conventional views of war and medicine. → Enacting: the fourth example, the Great Trigonometrical Survey of ­India, d ­ emonstrates how a visualization constructs an image of a ­territory that appears controllable, and, in doing so, enacts the ­territorial ­imperatives of that image, homogenizing or obliterating local ­understandings by becoming the common point of reference ­(­or ­obligatory passage point). → Identifying: the fifth example, the visualizations commissioned by African ­A merican scholar W.E.B. Du Bois for the 1900 Paris Expo, ­illustrates how data visualization became a means of identifying a people, re­purposing dominant forms to counter white supremacist narratives of the era, suggesting instead an authoritative and rigorous ­perspective on the state and being of African Americans. → Transformational: the final, sixth case study aims to ­recontextualize a familiar trope of information graphics — the Isotype — and ­foreground its participatory approach to social and political transformation, reposi­tioning the designer-researcher as a ‘transformer’.

75

Having demonstrated the cultural specificity of conventional histories of data visualization, and having begun to sketch an account of what a critical history might look like, we shall proceed in the next chapter with a critical exposition of forms of data, their implications and possibilities. █

Chapter II

220221_CritViz_Book_Full_revisions.indd 75

Critical Visualization

7/20/22 9:46 AM

Anna Ridler, Myriad (Tulips) 2018

76

Focus Figure 1.1 — A nna Ridler’s Myriad (Tulips), 2018. Anna Ridler

Focus Figure 1.2 — Detail from Anna Ridler’s Myriad (Tulips). Anna Ridler

Rethinking the Representation of Data 220221_CritViz_Book_Full_revisions.indd 76

7/20/22 9:46 AM

Focus

77

Anna Ridler’s Myriad (Tulips), a data set of 10,000 labelled photo­ graphs of unique tulips, subverts the dismissal of aesthetics as decoration (see Chapter 6) by opening up critical questions around data and representation. As a work of art, the images are immersive and affective, conveying the labour involved in gathering data through the immense scale of the project and its handmade qualities. As one of three projects probing the limits of data visualization, Myriad embodies a critique of data-power and popular conceptions of representation, while alluding to the canonical Iris flower data set introduced in 1936 by the British statistician Ronald Fisher. Ridler began the project from a poli­tical position on data ownership. ‘I wanted to push back against the ownership of data by large corporations, and that, as an artist working with data you are sometimes having to default to the data you’re given.’ She says: Normally computer vision data sets, such as ­ImageNet, are created by Mechanical Turkers, who label and match ­thou­sands, sometimes millions of images and inputs, some of which are scraped from the internet. Because the data sets are so large (Imagenet is 14 million images) —   it is almost impossible as an artist to look through and work out the cultural, social or political biases of the various different people who put them together, or the different competing pressures in making them. The alternative, for me, is to build the definitions from the ground up and train the machine learning model with your own viewpoint of the world. (A. Ridler, personal communication, September 23, 2020) The project echoes Fisher’s Iris data set, consisting of fifty samples from three species of the iris flower, each measured according to length and width of sepals and petals. Fisher’s iconic data set is used in machine learning as a method of finding a linear combination of features that charac­ terize classes of objects, but his legacy is also entangled with his tireless promotion of eugenics, which imposes a hierarchy on genetic traits,

and supports beliefs and practices aimed to improve the genetic quality of a population (Evans, 2020). The entanglement points to the situated cultural history behind machine learn­ ing, and flowers gain a more sinister connotation (see Chapter 4 for more on the categorization of populations). Inherent in the commitment to building a data set ‘from the ground up’ is Ridler’s  interest in exposing the human element 

Anna Ridler 220221_CritViz_Book_Full_revisions.indd 77

7/20/22 9:46 AM

78

Focus

Focus Figure 1.3 — Detail from Anna Ridler’s Myriad (Tulips). Anna Ridler

in machine learning that is hidden by algorithmic processes (A. Ridler, per­ sonal communication, September 23, 2020). Each of the tulip photographs —   taken over a period of three months in Utrecht during tulip season  —   is accompanied by a hand-written label and fixed on a painted black wall with magnets. The appearance of a machine-made project is thus unsettled on ­closer inspection, which reveals the uniqueness of each photograph, and slight errors and imperfections in the alignment of the grid and handwriting, and draws attention to the manual labour and human decisions involved in data selection and generation. ‘By creating my own data set I am pushing back and offering alternatives to the usual way of creating them (this was not a collection of flowers that I took for the sake of it, it was very much made to be a tool),’ says Ridler. ‘Also data sets

for machine learning are very difficult to see at scale — they exist really only as arrays of data — or as downloads in folders and files — so by presenting in this way we are able to see the scale of what is necessary to make these types of algorithms run’ (A. Ridler, personal communication, September 23, 2020). The 10,000 photographs, while exhibited as a standalone work, also form the handmade training data set for the algorithm that Ridler used to develop two accompanying works, Mosaic Virus (2018–19)—two video installations of tulips generated from the data set, and Bloemenveiling (2019)—a blockchain auction of the artificial-intelligence-generated tulips. As with a still life painting, which can show flowers from all seasons in bloom together at the same time, in the same vase, the generated tulips are botanically impossible, made up

Rethinking the Representation of Data 220221_CritViz_Book_Full_revisions.indd 78

7/20/22 9:47 AM

Focus

79

of combinations of Ridler’s Myriad image data set (Ridler, 2018). These two works develop a critical reflection on representation and economics in the age of machine learning. The seventeenth-century tulipmania speculation bubble was generated by the scarcity of tulip stripes, which are caused by the Mosaic virus. In Mosaic Virus, commissioned by the Dutch arts organization IMPAKT, a generative adversarial network (GAN) is trained to create videos of tulips blooming, linked to the value of the decentralized digital currency Bitcoin. ‘The virus weakens the bulb so eventually, once a tulip becomes striped, it will stop reproducing’ explains Ridler, ‘so Bitcoin scarcity is associated with the stripes’ (A. Ridler, personal com­ munication, September 6, 2019). The resulting installations comprise grids of pink and white tulips that bloom as the currency value rises and falls, and freezes when there is a ‘mode collapse’ — that is, when the GANs miss modes of input data. In Bloemenveiling, short video clips of impossible tulips are sold at auction via smart contracts on the cryptocurrency platform Ethereum. Like the striped tulips of the 1630s, the most valuable tulips in Bloemenveiling are artificially generated, drawing attention to the environmental cost of human systems for creating economic value (Ridler and Pfau, 2019). ●

Anna Ridler 220221_CritViz_Book_Full_revisions.indd 79

7/20/22 9:47 AM

220221_CritViz_Book_Full_revisions.indd 80

7/20/22 9:47 AM

220221_CritViz_Book_Full_revisions.indd 81

7/20/22 9:47 AM

III  Making Data

220221_CritViz_Book_Full_revisions.indd 82

7/20/22 9:47 AM

D

ata is a t er m t h at conj u r es i de a s of a r esou rce to be exploited or a material to be manipulated. Financial journalists state that data is the new oil (The Economist, 2017) and as such, can be extracted and commoditized. The hubris around data is propelled by technological advances in storage and networked devices, as well as the increase in online social life and the corporate strategies to track all transactions. With this enthusiasm for the potential of data to increase profits, there is also an enthusiasm to use data to inform decisions in business, government, research, etc. Visualization, along with other strategies, is used to aid in the analysis, or to represent the analysis, of important data.

83

As designers working in visualization, we often take data for ­granted. The visualization of data creates powerful representations of measurements and categories which themselves were created much earlier in the process and often in a context different from that of the designer. While this may mean that how the data is created is outside of the control of the designer, we are still responsible for using this data and reiterating and ­amplifying its status as a factual representation of reality. For this ­reason, as critical designers of visualizations, we need to pay attention to the conditions of production of these important components of our projects. Criti­ cality, in Foucault’s (1981/2002) account (see Chapter 1), involves paying attention to ‘what type of assumptions, of familiar notions, of ­established, unexamined ways of thinking the accepted practices are based’. As such, ­criticality can reveal injustice, oppression, misrepresentation and can foreground H ­ ara­way’s concept of situated knowledge. From this base, this chapter will attempt to address how to think critically about data as it is employed in visualization. By using data and representing it visually in our projects, we confirm its authenticity unless we explicitly undermine its validity, which is rarely done. We tend to affirm rather than negate the data when we visualize it. This is especially the case when the visualization is clean, minimalist and orderly. By erasing the traces of the parsing, cleaning, filtering and aggregation performed on data we effectively make the viewer forget that the data was produced. Or, as Lev Manovich (2002) reminds us, that data is generated. Data visualizations also gain power through how they juxtapose different data sets to draw out a relationship between different phenomena. For

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 83

7/20/22 9:47 AM

84

Figure 3.01 — Situated knowledge. Marc Ngui

Making Data

220221_CritViz_Book_Full_revisions.indd 84

7/20/22 9:47 AM

instance, your commute time from home to work over the last five years may have some interest, but when joined with the number of commuters on your route over the same period of time, a more interesting and compelling visualization is produced that directs our attention to the relationship between the two variables (i.e. trip duration and date). When you add government spending on public transit, property prices, zoning laws for business and residential usage you get a much richer visualization implying the relationship between all of these dimensions. Simultaneously, as we add more data sets to the visualization, we run the risk of de-emphasizing the specifics of how each data set was produced. Instead, our attention is drawn to the connections, and when we do this, we also tend to leave the site of each data set’s generation far away.

85

Every time we decide to use a particular data set in a project, we vouch for its accuracy and precision — we help reproduce its authority. One might ask why is it problematic to reproduce this authority if the data is not false, but instead accurate and precise? Part of the problem lies in the unintentional affirmation we as designers may perform. We affirm the validity of the data and prioritize this data and the reading of this data over other data. Despite our best efforts, what we do is never truly neutral, nor purely objective. One of the barriers to thinking about data as having a politics is the trope of data being a raw ingredient. Visualization researchers who assert that data must speak for itself (Tufte, 2001) reproduce this trope of data being raw, in addition to expounding the need for a visualization designer to be as neutral and objective as possible in their treatment of the material. This attitude can be traced back to the notion of scientific research which lets nature speaks for itself. As Daston and Galison (1992) point out, this was the moral attitude of positivism in the late eighteenth century, which demanded that scientists reduce their presence and use mechanical forms of observation and measurement that would eliminate the all-too-human errors of subjectivity or imprecision. The promise of raw data is that it is pure, unedited, and has a direct link to the phenomena being analysed. Gitelman and Jackson (2013) note that it has an understandable appeal, in that it appears to come before the fact and acts as a starting point. The reality is that data is ‘always already cooked’ (Bowker cited in Gitelman and Jackson, 2013) — in other words,

Chapter III

220221_CritViz_Book_Full_revisions.indd 85

Critical Visualization

7/20/22 9:47 AM

86

data is generated through human-directed processes and is thus subject to being framed in a particular time and place with a particular focus on what is important. If we look at the etymology of data and fact, we may see how our current usage entails a particular understanding. The word data comes from the Latin dare ‘to give’, and we can contrast this with fact from the Latin facere ‘to make’. While data are seen to be given to us by the world, facts are constructed out of this raw material. Data by definition are ‘that which is given prior to argument’ (Gitelman and Jackson, 2013, p. 7). We can ­therefore conceptualize visualizations as attempts to make facts through the use of data. Visualizing thus prioritizes the fact and pushes the data to ­the background. The process of visualizing has been modelled as an iterative process (B. Fry, 2008), starting at the acquisition of data, going through a process of analysis, representation and refinement, and ending in a visual and interactive representation. The initial step involving acquisition of data refers primarily to accessing available data, rather than the generation of data. In ­practical terms, a designer would only need to consider this to begin a project. As this is a model for a technical design practice, the generation of data is not interrogated and is treated solely as a resource to be accessed. We argue that the production of data is an important aspect of critical practice precisely because of the lack of attention normally accorded to it by designers. What is made invisible, in this case, the production of data, is often one of the most important aspects for a critical thinker to focus on. What becomes standardized, routine and efficient has become naturalized, and therefore unquestioned. But it is there, in the unquestioned areas, that so many ideologically, politically, culturally framed decisions have been made. As we stated earlier in the discussion regarding assemblages in Chapter 1, those decisions are made within a power/knowledge discourse that amasses power through the establishment of expertise, circulation of information, and the description of phenomena through standardized measures and categories. Perhaps it is easiest to imagine this dynamic when we think of the p ­ rocess of measuring and categorizing human populations. Categorization of human beings is very complex and fraught with issues of translating

Making Data

220221_CritViz_Book_Full_revisions.indd 86

7/20/22 9:47 AM

a ­person’s identity or lived experience into a measurable data point. Yet, this same dynamic can be seen within the natural sciences — especially ones that may have controversial implications. The production of scientific data is often challenged on the basis of methodology — that is, on the procedures and protocols used to assure the validity of basic research.

87

Take, for instance, the measurement of temperature, which in the ­present era is a very straightforward procedure involving inexpensive, simple and widely available sensors that can provide a quantity in degrees Celsius, degrees Fahrenheit, or kelvin units. If we measure the local temperature over the span of a year, we get a fairly reliable record of how warm it has been by using a stable technology (e.g. thermometer). But if we were to go back beyond one hundred years of temperature data, we would encounter growing inconsistencies of units, unreliable record-keeping and miscali­ brated instruments. If we were to go back over a thousand years, the col­ lection of historical data becomes even more difficult. In these cases, instead of modern thermometers, proxy indicators such as tree rings, ice core samples, etc. would be used to record temperature. A visualization showing temperature ranging from present-day to a millennia ago would therefore have to aggregate from several sources and emphasize the conti­ nuity of measurement over a long period. Since the focus is temperature, all temperature data are treated as valid sources. The so-called ‘hockey stick graph’ is such a visualization that ­represents the variation in temperatures over a thousand years. Mann et al. (1999) ­reconstructed global temperatures beyond the era where i­ nstruments ­recorded temperatures and discovered that the last few decades ­leading up to the year 2000 had been the warmest in millennia. A main ­innovation of Mann’s research and visualization was the bridging of proxy temperature data with instrument temperature data.

Figure 3.02 — Hockey stick graph (2014). Klaus Bittermann, licensed with CC BY-SA 4.0

Chapter III

220221_CritViz_Book_Full_revisions.indd 87

Critical Visualization

7/20/22 9:47 AM

88

Essentially, two categories of data were brought together in one visualiza­ tion. Their graph was assiduously designed to show the different data sources and the margin of error for each data point used. Nonetheless, it became ‘a central icon in the climate change debate’ (Mann, 2014, p. 22) and as such, a singular image that symbolized one thing: the sharp increase in temperatures over the last one hundred years. Representing data in visual form is a powerful rhetorical device p ­ recisely because it can collapse a complex array of data into a single strong and persuasive image. Visualizations, especially those that give a strong shape to the data, tend to have the effect of erasing the work that led to the making of the visualization. A similar phenomenon has been noted by John Pickles (2004) in the creation of maps, where the pencil marks, mistakes and h ­ uman judgements are erased in order to present a simplified and ­objective image. To non-expert viewers, the map or visualization can be extremely convincing because it looks scientific, orderly and objective. In the presentation of observations made through microscopes, we see a comparable process of refinement. Michael Lynch (1991) notes how blurry images that are difficult to distinguish for the untrained eye are progressively simplified in order to display the most salient features. The simplified and annotated images are often the ones that get circulated the most, presumably due to their directness, clarity and contextualizing information. Designing visualizations thus requires an acknowledgement of this powerful feature of data images and a responsibility to ensure the data is valid. The ‘hockey stick’ graph, and the analysis it represented, was perceived as a great threat to corporate and political interests which defended the continued use of fossil fuels and denied the human impact on the environment and climate. Many attempts were made to sow seeds of doubt into the analysis of the temperature data and reduce the power of the graph. Climate change sceptics, for instance, split the proxy data and instrument data in order to reiterate that an overall downwards trend in temperature was apparent (Mann, 2014). It is difficult to imagine that this response would have been as severe had the data not been so conclusive and the shape so simple.6 The graph that Mann’s team produced was based on decades of work done by paleoclimatologists, attempting to produce reliable records of

6 T  o learn more about the hockey stick graph controversy, read Mann’s The Hockey Stick Graph and the Climate Change Wars: Dispatches from the Front Lines.

Making Data

220221_CritViz_Book_Full_revisions.indd 88

7/20/22 9:47 AM

t­ emperatures spanning millennia. Mann describes the production of his, and other climate scientists’ work, as being based on an ‘extensive global network of long-term proxy data’ (Mann, 2014, p. 112). Researchers used tree rings from ancient forests and core samples for the polar ­regions as the material from which to derive these proxy temperatures. The research has been tested and expanded over a long period of time, and through the work of many international institutions and scientists. This underscores the fact that producing data can also be a very costly exercise, ­especially when considering scientific data, census data, big data, historical data, freedom of information requests, etc. Latour’s (1986) work on the ­networks of ­people, processes, and objects that are employed in the generation of scientific knowledge reminds us of the immense amount of work needed to ­produce data and inscriptions which transmit knowledge, and how this ­itself makes it difficult to counter these accounts. The criticism that Mann’s research received from Western Fuels Association-­sponsored climate change sceptics was largely on the basis of technical issues of protocol for entering this research into the ipcc report. The work needed by climate change sceptics to substantively challenge the decades-long research would be difficult to mount. It should be noted that Mann’s work, using the same data, has since been independently reproduced several times by different researchers and institutions, which have achieved similar results and made similar conclusions.

89

Data generation is part of a network of humans, artefacts, and ­processes that produces knowledge. As discussed in Chapter 1, Kitchin and ­Lauriault call this a data assemblage: ‘a complex socio-technical system, composed of many apparatuses and elements that are thoroughly entwined, whose central concern is the production of a data. A data assemblage consists of more than the data system/infrastructure itself, such as a big data system, an open data repository or a data archive, to include all of the technological, political, social and economic apparatuses that frame their nature, operation and work’ (2018, p. 8). This system is evidenced in the scientific network assembled around the production of climate data in the case of Mann’s research. It also helps us understand the parallel network of corporate actors, academics and media organizations that produces counter-arguments, although much less robust or rigorous, which ­increases uncertainty in the spheres of public opinion and public policymakers. As these interconnected networks demonstrate, facts that are derived from data require work to be maintained and work to retain their validity. This is not to say that they are any less real, but that they are constructed (Latour, 2005) through human (e.g. scientists) and non-human (e.g. trees, sensors) processes.

Chapter III

220221_CritViz_Book_Full_revisions.indd 89

Critical Visualization

7/20/22 9:47 AM

90

Figure 3.03 — Data generation is part of a network of humans, artefacts and processes which produces knowledge. Here are participants at a Data4Change Design Sprint in Nairobi (2019) tackling digital rights. Data4Change

Foucault’s (1980) concept of the dispositif is useful here to help u ­ nderstand the stakes involved in this sort of network or assemblage, for it describes the apparatus through which power/knowledge are connected. P ­ ower is secured through the establishment of a particular knowledge of a ­territory, a group of people, a phenomena, an activity, etc. Access to that knowledge and its creation and maintenance is necessarily restricted to those who uphold that power and are invested in maintaining that power. Kitchin and Lauriault (2018) note that data assemblages work similarly. These are infrastructures that operate within a larger context (e.g. politi­ cal, social, economic) and as such are transformed by them. Shaped by powerful institutions including government, industry groups, academia, scientific or health organizations, these data infrastructures take on many of the agendas (e.g. directions of research) and basic assumptions (e.g. methodology or sources of expert knowledge) that perpetuate their authority over a given field of knowledge. The difficulty in producing data and its attendant data ­assemblages also ­inherently reproduces a divide between those who are part of the a­ ssemblage and those who are not. Expert knowledge, access to funding, and technical skills are required in order to produce data. For those without access to such resources, this requirement effectively restricts access to the power of data generation. What emerges is a digital divide in (big) data where

Making Data

220221_CritViz_Book_Full_revisions.indd 90

7/20/22 9:47 AM

­ owerful organizations produce highly specialized, thoroughly d p ­ ocumented, standardized, and widely ‘trusted’ data sets while non-specialists who are usually at the receiving end of an analysis or policy decision have little access to making data or using it.

91

Big data exemplifies this dynamic. The complexity and scale of the data requires corresponding infrastructure (e.g. people, hardware, funding) and expertise to manage, use and analyze. One source of big data, social media (e.g. Facebook), amasses this material in order to understand ­consumer behaviour and advertise to its users. While a user’s own data may be downloaded in some cases, the accumulated data of all users is typically available only to the social media corporation and, in an aggregate form as an analytical tool for targeting users for its clients. Another example of this is census data, which is regularly collected by governments every few years in order to collect demographic data within its borders, and therefore, understand issues such as income levels, age distribution, family composi­tion, etc. Traditionally, this data has been collected and analyzed for government planning purposes, with certain derived products or services offered, for a fee, to researchers, organizations and corporations. QUA N TITATI V E A ND QUA LITATI V E DATA There are processes that make data conform to a set of conventions in order to allow their manipulation and combination at later stages. C ­ onventionally, in data visualization there are two general kinds of data that are used: quantitative and qualitative. Quantitative data is the kind of data that is in the form of discrete units that can be counted, categorized or analysed via mathematical means. Examples include sensor data, A/B test data, demographic data, etc. Qualitative data is typically descriptive and can be read or analysed via interpretation. Examples include text, sound, image, video, etc. Recorded interviews, journal writing, field notes, photography, or video documentation are the kind of qualitative data that is rich in information. While one kind of data is not inherently better than the other, there are practical and philosophical issues to consider when choosing whether to generate data that can be counted, or data that is descriptive. In practical application, we need to think of what it is that we are ­attempting to capture and what is the best way of doing it. A person’s height could be described in words (e.g. tiny, giant, tall, regular, big, diminutive, short,

Chapter III

220221_CritViz_Book_Full_revisions.indd 91

Critical Visualization

7/20/22 9:47 AM

92

­average, medium), or in numbers (e.g. 205 cm, 140 cm, 170 cm). One of these methods will allow you to analyse precise differences in height and the other will allow you to see the variety of terms we use to describe stature. The numbers (quantitative) will tell you little about the cultural attitudes towards stature and the words (qualitative) will tell you less about average heights of a given population — the question being asked determines the appropriateness of the data being collected and in turn the method being used. Card and McKinlay (1997) elaborated a basic framework of scales of data that has been fundamental in determining how data is used in i­ nformation and data visualization. The scales of data describe the variety of data types that allow for different forms of analysis. Quantitative data is typi­ cally thought of as either nominal (e.g. categories like married/single), ordinal (e.g. ordered scale like A+, A, A-, B+), interval (e.g. number where zero is arbitrary, like 4º c or -10º c), or ratio (e.g. number where zero is meaningful, like $100 or $0). Different mathematical operations can be ­performed with these kinds of data — simple operations like counting the frequency of a specific category (e.g. marital status), or comparing the ­difference between values (e.g. cost of housing). In contrast, qualitative data is difficult to categorize or quantify due to a number of factors — for ­instance, the variety of terms a text may have, or the ambiguity of an image. Often, qualitative data is translated into quantities by means of coding the data. For instance, someone’s opinion about a political candidate can be described by using words in an open-ended question. This same question could be framed as a closed question where the ­respondent chooses between favourable or unfavourable or no opinion. The coded re­ sponses (e.g. scale of 1 to 5) offer a quick and computable way of describing this opinion, but as you can see, there is potential for a drastic reduction of information in the response. More recently, there has been development in the automated analysis of text (e.g. responses) for semantic or ­affective content using artificial intelligence. For instance, descriptive words are counted or connections between noun, verbs, adjectives and adverbs ­identified in order to analyse sentiment. Other descriptions of types of data help us determine what can be done with the data. For instance, indexical data is unique (e.g. fingerprints, dna, social security number, passport number) and can therefore be used to identify an individual, an organization, etc. Attribute data is not unique

Making Data

220221_CritViz_Book_Full_revisions.indd 92

7/20/22 9:47 AM

(e.g. gender, height, country of origin, eye colour), but is still descriptive. Another type of data — metadata — describes the data itself. It plays an important role as the amount of data collection and storage increases, ­because it provides information (e.g. title, date, format, creator) that can be used to help curate, sort and validate data.

93

In philosophical terms, the difference between quantitative and qualitative data prompts us to consider how we perceive reality. If we think of data generation and analysis as part of an inquiry, we should look at how we frame and structure our inquiry (i.e. methodology). Creswell (1998), in his outline of the necessary components of a research design, notes that paradigms or worldviews deeply inform the basis of the inquiry itself. A worldview determines what you as a researcher and as a critical designer hold as valid ways of representing reality. Creswell lists five world views that shape our way of dealing with research and the creation of data: positivist (or naive positivist), post-positivist, social constructivist, participatory-advocacy (or transformative), or pragmatist. What distinguishes these paradigms is how each treats our ­understanding of reality and what apparatus is useful for recording phenomena. A positi­ vist approach holds that reality can be measured objectively with precise ­methods and instruments. A post-positivist approach acknowledges that objective knowledge is possible but not based on absolute truth, r­ ather through agreement by independent subjective observations. A social constructivist approach prioritizes how humans relate to their reality and holds that reality is shaped by human belief and culture. A participatory-­ advocacy approach sees the role of the inquirer as a transformative agent who makes social change and is influenced by the context and people being researched. A pragmatist approach combines all of these approaches to the extent that they help describe reality in complementary ways. By looking at how we generate data through these lenses we see what is p ­ rioritized in the process. For instance, how people use the natural environ­ment could be recorded in terms of the number of cars e­ ntering a park, aerial photography of a logging area, the number of toxic ­chemicals produced near a large body of water, photos uploaded to social media by campers, or the traditional Indigenous names used to describe a place. Each one of these data sets, and how an analysis uses the data set, is premised on a prioritization of a world view. For the purposes of a ­critical practice, the participatory-advocacy approach and pragmatist approach may be the most useful, as the first acknowledges the transformative potential of the inquiry itself, and the second triangulates between data sets by combining various approaches.

Chapter III

220221_CritViz_Book_Full_revisions.indd 93

Critical Visualization

7/20/22 9:47 AM

94

Iconoclasistas, a critical mapping group based in Buenos Aires, produces projects that illustrate the advantages of triangulating amongst ­different data sources when approaching the work from a transformative or parti­ cipatory-advocacy world view. Their process involves collaborative mapping in which they facilitate the collection and representation of knowledge in visual form. By working with community members in this way, Iconoclasistas turn around the power dynamic of researcher-subject to acknowledge the expertise of the participants. This knowledge comes in the form of stories, experiences, histories and eyewitness accounts. These are qualitative data that are rich in information and have resonances with specific audiences. For instance, a word describing a particular part of town may be known to the neighbourhood residents, but may be ambiguous to anyone else. While they don’t eschew quantitative data, the Iconoclasistas’ maps prioritize the identification of people, issues and features of a territory, so that a viewer may see where things are happening, who is being affected and who is responsible. Quantita­ tive data is often used as a base that gets problematized through the collaborative process. For instance, the location of streets and buildings offers an orientation to participant-mappers, but is annotated with text references to qualities such as smell, sight and sound that transgress the rigid locations derived from topographical maps and geolocations.

Figure 3.04 — Adding community knowledge to the re-presentation of territory: Iconoclasistas workshop, May 2019. Iconoclasistas

Figure 3.05 — Qualitative data as primary source: Iconoclasistas workshop, April 2019. Iconoclasistas

At the centre of Iconoclasistas’ practice is the work of making community knowledge accessible — to re-present in a collectively viewable platform a shared memory of place. This is accomplished through talking, writing, drawing and using icons that represent a set of identifiable topics to literally

Making Data

220221_CritViz_Book_Full_revisions.indd 94

7/20/22 9:47 AM

map out what is known in the community. A comparable approach using quantitative data is that of Data4Change, which works with non-profit groups around the world to gather new data and re-present existing data sets to shed light on pressing social, political and environmental issues, ­including internet shutdowns, displaced populations and drone strikes. (See Focus: Data4Change).

95

The qualitative data used by Iconoclasistas can also be thought of as coming from a primary source. That is, data that is derived directly from the source (e.g. interview) for the purposes of the research or project. We can contrast this with secondary sources that are published and include analysis of the data that is of interest (e.g. a review of an exhibition). There are similar distinctions for quantitative data where primary, secondary and tertiary data refer to the distance between the data and the researcher. Kitchin (2016) defines primary data as that which is produced by the researcher for their own purposes, whereas secondary data is already published, and tertiary data is published in a more aggregate form (e.g. an ­analysis is partially done and sources are anonymized). While primary data is very powerful, due to it matching very closely the needs of the investigation, much can be done with secondary (or tertiary) data. For critical visualization designers, access to the processes, instruments and labour needed to produce large amounts of primary data may be daunting or prohibitive. Using secondary and tertiary data may be a more accessible way of developing a data set. Mark Lombardi, for instance, developed an extensive series of network diagrams showing the relationships between corporate actors and government officials using news media, published reports, books, etc. Lombardi worked as a researcher, journalist and artist by developing his data set, performing the analysis and rendering a publicly viewable representation of his research. His method was a­ nalogue from beginning to end. He created index cards on particular actors,­detail­ ing important information and associations with entities on other index cards. This helped him organize and sort the nodes on his network, which he then sketched out in pencil. His final visualizations are large wall-size network maps produced using graphite and paper (Lombardi, 2004).

Chapter III

220221_CritViz_Book_Full_revisions.indd 95

Critical Visualization

7/20/22 9:47 AM

96

T H E ROL E OF CAT EG OR I Z AT ION Iconoclasistas and Mark Lombardi perform one important task very often in order to create their maps and visualizations, namely categorization. Categorization is a necessary step in the collection and analysis of data. It determines the identity of a particular thing according to a criterion or taxonomy. Bowker and Star define classification as a ‘spatio-temporal segmentation of the world’ and as ‘a set of boxes (metaphorical or literal) into which things can be put to then do some kind of work — bureaucratic or knowledge production’ (Bowker and Star, 1999, p. 10). Yet, difficulties inevitably arise when trying to place things into logical frames such as categories. William Kent (2000) reminds us that entities, which we need to identify in order to place in a category, can be highly ambiguous due to how they are defined by the people who use them. For example, a ‘book’ may refer to an author’s single self-contained work, or a copy of a book or a manual, or a publication of proceedings from a conference. Also, at which point does a book stop being a book? Does the number of pages or the thickness of the cover determine this? The simple term, ‘book’, reveals itself to be an ambiguous category once we try to establish what is included and what is not. The process of categorization becomes even more difficult when we try to frame parts of human society into orderly frames. If we take, for instance, the disparity in pay amongst genders, we are required to invoke a set of categories that would underpin such a representation or an analysis. The categories of work, pay and gender are designed through a combination of hegemonic concepts, such as recognized forms of work which often have omitted unpaid domestic work — which nonetheless is a major support for paid work — and concepts such as a male-female gender dichotomy, which erases any representation of trans people. Nevertheless, categories are extremely useful when identifying things and thinking about the relationships among things. The intuitive understanding of categories is that they contain entities with shared properties. So, for instance, cats are mammals that usually have four legs, a tail, whiskers and chase mice. This also fits the description of dogs. Yet we understand cats and dogs to be separate categories. We can also think of them both as belonging to the pets category. Categories, through human cognitive work, impose frames on the world so that we can think about, and manipulate, the world. These frames are highly influenced by context of use.

Making Data

220221_CritViz_Book_Full_revisions.indd 96

7/20/22 9:47 AM

Jorge Luis Borges’ fictitious taxonomy ‘Emporio celestial de ­conocimientos benévolos’ (Celestial Emporium of Benevolent Knowledge) is a good example of how categories impose human values on the world. Written in 1942, it is part of a short piece about John Wilkin, an English man of letters from the seventeenth century, who attempts to devise an analytical language capable of systematically and rationally describing the whole universe. It is noted that the attempt was doomed from the start.

97

These ambiguities, redundancies and deficiencies remind us of those which ­doctor Franz Kuhn attributes to a certain Chinese encyclopaedia e­ ntitled “­Celestial Emporium of benevolent Knowledge”. In its remote pages it is written that the animals are divided into: (a) belonging to the emperor, (b) ­embalmed, (c) tame, (d) suckling pigs, (e) sirens, (f) ­fabulous, (g) stray dogs, (h) included in the present classification, (i) frenzied, (j) ­i nnumerable, (k) drawn with a very fine camelhair brush, (l) et cetera, (m) having just ­broken the water pitcher, (n) that from a long way off look like flies. (­Borges, 1942/1999)

Foucault (1966/2007), in The Order of Things, uses Borges’ short story as a reminder of the foolhardiness of the attempt, the arbitrariness of the premise, and the compulsion of humans to categorize things in the world. Lakoff (1990) notes that Borges’ passage also reveals a deeper truth about Western attitudes towards non-Western thinking — that it is strange, fantastical and inscrutable, part of an orientalist tradition that treats non-European culture as an other that is either romanticized or vilified, or both. This also reproduces a pervasive trope that categories are self-­generating, that is, they are evident from the things themselves and it is up to the most advanced scientific expertise (e.g. Western knowledge) to properly name them. Yes, categories are necessary for thinking, but they are relative to their context (Lakoff, 1990) since they are culturally produced and any attempt at producing a universally valid system inevitably erases or marginalizes those things (including people) that do not fit. Why would categorization be important for a critical visualization designer? If we recall Foucault’s concept of the dispositif or apparatus, or Kitchin and Lauriault’s data assemblage, we see the inherent issue of power/­know­ledge in the creation of data and in the establishment of categories. These are processes that structure what we know about the world and its inhabitants and thus have an impact. The human work required to produce data about a human world also means that there is a mutually affecting relationship between that which is classified and that which classifies; Ian Hacking

Chapter III

220221_CritViz_Book_Full_revisions.indd 97

Critical Visualization

7/20/22 9:47 AM

98

(2006) calls this dynamic nominalism. This is when naming and categorizing result in making up people. Identities are literally created to describe parts of society according to particular criteria. For example, people may be categorized as autistic, affluent, bi-racial or healthy. Each of these categories require an individual to possess a set of characteristics in order to qualify. A definitive set of qualities is determined by researchers to be exemplary of that category. The use (in communications, planning, p ­ olicy, services, etc.) of the language and the category established to describe a person of this kind influences the person described. Hacking (2006) describes two processes that are at work in this d ­ ynamic: looping effects and engines of discoverability. The looping effect is the process through which data are organized (e.g. classification, the object of focus, the institution, knowledge, various experts) about a given subject (i.e. a person with particular characteristics), who is in turn affected by that description and thus changes in response. On the other hand, the engines of discoverability are the processes that establish a body of knowledge (e.g. biology, medicine, genetics) that attempts to frame the p ­ henomena in its own terms — as Hacking puts it, to biologize, to medicalize, or to ­geneticize, for example. Dynamic nominalism is especially salient when we consider how the human world is described. As social-cultural beings, humans are ­deeply affected by the language and the terms that describe their own e­ xistence, which effectively outline the range of possibilities of living their lives. Hacking (2006) uses examples of sexual orientation and ­psychological ­conditions which, once categorized and named in discrete terms, ­inevitably determine a field of experiences heavily shaped by those ­categories. We can use Hacking’s observations to understand how data through c­ ategorization and, more specifically for designers, the visual ­representation of these data and categories, help produce subjects. As a simple exercise, try to describe yourself by using standardized cate­ gor­ies from a census. For instance, Statistics Canada before 2016 used the ­following single response classifications for identifying race:

Making Data

220221_CritViz_Book_Full_revisions.indd 98

7/20/22 9:47 AM

99

1.1 Aboriginal (Inuit, Métis, North American Indian) 1.2 Arab/West Asian (e.g., Armenian, Egyptian, Iranian, ­Lebanese, ­Moroccan) 1.3 Black (e.g., African, Haitian, Jamaican, Somali) 1.4 Chinese 1.5 Filipino 1.6 Japanese 1.7 Korean 1.8 Latin American 1.9 South Asian 1.10 South East Asian 1.11 White (Caucasian) 1.12 Other This would prove difficult if you were Afro-Peruvian, because a respondent can only choose one response — you can select Latin American or Black, but not both. This leads to a significant lack of precision and also an erasure of a particular identity, diaspora or history. Translating from the realm of personal identity and culture that is immensely complex to a domain of countable units leads to the loss of rich information. If we take from Hacking’s observation the effect of categories on people, then we can see how a language (i.e. a system of labels for human identity and experience), which does not include one’s own, could affect one’s own way of expressing one’s identity and experience. These categories are not arbitrarily set, but are produced by a dominant cultural point of view. The official census reflects back to its residents, in this case racial categories, that are based on language already established — not necessarily at the individual or community level but at the hegemonic or dominant cultural level. Furthermore, as Hacking (2006) reminds us, once people are categorized, actions are taken upon them in the form of policymaking, investment, service delivery, etc. Charles Booth, nineteenth-century industrialist and social policy researcher, developed the Maps Descriptive of London Poverty (part of his Inquiry into Life and Labour of the People in London), which detailed socio-economic distribution in the city. Booth’s collection comprises twelve to thirteen maps that together form a grid of the city of London in the 1890s.

Chapter III

220221_CritViz_Book_Full_revisions.indd 99

Critical Visualization

7/20/22 9:47 AM

100

Figure 3.06 — Charles’s Booth Poverty Map as digitized by the London School of Economics with original categorization of neighbourhoods. Charles Booth and 2016 London School of Economics and Political Science

What is significant about this collection of maps is both the scale and the comprehensive nature of the endeavour. For our understanding of data and its function in categorizing human populations it is also significant. Booth used seven categories in his legend: →  Lowest class. Vicious, semi-criminal. →  Very poor, casual. Chronic want. →  Poor. 18s. to 21s. a week for a moderate family. →  Mixed. Some comfortable others poor. →  Fairly comfortable. Good ordinary earnings. →  Middle class. Well-to-do. →  Upper-middle and upper classes. Wealthy. Booth utilized the poverty line to delineate a basic division in the city ­between those who could barely survive and those who could survive at varying levels of comfort. Although extensive research was ­conducted through surveying neighbourhoods in order to develop a basis for the ­poverty line (Hacking, 2000), little is known about how exactly Booth established the criterion (i.e. 18 to 21 shillings a week). This categorization did have an effect, as Lauriault (2012) notes, on how poverty was measured in the colonies and the United States — in Canada it directly influenced the use of the Low Income Cut Off used by Statistics Canada.

Making Data

220221_CritViz_Book_Full_revisions.indd 100

7/20/22 9:47 AM

Compare Booth’s work with the identification of Priority Neighbourhoods in Toronto in 2004, later renamed and reconciled as Neighbourhood Improvement Areas (nia) in 2014. Out of the approximately 140 neighbourhoods in the city, thirteen priority areas were first identified, then this was expanded to thirty-one areas. Initially set up by using socio-economic indicators, proximity to services and the number of annual homicides (Doolittle, 2014), the category was reframed as nia in order to take into account many more factors, such as health, voter turnout and income level (Doolittle, 2014). The data used in these designations and measurements come from a variety of sources, including the 2011 Census, which collected information on housing, family structure, spoken languages and the 2011 National Household survey, which collected immigration history, ethnic origin, visible minority identification, education and income. These areas are designated at-risk and vulnerable due to lower income levels, higher homicide rate and poor access to services.

