# Figure Concept and Method

*Edited by* Celia Lury William Viney Scott Wark

Figure

# Celia Lury • William Viney • Scott Wark Editors Figure Concept and Method

*Editors* Celia Lury Centre for Interdisciplinary Methodologies University of Warwick Coventry, UK

Scott Wark Centre for Interdisciplinary Methodologies University of Warwick Coventry, UK

William Viney Department of Anthropology Goldsmiths, University of London London, UK

ISBN 978-981-19-2475-0 ISBN 978-981-19-2476-7 (eBook) https://doi.org/10.1007/978-981-19-2476-7

© Te Editor(s) (if applicable) and Te Author(s) 2022. Tis book is an open access publication.

**Open Access** Tis book is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this book are included in the book's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the book's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

Te use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specifc statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

Te publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. Te publisher remains neutral with regard to jurisdictional claims in published maps and institutional afliations.

Cover illustration: © Stefanie Posavec 2021

Tis Palgrave Macmillan imprint is published by the registered company Springer Nature Singapore Pte Ltd.

Te registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

# **Preface**

#### **Knights of the Oblong Table**

—Di Sherlock When I call them *Knights of Te Round Table* it's a spur to the collective wit. Te nomenclature derided, others are profered, dismissed, until, all things considered, someone comes up with *Knights of Te Oblong Table.* Tere we have it.

Te confederacy shifts dune-like, presence, absence confgure, reconfgure in the uncertain wind. Te Table a stout ship, the Crew vociferous – rifng, roaring, cursing, complaining, joking, jibing, expleting, explaining,

sparing, sparring, fooling, fnagling, loquacious, voracious, complicit, explicit, hopeful, doubtful, always respectful always remembrance.

No captains stowaways hostages tourists.

Passengers by invitation only.

# **Acknowledgements**

Te editors would like to acknowledge the help and support of their colleagues from the project 'People Like You': Contemporary Figures of Personalisation*—*Sophie Day, Helen Ward, Roz Redd, and Yael Gerson and the artists associated with this project—Felicity Allen, Stefanie Posavec, and Di Sherlock.

Tis volume was collated from presentations given at a conference called *Figurations: Persons In/Out of Data*, which was organised under the auspices of the 'People Like You' project and held at Goldsmiths, University of London, December 16—17, 2019. We would like to thank our keynote speakers, Wendy H. K. Chun, Jane Elliott, John Frow, and AbdouMaliq Simone, all of our presenters, and everyone who attended the conference and participated in discussions. In particular, we would like to thank our conference assistants, Avery Delaney, Stephanie Guirand, Ming-Te Peng, and Saba Zavarei, for their hard work, and Lizzie Malcolm and Daniel Powers from Rectangle for designing our conference poster and booklet.

Te 'People Like You' project is funded by a Wellcome Trust Collaborative Award in the Medical Humanities and Social Sciences, 2018-2022 (205456/Z/16/Z). We thank the Wellcome Trust for their support in making this volume Open Access.

# **Contents**


**ix**


# **Notes on Contributors**

**Felicity Allen** is an artist and writer whose current work is mainly focussed in two forms, the concept of the Disoeuvre (exhibited in 2019 at x-hibit, Vienna, and published by Ma Bibliothèque) and Dialogic Portraits, which she makes in series through watercolour on paper as well as textual, audio, and video recordings. Her frst series produced the twovolume artist's book *Begin Again nos 1–21*, collected by Tate and the Getty. Te third series, commissioned by Turner Contemporary, produced the flm *As If Tey Existed* (2015). She is currently making Dialogic Portraits with *Refugee Tales* and with the cross-disciplinary research project 'People Like You'. See also www.felicityallen.co.uk.

**Srishti Bhatnagar** is a medical anthropologist interested in health, gender, and intersectionality. She obtained an MSc from the University of Delhi, with a specialization in Social Anthropology and has held research positions at Te Institute of Economic Growth and Purple Audacity Research & Innovation, New Delhi. She has conducted ethnographic research on the social and cultural perceptions of asthma, air pollution, and sexual and reproductive health. Currently, she's pursuing a qualitative UI study on the impact of COVID-19 pandemic on food delivery applications.

**Liliana Bounegru** is Lecturer in Digital Methods at the Department of Digital Humanities, King's College London. She is also co-founder of the Public Data Lab and afliated with the Digital Methods Initiative in Amsterdam and the médialab, Sciences Po in Paris.

**Leila Dawney** is a cultural geographer and theorist of power, afect, and embodiment. Her research concerns the forms of experience and subjectivity that are produced in and through spaces of late capitalism. She is Senior Lecturer in Human Geography at the University of Exeter and a member of the Authority Research Network.

**Sophie Day** is Professor of Anthropology at Goldsmiths, University of London (https://www.gold.ac.uk/anthropology/staf/s-day/). She founded the Patient Experience Research Centre at Imperial College London with Helen Ward in 2011.

**Melody Devries** is a PhD candidate in the Communication and Culture Department at Toronto Metropolitan University. She uses theories of material relations, performativity, and real-making to examine how farright politics become compelling to folks through online practices. She is a recipient of a Joseph-Armand Bombardier (CGS) doctoral scholarship and lead editor of the recent volume *Rise of the Far-Right: Technologies of Recruitment and Mobilization* (with Judith Bessant and Rob Watts; Rowman & Littlefeld, 2021).

**Jane Elliott** is a professor in the Department of Sociology, Philosophy and Anthropology, University of Exeter. Her main research interests are gender and employment, healthy ageing, longitudinal research methodology, combining qualitative and quantitative research and narrative. Her frst book *Using Narrative in Social Research: Qualitative and Quantitative Approaches* was published in 2005.

**John Frow** is emeritus professor in the Department of English, Te University of Sydney. He works at the boundary between literary studies and cultural studies, and his books include *Marxism and Literary History* (1986), *Cultural Studies and Cultural Value* (1995), *Time and Commodity Culture* (1997), *Accounting for Tastes: Australian Everyday Cultures* (with Tony Bennett and Mike Emmison, 1999), *Genre* (2006, revised 2015), and *Character and Person* (2014). A collection of essays, *Te Practice of Value*, was released in 2013, and another book, *On Interpretive Confict*, in 2019.

**Emma Garnett** is a Research Fellow in the Social Science and Urban Public Health Institute at King's College London. Working across sociology and science and technology studies, her ethnographic work explores the socio-material practices of environmental health science, research, and practice, primarily around air pollution. She is interested in interdisciplinary methods and critical approaches to data and is currently working with a series of participatory public health projects deploying personal sensors and bio-sensing devices.

**CeliaLury** is Professor in the Centre for Interdisciplinary Methodologies, University of Warwick. She has a long-standing interest in the ways in which 'live' methods contribute to the enactment of social worlds. Her most recent book is *Problem Spaces: How and Why Methodology Matters* (2020).

**AbdouMaliq Simone** is a Senior Professorial Fellow, Urban Institute, Te University of Shefeld. His present research examines unconventional processes of urbanization across extended urban regions in South and Southeast Asia, exploring particularly the kinds of analytical and governance frameworks necessary to address the disparate conjunctions of landscapes, aspirations, and economies prevalent in these regions. He also explores the efects of globalized generic blackness as an organizing instrument of urban life and the kinds of political instruments that are entailed in circumventing racialized control systems.

**Jayne Smith** was a gardener full time from 1987 to 2015. She is now retired but still helps out a few former clients, friends, and neighbours on an ad hoc basis.

**Matt Spencer** is an associate professor in the Centre for Interdisciplinary Methodologies at the University of Warwick. His research interests concern the relationships between knowledge, trust and computation, and he is currently working on a UKRI Future Leaders Fellowship entitled Scaling Trust: An Anthropology of Cyber Security. His previous research examined practices of modelling in computational science.

**William Viney** is a Research Fellow in the Department of Anthropology, Goldsmiths, University of London, as part of the project, 'People Like You': Contemporary Figures of Personalisation. His most recent book is *Twins: Superstitions and Marvels, Fantasies and Experiments* (2021).

**Helen Ward** is Clinical Professor of Public Health and an NIHR Senior Investigator. She co-leads the Patient Experience Research Centre at Imperial College London with Sophie Day.

**Scott Wark** is a Research Fellow for the Wellcome-funded project, 'People Like You': Contemporary Figures of Personalisation. He is based at the Centre for Interdisciplinary Methodologies at the University of Warwick. He researches online culture, amongst other things.

**Esther Weltevrede** is Assistant Professor of New Media and Digital Culture at the Department of Media Studies at the University of Amsterdam.

# **List of Figures**


# **1**

# **Introduction: Figure, Figuring and Confguration**

**Celia Lury, William Viney, and Scott Wark**

# **Introduction**

Te word "fgure" refers to many things: numbers, characters in texts, representations of persons or other entities in images; turns of phrase; abstractions and personifcations; movement or series of movements; a diagram or a short succession of notes. Alongside these many everyday uses, the fgure has a long history as a concept, migrating across disciplines and felds of research, including literary and historical studies, art criticism and history, philosophy, politics, feminism, science and technology studies, information and computer science, mathematics, design, sociology and anthropology. We do not discuss all these understandings

Centre for Interdisciplinary Methodologies, University of Warwick, Coventry, UK e-mail: c.lury@warwick.ac.uk; s.wark@warwick.ac.uk

W. Viney

C. Lury (\*) • S. Wark

Department of Anthropology, Goldsmiths, University of London, London, UK e-mail: w.viney@gold.ac.uk

here, but consider a few that have been infuential and are relevant to the contributions in this collection.

We aim to address how fgures, fguring and confguration provide a way to study complex, contemporary problems and processes that require interdisciplinary approaches. We outline how individual contributions make use of fgures, fguring and confguration. We demonstrate what is at stake in the analysis of fgures, the practice of fguring, and the compositions of confguration.

#### **Part 1: Figure**

In his essay "Figura" (1938/1959), Erich Auerbach shows why beginnings and ends often meet in fgures. Te Latin word *fgura*, from which the English word follows, came into Latin via technical Greek discourses on *morphē* and *eidos*, *schēma*, *typos* and *plasis*—a constellation of words that plays with the subtle diferences between form and plastic shape, statue and portrait. By late antiquity, Auerbach argues, a tension emerged within *fgura*, which retained both its material (concrete) and immaterial (abstract) signifcance. Te classical meaning given to *fgura* encompassed forms, shadows and speculative appearances, which tied it to a vestigial materialism (Porter 2017). For Auerbach, these philological tensions between the material and symbolic became fundamental to how pagan fgures entered Christian doctrine and devotional practice.

Long into the medieval period, *fgura* signifed ways of knowing that connected signs to material and historical life. Te Old Testament "prefgures" the New Testament, past and future are symbiotically shaped in, and indeed incarnated by, typology: Word made fesh. What Auerbach calls "fgural representations" go beyond the work of allegory or metaphor, however. Figural representations involve an economy of prediction and fulflment, an event or person signifying both itself and a second that it involves or fulfls, with each retrospective analysis serving as an opportunity to read the present in a past, while each event or type of the past has the potential to join a phenomenon in the future. Such a time is lived conditionally and *in potentia*, inhabiting what Giorgio Agamben names the "already-and-not-yet" of the fgured future and the adopted past (2000: 74, 138–45).

Auerbach's fgural representations are adaptive signs, the work ascribed to them is multiple and their referential capacity varies according to different traditions of practice and innovation. For Auerbach, fgures intervene in and transform the referent: each fgure (p)refgures, making room for the work of fguration and subsequent confguration and linking practice, including the practice of analysis, with expressions of potential. Every fgure thus implies a serial creativity in that it contains "sign qualities, denoting an object, and thing qualities, which rather confect a 'fgure' to be contemplated" (Tygstrup 2021: 238).

Edward Said characterises this feature as an "essentially Christian doctrine for believers but also a crucial element of human intellectual power and will" (2013: xxii). It is this "but also" that has allowed Auerbach's analysis to branch from the semantic meaning of the word "fgure" into an analysis of a world, what he calls a "historical situation" (1938/1959: 97), while also enabling fgures to continue to be taken to be exemplary forms of humanistic reason. Historian Hayden White writes that it is this tendency to mix concept and method that allowed fguralism to underpin "Western culture's unique achievement of identifying reality as history" (1999: 96). As we shall see, other and alternative understandings of fgures and confguration have emerged to both support and undermine this sense of humanistic achievement and its proprietary enclosure of "culture," "reality" and "history" by Western forms of thought.