101

Figure 3.07 — Map of Neighbourhood Improvement Areas. City of Toronto, Social Development Finance and Administration, Social Policy Analysis and Research, October 2014

Similar to Booth’s maps, the data from surveys are mapped onto the physical representation of the city in order to identify concentrations of p ­ overty and thus help direct the government’s social and economic policy to improve the conditions and prospects of these residents. Data is generated, reaggregated and analysed for the purposes of governmentality — in other words, to make people governable (Porter, 1995).

Chapter III

220221_CritViz_Book_Full_revisions.indd 101

Critical Visualization

7/20/22 9:47 AM

102

Lauriault (2012) usefully extends Hacking’s making up people framework by introducing terms such as space (rather than people) and scientification (rather than biologizing) and adjusts it in order to account for the effects on spatial issues as well as inhabitants — what she terms making up s­ paces. Institutions such as Statistics Canada and Natural Resources Canada, through objects like the Atlas of Canada, perform a looping effect that results in making up spaces, according to Lauriault (2012). These representations of space and people produce and reproduce a geographic imaginary. Maps of low-income areas contribute to a geographic imaginary. For example, some city councillors reported not wanting nias identified because of the stigma associated with them — ostensibly responding to the residents not wanting to be further stereotyped by these designations. Others welcomed the designation as it would result in targeted funding for community projects, etc. We must conceive of data as always situated — meaning that they are always gathered at a certain time in a certain place for a certain reason. As Kitchin and Lauriault note, ‘how data are conceived, measured and employed ­actively frames their nature’ (2018). Gitelman and Jackson liken this to how a photographer frames a photo (2013). Donna Haraway (1988) ­suggests that situated is a useful term to contrast the objective, neutral gaze that ­scientific inquiry and knowledge (and mapping and ­visualization) is thought to have. The ideal of objectivity is also supported by the notion that a ­mechanical process is objective since data is generated by an apparatus that is non-­ human and therefore not subject to the limits of human subjectivity. As Daston and Galison (1992) have noted, the nineteenth century saw many advances in this through the development of photographic and instrumental methods of recording phenomena. Again, it was the notion of letting nature speak for itself that was prioritized. This coincides with how photography at the time was seen by many as a technical process that simply recorded things as they were. Henry Fox Talbot’s Pencil of Nature is one example of how photographic processes were seen to remove the human element in order to let the beauty of nature express itself. Visualization, when framed as ideally letting data speak for itself, as men­ tioned above, reproduces this same objective, technical and positivist

Making Data

220221_CritViz_Book_Full_revisions.indd 102

7/20/22 9:47 AM

attitude towards the apparatus of representing phenomena. Objectivity is used to secure the validity of a claim through a passionless and neutral observation of an event — stepping outside of our identity, desires, history, time and place. Historically, that ability, through machines of representation that produce a mechanical objectivity, has been exclusively granted to the white European male. We see this in what Scott (1998) describes as the ideo­logy of high Modernism, which takes from the legitimacy of science and technology, and demonstrates a desire for rational order, a lack of criticality or scepticism.

103

Haraway argues that what has resulted is a trope of masculinist k ­ nowledge that sets up a dichotomy in which the other type of knowledge is ­reduced to the feminine, the marked body, the non-white or the other. Rather than frame these alternate dimensions as lacking o ­ bjectivity, universality or validity, these positions can be framed in terms of producing a different kind of knowledge that is extremely valuable. This knowledge is situated in time and place, and is directed towards a thing — all of which informs what is known. She suggests that we should value the proximate, complicit and committed, and not disavow it in favour of the distant, innocent and objective. Similar to Haraway’s notion of situatedness, Loukissas (2016) urges that we should pay close attention to the local dimension of data (especially Big Data) so that we understand context and the meaning that specific characteristics of the data may contain. Much can be lost in the attempt to clean data, for instance. Dirty data, Loukissas explains, can contain important indicators of how the data was generated in the first place. The process of eliminating the noise or the gaps in the data also further decontextualizes it. The decontextualization and distancing from original sources can occur when data is aggregated (i.e. combined with other data) and/or re-mapped (i.e. translated from one context to another). Leonelli (2014) notes that the combination of very large data sets (e.g. scientific data), while part of the power of big data, also makes them potentially very vulnerable and difficult to disaggregate (i.e. take apart) and reuse or verify. Bates et al. (2016) use the term data journey to describe the different technical and social contexts through which data travel during their production, analysis and distribution. Data journeys contribute to that distance between source and

Chapter III

220221_CritViz_Book_Full_revisions.indd 103

Critical Visualization

7/20/22 9:47 AM

104

point of viewing or analysis, when many steps and processes take place in-between. This also reinforces the importance in thinking about the relational dimension of data, in the sense that there is an infrastructure (or assemblage) and a process (or journey). Haraway’s insistence on the value of situatedness also applies to the site of analysis and dissemination of data. One way to recontextualize or situate data again is to have subject area experts familiar with the site of data and its generation acquire the data and use it. What is important to define here is expertise. Expert knowledge in data science or on the particular topic would be an obvious start. Another dimension, respecting the importance of situatedness, would be lived experience regarding the data, which would result in a more proximate position. Rather than prioritize the distanced, unmarked, or neutral ideal of the scientist, we must acknowledge the importance of the proximate, affected and committed reality of the person working with the data. Take the open letter Yeshimabeit Milner (2018), from Data for Black Lives, addressed to Facebook, where she states that ‘data is destiny. For Black people, who have been disproportionately harmed by data-driven decision-making, this is especially true.’ Milner makes three demands in the letter: 1. 2. 3.

Commit anonymized Facebook data to a Public Data Trust. Work with technologists, advocates and ethicists to establish a Data Code of Ethics. Hire Black data scientists and research scientists.

Apart from providing access to the data in the form of a trust and establishing ethical standards, Milner emphasizes the importance of Black scientists doing the data analysis. Furthermore, as U.S. government data about issues that disproportionately affect Black lives become increasingly more difficult to access, Facebook data, according to Milner, can potentially fill a need for current data. The lived experience of the researchers and designers who create algorithms, aggregate data, visualize analyses and draw conclusions provides a dimension of care and knowledge that has been largely absent from dominant data infrastructures and assemblages.

Making Data

220221_CritViz_Book_Full_revisions.indd 104

7/20/22 9:47 AM

Visualization can reintroduce a kind of proximity to the source of data. Margaret Pearce is a cartographer and designer who works to maintain a situatedness in the maps she creates. In several of her projects, her data, per se, is the traditional knowledge maintained in Indigenous communities on Turtle Island (i.e. the North American continent). These are often in the form of place names that describe a place and embed a knowledge of the environment, the flora, the fauna and how they relate to people.

105

Denotative type symbols with canoe routes

canoe routes

Sample connotational combinations for point, line, and area old

Town

new

Town

carries

Denotative type symbols, connotative graphic symbols

Text and graphic symbol combinations for point, line, and area town

matάkəməssək

r

R i v e r

physical

cultural

Ri v e r

physic

al feature

wəskʷan

gwi

less important

wəlakesi

C

P it aube

important

men

aha

nuk

rive

Gluskabe leapt

calf began swimming across

Gluskabe landed

dog swam

I T Y

cultural feature

Figure 3.08 — Map details and illustrations. Margaret Pearce

Chapter III

220221_CritViz_Book_Full_revisions.indd 105

Critical Visualization

7/20/22 9:47 AM

106

Pearce (2014) notes that translation and representation can result in a sepa­ ration of knowledge from context, which results in a significant loss of meaning. Data, in the form of location and place names, needed to be translated (from its traditional usage and its oral form) in a highly nuanced way, in order to both keep the integrity of the knowledge represented on the map and simultaneously to be accessible to readers who may or may not speak the original Indigenous language. The mapping of I­ ndigenous knowledge highlights the issue of translating complex information (i.e. ­embodied knowledge that is interwoven with culture and territory) from one domain into another that reduces and separates data from its context. Criticality in visualization requires that we interrogate the data and information that we are re-presenting in visual form. This interrogation requires that we ask: →  How was the data produced? →  How/where is the data production situated? →  W hat biases may have framed it? →  How is the data aggregated and decontextualized? →  W hat is gained/lost when it is mapped into other contexts? →  W hat set of skills or knowledge is needed to understand the data from its original moment of creation/collection? →  W ho and what is erased when the data is produced? The definition and context of data can be explored in many ways, each focusing on a discipline, or technology, or usage or policy. In this chapter, we discussed the ways in which data is framed as an assemblage or infrastructure and we also discussed some critical visualization practices that address the ethical ways that data and visualizations are produced. Data does not simply arrive, nor is it something we discover. Instead, we make data from people and things that we want to represent. It is a material born out of these relations with sources.

Making Data

220221_CritViz_Book_Full_revisions.indd 106

7/20/22 9:47 AM

107

Figure 3.09 — T he SizeChina project was initiated to address the exclusion of Asian head shapes in the dominant data set derived from US Air Force personnel. Roger Ball

The ease with which we can access data and combine data can simultaneously make the system that supports this access and interoperability invisible. When things become invisible they get normalized and less scrutinized for what they reproduce (e.g. what is prioritized and what is omitted). This is precisely the moment when criticality is necessary. The critical stance prompts the questions: ‘Who is not here?’, ‘What has been excluded?’, ‘Why was it not included?’ and ‘How is the status quo ­reproduced? Data generation is part of knowledge production. It is the generation of material that can help frame a debate or dispel a myth. As designers of ­visualizations, we choose what data and which stories are amplified through our work. As critical visualization designers, we assume a responsibility for producing those stories and for the ways they were produced. █

Chapter III

220221_CritViz_Book_Full_revisions.indd 107

Critical Visualization

7/20/22 9:47 AM

Data4Change

108

Rather than focus on data visualization per se, the Londonbased non-profit Data4Change works to ‘harness the power’ of data, in collaboration with local civil society organizations around the world in gathering, organizing and presenting data to achieve lasting change. The group’s co-founder Bronwen Robertson started Data4Change in 2014 with journalist Stina Bäcker, based on the knowledge that ‘troves of data are locked away in PDF reports that could help create powerful counternarratives to data sets that pretend to speak on behalf of populations’ (B. Robertson, personal communication, October 10, 2020). The group has worked on nearly fifty projects in over ninety countries since 2014, including the three projects shown here, which set out to: 1. visualize how internet shutdowns are impacting human rights, 2. advocate for people with visual impairments in Ethiopia, and 3. better represent and improve the living conditions of people in Yemen. The primary tool for Data4Change is the three-to-five-day sprint, following the UK Design Council’s ‘double diamond’ framework for innovation — discover-define-develop-deliver. However, as the organization’s network has grown, it has begun to explore the possibilities of a decentralized network of hubs to increase and amplify existing capacities in the various regions in which it works.

Focus Figure 2.1 — Data4Change sprints make sure that data is readable and available for targeted groups and communities. Data4Change

Rethinking the Representation of Data 220221_CritViz_Book_Full_revisions.indd 108

7/20/22 9:47 AM

Focus

109

1. KeepItOn Strategically shutting down internet access has become a political weapon used increasingly by governments in recent years. Access Now, an advocacy organization focused on defending and extending the digital rights of users at risk, established the KeepItOn interactive platform to draw attention to the impact of shutdowns on human rights, businesses, emergency services and journalism around the world. At a Data4Change sprint in Nairobi, Kenya in 2019, the #KeepItOn coalition worked with a creative team to build a new interactive visualization of data from three years of internet shutdown instances and crowdsourced stories submitted to #KeepItOn.

The project has focused on helping Access Now’s #KeepItOn coalition develop a sustainable platform for data gathering, by linking shared spreadsheets to the website so that live data can update the visualizations with minimal effort and cost. Accord­ ing to a spokesperson for Access Now, ‘the new data entry system has streamlined our data collection and documentation process and, as a result, has increased our data accuracy, reliability and verification process’ (B. Robertson, personal communication, October 10, 2020).

Focus Figure 2.2 — Internet shutdowns by region, KeepItOn (2019). Access Now/Data4Change

Data4Change 220221_CritViz_Book_Full_revisions.indd 109

7/20/22 9:47 AM

110

Focus

Focus Figure 2.3 — Internet shutdowns tracked in the Democratic Republic of the Congo between 2016 and 2018. Access Now/Data4Change

Rethinking the Representation of Data 220221_CritViz_Book_Full_revisions.indd 110

7/20/22 9:47 AM

Focus Robertson points out that the data is often difficult to obtain, because governments are not usually forth­ coming about internet shutdowns, and will sometimes use surreptitious methods to restrict access such as throttling. This describes a tactic where a government issues a mandate to internet service providers to arti­ ficially slow down internet traffic, which is more difficult to identify and track than a full network shutdown. ‘In a full network shutdown, people very quickly realize they can’t access communications technologies, but with throttling, it can be much less apparent.’

111

A particular focus for Access Now’s #KeepItOn coalition is developing a community standard to reduce the potential harm that its measurement tools might cause to users at risk. ‘We are slowly trying to build con­ sen­sus around the criteria and agree on an implementation mechanism that does not necessarily punish tool developers,’ reports the organ­ization, ‘but rather help them meet these standards and continue providing essential services’ (KeepItOn coalition and B. Robertson, personal communi­ cation, October 6, 2020).

2. Hear the Blind Spot

Focus Figure 2.4 — Data sonifications were developed to support a campaign about visual impairment in Ethiopia and the impact of digital exclusion. Together! Ethiopia/Kim Albrecht/Data4Change

Launched at the Forum on Internet Freedom in Africa in 2019, Hear the Blind Spot uses ‘data sonifications’, developed by designer Kim Albrecht (see also The Hairs of Your Head are Numbered in Chapter 4), to convey the findings of a survey of visually impaired Ethiopians about their access to technology, education and employment — as well as their feelings of social and digital exclusion. Case studies are conveyed as spoken  narratives accompanied by

sonifications of relevant data sampled from notes played on a flute. The survey was conducted with Together! Ethiopian Residents Charity and found that while one in twenty people in Ethiopia has a visual impairment, they face considerable challenges in accessing digital technology due to insufficient training, inaccessible websites and limited access to tools and resources like screen readers (Data4Change, 2020).

Data4Change 220221_CritViz_Book_Full_revisions.indd 111

7/20/22 9:47 AM

112

Focus

Focus Figure 2.4 — Data sonifications were developed to support a campaign about visual impairment in Ethiopia and the impact of digital exclusion. Together! Ethiopia/Kim Albrecht/Data4Change

The sonifications were developed was very positive, one challenge throughout a five-day sprint in Nairobi, with data sonification is that it is Kenya by German designer and difficult to achieve aesthetically researcher Kim Albrecht, as part of the appealing outcomes while retaining experimental push of a Data4Change accuracy. ‘It works well when you sprint. While the various dimensions use a single note or pitch to convey of visual mapping are at least well the up and down movement of a defined since Bertin’s Semiology of line graph, e.g. translating the data Graphics (1967/2011), the translation directly onto pitch, but attempts to between data and sound was, at least then layer in other data using chords for Albrecht, still an untold territory. or other harmonic elements, can He experimented with mapping make it overly complicated and numeric values to different audio lose the specificity’ (B. Robertson, elements in the same way that data personal communication, October can be mapped to various visual 10, 2020). Going forward, the project elements. Instead of using colour is being rolled out through more intensity, size or shape to convey live performances in schools, after data variables, he experimented with delays due to the COVID-19 pandemic. the tone of various instruments, the The plan was also modified into length of notes, pitch variability, an audio broadcast of key health volume levels and trigger attacks. information about the virus that was As Albrecht notes, ‘The aesthetic of played via a loudspeaker attached sound is entirely different from images. to a roving four-wheel-drive car. The mediation of an image functions ‘One key takeaway,’ adds Robertson, in space while sound functions ‘is not to only think about “new” throughout time.’ (Backer, 2020). technologies when embarking on data projects, as sometimes it’s more Robertson notes that while the feed­ effective to “hack the box” and use back from a live performance of Hear existing opportunities’ (personal the Blind Spot at the 2019 forum communication, October 10, 2020).

Rethinking the Representation of Data 220221_CritViz_Book_Full_revisions.indd 112

7/20/22 9:47 AM

Focus

113

Focus Figure 2.5 — Perceiving Yemen is a 5MB downloadable website of data on living conditions across Yemen, designed to minimize energy and data use. Yemen Polling Center/Data4Change

3. Perceiving Yemen Researchers working for the Yemen Polling Center (YPC), an independent research group established in 2004, face considerable risks in conducting in-person interviews and delivering surveys amid the ongoing conflict. At a 2018 sprint with Data4Change in Bierut, Lebanon, the assembled team explored ways to make two important recent YPC survey results accessible to policymakers and the general public. The first survey, a face-to-face poll of

1,500 young people aged fifteen to twenty-five, included questions on media use and the security situation; the second survey consisted of 4,000 responses from Yemenis about their perception of living conditions and security. At the same event, the sprint team developed a bespoke interface and survey creator, and a system of Wi-Fi hubs as a way to make data collection safer and more efficient (Data4Change, 2020).

Data4Change 220221_CritViz_Book_Full_revisions.indd 113

7/20/22 9:47 AM

114

Focus

As well as the dangers of being in war zones, where smartphones and tablets used by researchers can be regarded as tools for spying, YPC faces the challenges of providing access to collected data in a nation where electricity supply can be intermittent, the cost of (satellitebased) internet is high, and literacy rates are low (thirty-five per cent of the population is illiterate). Taking this into consideration, the sprint team developed a bilingual platform that provides an interactive map which allows users to compare and filter the survey results by topic or governorate, and share with others via Twitter, Bluetooth or WhatsApp. To avoid draining battery life for users with limited electricity, the site uses a dark screen background and only one image file, keeping the site to under 5MB, which enables people to download the entire Perceiving Yemen website to their phones or computers.

ʻThe team learnt better ways related to data visualizing that enable the organization to present the collected data with easy read methods, as well as making sure these data are readable and available for more targeted groups and people in the communityʼ (B. Robertson, personal communication, October 10, 2020). ●

While the Wi-Fi hubs have yet to be rolled out across the country (due to the war and COVID-19 pandemic) a spokesperson for YPC reports that the interaction with Data4Change proved particularly valuable for sharing ways to make data more accessible and understandable:

Rethinking the Representation of Data 220221_CritViz_Book_Full_revisions.indd 114

7/20/22 9:47 AM

220221_CritViz_Book_Full_revisions.indd 115

7/20/22 9:47 AM

IV Data and the Self

220221_CritViz_Book_Full_revisions.indd 116

7/20/22 9:47 AM

S

ince its inaugur a l ev ent in 2011, the qua ntified self (qs) Conference has brought together self-tracking enthusiasts to showcase and discuss their technology and practices of tracking and visualizing personal data. Using an array of biometric sensors, mobile devices, and sharing media, presenters have tracked their food consumption, sleep patterns, mood, activity, heart rate, blood-sugar levels, location, local air quality and so on. In 2020, the Quantified Self website had inspired a covid-19 self-tracking spin-off, Quantified Flu, based on the hunch that heart rate data might be predictive of oncoming illness. Founded in California by Gary Wolf and Kevin Kelly (of Wired magazine) in 2008, the Quantified Self organization hosts ‘international meetings, conferences and expositions, community forums, web content and services, and a guide to self-tracking tools’. Its Meetups page boasts nearly 100,000 members, and its expos and conferences attract an array of sponsors who sell the biometric sensors, wearable technology and online platforms on which self-trackers upload, share and compare their personal data.

117

At the heart of the QS ethos is a somewhat utopian faith in ­technology’s capacity to shed light on human health and well-being through the ­power of data, what Geoffrey Bowker calls an ‘oddly worshipful effort to c­ elebrate’ quantification (Bowker, 2013). In a New York Times article in 2010, Wolf argues that ‘we tolerate the pathologies of quantification — a dry, abstract, mechanical type of knowledge — because the results are so powerful’. Whereas for a long time, he writes, people’s personal lives were apparently immune to quantification, gradually and imperceptibly, ‘numbers are infiltrating the last redoubts of the personal. Sleep, exercise, sex, food, mood, location, alertness, productivity, even spiritual well-being are being tracked and measured, shared and displayed’ (Wolf, 2010). The optimistic tone in QS discourse is characterized by sociologist ­Deborah Lupton as the belief that ‘digital data generate new forms of ­knowledge and new insights into people’s bodies and selves’ (and by critics as ‘data fetishism’). The QS organization’s tagline of ‘self knowledge through ­numbers’ is elaborated in the numerous self-tracking reports and testimonials on its blog and conference proceedings, where the self is typically presented as an entity with inputs and outputs. Paraphrasing Lord Kelvin, Kevin Kelly writes, ‘Unless something can be measured it cannot be improved’ (cited in P. Moore and Robinson, 2016, p. 2780).

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 117

7/20/22 9:47 AM

118

At the 2018 Quantified Self Conference in Portland, Oregon, over half of the presentations were on topics related to personal health, diet and exercise: a runner had tracked his sneezes to ascertain which grasses aggravate his allergies; another presenter surveyed the services that analyse dna swabs to determine health risks. This preoccupation appears to accurately represent the interests of the particular demographic that has taken up the practice of self-tracking, and, perhaps, the economic driver for the movement: the technology. Indeed, the conference exhibit provides an opportunity for toolmakers to showcase their technologies to ‘highly engaged advanced users and the self-tracking public’ (Quantified Self, 2018). The question of what is being visualized, and how, begins to reveal motivations. The visual display of personal data ranges from the default outputs of various data visualization softwares, to experimental programmable sculptures, as in Stephen Cartwright’s datascapes of his hourly location, in latitude, longitude and elevation. A tendency to present personal data without critical analysis, in the spirit of sharing to gain greater insights, is a Quantified Self-er trait acknowledged and upheld by Wolf: ‘For many self-trackers, the goal is unknown. Although they may take up tracking with a specific question in mind, they continue because they believe their numbers hold secrets that they can’t afford to ignore, including answers to questions they have not yet thought to ask’ (Wolf, 2010).

Figure 4.01 — Mesh 1 (Bicycling mileage 1997–2009), 2010; Acrylic, hardware; 78 × 78 × 20 inches. Stephen Cartwright

Data and the Self

220221_CritViz_Book_Full_revisions.indd 118

7/20/22 9:47 AM

Tiffany Qi’s ‘College Performance’ project is an example: according to the QS site, during her four years of undergrad, Qi tracked her time, organiz­ing and colour-coding it into categories, e.g., ‘exercise’, ‘class’, ‘games/time-wasting’ and ‘homework and studies’ analysing the data with questions such as:

119

→  Did Qi’s commitment to her studies wane over time? →  How much did time spent studying matter for Qi’s fi ­ nal grade? →  Did the amount of time spent on fun help or hinder her grades? →  Did having a job or other job-like responsibilities lower Qi’s grades? The productivist value system behind these research questions is also ­reinforced in Qi’s bar chart visualizations. Prominent colours (red and orange) are reserved for class, studies and exercise, while the ‘games/ time-­wasting’ category is rendered in a dull brown. The QS impulse to bare all and let the data ‘speak for themselves’ (see Chapter 3) not only recalls the Tufte-Shneiderman investment in the notion of data’s neutrality, it gives many presentations and accompanying visualizations the air of a public confession (Sharon and Zandbergen, 2017, p. 1704). Aside from tracking diet and exercise, QS-ers have presented visualizations of their sleep patterns, headaches, moods, ovulatory cycles and shopping habits, body scars and their causes. Here, confesses the Quantified Self to her compatriots, is what the data tells you about me: based on this, what can you tell me about me?

Figure 4.02 — Weekly calendar. Tiffany Qi

Chapter IV

220221_CritViz_Book_Full_revisions.indd 119

Critical Visualization

7/20/22 9:47 AM

120

The confession analogy might be pertinent in light of the twentieth-century substitution of the confessional for the psychiatrist’s couch. Citing examples of self-trackers who monitored the beneficial effects of specific dietary supplements on their concentration and insomnia, and another who took control of mood swings through self-tracking, Wolf is inspired to suggest that a new form of therapy is emerging: The contrast to the traditional therapeutic notion of personal development is striking. When we quantify ourselves, there isn’t the imperative to see through our daily existence into a truth buried at a deeper level. Instead, the self of our most trivial thoughts and actions, the self that, without technical help, we might barely notice or recall, is understood as the self we ought to get to know. Behind the allure of the quantified self is a guess that many of our problems come from simply lacking the instruments to understand who we are. (Wolf, 2010)

The question of how personal data is captured and visualized is ­perhaps the most prominent topic of discussion among self-trackers, whose ­interest in methods and biometric, sensing and sharing technologies is ­indicative of what Tony Fry terms a functionalist imperative: ‘a p ­ osition which ­effectively reduces the question of why something exists to its f­ unction’ (T. Fry, 1999, p. 37). A pioneer in the field of self-tracking who, in some ways, helped legitimize the visual datafication of the personal, is the graphic d ­ esigner Nicholas Felton. Between 2005 and 2014, Felton produced ‘personal annual reports’, named for an imaginary organization called ­Feltron. These first appeared as a kind of parody of the ‘aesthetics of administration’ (Buchloh, 1990) — the visual and textual language of the corporate annual report — while conveying elements of the modern-day lifestreamer’s narcissism. The debut Feltron report quantified in statistical charts everything from kinds of meals eaten, miles run, how many emails and texts he had sent, which books he had read, photographs taken per country visit, to the amount of time spent at work and play. These were initially documented without purpose-built self-tracking tools, wearables or sensors. The Feltron 2008 Annual Report reveals an infatuation with the visual ­display of data symptomatic of the era (see, for example, the oversized ­figures introduced into the Bloomberg New York headquarters by Pentagram Design (Hall, 2005)). This report conveyed everything in distances, so that Felton’s year could be expressed as a single, individual number: 38,524 miles. He explained that, ‘Included in this distance are miles flown, ­driven, walked, swam, run alongside those travelled in the video game Grand Theft Auto IV’ (Felton, 2008, p. 3).

Data and the Self

220221_CritViz_Book_Full_revisions.indd 120

7/20/22 9:47 AM

121

Figure 4.03 — Feltron Annual Report (2014). Nicholas Felton

In the tenth and final Feltron Annual Report 2014, Felton was able to produce the year’s data using entirely off-the-shelf self-tracking methods, rather than the analogue approach with which he had begun. Felton noted that the state of self-tracking had transformed in ten years from homegrown, to niche industry of self-tracking equipment and platforms: ‘When the project began in 2005, the music-tracking website Last.fm was the only service capable of automatically capturing a category of personal data. Ten years later, listening habits can be augmented with persistent location data, categories and amounts of physical activity, sleep, weight, continuous heart rate, blood-alcohol levels, driving habits and computer usage. This report attempts to merge all of this information in a format that reveals connections, provides context and suggests correlations’ (Felton, 2014).

Chapter IV

220221_CritViz_Book_Full_revisions.indd 121

Critical Visualization

7/20/22 9:47 AM

122

Figure 4.04 — Feltron Annual Report (2005). Nicholas Felton

TAY L OR IS M W I T H I N? It is productive at this point to challenge the functionalist imperative and (following the critique prompt outlined in Chapter 1), ask why self-trackers are driven to quantify and share details of their personal lives in data visualizations. In his chapter-long skewering of the advent of the ‘datasexual’, cultural critic Evgeny Morozov singles out Felton’s annual reports and healthy lifestyle as evidence of a latter-day, indivi­ dualized scientific management he calls ‘Taylorism within’, noting Felton’s subsequent recruitment by Facebook (Morozov, 2013, p. 228). The graphic designer Michael Johnson (2008) described Felton’s a­ nnual report as a classic example of ‘me-projects’, functioning as a ‘sort-of typographic Truman Show’, but ultimately serving to promote his design ­practice — a ­narcissism-as-business-model critique that is echoed in the popular press, where self-tracking with technology has been characterized as ‘extreme navel-gazing’ (Hill, 2011). The ‘navel-gazing’ and self-promotion argument is rebuffed by Quantified Self-ers who argue that tracking and sharing personal data is a new kind of knowledge production. Wolf notes that more than 30,000 new personal tracking projects are started by users every month on one of the largest

Data and the Self

220221_CritViz_Book_Full_revisions.indd 122

7/20/22 9:47 AM

internet forums for health information, OnMedHelp (Wolf, 2010). Felton subscribes to a similar position, stating in one interview that tracking and measuring his life has provided unexpected insights: ‘The Reports once inspired me to be more adventurous and to say “yes” to activities that I would naturally decline as it might make for an interesting story at the end of the year. [But] now that they have become so ingrained in my behaviour, I am far less likely to be swayed by their influence. In general, I think the Reports have made me much more aware of my routines and grateful when I can break from them’ (Houston, 2012).

123

A why critique would acknowledge Felton’s insights and the flurry of health-­ information sharers and seekers online but then ask, ‘In whose interests is this trend?’ To critique the visualization, as noted previously, requires that we unpick the assemblage in order to understand how various interests are assembled in the very act of recording one’s life. Morozov’s reference to ‘Taylorism within’ indicates that the impulse to measure one’s activity to the smallest detail in the interest of self-improvement is a practice inherited from the productivism of the late nineteenth century. The historian Anson Rabinbach has argued that modern p ­ roductivism  — ­ ‘the belief that human society and nature are linked by the primacy and ­ultimate interchangeability (convertibility) of all productive activity’ (Rabinbach, 1992, p. 22) — first arose from scientific discoveries of the nineteenth century, particularly thermodynamics. The metaphors and images used in the scientific theories of Hermann von Helmholtz, Lord Kelvin and Rudolf Clausius not only reflected theological and social conceptions of the world, but presupposed that the forces of nature were forms of a ­single, universal energy that could not be added to or destroyed. Mankind’s imperative was to mirror nature’s own ‘unlimited capacity for production’ (Rabinbach, 1992, p. 2), reflecting both the socialist’s and industrialist’s vilification of idleness. If the purpose of nature was to yield ‘work’, the human body yielded ‘the work of the nerves, the muscles and the o ­ rgans, which were subject to the same laws of nature as any other ­machine’ (Rabinbach, 1992, p. 46). Hence the emerging metaphor of the ‘human ­motor’, which could be cultivated to a state of maximum productivity. Referring to the philosopher Michel Serres, Rabinbach makes a distinction between the eighteenth-century Newtonian — mechanical — machine and the ­nineteenth-century motor, a thermodynamic engine capable of converting fuel into heat and heat into mechanical work (Rabinbach, 1992, p. 52).

Chapter IV

220221_CritViz_Book_Full_revisions.indd 123

Critical Visualization

7/20/22 9:47 AM

124

Many aspects of the self-tracking movement link back to the nineteenth-­ century formalization of methods for capturing human and animal locomotion (Daston and Galison, 2010). The physiologist Étienne-Jules Marey’s photographic depiction of the body as a ‘theatre of motion’ played a significant part in the emergence of a visualization with a claim to objectivity. Marey developed a series of technical apparatuses in the late nineteenth century to delegate to machines the recording and observation of the functioning of the human body, including the myograph (for measuring the force produced by a contracting muscle), pneumograph (for measuring breath), and thermograph (for heat). Rabinbach describes the use of such instruments as a ‘kind of automatic writing [that] united the body’s own signs (pulse, heart rate, gait, flapping of wings) with a graphic means of technical representation’ (Rabinbach, 1992, p. 97).

g of the human body, including the myograph (for measuring the force produced by a contracting muscle), pneumograph (for measuring breath) and thermograph (for heat). Rabinbach describes the use of such instruments as a ‘kind of automatic writing [that] united the body’s own signs (pulse, heart rate, gait, flapping of wings) with a graphic means of technical representation’ (Rabinbach, 1992, p. 97).

Figure 4.05 — Geometric chronophotograph of the man in the black suit (1883). Étienne-Jules Marey

Measurement of motion, muscles, breath and heat were linked to a machinic and moral framework that conceived of a universal energy to be harnessed. Its counterpart, entropy, described the loss or diffusion of energy with the irreversibility of heat flow in the second law of thermodynamics. Entropy was characterized in the form of ‘fatigue’, which Rabinbach describes as a preoccupation that exemplifies nineteenth-century positivism: ‘attributing an objective basis to highly subjective states’ (Rabinbach, 1992, p. 44). This moral dialectic of energy and fatigue underlies the pioneering visualization work of the time. The Italian physiologist Angelo Mosso believed that fatigue was a ‘poison’ and produced hundreds of visualizations of ‘­fatigue curves’ in the late 1880s using the first ergograph, a metal glove that allowed the wearer only to move the middle finger, to which was attached a string a weight connected to a registering apparatus. Both Marey and his US-based counterpart Eadweard Muybridge were, independent of each other, pursuing the visualization of animal movement, in Muybridge’s case,

Data and the Self

220221_CritViz_Book_Full_revisions.indd 124

7/20/22 9:47 AM

famously to settle a bet and determine whether a galloping horse’s four hooves ever all left the ground at once. While Muybridge used cameras trigged by trip wires to demonstrate a horse’s gallop in freeze-frame, Marey had been using a ‘graphic method’ of attaching inscriptors to the horse’s hooves to capture their motion. Muybridge’s photographs were an instant success when shown in Paris, and, as Marey recognized, more impactful than his own inscription studies. However, Muybridge’s method was not capable of measuring the time of a horse’s movement, and this deficit led Marey to develop chronophotography: the automatic, timed photography of humans and animals in motion.

125

In 1882, Marey’s photographic gun and physiological station — two concentric circular tracks and wagon to carry a photographer and a three-sided camera obscura — systematized a method for capturing twelve images per second of a moving man or animal. The productivist ethos behind Marey’s experiments is evident in his 1886 description of the work in an address to the Association Française pour l’Avancement des Sciences: ‘As we regulate the use of machines in order to obtain a useful result with the least exertion of work, so man can regulate his movements ... with the least fatigue possible’ (Rabinbach, 1992, p. 116). In these early experiments in tracking motion, then, are assumptions that came to be enframed in the emerging technology: a productivist imperative and a glorification of speed. By 1883, Marey had begun to use a ­moveable film ‘ribbon’ in his camera, anticipating the arrival of c­ inematography. The resulting imagery had a significant cultural impact, inspiring the Modern­ ists’ efforts to represent dynamism in art, codifying the visual forms of motion capture, and establishing the basis of what the ­philosopher ­Henri Bergson later termed the ‘cinematographic illusion’ (Deleuze, 1986, p. 1). The ‘partial’ photography that Marey developed in 1883 to ­overcome the blurriness of photographed figures in motion (his human subjects wore a specially designed suit with silver stripes to highlight the parts ­relevant to the photographic study) achieved an abstracted mode of r­ epresenting motion that influenced the Futurist art of the subsequent century. ­Duchamp’s indebtedness to Marey for the inspiration behind his ­painting Nude ­D escending a Staircase is made evident in an 1967 interview where ­Duchamp describes Marey’s chronophotography of men fencing and ­horses galloping with ‘a system of dotted lines delineating the different movements’ (Rabinbach, 1992, p. 115).

Chapter IV

220221_CritViz_Book_Full_revisions.indd 125

Critical Visualization

7/20/22 9:47 AM

126

The sense in which Marey’s sequences of images both do and do not present an experience of movement has been the subject of considerable philoso­phical debate since. Bergson famously invokes Marey’s images in several essays of the late nineteenth and early twentieth centuries, in which he is working out the concept of duration (durée), as illustrated in the discrepancy between innate or experienced time (duration) and remembered and recorded time. According to Bergson, the representations of movement in Marey’s chronophotographs are illusory in that they are discontinuous fragments that our consciousness spatially fixes: ‘Instead of attaching ourselves to the inner becoming of things, we place ourselves outside them in order to recompose their becoming artificially. We take snapshots, as it were, of the passing reality’ (Bergson, 1911 cited in ­Rabinbach, 1992, p. 112).

Figure 4.06 — Chronophotograph: ‘Movements in Pole Vaulting’ (1885/1895). Étienne-Jules Marey

For the philosopher Gilles Deleuze, it is through the recomposition of ­images seen through the inhuman eye of the camera, that seeing is liberated from an organizing viewpoint, and the ‘movement-image’ becomes ­possible. In this sense, the chronophotographs present a diagram of the cinemato­graphic function, revealing its radical possibilities. Marey’s camera moves as the runner moves, and movement is no longer the synthesis of points in a single line of time, but ‘movement itself, in all its diversity, from which single points of view are composed’ (Colebrook, 2002, p. 33). If Marey’s visualizations seeded the radical possibilities of cinema, they also expressed and encoded the logic of capitalism. For the capitalist to buy a portion of the labourer’s time in order to produce surplus ­value,

Data and the Self

220221_CritViz_Book_Full_revisions.indd 126

7/20/22 9:47 AM

‘that time must be measurable and therefore divisible’ (Doane, 2002, p. 8). The emergence and internalization of divisible, mechanical time are tracked by historian E.P. Thompson through the early stages of industrial­­ ization: he records various conflicts over efforts to standardized time in eighteenth- and nineteenth-century Britain, and the long-standing agrarian and seafaring, task-based systems that were practised in response to the tides and passage of the sun. The day in pre-industrial societies was organized around tasks that needed to be completed within the available daylight, the tides or the ‘cattle clock’ (depending on the work). Pre-­industrial approaches to expressing duration reflected local customs and belief systems:

127

Thus in seventeenth-century Chile time was often measured in “credos”; an earthquake was described in 1647 as lasting for the period of two credos; while the cooking-time of an egg could be judged by an Ave Maria said aloud. In ­Burma in recent times monks rose at daybreak “when there is light enough to see the veins in the hand”. The Oxford English Dictionary gives us English ­examples — “pater noster wyle”, “miserere whyle” (1450), and (in the New ­English ­Dictionary but not the Oxford English Dictionary) “pissing while” — a somewhat arbitrary measurement. (Thompson, 1967, p. 58)

The impact of ‘time discipline’ described by Thompson was insidious and far-reaching, extending from bells and clocks to curfews, preachings and schoolings, the suppression of sports, and the division and supervision of labour. ‘New labour habits were formed and a new time-discipline was imposed’ (Thompson, 1967, p. 90). Pre-industrial labour mixed work and play somewhat indiscriminately, but industrial labour required punctuality; machines set the pace at which people worked, and time became a commodity that could be ‘saved’, ‘spent’ or ‘wasted’ (Adas, 1995, p. 242). The cultural specificity of time discipline became more pronounced when contrasted with the cultures of the unprepared subjects of the expanding European empires. According to historian Michael Adas, the European explorers and missionaries who went out to Christianize Africa and Asia found the pace of life exasperating. Nineteenth-century observers ­insisted that Africans had no sense of time beyond that of nature’s rhythms, marked by the cycle of the wet and dry seasons, the transitions from day to night, birth to death, or the waxing and waning of the moon (Adas, 1995, p. 244). Because the Africans they encountered had not developed ­writing, the Europeans assumed they had no history: Edmond Ferry claimed that

Chapter IV

220221_CritViz_Book_Full_revisions.indd 127

Critical Visualization

7/20/22 9:47 AM

128

the peoples of the Sudan had no verb forms to express the past tense (Adas, 1995, p. 244). Adas sums up the European attitude to the so-called primitive colonized cultures: ‘Textbooks used in colonial schools urged native school children to reject the slothful ways of their ancestors and embrace the work ethic of their conquerors’ (Adas, 1995, p. 258). Many nineteenth-century writers at the time equated the imperialist ­projects of Europe with the ‘triumph of science and reason over the forces of super­stition and ignorance which they perceived to be rampant in the non-­industrialised world’ (Adas, 1995, p. 204). The control of populations was precisely the challenge of the larger political project of colonization, and time discipline, categorization and enumeration were the empiricist’s tools for this. Edward Said’s seminal book, Orientalism (1978), describes the ­ideo­logies that supported the European colonial project of domination and extraction. Orientalism, argues Said, is also ‘anatomical and enumerative, to use its vocabulary is to engage in the particularizing and dividing of every­thing Oriental into manageable parts’ (Said, 1978, p. 72). If the ­Algerian colonial project combined feminization with architectural intervention in an attempt to achieve control of social units, the British ­project in India was characterized by the generation of a ‘vast ocean’ of data, count­ ing and measuring land, fields, crops, forests, castes, tribes and so on. Back in the UK, US and Europe, the work ethic of the conquerors began to reach industrial proportions. The Scientific Management movement, which drew from Marey’s chronophotography and its rationalization of the movements of the body, introduced the concept of industrial efficiency in the production lines of the twentieth century. Formal and m ­ ethodological similarities in the visualizations emerging from Marey and those who ­followed him two decades later are identified in Siegfried Giedion’s influential 1948 history of ‘anonymous’ design Mechanization Takes ­Command. The 1894 study of a blacksmith wielding a hammer at the anvil, for e­ xample, in a multiple exposure photograph subsequently schematized in a ­drawing by Charles Fremont in Marey’s laboratory, predates by twenty years the ‘cyclographs’ of Frank and Lillian Gilbreth, which sought to reduce the motions involved in a given task by attaching small electric lamps to s­ ubjects’ fingers and capturing their movements with long exposures (Gainty, 2016).

Data and the Self

220221_CritViz_Book_Full_revisions.indd 128

7/20/22 9:47 AM

129

Figure 4.07 — Motion efficiency study by Frank Gilbreth (1914). Frank and Lillian Gilbreth Collection, Archives Center, National Museum of American History, Smithsonian Institution

Where the Gilbreths used the camera to study human movement and ­eliminate inefficiency, the mechanical engineer Frederick Winslow T ­ aylor’s primary tool was the stopwatch. And where physiologists like Marey were concerned with how a worker’s movements might be optimized to reduce fatigue, Taylor’s interest was in maximizing productivity regardless of the physiological impact. Taylorism, the name given to distinguish the US-based Efficiency Movement from parallel trends in Europe, was formally established with the publication of Taylor’s 1911 Principles of ­Scientific ­Management. The book describes how ‘thousands of stop-watch ­observations’7 were made of labourers at the Bethlehem Steel Company in Pennsylvania, to determine how quickly a labourer could push a shovel into a pile of materials and draw it out fully loaded, with different shovel sizes and load weights tried in order to establish maximum efficiency. In ­another experiment, a bricklayer’s motions were reduced from eighteen to five, to increase speed and output. The book effectively handed management a means of controlling the entire production process. Scientific Management met with fierce resistance, much as the clock and mechanized time brought ‘far-reaching conflict’ in Thompson’s ­account:

7 https://www.marxists.org/reference/­subject/ economics/taylor/principles/ch02.htm (­accessed 29 May 2021).

Chapter IV

220221_CritViz_Book_Full_revisions.indd 129

Critical Visualization

7/20/22 9:47 AM

130

‘the historical record is not a simple one of neutral and inevitable technological change, but also one of exploitation and of resistance to exploitation’ (Thompson, 1967, p. 94). During Frank Gilbreth’s lifetime, efforts to ­rationalize workplaces through motion studies were thwarted by strikes and cancelled contracts (Price, 1992, p. 69). The physiologist Jean-­Maurice Lahy published a series of articles beginning in 1913, that were highly critical of Taylor’s studies, comparing them unfavourably with those of Marey, finding inadequate science in Taylor and a rudimentary method of chronometry, which in reducing measurement to time passed and the speed of work, subscribed to what Lahy called a ‘sort of scientific fetishism without positive value’. Taylor’s ‘ignorance’ of basic human physiology and his blinkered approach to obtaining maximum efficiency out of workers jeopardized workers’ health and safety, according to Lahy (Rabinbach, 1992, p. 250). Nevertheless, with the gradual softening of its more utopian strains (­dropping its extensive wage incentive formulas), and the absorption of the Gilbreths’ emphasis on the human factor in their motion studies, Taylor­ism gained considerable ground among automotive and electrical firms at first and gradually spread its influence. Taylor’s 1911 book was ­described as ‘the most influential management book of the twentieth ­century’ in 2001 (Bedeian and Wren, 2001). COM IC CR I T IQU E The charge from Morozov, then, that self-tracking represents a form of internalized Taylorism, suggests that the adoption of productivist values by self-trackers goes unchallenged, unconscious or involuntary. ­Certainly, this kind of uncritical instrumentalism would account for the bare-all approach of many Quantified Self-ers, and points back to the relentless involuntary rhythm of the assembly line. Charlie Chaplin’s film Modern Times (1936) sharply satirizes the involuntary effects of time-and-motion discipline quite soon after the emergence of Taylorism: Chaplin’s character is an assembly line worker who performs the same repetitive motion until the world is transformed into nuts to be turned by his wrench. The key features of the associated economic system are encountered by Chaplin’s hapless character in sequence: the worker’s bodily movements are reconstituted according to the rhythms of the machinery, he is separated from the product of his labours, an authoritarian structure (represented by the assembly line foreman) creates friction and competition between ­workers, and a surveillance system (represented by a screen representation of a factory executive on the bathroom wall) ensures worker time is monitored and (scientifically) managed to ensure maximum productivity.

Data and the Self

220221_CritViz_Book_Full_revisions.indd 130

7/20/22 9:47 AM

131

Figure 4.08 — Modern Times (1936). Modern Times, d.Charlie Chaplin (1936) © Charles Chaplin Productions.

To compare a more recent comic reflection on ‘Taylorism within’, David Sedaris’ account of adapting his life to the demands of a new Fitbit fitness tracker — which delivers a gratifying vibration to the wrist once a target number of steps has been taken by its wearer — relates his progression from an initial default Fitbit target setting of 10,000 steps a day to 60,000 steps. Sedaris describes a state of device-dependence reminiscent of Chaplin’s involuntary wrench-turning motion: I look back on the days I averaged only thirty thousand steps, and think, “­Honestly, how lazy can you get?” When I hit thirty-five thousand steps a day, ­Fitbit sent me an e-badge, and then one for forty thousand, and forty-five ­thousand. Now I’m up to sixty thousand, which is twenty-five-and-a-half miles. Walking that distance at the age of fifty-seven, with completely flat feet while lugging a heavy bag of garbage, takes close to nine hours — a big block of time but hardly wasted. I listen to audiobooks and podcasts. I talk to people. (­Sedaris, 2014)

When Sedaris’ first Fitbit stopped working he describes a ‘great sense of freedom’ that nevertheless made walking twenty-five miles suddenly seem pointless without the steps being counted and registered. ‘I lasted five hours before I ordered a replacement, express delivery’ (Sedaris, 2014). The enthusiasm with which self-trackers share their personal data would suggest that the internalization process has reached new heights in the age of the ‘datasexual’. Morozov argues that a ‘fundamental assumption’ of the Quantified Self movement, is that ‘the numbers can reveal a core and stable self — if only we get the technology right’ (Morozov, 2013, p. 232). The emphasis again on the technology, the ‘how’ of self-tracking, suggests a consumerist version of Chaplin’s assembly line, with data producers on

Chapter IV

220221_CritViz_Book_Full_revisions.indd 131

Critical Visualization

7/20/22 9:47 AM

132

the treadmill now voluntarily uploading the very commodity — personal data — that technology and social media corporations need in order to predict trends, focus their marketing and target sales of their products. A visualization project that draws critical attention to the exploitative ­aspect of self-tracking is, significantly, not a stand-alone piece, but part of an im­mersive theatre performance by the Berlin-based production company doublelucky productions, whose founders Chris Kondek and Christiane Kühl are interested in the ‘invisible infrastructures that define our current social, economic and private living environment’. Their performance The Hairs of Your Head are Numbered engages the audi­ ence in a seemingly open experiment in communal use of self-tracking technologies, where participants are each given a number and gloves equipped with biometric sensors that convey their heart rates to a live visualization on a large monitor, cheered on by performers playing the part of ted-talk-style cheerleaders. After one performer leads a relaxation ­cycle, another encourages the group to achieve a communal heart rate, and begins singling out individuals by number for not managing to get their heart rates in synch with the larger group; those individuals have their gloves removed. According to Kim Albrecht, the designer of the on-screen visualization for the performance, ‘your first impression is of techno-­ positivsm — that this is about a new way of understanding ourselves — and then more and more you get a hint that something is totally wrong’ (K. Albrecht, personal communication, October 10, 2018). By the end of the performance, says Albrecht, non-compliant self-trackers have been asked to leave the stage while the synchronized elite are brought to a champagne celebration party where they are advised that their data has been uploaded to the database of a large insurance company. Albrecht notes that to design a visualization that was critical was ‘utterly difficult’, but in The Hairs of Your Head are Numbered, the data was embodied, live and incorporated into a critical narrative. The audience ‘were seeing data screens, they had it on their bodies and saw visualizations: It’s all happening simultaneously in one space and performers put the story on top of it’ (K. Albrecht, personal communication, October 10, 2018).

Data and the Self

220221_CritViz_Book_Full_revisions.indd 132

7/20/22 9:47 AM

133

Figure 4.09 — T he Hairs of Your Head are Numbered. Kim Albrecht/doublelucky productions

Figure 4.10 — T he Hairs of Your Head are Numbered. Kim Albrecht/doublelucky productions

In their spirited defence of self-tracking, researchers Sharon and ­ andbergen argue that their own ethnographic work among Quantified Z Self-ers revealed a much less docile, compliant and naïve individual than Morozov and other critics have portrayed in their attack on ‘data fetishists’ enamoured with the authority of numbers. They outline three other forms of ‘meaning-making’ that characterize what they see as a heterogeneous network of people actively exploring different effects and affects of tracking practices: self-tracking as a practice of mindfulness; as a means of resistance against and a remaking of dominant social norms and conventions; and as a narrative and communicative practice that can articulate experiences at the boundaries of different domains of knowledge. (Sharon and Zandbergen, 2017)

Chapter IV

220221_CritViz_Book_Full_revisions.indd 133

Critical Visualization

7/20/22 9:47 AM

134

The resistance and remaking of dominant social norms are exemplified by Sharon and Zandbergen in the form of discussions hosted by QS about data privacy, how data is obtained and how autonomy is maintained, and a ‘soft resistance’ to the biopolitics of the technology industry. In the tradition of Silicon Valley, this resistance takes the form of hacking or tweaking the given tracking technologies to challenge and remodel ‘the assumptions, norms and categories that are built into tracking devices’ (Sharon and Zandbergen, 2017, p. 1703). An ability to challenge norms would seem to distinguish data fetishism from a critical project of self-tracking, but Sharon and Zandbergen’s individualized examples can be equally characterized as outcomes of a neoliberal era, in which ‘macro-level, reform-based solutions to problems are discarded in favour of carefully delineated action by atomized individuals’ (Morozov, 2013, p. 237). An example is one of Sharon and ­Zandbergen’s QS study participants, who describes tracking her own weight as ‘­liberating’ compared with a doctor putting her on a scale. The sense of patient empower­ment and control over her health is an important aspect of contemporary discussions in healthcare, but so, too, is the infiltration of neoliberalism in market-driven healthcare systems which emphasize inefficiencies in state-owned healthcare and place the burden on individuals to reduce costs through personal preventive care (e.g. Horton, 2007). The upbeat language of self-help literature (and ted talks) that urges citizens to take ownership of their health and their bodies is mimicked in the texts in The Hairs of Your Head, even appropriating the QS tagline ‘self knowledge through numbers’ (Kondek and Kühl, 2018). We do not just want to be, we want to be good and get better, on all ­levels. That’s why you too are here tonight: Because the theatre of society holds up a ­mirror and promises to bring us closer to the true, the beautiful, the good in the ­cathartic moment of knowledge. To make better people out of us. The ­theatre and ­Silicon Valley are so close: make this world a better place. (Kondek and Kühl, 2018)

Data and the Self

220221_CritViz_Book_Full_revisions.indd 134

7/20/22 9:47 AM

135

W H AT IS NOR M A L? We shall return later to Sharon and Zandbergen’s other two ‘meaning-­ making’ practices of mindfulness and ‘articulating experience at the boundaries of different knowledge domains’, but the notion of ­challenging ‘norms’ through data collection and visualization deserves further explica­ tion. This can be helped again by unpacking a historical trajectory of practices that arrives at mid-nineteenth-century data visualizations, in this case, connected to a concern with understanding deviancy. Ian Hacking has written of the ‘avalanche of printed numbers’ that accom­ panied the arrival of statistics (aligned with Europe’s second Industrial Revolution). As noted previously, data are always collected for a purpose, and in the first half of the nineteenth century, states became interested in accumulating and analysing data on populations and birth rates, as well as perceived hindrances to the strength of a state, that is degeneracy: mortality, disease, crime, prostitution and suicide. The collection of data on suicide came, according to Hacking, from the French ‘fascination with deviants, especially those who were degenerate or could not contribute to the growth of the French population’ (Hacking, 1990, p. 64). A highly significant move in the contemporary use of statistics was initi­ ated at this time by the Belgian astronomer Adolphe Quetelet, who transferred the Gaussian distribution, or curve of errors, from astronomical measurement to human and social measurement. First described in the early 1800s, the Gaussian curve was used to correct variation in the observations of an astronomical body on different days by different people, establishing a central mean with which to determine the position of the astronomical body. Quetelet argued that the same curve was to be found in data recording the measurements of soldiers in a Scottish regiment and concluded that the same statistical law was at work — that, just as the range of measurements of an astronomical body gave us an accurate position, so too the measurements of a range of men would give us an accurate measure of the ‘average man’ or homme type. Whereas previously, people could be grouped by culture, geography, language or religion, the average man allowed Quetelet to characterize the measurements of physical and moral qualities in a race. Hacking concludes: This is half of the beginning of eugenics, the other half being the reflection that one can introduce social policies that will either preserve or alter the average qualities of a race. In short, the average man led to both a new kind of information about populations and a new conception of how to control them. ­(Hacking, 1990, pp. 107–108)

Chapter IV

220221_CritViz_Book_Full_revisions.indd 135

Critical Visualization

7/20/22 9:47 AM

136

Quetelet’s notion of the average man, as Doane has noted, reduces a host of individualities to a single figure, ‘a quantity/number that exists ­nowhere but that performs work, resulting in a vast reconceptualization of ­reality’ (ultimately as a predictable, manageable risk) (Doane, 2002, p. 127). For example, to advance his theory of eugenics, the English statistician Francis Galton developed a method of making composite portraits by overlaying on one photographic plate several exposures of different faces. In this way, Galton hypothesized, ‘the true physiognomy of a race’ could be determined by isolating its common features while cancelling out individual peculiarities. Galton brought to this visualization project many of the prevalent concerns and prejudices of Victorian politics: that cultivating the desirable features of the English ‘race’ would counter the logic of e­ ntropy, which Galton cavalierly applied from thermodynamics to a perceived ­tendency toward social decline and degeneracy (Doane, 2002, p. 128). He also used composite portraiture to develop his theories of a shared ­genetic physiognomy among lunatics, Westminster schoolboys, Jews, t­ uberculosis patients and criminals. In a lecture to the ­A nthropology ­Section at the British Association in 1877, Galton showed images of ‘the ­ideal criminal’ to illustrate his eugenic hypothesis: The ideal criminal has three peculiarities of character; his conscience is almost deficient, his instincts are vicious, and his power of self-­control is very weak ... It becomes an interesting question to know how far these ­peculiarities may be correlated with physical characteristics. (Green, 1985, p. 11)

Galton’s source material was a series of photographic portraits, provided by the Home Office, of inmates at Pentonville and Millbank prisons, classified into three groups of convictions: murder and manslaughter, felony, and sexual offenses (Green, 1985, p. 11). The hypothesis that physical characteristics may correlate with criminal propensities had an immediate practical application in the first anthropometric system, Bertillonage, developed by French police officer and researcher Alphonse Bertillon, who began photographing, measuring and classifying physical features of convicted criminals. In seeking to identify repeat offenders, Bertillon developed a system of eleven measurements that could, he believed, overcome the professional criminal’s expertise at developing disguises, alibis and false identities (Sekula, 1986, p. 27). B ­ ertillon thought that a person could be identified by the ear, for example, and it is

Data and the Self

220221_CritViz_Book_Full_revisions.indd 136

7/20/22 9:47 AM

thanks to him that visitors to the US are required to be photographed with their right ear visible (Hacking, 1990, p. 187). B ­ ertillon’s studio set-up, with standard focal length, consistent lighting, plain background, and expression and position of the subject in the portraits established a visual form that persists today, for example, in police ‘mugshots’ and passport and immigration photographs.

137

Figure 4.11 — Composite portraiture: A product of eugenics theory premised on nineteenth-century notions of ‘average’ and ‘normal’. Francis Galton

Chapter IV

220221_CritViz_Book_Full_revisions.indd 137

Critical Visualization

7/20/22 9:47 AM

138

Figure 4.12 — T he Bertillon system (1879). Alphonse Bertillon

Bertillon translated the visual signifiers into a text-based system, and combined this in criminal records with numerical data from the measure­ ments, following Quetelet’s principle of the ‘average man’, into subdivisions based on below-average, average and above-average figures for the eleven measurements (height, chest, arm span, head size, foot size, etc). Using this method with 120,000 criminal records documented between 1883–1893, Bertillon was able to ‘infallibly’ identify 4,564 repeat offenders (Sekula, 1986, p. 29). Shortly after Bertillon developed this apparatus and method for identifying criminals, the Italian criminologist and physician Cesare Lombroso began theorizing that criminality was inherited and could be identified by physical ‘defects’. Lombroso even imagined that he had discovered anato­ mical similarities in criminals recalling an apish past, reflecting the most pernicious and influential of the adaptations of Darwin’s theory of evolu­ tion: recapitulation. The hypothesis that an individual ‘climbs his own family tree’ (Gould, 1980/2008, p. 114) while developing as an embryo, and goes through stages representing successive stages in the evolution of its ancestors, was summarized by Ernst Haeckel as ‘ontogeny recapitulates phylogeny’ and was visualized in Haeckel’s drawings. The idea was ­elaborated into a theory of biological determinism and race and gender hierarchies — which typically placed northern European whites at the

Data and the Self

220221_CritViz_Book_Full_revisions.indd 138

7/20/22 9:47 AM

top of the tree, which fed, via Haeckel, directly into Nazi policies (Gould, 1980/2008, p. 115). Lombroso wrote treatises on the criminality of animals and those whom he nefariously claimed were inferior people.

139

The roots of modern-day anthropometry and biometry are thus entangled with racism and eugenics — the idea that humans could be ‘designed’ to breed out ugliness, genetic deficiency and disease. The role of visualization was to provide evidence, in Galton’s composite portraiture, Bertillon’s photographic apparatus, Lombroso’s curated photographs, and Haeckel’s drawings of embryos, which overemphasized similarities between related species. The perception of the camera as an objective instrument in this project parallels the use of a statistical construct, the average man, in transforming properties into probabilities. Sekula notes how, contrary to Bertillon’s aspirations, Bertillonage actually eroded the uniqueness of the individual in its dependence on deviations from the norm. As Bertillon himself noted, the logic of the curve of errors, was such that ‘each observation or each group of observations is to be defined, not by its absolute value, but by its deviation from the arithmetic mean’ (Sekula, 1986, p. 34). BIOM E T R IC S A N D R ISK-PROF I LI NG A clear trajectory can then be drawn from Bertillon’s anthropometric classifications and Galton and Lombroso’s biological determinism to present-day risk profiling. By the early twentieth century, Bertillonage was being superseded by fingerprinting (invented by Galton), which took less skill and time for clerks to gather than the eleven-measurement system. Fingerprints have not simply persisted as a means of identification, they have become part of an entire dataveillance programme, introduced after the terror attacks of 11 September 2001. Geographer Louise Amoore tracks the $10 billion US restructuring of air, land and seaport security based on the concept of the virtual or biometric border, which reaches beyond physical borders, governing mobility, delegating authority to private security firms and citizens, and  — most relevant for the link to Galton — using biometrics to link the body to predictive policing, to prevent would-be terrorists from even reaching US borders. Amoore counts over twenty databases in the US ‘smart border solution’, including ident, which stores fingerprints; adis, which stores entry and exit data for travellers; apis, which contains passenger manifests; sevis, which contains data on foreign and exchange students to the US; and claims3, which has information on foreign n ­ ationals claiming benefits. These databases combine to ‘pre-emptively’ fight the ‘war on terror’ through predicting future acts: The risk-based identity of the person who attempts to cross an international border is in this way encoded and fixed far in advance of reaching the physical border — when, for example, he leaves the electronic traces of buying an air ­ticket, applying for a visa, using a credit card, and so on. (Amoore, 2006, p. 340)

Chapter IV

220221_CritViz_Book_Full_revisions.indd 139

Critical Visualization

7/20/22 9:47 AM

140

The actions of the self-tracker would seem to play directly into the hands of this dataveillance system, supporting Foucault’s description of biopower as a ‘far more subtle’ alternative to the disciplinary power being exerted on individuals or populations, since it is focused on the ‘promotion of self-­ regulation and self-management’. Or as Zygmunt Bauman puts it: ‘just as snails carry their homes, so the employees of the brave new liquid modern world must grow and carry their personal panopticons on their own bodies’ (Bauman and Lyon, 2013, p. 59). It is possible, nevertheless, to push self-­ tracking to an extreme and disrupt the surveillance systems. American Bangladeshi artist Hasan Elahi began to use this tactic after being apprehended at Detroit airport on his return from a visit to the Netherlands. Falsely suspected of hoarding explosives by fbi agents who had received a ‘tip’, Elahi, who is an associate professor at the U ­ niversity of Maryland, was subjected to six months of interrogations before his name was cleared. In response to this experience, Elahi created a website tracking his daily movements, sometimes a hundred per day. The data visualizations provide an interface to this material and recall the dazzlingly complex network maps of the internet that characterized early net art. Their use of data overload is subversive, however. Thousand Little Brothers (2014), a composite image of 32,000 images of daily life, provides a mesmerizing account in myopic detail: literally too much information. ‘By putting everything out there, the government’s data no longer means very much,’ Elahi says. ‘I can tell them what I ate, where I slept, exactly what I was doing on every day over the last decade’ (Urist, 2015).

Figure 4.14 — Detail from Thousand Little Brothers (2014). Hasan Elahi

Fig. 4.13 — T housand Little Brothers (2014). Hasan Elahi

The composite portrait plays a key role in the philosophy of data visuali­ zation because it demonstrates how evidence is turned into data: a photograph of one person is evidence, but when four photographs are overlaid, we are able to abstract what is common. This demonstrates how general ideas are formed. In his Critique of Judgement, philosopher I­ mmanuel Kant argued (against empiricists) that to abstract commonalities we

Data and the Self

220221_CritViz_Book_Full_revisions.indd 140

7/20/22 9:47 AM

must first have a concept (or conceptual framework) that we bring to the evidence (Conant, 2005). In the case of Galton’s theory that p ­ hysiognomy could predict criminality, the overlaid images did not in fact provide him with the telltale features of a criminal: instead more images produced more idealized composites: ‘everything starts to look like a Romantic portrait of Lord Byron’, according to philosopher James Conant (2005). Elahi’s 32,000 images are not overlaid, but presented in a grid whose sheer volume and organization into dominant tones, create a portrait as a smokescreen, or a thousand little alibis.

141

Elahi’s attempt at visualizing uncertainty or blurriness through sheer volume echoes a seminal experiment with photography by the philosopher Ludwig Wittgenstein in the 1920s. Based on his philosophical interest in Galton’s composite portraiture, Wittgenstein created a composite of an image of himself and his three sisters, but rather than manipulating photo­graphs to make features align, as Galton did, Wittgenstein simply had each person sit in an identical position in front of the camera. The contrast between the two men’s composite portraits is that Wittgenstein’s appears descriptive, whereas Galton’s appears prescriptive. Wittgenstein’s aim, ­according to the keeper of his archives, was in part to show that ­Galton’s claim to authority was based on ‘a probability’ — ‘one image frozen o ­ utside of time’. Wittgenstein’s composite image, on the other hand, shows ‘all manner of possibilities’ (University of Cambridge, 2011). Wittgenstein’s composite image effectively co-opts Galton’s method to demonstrate a more open-ended explanation of why people look alike than Galton’s essentialist claim to genetic determinations of superiority or inferi­ ority. The photograph of the Wittgensteins illustrates the philosopher’s important later concept of ‘family resemblances’, whereby no single feature is common to all the Wittgensteins pictured, but as a group, they are connected by a series of overlapping similarities — analogous to the way a rope is composed of strands, none of which run the entire length of the rope. The critical trajectories of composite and physiognomic photo-portraits, then, might be summarized as follows: the biomechanical study of Marey with its productivist connotations (e.g. Gilbreth); the identification system of Bertillon with its implications for biometrics, racial profiling and predictive policing; the eugenic hypotheses of Galton and Lombroso with their racist connotations; and the anti-essentialist composite of Wittgenstein with its suggestion of alternative conceptual systems.

Chapter IV

220221_CritViz_Book_Full_revisions.indd 141

Critical Visualization

7/20/22 9:47 AM

142

In an era when manipulation of photographs is available to anyone with a computer, composite portraiture has maintained currency as a means of visualizing the dynamic possibilities of a portrait and of disrupting categorization. In a league with the Quantified Self-er is the daily selfie practitioner — such as twenty-two-year-old design student Eadington, whose eight years of selfies in a 3.5 minute time-lapse provide an index of ageing, health or mood: eight years into her project, Eadington told the Press ­Association that she hoped it would ‘continue to motivate me to improve myself’ (Irish News, 2018). Potentially more revealing are the composite photography works of Jason Salavon, which deploy code to amalgamate multiple commonplace images — of real estate photographs of homes for sale, pornographic images or high school graduation photos, for example — in blurry composites that aim to reveal the trends and formulas at work in the framing, styling, lighting and presenting of these ‘rites of passage’. Other projects to confront selfie culture include Lancel and Maat’s installation Saving Face (Kitchin and Perng, 2016) and Tega Brain and Surya Mattu’s ‘Unfit bits’ efforts to fool fitness tracking tools ‘Unfit Bits’ and lower insurance premiums (­ ­unfitbits.com). CH A L L E NGI NG NOR M S In her account of the biometric border, Amoore places considerable emphasis on the ‘satirical and playful practices of the arts’ to disrupt the governing of mobility. Playful projects like Alex Rothera and James ­K rahe’s installation Playful Self: The Anxiety of Data, Calmed with Tea (2015) poke at the typical performance-based, sports domains occupied by self-­trackers and technology makers, instead presenting a record player and tea-making set that respond to the user’s heart rate (the teabag and sugar pulsate and the record plays Sinatra’s ‘Under my Skin’ at a matching speed). Amoore cites the artist Heath Bunting, whose Status Project visually maps the relationships between databases to reveal the o ­ ntologies that link legal, political and social statuses: according to one data map, ‘a blood donor cannot be hiv positive, an injector of drugs, taking antibiotics, a prostitute, gay or less than one year from having a piercing’ (see p. 180 Focus: Heath Bunting: Status Project).

Data and the Self

220221_CritViz_Book_Full_revisions.indd 142

7/20/22 9:47 AM

143

Figure 4.15 — Playful Self (2015)(2015): Objects, when touched, reflect visitors’ personal body data such as heart rate. Alex Rothera and James Krahe

Figure 4.16 — Playful Self (2015): An interactive exhibition piece exploring biometric body-data transfer. Alex Rothera and James Krahe

Other satirical and playful artistic approaches to biometric surveillance include work by the Mexican Canadian artist Rafael Lozano-Hemmer, whose projects appropriate technologies elsewhere championed as omni­ potent biotechnical tools of security and surveillance. Lozano-Hemmer’s 2014 installation Zoom Pavilion, made in collaboration with Krzysztof ­Wodiczko, deploys twelve surveillance systems and face recognition algorithms to detect participants and their relationships within the ­exhibition space. Using the familiar authoritative language of ­surveillance technology, walls of the exhibition feature continuously changing projections from the robotic cameras as they zoom in to individuals’ faces,

Chapter IV

220221_CritViz_Book_Full_revisions.indd 143

Critical Visualization

7/20/22 9:47 AM

144

zoom out to a landscape of participants and overlay detection rectangles and lines that connect individuals across the room. The addition of terms that appear to classify the relations between apparent strangers, such as ‘distant/­attraction/interest’ suggests that the machine algorithms are able to s­ peculate or detect c­ onnections between people — even where there might be none. In the context of an art fair or exhibition, the visualizations ­naturally prompt amusement, but Lozano-Hemmer explains in an interview that Zoom Pavilion is at the same time a ‘terrifying’ piece, ‘because it’s the ­surveillance of relationship’ (Lozano-Hemmer et al., 2015, p. 219).

Figure 4.17 — Zoom Pavilion (2014). Rafael Lozano-Hemmer in collaboration with Krzysztof Wodiczko; photo by Mariana Yañezr

Lozano-Hemmer’s work in this sense exposes the limits and biases — and power — inherent in biometric systems. In Level of Confidence (2015), which commemorates the forced disappearance of forty-three student teachers from Ayotzinapa, Mexico, a facial recognition system is programmed to search for the students, and when confronted by a visitor to the exhibition, looks for shared physical traits between the students’ photographs and the face of the visitor. Its repeated failure to find a match becomes a haunting memorial that also implicates the viewer in the process, challenging the notion of an innocent bystander.

Data and the Self

220221_CritViz_Book_Full_revisions.indd 144

7/20/22 9:47 AM

145

Figure 4.18 — L evel of Confidence (2015). Rafael Lozano-Hemmer; photo by Guy L’Heureux

The possibility of machine error and algorithmic bias in biometric surveil­ lance systems is explored in a study by Shoshana Magnet (2011), which finds that manual labourers often cannot be successfully fingerprinted because their fingers are worn, and that eye recognition readers often fail on people in wheelchairs or people who are too tall, because the iris recognition only works within a range of angles. Simone Browne’s extensive study (2015) of the intersection of race and surveillance elaborates on how ‘norms’ are racialized, as in the example of how street behaviours of white men standing around are coded as normal and the same behaviours of Black men are not. Browne contends that within the field of surveillance studies race remains undertheorized. ‘Blackness’ argues Browne, is ever present in the subjection of black motorists to a disproportionate number of traffic stops (driving while black), stop-and-frisk policing practices that subject black and Latino pedestrians in New York City and other urban spaces to just that, cctv and urban renewal projects that displace those living in black city spaces, and mass incarceration in the United States where, for example, black men between the ages of twenty and twenty-four are imprisoned at a rate seven times higher than white men of that age group, and the various exclusions and other matters where blackness meets surveillance and then reveals the ongoing racisms of unfinished emancipation. (Browne, 2015, p. 13)

The implications of statistical norms and their negative impact on peoples’ lives are felt across the design disciplines, and opportunities to resist them are not limited to the ‘satirical and playful practices’ of the arts. A visualization project that performs the function of challenging and reinventing the assumptions of early anthropometry is SizeChina, a functional database of head shapes in China. Initiated by product designer Roger Ball while working as a researcher at Hong Kong Polytechnic University, the 3D anthropometric database is the result of fieldwork in six provinces of mainland China to scan the heads of 2,000 Chinese civilians. In the

Chapter IV

220221_CritViz_Book_Full_revisions.indd 145

Critical Visualization

7/20/22 9:47 AM

146

project documentation, extensive photography illustrates the situatedness of the data-gathering and the process by which the 3D scans were cleaned up and amalgamated to create a set of anthropometric measurements for designers of eyewear and headgear for Asian markets. The project was initiated by Ball — and funded by the Chinese government — after it became apparent that the ‘standard’ data set for head shapes was not universal as its status suggested, but based on measurements of 1940s US military personnel: Designers at the US-based Henry Dreyfuss Associates had been tasked with designing equipment for the cockpits of fighter planes, and hired Alvin Tilley in 1946, to draw the famous anthropomorphic charts of Joe and Josephine, ‘typical Americans’ (Flinchum, 1997, p. 87). Tilley’s sources were not exactly clear, but the charts resulted in the publication in 1960 of the book The Measure of Man, republished and updated as The Measure of Man and Woman — and still used as a reference text by designers today. Ball discovered that both eyeglasses and headgear were ill-suited for Asian head shapes, because of the cultural specificity of the socalled ‘universal’ data, after a group of Japanese s­ nowboarders e­ xplained to him that the helmets he designed for the US company B ­ urton didn’t sell in Japan, because they gave a headache to the wearers, who typically had rounder-shaped heads than US military personnel.

Figure 4.19 — Scans from SizeChina. Roger Ball

Data and the Self

220221_CritViz_Book_Full_revisions.indd 146

7/20/22 9:47 AM

A comparably flawed claim to universality is to be found in the algorithms behind facial recognition technologies, whose failure to recognize race and gender can be traced to the prejudices written into the biometric algorithms (Magnet, 2011, pp. 34–43). Among the repeated failures are miscategorization of sex based on hair length, make-up, expression of emotion (women smile more than men) and clothing. In one experiment, a computer was taught to assign race based on ‘component’ analysis and two-dimensional vectorized models of photographs, then correlated with a database of manually labelled photographs, which produced a 17.5 per cent failure rate. Magnet finds the racist roots of anthropometry highly problematic, and notes that among those misrecognized by biometric techno­logies are mixed-race individuals:

147

As a result of biometric science, entire communities become what Leigh Star and Geoff Bowker called “boundary objects” Donna Haraway has called “monsters” and Kal Alston has called “unicorns” — uncategorizable objects that do not fit into any classification system. (Magnet, 2011, p. 44)

T H E E X A M I N E D LI F E Earlier in this chapter, it was suggested, by defenders of the Quantified Self movement, that a more heterogenous ethnography of self-trackers reveals ‘meaning-making’ through mindfulness and in the communicative and narrative practice of articulating experiences at the boundaries of different domains of knowledge. Sharon and Zandbergen seem to suggest that in the process of quantifying their lives, practitioners achieve a level of self-­ reflection that is also expressed in conversations and presentations among the community — notably at QS gatherings. Indeed, they find, amid the confessional aspect of the QS presentations of data, an emerging concept of the ‘qualified self’ — a term used to describe the processes by which quantified data are interpreted and transformed into qualitative narratives that create meaning (Sharon and Zandbergen, 2017, p. 1705). In closing, we will explore this claim that to quantify oneself contributes to ‘knowing thyself’, to cite the Socratic aphorism. Much of this chapter has focused on the way in which the ‘self’ as we know it, is not a raw or pure underlying essence that can be revealed through a lens of numbers, but is constituted by the aims, ambitions, decisions and agendas embedded in the technologies we use and have inherited. Lupton usefully identifies three bodies that constitute the quantified self: the body as machine, the private body (‘Not overcoming difficulties becomes firmly positioned as the fault of the private individual rather than of their relative social and economic advantage’ (Lupton, 2016, p. 51).) and the spectacularized body.

Chapter IV

220221_CritViz_Book_Full_revisions.indd 147

Critical Visualization

7/20/22 9:47 AM

148

This transition is loosely chronological, from Rabinbach’s machine or ­motor body, to the body that is a part of a digital information system, ­subject to communication errors causing illness and disease (Haraway, 1991), to a spectacular body, ‘in which the inner workings are displayed and made visible’ (Lupton, 2016, p. 54). The spectacular body, which Lupton’s genealogy suggests is a development of the private, or neoliberal body, attracts a rhetoric of religious proportions from Quantified Selfers, suggesting the singularity is nigh: it is a digital, virtual body that in its state of data transparency somehow transcends our fleshy selves: ‘as we get more and more connected, our feeling of being tied into one body will also fade, as we become data creatures, bodiless, angelized’ (Blumtritt, 2014, as cited in Lupton, 2006, p. 79). Other self-trackers voice a more circumspect account of the benefits of the always-­on, shared, transparent self. The digital model of himself that Kaiton Williams constructed as part of his PhD research at Cornell ­University prompted something of an identity crisis: ‘We (The Apps and I) had co-constructed a digital model of myself, and here I was, ­managing myself, it seems, by proxy. The feedback from that digital model often took ­precedence over how I physically felt’ (Williams, 2014).

Figure 4.20 — Dual-energy X-ray absorptiometry (dx a) scan from ‘An Anxious ­A lliance’. Kaiton Williams

Data and the Self

220221_CritViz_Book_Full_revisions.indd 148

7/20/22 9:47 AM

Indeed, a Socratic critique would surely argue that to quantify one’s life is not the same as examining it. The examined life is a process of questioning our most cherished beliefs through dialogue; the unexamined life, Socrates famously argued at his trial, is ‘not worth living’. To leave the examination to machines and the wisdom of the (QS) crowd would surely detach us from this process. In a bbc radio programme devoted to the question ‘Can Life be Measured?’ the claims made by Wolfe and QS advocates are contextual­ ized with an array of literary wisdom and expert knowledge. Psychoanalyst and literary theorist Josh Cohen argues that quantifying the self is the ‘sub­ con­tracting of the monitoring process’. To the argument that life-tracking, auto-analytics and sharing the resulting higher levels of description can help us navigate the complexities of existence and support the attainment of health, happiness and self-knowledge, Cohen responds that its idealism neglects the broader spectrum of human experience: ‘How can a self which is fully quantified … allow for those currents that don’t make themselves available to quantification? What happens to the full range of the inner life? It’s terribly unfashionable to speak up for unhappiness’ (bbc Radio 4, 2013).