Although Auerbach did not intend his methods to be either sociological or political,1 thinking with fgures has accompanied a variety of approaches in the social sciences, the humanities and political practice. For example, Georg Simmel's sociology used the fgures of the stranger, the poor and the adventurer to illustrate a more general condition, whereby "each person is called to realize his own, his very own prototype" (1971: 557). Norbert Elias developed a sociology in which great signifcance is attached to process and the interdependence of persons. For Elias, it is the social scientist's role to understand "the changing confguration of all that binds them to each other" (2007/1987: 79). Figurational

<sup>1</sup>Replying to critics of *Mimesis*, Auerbach felt others had "ascribed to the book, in praise or blame, tendencies that were far removed from me: that the method of the book is sociological, even that the tendency was socialist" (2013: 570).

sociology produces an understanding of persons as relational, provisional, performed and in-process. It aims to show how fgures articulate intermingled processes operating beyond the scale of singular or unitary entities such as the individual or society, towards their dynamic and continuous fguration.

Figuration ofers other thinkers a creative means of blending and transitioning between units, scales, orders or magnitudes of time and space. For example, fgures are central to the work of Walter Benjamin, who used them to arrest world-historical processes of modernisation. His dialectical treatment of Charles Baudelaire's poetry of a new urban modernity is representative. Baudelaire is often remembered for celebrating the fgure of the *fâneur*, the disinterested urban aesthete and observer. For Benjamin, it is with the *fâneur*'s emergent and collective opposite, the crowd, that the true "sensation of modernity" becomes apparent. In the fgure of the crowd, Benjamin argues, one can experience "the disentegration of the aura" (2003: 339), the uniqueness of things in the world, in what he calls "immediate shock experience" (2003: 343). Handled dialectically, fgures like these—others include the storyteller, the angel and the collector—ofer Benjamin a means of specifying what is both new and signifcant about modernity rather than what is simply novel.

Importantly, however, many scholars have been suspicious of the power invested in fgures and critical of the historical, political, racial and technological presumptions and prejudices of those that speak for them (see Dawney, this volume). Te arbitrarily coherent—and white and male—canon detailed above is continuous with declarations that the use of fgural representation is "Western culture's unique achievement of identifying reality as history" (White 1999: 96). Some have pointed out that fguring involves both inclusion and exclusion; for example, the fgure of "man" fgures who gets to be considered *human* by means of a series of constitutive exclusions (Mbembe 2017). Does one have to be male to count as "man"? White? Western? Wealthy? Able-bodied? For Alexander G. Weheliye, "racialisation" is crucial to this fgure's constitutive exclusions: in his terms, it "fgures as a master code within the genre of the human represented by western Man" (2014: 27). Weheliye argues that focusing on how this fgure is constituted and who it excludes allows us to take "humanity" itself as an object of knowledge. Tis particular fgure can function as a "heuristic model" for refecting on and critiquing how we produce knowledge about the world (2014: 8).

In ofering us a means of connecting word to world, fgures are doubled. Tey inform: that is, they participate in knowing, containing qualities that shape how knowing is known. But their tendency to eschew specifcation or determination also leads us to the very limit of expression and representation. Among philosophers such as Gilles Deleuze, Felix Guattari and Jean-François Lyotard, singular fgures are illustrations to be contrasted to the *fgural*, a disruptive force that is irreducible to systemic and linguistic approaches to representation and whose movement is a portal to pure sensation and becoming (Deleuze and Guattari 1994; Lyotard 2011). For Michel Serres, by contrast, "fgures of thought" are quasi-algorithmic, providing mobile protocols or operations that turn thinking into a set of parameters to be performed on something (Watkins 2020: 22-3). Serres' fgures are both natural phenomena and literary and mythological in process. Figures as various as a fox or the Challenger space craft, the Greek god Hermes or the movement of a rugby ball are equivalent in that they carry out and participate in the emergence of concepts. Te very movement of fgures makes them useful to think with.

#### **Part 2: Figuring**

Once you start hunting for fgures, it's hard not to see them everywhere. Tey inform research into all manner of things, objects and processes across disciplines and modes of scholarly enquiry. A wellcrafted fgure can lend consistency to thought, drawing together its disparate threads; indeed, for Paul De Man, "fgurality" is an essential component of philosophical speculation (1988: 13). What makes fgures so compelling to think with is that the shape they lend to thought contains an imperative—to put thinking to work. To invoke or propose a fgure *of* something, to fgure *with* something or to declare that you or someone else should fgure something *out* is to suggest, tacitly or not, that fgures involve the work of fguring. Te recourse to using fgures to illustrate a conceptual claim or to specify what's really at stake in our research has methodological implications: it prompts us to ask not only, with De Man, what thinking with fgures does to our thinking, but how it shapes our methodological engagements with the objects of our thought.

Taken as a method, fguring has productively diverse connotations. It can involve giving shape to something or, alternatively, to apprehend the shape that something already has. It can mean to calculate, solve or discover something, as in to fgure something *out*. Figuring can also mean to play a role *in* an event or happening. Te expanded conception of fguring proposed in this collection encompasses each of these disparate processes.

Figure's most obvious (and original etymological) sense is spatial, but not in the sense of form imprinted into matter: the fgure is shaped by and shapes its grounds. Tis distinction between a fgure and what surrounds it is fundamental in disciplines that study visual objects, most notably art history, though it was also the subject of Edgar Rubin's psychological research into perception (1958) and further formalised by Gestalt psychologists. Tough colloquially conceived of as opposed elements of an image or scene, Rubin's fgure and ground are intimately related because the distinction between them is articulated by what he called their "contour," or shared border. Trough experiments with images or objects containing components that reverse the relation between fgure and ground, Rubin argued that fgures that emerge from grounds exhibit something like a "shaping efect" (1958: 194-5). For W. J. T. Mitchell, the image that Rubin used to most arrestingly illustrate this efect—the eponymous "Rubin vase," which can be seen as a decorative vase on a dark background or two faces on a light background refects on its own conditions of emergence: it is what Mitchell calls a "metapicture" (2008: 9-10).

If fgures are metapictures that draw attention to their conditions of emergence, they also inform our engagements with things and processes beholden with, through or by them. Diagrams do this in a particular way. For Charles Sanders Peirce, what defnes diagrams is their capacity to depict both "a set of rationally related objects" *and* "the relations between" these objects (Peirce 1976: 316-7). By inscribing these relations, diagrams render the objects of thought operable in new domains—such as when spatial relations re-present algebra, as, for example, in Cartesian coordinates (Krämer 2010).