149

Lupton finds in Quantified Self and self-help literature the prevailing ­notion of individuals as ‘atomized actors who are expected and encouraged to work upon themselves in the quest to achieve health, productivity and happiness’. With this holy trio — health, productivity and happiness — ­Lupton suc­ cinctly identifies what have become the individualized values of t­ wentyfirst century neoliberalism. The interest in personal health, diet and ­exercise espoused by self-trackers, features, commonly, a ­prevailing ­metaphor of the body-as-machine. Outside of the QS techno-utopian hothouse rhetoric, however, there are life-tracking practitioners whose outlook does suggest a more reflective approach. One of Sharon and Zandbergen’s examples of self-tracking as a mindfulness practice is the artist Alberto Frigo, whose 36-year, ongoing ‘life-stowing’ experiment records ‘details of his dreams, the songs he listens to, the external surroundings in which he moves each day, people he meets, new ideas, cloud shapes and the daily weather’ (Lupton, 2016, p. 12). The seemingly arbitrary and idiosyncratic methods and classifications ­invented by Frigo sit strangely with the obsessive nature of the project and the sheer volume of the resulting data. Frigo’s system seems to defy the notion of transferable knowledge, yet the personal insights are apparent. Sharon and Zandbergen argue that for Frigo, ‘the act of logging the data becomes more meaningful — and therapeutic — than the actual data-as-memorabilia

Chapter IV

220221_CritViz_Book_Full_revisions.indd 149

Critical Visualization

7/20/22 9:47 AM

150

that is its content’ (Sharon and Zandbergen, 2017, p. 1701). In an interview, Frigo describes a process that is analogue and ritualistic. ‘I believe I have become more stoic, the love that I put into the curation of my lifelog makes me a happier and more helpful person to others’ (Frigo, 2015).

Figure 4.21 — Cube installation documenting a 36-year ‘Life Stowing’ project. Each square corresponds to one of 36 ‘life works,’ made up of 432 hand-welded pixels that can be scanned by visitors. Alberto Frigo

In the serial documentation of one’s daily activity, it would seem, is something life affirming, that goes beyond quantification and measurement. This might be better understood in terms of the post-war art, film and literature movements that took as their subject the everyday, such as the French Oulipo group, whose stripped-back literary experiments used as their subject matter the ‘most obvious, most common, most colourless’ (Perec, 1974/1997, p. 50), or the artist On Kawara, who for eleven years sent friends a daily postcard recording the time he awoke, his geographical ­location and details of the people he had met, places he had visited and books he had read.

Data and the Self

220221_CritViz_Book_Full_revisions.indd 150

7/20/22 9:47 AM

151

Chapter IV

220221_CritViz_Book_Full_revisions.indd 151

Figure 4.22 — Signs of Life (1994). Alfredo Jaar. Courtesy of the artist, New York

Critical Visualization

7/20/22 9:47 AM

152

The ritualistic aspects of self-tracking begin to suggest a very different notion of self-hood when situated in close proximity to death and trauma. The artist Alfredo Jaar took inspiration from Kawara to make his Signs of Life postcard project after visiting Rwanda following the 100-day genocide in 1994, which led to the deaths of one million Rwandans. In August of that year, Jaar travelled to Rwanda to bear witness to the aftermath of the genocide, which has been called the ‘failure of humanity’ (Dallaire and Beardsley, 2004) — the failure of the international community to respond, its absence in Western media, and the result of systemic ­racism. Jaar’s Rwanda Project, created between 1994 and 2000, comprises twenty-­ five works including Signs of Life, the postcards Jaar found in a destroyed post office in Kigali, and mailed, each with the name of a person he had encountered during his visit and the handwritten message ‘___ is still alive!’ (Duncombe and ­Lambert, 2018). The project, described as both ‘a dirge and an exaltation’ (Gervat, 2019, para. 6), was followed by ­S earching for Africa in life, a visualization made up of all 2,128 covers of the US photo­graphic news magazine life, published between 1936 and 1996 (see ­Figure 1.13). Notably absent from the covers is any evidence of coverage of ­human life and culture on the African continent, revealing a blind spot in the magazine’s supposed world view. The significance of the everyday in the arts, politics and philosophy has been extensively discussed, but its relevance to data visualization sheds a useful light on the Quantified Self movement. If self-tracking at one level suggests a desperately techno-utopian bid to enact an individualistic productivism that plays into the hands of a biometric surveillance society, at another level it holds the potential to re-enact the selfhood that refuses to be commodified, surveilled and homogenized. Amoore’s investment in ‘satirical and playful’ practices in the arts has a grounding in the work of Henri Lefebvre, among other theorists to have focused on the politics of the everyday, particularly in the context of the post-war years. In his Critique of Everyday Life, Lefebvre argues that art, like play, is ‘transfunctional — it is both useless and has many uses, acting as a kind of “play-generating yeast” in the everyday, fermenting, ­agitating and disrupting’ (Lefebvre, 1961/2008, p. 14). A visualization practice that is able to ferment, agitate and disrupt is quite a different p ­ roposition from the positivist mantra of the QS life-tracker, but perhaps closer to what Sharon and Zandbergen were leaning towards in their case for meaning-­making through narrative practices of articulating experiences at the boundaries of knowledge domains. In its distrust of the heroic and specta­cular, Lefebvre’s account of the everyday confronts ‘the bureaucracy of controlled consumption’ (Harris and Berke, 1997, p. 3). There is no better rejoinder to the charge that self-tracking is only a practice of self-­ commodification and self-­surveillance.  █

Data and the Self

220221_CritViz_Book_Full_revisions.indd 152

7/20/22 9:47 AM

220221_CritViz_Book_Full_revisions.indd 153

7/20/22 9:47 AM

154

Margaret Pearce and Michael ­Hermann — ‘They Would Not Take Me There: People, Places, and Stories  from Champlain’s Travels in Canada, 1603–1616’

Focus Figure 3.1 — Margaret Pearce and Michael Hermann, ‘They Would Not Take Me There: People, Places, and Stories from Champlain’s Travels in Canada, 1603–1616’. Margaret Pearce

Designed by cartographers Margaret Pearce and Michael Hermann, ‘They Would Not Take Me There: People, Places, and Stories from Champlain’s Travels in Canada, 1603–1616’ is a narrative map that complicates and adds nuance to one of the foundation stories of New France. Commissioned for the 400th anniversary of the founding of Quebec by Samuel de Champlain at the Canadian American Centre at the University of Maine, the two-sided 40 × 60 inch bilingual map, tells of the colonist’s journey and encounters with Indigenous peoples over a ten-year period. The map uses text from Champlain’s own journal as a source for creating a narrative that spans time and territory. Hermann explains that they ‘were interested in the idea of mapping narrative using

Rethinking the Representation of Data 220221_CritViz_Book_Full_revisions.indd 154

7/20/22 9:47 AM

Focus the written record of Champlain’s explorations as the primary source document’ (cited in Bangor Daily News , 2009). Pearce also notes that the map inverts the dynamic where typically only the written account gets mapped, represented and subsequently reproduced as history. Indigenous voices are included in the map as imagined counterparts, correlating to the written account by Champlain.

Focus Figure 3.2 — Detail of legend from Margaret Pearce and Michael Hermann, ‘They Would Not Take Me There: People, Places, and Stories from Champlain’s Travels in Canada, 1603–1616’. Margaret Pearce

Typography is used to distinguish the difference voices floating in the territory. Garamond in blue for Champlain’s voice, in green for Native voices, in brown for introductory text from the cartographers; Univers italic in black for the cartographers’ voices that comment and translate. Pearce states that the levelling of visual and narrative importance between the written record and imagined voice works towards decolonizing the map.

155

What is highlighted through the use of missing voices and Champlain’s own experiences, is how dependent the French colonist was on the knowledge and help of the Algonquin, Went, Wabanaki and Innu. Rather than adding another hagiographic account to the pile of nation-building narratives of Canada and Quebec, this map artfully reduces the hero to a man utterly dependent on the help of others and who was ultimately refused access to the Northwest Passage, the James Bay — referred to in the title of the map. Champlain’s desire to find this passage that would allow French trade with Asia was frustrated for over a decade when finally he loses faith. The narrative in the map offers a linear, spatial and rhizomatic organization. Readers can follow a temporal structure by following the sequence of numbers and dates, but they can also go to parts of the map that are most relevant to their own interests. Ultimately, the reader can follow a variety of narrative paths that at times follow linear time, contours of the land and water, or intensity of information. This narrative map, reiterates as do many Indigenous-led mapping projects,  the fact that the territory of Turtle Island was rich with inhabitants, culture and ways of living — not empty or needing to be civilized or Westernized. While it documents and problematizes the conventional discovery narrative of colonialism, it also shows how land and bodies of water were (and continue to be) living with spoken words that described their importance and rela­ tion to the original inhabitants. ●

Margaret Pearce and Michael Hermann 220221_CritViz_Book_Full_revisions.indd 155

7/20/22 9:47 AM

156

Focus

Focus Figure 3.3 — Detail of map from Margaret Pearce and Michael Hermann, ‘They Would Not Take Me There: People, Places, and Stories from Champlain’s Travels in Canada, 1603–1616’. Margaret Pearce

Rethinking the Representation of Data 220221_CritViz_Book_Full_revisions.indd 156

7/20/22 9:47 AM

220221_CritViz_Book_Full_revisions.indd 157

7/20/22 9:47 AM

V Data and the City

220221_CritViz_Book_Full_revisions.indd 158

7/20/22 9:47 AM

T

he so-called ‘smart city’ is championed by governments, technologists, urbanists and designers as a prime locus for the cultural, demographic and environmental challenges of the twenty-first century. Data visualization is often identified as a vital tool of the contemporary city, and it plays a key role in many visions of the smart city. In the rhetoric of urban design and planning advocates, smart cities are promoted as places where citizens are informed about, networked and engaged in — via technology — the relationship between their activities and the broader urban environment. Citizens, a term used b ­ roadly despite its exclusionary connotations of state membership (Kivisto and Faist, 2007) are also encouraged to see the city as something they can ‘collectively tune’ — unlike the more monolithic, inflexible structures of twentieth-century cities (Arup, 2010). Faced with the challenges of population growth, extreme weather, climate change, urbanization and resource depletion, smart cities can better use urban informatics to monitor infrastructure, buildings and activities and make invisible processes visible — ‘engaging, informative, even beautiful’ (Arup, 2010). In this chapter, we will explore the role and challenge of visualization in the smart city, and what it might learn from the rich histories of counter-­ mapping, critical cartography and critical data studies.

159

An early promoter of the smart or ‘sensing’ city is the Senseable City Lab at Massachusetts Institute of Technology, where projects began in the mid2000s using aggregated data from mobile phones, and mapping them onto the geography of the city. In the lab’s project Real Time Rome, launched at the Venice Biennale in 2007, mobile phone data was correlated with the gps locations of city buses and taxis, to create real-time visualizations of city flows during specific events such as the 2006 fifa World Cup Final (Duarte and Ratti, 2019). More recent projects such as Urban Exposures use phone data and measurements of air pollution to provide a more nuanced and accurate mapping of peoples’ exposure to pollutants. The lab’s Trash|Track project (discussed in Chapter 1) collected 3,000 objects about to be discarded from 500 volunteers in Seattle and fitted them with trackers, a project which unexpectedly highlighted the illegal trafficking of electronic waste. In one case, the resulting visualization had a demonstrable effect on the behaviour of a participating researcher, who gave up using bottled water after seeing how ‘quick decisions made out of convenience led to long-lived negative consequences — plastic bottles sitting for eternity in a landfill’ (Vanky, 2016, p. 18). Such experimental projects reveal both the promise and the hype around the visualization-empowered twenty-first-century citizen. One researcher giving up bottled water may provide anecdotal evidence of a behaviour change shift resulting from visualization, but not quite the ‘collectively tuned’ smart city. According to Carlo Ratti, director of the Senseable City Lab, the visualization of big data is a way of showing real-time patterns of

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 159

7/20/22 9:47 AM

160

people living in a city — ‘the signature of humanity’. But Ratti argues that for big data to have a big impact it needs to be ‘analysed and digested’: Today, that is done by visualization experts, who might make an image, a graph or a video; what we are now working on at the lab is making a tool that puts data at your fingertips ... When everyone — f rom scientists to ­politicians to everyday commuters — has the opportunity to explore invisible dimensions of the city around them, we will see radical behavioural change. When the city is revealed, particularly in real time, we all will become smart citizens. (­R atti and ­Claudel, 2018)

The implication here is that with the forward march of sensing and visualization technologies, the smart city will arrive at a near-synchronous state of monitoring-visualizing-tuning. A hurricane is forecast, evacuation plans are e-distributed among the populus and the flood barriers automatically rise; air quality approaches a dangerous level in the city centre and traffic is re-routed. Speculative visualization projects such as Real Time Rome, or Arup’s Melbourne Smart City (Arup, 2010) — which hypothesized a floating ‘net’ of data to communicate patterns of urban activity (Moere and Hill, 2012, p. 34) — seem to suggest that the feedback loop between the technology and its citizens will be increasingly streamlined to the point where humans are fed a stream of just-in-time, accessible information (via floating data nets or embedded displays), or cut out of the loop completely. However, such envisioning tends to glide over the point that, as Orit H ­ alpern puts it, a visualization is ‘non-synchronic’ with the event it is depicting and translating. Between the event and its visual representation (even if it is animated in ‘real-time’) is a lag: This non-synchronicity preoccupies our imaginings of “real-time” ­i nteractivity and data visualization, driving a constant redefinition of the ­temporal lags b ­ etween collecting, analysing, displaying and ­using ­i nterfaces. (­Halpern, 2014, p. 22)

Arguing that vision and reason have become intertwined with governmentality, Halpern introduces the case study of the South Korean smart city Songdo, which is being built from the ground up on drained land from the Yellow Sea south of Seoul. A collaboration between the South ­Korean government with US property developer Gale International, architect Kohn Pedersen Fox and network infrastructure provider Cisco Systems, Songdo is conceived as a ‘testing grounds for the future of human habitation’, and part of a free economic zone integrating an airport, ­infrastructure, ­technology and housing (Halpern, 2014, p. 2). Songdo’s eco-friendly techno-­infra­structure includes rainwater collection, remote heating, ventilation and air conditioning control, and waste disposal via pneumatic tubes to an underground waste facility. Flow diagrams on k pf’s website are used to contrast Songdo infrastructure with conventional systems (k pf, n.d.; Lobo, 2014).

Data and the City

220221_CritViz_Book_Full_revisions.indd 160

7/20/22 9:47 AM

The assumption that ‘life itself’ can be managed by bandwidth, and that data presents ‘stability, wealth and sensorial pleasure’ is built into the foundations of Songdo, which is constructed over an infrastructure of tunnels for fibre optic cabling, and comprises networked buildings whose surfaces function as interface, wired with sensors to track behaviours and transfer information. According to Halpern, Songdo represents a current obsession in urban planning for cities made to hold millions of people, cities that are sold by engineers and consultants as ‘commodities’ built on a ‘biopolitical’ hypothesis that data can be gathered and traded in ‘recombinable units of attention, behaviour and credit’ (Halpern, 2014, p. 5).

161

Where the machine-age city offered the promise of frictionless efficiency and precision in the movement of goods, the twenty-first-century smart city promises security through predictive policing and smart clouds, with zero waste helping the smart city ‘do its bit’ for the environment. As part of an ongoing Google Earth Outreach special project on air quality in cities, air pollution sensors have been installed on selected Google Street View vehicles to produce data for a map of Copenhagen, developed in partnership with the City of Copenhagen and Utrecht University. The map shows the street level concentration of black carbon and ultrafine particle pollution, which, according to Google’s promotional material, is being used by the city to work with architects and designers ‘to rethink the city for the future’ (Powell, 2019). Again, details on how Copenhagen’s officials, designers and Google are ‘rethinking’ the city are not forthcoming. This may be simply because it is too early to spell out how monitoring and visualizing air quality might lead to steps to improve it, but our point is to show how the rhetoric of smart city visualization tends to lean rather heavily on the notion that ­visualizing the problem is tantamount to addressing the problem. The same critique might be applied to any number of large-scale visuali­za­ tion projects in which important environmental information is r­ epresented continuously on large-scale surfaces. A recent project by rndr in the Hague indicates that there is a significant amount of work for designers in making invisible flows visible through experiential visualization. The ­tunnel over the A2 highway in Utrecht (part of the urban development master plan for Leidsche Rijn by Maxwan which by 2030, will bring 110,000 people to live in the area) makes the district into a single e­ ntity, hiding the cars on the motorway, and keeping their noise and exhaust fumes, underground. rndr’s light installation open highway — 14 light bars (11 m long) placed to correspond with the lanes below — makes the hidden highway visible again and ‘shows that you can make something beautiful out of something intrusive’ (rndr, 2019).

Chapter V

220221_CritViz_Book_Full_revisions.indd 161

Critical Visualization

7/20/22 9:47 AM

162

Figure 5.01 — Open Highway installation by rndr. rndr —  Jeroen Barendse (traffic flows) and Juri Hiensch (installation)

Beautifying data is a function of visualization, discussed in C ­ hapter 6. Making it also actionable is more central to the question of this ­chapter — how might a critical visualization respond to the challenges of the ­contemporary city? In light of the catastrophic bush fires in Australia, such p ­ roposals as Arup’s data ‘net’ could take on the potential role of emer­gency signage rather than a gentle nudge to ‘look after the earth’. As cities and megacities begin to take steps to protect themselves against anticipated extreme weather events as a result of climate change, visualization also plays an important role in explaining and persuading various publics that proposed planning projects are in the interests of stakeholders. An instructive example is the array of renderings, maps and visualizations presented for New York’s $930 million Rebuild by Design project by ­Bjarke Ingels Group (big) and Dutch firm One Architecture & Urbanism. The images depict the plan for a ten-mile-long flood barrier to protect areas at the low-lying southern tip of Manhattan from floods and storm surges, described by architect Bjarke Ingels, founder, and creative partner of big, as a series of public spaces and viewpoints, or ‘a string of pearls of social and environmental amenities tailored to their specific neighbour­ hoods, which also happens to shield their hinterlands from flooding’ (Gilbert, 2014). The scheme, named eponymously (with a rhetorical finger beckoning its publics) ‘big u’, is strategically conceived not as a barrier but as a berm, with renderings of future citizens walking, sightseeing and even wading through flooded terraces planted with salt-water-tolerant ­flora. To make such a scheme ‘urbanistically successful’ Ingels told the design blog Dezeen, ‘It needs to happen rooted in a dialogue with the ­different communities’ (Howarth, 2018).

Data and the City

220221_CritViz_Book_Full_revisions.indd 162

7/20/22 9:47 AM

163

Figure 5.02 — T he Rebuild by Design project, a $930 million flood barrier plan for lower Manhattan. Bjarke Ingels Group (big) and Dutch firm One Architecture & Urbanism

Unlike New York’s grand planning projects of the twentieth century, Rebuild by Design was premised on a promise of early-stage community consultation. But which communities? In its striking maps of sea currents, flooded areas and renderings of resilient New Yorkers calmly participating in a kind of flood-friendly blue-skied utopia, big u adopts a surprisingly parochial position vis-à-vis the larger context of another storm surge like that experienced with Hurricane Sandy in 2012. In his 2017 critique of the project in the book Extreme Cities, Ashley Dawson cites a report by a climate scientist who argues that the flood barrier increases rather than decreases risk, providing a false sense of security (like the New Orleans levees) and displacing danger into other areas: Where will the water that the big u turns aside go? It is likely to end up in adjacent communities with large poor populations such as Red Hook, where ­hurricane Sandy hit public housing particularly hard. Interestingly, big looked at flood defences for Red Hook in its research, but this was not part of the final proposal. It should not be particularly surprising that ­defense of Wall Street garnered more attention and funding than defense of a ­historically poor (if rapidly gentrifying) neighbourhood such as Red Hook. (Dawson, 2017, p. 160)

Urban planning for extreme weather, it seems, is being performed with the promise of resilience among a small, fortunate elite — in this case, those living and working in lower Manhattan, an area which is home to the world’s two largest stock exchanges, and whose real estate is among the most expensive in the world. As has become quite common in contemporary geoand environmental politics, the sense of a shared world that dominated visualizations of the last century — for example, B ­ uckminster Fuller’s ‘geoscope’ proposal for a 60-metre globe lined with screens graphically representing an inventory of the earth’s resources and developmental trends — has been supplanted by images of model city quarters for a surviving few (Wigley, 1997). Latour characterizes a position of ‘post-­politics’ that seems to account for the impulse to stake out and protect a defined gated community whose future is, by definition, short term, since all territory is now threatened by

Chapter V

220221_CritViz_Book_Full_revisions.indd 163

Critical Visualization

7/20/22 9:47 AM

164

ecological degradation. Post-politics is a politics with no object, ‘since it rejects the world that it claims to inhabit’ (Latour, 2018, p. 38). In the sense that it represents a phantom city that nevertheless e­ ncapsulates an object of desire, and a belief system, the image of the smart city can be seen as an outgrowth of the mid-century ‘spectacle’ as famously d ­ escribed in Guy Debord’s Society of the Spectacle wherein ‘everything that was ­directly lived has receded into a representation’ (Debord, 1967/1994, p. 32). At the root of this spectacle of represented life, argued Debord, is the oldest social specialization, the specialization of power. The spectacle of the smart city, by extension, recalls what Debord called ‘the diplomatic representation of hierarchic society to itself’ (Debord, 1967/1994, p. 37). If visualization works to re-present lived existence in the form of a life in a smart city, promising a better life for all while serving a lucky few, might not a critical visualization be able to support a counter project, to reaffirm and reimagine everyday life in a way that creates new unities and resists commodification? This would be to adapt the Situationist project and that of Henri Lefebvre, a critique of everyday life that is ‘the critique of existing needs and the creation of new desires’ (Wark, 2013, p. 25). A prominent historical example of a visualization that both embodies a ­critique of the city and a reimagining of everyday life, the creation of new situations and desires, is the psychogeographic map of a dérive assembled by Debord and Asger Jorn from cut-up sections of the Paris city map. Debord and Jorn’s Guide psychogéographique de Paris. Discours sur les passions de l’amour, (1957) was created after the authors undertook a dérive through Paris, and set out to represent ‘the spontaneous direction a subject takes while traversing this environment without taking into account the practical sequences — involving work or distraction — which usually ­condition his behaviour’ (Gilles Rion, cited in Moro, 2021, p. 80). To reify the psychogeographic map alone would be to risk slipping into the trap of the spectacle, of course, which repackages everyday life as an image and sells it back to us — or, as Debord states in Society of the ­S pectacle, ‘The spectacle corresponds to the historical moment at which the commodity completes its colonization of social life’ (1967/1994, p. 29). Debord and Jorn had pointedly satirized the emptiness at the heart of the ­spectacle in their 1957 book End of Copenhagen, a détournement (­hijacking) of i­ mages and text from newspapers and magazines collected on a visit to a Danish newsstand; the Guide psychogéographique de Paris similarly satirizes the functionality of a map with its documentation of ‘ambiances’ (Sadler, 1998, p. 27). The critical point of Debord and Jorn’s visualization was to pose an event — a dérive — as an alternative to the functionalist city promoted by Modernist planning zealots in the 1933 Athens Charter, a manifesto issued by the Congrès Internationaux d’Architecture M ­ oderne (ciam), which

Data and the City

220221_CritViz_Book_Full_revisions.indd 164

7/20/22 9:47 AM

advocated rigid functional zoning of cities into work, ­recreation and ­housing zones, and high-density a­ partment blocks connected by highspeed transportation corridors for private cars (Sadler, 1998, p. 22). The Guide psychogéographique de Paris evoked instead the ­indigenous zones, the subconscious pushes and pulls of the city’s clusters of ­buildings and activities. Satirizing the Athens Charter’s embrace of t­ raffic as a fourth function that brought together the zones of work, r­ ecreation and housing, Debord promoted a ‘unitary urbanism’ that ‘acknowledges no boundaries. It aims to form a unitary human milieu in which separations such as work/leisure or public/private will finally be dissolved’ (Sadler, 1998, p. 25).

165

The trajectory of practices derived and inspired by the Situationist dérive and the Guide psychogéographique de Paris included a revival and blossom­ ing of counter-cartographies in the early 2000s with the availability of satellite navigation, following the US government’s discontinuation of ‘selective availability’, making global positioning available to civilians to an accuracy of five metres. Today’s L5 satellite signals and new global navigation satellite systems chips in smartphones promise centimetre-level accuracy. Artists such as Esther Polak and Christian Nold détourned satellite navigation by inviting participants to undertake dérives equipped with global positioning. The resulting projects suggested the advent of a form of citizen-led cartography: Polak’s Amsterdam RealTime (2002) traced participants’ movements through Amsterdam via a central server linked to a large screen at the Waag Society, which compiled images as sinewy traces on a black screen to render a live map of the city that was ‘created’ by its residents. Nold’s Bio Mapping projects engaged participants (initially gallery audiences) in generating subjective and collaborative maps of areas in which they lived or worked by equipping them with satellite navigation and a galvanic sensor that measured changes in sweat level on the wearer’s finger as an indicator of emotional intensity; global positioning enabled these points of intensity to be mapped so that, after the dérive, participants could annotate them along with their routes, foregrounding the psychogeographic flavour of the mappings. In the 2007 Stockport E ­ motion Map, for example, which was developed over six public mapping events, participants were asked to sketch their responses to a variety of s­ erious and humorous provocations about their daily lives, such as what really annoyed them about their town, where they met their friends, and who were the most important and most dangerous people. Nold framed these richly layered, subjective maps as a ‘critical reaction towards the currently dominant concept of pervasive technology, which aims for computer “intelligence” to be integrated everywhere’ (Nold, 2009, p. 3), but as Brian Holmes has noted, these neo-psychogeogra­phic, satellite navigation enabled projects were ‘fraught with ambiguity’:

Chapter V

220221_CritViz_Book_Full_revisions.indd 165

Critical Visualization

7/20/22 9:47 AM

166

The individual’s wavering life-line appears at once as testimony of human ­singularity in time, and proof of infallible performance by the satellite ­mapping system. (Holmes, 2006, p. 25)

Writing in Nold’s Emotional Cartography book, the Raqs Media C ­ ollective positions biometric technologies both as tools of state control and as a means to ‘hyperlink aspects of being — cross-referencing acts, attributes and attitudes that constitute the database of a person’s becoming over time’ (Nold, 2009, p. 8). Such ‘cross-referencing’ of acts, attributes and a­ ttitudes suggests the potential of the Bio Mapping to aggregate the s­ ituations, or ­desires, that defy the colonizing logic of the spectacle.

Figure 5.03 — Detail from Stockport Emotion Map (2007). Christian Nold and Daniela Boraschi

In a sense, the ambiguity inherent to tracking a psychogeographic wandering with military-grade satellite signals embodies a paradox that emerges as soon as practitioners seek to ‘recuperate’ — to use McKenzie Wark’s term — the Situationist project. As Nold’s project evolved, its applications moved away from gallery audiences toward community groups, where discussions might move from noticing the effects of car traffic on one’s own body, to discussion of ‘the lack of public space and identifying its social and political causes’ (Nold, 2009, p. 7). On the one hand, Bio Mapping presented a platform and mediating object with which awareness might be raised of otherwise-concealed issues, potentially informing the planning and development of a city that better accommodates a diverse range of citizens. On the other hand, a technology-enabled participatory mapping project quickly becomes a token gesture for planners seeking to tick the ‘community consultation’ box in order to proceed with a controversial regeneration plan. The image or visualization of participation becomes a stand-in for the messy and agonistic process of consultation, and an appropriation of the everyday practices of people that, as Michel de Certeau (1984) argued actually compose the city. To adopt Wark’s argument: Today … everyday life has been so colonized by the spectacle of the ­commodity form that it is unable to formulate a new relation between need and desire. (Wark, 2013, p. 10)

Data and the City

220221_CritViz_Book_Full_revisions.indd 166

7/20/22 9:47 AM

167

Figure 5.04 — Stockport Emotion Map (2007) — Christian Nold, designed by Daniela Boraschi, supported by Stockport Council and Lend Lease, a two-month project involving approximately 200 people in mapping their local area. Christian Nold and Daniela Boraschi

The problem is compounded by the circumstances of the present-day ­climate crisis, with its threat of extreme weather, overcrowding and ­resource de­ple­tion, and in which the capacity of everyday life to f­ormulate new relations and situations is rendered powerless by the vast scale and apparently irreversible process of environmental degradation. Wark ­describes a third phase of the spectacle, that follows Debord’s twophase passage of the concentrated spectacle of fascism to the diffused spectacle of consumerism. According to Wark, the diffused spectacle has been succeeded by the spectacle of disintegration, symbolized by the ­swirling garbage patch in the North Pacific ocean, an emblem of ‘the ­passing show of contemporary life, with its jetsam of jostling plastic ­artefacts, all twisting higgledy-piggledy on and below the surface of the ocean’ (Wark, 2013, p. 8). Wark adds: The command of the concentrated spectacle was obey! The command of the diffuse spectacle was: bu y! the command of the disintegrating spectacle is: r ecycle! (Wark, 2013, p. 10)

Yet the depressing image of a swirling patch of consumerist effluence offers nothing in terms of a programme for surviving within this precarious future. So far, our discussion of counter-visualization in the smart city has privileged the visualization of data by and for humans. But with the ­spectre of climate crisis, the centuries-old schema that placed humans on the centre of the earth and cast ‘nature’ as an infinite storehouse to supply the materials for human productivity has been destabilized. Latour characterizes this shift as follows:

Chapter V

220221_CritViz_Book_Full_revisions.indd 167

Critical Visualization

7/20/22 9:47 AM

168

Humans have always modified their environment, of course, but the term ­designated only their surroundings, that which, precisely, encircled them. They remained the central figures, only modifying the decor of their dramas around the edges. Today the decor, the wings, the background, the whole building have come on stage and are competing with the actors for the p ­ rincipal role. (Latour, 2018, p. 43)

With a no-longer passive nature, it becomes more important to be able to represent the ways in which the environment is behaving. One pioneer in this effort is the New York-based researcher and artist Natalie J­ eremijenko, whose experimental OneTrees project both stretched the definition of a map and critiqued the scientific-objectivist view of visualization. Working with the non-profit group pond in California, Jeremijenko planted pairs of genetically cloned trees in the differing neighbourhoods of the Bay Area. The idea was for the trees themselves to become sensors, recording, registering and thus mapping over time the different microclimates, cultural attitudes toward the care of trees, and socio-economic conditions of the neighbourhoods. In addition, the tree-map disrupted conventions of the time frame of cartography and what qualifies as data. A printed map produced to locate the trees and introduce the project reinforced the critique. The map of the mapping, designed with ­architecture practice TerraSwarm, juxtaposes a usgs Landsat 7 aerial image of the Bay Area with lay knowledge, such as the locations of bike trails, common hawk flight paths and the habitat of the endemic song sparrow. The implicit critique is of culturally entrenched hierarchies of information, which, for example, prioritize satellite views and expert, institutional knowledge over the knowledge of laypeople.

Figure 5.05 — In OneTrees, cloned pairs of trees become a map of socio-environmental conditions, while a map of the trees juxtaposes expert and lay data. Terraswarm, Natalie Jeremijenko, pond

Jeremijenko’s more recent Phenology Clock, initiated at the ­Environ­mental Health Clinic at New York University, adopts and subverts the form of a ­mechanical clock, an invention whose history is tied to ­industrialization, repurposing it to display when local organisms are budding and ­blooming in the calendar year. Aside from providing a useful indicator of the pollin­ ation migration and flowering of local organisms, the ­Phenology Clock

Data and the City

220221_CritViz_Book_Full_revisions.indd 168

7/20/22 9:47 AM

f­ unctions as a critical design which re-conceptualizes time as s­ omething defined by biogeochemical processes, the processes that create microclimates upon which humans depend for food, nutrients, air and water ­quality and health. As noted in the supporting text for the 2015 version of the Phenology Clock in Milwaukee:

169

Phenology is our most sensitive indicator of climate destabilization and under­ standing this will inform the creative collective work of reimagining and redesigning our collective relationship to natural systems. In the time of a ­pollinator crisis, threatened ecosystem collapse and climate ­destabilization, the temporal structure of the ecosystem is a valuable way to represent ­organisms and coordinate our collective attention toward the workings of ­natural systems. (Hannah and Krajewski, 2015, p. 133)

Since one aim of this chapter is to better understand how a critical visuali­ zation practice might build on the flourishing discourse of critical data studies and critical cartography, we shall set out to find trajectories from psychogeography to the everyday practices of the contemporary city. To the extent that Nold’s mappings move to draw participants’ attention to the effects of car traffic on their bodies, and the causes behind a lack of public space, Bio Mapping presents a transition between the Situationist interest in recomposing the city by drifting through its grid, and hijacking and repurposing its commodity logics, and the planning profession’s interest in participatory design. The link between psychogeography and planning, while rejected by Debord, has produced its own fertile trajectory.

Figure 5.06 — T he Phenology Clock (2015) reimagines the measurement of time in terms of the budding and blooming of local organisms in the calendar year. Natalie Jeremijenko, Tega Brain, Jake Richardson and Blacki Migliozzi

Denis Wood, for example, has drawn a useful comparison between the revolutionary psychogeography of Debord and colleagues in France with the more pragmatic psychogeography that emerged in the usa — ­initially independent and unaware of the French version — at the University of Chicago and Clark University in Massachusetts in 1967. In that year at Clark, psychologist David Stea offered the first University course in psychogeo­graphy, drawing from the ‘cognitive mapping’ methods of city

Chapter V

220221_CritViz_Book_Full_revisions.indd 169

Critical Visualization

7/20/22 9:47 AM

170

planner Kevin Lynch, whose influential book The Image of the City was a core text. Unlike the contestatory dérive, Lynch’s method was to study the ‘psychological and sensual effects of the physical form of the city’ by asking residents to draw maps which he then analysed to gauge a sense of a city’s legibility — through aggregating residents’ mental images of identifiable districts, confusing nodes and impermeable boundaries. As Wood notes: The obvious difference is in their situations: the Situationists outside the ­plan­n ing profession and so free to think about the problem as dictated by their roots in Surrealism and their commitment to dialectical materialism; Lynch within the profession, and so shackled to thinking through the ­problem from the perspective of city government, with its grab-bag of service p ­ rovision (­especially of roads, sewer and water), condemnation, ordinances and i­ ncentives. (D. Wood, 2010, p. 194)

As two very different means of conveying subjective views of the city, both French and North American psychogeographic practices used visual­ization approaches: the cut-up maps of the Situationists and the cognitive maps of Lynch and colleagues. Today, the ­complexity of city-planning issues is increasingly identified as a challenge that can be addressed using visualization tools in smart city development ­technologies, where the representation of diverse viewpoints is being swept into a black box of automated planning procedures. Sidewalk Labs, a subsidiary of Alphabet (Google’s parent company), for e­ xample, is ­currently developing a generative design tool that claims to do two things that existing planning tools cannot: using machine learning and computational design, it can ‘help planners generate not just one or two but millions of comprehensive planning scenarios’ and second, it can ‘help evaluate all kinds of impacts these different scenarios could have on key quality-of-life m ­ easures, producing a set of options that best ­reflects a community’s priorities’ (­W hitney and Ho, 2019). The challenge inherent to such rhetoric is how it fares when terms like ‘community’ are unravelled to reveal actual communities that do not ­comply with the often unexamined agendas embedded in the machine learning and computational design. Sidewalk Labs was recently at the centre of a planning controversy in Toronto, where leaked early details of its proposed Quayside development included plans to privatize parts of ­Toronto’s lakeshore with privately regulated roads, charter schools and a ‘continuous surveillance’ data collection system with a penalty of r­ educed services for citizens who opt out of the system. Sidewalk Labs pulled out of the project after an open letter was sent to Waterfront Toronto, the a­ gency overseeing land development, by Indigenous elder Duke R ­ edbird and architect Calvin Brook, accusing Sidewalk Labs of ‘hollow and t­ okenistic’ consultation that completely ignored their r­ ecommendations (Doctorow, 2019; Warren, 2019). A tokenistic approach to community consultation can thus undermine the whole premise of the smart city, ‘­fine-tuned’ by its residents.

Data and the City

220221_CritViz_Book_Full_revisions.indd 170

7/20/22 9:47 AM

As Adam Greenfield has observed, one issue with smart city narratives is that they work better when used to describe generic cities, rather than actual ones.

171

Only by proposing to install generic technologies on generic landscapes in a generic future can advocates avoid running afoul of the knotty complexities that crop up immediately any time actual technologies are deployed in actual places. (Greenfield, 2013, p. 31)

Underlying Sidewalk Labs’ promise to represent myriad ­stakeholder inter­ ests in a million permutations of an urban plan, is the ­philosophical as­ sumption that all useful knowledge can be digitized and ­translated into artificial intelligence that far surpasses all human intelligence — the ­imminent technological singularity promoted by Ray Kurzweil among others. This is played out in smart city discourse in the hypothesis that urban planning is too complex for human beings to understand, as Greenfield has observed, and artificial intelligence is needed to do the job. Other efforts to digitize and automate aspects of planning indicate the ­level of investment in the notion of a technologically enabled ­citizenry that is actively engaged in the governance and planning of its urban ­environment. Since the Open Government Partnership began promoting citizen partici­pation and open data in 2011, several urban authorities have made public large data sets, with the stated aim of facilitating ­accountable governing and participatory citizenship (Lauriault and Francoli, 2017; Moere and Hill, 2012). Interest in the smart city has spread so rapidly across global regions, in fact, that it has become a ‘major paradigm of urban ­policy, planning and development’ (Joss et al., 2017, p. 29). The Smart City M ­ ission in India, the Smart City Challenge in the US, the European ­Commission-funded Smart City Solutions and the £24 million Future ­Cities Demonstrator initiative in the UK — awarded to Glasgow in 2013 — are an indication of the trend. In their analysis of the language used to promote smart cities, Joss, Cook and Dayot (2017) note how the terms ‘citizen’ and ‘citizen-centric’, which crop up frequently in the marketing literature, are inconsistently ­defined. The language used — for example, by the British Standards ­Institutions to promote the smart city — suggests that the challenges facing the contempo­ rary city, including the failures of representation in traditional ­governance, can be met by an organizational change informed by real time, compre­ hensive city data, informing a ‘more open and inclusive’ governance ‘allow­ ing citizens, policymakers and businesses to work together to manage the life of the city, for the benefit of all’ (Joss et al., 2017, p. 36). But quite how the smart technologies translate into a new, participatory form of govern­ ance is not made clear. The lack of realistic scenarios for collaborative decision-making involving stakeholders perpetually risks repeating the twentieth-century positivist planning traditions by advancing ‘a smooth

Chapter V

220221_CritViz_Book_Full_revisions.indd 171

Critical Visualization

7/20/22 9:47 AM

172

techno-bureaucratic governance mode which assumes an unproblematic community and benign politics’ (Joss et al., 2017, p. 44). PA RT ICI PAT ORY PL A N N I NG : H EC T OR Visualization need not be solely aligned with technological black ­boxing, however. Instead, it can be used to surface the tensions and issues at stake in a given project area or proposal. The Newark-based urban design, planning and civic arts practice hector takes a creative, participatory approach to making design useful for community organizing that aims to expose differences of interest and tensions around urban development, via workshops, low-tech models and large-scale three-dimensional city plans. ‘For most of the projects we’ve worked on at hector,’ says Damon Rich, who co-founded the practice with Jae Shin in 2013, ‘unless there’s some way to put existing antagonisms in the middle and make them visible and something people can talk about, the project’s not going to work’ (D. Rich, personal communication, December 1, 2019). hector recently worked on a project in South Philadelphia, for example, for a coalition of resident organizations looking to mobilize local communities in a campaign for city officials to invest in the neighbourhood’s aging park, in an area whose history had seen waves of immigration from Italy, ­Ireland, the southern United States, and, since the 1970s, Cambodia, Vietnam, Laos, Burma and Bhutan. Among other forces that made it challenging for the organizing coalition to succeed was an abusive narrative that more recent immigrants from South East Asia were responsible for crime and litter. As Rich parodies the view, ‘Things were idyllic and perfect until that most recent group showed up’. To confront and undermine this false narrative, hector was commissioned by Philadelphia Mural Arts to organize a Summer Youth Employment Program that, among other activities, conducted archival research uncovering newspaper articles documenting anti-litter campaigns and occasional violence throughout the park’s century of existence. By historicizing the current regime of racial domination, largescale drawings, produced by thirteen teenagers with hector’s support, prodded and unravelled some unfounded neighbourhood myths in o ­ rder to set the table for productive deliberations on its physical future. Rich’s point is that planning is complex, layered and negotiated between groups of ­humans, a fact that the smart city rhetoric seems to neglect as much as the power broker rhetoric of top-down, planning in the Modernist era. I’ve read enough history of planning to know that any tendencies to ­suggest that we’re going to be able to displace politics with genius ­planning don’t seem like good ideas in retrospect. The question becomes how can ­antagonisms get ­represented in a productive way? (D. Rich, personal ­communication, ­December 1, 2019)

Data and the City

220221_CritViz_Book_Full_revisions.indd 172

7/20/22 9:47 AM

In his suggestion that antagonism is productive rather than a problem to be swept away by technology, Rich’s approach recalls Chantal Mouffe’s concept of agonistic pluralism: ‘Within the context of the political community, the opponent should be considered not as an enemy to be destroyed, but as an adversary whose existence is legitimate and must be tolerated’ (Mouffe, 1993, p. 4).

173

Figure 5.07 — A s part of hector’s work in Philadelphia, student investigators study conditions on the ground in Mifflin Square Park before creating analytic drawings for the Mifflin Square Action Poster. hector

A practice of counter cartography, or counter-visualization, then, is one that represents antagonisms and productive tensions, a task that, historically, visualizations have tended to avoid. To foreground conflict in urbanism is to move the discourse out of removed visionary practices toward messy, adaptive and accidental ones. This is a premise of the Conflict Urbanism initiative at Columbia University’s Center for Spatial Research, whose researchers argue that ‘cities are not only destroyed but also built through conflict’ (Saldarriaga et al., 2017). Like her earlier work challenging the authority of the satellite view (see Chapter 1), Laura Kurgan’s work at the Center for Spatial Research, which she co-founded in 2015, supplements the synoptic ‘view from nowhere’ with street-level data. In an interactive map focused on the damage caused by the civil war in the city of Aleppo since it fell to Syrian Government forces in 2016, for example, both satellite and street-level video data are layered in a gis interface, representing a credible account of events from different perspectives and at different temporal rates. The high cost, high-resolution satellite imagery analysed by the unitar-­u nosat, the branch of the UN dedicated to manual examination of satellite images to identify bomb damage in war zones, represents change, but at only four moments during the five-year period of the war. The low-resolution Landsat imagery, which can track change every two weeks; and videos uploaded to YouTube by witnesses of the bombings provide a street-level corroboration of events. Over a million videos of the conflict have been

Chapter V

220221_CritViz_Book_Full_revisions.indd 173

Critical Visualization

7/20/22 9:47 AM

174

Figure 5.08 — Named in homage to South Philadelphian Meek Mill’s 2013 song ‘Levels’, this poster reports what hector and Park Powers students discovered about who makes the decisions about Philadelphia’s parks, and compares the resources provided for Mifflin Square Park to those provided for Rittenhouse Square Park, which is located in one of the city’s most affluent areas. hector & Park Powers students (Jae Shin & Damon Rich with Nashay Blancher, Abou Diallo, Abdou Drabo, Breyelle Gupton, Beyoncé Harmon, Naquan Harmon, Devonne Miller, José Pantoja, Rakim Perry, Antoine Rainey, Noelia Rodriguez-Ruiz, and Paige Scott-Cooper), Mural Arts Philadelphia Restored Spaces (Julius Ferraro, Shari Hersh, Margaret Kearney) and Art Education (Patricia Barrera, Ambrose Liu, and Christine Rhodes), Park Powers Participating Organizations: seamaac, Cambodian Association of Greater Philadelphia, Bhutanese American Organization –Philadelphia, 7th Street Civic Community Association, United Communities Southeast Philadelphia, Philadelphia Parks and Recreation, and Philadelphia Water Department.

Data and the City

220221_CritViz_Book_Full_revisions.indd 174

7/20/22 9:47 AM

uploaded since 2012, and major news agencies now use YouTube as a primary source in their reporting. By cross-referencing the three sources of data, the Conflict Urbanism: Aleppo project was able to reveal which areas of the city suffered the most damage and the strategic nature of the bombing, and the project was also able to provide powerful evidence that bomb damage to the city, including some of its heritage and residential sites, was intentional and not just a by-product of military operations as claimed (Saldarriaga et al., 2017).

175

To imagine smart cities informed by the multi-layered, multi-sourced visuali­ zation of conflict investigated by groups like Center for Spatial Research (and the London-based project Forensic Architecture) is to enrich the urban discourse with data from the kinds of human experiences traditionally seen as beyond the scope of future city visionaries. It would be to embrace what Forensic Architecture architect and researcher Eyal Weizman has called the ‘forensic turn’ in human rights, whereby the urban fabric can ­become a reliable witness to contested events (Weizman, 2017, p. 78). Here, the legacy of the Situationist project is apparent, to counter top-down, synoptic envisioning with the practice of showing a city’s granular transformation over the passage of time, including the forensic analysis of damage and trauma. This is the distinction between the ‘map’ and the ‘mapping’ — the latter term denoting a continuing process of discovery (Hall, 2012). In critiquing the rhetoric around smart cities, Paul Dourish has adopted the phrase ‘accidentally smart’ to describe how cities — in contrast to the rhetoric — become smart in a piecemeal way, under the control of various groups, and shaped by politics, ‘without a master plan, and with a lot of patching, hacking, jury-rigging and settling’ (Dourish, 2016, p. 37). As well as puncturing a hole in the bubble of a promised seamless ­connectivity, the ‘accidentally smart city’ provides an account of how institutional infrastructures are adapted or hacked, as in the diy networking of the ­mid-2000s, which set out to create new internets. Joss et al. provide comparable accounts of digital technologies being used to support crowdsourced constitutional drafting in Iceland and socio-technical innovations linked to hackathons and a maker culture (Joss et al., 2017, pp. 31–32). A picture emerges of a messier, situated, hacked and patched version of innovation, governance and even visualization in the accidentally smart city.

Chapter V

220221_CritViz_Book_Full_revisions.indd 175

Critical Visualization

7/20/22 9:47 AM

176

This picture is elaborated by the anthropologist Anna Lowenhaupt Tsing’s evocation of the precarious livelihoods and environments of twenty-­fi rst-­ century capitalism thwarted by ecological catastrophe: in place of a dream of continuous modernization and progress is a picture of ‘patchy’ capitalism in which ‘the concentration of wealth is possible because value produced in unplanned patches is appropriated for capital’ (Tsing, 2015, p. 5). Tsing’s book, subtitled ‘the possibility of life in capital­ist ruins’ follows the trail of the matsutake mushroom, which grows in human-­ disturbed forests and is said to have been the first living thing to grow in the blasted landscape of Hiroshima. In light of patchy capitalism and a situated mapper, it seems p ­ ertinent to consider in closing how data visualization is providing support for desirable shifts in behaviour. In the simplest terms, public displays of real-time data are an increasingly common means of representing ­statistical trends to urban citizens, and, in some cases, a trigger to escalate those trends. Nuage Vert by artist group HeHe, the winner of Ars Electronica in 2008, détourned the cloud emerging from a chimney at an electricity plant, turning it into a projection surface. As energy was saved by city residents, larger areas of the cloud were lit up by green laser light, functioning as a motivational visualization. Arguably, more actionable is the bicycle barometer, a real-time visualization that clocks how many cyclists have passed the sign that day and year, launched in Odense, ­Denmark in 1999, and is now installed in cities around the world. Its function is simple: to motivate more people to use bicycles (­k nowing that they are not alone) and to equip policymakers with data to ­inform decisions around cycling infrastructure. In ­effect, the bicycle barometer uses the ‘nudge’ theory of ­behavioural design ­documented and ­advocated by Thaler and Sunstein (2009), which became the basis of a British ­Government unit in 2010. In their survey of practices in the public visualization of urban data, a useful distinction is made by Andrew van de Moere and Dan Hill (2012), between three aspects of urban visualization: functional visualizations that serve to provide a ‘trustworthy and persuasive’ experience to a large audience, informative visualizations that create a direct feedback loop between the actions of city dwellers and their environment, and situated visualizations that take into account the characteristics of their physical location, reflecting on relevant local issues. The bicycle barometer, curiously, satisfies all three of Moere and Hill’s criteria, in providing trustworthy local information that creates a direct feedback loop.

Data and the City

220221_CritViz_Book_Full_revisions.indd 176

7/20/22 9:47 AM

177

Figure 5.09 — For Nuage Vert, emissions from a Helsinki coal-burning power plant (2008) and a Paris incinerator (2010) were illuminated with a high-power green laser animation. HeHe/Antti Aahonen

In conclusion, it would seem that the missing ingredient in smart city ­rhetoric, is one that takes on board the topic of everyday life as a site of potential political transformation, as articulated by Henri Lefebvre: ... it is in everyday life and starting from everyday life that genuine creations are achieved, those creations which produce the human. (1961/2008, p. 31)

In his book Down to Earth, Bruno Latour (2018) argues for a r­ edefinition of politics as what leads toward the earth rather than the global or ­national. Ecological degradation on a planetary scale has become increasingly visible, he writes, its effects visible in three simultaneous political phenomena: vertiginous escalation of inequalities, massive deregulation, and climate change denial. One effect of ecological degradation is that the historical colonizer and the colonized are thrown together in something like a new, bizarre bond, replacing the classic conflict. The colonizer asks: “How have you managed to resist and survive? It would be good if we too could learn this from you.” Following the questions comes a muffled, ironic response: “Welcome to the club!” (Latour, 2018, p. 23)

Historically, colonizers have often depended on Indigenous peoples for survival, only to disavow their dependence in order to displace them. But Latour’s belief that the climate emergency casts a new light on this relation­ ship underlines exactly why decolonizing design is both an imperative and awkward, humbling project for Westerners, as we try to turn our critical tools on ourselves, embrace a praxis-based learning and invite those we have previously excluded, to join us in the project. The distrust and weariness at our relentless narcissism can only be countered with the reminder that we’re all in this together. █

Chapter V

220221_CritViz_Book_Full_revisions.indd 177

Critical Visualization

7/20/22 9:47 AM

Heath Bunting — Status Project

178

Focus Figure 4.1 — Status Project — A 1003 — a Terrorist. Heath Bunting

Heath Bunting’s ongoing Status Project (2008–present) comprises a growing catalogue of lists and network diagrams that map the conditions required for a person to be classified with a particular identity (e.g. a terrorist, a homeless person, a transgender person, a citizen of a Schengen agreement nation state). The complex but austere diagrams hide a ‘wayward disposition’ as Colin Perry (2010) has observed, probing the relationship between a person, their class and their constructed identities. Bunting contends that the capacity to invent additional identities is a product of class privilege. While upper-class human beings often possess multiple identities and organizations (which Bunting refers to as natural and artificial persons), a homeless person, for example, will typically possess only one identity. As such, the Status Project aims to pro­vide a guide to help people navigate the structures within which status is managed and produced, and, potentially, create an identity that facilitates movement from, say, a homeless person to a bank manager. As Bunting told Annet Dekker (2011), ‘The vast majority of people share the same rights,

Rethinking the Representation of Data 220221_CritViz_Book_Full_revisions.indd 178

7/20/22 9:47 AM

Focus

179

Focus Figure 4.2 — Detail of Status Project. Heath Bunting

but they don’t exercise them, because they are convinced that they don’t have the same rights’. The project was catalysed by a com­ mission from the Tate in 2010, follow­ ing Bunting’s experience of being classified by UK authorities as a terrorist, in connection to a bomb hoax. Bunting believes that he was framed by security services because of his prior, well-known project BorderXing, which documented attempts by Bunting and fellow artist Kayle Brandon to cross European borders without official papers. BorderXing provided an online guide for would-be undocumented migrants: detailed routes, photographs and anecdotal knowledge (‘don’t sit on barbed wire in cold weather’). The Tate-commissioned ‘Terrorist’ map in Status Project sets out a

comprehensive account of Bunting’s efforts to understand what conditions are necessary for a person to be classified as a terrorist — and, by the same token, declassified. Incorporat­ ing Bunting’s own direct experience of the heightened level of state surveil­ lance in the wake of the terrorist attacks of September 11, 2001, the project is both a personal and a political mapping. At points, Status Project exposes the less-than-watertight conditional logic of the decision trees it maps: if a person  is classified as a blood donor, then he or she is ‘not HIV positive, an injector of drugs, taking antibiotics, a prostitute, gay, or less than one year from having a  piercing’ according to the map (Amoore,  2006). This logic is then exploited by Bunting, notably at associated work­ shops at the Tate, in which participants 

Heath Bunting 220221_CritViz_Book_Full_revisions.indd 179

7/20/22 9:47 AM

180

Focus

construct their own identities. The Tate’s own identity is subject to the same Hobbesian logic as other actors. ‘The Status Project could be an extension of institutional critique’ says Tate research fellow Lucy Bayley (personal communication, February 14, 2020), who observes that Bunting’s work reflects on the role of the troublemaker in society, a figure associated with those lower down in class hierarchies, whose work both uses and reflects on institutions such as the Tate. Bunting is candid about his exploitation of a position at a threshold between institutional, covert and autonomous spaces: ʻas an artist you have to exist in these spaces, be secure, unstable, upset everybody including yourselfʼ (H. Bunting, personal communication, July 24, 2019). This threshold position is illuminated by Dieter Thomä’s book-length account of the ʻpuer robustusʼ, a figure ­identi­ fied by Hobbes that has recurred through history: ʻThe puer robustus who roams around on the threshold is not caught between two different political borders; instead, he moves along the edge of a single world that is defined by the reach of its power. This edge is not a different place, it is actually a non-place. The puer robustus does not belong in one place or the other; he is the very epitome of non-belongingʼ (Thomä, 2019, p. 5). ●

Rethinking the Representation of Data 220221_CritViz_Book_Full_revisions.indd 180

7/20/22 9:47 AM

220221_CritViz_Book_Full_revisions.indd 181

7/20/22 9:48 AM

VI  B  eyond Aesthetics 

and Representation

220221_CritViz_Book_Full_revisions.indd 182

7/20/22 9:48 AM

A

v isua li z ation or m a p goes beyon d bei ng a n obj ect t h at represents data  — it also works to aggregate, simplify and distil. It is a tool that performs a function that helps us think through data. It is not passive, but actively engages our cognition. In this sense, visualizations have a given agency — they can affect how we perceive the world. Charles Joseph Minard, a civil engineer credited with advancing the science of visualizing data, points to this form of interaction between human cognition and the work that maps do: ‘I have heard, regarding my maps, that for a long while we have been making talking maps; not only do my maps speak, but, moreover, they count, they calculate with the eye; this is the main point; this is the improvement which I have introduced into my thematic maps by the width of zones, and in my graphic tables by rectangles’ (Minard, 1862, p. 4) (Translation).8

183

Visualizations act on our cognition and on our sense of the world. On a basic level, human cognition is channelled through the eyes working together with graphic shapes. On a higher level, they gather data and make connections and correlations. Of all the different ways that we sense the world, ­visualizations are a powerful prosthetic to summarize and view a wide range of phenomena that are either too far away, too small, too large, too fast or too slow for us to perceive unaided. Visualization performs a paradoxical role of making visible selected aspects of data drawn from our reality, while simultaneously making invisible the way the data was produced. How this dual role of giving voice and silencing is ethically carried out is part of the mandate of a critical visualization practitioner. In addition, the designer translates the data into a visual treatment that makes shapes in the data easier to see and imbues specific ­connotations. These are important functions that can empower or alienate, and for this reason, we need to think of aesthetics and representation in terms of what is made available to our perception and who or what is represented — not just the visual form or the translation of data to visual form. These two concepts have significant overlaps, but we will tease out defini­ tions and examples of aesthetics and representation while also marking many connections between the two along the way. We will conclude by making a case for considering them together in order to better understand

8 Original: « J’ai entendu dire, à l’occasion de mes cartes, qu’il y avait bien longtemps qu’on avait fait des cartes parlantes ; non-seulement mes cartes parlent, mais, de plus, elles comptent, elles calculent par l’œil ; c’est là le point capital ; c’est là le perfectionnement que j’ai ­introduit dans mes cartes figuratives par la largeur des zones, et dans mes ­tableaux graphiques par les rectangles. »

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 183

7/20/22 9:48 AM

184

critical visualization practice. Rather than treat them separately, it is more productive to move between one and the other in order to elaborate their definitions in relation to each other. If we look to its ancient Greek root, aesthesis relates to ‘the percep­tion of the external world by the senses’ (­Oxford English Dictionary, 2020). This is a useful expansion of the term, as it helps us look at how beauty is used to engage the viewer, how perception is managed to augment cognition, and what is made perceptible or visible. This links significantly with the idea of representation in terms of how we make sense of the world based on what is (re)presented and what is not. In turn, representation has a further complica­tion in that in an ideal sense it aims to re-present, that is, present again  —  a promise that is fraught with problems of contextualization and translation. This has led to the following discussion having a meandering quality, because it attempts to touch on aspects of philosophy, art, sociology, cogni­ tive science, social justice and culture, while tying together these two related concepts of aesthetics and representation. For instance, computer science and cognitive science have subfields dedicated to understanding how we sense and interact with visuals to aid our cognitive work. Related research (Bederson and Shneiderman, 2003; Stuart K. Card et al., 1999; Tufte, 2001; Ware, 2004) and cartographic research (Bertin, 1967/2011) have also developed insights on the design of the visual symbols, ­codings and arrangements that best represent the data and which are most decipher­ able by a human user. These provide useful starting points to evaluate the feedback loops between thinking and seeing. But we also need to move from the lab to the field where history, context and agendas ground the ways that visual representations are used to help us think and act. A E S T H E T IC S AND F U NC T ION Conventionally, when we hear of aesthetics, the primary reference is to notions of beauty, style, and of surface or form. Beauty is often defined in terms of purity, elegance, simplicity. These are culturally based notions of beauty that are built on top of particular tendencies in human perception. Designers and artists work with and against these conventions in order to produce particular meanings, to contextualize, and to attract and persuade a viewer or user. The simplicity of lines drawn with pencil and paper, like those of Mark Lombardi, may appeal to the gracefulness of a steady hand drawing a diagram. The complexity of ­Bureau d’études’ World Government diagrams work in an opposite manner, where a dense and overwhelming number of entities are captured, ­enumerated and linked.

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 184

7/20/22 9:48 AM

185

Figure 6.01 — Gouvernement mondial (2004). Bureau d’études

Figure 6.02 — Gouvernement mondial (2005). Bureau d’études

Chapter VI

220221_CritViz_Book_Full_revisions.indd 185

Critical Visualization

7/20/22 9:49 AM

186

Figure 6.03 — Gouvernement mondial (2013). Bureau d’études

Figure 6.04 — Detail of Gouvernement mondial (2013). Bureau d’études

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 186

7/20/22 9:49 AM

187

Figure 6.05   —   Detail of Gouvernement mondial (2013). Bureau d’études

It has become a truism in recent years that beauty is contingent on the viewer’s own context, priorities, tastes and cultural backgrounds. As ­Umberto Eco notes, ‘The mass media no longer present any unified model, any singular idea of beauty’ (2004, p. 426). This aligns with the ­dichotomy where subjectivity is valued less than objectivity — where the former is variable and deceptive, and the latter is constant and true. In their panoramas, Beehive Design Collective uses intricately hand-drawn figures to tell their story. These are meant to be pleasurable for the ­viewers while the stories told are of struggle, resistance and violent oppression. There is a beauty to the story maps that helps engage the audience in much the same way as Mexican muralists (e.g. Rivera, Orozco, Siquieros) use composition, scale, colour and figurative drawing in complex ­panoramic murals to address people in different spaces (i.e. pedestrians, visitors, workers). Aesthetics as beauty are necessary dimensions of e­ xperience. As bell hooks (2007) reminds us, aesthetics and beauty bring pleasure and are necessary for surviving poverty and hard living.

Chapter VI

220221_CritViz_Book_Full_revisions.indd 187

Critical Visualization

7/20/22 9:49 AM

188

Aesthetics is also frequently disparaged as being solely a matter of subjectivity; taste involving the use of style for making form or decorating a surface. It is also gendered, as aesthetics in the form of beauty products and services is largely focused on feminine gender expressions. In fields such as software development, still a male-dominated sector, it was once popular to refer to the visual design of a software interface as the ‘lipstick’. The dismissal of the aesthetic as not deep, real or functional is supported by legacies of patriarchy, colonialism and racism. Aesthetics in the form of decoration were repudiated by Adolf Loos (1908/2002), for instance, on the grounds that ornament was wasteful and backwards. ­Functionalism, as ­expressed in the mantra coined by architect Louis Sullivan (1896), ‘form ­follows function’, sought to replace the language of surface effects with a more essential and pragmatic source of utility thereby creating a direct correspondence between original function and its formal representation. Yet aesthetics as visual style cannot be so easily detached from function. It is used more instrumentally to appeal to a given community of viewers and link the work to other forms that carry meaning. Sometimes referred to as ‘visual language’, the style of presentation is wielded like a tool to address a viewer or user. It can be used to locate the image in a particular tradition, history or ter­ ritory. These are visual codes that are meaningful to specific communities of people and require knowledge to understand. We may connect very deeply to an image because of this shared understanding between maker and viewer. Drawing style can also be used to make the data representations more ­approachable to a viewer. It can remind the viewer that the map or visuali­ zation was human-made and therefore subject to bias or error. For ­instance, J. Wood et al. (2012) found that participants in an experiment were more willing to engage with and critique visualizations that had a hand-sketched style rather than a clear line style.

Figure 6.06 — Bar chart displayed using the Handyrenderer for Processing. J. Wood et al. (2012) found that participants in an experiment were more willing to engage with and critique visualizations that had a hand-sketched style, rather than a clear line style. Wood et al.

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 188

7/20/22 9:49 AM

189

Figure 6.07 — Map showing bicycle availability using Handyrenderer, J. Wood et al (2012). Wood et al.

Visual style can therefore communicate certainty, or in some cases, ­authority. Clean lines and a minimalist aesthetic may look more scientific and objective. The modern minimalist aesthetic epitomizes a modern attachment to universalism and function devoid of culturally specific ornamentation. This coincides with a communication model (formalized later by Shannon and Weaver, 1964) that prioritized the integrity of the message and aimed to reduce noise in the signal. Elements that are extraneous to the essence of the message are seen as either noise that makes the message less clear or, worse, communicates its own cultural specificity which runs counter to a virtue of universal accessibility. This was taken up within c­ artography research in the first half of the twentieth century as a scientific way of ­improving cartographic communication; that is, the efficient transmission of knowledge through a map medium. Cartography communication science (MacEachren, 2004) had great appeal to many in the field, as it promised an objective and testable way of increasing the efficiency of spatial data representation — an advantage to planning and analysis for a variety of potential users (e.g. government, industry, science and military). Cartographers employed a minimalist visual language and the simplicity of lines and geometry to lend an air of objectivity, universality and ­clarity. The sparse treatment of visuals suggests a more direct correspondence between the data and the representation, and less human involvement. It communi­cates that the image has been reduced to its bare minimum. It has been polished through successive passes to remove the unnecessary and the contingent. And in doing so, it indicates something essential and closer to a transcendent type, or perhaps an ideal.

Chapter VI

220221_CritViz_Book_Full_revisions.indd 189

Critical Visualization

7/20/22 9:49 AM

190

The minimalist aesthetic is at least partially rooted in the notion that there is a direct relationship between the visual stimuli and a u ­ niversal 9 or physiologically based response. Gestalt principles of visual percep­ tion hold that the basic visual forms can have a cognitive effect and that a certain kind of simplicity is desired by the human eye (e.g. law of ­pragnatz) that spans cultures, age, gender, etc. Gestalt p ­ sychology (­initiated in the early 1900s by a group of German psychologists ­including Max Wertheimer, Kurt Koff ka and Wolfgang Kohler) made a set of laws of perception still used today as visual design principles that help describe how humans perceive visual phenomena (Ware, 2013). Gestalt principles of continuity and proximity, for instance, are useful in picking out a trend or cluster in the data. Allowing these basic visual-­ cognitive processes to occur would require stripping away all but essential elements representing data. Messiness or extraneous visual elements would potentially introduce noise and hinder the process of pattern recognition. Unsurprisingly, the minimalist aesthetic continues to be very popular in contemporary visualizations, especially in areas such as scientific visualization, data journalism or business analytics. We can speculate that the endurance of this aesthetic, apart from a certain amount of efficiency of communication and the establishment of a convention, is also due to the desire for producers of visualizations to communicate a level of authority, seriousness and certainty in the data being represented. The use of a minimalist and functional aesthetic evokes objectivity, univers­alism and scientific rigour. These visual traits in visualizations have been shown to deter critique and imbue the image with more authority. On the ­other hand, experiments have shown that sketchy (i.e. hand-drawn) visualizations can shift the viewer’s perception of the data. Wood et al. (2012) hypothesize that the imperfect lines of the visualizations in a series of experiments may reinforce the intention of an author, portray an informality and simplicity to the work, suggest an u ­ ncertainty to the claims made, or even attract the viewer due to its artistic style. This would indicate that a sketchy rendering of shapes in a visualization would potentially encourage a viewer to engage more ­critically with the claims of the diagrams. By imitating the (human) ­process of making, the visualizations express a degree of uncertainty.10

9 Gestalt basically means pattern (Ware, 2013). Drucker summarizes Gestalt as certain principles in human perception: proximity, similarity, closure, continuation, common fate and good form (2014, p. 41).

10 See MacEachren et al. (2005) on representing uncertainty in geospatial visualization and Lombardi’s (2004) use of pencil in his diagrams to connote uncertainty.

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 190

7/20/22 9:49 AM

A useful contrast to the more conventionalized ‘clean’ form of visualization is the work of Philippe Rekacewicz, which is often hand-drawn. His Cartes de Colère (Rekacewicz, 2013b) clearly represent complex information, but it is evident that the coloured pencils and hand-drawn borders evoke the sense of everyday creativity and graphic production with tools that are relatively easy to acquire and use. Although the information represented in the maps required a significant amount of knowledge and research to produce, the presentation of it suggests that it is accessible and should perhaps be commonly known.

191

Figure 6.08 — ‘ Sanctuarization of the world’ by Philippe Rekacewicz, visionscarto.net — this map was created and drawn for the exhibition ‘Frontières’, musée des confluences, Lyon, 2006. Philippe Rekacewicz

What is implied most of all in the aesthetic of Rekacewicz’s maps is the process by which the work is produced. The human hand, and by extension the human mind, has created this representation of reality. It is not the bodiless, Apollonian view from nowhere, that is signalled with the drawings, but an unfinished work that may contain omissions or errors ­because of its all-too-human origins. This is generally unusual behaviour

Chapter VI

220221_CritViz_Book_Full_revisions.indd 191

Critical Visualization

7/20/22 9:49 AM

192

for a cartographer. John Pickles notes how a map makes a ‘representation­al claim [that] was always predicated on making the many specific decisions, compromises, adjustments, generalizations, and slippages that characterize all knowledge production disappear. That is, it erases the work the map actually does’ (Pickles, 2006, p. 349). It is the clean aesthetic of maps that intensifies the erasure of the messy traces of human work and judgement performed. Framed in the influential terms of Shannon & Weaver as a communication process, the visualization should represent the ­concept or number (i.e. clarity of signal), rather than display the residue of its ­construction (i.e. noise in the channel). A E S T H E T IC S A N D PE RCE P T ION Visual representation is the translation and amplification of how a site is made legible to a viewer or user. We represent the data in visual form in order to help our eyes make connections and see patterns. As information visuali­zation researcher Ben Shneiderman (1996) states, ‘The eyes have it’ — the eyes make great use of visual form to help gain insights. As such, this form of representation also affords an aesthetic experience —­ one in which the body, through the eyes, senses shapes, distances, colours and patterns. Beyond Gestalt principles of visual perception subsequent ­psychophysics research has discovered how preattentive processing works to helps us distinguish between different visual stimuli (Ware, 2013). The e­ xperience of preattentive processing is that of immediate recognition of a s­ patial arrangement of visual elements (i.e. the grouping of similar objects, ­connecting nodes in a network). This is used as a building block for ­designing visuali­ zations and it is in this way that Shneiderman claims that the eyes do much work in finding shapes within data visualization. Framed as a support for thinking, visualization research has ­largely focused on the technical and cognitive aspects of the mental tasks of ­analysing data by visual means. Work in the form of cognitive load is ­understood as a quantity that needs to be reduced so that cognition ­increases in capacity, speed and efficiency. Take for instance the ‘Visual Information-Seeking Mantra’, which describes an optimum way of interacting with data: ‘overview first, zoom and filter, then details on demand’ (Shneiderman, 1996). This is a useful rule-of-thumb when thinking about how a user might best work with a data visualization with an interactive interface. Managing the appropriate amount of ­relevant information is at the heart of this design process. What does the user need in order to find the information or the pattern in the data? How does the

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 192

7/20/22 9:49 AM

user understand details within the context of the large data set? How can the user easily compare different data points? How does the design minimize irrelevant data (i.e. noise) and maximize relevant data (i.e. signal)?

193

The focus on developing a more efficient interaction between the user and a visualization echoes the communication paradigm of ­cartographic ­science discussed earlier. Developing better ways of representing data types and structures so that humans can search and find information has been a core focus of much scholarly research in information visualization. What is often left out is the context and meaning generated at the site of produc­tion and the site of reception of a visualization where the sender and re­ceiver are located. It is useful to note that MacEachren, a one-time proponent of a cartography-­ as-communication paradigm, has since critiqued the narrow focus of this approach and has instead argued for a mixed method that acknowledges the technical precision and potential advancement in design of maps with the social, political, cultural aspects of maps that prioritize interpretation and context (MacEachren, 2004, p. 12). To advocate for a mixed method, MacEachren makes an important distinction between communication and representation. Where communication is a technical problem between a sender (cartographer) and receiver (user) through a medium (map), a representation is a non-linear process in which meaning is constructed through an interaction among components such as designer intention, map, map context and user need.

Figure 6.09 — A lan MacEachren’s 1979 view of cartography as a process of graphic communication. Redrawn from MacEachren’s 2004 book How Maps Work

Chapter VI

220221_CritViz_Book_Full_revisions.indd 193

Critical Visualization

7/20/22 9:49 AM

194

R E PR E S E N TAT ION A S T R A NSL AT ION Representation takes place through translation. An example would be the translation of a living thing into an image of the thing. If we think of a territory translated into a map, we can see that the endless complexity of a living system can be translated into a static two-dimensional image. These translations have a great ability to be copied, distributed, taken apart and reassembled — all while having a claim to representing the original territory. Latour developed the term immutable mobiles to describe the documents, maps, illustrations, charts, etc. that point to a thing in the world (e.g. the contours of an island, the temperatures recorded during a scientific experiment). The data and the accompanying visual representations are mobile in that they can be shared widely and at relatively low cost. They can also be collected and assembled into larger documents that stake a claim of knowledge over a thing, a place, an event or a people. In visualization, representation starts not in the visual artefact but in the counting and measuring. In other words, quantification is representation (Stray, 2016). It is a process designed to select part of a complex field of interacting phenomena and translate it to a stable and computable form. In other words, before there is visual representation there is conceptual representation. If we extend the range of meanings of representation we can also see its political dimension. Representation is also a form of administration when the things counted are populations of living beings. Just as in political representation, where someone is a representative of another person or group of people, representation in data visualization or mapping is a stand-in or proxy for the thing we are inspecting. James Corner (1999) describes a double-sided character of mapping in which there is an analogous relationship to physical space and an abstract (or conceptual) relationship. This is the power of maps to both allow us to point at a feature that feels naturally to be the place we may inhabit while also allowing a play of concepts that far outstrips this physical reality — deep history, stories, people and connections to other territories. The bridge between aesthetics and representation reaches between what is sensible and what/who is made available to the senses through visualization. Given that historically, mapping, visualization and quantification have so often been used for the purpose of command and control, what do critical practitioners do with this legacy? Some play with the utility of the map (e.g. Jorn and Debord). Others use critical maps to combat dominant maps and dismiss essentialist ideas of the map as a tool of domination (e.g. ­Bureau d’études, Philippe Rekacewicz). Still, others develop a c­ ritical practice in visualization which acknowledges that the objects created

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 194

7/20/22 9:49 AM

195

Figure 6.10 — a nt. Marc Ngui (see p.61)

Chapter VI

220221_CritViz_Book_Full_revisions.indd 195

Critical Visualization

7/20/22 9:49 AM

196

­extend the voices of those represented. How the ethics of this representation is cared for becomes an important part of the visualization assemblage (e.g. Anti-­Eviction Mapping Project). Visualization’s role in colonization is countered in order to decolonize knowledge of territory (e.g. Margaret Pearce) — this is discussed more in the final chapter. Cartography, and information and data visualization are ­fundamentally representational practices. They translate concepts, hidden attributes and invisible quantities into visual form. They re-present data in order to construct facts. By extension, they also re-present things and people that produced the data. For this reason, representation for critical visualization practice needs to be thought of in at least two important, interrelated ways. First, the translation of a thing or person into another form (e.g. the datum or the visual symbol). Second, the translation of presence and power from a thing or person into another sphere (e.g. political). While one focuses on the visual form used to depict data, the other focuses on the articulation of a power relationship used to generate that visualization. Aesthetics plays an important part in the process of visualization. One dimension of aesthetics is in terms of beauty, elegance and d ­ esirability. These are important attributes for the rhetorical function of ­visualization. Another dimension, as a term recuperated by Rancière (2010) from a more archaic notion of aesthetics, is the sensible or what can be sensed. This aligns in part with the process of perception. A central concern in the field of information and data visualization has been the design of tools to aid human visual perception of data and thereby augment cognition. Yet Rancière is more concerned with the politics of how what is available to be sensed is distributed in society — that is, what we are allowed to sense. If applied to what artists and designers do with visualization, this prompts us to think about what images are available to use to help us understand and feel. Think, for instance, of the Peters Projection (see Chapter 2), which is intended to represent the landmass of continents more accurately, thereby recalibrating the relation of power between former colonies and the centres of empire. The intention was for this map to be used in classrooms in the s0-called Third World where children could sense and think through this representation (Ashmore, 2003). This map is a good example of the problem of visualization in general. The old saying, ‘All maps lie flat, therefore all maps lie’ (Henrikson, 1999, p. 98) reminds us of the compromise made in making a projection. The projection of a sphere onto a flat surface — the main problem of ­representing the globe on paper or on a screen — requires that certain distortions take place. Some areas of the globe will be shrunk and others expanded ­resulting in an inaccurate size relationship between some countries.

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 196

7/20/22 9:49 AM

197

Figure 6.11 — Example of Mercator map projection. Daniel R. Strebe, licensed with CC BY-SA 3.0.

Peters was a historian with a keen interest in the way that geography represents the world to itself and can thus be used to promote specific political interests. Brotton (2014) notes that the Peters Projection was massively popular around the world immediately after its publication despite the harsh criticism it received from the professional cartographic community. More recently, the map projection’s enduring popularity is evidenced by public schools in Boston, usa replacing their Mercator maps with Peters maps (Walters, 2017). One of the problems of representation is the impossibility of true representation. The representation always exists next to the original, after the original, and in a different form than the original, therefore betraying it in the process. Yet we often expect the representation to be ‘true’ to the original and judge it based on its fidelity. It is a judgement of quality and of ­morality that results in a representation being good or bad. This is especially true with visualization. Representation is thus part of a process of creating a correspondence between the source and the proxy. For ­instance, when Munster (2009) states that ‘what we see as patterns, visualizations and diagrams are the perceptible end of data’, she conjures a linear trajectory between the generation of data and its representation. This is also reflected in how Edward Tufte (2001) implores designers to let the data speak in order to be as faithful to the original source as pos­sible — to minimize visual confections so that the viewer can grasp the shape of the data as quickly as possible.

Chapter VI

220221_CritViz_Book_Full_revisions.indd 197

Critical Visualization

7/20/22 9:49 AM

198

In this formulation, to be a good visualization designer is to get out of the way of the data, or at the very least simply provide a neutral conduit through which to convey the informational value of the collected data. This would suggest that the data is close to objective reality and is thus raw. Yet, as Laura Kurgan reminds us, ‘There is no such thing as raw data. Data are always translated such that they might be presented. The i­ mages, lists, graphs and maps that represent those data are all interpretations. And there is no such thing as neutral data. Data are always collected for a specific purpose, by a combination of people, technology, money, commerce and government’ (Kurgan, 2013, p. 35). Kurgan goes further still and suggests that ‘the phrase “data visualization” in that sense, is a bit redundant: data are already a visualization’ (2013, p. 35). In effect, there is an impossibility to representation — meaning to re-present or present again, as if the original is simply made available again to a viewer. Context is always shifting and translation, by device and by human agency, from one mode to another, results in the transformation of the original via the omission of data, the introduction of other data, and the contextualization of the data. Although forms of visualization and diagramming have been present in cultures around the world for millennia (Drucker, 2014), they have found perhaps their most powerful expression as a particularly modern activity that aids communication and centralized control. This activity is, in turn, based on the generation of data in order to ‘see’ things that are normally difficult or impossible to see with the naked eye. Arjun Appadurai calls this a numerical gaze (1996b, p. 121), which has a representational effect on knowledge of people and place. In the form of demography, Appadurai states that ‘statistics are to bodies and social types what maps are to territories: they flatten and enclose’ (1996b, p. 133). In a cultural context where the rationalization of labour, social connection, leisure, consumption, etc. is taken for granted, visualization finds a natural position to express the quantification of various dimensions of life. It is the tool used to demystify social complexity and instrumentalize the insights drawn from it. This modernist agenda to quantify has been analysed in a variety of ways in critical theory and sociology (Adorno and Horkheimer, 1947/2009; Appadurai, 1996b; Scott, 1998), and it has been shown as an intentional tool to command and control that effectively disempowers, alienates or dehumanizes its subjects.

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 198

7/20/22 9:49 AM

The power to control bodies through knowledge structures and policies requires categorization and quantification. Indivi­dual persons become an essential unit for calculation and comparison. Collectivities or groups associated with a territory were not compatible with how a colonial power sought to govern.