In his philosophical discussion of the use of fgures in mathematics, Gilles Châtelet suggests that diagrams provide us with a way of apprehending thought in the act. Figures "trac[e] contemplation" by materialising how problems are worked out (2000: 8). More generally, they capture the "gestures" that give thinking its texture or shape. Tese gestures might include tracing lines or plotting points, but they also include more complex or embodied manoeuvres, like cutting shapes out or articulating contexts—as when a fgure of a circuit conveys the sense of an electromagnetic feld's encompassing, spatial "around" and, along with it, "a new type of intuition linked with the domination of oppositions by loops and bends" (2000: 154). In this conception, fgures aren't just representations or depictions, or a "subsidiary 'tool'" of mathematical reasoning. Tey have what Châtelet describes as an "ontological dignity" which makes certain kinds of mathematical operations possible before the theory behind them is fully understood (2000: 11). For Châtelet, it is not only that fgures like diagrams operate or that they're one of thought's enabling "cultural techniques" (Krämer 2010: 2), but that, in fgur*ing*, they make it possible to apprehend the production of knowledge. In this conception, fguring precedes and succeeds distinctions, ordering—after Rubin, we might say *contouring*—relations between fgures and grounds.

It is by fguring air, for example, that what might otherwise be taken for granted can be acknowledged, allowing us to appreciate its place at "the foreground of our perception as both object and condition of perception" (Horn 2018: 23). Te installation *Yellow Dust* instantiates this process and demonstrates how it works. By translating data about air quality into a mist that could be seen, felt, and stepped into and out of by participants, as Nerea Calvillo and Emma Garnett suggest, interventions like these allow those who engaged with them to "[a]ttend[] to corporeal processes of practising air" (2019: 344). As Châtelet might put it, this fgure traces comprehension: fguring air fgures air *and* how air can be thought.

#### **Part 3: Confguration**

Te word "fgure" can be conjugated with a variety of prefxes and suffxes: prefgure, confgure, disfgure, the fgural, the fgurative and fguration. But these "fxes" do not secure the object or entity in place. In Claudia Castañeda's terms, fguration incorporates "a double force: constitutive efect and generative circulation" (2002: 3). And this double force is why fgure and its "fxes" have acquired a special value in helping us to understand our contemporary situation. By speaking to the relations in and by which fgures fgure, fgure and its "fxes" provide us with a means of understanding and analysing problems that emerge in and through complex relations: of confguring.

Confguring refers to a joining of diverse elements that is never fnal or closed, even as it is stabilised. In practices of system design, engineering and information systems processing, confguration is not the fnal arrangement of hardware and software components, but refers instead to a provisional implementation of organisational infrastructures across myriad and often incommensurate practices. Because of its emphasis on the activity or work of relating the elements of a fgure in movement, confguration has found particular application in science and technology studies, which has developed it to encompass the refexive delineation of the bounds and composition of an object of analysis. As Lucy Suchman says, confguration is "[a]t once action and efect" (2012: 49): it both holds things together and enables potential transformation. Confguration comprises a method through which things are made *and* a resource for their analysis and/or un/remaking, both "a mode of ordering things in relation to one another" and "the arrangement of elements in a particular combination that results" (Suchman 2012: 49). An arrangement may in turn—become a mode of ordering. Tis "double force" is why confguration is particularly useful for analysing novel kinds of ordering associated with the rise of digital technologies, the more-than-human dynamics of ecological crisis and emergent socio-political formations.

D. N. Rodowick (2001) describes new media as technologies of the fgural by drawing on Michel Foucault's notion of similitude: whereas "resemblance presupposes a primary reference that prescribes and classes … the similar is unleashed in a temporal continuum without origin or fnality … governed only by seriality, the similar multiplies vectors … that can be followed as easily in one direction as another, that obey no hierarchy, but propagate themselves from small diferences among small diferences" (Foucault 1983: 44). In enabling unprecedented control over strategies of ordering in time and space, he suggests, contemporary media expand the possibilities of fguration as similitude. Frederik Tygstrup describes the set of new objects-made-out-ofinformation as having a fgural force:

Intuitively, we would probably say that information is something predicated of discernible objects in the world. In the information society, however, the hierarchical relation between objects and information tends to get reversed. On the one hand, what seems to be information about an individual object increasingly stands out as the construction of a new, dividual object. And on the other, the aggregation of information about decoded, endlessly divided objects allows the recoding of completely new, transversal objects. (Tygstrup 2021: 237)

Tese objects have a "two-pronged expressive capacity, sometimes referring back to something existing and sometimes instantiating an image of something new" (Tygstrup 2021: 238; see also Cellard, 2022). Underpinning these transformations are technical platforms that, as Adrian Mackenzie (2018) argues, have an essentially confgurative modality characterised by "confgurative dynamism," "confgurative diferentiation" and "confgurative growth." Ordered by the platform, digital media confgure people and things in constantly varying and experimentally modulated relations.

Feminist science studies in general, and the work of Donna Haraway in particular, have engaged these possibilities of instantiating "something new" beyond (and before) the digital. *A Cyborg Manifesto* (1985) mobilises a politics of the fgure that "rests on the construction of the consciousness, the imaginative apprehension, of oppression, and so of possibility" (1991: 149). Troughout her career, Haraway has presented various "material-semiotic nodes or knots in which diverse bodies and meanings co-shape one another" (2008: 4). Tese she calls "fgures" (sometimes "string fgures"), but named diferently according to the node or knot, the (con)fguration of diferent diverse bodies and meanings. Best known are the cyborg, oncomouse, Terrapolis, chthulu, which are "performative images that can be inhabited" (1997: 11; see also 2016). Teir collective work serves to divert political energy from traditional fgures of sociological, political and psychoanalytic thought—the mother, child, terrorist, immigrant, schizoid or hysteric—towards feminist fgures that, for Rosi Braidotti, "materially embody stages of metamorphosis of a subject position towards all that the phallogocentric system does not want it to become" (Braidotti 2002: 13). Braidotti, Haraway and other feminist makers of fgures and practitioners of fguration do not stand outside the world they describe. Figures are to be inhabited; they are historical entanglements to be felt, reckoned with, struggled over and occupied (see Braidotti 2006: 170; Bastian 2006: 1038).

Tis tradition of making creative, concrete, multiple and playful fgures has inspired scholars that seek alternative ways of confronting complex relational confgurations and, perhaps, imagining them otherwise. In their quest to "denaturalise humanist conceit" (Giraud et al. 2018: 64), for example, scholars in environmental humanities have taken up the challenge fgures pose to normative separations between animate and inanimate, nature and culture, animal and human. Te work of these scholars acknowledges that there are "dangers associated with particular fgurations" (2018: 74). Yet it also fnds a critical, even hopeful potential for alternate settlements between peoples and planet: if indeed "[w]e *are* certainly quite a crowd," then "the ways in which we meet as particular species, and how these entanglements mesh with non-anthropocentric thought, deserve still further fguration" (ibid.). In this work, fgures become critical diagnostic as well as prognostic tools of speculation images or personas that can be used to understand and contest the social, political and conceptual confgurations that we have inherited.