199

Appadurai (1996b) notes that numbers and counting have a referential role in representing a given territory and its people to the British colonial state. They also had a rhetorical role that allowed comparison between things that were otherwise very different, and enabled concise, powerful compression or reduction of a variety of social characteristics. He states that ‘numbers permitted comparison between kinds of places and people that were otherwise different, that they were concise ways of conveying large bodies of information, and that they served as a short form for capturing and appropriating otherwise recalcitrant features of the social and human landscape of India’ (1996b, p. 120). While Appadurai’s analysis was focused on the surveys of India ­under ­British rule, his insight can be applied to visualization as the visual manifest­ ation in a process of making and presenting data. Reducing things first to number and then to simplified shapes further allows for the comparison of things — especially in the service of administration. Representation, as a goal, is predicated on a notion of correspondence between the real and the image. Or the person and the representative. From a critical visualization designer’s position this may take the form of representing one’s relationship to the whole — the individual to society or history. Frederic Jameson argues that our inability to ultimately resist the oppressive power of capital is partly due to our inability to cognitively map our reality at a scale that can account for its level of complexity and scope of time and space — what he calls representability (1992, p. 4). Representability is a central question in critical visualization practice because we are concerned with how to record and communicate the ways in which oppression takes place. Colonialism and capitalism are networked systems of oppression that dislocate actors between a centre and a periphery, which privilege individual action and choice rather than solidarity and allyship. Jameson spurs us to develop an aesthetic of cognitive mapping that is up to the task of representing a social totality such that we do not imagine ourselves isolated, disconnected and powerless in the face of global networks of power. Jameson states that, ‘It is a question which necessarily opens out onto the nature of the social raw material on the one hand (a raw material which necessarily includes the psychic and the subjective within itself) and the state of the form on the other, the aesthetic technologies available for

Chapter VI

220221_CritViz_Book_Full_revisions.indd 199

Critical Visualization

7/20/22 9:49 AM

200

the crystallization of a particular spatial or narrative model of the social totality’ (1992, p. 4). It is this totality that is not accessible at an ­immediate sense level due to its incomprehensible scale. According to Jameson, an aesthetic of cognitive mapping is necessary for an emancipatory politics (1988/2000, p. 283). The scale of the social totality under global capitalism that Jameson identi­ fies is so beyond immediate experience that it defies representation. This impossibility is akin to the sublime — the thing that surpasses our human ability to completely grasp. Mapping and visualization is the attempt to represent this type of entity. New media theorist Lev Manovich called data visualization the anti-sublime precisely because it tries to contain a totality that is beyond our human capacity to sense. It should be noted that Jameson’s notion of cognitive mapping was used to analyse works of art and literature. In fact, he dismissed the attempt to think this through actual cartographic (or visualizing) terms, ‘cognitive mapping cannot (at least in our time) involve anything so easy as a map; indeed, once you knew what “cognitive mapping” was driving at, you were to dismiss all figures of maps and mapping from your mind and try to imagine something else’ (Jameson, 1991, p. 409). It is unclear exactly why Jameson dismisses actual maps, although Toscano and Kinkle (2015) point to their tendency to be fetishized. Perhaps they identify a tendency to treat visualizations as artefacts capable of magical feats of representation. Conversely, to more fully appreciate the ways in which these objects attempt some kind of visualization of a social totality, it is necessary to acknowledge how the practitioners engage the processes, methods, media, users and participants that make up the visualization assemblage. McKenzie Wark (2015) notes that the issue with Jameson’s cognitive mapping is that it relies on a logic of essence (theory) and appearance (aesthetics) where representation is the key. Wark also problematizes the premise that the philosopher (or artist/designer in our case) has a privileged position better suited to seeing the totality than the impoverished and inveigled everyday workers caught in the system. Aesthetic work is secondary to the work of theory in this logic. Instead, a logic of affect or embodiment may be better suited to apprehend the social totality.

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 200

7/20/22 9:49 AM

The separation of essence and appearance is sidestepped as an issue when the representation is not fetishized but rather taken up, contextualized, ripped apart, reassembled, narrativized, etc. Wark states that ‘what we need, then, are not so much cognitive maps (theory controlling aesthetics) as a new kind of proletkult, or a new kind of dérive, a new way of collectively experimenting in the act of mapping, as an ongoing practice, rather than an aesthetic work for contemplation’ (2015, para. 40).11 Wark points to a radical aesthetics that helps us feel and understand the social totality.

201

One historical example of artistic visualization that attempts this form of remapping is Guy Debord’s and Asger Jorn’s collage maps. They reassembled versions of a territory through copying and juxtaposing incon­gru­ ent parts. As part of the larger Situationist movement, they railed against the fashion of machine aesthetics and rationalist form (Sadler, 1998, p. 10). Their maps and diagrams of urban space drew on collage, sketching and writing, not so much to represent a given built environment, but to document a process of moving through the city and to project a form of destabilizing the form of the city. Coming into contact with the social totality was not the aim of these experi­ ments, but rather the aim was to create aesthetic experiences that disrupted convention­alized ways of navigating space. As assemblages of disparate locales that disregarded distances, these maps resonate with cognitive ­psychology ­researcher Barbara Tversky’s model of spatial memory as collage map (1993). Rather than a contiguous cognitive map that contains comprehensive information of whole areas of space, Tversky’s experiments suggest that we recall a territory in related chunks. While these maps are interesting strategies for working beyond a representational logic, it is their potential use that is more radical. Resistance is performed rather than contemplated. One of the ­tactics of resistance is education — to produce an awareness of one’s ­position within a social totality. In the Isotype work of Otto Neurath, ­Marie ­Reidemeister and Gerd Arntz, visual representation of complex inform­ation was used as pol­itical tool of public education, for a population with a wide range of literacy levels (see Chapter 2). As a form of popular e­ ducation, ­Isotype was concerned with making the complexity of ­economy, ­industry and ­demography acces­sible and navigable to a wider range of p ­ eople — ­especial­ly non-experts. Isotype, for instance, at first showed their work in museums

11 Dérives, as defined by Debord (1956) are action/­ experience-oriented and not abstract/contemplative. It is a creative method by which normative logics and conventionalized experiences of urban space can be decentred in order to notice the ways in which the built environment frames our everyday experiences.

Chapter VI

220221_CritViz_Book_Full_revisions.indd 201

Critical Visualization

7/20/22 9:49 AM

202

(e.g. Museum of Society and Economy in ­Vienna). Many critical visualiza­ tion projects also function, or perhaps mainly ­function, as educational projects.

Figure 6.12 — 5 types of man. International Foundation for Visual Education, ‘Signs for the 5 groups of men’, in Otto Neurath, International Picture Language, London: Kegan Paul, Trench, Trubner. & Co., 1936.

In these visualizations, representational function was aligned with clear communication and simplification. For instance, the design of symbols in the Isotype system was intended to represent data in a way that could be efficiently read by a viewer. To this end, the familiar style of clean lines and minimalist presentation, which developed around Gerd Arntz’s linocut pictograms, dominated their work. If we compare this with Iconoclasistas images, we see a similarity in ­agenda but a shift in both methodology and the political imaginary. A ­similar ­visual aesthetic is shared between the Isotype and Iconoclasistas images — minimalist, mostly monochromatic, bold images. They therefore achieve a generic look with a utilitarian purpose rather than an expressive or idiosyncratic feeling. This style of drawing reinforces their status as tools for either understanding categories (Isotype) or communicating personal knowledge (Iconoclasistas). In contrast, Iconoclasistas takes the same minimalist aesthetic, yet more as an ironic evocation of the ­utilitarian, and promotes a non-linear relationship between image and identity, where

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 202

7/20/22 9:49 AM

combination and hybridity are emphasized. Iconoclasistas’ images attempt to represent subaltern identities (e.g. militant women, racialized workers, protestors, queer activists) — the people and experiences that are ignored by dominant narratives of a society.

203

Julia Risler and Pablo Ares, the duo behind Iconoclasistas, mix ­signifiers to thereby represent the constant hybridizing that occurs in human ­societies. The conical straw hat, for instance, is used to denote a ­person from Asia, both by Neurath and Iconoclasistas. While in the Isotype visualizations it was used to show one of the ‘five races of man’ — at the time a ­popular categorization of world ethnicities and reiteration of race pseudo-­science — in the hands of the Iconoclasistas, the images are ­elements to mix and match, and for participants to select if they speak to their experience or aspects of their identity. Other images from the Icono­clasistas collection include people wearing a hijab, a niqab, a bindi, a moustache, a ponytail or having curly hair.

Figure 6.13 — T hese icons were developed by Iconoclasistas to be used in community mapping projects, ­Diversidad / organización / articulación / ética del cuidado. Iconoclasistas

While the Isotype images reduce complexity to conventional c­ ategories (e.g. race, industry, labour, class), the Iconoclasistas images reveal complex­ ity (e.g. gentrification, diversity, popular resistance, precarity, extraction) — a complexity that is inherent in an intersectional analysis that shows how a multitude of systems impact individuals. To design representations of intersectionality in this way is important for the purpose of giving partici­ pants low-cost tools to animate discussions on local issues in their collect­ ive mapping projects. The images are also released under a Creative

Chapter VI

220221_CritViz_Book_Full_revisions.indd 203

Critical Visualization

7/20/22 9:49 AM

204

Commons license and can be freely downloaded with the caveat that they be used by communities or organizations focused on social movements or neighbourhood organizing. This is perhaps the radical aesthetics that works beyond the theory-aesthetics dichotomy in cognitive mapping identified by Wark (2015). Participatory action involving data and mapping is relational, not focused on contemplation so much as on specific goals for mobilizing knowledge and organizing resistance.

Figure 6.14 — Mobilizing knowledge and organizing resistance: Iconoclasistas workshop, 25 May 2019. Iconoclasistas

Beehive Collective takes an almost diametrically opposite approach to the map-communication-model that is still prevalent in dominant paradigms of data visualization. Rather than prioritize an efficiency of communication and an immediacy of visually coding data on a map, the group emphasizes process, slowness and participation. In the presentation of their visualizations, the collective often takes a teach-in approach where knowledge is shared orally and visually. Communities or collectivities are reproduced via participation in the event. Again, the relational is emphasized through the work of physically coming together and spending time. The collective’s narrative posters (i.e. visual story maps) are dense and overwrought with detail. This produces the effect of the viewer moving across a landscape and focusing on a detail where a local story emerges. The feeling is that everything is local and yet everything is connected. In Beehive Collective’s work, clarity and navigability of data gives way to comprehensiveness and the impression of complexity and interrelatedness. Also, the techniques used are those of artists either using experimental forms or poetic/metaphorical and illustrative techniques to represent ­actors, territories or relations.

Beyond Aesthetics and Representation 220221_CritViz_Book_Full_revisions.indd 204

7/20/22 9:49 AM

205

Figure 6.15 — T he Story of the True Cost of Coal Graphics Campaign. Beehive Design Collective

Figure 6.16 — Detail from The Story of the True Cost of Coal Graphics Campaign. Beehive Design Collective

Chapter VI

220221_CritViz_Book_Full_revisions.indd 205

Critical Visualization

7/20/22 9:49 AM

206

Figure 6.17 — Two Beehive Collective presenters perform a ‘Plan Colombia’ show at pnca in Portland, OR. Beehive Design Collective, licensed with CC BY 3.0

In a similar fashion, Bureau d’études, a collective formed by Léonore ­Bonaccini and Xavier Fourt, produces a variety of visualizations that map social-technical-political processes and networks of power. Instead of exclusively using figurative drawing, like Beehive, to represent actors and events in the narrative, Bureau d’études creates more diagrammatic maps of relations. Their maps don’t simplify a complex phenomenon, instead they are massively complex diagrams, such as the World Government series, which both invite the viewer to delve into the topic or to be overwhelmed by the relatedness of so many aspects of global power. The panorama form (which is appropriated in Bureau d’études’ maps) along with the god’s-eye-view has been critiqued as promoting an all-­ seeing or totalizing view of the world. The viewer is positioned at the best vantage point to strategize movements and calculate relationships between ­different a­ reas, and this form minimizes the on-the-ground or local ­perspective. This ‘­synoptic’ view replicates a command-and-­ control aesthetic reminiscent of military strategy, corporate planning and ­colonial ­expansion. Mid-twentieth-century pictorial maps like Hermann ­Bollmann’s Vogel­schaukarten, for example, replicated what Cosgrove characterizes as a ‘distanciated spatial vision’ of urban planning in European cities that p ­ rovoked the subsequent critiques and counter-mappings of the S ­ ituationists (D. Cosgrove, 2005, p. 41). In an interview (Dávila, 2019) Bureau d’études suggested that the nature of totalizing views or panoramas was less problematic than the absence of counter-mappings. In other words, if a set of representations that supports the status quo and dominant interests exists everywhere, then it is necessary to produce and distribute representations that counter it. What ­is available to the senses (in the form of visualized information),

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 206

7/20/22 9:49 AM

and ultimately to how we think and feel about our conditions of life, needs to be supplanted with critical voices that make available under-represented realities. Of course, these are supplemental to the knowledge gained by lived experience and offer the presentation of data outside of everyday experience through which we might connect more dots.

207

Along with Bureau d’études, Iconoclasistas’s A quien pertenece la tierra / Who owns the land? and Beehive Collective’s Mesoamérica Resiste both attempt to map an extremely difficult thing to represent — capital and transnational forms of flow and resistance. Perhaps what these visualizations are also trying to do is provide a ‘You are here’ indicator on maps typically seen in a mall in North America, or at international airports where the layout of the environment is designed to make you go through areas of optimum exposure to products and messaging (Rekacewicz, 2013a). Instead, these maps are meant to orient the viewer with a much larger entanglement — either networks of global corporate power (e.g. Bureau d’études), international scope of Indigenous resistance (e.g. Iconoclasistas), or the deep history of exploitation and resistance in Central America (e.g. Beehive Collective). Where the viewers choose to locate themselves acknowledges their complicity, or identifies the ‘structuring structures’ (Bourdieu, 1990) or the visible and hidden forces of oppression that affect their lives. Maps and visualizations are used for communicating, teaching and thinking. While they are often made for wide circulation so that specific ­knowledge is disseminated to a larger group of viewers, at other times they are made for the self-organization of a community. Grupo de Arte ­Callejero, an activist-­academic-artist group based in Buenos Aires, use the term ­pensamiento cartografico (trans. cartographic thinking) to describe how they use this form of representation in their own organization’s efforts. They state that ‘Estos mapeos tienen un valor hacia adentro del grupo, y no suelen mostrarse salvo a quienes participan de la acción’ (Grupo de Arte ­Callejero (­A rgentina), 2009, p. 40) (trans. ‘These maps are valuable within the group, and are not often shown except to people who ­participate in a direct ­action’). Maps are used for internal planning by activists, as they are supports f­or ­collaborative thinking: ‘Entonces, los mapas no son sólo un objeto de r­ epresentación, ni una acumulación de información: son también formas de ­ampliar la p ­ ropia mirada a través de un diálogo’ (Grupo de Arte C ­ allejero (­A rgentina), 2009, p. 42) (trans. ‘Maps are not only a form of representation nor a gathering of information — they are also ways of widening our attention through a dialogue’). In these cases, the community or group itself is the centre of the activity, rather than a researcher or designer.

Chapter VI

220221_CritViz_Book_Full_revisions.indd 207

Critical Visualization

7/20/22 9:49 AM

208

Iconoclasistas, while dedicating a good amount of effort to the production of comprehensive maps (e.g. critical mappa mundi), direct their focus to a­ spects of community-led mapping. In this way their aesthetics are concerned with the representational and the relational. We also see this dynamic in Beehive Collective’s projects which are produced over a long period of time, often with residents from the areas they are representing directly involved in the production (e.g. Detroit). They are narrative and explanatory diagrams, which don’t use geometric forms and labels but figurative drawings. They are designed for an interlocutor to work with audiences for the visualizations to come to life. They are not fully self-­ sufficient forms of communication, but rather tools to think c­ ollaboratively. Representation thus takes a turn towards performance and affect. B ­ eehive Collective creates communities of production and presentation for the work where the narrative map is inextricably attached to the process of ­visualizing. Through the work of these collectives, participation in the creation of visualizations is a form of political representation. They embody an ethos of ‘nothing about us without us’, where there is a greater correspondence between representations of a community and/or territory and the people who are part of it. For instance, Iconoclasistas’ visualization offers a great example of inclusion and participation for the purposes of representing knowledge of specific neighbourhoods. One of the reasons why groups like Iconoclasistas and Beehive Collective created visualization projects for communities is to combat dominant stories that continue to marginalize these same communities. They offer hidden histories and situate data within radical frameworks (e.g. feminism, indigeneity, resistance). Cultural theorist Stuart Hall developed a useful framework for understanding representation in media and popular culture that can be used to think about how we consume dominant images about ourselves. This can also be extended to think about visualizations and maps. For instance, Hall notes that representation has a notion of re-presenting something that was already there — as if it’s just presented again and therefore is the thing in itself (in fact, he argues, the process of representation itself constitutes the world it aims to represent). This is similar to how data also has a notion of being raw and as being an index of something already there — as discussed earlier in this book. Yet, representation occurs when meaning is given to that which is depicted either as contextualization or framing. In other words, both images and data have the effect of structuring the world they re-present (S. Hall, 1997).

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 208

7/20/22 9:49 AM

Hall explains that the ‘event’ has no meaning until it has been ­represented, therefore when we understand an event, representation has become an integral part of it. We can reapply this notion to think of data as having no meaning by itself until it is represented, framed, visualized, interpreted. Representation is often seen as a dichotomy between accurate or ­distorted (i.e. true or false), but it is more complex and better understood as a ­creative-active process.

209

We can also contrast this notion of representation with the communication model used in mapping and visualization. While in a simplistic communication model there is a focus on the correspondence between the original message and the received message, in representation, as elaborated by Hall, a message is necessarily interpreted in many ways and gains meaning not via the source but through the context of reception. The importance of Hall’s insight is that the process of meaning-making takes place through other cultural contexts and ideologies that impose structures of ­meaning that reproduce racism, sexism, homophobia, amongst other forms of oppression. It is not only the denotative content of an image (e.g. this is an image of person) but also the connotative level (e.g. this is an image of a hero). We can use Hall’s concept to analyse the trajectory of visualization work done by W.E.B. Du Bois, for instance. The maps and visualizations prepared by Du Bois and his team for the Paris Exposition in 1900 sought to fix the interpretation of what Black life is like in the United States for the audience attending the international event. The charts displayed as part of the American Pavilion at the exposition were intended to clearly show the world the progress of the African American standard of living after the civil war, as well as the high level of economic output and c­ ultural production. Du Bois states that he sought ‘to show: (a) The history of the American N ­ egro. (b) His present condition. (c) His education. (d) His ­literature’ (Du Bois, 1900). This exposition worked to combat white ­supremacist stereotypes, conceptions of difference and ongoing dehumanization of non-white people. Du Bois was concerned with using data and representing facts to ­address conceptions of life in the usa. Du Bois, in his essay reflecting on the expo­ sition, laments the poor state of scientific rigour in the ­description of sociological facts and therefore presents his own research to ­demonstrate methodological precision (Du Bois, 1900). This concern for the ­denotative accuracy in his charts and maps was coupled with his ­concern for the connotative power of his exposition. Du Bois displayed data ­gathered and visualized by himself and his collaborators, along with a ­collection of ­portraits of Black life, books and periodicals. In this way, his work was very innovative. In the presentation of demographic data as a form of public education, it predates the work of Isotype. Du Bois ­contextualized

Chapter VI

220221_CritViz_Book_Full_revisions.indd 209

Critical Visualization

7/20/22 9:49 AM

210

the charts on display with other forms of ‘data’ (i.e. ­images of the people represented in data, as well as other materials detailing the cultural and economic production of African Americans at the time). To adapt Hall’s terms, Du Bois’s data was put within a framework to combat a dominant reading of images of Black life in the usa. The data was situated within a counter form of intertextuality where other images supported a preferred reading of the images and information.

Figure 6.18 — Exhibit of American Negroes at the Paris Exposition. W.E.B. Du Bois

Du Bois’s team also created many unique forms of visualization that have since been lauded for their aesthetic. These charts were hand-rendered in ink and watercolour. They are effectively data paintings that remind us of their human construction in a pre-digital age. Theaster Gates, an interdisciplinary artist based in the usa, refers back to Du Bois’s charts in a series of his own paintings. Here, the numbers and text are omitted in order to show the graphic form alone. The utility of a chart showing numeric data is transformed into an aesthetic object that honours the work of Du Bois, echoing it back as African American visual culture and history. Gates contextualizes the paintings with an exhibition titled ‘But to Be a Poor Race’, after a quote from Du Bois’s The Souls of Black Folk, published in 1903, such that the viewer is presented with the possibility of experiencing a variety of archival media (e.g. magazines, records) much in the same way that Du Bois presented an archive of Black life in the usa at the turn of the previous century.

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 210

7/20/22 9:49 AM

211

Figure 6.19 — Occupations of Negroes and whites in Georgia. W.E.B. Du Bois

Figure 6.21 — Illiteracy. W.E.B. Du Bois

Figure 6.20 — T he Amalgamation of the white and Black Populations in the United States. W.E.B. Du Bois

Chapter VI

220221_CritViz_Book_Full_revisions.indd 211

Critical Visualization

7/20/22 9:49 AM

212

In the original charts, Du Bois is providing proof of Black people’s humanity and equality to white people, but Gates does something else by stripping away the specificity of the data. He brings it back to one of the fundamental moves in visualization — abstraction. He utilizes a certain ambiguity to help the viewer create meaning with the image. He states that, ‘In this situation, people are given an opportunity to imagine their own content — they can layer themselves upon my abstractions’ (Miranda, 2017). Beyond this immediate potential effect for a viewer, the paintings also reinscribe Du Bois within African American culture and take up his visualizations as an aesthetic that resonates with meaning. Du Bois’s works of data representation are reworked as visual aesthetic experiences. In another case, the charts are referenced in order to show the continuation of race-based class oppression in the usa, as in the work done by The Guardian data journalist Mona Chalabi (2017). Chalabi uses the same, or very similar, categories of data to produce the charts in order to allow a comparison between then and now. The journalist also meticulously recreates the colours, composition and line work of the originals. This reproduction has two outcomes: aesthetics become meaning­ ful and function in the process of representation — first of data describing the shape of human lives, but also of an important history of cultural production. It also has another, perhaps deeper, effect, that is, to honour the work of a prominent Black scholar by putting his work back in the spotlight and literally building upon his work for contemporary analysis.

Figure 6.21 — Occupation of Blacks and whites in the United States. Mona Chalabi / Copyright Guardian News & Media Ltd 2021

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 212

7/20/22 9:49 AM

Figure 6.22 — Illiteracy. Mona Chalabi / Copyright Guardian News & Media Ltd 2021

Figure 6.23 — T he racial elements of the population of the United States. Mona Chalabi / Copyright Guardian News & Media Ltd 2021

Chapter VI

220221_CritViz_Book_Full_revisions.indd 213

Critical Visualization

7/20/22 9:49 AM

214

Representation for a critical practitioner is about making certain things that are invisible, or that have been made invisible, into visible things that can be pointed at and witnessed (e.g. a disappeared political a­ ctivist, an un­acknowledged traditional place name or a local effect of global forces). The visualized artefacts produced and the processes undertaken create their real effects. As Jean Baudrillard (1983) says, the map precedes the ­territory. Representations as simulations of a territory are not mirrors of an objective reality, but a force that shapes the world. Visualizations inscribe boundaries and construct objects which in turn become our ­realities, cadastres give legal and material form to the new landscapes of private property, geomorphological mappings do not mirror the ­physical world but create textual abstractions (Pickles, 2004, p. 145). Visualizations are powerful and often enthralling artefacts that can p ­ roduce real effects. Yet to fetishize data, maps and visualization would be to accord these vehicles with more power over our imaginations and realities than they actually have. Focusing on their aesthetics, rich information and ability to engross us in a topic can also decontextualize the objects generated by artists and designers, thus making them appear as beautiful ephemera. We must not mistake the artefact for the real thing and forget the actual visualizing or mapping itself. For this reason, we need to focus on the process as well as the various material outcomes of ­visualization. Visualization should be thought of as an action (i.e. ­visualizing), rather than an object (Hall, 2012). This is proposed as a continuation of Pickles’ (2004) suggestion that maps should be thought more as mapping. It has been generative to blur the lines between a map, a diagram, an inform­ ation visualization and a data visualization. Different kinds of visualizations will be better suited to do some things rather than others (e.g. mapping for spatial issues or information visualization of text and image-based data sets). Yet, for the purposes of the discussion in this book it is more useful to focus on the overlaps between these kinds of visualizations, rather than reiterate their specific differences. We have found that designers, artists, researchers and activists have very often creatively mixed genres or problematized conventional representations of data. We looked at how aesthetics and representation are imbricated in the process of visualization and how a critical practice takes on these issues by problematizing or emphasizing them. Critical visualization designers, in order to be good designers, must embrace their agency in the process and the agency of the artefacts they produce to shape the world they are aiming to represent. █

Beyond Aesthetics and Representation

220221_CritViz_Book_Full_revisions.indd 214

7/20/22 9:49 AM

220221_CritViz_Book_Full_revisions.indd 215

7/20/22 9:49 AM

VII Beyond Critical  Visualization ­Practice

220221_CritViz_Book_Full_revisions.indd 216

7/20/22 9:49 AM

T

o cr eate a genea logy of cr itica l v isua liz ation w e need to decide on the characteristics of the artefacts that provide a thread between different exemplars. Perhaps the first thing to note is that the focus is not just on the visualization object, but also on the social process and political context from which the artefact emerges. This runs counter to a dominant narrative where visualization, like many technical fields, has continued the trajectory of the ‘cartography of progress’. As a science and an art, visualization has become more usable, more efficient and more capable of compressing complexity into an object that invites insight. Techniques of human-computer interaction that better utilize human perception and cognition through the creation of novel visual forms, or the design and orchestration of interactive tools and processes, are the dominant research agendas of academic researchers.

217

Beyond a linear conception of developing visualization techniques for creating more efficient and more precise representations of data, in this book we have promoted moves that are more lateral and sometimes jump back to revisit and learn again. It is useful to think of Adorno’s (1966/1983) n ­ otion of constellation as a way of noticing relations among different objects which offers us a way of thinking across not just historical c­ onceptions of visualization or visualization artefacts, but also processes and contexts of use and creation. For this reason, we have looked at makers of critical visualization who occupy or ally themselves with a marginalized position and practice a counter-hegemonic critique. The critical approach espoused in this book should also be seen within a context of critical studies and critical theory. This may be an obvious point to some, but it needs to be stated that the critical project is one that has come from a specific commitment to a critique of the Enlightenment and mass culture — two targets that conceive of modernity as an apex of Western thinking and the concentration of cultural production within a capitalist economy. While indebted to this form of critical ­thinking, we also realize the ways we must expand beyond these traditions if we hope to stay relevant to contemporary struggles against oppression. In particu­lar, we need to pay attention to, learn from, and participate in forms of thinking and practice that decentre Western epistemologies. Boaventura de Sousa Santos puts the failings of this tradition in very clear terms when he states that Frankfurt School critical theory is not sufficient

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 217

7/20/22 9:49 AM

218

for a decolonial praxis. It has, in fact, ‘failed to account for the emancipa­tory struggles of our time, in part at least because it shares with the bourgeois thinking it criticizes the same epistemological foundations that suppress the cognitive dimension of social injustice, and thus renders universal the Western understanding and transformation of the world. Moreover, it sees itself as a vanguard theory that excels in knowing about, explaining and guiding rather than knowing with, understanding, facilitating, sharing and walking alongside’ (Santos, 2016, p. ix). As visualization is a process that embeds data and representation within cognitive and cultural processes, we take Santos’ challenge to heart. One of the dimensions we’ve chosen to focus on in this book is the work of making and maintaining relations within the process of visualization. This aligns with Santos’ working with and sharing in a struggle. It prompts us to find a range of ways of making futures that de-prioritize Western frameworks — even critical traditions — which Santos and others are ­articulating through notions such as epistemologies of the south, decolonization and pluriversal thinking. This, we believe, is where the future work of critical visualization practice must be directed. One of the reasons we think this is needed is due to the continuation and extension of colonization by means lauded as progress. In the past decade, we have seen a massive increase in the collection of data from users of internet media (e.g. video/photo sharing, online shopping, social media, video/music streaming). This is often accepted as part of the deal, i.e. that data is collected as ‘payment’ for free searching, navigating, mapping and so on, but it also amounts to a form of data colonization (Couldry and Mejias, 2019) that is reproducing and intensifying power asymmetries by extracting value from work and leisure. Extraction is a key logic within colonization and capitalism. ­Historically, this has taken the shape of extending empires from a centre of power (e.g. Britain, France, Spain, Belgium), by subjugating through military ­violence and cultural assimilation territories and peoples valued as resources (i.e. raw materials and labour). More recently, this has taken the shape of continued settler colonialism, which has further dispossessed Indigenous communities from their own territories and ways of life (e.g. Canada, usa, Australia, Brazil). This works in tandem with the t­ ransformation of global relations between rich countries and poor ­countries where asymmetric trade agreements (as well as military intervention, when necessary) are used as a tool for continuing the extraction of value.

Beyond Critical Visualization Practice

220221_CritViz_Book_Full_revisions.indd 218

7/20/22 9:49 AM

Extraction, rather than sharing with and walking alongside, is a ubiquitous logic that permeates so much of what we do in the everyday. We’ve described how this logic is also present in the making of data. Data is so often seen as a resource that can be converted into insight both for good and for profit. The invocation that data is the new oil, as mentioned in C ­ hapter 3, further reminds us of the inherent logics of appropriating, amassing, ­abstracting, transforming and commodifying.

219

Macarena Gómez-Barris (2017) notes that to resist extractive logics we must recover a basic human faculty. This faculty is part of her own work — she states that, ‘by making visible micro spaces of interaction and encounter within geographies where coloniality has left and continues to leave a deep imprint, […] renewed perception offers a method for decolonized study’ (p. 2). It is important to acknowledge that Gómez-Barris focuses on parts of South America where colonization has officially ended but extractive power relations, as stated above, continue. These are submerged and non-disciplinary perspectives that are situated and ‘saturated in ­coloniality’ (p. 11). For the notion of a renewed perception of relation, Gómez-Barris draws on Éduoard Glissant’s (1997) conceptualization of relation and difference as a generative force. Glissant promotes a search for opacity (due to complexity) over a transparency (due to simplicity) (Glissant quoted in Diawara, 2017). This prioritizes an aesthetics and composition that always alerts the reader that the process of highlighting one thing necessitates the obscuring of another. The relation amongst things is reasserted even if immediate clarity is sacrificed. Where conventional visualization practice may be geared towards simplification of complexity, Glissant points us in another direction — to maintain the complexity and not succumb to the lure of absolute meaning. Glissant urges us to move towards poetry because of the generative and irreducible character of its construction and performance. Complexity and the production of difference is a result of myriad connect­ ions and relations of varying strength, duration and character. The empha­sis on relations is the counter to extractive logics. This direction may produce a messy problem for dominant notions of visualization, which have been obsessed with visibility as a form of bringing light to a ­poorly understood reality. In this way, the dominant notions reproduce an E ­ nlightenment agenda that builds a correspondence between unity,

Chapter VII

220221_CritViz_Book_Full_revisions.indd 219

Critical Visualization

7/20/22 9:49 AM

220

­rationality, measurement, truth and transparency. Perhaps the pursuit of justice also requires another ally beyond transparency and visibility. This may be where praxis is key. What is created through the process of visualization? A ­community? Accomplices? Allies? The process of documenting, ­recording, translating, representing is its own end when done right. What is created o ­ utstrips transparency or representability as the only criteria — relations of a­ ccountability. The density of endless connections, difference and worlds does not pose a design problem, but is perhaps already a strategy for survival and a result of struggle, displacement, migration, geography, history or context. The pluri­ versal, or multiple worlds, are a necessity (Escobar, 2018). The role of visualization is to be a particular manifestation and representation of those worlds. If we think of some of the contemporary struggles for autonomy, territory and freedom from extraction there emerges an alignment of these principles. Brothers and Sisters: Many words walk in the world. Many worlds are made. Many worlds are made for us. There are words and worlds which are lies and injustices. There are words and worlds which are truths and truthful. We make true words. We have been made from true words. In the world of the powerful there is no space for anyone but themselves and their servants. In the world we want everyone fits. In the world we want many worlds to fit. The Nation which we construct is one where all communities and languages fit, where all steps may walk, where all may have laughter, where all may live the dawn. We speak of unity even when we are silent. Softly and gently we speak the words which find the unity which will embrace us in history and which will discard the abandonment which confronts and destroys one another. Our word, our song and our cry, is so that the most dead will no longer die. So that we may live fighting, we may live singing. Long live the word. Long live Enough is Enough! Long live the night which becomes a soldier in order not to die in oblivion. In order to live the word dies, its seed germinating forever in the womb of the earth. By being born and living we die. We will always live. Only those who give up their history are consigned to oblivion. We are here. We do not surrender. Zapata is alive, and in spite of everything, the struggle continues. From the mountains of the Mexican Southeast. Subcomandante Insurgente Marcos Indigenous Clandestine Revolutionary Committee General Command of the Zapatista Army of National Liberation Mexico, January of 1996. (Marcos, 2004, p. 669)

Words that walk, in our terms, are akin to maps and visualizations that walk and see, and guide our imaginaries and our actions. Similar to how Minard (1862) referred to his data representations as objects that speak,

Beyond Critical Visualization Practice

220221_CritViz_Book_Full_revisions.indd 220

7/20/22 9:49 AM

count and calculate, visualizations act. They have a particular agency to affect us and the environment. Visualization objects and processes are designed to act in the world, and these designed things design us as well (Willis, 2006).

221

Whose words? Whose visualizations? Frantz Fanon recognized the incredi­ble power of representation and language in the formation of ­identity. When he states that, ‘A man who has a language consequently possesses the world expressed and implied by that language’ (Fanon, 2008, p. 9), he makes clear that power is wielded by those who wield the terms. Consider the image of Canada. Recently, it has become more common to say ‘so-called Canada’ in order to signal the contingency or ­positionality involved in naming a place. A government website relates the story of explorer Jacques Cartier adapting the word ‘kanata’, meaning ‘village’ in Huron-Iroquois, given to him by two teenage boys (Canadian Heritage, 2017). Candidate names for the country were also Albertland, Britannia, Victorialand and Colonia, among others (Canadian Heritage, 2017) — perhaps the imaginary that describes Canada as a British colony would have been significantly aided if one of these other names had been chosen. Many Indigenous writers, scholars and activists instead refer to the territory as Turtle Island, following the creation story shared by many Native peoples in the area. Other names for territories in the Americas include: Abya Yala (by the Kuna-Tule people), Anahuac (by many Indigenous peoples in the North), and Aztlán (by the Aztec peoples). These are strategies to take back a conceptualization of the territory. Iris Zavalla tells us that naming is a form of cartography that fixes an idea of what a place is comprised of and thus suppresses previous identifications held within that place (Mignolo and Walsh, 2018, p. 22). Naming a place covers a place with a web of meanings — and we might ask, whose web? If we draw an analogous relationship to data and information representation, then the arrangement, contextualization, identification are also forms of covering events, people, things and places with particular sets of signifiers. Margaret Pearce’s Coming Home map, which places traditional Indigenous names on the Canadian landmass, does this kind of work in two ways: building relations of accountability to knowledge keepers; ­representing ­visually/textually the relationships of orality to place. Pearce, in the

Chapter VII

220221_CritViz_Book_Full_revisions.indd 221

Critical Visualization

7/20/22 9:49 AM

222

­creation of the map, employed a protocol by which she made contact with all constituents who were custodians of traditional knowledge of a given place. This meant that trust between visualization designer and participant-as-source-of-data needed to be established and cared for. The orality-based community knowledge of place was in part represented through traditional names — names that rarely, if ever, appear in maps but instead in stories retold over millennia.

Figure 7.01 — Coming Home to Indigenous Place Names in Canada (2017). Map design © 2017 Canadian-American Center, University of Maine. Place names remain the property of the First Nations and Indigenous communities whose names they are.

Pearce hence does the work of vincularidad. Mignolo and Walsh write, ‘incularidad is the awareness of the integral relation and interdependence amongst all living organisms (in which humans are only a part) with territory or land and the cosmos. It is a relation and interdependence in search of balance and harmony of life in the planet’ (2018, p. 1). Our inter­ dependence requires accountability in our design processes. Pearce’s method embeds this logic and further directs the project to keep doing the work of telling the story of interdependence between humans, non-­ humans and land.

Beyond Critical Visualization Practice

220221_CritViz_Book_Full_revisions.indd 222

7/20/22 9:49 AM

Many critical visualization projects embed an educational p ­ urpose. As we saw, Beehive Collective expressly uses its large and incredibly d ­ etailed maps in storytelling and teaching. Iconoclasistas finds collaborative ­methods of mapping that both rely on existing participant knowledge and s­ haring of knowledge. Rather than treat the participants as empty vessels where you can deposit information (i.e. the ‘banking’ concept of pedagogy), these ­projects emphasize reciprocity and acknowledgement of the lived experience and position of the participant. Freire’s (1968/2017) highly influential notion of critical pedagogy, opposed the ‘banking’ model of teaching/ learning and instead promoted the co-production of knowledge and ­acknowledging the authority of the teacher as a tension in the classroom.

223

Figure 7.02 — W ho owns the land? (2017). Iconoclasistas

Janet Csontos’ project — Land Acknowledgement: Place Based H ­ istories and Indigenous Leadership — works to build knowledge collectively through the addition of image, text and story to specific locations on ­Turtle Island (2019). Csontos uses the visualization as a tool to let a knowledge ­sharing and validation process emerge. The map served as a proxy for land and aided her student-centred approach, which directly linked ­history to territory rather than linear time. She states that, ‘Inspired by Vine D ­ eloria’s writings, the shift from time-based history to place-based history was intended to evoke a more concrete understanding (than what would be

Chapter VII

220221_CritViz_Book_Full_revisions.indd 223

Critical Visualization

7/20/22 9:49 AM

224

learned from memorizing events on a timeline) of Indigenous peoples’ profound connec­tion to the lands here, and of Canada’s colonial history’ (Csontos, 2019). Csontos’ visualization project mediates a process within a larger process of decolonization. Within a settler-colonial context like North America, this requires a restitution of territory to the ­Indigenous peoples who continue to live here. It is important to point out this specificity due to the ­increasing use of the word ‘decolonization’. As Eve Tuck explains, ‘Land is what is most valuable, contested, required. This is both because the settlers make Indige­nous land their new home and source of capital, and also because the disrup­tion of Indigenous relationships to land represents a profound epistemic, ontological, cosmological violence’ (Tuck and Yang, 2012, p. 5). Land and territory is a source of traditional and spiritual knowledge and thus is part of how a people survives. When advocating for a decolonization, as non-Indigenous authors, we are also careful to not reproduce the settler logic described by Tuck and Yang. ‘The settler intellectual who hybridizes decolonial thought with ­Western critical traditions (metaphorizing decolonization), emerges superior to both Native intellectuals and continental theorists ­simultaneously’ (Tuck and Yang, 2012, p. 16). This is a profound challenge for us, in that we solidarize with decolonial Indigenous movements and learn from their thinkers and activists, but need to reflect very carefully on how we parti­ cipate in these struggles from the outside. As Tuck and Yang explain, decolonization is not a metaphor, ‘it is not an approximation of other experiences of oppression’ (2012, p. 3). When we think of decolonization in visualization practice, we are attempting to articulate a way of supporting these struggles through terms established by those most affected and those who are members of these communities. We recognize that decolonization may mean something different in other contexts of territory and history. As visualization practitioners, it is therefore our responsibility to these communities to reflect their priorities in their particular conditions. When representations of territory have been instrumentalized to ­circulate forms of knowledge of a space, it has been historically in the service of colonial governments and corporate interests where useful i­ nformation is encapsulated in calculable form and easily communicated to ­decision­makers far away. Visualization and mapping did not originate in modern­ity, but our experience of these techniques is very much rooted in modern­ity. Modernity is characterized by a commitment to rationality, transparency, and universalism — visualizations have sought to aid in this pursuit. Maps translate an unfriendly, dense, lively and overwhelming territory into a bird’s eye view that is readily graspable due to its stable form.