Similarly, work by Elizabeth Povinelli (2016) and Michelle Murphy (2017) is explicitly driven by the need to compose fgures equal to contemporary political confgurations. Each draws on Foucault's four fgures of biopower—the hysterical woman, the Malthusian couple, the perverse adult and the masturbating child. Povinelli uses fgures to identify what she calls the "governing ghosts" of late liberalism: the Desert, the Animist and the Virus (2016: 15). Murphy characterises emergent fgures as the "*phantasmagrams* of economic life" whose spectres of non-life, haunting the social reproductive consequences of the calculations of Gross Domestic Product, converge in the fgure of Te Girl: "the felt and astral consequences of social science quantitative practices, such as algorithms, equations, measure, forecast, models, simulations, and cascading correlations" (2017: 24).

Tis work invites us to refect on how we use fgures in the humanities and social sciences—and how we might make them knowingly and with responsibility. As Murphy writes, the girl is a "generic fgure," but she is assembled from a broad range of practices, including "quantifcation, speculation, and afect … 'fgured out' from a variegated patchwork of social science correlation and wishful speculation, of linked probabilities painted pink with tropes of agency imported from liberal feminism for a North American audience" (2017: 120). Figures, fguring and confguration, as Murphy reminds us, are historical accretions that now no longer rely only on the philological movement from word to historical situation but upon varieties of method and media, prefguring, confguring and disfguring.

It is this variety of method and media that this collection at once comments on and participates in. Our contributors identify fgures, fguring and confguration as a means to query positions, political commitments and know-how. Te collection experiments in alternate ways of knowing and living, fnding and wrestling with fgures that are both symptomatic of and can be used in the diagnosis of the relations that constitute the contemporary situation. Te fgure's confgurative double force is what makes it something that can be engaged with *and* used as concept, methodological prompt and heuristic point of departure for creative and analytical engagements with thorny problems and tangled relations. It is the (im)*mediacy* of the fgure that we aim to capture in this volume.

#### **Part 4: Go Figure!**

Te chapters in this volume are collected here to entice others to "go fgure!"—to show something of what the fgure and fguring can do. Each chapter considers fgures in specifc contexts and traces the efects of fgures according to diferent critical perspectives and standpoints: no single way of fguring is advanced here. Instead, we draw attention to the many ways in which fgures work, with the hope that you will be encouraged to "go fgure" for yourself.

Te next chapter is by Leila Dawney, who suggests that fgures can play a central role in cultural politics—that is, in contestations over power, values and worth that play out in and through the production of culture. Recognising that fgures have sometimes occupied a marginal role in the disciplines concerned with studying these processes—like sociology, political science and cultural studies—her chapter draws on the work of Michel Foucault, Erich Auerbach and Donna Haraway to propose a synthetic concept of fgures equal to their cultural-political signifcance. With Foucault, her fgures are "technologies of power" that order politics and society. With Haraway and Rosi Braidotti, she argues that insofar as fgures are "performative images that can be inhabited," they are necessarily unstable, "labile" and in need of "care." Because fgures are never fxed, it's incumbent on us not only to study them but also to remain attentive to their political force.

Scott Wark engages in a dialogue with Haraway's understanding of fgures as performative images to be inhabited to argue that fgures such as "the cloud," "platforms" or "the stack" allow for the apprehension of what is incommensurable in contemporary media: the speed, the complexity and heterogeneity of scales—both large and small. Rather than as images, however, he suggests that they do so by engaging the potential for refexivity within media, understood as both instruments that mediate perception and cognition and *milieu*—literally, middle places—or *environments*. In this capacity, he suggests, fgures have a unique capacity to help us understand how we live with, through, and in media-technical systems.

Te chapter that follows is an interview between the artist Felicity Allen and Celia Lury. Tey discuss Allen's practice of Dialogic Portraits and the flm *Figure to Ground—a Site Losing its System* (2021), which was produced as part of Allen's residency for a research project (https://peoplelikeyou.ac.uk/). Her dialogic practice allows for an exploration of both relations between the painter and her subjects, and the relations between fgure and ground. It creates a double or multiple space in which the fgure of a person emerges. As Allen says, "In painting one might make a representation which has a background and speak of fgure to ground, but 'ground' is also the sizing treatment and base colour on the canvas, for the picture itself. In this sense the picture itself is the fgure" (insert pg ref this volume).

Te issue of how to understand media by re-purposing media is confronted by Liliana Bounegru, Melody Devries and Esther Weltevrede. Identifying the difculties of studying the lived experience of participating in information fows, the authors propose the novel method of the research persona. Rather like an avatar, the research persona is designed to fgure out how users experience personalised information fows, but it does so by enabling researchers to inhabit the position of fctional users on social media platforms such as Facebook. Te authors show how the fgure of a persona can be used to make visible how platforms confgure Internet users through the use of digital, ethnographic and speculative methods.

For Matt Spencer, the feld of "confguration management," or the task of confguring large and heterogeneous computational systems, provides a rich site for refecting on the role that fgures play in ordering our relations to technical systems. Spencer's chapter focuses on the emergence of "promise theory," a little-studied area of systems management that formalises the "intent" embedded within technical systems by their designers in the form of—tacit or explicit—"promises" that a system will act in a particular way. Spencer suggests that the development of this pragmatic technique for "confguring" systems is of signifcance for the study of computation, in particular, and for social scientists, in general, because it marks a moment in which our relationship to complex technical systems shifts. In it, he suggests, we fnd the emergence of a diferent kind of relational fgure of technical systems: one that recognises that to recognise the "intent" of such systems is to realise that using them entails a form of cooperation rather than mastery. In the fgure of "confguration management," then, we fnd not only an under-appreciated moment in systems management's recent history, but also an example of a pragmatic shift in how technical systems are fgured that, perhaps, betokens a more realistic, open and cooperative means of conceiving how we live in and with technology.

Promise is also central to William Viney and Sophie Day's discussion of personalised medicine. In their chapter, they consider how a research study fgured relapse of disease for patients treated for cancer and classifed as being of high risk for metastatic recurrence. Focusing on the promise of personalised medicine through a multi-perspectival account of this study, they suggest that fgures are used as an empirical proof in the research study they followed, while also forming promises in ways that are at once confrmatory and confounding. Drawing on the work of Auerbach, Viney and Day highlight the temporal dimension of the promise to show the ways in which the fguring of the disease in the research study encompasses multiple temporalities. Personalised blood monitoring of circulating tumour DNA uses novel genomic sequencing technologies but also follows an archaic analytic structure, insofar as it relies on serial fgurations of something unresolved: a (yet to be defned) disease-in-progress.

Sophie Day, Jayne Smith and Helen Ward use a method of fguration to identify how diferent data and samples have been grown, cultivated, studied and propagated in a research hospital. Following Smith, a patient at the hospital and the titular "gardener" of their chapter, and "Grumpa," her tumour, enables the authors to track health data as it moves between the technical environments in which samples and data have been used. Teir collaborative and investigatory work into how samples and data have been fgured helps to identify the cross-cutting relations between care and research that are enabled—or disabled—by data's movement.

Jane Elliott also addresses how the fguring of time is integral to the realisation of the methodological potential of data. Discussing both selftracking or personal informatics and methods of longitudinal research in social science in terms of how they fgure the individual, she identifes the benefts of conceiving fgure and ground in temporal terms, noting that while self-tracking practices rely on a cyclical and repetitive conception of time in order to observe, record and modify behaviour on a daily basis, longitudinal studies rely on a more linear conception of time. In her analyses of cohort studies, it is life events and key transitions that fgure the individual against a taken-for-granted ground of everyday experience, whereas in self-tracking there is the potential for the individual to refexively engage with their everyday lives. Elliott concludes with the suggestion that "we therefore need to attend to more than the contrast between (or mutual constitution of) fgure and ground, but their mutual constitution in cyclical and linear time" (191).

In their chapter about tracking and modelling air pollution, Emma Garnett and Srishti Bhatnagar also highlight the ways in which identifying a fgure enables the relations between the objects and subjects of research to be problematised. Drawing on their ethnographic feldwork as researchers in an interdisciplinary research project conducted in Delhi, the authors explore two occasions when the method of "person-centred environments" was troubled, revealing some of the underlying assumptions of disciplinary methodological and epistemological practices.

Celia Lury's chapter identifes and unpicks three fgures of speech associated with contemporary political campaigns. Tese fgures of speech, "Not in Our Name," #MeToo and #JeSuisCharlie, have been used by people to identify and associate with each other, but the fgures themselves contain personal pronouns that are crucial components to how identifcation and association are achieved. Te focus of Lury's concern is a personal pronoun—"our," "je," "me" and "you"—and the analysis centres on the shifting distribution of the collectives the pronouns call into existence. Lury suggests that the disjuncture between "participating in/ being part of" produced by media-specifc uses of pronouns raises issues of social and political inclusion and exclusion, as well as challenging ideas of truth and individual identity. Accordingly, the chapter indicates how the multiply mediated, pronominal iteration of fgures of speech expose both the limits and the possibilities of a non-representational politics.

John Frow is specifcally concerned with the pronoun "you" that is characteristic of the personalising address of the Internet; as he says, "the pronoun 'you' is silently embedded in an imperative that works ambiguously as both an order and an invitation" (252). He argues that uncertainty of deictic reference is at the heart of the interpellation efect, captured in Althusser's discussion of a policeman calling out "Hey, you there" (Althusser 2001). Frow describes the signifcance of the ways in which while digital or algorithmic personalisation generate a "you" that is not based on fxed markers of identity, these imaginary fgures are constantly being "contextually specifed through acts of rigid designation that seek to tie them to a name and a legally established identity" (255). Tese *points de capiton* pin the digital self to administrative and legal documents that comprise an individual's ofcial identity. But rather than seeing the relation between real and algorithmic personhood as dichotomous, Frow supports instead the idea that there has been a fundamental change at the level of ontology, since "interaction with data, whether voluntary or involuntary, witting or unwitting, is integral to the actuality of our selfhood" (257-258). Proposing that fguring means both calculating and performing the form of the person, he concludes that no single form provides a ground.

Rather than the "you" of personalisation, AbdouMaliq Simone asks us to consider as fgures those who are "something else besides," or rather, to think of fguring as "involving accompaniment or as always also accompaniment: something that does not discernibly alter the visual and sensual dimensions of an event or entity, that remains apparently aloof from its confguration, but which nevertheless prompts a reorientation of view and engagement; which at least raises a degree of uncertainty about what it is we are confronting in an appearance that otherwise has all the hallmarks of an integrity and coherence" (265). Figuring as accompaniment does not create an obligation or a debt; it does not even require mutual recognition or desire. Instead, Simone suggests, it is an enactment of agency not bifurcated by self and other, human and nonhuman, but an intersecting of multiple operations. It is "the restitution of spaciousness" (p. 282, this volume).

### **Coda**

Te contributions we have just described draw on a variety of approaches to the concept of the fgure, extending beyond those we outline in the frst half of this chapter. Many deploy the concept of the fgure to consider contemporary forms of the person and relations of personhood. In these contributions, a person is sometimes distinguished from the individual: as the fgure of a child with asthma (Bhatnagar and Garnett), as a singular and plural fgure of speech (Lury), as data extracted from a self that moves between walled gardens (Day, Smith and Ward) or as data that accompanies or is integrated in a self (Frow). Other contributions (Bounegru, Devries and Weltevrede; Viney and Day; Allen and Lury) share an interest in fguring as a research or artistic method, working across disciplines with numbers, narrative, diagrams and images, highlighting recursion, dialogue and the putting into time of fgures of thought. In some chapters, the individual is recognised to be constituted as a specifc kind of person, distinguished as such in time in relation to a ground (Elliott). Others still (Wark, Spencer, Simone) address issues of fgure and ground, of fguration and confguration, of what it means to inhabit a milieu, a surround or surroundings. In doing so, they enable the worlds built into fgures such as "the cloud" to be acknowledged; they ofer the promise of a restitution of time and space.

What all the contributions share is recognition of a doubling that is intrinsic to fgure. Both noun and verb, a fgure is always fguring, sometimes as part of a confguration. So a fgure may indeed be a number, a character in a text, a representation of a person or another entity, as well as a knot and a node, a turn of phrase, a movement, a diagram or a sequence of notes. But to describe each of these things as a fgure is to indicate that it is both the end-point and the beginning of a fgur*ing*, an activation of the multiple temporalities of the (historical and futureoriented) present tense (Lury 2019). In the relations between subjects (who or what is doing the fguring) and objects (who or what is being fgured), that is, in the (im)mediacy of the relations between doing and being, are the cultural, political and methodological possibilities of fguring: a fgure and its confgurations.