Beyond Critical Visualization Practice

220221_CritViz_Book_Full_revisions.indd 224

7/20/22 9:49 AM

Haraway calls these views provided by visualizations apollonian after the Ancient Greek God Apollo, associated with reason, culture, ­harmony and r­ estraint as well as with the sun that travels across the sky. It is this ­distance from above and the emphasis of reason that Haraway ascribes to dominant forms of described reality. Haraway (1988) calls this the ­god-trick, the infinite view from nowhere, innocent, not attached and therefore guaranteeing critical distance and objectivity.

225

As Cosgrove (2003) reminds us, the Apollonian eye, the disinterested view from above earth also promotes a view of the world as a unified whole. A pluriversal approach works against this one world narrative that invites hierarchical order and mastery. As an antidote to the totalizing vision, Haraway offers a partial vision, situated in the observed, aware of its generativity and therefore ­responsible and accountable to what is seen. This is, in part, what informs the ­multiple worlds as non-exclusive forms of understanding reality. While the multiplicity of positions and knowledges may conjure images of relativism, it is ultimately a false association. Haraway points to how relativism ­guarantees the ability to be everywhere by reiterating a connection to nowhere. I­ nstead, she states, ‘The alternative to relativism is partial, locatable, ­criti­cal knowl­ edges sustaining the possibility of webs of connections called ­solidarity in politics and shared conversations in epistemology’ (­Haraway, 1988, p. 191). Thus, a critical visualization practice can operate by ­prioritizing subaltern, marginalized, counter-hegemonic, situated ­knowledges that ­acknowledge their partial views, by acting as an ally to those who maintain these ­epistemologies. In the pursuit of revealing the wages of modernity, or colonialism or ­capitalism, it has often been the case that the role of visualization has been devoted to the presentation of evidence of this violence and history of extraction. While these efforts are necessary, they do not represent the only task required of visualization. The question of what authority is being appealed to with the representation of evidence, and whether that status quo will change as a result, is especially pointed within colonial conditions where the authority has a vested interest in maintaining the current relations of power. Other directions for representation with data (and other material) can be imagined that work outside of this limited economy. Tuck (Tuck, 2009; Tuck and Yang, 2012) provides a useful framework to think beyond too-narrow a logic of criticality by articulating the difference between pain-centred research and desire-centred research.

Chapter VII

220221_CritViz_Book_Full_revisions.indd 225

Critical Visualization

7/20/22 9:49 AM

226

Tuck identifies two forms of committed scholarship that have two very different outcomes. Damage-centred research, Tuck says, is ‘research that operates, even benevolently, from a theory of change that establishes harm or injury in order to achieve reparation’ (Tuck, 2009, p. 413). This approach uses a theory of change, of reparation, that stems from a litigation framework where providing evidence of wrong-doing will bring about justice (Tuck, 2009, p. 413). Visualization practices, as forms of witnessing and providing evidence, often work in this mode. Through the p ­ resentation of new data in forms that reveal new patterns, it is hoped that either a ­critical mass of viewers, or specific authorities, will finally see the truth and therefore act in the service of justice. The second form of committed scholarship identified by Tuck (2009) is ‘desire-centred research’. Tuck describes desire as ‘a thriving of the ­dicho­tomized categories of reproduction and resistance’ (Tuck, 2009, p. 419), where reproduction is the affirmative and the reiteration of the ­status quo, and resistance is the opposition to oppressive forces. The ­question that desire posits is how does the work help heal and develop a  futurity of a community? Tuck states that, ‘Desire, yes, accounts for the loss and d ­ espair, but also the hope, the visions, the wisdom of lived lives and c­ ommunities. Desire is involved with the not yet and, at times, the not anymore’ (2009, p. 417). The work of Pearce and Csontos are natural examples of visualization that operates through desire, healing and futurity. This also resonates with what a graduate student once told us about her own mapping thesis project concerning the colonial process in Canada. As a settler, she had come to understand her implicatedness in the dispossession of First Nations, specifically in the territory in which she had grown up. As with many territories in Canada, Indigenous communities were progressively removed from their land, either through i­ nterpretations of treaties, structural or military violence, or contracts made in bad faith. To present and receive feedback on her archival research detailing a specific case of dispossession, she held visiting hours at a community centre. Residents would wander in to peruse the materials and maps and sometimes enter into a discussion. On one occasion, she talked with a ­community elder about the role of maps. After looking through the visual­ izations and tracing the history it revealed, he suggested that these maps could be a tool for healing. Healing through the presentation of the roots of intergenerational pain, as well as through the sharing of experiences.

Beyond Critical Visualization Practice

220221_CritViz_Book_Full_revisions.indd 226

7/20/22 9:49 AM

The function of critical visualization as a decolonizing tool/process needs to reflect its context. Decoloniality is not one universal term from the outside. On Turtle Island and in a settler-colonial context, it is directly linked to sovereignty and restitution of territory. In other, interrelated contexts, such as Latin America, this may also include both Indigenous and mestizo epistemologies marked by modernity. Decolonial can be understood not as post-colonial — focused on the various histories of colonies that gained independence — but rather an interrogation and disruption of the continued colonialist relations between powerful states and Indigenous peoples, and the cultural logics that reinforce these relations. Mignolo and Walsh remind us that ‘coloniality is constitutive, not derivative, of m ­ odern­ity’ (2018, p. 4). Colonization has also meant a continued asymmetrical relationship between the Global South and the Global North. Joaquín Torres-García’s América Invertida signals this by imagining our sense of North-South as corresponding to up-down and correlating to good-bad. Alfredo Jaar’s A Logo for America similarly disrupts US-centric naming conventions and reminds us of the presence of South and Central America in the designation ‘America’. Reminiscent of associations of white and black to good and bad, and reproduced and applied by white supremacist thinking to skin colour, the up-down/good-bad/master-slave dichotomies play out in our geographical imagination. Iconoclasistas employs this now-famous gesture as a pedagogical move in their mappamundis showing Indigenous and feminist resistance to environmental violence.

227

Figure 7.03 — América Invertida. Joaquín Torres-García

Chapter VII

220221_CritViz_Book_Full_revisions.indd 227

Critical Visualization

7/20/22 9:49 AM

228

Figure 7.04 — Decolonizing Knowledge. Marc Ngui

Beyond Critical Visualization Practice

220221_CritViz_Book_Full_revisions.indd 228

7/20/22 9:49 AM

229

Figure 7.05 — A Logo for America. Alfredo Jaar. Courtesy of the artist, New York

Decoloniality, in these terms, is concerned with both unveiling coloniality-­ modernity in its many forms, including the dominance of technoscientific knowledge, as well as expressing other worlds. As Mignolo and Walsh state, ‘… decoloniality seeks to make visible, open up, and advance radically distinct perspectives and positionalities that displace Western rationality as the only framework and possibility of existence, analysis and thought’ (2018, p. 17). This pluriversal conception, expressed through visualization, can counter dichotomies and offer differently rooted knowledges. This also requires us to not just tell how decoloniality happens but also reveal it in process (Gloria Anzaldúa, cited in Mignolo and Walsh, 2018, p. 20) — in other words work from and with community, resonating with Haraway’s call for a partial and situated knowledge practice. Critical visualizations can do this if we commit to thinking in terms of process — assemblages of visualization that act, share, produce, imagine. As we stated earlier, visualization and mapping do not originate in modern­ ity, but our experience of these techniques is very much rooted in modernity. Visualization is complicit in its historical commitments to ­technoscientific knowledge and the extractive processes in colonialism and capitalism. We believe that critical visualization practice must go towards decolonial futures in order to address the deep entanglement between data making, representing, shaping and circulating. Since we act in the world with the knowledge we hold, the role of visualization as a process of knowledge transfer and reproduction is particularly important. Future ­directions of critical visualization must take on the challenge of decolonization particu­ larly because of the historical and continued role of data collection and representation in the control and dispossession of people from their land, as well as the potential for telling stories that bring together histories of ­resistance and solidarity. █

Chapter VII

220221_CritViz_Book_Full_revisions.indd 229

Critical Visualization

7/20/22 9:49 AM

Sources

220221_CritViz_Book_Full_revisions.indd 230

7/20/22 9:49 AM

AB

231

A



Achinstein, P. (2004). Science Rules: A Historical Introduction to Scientific Methods. jhu Press. Adas, M. (1995). Machines as the Measure of Men: Science, Technology and Ideologies of Western Dominance (5. paperback print). Cornell University Press. Adorno, T. W. (1983). Negative Dialectics. Continuum. (Original work published 1966). Adorno, T. W., and Horkheimer, M. (2009). The Culture Industry: Enlightenment as Mass Deception. In M. G. Durham and D. M. Kellner (Eds.), Media and Cultural Studies: KeyWorks (pp. 41–72). (Original work published 1947). Albrecht, K. (2018, October 10). Interview with Peter Hall [Audio call and email]. Amoore, L. (2006). Biometric borders: Governing mobilities in the war on terror. Political Geography, 25(3), 336–351. https://doi.org/10.1016/­ j.polgeo.2006.02.001 Appadurai, A. (1996a). Global Ethnoscapes: Notes and Queries for a Transnational Anthropology. In Modernity at Large: Cultural Dimensions of Globalization (­pp. 48–65). University of Minnesota Press. Appadurai, A. (1996b). Modernity at Large: Cultural Dimensions of Globalization (Vol. 1). University of Minnesota Press. Arup. (2010). Smart Cities: Transforming the 21st Century City via the Creative Use of Technology. https://www.arup.com/ perspectives/publications/research/ section/smart-cities

Ashmore, B. (2003). Arno Peters changed the World! Development Education and the Peters’ Projection. The Cartographic Journal, 40(1), 57–59. https://doi.org/10.1179/000870403 235002097 Backer, S. (2020, December 1). Hear the Blind Spot: Visualizing Data for Those Who Can’t See. Medium. https://medium.com/nightingale/ hear-the-blind-spot-visualizing-datafor-those-who-cant-see-b370e6ec0b9e Bangor Daily News. (2009, January 14). Map Tells Story of Champlain’s Explorations. Bangor Daily News. https://www.bangordailynews. com/2009/01/14/news/bangor/ map-tells-story-of-champlainrsquosexplorations/?amp Bates, J., Lin, Y.-W., and Goodale, P. (2016). Data journeys: Capturing the sociomaterial constitution of data objects and flows. Big Data & Society, 3(2), 205395171665450. https://doi. org/10.1177/2053951716654502 Battle-Baptiste, W., and Rusert, B. (Eds.). (2018). WEB Du Bois’s Data Portraits: Visualizing Black America. Chronicle Books. Baudrillard, J. (1983). Simulations. Semiotext(e), Inc. Bauman, Z., and Lyon, D. (2013). Liquid Surveillance: A Conversation. Polity. Bayley, L. (2020, February 14). Interview with Peter Hall [Personal communication]. bbc. (2010, August 9). Information Graphics. In Newsnight. https://www.youtube. com/watch?v=q2Wnu1SOhKs&t=357s

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 231

7/20/22 9:49 AM

B 232



bbc Radio 4. (2013, June 16). The Quantified Self: Can Life Be Measured? British Broadcasting Corporation. https:// www.bbc.co.uk/programmes/ b02mfzrw Bedeian, A. G., and Wren, D. A. (2001). Most Influential Management Books of the 20th Century. Organizational Dynamics, 29(3), 221–225. https://doi. org/10.1016/S0090-2616(01)00022-5 Bederson, B., and Shneiderman, B. (2003). The Craft of Information Visualization: Readings and Reflections. Morgan Kaufmann. Bertin, J. (2011). Semiology of Graphics: Diagrams, Networks, Maps (1st ed.). esri Press; Distributed by Ingram Publisher Services. (Original work published 1967). Bessa, P. (2008). Skittish Skirts and Scanty Silhouettes: The Tribulations of Gender in Modern Signage. Visible Language, 42(2), 119. Beynon-Davies, P. (2009). Significant Threads: The Nature of Data. International Journal of Information Management, 29, 170–188. Bini, E. (2014). Drawing a Global Color Line: ‘The American Negro Exhibit’ at the 1900 Paris Exposition. In G. Abbattista (Ed.), Moving Bodies, Displaying Nations: National Cultures, Race and Gender in World Expositions: Nineteenth to Twenty-first century (pp. 39–65). eut Edizioni Università di Trieste. Boddy-Evans, A. (2018). The Trans-Atlantic Slave Trade. In Thought Co. https:// www.thoughtco.com/the-transatlantic-slave-trade-44544 Borges, J. L. (1999). Selected Non-fictions (E. Weinberger, Ed.). Viking. (Original work published 1942).

Bourdieu, P. (1990). The Logic of Practice (Reprinted). Stanford University Press. Bowker, G. (2013). Data Flakes: An Afterword. In L. Gitelman (Ed.), ‘Raw Data’ is an Oxymoron. mit Press. Bowker, G., and Star, S. L. (1999). Sorting Things Out: Classification and Its Consequences. The mit Press. Brotton, J. (2014). A History of the World in Twelve Maps. Penguin. Browne, S. (2015). Dark Matters: On the Surveillance of Blackness. Duke University Press. Buchloh, B. H. D. (1990). Conceptual Art 1962–1969: From the Aesthetic of Administration to the Critique of Institutions. October, 55, 105. https:// doi.org/10.2307/778941 Bunting, H. (2019, July 24). Interview with Peter Hall [Personal communication]. Burke, C. (2012). Isotype Revisited: Society and Economy. Isotype Revisited. https:// isotyperevisited.org/2012/08/societyand-economy.php

BC

Callon, M. (1986). Some Elements of a Sociology of Translation: Domestication of the Scallops and the Fishermen of St Brieuc Bay. In J. Law (Ed.), Power, Action and Belief: A New Sociology of Knowledge? (pp. 196–223). Routledge. Canadian Heritage. (2017, August 15). Origin of the name ‘Canada.’ Aem. https:// www.canada.ca/en/canadianheritage/services/origin-namecanada.html

Sources

220221_CritViz_Book_Full_revisions.indd 232

7/20/22 9:49 AM

CD

233

C



Card, S.K., and Mackinlay, J. (1997). The Structure of the Information Visualization Design Space. Proceedings of viz ’97: Visualization Conference, Information Visualization Symposium and Parallel Rendering Symposium, 92–99. https://doi. org/10.1109/INFVIS.1997.636792 Card, S. K., Mackinlay, J. D., and Shneiderman, B. (1999). Readings in Information Visualization: Using Vision to Think. Morgan Kaufmann Publishers. Chalabi, M. (2017, February 14). W.E.B. Du Bois: Retracing his attempt to challenge racism with data. The Guardian. https://www.theguardian. com/world/2017/feb/14/web-du-boisracism-data-paris-african-americansjobs Chaplin, C. (Director). (1936). Modern Times [Film; 87 mins]. Charles Chaplin Productions. Christianson, S. (2012). 100 Diagrams that Changed the World: From the Earliest Cave Paintings to the Innovation of the iPod. Penguin Books. Codebò, A. (2019). Decolonizing the Landfill: Counter-Maps of Waste in Buenos Aires. Journal of Latin American Geography, 18(3), 30–53. https://doi.org/10.1353/lag.2019.0057 Cohen, B. (2014). Burning Cars, Caricatures and Glub. Third Text, 28(2). https://doi. org/10.1080/09528822.2014.891829 Colebrook, C. (2002). Gilles Deleuze. Routledge. Conant, J. (2005). Family Resemblance, Composite Photography, and Unity of Concept: Goethe, Galton, Wittgenstein. Wittgenstein Repository. http://wittgensteinrepository.org/ agora-wab/article/view/3148

Corner, J. (1999). The Agency of Mapping: Speculation, Critique and Invention. In D. Cosgrove (Ed.), Mappings. Reaktion Books. Cosgrove, D. (2005). Maps, Mapping, Modernity: Art and Cartography in the Twentieth Century. Imago Mundi, 57(1), 35–54. https://doi-org.arts.idm.oclc. org/10.1080/0308569042000289824 Cosgrove, D. E. (2003). Apollo’s Eye: A Cartographic Genealogy of the Earth in the Western Imagination. Johns Hopkins University Press. Couldry, N., and Mejias, U. A. (2019). The Costs of Connection: How Data is Colonizing Human Life and Appropriating it for Capitalism. Stanford University Press. Crampton, J. (1994). Cartography’s Defining Moment: The Peters Projection Controversy, 1974–1990. Cartographica: The International Journal for Geographic Information and Geovisualization, 31(4), 16–32. Crampton, J. W. (2010). Mapping: A Critical Introduction to Cartography and gis. Wiley-Blackwell. Creswell, J. W. (1998). Qualitative Inquiry and Research Design: Choosing among Five Traditions. Sage Publications. https://uk.sagepub.com/en-gb/eur/ qualitative-inquiry-and-researchdesign/book246896 Csontos, J. (2019). The Map Challenge. Nandogikendan. https:// nandogikendan.com/map/ Dallaire, R., and Beardsley, B. (2004). Shake Hands with the Devil: The Failure of Humanity in Rwanda. Arrow Books.

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 233

7/20/22 9:49 AM

D 234

D

Dalrymple, W. (2015, March 4). The East India Company: The Original Corporate Raiders. The Guardian. https://www.theguardian.com/ world/2015/mar/04/east-indiacompany-original-corporate-raiders Daston, L., and Galison, P. (1992). The Image of Objectivity. Representations, 40, 81–128. https://doi. org/10.2307/2928741 Daston, L., and Galison, P. (2010). Objectivity. Zone Books. Data4Change. (2020). Our Work. https:// www.data4chan.ge/our-work Dávila, P. (2019). Diagrams of Power: Visualizing, Mapping, and Performing Resistance. Onomatopee. Dawson, A. (2017). Extreme Cities: The Peril and Promise of Urban Life in the Age of Climate Change. Verso. Debord, G. (1994). The Society of the Spectacle. Zone Books. (Original work published 1967). Debord, G., and Jorn, A. (1957). Guide psychogéographique de Paris. Discours sur les passions de l’amour. De Certeau, M. (1984). The Practice of Everyday Life (S. Rendall, Trans.; Nachdr.). University of California Press. Dekker, A. (2011, May). Heath Bunting: The World of an Administrative Content Provider. Aaa.Net. http://aaaan. net/heath-bunting-the-world-of-anadministrative-content-provider/ Deleuze, G. (1986). Cinema 1: The MovementImage (H. Tomlinson and B. Habberjam, Trans.). University of Minnesota. Deleuze, G., and Guattari, F. (1983). AntiOedipus: Capitalism and Schizophrenia. University of Minnesota Press. Descartes, R. (1998). Meditations on first philosophy. Donald A. Cress (Trans.), Hackett Press.

Descartes, R. (1986/1641) Meditations on First Philosophy, (J. Cottingham, Trans.). Cambridge University Press. Description of a Slave Ship. (1789). Princeton University Library. https://dpul. princeton.edu/wa/catalog/qj72p788s Diawara, M. (2017). Édouard Glissant’s Worldmentality: An Introduction to One World in Relation. https:// www.documenta14.de/en/ south/34_edouard_glissant_s_world_ mentality_an_introduction_to_one_ world_in_relation Dick, M. (2020). The Infographic: A History of Data Graphics in News and Communications. The mit Press. D’Ignazio, C. (2017). Creative Data Literacy. Information Design Journal, 23(1), 6–18. Doane, M. A. (2002). The Emergence of Cinematic Time: Modernity, Contingency, the Archive. Harvard University Press. Doctorow, C. (2019, October 30). Leaked document reveals that Sidewalk Labs’ Toronto plans for private taxation, private roads, charter schools, corporate cops and judges, and punishment for people who choose privacy. Boing Boing. https:// boingboing.net/2019/10/30/citizenscores-eh.html Doolittle, R. (2014, March 9). Toronto to expand ‘priority’ neighbourhoods to 31. The Toronto Star. Dourish, P. (2016). The Internet of Urban Things. In R. Kitchin and S.-Y. Perng (Eds.), Code and the City (pp. 27–45). Routledge, Taylor & Francis Group. Drucker, J. (2014). Graphesis: Visual Forms of Knowledge Production. Harvard University Press. Drucker, J., and McVarish, E. (2010). Graphic Design History: A Critical Guide. Pearson Prentice Hall.

Sources

220221_CritViz_Book_Full_revisions.indd 234

7/20/22 9:49 AM



DE

Duarte, F., and Ratti, C. (2019). Designing Cities within Emerging Geographies: The Work of Senseable City Lab. In T. Banerjee and A. Loukaitou-Sideris (Eds.), The New Companion to Urban Design (pp. 561–570). Routledge, Taylor & Francis Group. Du Bois, W.E.B. (1897, August). Strivings of the Negro People. The Atlantic. https://www.theatlantic.com/ magazine/archive/1897/08/strivingsof-the-negro-people/305446/ Du Bois, W.E.B. (1900, November). The American Negro at Paris. American Monthly Review of Reviews, XXII(5), 575–577. Duncombe, S., and Lambert, S. (2018, March 2). Alfredo Jaar. The Center for Artistic Activism. https://c4aa. org/2018/03/alfredo-jaar Eco, U. (Ed.). (2004). On beauty (A. McEwen, Trans.). Secker and Warburg. Edney, M. (1993). The Patronage of Science and the Creation of Imperial Space: The British mapping of India, 1799– 1843. Cartographica: The International Journal for Geographic Information and Geovisualization, 30(1), 61–67. Edney, M. (1999). Mapping an Empire: The Geographical Construction of British India, 1765–1843. University of Chicago Press. Escobar, A. (2018). Designs for the Pluriverse: Radical Interdependence, Autonomy, and the Making of Worlds. Duke University Press. Evans, R. J. (2020, July 28). RA Fisher and the Science of Hatred. New Statesman. https://www.newstatesman.com/ long-reads/2020/07/ra-fisher-andscience-hatred

F

235

Fanon, F. (2008). Black Skin, White Masks (1st ed., new ed). Grove Press; Distributed by Publishers Group West. Felski, R. (2015). The Limits of Critique. The University of Chicago Press. Felton, N. (2008). The Feltron 2008 Annual Report. http://feltron.com/FAR08. html Felton, N. (2014). 2014 Feltron Annual Report. http://feltron.com/FAR14.html Feyerabend, P. (1975). Against Method: Outline of an Anarchistic Theory of Knowledge. Verso. Finley, C. (2018). Committed to Memory: The Slave Ship Icon. Princeton University Press. Flinchum, R. (1997). Henry Dreyfuss, Industrial Designer: The Man in the Brown Suit. Cooper-Hewitt, National Design Museum, Smithsonian Institution and Rizzoli. Foucault, M. (1980). Power/knowledge: Selected Interviews and other Writings, 1972–1977 (C. Gordon, Ed.; 1st American ed). Pantheon Books. Foucault, M. (2002). Power: Essential Works of Foucault (1954–1984), Volume 3 (J. D. Faubion, Ed.). Penguin. (Original work published 1981). Foucault, M. (2007). The Order of Things: An Archaeology of the Human Sciences. Routledge Classics. (Original work published 1966). Freire, P. (2017). Pedagogy of the Oppressed. Penguin Books. (Original work published 1968). Friendly, M. (2008). A Brief History of Data Visualization. In C.-H. Chen, W. Härdle, and A. Unwin. (Eds.), Handbook of Data Visualization. (pp. 15–56). Springer.

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 235

7/20/22 9:49 AM

236

FG

Frigo, A. (2015, February 13). Lifelogging [Interview]. https://dublin. sciencegallery.com/lifelogging Fry, B. (2008). Visualizing Data. O’Reilly Media, Inc. Fry, T. (1999). A New Design Philosophy: An Introduction to Defuturing. unsw Press. Fry, T. (2012). Becoming Human by Design. Bloomsbury Academic, an imprint of Bloomsbury Publishing Plc. Funkhouser, H. G. (1937). Historical Development of the Graphical Representation of Statistical Data. Osiris, 1, 269–405. Gainty, C. (2016). Mr. Gilbreth’s Motion Pictures — The Evolution of Medical Efficiency. New England Journal of Medicine, 374(2), 109–111. https://doi. org/10.1056/NEJMp1514048 Galison, P. (2002). Images Scatter into Data; Data Gather into Images. In P. Weibel and B. Latour (Eds.), Iconoclash: Beyond the Image Wars in Science, and Art. mit Press. Gervat, B. (2019). The Penetrating Epic Poem. A Conversation with Alfredo Jaar. Art Pulse. http://artpulsemagazine.com/ the-penetrating-epic-poem-aconversation-with-alfredo-jaar Gibson, J. J. (2011). The Ecological Approach to Visual Perception (17th pr). Psychology Press. (Original work published 1979). Gilbert, H. (2014, June 3). big Scoops Funding for New York Hurricane Protection. Architects’ Journal. https://www.architectsjournal.co.uk/ news/big-scoops-funding-for-newyork-hurricane-protection



GH

Gitelman, L. (Ed.). (2013). ‘Raw data’ is an Oxymoron. The mit Press. Gitelman, L., and Jackson, V. (2013). Introduction. In L. Gitelman (Ed.), ‘Raw data’ is an Oxymoron (pp. 1–14). The mit Press. Glissant, É. (1997). Poetics of Relation (B. Wing, Trans.). University of Michigan Press. Goetz, K. (2010, July 13). Promoting Social Justice One Poster at a Time. Fast Company. https://www.fastcompany. com/1661896/promoting-socialjustice-one-poster-at-a-time Gómez-Barris, M. (2017). The Extractive Zone: Social Ecologies and Decolonial Perspectives. Duke University Press. Gould, S. J. (2008). The Mismeasure of Man (Rev. and expanded, with a new introduction). W.W. Norton. (Original work published 1980). Graham, R. (2014). A History of Search Engines: Mapping Technologies of Memory, Learning and Discovery. In R. König and M. Rasch (Eds.), Society of the Query Reader: Reflections on Web Search (pp. 106–120). Institute of Network Cultures. Green, D. (1985). Veins of Resemblance: Photography and Eugenics. Oxford Art Journal, 7(2), 3–16. Greenfield, A. (2013). Against the Smart City (The city is here for you to use Book 1). Amazon Media. https://www. amazon.com/Against-smart-cityhere-Book-ebook/dp/B00FHQ5DBS Grupo de Arte Callejero (Argentina) (Ed.). (2009). gac: Pensamientos, prácticas, acciones. Tinta Limón Ediciones. Hacking, I. (1990). The Taming of Chance. Cambridge University Press.

Sources

220221_CritViz_Book_Full_revisions.indd 236

7/20/22 9:49 AM

237

H

H

Hacking, I. (2000). Façonner les gens: Le seuil de pauvreté. In J.-G. Prévost and J.-P. Beaud (Eds.), Systèmes statistiques et traditions nationales: Statistical Systems and National Traditions (pp. 15–35). Les Presses de l’Université du Québec. Hacking, I. (2006). Making Up People. London Review of Books, 28(16). https://www.lrb.co.uk/the-paper/ v28/n16/ian-hacking/making-uppeople Hall, P. (2005, November 1). By the Numbers. Metropolis. https://metropolismag. com/programs/by-the-numbers/ Hall, P. (2008). Critical Visualization. In P. Antonelli and Museum of Modern Art (New York, N.Y.) (Eds.), Design and the Elastic Mind (pp. 122–131). Museum of Modern Art : Distributed in the U.S. and Canada by D.A.P./ Distributed Art Publishers. Hall, P. (2011). Bubbles, Lines and String: How Visualization Shapes Society. In A. Blauvelt, E. Lupton, and R. Giampietro (Eds.), Graphic Design: Now in Production (1st ed). Walker Art Center. Hall, P. (2012). Afterword: On Mapping and Maps. Design Philosophy Papers, 10(2), 157–167. http://www. thestudioattheedgeoftheworld.com/ uploads/4/7/4/0/47403357/07afterwo rdmapping-mapspeterhall.pdf Hall, P. (2019). Of Neon, Road Signs and Head Shapes: A Case for Generative Criticism. In C. Brisbin and M. Thiessen (Eds.), The Routledge Companion to Criticality in Art, Architecture, and Design. Routledge, Taylor & Francis Group. Hall, S. (Ed.). (1997). Representation: Cultural Representations and Signifying Practices. Sage in association with the

Open University. Halpern, O. (2014). Beautiful Data: A History of Vision and Reason since 1945. Duke University Press. Hannah, D., and Krajewski, S. (2015). Placing the Golden Spike: Landscapes of the Anthropocene. Publication Studio. Haraway, D. (1988). Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective. Feminist Studies, 14(3), 575. https://doi.org/10.2307/3178066 Haraway, D. (1991). Simians, Cyborgs, and Women: The Reinvention of Nature. Routledge. Harley, J. B. (1988). Maps, Knowledge and Power. In D. E. Cosgrove and S. Daniels (Eds.), The Iconography of Landscape: Essays on the Symbolic Representation, Design, and Use of Past Environments. Cambridge University Press. Harley, J. B., and Woodward, D. (Eds.). (1987). The History of Cartography. University of Chicago Press. Harley, J. B. (2002). The New Nature of Maps: Essays in the History of Cartography. jhu Press. Harris, S., and Berke, D. (Eds.). (1997). Architecture of the Everyday. Princeton Architectural Press. Harvey, F. (2019). Jacques Bertin’s Legacy and Continuing Impact for Cartography. Cartography and Geographic Information Science, 46(2), 97–99. https://doi.org/10.1080/15230 406.2019.1533784 Henrikson, A. (1999). The Power and Politics of Maps. In G. J. Demko and W. B. Wood (Eds.), Reordering the World: Geopolitical Perspectives on the Twentyfirst Century (2nd ed, pp. 94–116). Westview Press.

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 237

7/20/22 9:49 AM

H 238



Hill, K. (2011, April 7). Adventures in Selfsurveillance, aka the Quantified Self, aka Extreme Navel-gazing. Forbes. https://www.forbes.com/ sites/kashmirhill/2011/04/07/ adventures-in-self-surveillance-akathe-quantified-self-aka-extremenavel-gazing/?sh=43c405b6773b Hodge, R., and Kress, G. R. (1988). Social Semiotics. Cornell University Press. Holmes, B. (2006). Counter Cartographies. In J. Abrams, P. Hall, and Design Institute (Eds.), Else/where: Mapping: New Cartographies of Networks and Territories; [mapping networks, mapping conversations, mapping territories, mapping mapping] (2. print). University of Minnesota Press. hooks, bell. (2007). An Aesthetic of Blackness: Strange and Oppositional. In J. Livingstone and J. Ploof (Eds.), The Object of Labor: Art, Cloth, and Cultural Production (pp. 315–332). School of the Art Institute of Chicago Press ; mit Press. Horton, E. S. (2007). Neoliberalism and the Australian Healthcare System (Factory). Proceedings 2007 Conference of the Philosophy of Education Society of Australasia, 1–7. Houston, T. (2012, March 1). 5 Minutes on The Verge: Nicholas Felton. The Verge. https://www.theverge. com/2012/3/1/2831155/5-minutes-onthe-verge-nicholas-felton Howarth, D. (2018). Big U Flood Defences for Manhattan Move Forward. Dezeen. https://www.dezeen. com/2018/07/20/big-u-storm-flooddefences-east-side-coastal-resiliencymanhattan-move-forward/ Hutchins, E. (1995). Cognition in the Wild (Nachdr.). mit Press.

Ingold, T. (2000). The Perception of the Environment: Essays on Livelihood, Dwelling and Skill. Routledge. Irish News. (2018, April 30). A Student Took a Selfie Every Day for Eight Years and it’s the Best Timelapse. Irish News. https://www.irishnews. com/magazine/daily/2018/04/30/ news/a-student-took-a-selfie-everyday-for-eight-years-and-it-s-the-besttimelapse-1317091/

IJK

Jameson, F. (1991). Postmodernism, or, the Cultural Logic of Late Capitalism. Duke University Press. Jameson, F. (1992). The Geopolitical Aesthetic: Cinema and Space in the World System. Indiana University Press ; bfi Pub. Jameson, F. (2000). Cognitive Mapping. In M. Hardt and K. Weeks (Eds.), The Jameson Reader (pp. 277–287). Blackwell. (Original work published 1988) Jenkins, K. (2008). At the Limits of History: Essays on Theory and Practice. Routledge. Johnson, M. (2008, December 18). M ­ e-Projects. https://www.johnsonbanks.co.uk/ thoughts/me-projects Joss, S., Cook, M., and Dayot, Y. (2017). Smart Cities: Towards a New Citizenship Regime? A Discourse Analysis of the British Smart City Standard. Journal of Urban Technology, 24(4), 29–49. https://doi.org/10.1080/ 10630732.2017.1336027 KeepItOn coalition, and Robertson, B. (2020, October 6). Access Now [Personal communication]. Kent, W. (2000). Data and Reality. 1stBooks Library.

Sources

220221_CritViz_Book_Full_revisions.indd 238

7/20/22 9:49 AM

KL

239

K



Kitchin, R. (2014). The Data Revolution: Big Data, Open Data, Data Infrastructures and their Consequences. sage Publications. Kitchin, R. (2016). Big Data. In D. Richardson, N. Castree, M. F. Goodchild, A. Kobayashi, W. Liu, and R. A. Marston (Eds.), International Encyclopedia of Geography: People, the Earth, Environment and Technology (pp. 1–3). John Wiley & Sons, Ltd. https://doi.org/10.1002/9781118786352. wbieg0145 Kitchin, R., and Dodge, M. (2007). Rethinking Maps. Progress in Human Geography, 31(3), 331–344. https://doi. org/10.1177/0309132507077082 Kitchin, R., and Lauriault, T. P. (2018). Toward Critical Data Studies: Charting and Unpacking Data Assemblages and Their Work. In J. Thatcher (Ed.), Thinking Big Data in Geography: New Regimes, New Research (pp. 3–20). University of Nebraska Press. Kitchin, R., and Perng, S.-Y. (Eds.). (2016). Code and the City. Routledge, Taylor & Francis Group. Kivisto, P., and Faist, T. (2007). Citizenship: Discourse, Theory, and Transnational Prospects. Blackwell Publishing. Kondek, C., and Kühl, C. (2018). The Hairs of Your Head are Numbered. Excerpt — Intro to the Devices. https://vimeo.com/260943144 kpf. (n.d.). New Songdo City Songdo International Business District. Retrieved November 8, 2020, from https://www.kpf.com/projects/newsongdo-city Krause, H. (2019, June 7). Why We Need to be Data Detectives. We All Count. https://weallcount.com/2019/06/07/ why-we-need-to-be-data-detectives/

Kuhn, T. (2012). The Structure of Scientific Revolutions. University of Chicago Press. (Original work published 1962). Kulke, H., and Rothermund, D. (2016). A History of India (Sixth edition). Routledge, Taylor & Francis Group. Kurgan, L. (2013). Close Up at a Distance: Mapping, Technology, and Politics (First hardcover edition). Zone Books. Lakoff, G. (1990). Women, Fire, and Dangerous Things. University of Chicago Press. Latour, B. (1986). Visualisation and Cognition: Drawing Things Together. In H. Kuklick (Ed.), Knowledge and Society Studies in the Sociology of Culture Past and Present (pp. 1–40). Jai Press. Latour, B. (1992). Where are the Missing Masses? In W. E. Bijker and J. Law (Eds.), Shaping Technology/Building Society: Studies in Sociotechnical Change (pp. 225–258). Cambridge, Mass.: mit Press. Latour, B. (2004). Why Has Critique Run out of Steam? From Matters of Fact to Matters of Concern. Critical Inquiry, 30(2), 225–248. Latour, B. (2005). Reassembling the Social: An Introduction to Actor-Network Theory. Oxford University Press. Latour, B. (2018). Down to Earth: Politics in the New Climatic Regime (English edition). Polity. Latour, B. (2020, March 26). Is This a Dress Rehearsal? In the Moment. https:// critinq.wordpress.com/2020/03/26/ is-this-a-dress-rehearsal/ Lauriault, T. P. (2012). Data, Infrastructures and Geographical Imaginations [Doctoral Dissertation]. Carleton University.

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 239

7/20/22 9:49 AM

L 240



Lauriault, T. P., and Francoli, M. (2017). Openness, Transparency, Participation. In R. Kitchin, T. P. Lauriault, and M. Wilson (Eds.), Understanding Spatial Media (1st edition). sage Inc. Law, J. (2007). Actor-Network Theory and Material Semiotics. Lefebvre, H. (2008). Clearing the Ground. In S. Johnstone (Ed.), The Everyday. Whitechapel; mit Press. (Original work published 1961). Lefebvre, H., and Lefebvre, H. (1991). Critique of Everyday Life: Foundations for a Sociology of the Everyday. Verso. (Original work published 1977). Leonelli, S. (2014). What Difference does Quantity Make? On the Epistemology of Big Data in Biology. Big Data & Society, 1(1), 205395171453439. https:// doi.org/10.1177/2053951714534395 Levering Lewis, D., and Willis, D. (2010). A Small Nation of People: W.E.B. Du Bois and African American Portraits of Progress. Zondervan. Liu, Z., Nersessian, N., and Stasko, J. (2008). Distributed Cognition as a Theoretical Framework for Information Visualization. ieee Transactions on Visualization and Computer Graphics, 14(6), 1173–1180. https://faculty.cc.gatech.edu/~john. stasko/papers/infovis08-dcog.pdf Lobo, R. (2014, January 21). Could Songdo be the World’s Smartest City? World Finance. https://www.worldfinance. com/inward-investment/couldsongdo-be-the-worlds-smartest-city Lombardi, M. (2004). Mark Lombardi: Global Networks (R. C. Hobbs, Ed.). Independent Curators International.