#### **References**


**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/ licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **2**

# **The Work That Figures Do**

**Leila Dawney**

## **Introduction**

In Margaret Atwood's near-future science fction novel *Te Testaments*, the sequel to *Te Handmaid's Tale*, the fgure of Baby Nicole props up two opposing political regimes. "'So useful, Baby Nicole,' Aunt Lydia observes. 'She whips up the faithful, she inspires hatred against our enemies, she bears witness to the possibility of betrayal within Gilead and to the deviousness and cunning of the Handmaids, who can never be trusted'" (Atwood 2019: 33). Smuggled across the border from the misogynist theocracy of Gilead to Canada by her Handmaid mother ffteen years ago, she stands there for the success of the refugee programme from Gilead and the liberation it ofers Gilead's women. In Gilead, on the other hand, she signifes the evils of Canada and the potential enemy within of traitorous Handmaids. Her image, still a baby ffteen years

L. Dawney (\*)

Geography, College of Life and Environmental Sciences, University of Exeter, Exeter, UK

e-mail: l.a.dawney@exeter.ac.uk

later, adorns the walls of the training centre for novice Aunts—the women who uphold Gilead's regime—where she is prayed for daily. In Canada, her face is held aloft on the placards held by refugees from Gilead and their supporters.

Figures, as "performative images that can be inhabited" (Haraway and Randolf 1997: 11), are central to the workings of cultural politics. Tey are technologies of power that work through the afects: ofering spaces for inhabitation and emerging subject positions, stirring up intensities, and fostering attachments and hostilities. In a context where afect and emotion are powerful agents, fgures give imaginative and material form to structures of feeling and political formations. Figures mediate power and authority and personify mythologies, generating cultural and political forms of life. Te power of the fgure of Baby Nicole to harness afects comes from her ability to refer outside of herself: to signify life, innocence, futurity, vulnerability and disloyalty, paradoxically in support of two opposing regimes. Whether individualised, like Baby Nicole, or "types," like the expert, the whistleblower and the migrant, they are at once social imaginary, media image, archetypal form and locus for public feelings.

Where institutional modes of authority such as the church, state and academy are increasingly questioned and public trust in these institutions declines, fgures that challenge these traditional institutions can become new loci for authority around which public afects circulate. In the public sphere, fgures such as Jordan Peterson and Greta Tunberg gain authority through their personifcation of inchoate political and collective feelings, and abject fgures like the terrorist, or the benefts cheat, participate in regimes of control, uncertainty and paranoia. Tey gain traction through reference to master narratives and cultural myths, generating new mythologies in the process. Similarly, fgures that were historically a focal point for collective politics, such as the heroic male worker, are losing their grip in an increasingly precarious and fragmented economy.

While the afective power of public fgures is increasingly recognised in critical scholarship, little attention has been paid to their theorisation. Tere is a need for conceptual attention and precision to understand, frstly, how fgures operate as afective technologies of power by tapping into public feelings, and secondly, their potential to organise alternative forms of life. Tis chapter sets a conceptual agenda for exploring the political work that fgures do in contemporary cultural politics. Understanding this work is, I argue, a vital aspect of the study of political life: the power of fgures to shape public moods demands their urgent theorisation and critical attention, while attending to this power highlights their potential to guide other subjectivities and allegiances. Tis agenda is based on three initial premises. Firstly, fgures are materialsemiotic signposts towards ways of knowing, understanding and inhabiting the world. Tis means that they carry with them, and point to, sets of ideas, feelings and positions: they shortcut to what Raymond Williams calls "structures of feeling" (Williams 1977). Tey structure the world by giving substance to cultural ways of being in the world. Secondly, they are technologies of power that work through afective capacities of specifcally historied bodies. In other words, they act on our bodies, generating emotional and afective responses and feelings and as such can be mobilised for particular political ends. Te bodies that they act upon are already entrained to respond in particular ways: they are gendered, classed, racialised, and embedded in histories and cultures. Tey buy into certain life narratives, hopes and dreams, and it is in fgures' interaction with already-entrained bodies that their afective work is undertaken. Finally, as a critical practice, fguration involves both invoking and thinking with fgures. If we accept the frst and second premises, and acknowledge that fgures do indeed wield cultural power and that this power works on our afects, sensibilities and emotions, then, as critical scholars, we too can work with fgures to bring about change, to question and to amplify other ways of being and living.

Tis chapter brings three writers whose work engages with the politics of fgures and fguration into dialogue to provide a conceptual outline for thinking about the work that fgures do. In diferent but overlapping ways, these three writers exemplify and develop understandings of fgures that acknowledge and attempt to explain their cultural power. Tey demonstrate the importance of meaning making, storytelling and fguration in the shaping of social life; moreover, they begin the work of showing both how this takes place and its implications for understanding the relationship between power, bodies and our imaginary worlds.

Te chapter begins with an outline of the concept of fguration in the German literary critic Eric Auerbach's *Mimesis: Te Representation of Reality in Western Literature*, focusing on the power of fgures to make stories and narratives "make sense" through their teleological and portentous capacities. It then discusses how fgures are understood in Michel Foucault's four fgures of biopolitics in the *History of Sexuality 1.* In this volume, Foucault demonstrates the part fgures play in securing regimes of governance; how they illustrate and exemplify what is counted as "normal" in a particular historical condition. Finally, this chapter discusses how the work of both of these thinkers plays out and emerges as critical practice through the philosopher of science Donna Haraway's "menagerie" of fgurations. By reading the work of these three thinkers difractively—against and through each other—and witnessing what arises from their interaction, I highlight how each develops conceptual tools for thinking about the relationship between fgures and the political making of the world. In Auerbach, we see how fgures become attached to myths of redemption, which confrm and reify cultural narratives. In Foucault, we see the emergence of fgures as object-targets of biopower, demonstrating how they act as technologies (techniques) that work in the service of particular formations of power. Haraway uses both of these aspects of fgural critique to intervene in the world, drawing a series of alternative, feminist fgurations that challenge dominant narratives that defne the human. Reading these texts together builds an agenda for studying how fgures work as technologies of afective power and explores the potential for fgures to destabilise normative ideas. Tis chapter proceeds by looking at how fgures have been put to work to dismantle the very symbolic and structural orders that they suture, and it explores the relationship between fguration and afect. Finally, it discusses the cultural and temporal specifcity of fgurations, pointing to their need to resonate with and refect existing cultural forms and material practices, and their vulnerability to being seized, or appropriated, like Baby Nicole, for other agendas.