Loos, A. (2002). Ornament and Crime. In B. Miller and M. Ward (Eds.), Crime and Ornament: The Arts and Popular Culture in the Shadow of Adolf Loos (pp. 29–36). yyz Books. (Original work published 1908). Lorde, A. (2018). The Master’s Tools Will Never Dismantle the Master’s House. Penguin Books. (Original work published 1984). Loukissas, Y. A. (2016). Taking Big Data Apart: Local Readings of Composite Media Collections. Information, Communication & Society, 20(5), 651–664. https://doi.org/10.1080/136 9118X.2016.1211722 Lozano-Hemmer, R., Barrios, J. L., and Labastida, A. (2015). Conversation on the Layout of the Exhibition. In Rafael Lozano-Hemmer: Pseudomatismos (pp. 184–221). muac, Museo Universitario Arte Contemporáneo, unam. https:// www.lozano-hemmer.com/texts/ bibliography/monographs/catalogue_ pseudomatisms.pdf Lupton, D. (2016). The Quantified Self: A Sociology of Self-tracking. Polity. Lynch, M. (1991). Science in the Age of Mechanical Reproduction: Moral and Epistemic Relations between Diagrams and Photographs. Biology & Philosophy, 6(2), 205–226. https:// doi.org/10.1007/BF02426838 Lyotard, J.-F. (2005). The Postmodern Condition: A Report on Knowledge (Repr). Manchester University Press.

LM

MacEachren, A. M. (2004). How Maps Work: Representation, Visualization, and Design (Pbk. ed). Guilford Press.

Sources

220221_CritViz_Book_Full_revisions.indd 240

7/20/22 9:49 AM

241

M

M

Magnello, M. E. (2012). Victorian Statistical Graphics and the Iconography of Florence Nightingale’s Polar Area Graph. bshm Bulletin: Journal of the British Society for the History of Mathematics, 27(1), 13–37. https://doi. org/10.1080/17498430.2012.618102 Magnet, S. (2011). When Biometrics Fail: Gender, Race, and the Technology of Identity. Duke University Press. Mann, M. E. (2014). The Hockey Stick and the Climate Wars: Dispatches from the Front Lines (Paperback edition). Columbia University Press. Mann, M. E., Bradley, R. S., and Hughes, M. K. (1999). Northern Hemisphere Temperatures during the Past Millennium: Inferences, Uncertainties, and Limitations. Geophysical Research Letters, 26(6), 759–762. https://doi. org/10.1029/1999GL900070 Manovich, L. (2002). The Language of New Media. mit Press. Marcos. (2004). Ya Basta! Ten years of the Zapatista Uprising (Ž. Vodovnik, Ed.; 1st ed). AK Press. Markley, R., and Cohen, T. (2012). Time. In Telemorphosis: Theory in the Era of Climate Change. Open Humanities Press. McCandless, D. (2012). Information is Beautiful. Collins. McCandless, D. (2015). Welcome to our Office: David McCandless, renowned data journalist and speaker — YouTube. https://www.youtube.com/ watch?v=BkQEUJn86MU McDonald, L. (2006). Florence Nightingale on Society and Politics, Philosophy, Science, Education and Literature: Collected Works of Florence Nightingale, Volume 5. Wilfrid Laurier University Press.

Menking, A., and Rosenberg, J. (2020). wp:not, wp:npov, and Other Stories Wikipedia Tells Us: A Feminist Critique of Wikipedia’s Epistemology. Science, Technology, & Human Values, 016224392092478. https://doi. org/10.1177/0162243920924783 Mignolo, W., and Walsh, C. (2018). On Decoloniality: Concepts, Analytics, Praxis. Duke University Press. Milner, Y. (2018, April 7). An Open Letter to Facebook from the Data for Black Lives Movement. Medium. https://medium. com/@YESHICAN/an-open-letterto-facebook-from-the-data-for-blacklives-movement-81e693c6b46c Minard, C.-J. (1862). Des tableaux graphiques et des cartes figuratives. impr. de Thunot. Miranda, C. (2017, January 27). Artist Theaster Gates on W.E.B. DuBois and what Donald Trump doesn’t get about Chicago — Los Angeles Times. https://www.latimes.com/ entertainment/arts/miranda/la-etcam-theaster-gates-regen-projects20170127-story.html Moere, A. V., and Hill, D. (2012). Designing for the Situated and Public Visualization of Urban Data. Journal of Urban Technology, 19(2), 25–46. https://doi.org /10.1080/10630732.2012.698065 Mollerup, P. (2015). Data Design: Visualising Quantities, Locations, Connections. Bloomsbury Academic, an imprint of Bloomsbury Publishing Plc. Moore, P., and Robinson, A. (2016). The Quantified Self: What counts in the Neoliberal Workplace. New Media & Society, 18(11), 2774–2792. https:// doi.org/10.1177/1461444815604328

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 241

7/20/22 9:49 AM

242

MNO

Moro, S. (2021). Mapping Paradigms in Modern and Contemporary Art: Poetic Cartography. Routledge. Morozov, E. (2013). To Save Everything, Click Here: Technology, Solutionism and the Urge to Fix Problems that Don’t Exist. Allen Lane. Morton, T. (2013). Hyperobjects: Philosophy and Ecology after the End of the World. University of Minnesota Press. Morton, T. (2018). Being Ecological. Müller, M. (2015). Assemblages and Actornetworks: Rethinking Sociomaterial Power, Politics and Space: Assemblages and Actor-networks. Geography Compass, 9(1), 27–41. https://doi.org/10.1111/gec3.12192 Munster, A. (2009). Data Undermining: The Work of Networked Art in an Age of Imperceptibility. New Radio and Performing Arts, Inc. https://doi. org/10.7551/mitpress/8982.003.0007 Neurath, M., and Kinross, R. (2009). The Transformer: Principles of Making Isotype Charts. Hyphen Press. Noble, S. U. (2018). Algorithms of Oppression: How Search Engines Reinforce Racism. New York University Press. Nold, C. (2009). Emotional Cartography: Technologies of the Self. Publisher not identified. http:// emotionalcartography.net/ O’Neil, C. (2017). Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy. Penguin Books. Oxford English Dictionary. (2020, September). “aesthesis | esthesis, n.”. https://www. oed.com/view/Entry/3234?redirected From=aesthesis+

P New Personalized Web is Changing What We Read and How We Think. Penguin Press. Pearce, M. W. (2014). The Last Piece Is You. The Cartographic Journal, 51(2), 107– 122. https://doi-org.arts.idm.oclc.org/ 10.1179/1743277414Y.0000000078 Pearce, M. W. (2017). Coming Home to Indigenous Place Names in Canada [Map]. Canadian-American Center, University of Maine. https://umaine. edu/canam/publications/cominghome-map/coming-home-indigenousplace-names-canada-pdf-download/ Perec, G. (1997). Species of Spaces and Other Pieces (J. Sturrock, Ed.). Penguin Books. (Original work published 1974). Perry, C. (2010). Review of Heath Bunting 0–9 The System Deluxe, Helix Arts, 2009. Art Monthly, 334, 30. Pevsner, N. (2005). Pioneers of Modern Design: From William Morris to Walter Gropius (4., rev.expanded ed). Yale University Press. (Original work published 1936). Pickles, J. (2004). A History of Spaces: Cartographic Reason, Mapping, and the Geo-coded World. Routledge. Pickles, J. (2006). On the Social Lives of Maps and the Politics of Diagrams: A Story of Power, Seduction, and Disappearance. Area, 38(3), 347–350. Plasa, C. (2014). Ekphrastic Poetry and the Middle Passage: Recent Encounters in the Black Atlantic. Connotations, 24(2), 290–324.

Pariser, E. (2012). The Filter Bubble: How the

Sources

220221_CritViz_Book_Full_revisions.indd 242

7/20/22 9:49 AM



PQR

Playfair, W. (1801). The Statistical Breviary: Shewing, on a Principle Entirely New, the Resources of Every State and Kingdom in Europe; Illustrated with Stained Copper-plate Charts the Physical Powers of Each Distinct Nation with Ease and Perspicuity: To which is Added, a Similar Exhibition of the Ruling Powers of Hindoostan. T. Bensley, Bolt Court, Fleet Street. Polak, E. and Waag Society. (2002). Realtime Amsterdam. Porter, T. M. (1995). Trust in Numbers: The Pursuit of Objectivity in Science and Public Life. Princeton University Press. Powell, J. (2019). Google Releases New Tool to Gauge Pollution. China Daily Global. https://global. chinadaily.com.cn/a/201910/11/ WS5d9fd824a310cf3e3556fd94.html Price, B. (1992). Frank and Lillian Gilbreth and the Motion Study Controversy, 1907–1930. In D. Nelson (Ed.), A Mental Revolution: Scientific Management since Taylor (pp. 55–78). Ohio State University Press. Quantified Self. (2018, September 22). Be An Exhibitor. QS18 Quantified Self Conference. http://qs18. quantifiedself.com/exhibitor/ Rabinbach, A. (1992). The Human Motor: Energy, Fatigue, and the Origins of Modernity. University of California Press. Rancière, J. (2010). Dissensus: On Politics and Aesthetics (S. Corcoran, Trans.). Continuum.

R

243

Ratti, C., and Claudel, M. (2018). senseable City — Conversation. In M. Del Signore and G. Riether (Eds.), Urban Machines: Public Space in a Digital Culture (pp. 208–214). ListLab. Rediker, M. (2008). The Slave Ship: A Human History. John Murray. Rekacewicz, P. (2013a, March 1). The Airport Malls. Le Monde Diplomatique. https://mondediplo.com/2013/03/ 12mapmaking Rekacewicz, P. (2013b, December 22). Cartes en colère. Le Monde diplomatique. https://mondediplo. com/2013/03/12mapmaking Rich, D. (2019, December 1). Interview with Peter Hall [Audio call and email]. Ridler, A. (2018). Myriad (Tulips). http:// annaridler.com/myriad-tulips Ridler, A. (2019, September 6). Interview with Peter Hall [Personal communication]. Ridler, A. (2020, September 23). Correspondence with Peter Hall [Personal communication]. Ridler, A., and Pfau, D. (2019). Bloemenveiling. Bloemenveiling. https://bloemenveiling.bid/ rndr. (2019). Open Highway: A Real Time Light Instal­lation Visualizing Highway Vehicles in Scale 1:1. https://rndr. studio/projects/open-highway/ Robertson, B. (2020, October 10). Interview with Peter Hall [Audio call and email]. Robinson, A. H. (1982). Early Thematic Mapping in the History of Cartography. University of Chicago Press. Rosenberg, D. (2013). Data Before the Fact. In L. Gitelman (Ed.), ‘Raw data’ is an Oxymoron (pp. 15–41). The mit Press. Rosenberg, D., and Grafton, A. (2010). Cartographies of Time (1st ed). Princeton Architectural Press.

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 243

7/20/22 9:49 AM

244

RS

Rosenberg, T. E. (2007). Designs on Critical Practice? In Reflections on Creativity: Exploring the Role of Theory in Creative Practices. Dundee: Duncan of Jordanstone College. Ross, R. (2006). Perils of Precision. In J. Abrams, P. A. Hall, and Design Institute (Eds.), Else/Where: Mapping — New Cartographies of Networks and Territories (pp. 184–189). University of Minnesota Press. Sadler, S. (1998). The Situationist City. mit Press. Said, E.W. (1978). Orientalism. Routledge & Kegan. Saldarriaga, J. F., Kurgan, L., and Brawley, D. (2017). Visualizing Conflict: Possibilities for Urban Research. Urban Planning, 2(1), 100–107. https:// doi.org/10.17645/up.v2i1.880 Santos, B. de S. (2016). Epistemologies of the South: Justice Against Epistemicide. Routledge. Scott, J. C. (1998). Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. Yale University Press. Sedaris, D. (2014, June 30). Stepping Out. Living the Fitbit Life. The New Yorker. https://www.newyorker.com/ magazine/2014/06/30/stepping-out-3 Sekula, A. (1986). The Body and the Archive. October, 39, 3. https://doi. org/10.2307/778312 Serres, M., and Latour, B. (2008). Conversations on Science, Culture, and Time (5. print). University of Michigan Press. (Original work published 1995). Shannon, C. E., and Weaver, W. (1964). The Mathematical Theory of Com­munication. University of Illinois Press.  



ST

Sharon, T., and Zandbergen, D. (2017). From Data Fetishism to Quantifying Selves: Self-tracking Practices and the Other Values of Data. New Media & Society, 19(11), 1695–1709. https://doi.org/ 10.1177/1461444816636090 Shneiderman, B. (1996). The Eyes Have It: A Task by Data Type Taxonomy for Information Visualizations. Proceedings of the 1996 ieee Symposium on Visual Languages, 336-. http://dl.acm.org/citation. cfm?id=832277.834354 Southworth, M., and Southworth, S. (1982). Maps, a Visual Survey and Design Guide (1st ed). Little Brown. Stake-Doucet, N. (2020). The Racist Lady with the Lamp. Nursing Clio. https:// nursingclio.org/2020/11/05/the-racistlady-with-the-lamp/ Stray, J. (2016, March 24). The Curious Journalist’s Guide to Data. Columbia Journalism Review. https://www.cjr. org/tow_center_reports/the_curious_ journalists_guide_to_data.php/ Sullivan, L. (1896, March). The Tall Office Building Artistically Considered. Lippincott’s Magazine, 403–409. Thaler, R. H., and Sunstein, C. R. (2009). Nudge: Improving Decisions about Health, Wealth and Happiness. Penguin Books. The Economist. (2017, May 6). Data is Giving Rise to a New Economy. The Economist. https://www. economist.com/briefing/2017/05/06/ data-is-giving-rise-to-a-new-economy Thomä, D. (2019). Troublemakers: A Philo­sophy of Puer Robustus. Polity Press. Thompson, E. P. (1967). Time, Work-discipline, and Industrial Capitalism. Past and Present, 56–97.

Sources

220221_CritViz_Book_Full_revisions.indd 244

7/20/22 9:49 AM



TU

Toscano, A., and Kinkle, J. (2015). Cartographies of the Absolute. John Hunt Publishing. Tsing, A. L. (2015). The Mushroom at the End of the World: On the Possibility of Life in Capitalist Ruins. Princeton University Press. Tuck, E. (2009). Suspending Damage: A Letter to Communities.(Essay). Harvard Educational Review, 79(3), 409–427. https://doi.org/10.17763/ haer.79.3.n0016675661t3n15 Tuck, E., and Yang, K. W. (2012). Decolonization is not a Metaphor. Decolonization: Indigeneity, Education & Society, 1(1). Tufte, E. R. (2001). The Visual Display of Quantitative Information. Graphics Press. Turnbull, D., and Watson, H. (1993). Maps are Territories: Science is an Atlas: A Portfolio of Exhibits (University of Chicago Press ed). University of Chicago Press. Tversky, B. (1993). Cognitive Maps, Cognitive Collages, and Spatial Mental Models. In A. U. Frank and I. Campari (Eds.), Spatial Information Theory: A Theoretical Basis for gis, Proceedings cosit ’93 (pp. 14–24). Springer. University of Cambridge. (2011, July 5). Wittgenstein’s Camera. https:// www.cam.ac.uk/research/news/ wittgenstein%E2%80%99s-camera Urist, J. (2015, May 14). From Paint to Pixels. The Atlantic. https://www.theatlantic. com/entertainment/archive/2015/05/ the-rise-of-the-data-artist/392399/



245

VW

Vanky, A. (2016, Spring). Make Data Make Sense: The Importance of  Visualization in Data Analytics. iqt Quarterly, 7(4), 17–19. Vossoughian, N. (2011). Otto Neurath: The Language of the Global Polis (D. Camp, Ed.). NAi Publishers; Art Data [distributor]. Walters, J. (2017, March 23). Boston Public Schools Map Switch Aims to Amend 500 Years of Distortion. The Guardian. https://www.theguardian. com/education/2017/mar/19/bostonpublic-schools-world-map-mercatorpeters-projection] Ware, C. (2004). Information Visualization: ­Perception for Design. Morgan Kaufman. Ware, C. (2013). Information Visualization: Perception for Design (Third edition). Morgan Kaufmann. Wark, M. (2013). The Spectacle of Disintegration (First edition). Verso. Wark, M. (2015, May 1). Cognitive Mapping. Public Seminar. https://publicseminar. org/2015/05/cognitive-mapping/ Warren, M. (2019, October 25). Indigenous Elder Slams ‘hollow and tokenistic’ Consultation by Sidewalk Labs. The Star. https://www.thestar.com/ news/gta/2019/10/25/indigenouselder-slams-hollow-and-tokenisticconsultation-by-sidewalk-labs.html Weizman, E. (2017). Forensic Architecture Violence at the Threshold of Detectability. Princeton University Press.

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 245

7/20/22 9:49 AM

W 246 Whitney, V., and Ho, B. (2019, December 11). A First Step Toward the Future of Neighborhood Design. Medium. https://medium.com/sidewalktalk/a-first-step-toward-the-future-ofneighborhood-design-a2777ad69550 Wigley, M. (1997). Planetary Homeboy. any, 17, 16–73. Williams, K. (2013). The Weight of Things Lost: Self-knowledge and Personal Informatics. Personal Informatics. http://v1.personalinformatics.org/ docs/chi2013/williams.pdf Willis, A.-M. (2006). Ontological Designing. Design Philosophy Papers, 4(2). https://doi.org/10.2752/14487130 6X13966268131514 Wolf, G. (2010, May 2). The Data-Driven Life. The New York Times. https:// www.nytimes.com/2010/05/02/ magazine/02self-measurement-t. html?searchResultPosition=1 Wood, D. (2010). Rethinking the Power of Maps. Guilford Press. Wood, J., Isenberg, P., Isenberg, T., Dykes, J., Boukhelifa, N., and Slingsby, A. (2012). Sketchy Rendering for Information Visualization. Visualization and Computer Graphics, ieee Transactions On, 18(12), 2749– 2758. https://doi.org/10.1109/ TVCG.2012.262 Wood, M. (2000). Blind Memory: Visual Representations of Slavery in England and America 1780–1865. Manchester University Press.

Sources

220221_CritViz_Book_Full_revisions.indd 246

7/20/22 9:49 AM

220221_CritViz_Book_Full_revisions.indd 247

7/20/22 9:49 AM

Index

220221_CritViz_Book_Full_revisions.indd 248

7/20/22 9:49 AM



AB

Note: Page locators in italic refer to figure captions. Aboriginal land rights 40–41 maps 54 Actor-Network Theory (ant) 12–14, 16, 36, 61, 197 Adas, M. 127, 128 aesthetics 198, 202–203, 206, 214 etymology 186 function and 186–194 perception and 194–195 affordances 34–35 air quality visualizations 161 Albrecht, K. 111–112, 132, 133 Aleppo 175–177 América Invertida 231, 231 Amoore, L. 139–140, 142, 152, 181 Amsterdam Real Time 165 anthropometry 137–138, 139, 145–147 Apollonian visualizations 229 Appadurai, A. 63, 200, 201 Arntz, G. 69, 72, 204, 205 arts, satirical and playful practices of 142–145, 152 Arup 159, 160, 163 assemblages 12, 13, 14, 16 data 11, 12, 13, 14, 22, 89–90, 97 Athens Charter 165–166 attribute data 92 ‘average man’ 135–136, 138 Ball, R. 106, 145–146, 146 Baudrillard, J. 52, 216 beautifying data 16, 162, 186, 189 Beck, H. 70 Beehive Collective 189, 206–207, 208, 209, 210, 227 behaviour shifts 160, 178 Bentham, J. 58 Bergson, H. 49, 125, 126 Bertillon, A. 137–138, 139



249

BC

Bertillonage 137–138, 139 Bertin, J. 18–19, 112, 186 bicycle barometers 178 big data 73, 90–91, 103, 160 big u 163–164 Bio Mapping projects 166–168, 170–171 biometrics 139–142, 168 challenging norms of 142–147 biopower 140 Bjarke Ingels Group (big) 163–164 Black American exhibit, Paris Exposition 64–68, 75, 211–214 Lives, Data for 104 oppression in US 214, 215, 216 Bloemenveiling 78, 79 bodies of quantified self 147–148 Booth, C. 99–100 BorderXing 181 Borges, J.L. 97 Brazilian timber exports 38, 39 Brody, N. 16 Brookes slave ship diagram 55–59, 75 Browne, S. 59, 145 Bunting, H. 142–143, 180–182 Bureau d’études 38, 186–189, 208–209 Cadman, T. 37, 38 Callon, M. 14, 61 Canada census categories 98–99  Coming Home to Indigenous Place Names in Canada 40, 226–227 Indigenous people and connection to land 225–231 Neighbourhood Improvement Areas 101–102 planning controversy 172 ‘They Would Not Take Me There’ 153–155 canon, Western 45–46

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 249

7/20/22 9:49 AM

C 250



capitalism extraction and 222–223, 229–230, 233 measurement of time 126–127 ‘patchy’ 178 Scientific Management 128–130 Cartesian dualism 31–32 cartographic silences 24 cartographic thinking 209–210 cartography 18–19, 20, 49, 51, 198–199 challenging Western concepts of 54 communication science 191, 195, 195 critical 17, 50–52 cartoneros project 26, 26, 27 Cartwright, S. 118, 118 categorization of populations 63–64, 77, 86–87, 96–107, 201 ‘Celestial Emporium of Benevolent Knowledge’ 97 census Canadian categories 98–99 data 91, 101 Indian population 63–64 Center for Spatial Research 175, 177 Center for Urban Pedagogy (cup) 74 Chalabi, M. 214, 215, 216 Champlain, S. de 153–155 Chaplin, C. 130, 131 cinematography 125, 126 cities, data and 159–179 Clarkson, T. 55, 58 Claudel, M. 160 cleaning data 83, 103 climate change 36–40, 168, 169 increases in temperature 87–89 nature projects 169–170, 171 planning for extreme weather 163–164, 168 unexpected alliances 179 Codebò, A. 27, 28 cognitive load 34, 194–195 cognitive mapping 38, 40, 54, 171, 202–203 cognitive systems 32–35 Cohen, B. 68, 72–73 Cohen, J. 149

collage maps 165, 203 collective mapping projects 94–95, 166, 205, 206, 210, 227–228 College Performance project 119, 119 colonialism asymmetrical relationships 231, 231, 233 extraction and 222–223, 229–230, 233 in India 62–64 narrative map of a colonist's journey 153–155 numbers and counting 63–64, 201 settler 222, 228, 231 time discipline of subjects of 127–128 Coming Home to Indigenous Place Names in Canada 40, 226–227 communities, maps within 209–210 community-building 68–74 community mapping projects 94–95, 205, 206, 210 composite portraiture 136–137, 141–142 concealment 17, 185 Conflict Urbanism 175–177 consumerism, diffused spectacle of 168–169 Copenhagen 161–162 Corner, J. 69, 196 coronavirus pandemic 9, 9 counter-mapping 27–28, 166, 175, 208, 209 Crampton, J. 17, 20, 28, 51 creator of a work 47 Creswell, J.W. 93 Crimean War 59–62, 75 criminals, physical characteristics of 137–139, 141 critical cartography 17, 50–52 critical framework 11–17 critical theory 221–222 criticality 35, 83, 106–107 critique 35–41 Csonto, J. 228

CD

damage-centered research 230 data assemblages 11, 12, 13, 14, 22, 89–90, 97 ‘data biographies’ 14 data, etymology of 21–22, 86

Index

220221_CritViz_Book_Full_revisions.indd 250

7/20/22 9:49 AM



DE

data fetishism 117, 134 Data for Black Lives 104 data journeys 103–104 data literacy 9, 16 data paintings 214 data sets 76–79 data sonifications 111–112 Data4Change 95, 108–114 Dávila, P. 18, 40, 41, 209 Dawson, A. 164 Debord, G. 164–165, 166, 168, 171, 203, 203n decoloniality 24, 55, 231–233 decolonization 179, 222, 223, 228, 233 of knowledge 155, 198, 232 ‘delinking’ data visualization from Western paradigms 24–28, 221–233 dérives 165–166, 203n Descartes, R. 31–32, 46 desire-centred research 230 deviancy, understanding 135–139 Dick, M. 17, 18, 19 D’Ignazio, C. 12, 16 dimensions of critical visualization 75 dispositif 11, 90, 97 distributed cognition 34 dot maps 50 double-sided character of mapping 69, 196 Dourish, P. 30, 177 Drucker, J. 18, 24, 33–34, 36, 45, 200 Du Bois, W.E.B. 64–68, 75, 211–214 Duarte, F. 26, 159 dynamic nominalism 98 Eadington 142 ecological degradation 36, 38, 164, 179 Edney, M. 63, 72 ‘efficiency principle’ 18 Elahi, H. 140, 140, 141 empowerment/disempowerment 10, 11, 16 engines of discoverability 98 Ethiopia, visual impairment in 111–112 eugenics 136–139 everyday life 152, 165, 168, 179 examined lives 147–152



251

EFG

extraction 222–223, 229–230, 233 extreme weather planning 163–164, 168 Facebook 91, 104, 122 facial recognition technologies 144, 145, 147 facts 21–22, 86 Fanon, F. 225 Farocki, H. 72–73, 73 fatigue 124, 125 Felton, N. 120–122, 123 Feltron Annual Reports 120–121, 122 field of data visualization 17–24 fingerprints 139, 145 Fisher, R. 77 Fitbits 131 floating data nets 160 Forensic Architecture 177 In-Formation 72–73, 73 Foucault, M. 11, 22, 35, 58, 83, 90, 97, 140 framework, critical 11–17 Freire, P. 35, 227 Friendly, M. 46, 48–49, 50 Frigo, A. 149–150 Fry, T. 18, 120 function and aesthetics 186–194 Funkhouser, H.G. 45–46 Gall-Peters projection 51, 51, 198–199 Galton, F. 136–137, 139, 141 Gates, T. 214 Gaussian curve 135 gender 68, 72, 96, 147, 190 generative criticism 36 geographic imaginaries 102 Geographic Information Systems (gis) 18, 19, 20, 175 Gestalt principles 192 Gibson, J. 32, 34–35 Gilbreth, F. 128, 129, 129, 130 Gilbreth, L. 128, 129, 129, 130 Gitelman, L. 23, 85, 86, 102 Glissant, É. 31, 223 Global Positioning System (gps) 19–20, 25, 159, 166

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 251

7/20/22 9:49 AM

252

GH

‘god trick’ 12, 47, 229 Gómez-Barris, M. 223 Google 15, 16 Google Earth Outreach 161 governmentality 22, 101, 160 Grafton, A. 45, 49 graphesis 24, 33–34 Great Trigonometrical Survey (gts) 62–63, 75 Great War Isotype 70–71, 71 Green, D. 136, 137 Greenfield, A. 172 Greenpeace 38, 39 Grupo de Arte Callejero 209–210 Guide psychogéographique de Paris 165–166 Hacking, I. 11, 60, 98, 99, 100, 102, 135, 136, 137 Haeckel, E. 139 The Hairs on Your Head are Numbered 132, 133, 134 Hall, P. 18, 20, 36, 121, 177, 217 Hall, S. 210–211 Halley, E. 20, 20, 46 Halpern, O. 22, 23, 160–161 Hamilton, P. 38, 39 hand-sketched style visualizations 189, 190, 190, 191, 192–194, 214 Hannah, D. 170 haptic visualization 52–55 Haraway, D. 12, 102, 103, 104, 147, 229 Harley, J.B. 24, 32, 52 head shapes 106, 145–146, 146 health, self-tracking of 117, 118, 123, 134, 149 Hear the Blind Spot 111–112 hector 174–175, 175, 176 HeHe 178, 179 Hermann, M. 153–155 Hill, D. 160, 173, 178 histories, disruptive 45–79 ‘hockey stick graphs’ 87–89 Holmes, B. 38, 167 human-computer interaction (hci) 18, 33–34 ‘humanistic’ design approach 24, 33–34



HIJ

Hutchins, E. 34 hyperobjects 36 ibm pavilion, World’s Fair 1964 22–23, 23 Iceland 177 Iconoclasistas 94–95, 96, 204–206, 210, 227, 231 La República de los cirujas 26, 26, 27 Who owns the land? 209, 227 identities classifications 180–182 formation 225 making up 11, 98 ImageNet 77 indexical data 92 India 201 Great Trigonometrical Survey (gts) 62–63, 75 population census 63–64 Indigenous knowledge 40–41, 155, 179 mapping 40–41, 105–106 people and connection to land 225–231 place names 40, 105, 106, 225–226, 226 voices in a colonist’s story 154–155 infographics 17–18 Ingold, T. 32, 54 internet data collection from users of 222 data map 21 shutdowns 109–111 interrogation of data 14–15, 16, 24, 106 Iris data set 77 Isotype (International System of Typographic Picture Education) 68–74, 75, 203–205 Jaar, A. 41, 150–152, 231, 233 Jackson, V. 85, 86, 102 Jameson, F. 36–38, 201, 202 Jenkins, K. 29, 52, 53 Jeremijenko, N. 169–170, 171 Jorn, A. 165, 203 Joss, S. 173, 177

Index

220221_CritViz_Book_Full_revisions.indd 252

7/20/22 9:49 AM



KLM

KeepItOn 109–111 Kinross, R. 70, 71–72, 74 Kitchin, R. 11, 17, 89, 90, 95, 97, 102, 142 knowledge collective building of 227–228 decolonization of 155, 198, 232 Indigenous 40–41, 155, 179 masculinist 103 power and access to 90–91 separation of context and 106 situated 12, 40, 84, 229, 233 visual production of 24, 33–34 Kondek, C. 132, 134 Krahe, J. 142, 143 Krajewski, S. 170 Krause, H. 14 Kühl, C. 132, 134 Kurgan, L. 19–20, 175, 200 La República de los cirujas (The Republic of Garbage Pickers) map 26, 26, 27 ‘Land Acknowledgement: Place Based Histories and Indigenous Leadership’ 228 Latour, B. 9, 35–36, 50, 89, 164, 169, 179, 196 Lauriault, T.P. 89, 90, 97, 100, 102, 173 Lefebvre, H. 30, 152, 165, 179 Level of Confidence 144, 145 ‘Life Stowing’ project 149, 150 linear history 45, 48–50, 52 Liu, Z. 33, 34 lived experience regarding data 104 A Logo for America 231, 233 Lombardi, M. 95, 96, 186 Lombroso, C. 138, 139, 142 London poverty maps 99–100 London Underground diagram 70 looping effects 11, 98, 102 Lozano-Hemmer, R. 143–144, 145 Lupton, D. 117, 147, 148, 149 Lynch, K. 171 MacEachren, A. 19, 191, 195, 195 Magnet, S. 145, 147 making data 83–107



MN

253

‘Making Policy Public’ programme 74 ‘making up people’ 11, 98 Mann, M.E. 87, 88, 89 Maps are Territories: Science is an Atlas 40, 54 Marcos 224 Marey, É.-J. 124–126, 128 masculinist knowledge 103 McCandless, D. 15–16 McVarish, E. 45 Melbourne Smart City 160 Mercator projection 50–51, 199, 199 metadata 93 Micronesian mariners 54 Mignolo, W. 24, 28, 32, 55, 225, 226–227, 231, 233 Milner, Y. 104 Minard, C.J. 46, 225 mindfulness, self-tracking as 134, 147, 149–150 Modern Times 130, 131 Moere, A.V. 160, 173, 178 Morozov, E. 122, 131 Morton, T. 36, 38 Mosaic Virus 78–79 motion capture and measurement of 124–126, 128 studies 128–129 Muybridge, E. 124–125 Myriad (Tulips) 76–79 names, place 40, 105, 106, 225–226, 226 disrupting conventions 231 narrative maps 153–155, 189, 207, 207, 210 nature 45, 169–170, 171 navigation 54 Neighbourhood Improvement Areas (nia) 101–102 Neurath, M. 27, 68, 71–72, 73, 74 Neurath, O. 27, 68–69, 70, 72, 73, 203–204, 204, 205 neutrality of data 9–10, 85, 119, 200 New York ‘Making Policy Public’ programme 74 Rebuild by Design project 163–164

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 253

7/20/22 9:49 AM

254

NOP

Nightingale, F. 59–62 Nold, C. 166–168 normal, defining 135–139 ‘norms’, challenging 134, 142–147, 180–182 Nuage Vert 178, 179 numerical gaze 200–201 objectivity 47, 102, 103, 192 O’Neil, C. 17 OneTrees project 169, 170 opacity and transparency 223–224 Open Government Partnership 173 open highway 162, 162 Opte Project 21 Orientalism 128 paintings, data 214 panopticon 58 panoramic views 208, 209, 229 Paris Exposition 1900 64–68, 75, 211–214 psychogeographic map 165–166 park project in Philadelphia 174, 175, 176 participatory mapping 27, 40, 168, 206 planning 174–179 Pearce, M. 40, 41, 105–106, 153–155, 198, 226–227, 230 Perceiving Yemen 113–114 perception and aesthetics 194–195, 198 Peters projection 51, 51, 198–199 Phenology Clock 170, 171 photography 102, 125, 126, 128 composite portraiture 136–137, 141–142 Pickles, J. 88, 194, 216, 217 place names 40, 105, 106, 225–226, 226 disrupting conventions 231 Playfair, W. 46, 47 Playful Self 142, 143 pluriversal thinking 224, 229, 233 Polak, E. 166 Polar area graphs 59–62, 75 positivism 46–47, 85, 124 poverty maps 99–102



PQR

power 12–14, 22, 36, 83–85, 117, 198 and access to knowledge 90–91 Brookes diagram as a mechanism of 58 data assemblages and exercise of 11, 12, 14, 22, 86 diagrams of 18 extractive relations of 222–223, 229–230, 233 imbalances of 16, 90–91, 222–223, 231 mapping networks of 208, 209 networked systems of 201–202 praxis 24, 55, 222, 224 preattentive processing 194 Priestley, J. 45, 48, 49 primary data 95 production of data 11, 83, 86–89, 107, 185 progress, history of 48–50 psychogeographic maps 165–168, 170–172 public visualization of urban data 178, 179 puer robustus 182 Qi, T. 119, 119 qualitative and quantitative data 45, 91–95 Quantified Self (QS) 117–124, 147–152 critique 130–139 Quetelet, A. 135, 136, 138 quipu 52–55, 75 Rabinbach, A. 123–124, 125, 126, 130 race census classifications 98–99 facial recognition technologies and 147 hierarchies of gender and 139 intersection of surveillance and 58, 145 physiognomy 135–136 stereotypes 72, 73, 204–205, 204 Ratti, C. 26, 159, 160 raw data 85–86, 200 ‘real time’ data, public displays of 178, 179 visualizations of cities in 159, 160 Real Time Rome 159, 160 Rebuild by Design project 163–164 Rekacewicz, P. 193–194, 209

Index

220221_CritViz_Book_Full_revisions.indd 254

7/20/22 9:49 AM



RS

representability 201 representation as translation 196–217 Rich, D. 174, 176 Ridler, A. 76–79 risk profiling 139–140 rndr 162, 162 road signs 71–72 Robertson, B. 108, 109, 111, 112, 114 Robinson, A. 20, 49, 50, 51 Rosenberg, D. 21, 22, 45, 49 Ross, R. 10 Rothera, A. 142, 143 Rwanda Project 150–152 Said, E. 128 Salavon, J. 142 Santa Efigênia Agropecuária 38, 39 Santos, B. de S 221–222 satellite imaging 19–20 of Aleppo 175, 177 satellite navigation 166–168 scales of data 92 scientific data, production of 87–89 Scientific Management 128–130 scientific method 46–47 Searching for Africa in life 41, 152 secondary data 95 Sedaris, D. 131 self, data and 117–155 self-tracking 117–124, 140, 147–152 critique 130–139 selfies 142 Senseable City Lab 160 Real Time Rome 159, 160 Trash|Track project 24–25, 25, 26, 27, 159 Serres, M. 50 Sharon, T. 119, 133–134, 147, 149 Shin, J. 174, 176 Shneiderman, B. 19, 186, 194 Sidewalk Labs 172–173 Signs of Life 150–152 situated data 12, 27, 102, 103, 104, 146 knowledge 12, 40, 84, 229, 233



ST

255

Situationist project 165, 168, 170–171, 172, 177, 203, 208 SizeChina 106, 145–146, 146 slave ship diagram 55–59, 75 smart cities 159–165, 172–173, 179 ‘accidentally smart’ 177 sociology of translation 61 Songdo, South Korea 161 spaces, making up 102 speaking for itself, data 85, 102–103, 199–200 the spectacle 164–165, 168–169 spectacular bodies 148 statistics 9, 20, 45, 49–50, 60, 135, 200 Statistics Canada 98–99, 100, 102 Status Project 142–143, 180–182 stereotypes, cultural and racial 73, 72–73, 204–205, 204 Stockport Emotion Map 166, 167 story maps 153–155, 189, 207, 207, 210 Street Vendor project 74 surveillance systems 58, 143 challenging biometric 143–145 disrupting 140 Taylor, F.W. 129, 130 Taylorism 129–130 ‘Taylorism within’ 122–123 temperature data 87–89 terminology 21–23 ‘Terrorist’ map 180, 181 tertiary data 95 thematic maps 20, 49, 51 ‘They Would Not Take Me There’ 153–155 ‘Think’ 22–23, 23 Thomä, D. 181 Thompson, E.P. 127, 129–130 Thousand Little Brothers 140, 140 Tilley, A. 146 timber exports 38, 39 time discipline 126–128 timelines, linear 45, 48, 49, 50

Critical Visualization

220221_CritViz_Book_Full_revisions.indd 255

7/20/22 9:49 AM

256

TUVW

Toronto Neighbourhood Improvement Areas 101–102 planning controversy 172 Torres-García, J. 231, 231 totalizing views 208, 209, 229 trade winds map 21, 21 transformer, role of 69–70, 74, 75 transparency and opacity 223–224 Trash|Track project 24–25, 25, 26, 27, 159 tree maps 169, 170 Tsing, A.L. 178 Tuck, E. 31, 228, 230 Tufte, E. 10, 17, 18, 20, 46, 48, 85, 186, 199 Turnbull, D. 40, 54 Tversky, B. 203 United Nations Framework Convention on Climate Change (unfccc) 37, 38 urban data, public visualizations of 178 urban planning 161, 172–173, 208 community consultation 168, 172 for extreme weather 163–164 participatory 174–179 visualization tools to address complexity of 172, 173



WYZ

who/when/why/what questions 14–15, 16, 24, 106 Wikipedia 15, 16 Williams, K. 148, 148 Wittgenstein, L. 141 Wolf, G. 117, 118, 120, 122–123 Wood, D. 20, 171 Wood, J. 190, 190, 191, 192 Wood, M. 56, 57, 58 World Government 186–189, 208 world views 93 Yang, K.W. 228, 230 Yemen Polling Center (ypc) 113–114 Zandbergen, D. 119, 133–134, 147, 149 Zoom Pavilion 143–144

vincularidad 226–227 visual impairment 111–112 ‘Visual Information-Seeking Mantra’ 194 visual style 190–194 Walsh, C. 24, 28, 32, 55, 225, 226–227, 231, 233 Wark. M. 165, 168, 169, 202–203, 206 waste tracking projects 24–27, 159 Watson, H. 40, 54 Western attitudes to non-Western thinking 97 canon 45–46 paradigms, visualization beyond 24–28, 54, 221–233 Who owns the land? 209, 227

Index

220221_CritViz_Book_Full_revisions.indd 256

7/20/22 9:49 AM