# **Auerbach: Figuration as Tropic Device**

Te early twentieth-century literary historian and critic Eric Auerbach provides an important and seminal resource for thinking about the work that fgures do. In his magnum opus, *Mimesis: Te Representation of Reality in Western Literature*, Auerbach ties the practice of fguration to the history of Western literature and representation. His work falls within a German humanistic tradition of literary criticism that understands literature as an expression of lived experience, and as such positions criticism as a dialogue between author and critic: an attempt to live the author's reality and experience the *geist* of the work. In *Mimesis*, Auerbach discusses how the term *fgura* was taken up by the early Christians, notably Augustus, to describe the practice of reading the Old Testament in relation to the New. Unlike allegory, which points outside of itself to an abstract form, *fgura* remains within the realist tradition, referring instead to other forms, or echoes, of itself. In the hands of the early Christian Church, and in Medieval Christianity, the New Testament becomes a "fgural, and he adds, carnal (hence incarnate, real, worldly) realisation or interpretation of the Old Testament," enabling the Old Testament to be read as a precursor of what is to come and tying both together in an overarching salvation narrative of Fall, Sacrifce and Last Judgement (Auerbach and Said 2013: xxi). In late antiquity, the Old Testament's role became fgurative: as a prophetic announcement or anticipation of the coming of Jesus. Auerbach illustrates this through Augustine, who maintains that the sacrifce of Isaac prefgures the sacrifce of Christ. Auerbach thus highlights the "vertical connection" of disparate elements of the Old Testament, enabling its reading as though "God chose and formed these men to the end of embodying his essence and will."1 Here, the fgure is central in constructing both ideas of time and historicity through its repetition and refraction across timescales, making connections and links between epochs and tying them together in overarching narratives of redemption and fulflment. Tragedies were thus seen as trifes in this great scheme: "however serious the events of earthly existence might be, high above them stood the towering and all-embracing dignity of a single event, the appearance of Christ, and everything tragic was but fgure or refection of a single complex of events into which it necessarily fowed, at last: the complex of the Fall, of Christ's birth and passion, and the Last Judgment" (Auerbach and Said 2013: 317).

<sup>1</sup> It is worth noting here that *Mimesis* was published in the 1930s and constituted in part a project designed to revalue Old Testament scripture and as a means of refuting Aryan philology.

By reading one text through another, Auerbach's resurrection of fgural realism insists on the referential and relational substance of fgures. Figural representation "establishes a connection between two events or persons in such a way that the frst signifes not only itself, but also the second, while the second involves or fulfls the frst. Te two poles of a fgure are separate in time, but both, being real events or persons, are in temporality. Tey are both contained in the fowing stream which is historical life" (Auerbach 1984: 53). Figuration connects events, characters and objects to broader narratives. It incorporates them into cultural myths that self-reinforce through their own fguration. In other words, Auerbach demands we take seriously the referential work of fgures and how they participate in the production of myths and master narratives of salvation and redemption. It demands that we read texts not in isolation, but in relation to broader patterns, discursive regimes and master narratives. When we do so, we bring to the table those related stories and master narratives and do the work of joining the dots that allows them to fulfl one another, in turn augmenting their cultural power. Auerbach's fgures make sense in their potentiation for future reading and fulflment. His fgures are tropic: they turn towards other fgures, objects and narratives. Tey gesture towards something greater than themselves, acting as both signs and referents in a perpetual play of associations. Unlike metaphor or allegory, they do not represent or stand in for ideas on their own. Rather, they gather the stories to which they refer into a coherent narrative of fulflment, which is gestured at rather than made explicit, requiring the reader to do the work of tying them together.

# **Foucault's Figures as Objects and Targets of Power**

Where Auerbach helps us to understand how, in Western literature, fgures are caught up in networks of potentiation and referral that produce narratives of time, history and redemption, the work of Michel Foucault demonstrates how fgures have also been mobilised in civic institutions and medical discourse as a means of bolstering biopolitical control. In the frst volume of *Te History of Sexuality, Te Will to Knowledge*, Foucault delineates his concept of biopower, drawing attention to the relations, modes of organisation, technologies and practices that produce biopolitical subjectivities and forms of life. *Te Will to Knowledge* addresses biopower as a mode of governing—a political rationality—that focuses on populations as a whole and the managing of life itself, rather than the behaviour of individuals. Biopower involves the management of sex and reproduction, mortality, health and illness. Yet this management is indirect: it does not come from the diktat of a sovereign, but rather is distributed across institutions such as psychiatry, the family, education, and welfare provision. It produces knowledges and practices (technologies) which, among other functions, dictate what is normal and what is deviant. In *Te Will to Knowledge*, he identifes four "strategic unities," or trends in governance, that emerged during the eighteenth and nineteenth centuries and that produced and organised bodies and sex in the service of biopower (Foucault 1978). Tese strategic unities were the hystericisation of women, the pedagogisation of child sexuality, the socialisation of procreative behaviour and the psychiatrisation of perverse pleasure. Tey found their objects, and their targets, in four fgures around which they coalesce: the hysterical woman, the masturbating child, the Malthusian couple and the sexual deviant. Emerging from sets of discourses around sexuality, procreation and social reproduction in institutions such as the psychiatric clinic, these four fgures appear as personifcations of what must be regulated and controlled (sexuality in particular) and in doing so police the borders of what is considered normal and healthy. As such, the four fgures occupy spaces at both the centre and the limits of power: they are central to how the biopolitical ordering of sex is organised, yet they also occupy limit conditions. Tey are troubling to the forces that produce them and are subject to techniques and technologies that create them as fgures, that target them as objects, that produce them as subjects and that draw on them to augment the forms of knowledge/power they reproduce.

Although the idea of the fgure as a technology of power was not elaborated at length in Foucault's work, his four fgures of biopolitics nevertheless exemplify the work that fgures do in relation to power. Foucault's fgures are, on the one hand, the objects and targets, as individuals, of disciplinary modes of biopower and, on the other hand, vehicles for the generation of collective afects that shape bodies, desires and sensations and order sexual and reproductive life. Individual bodies, biopolitically fgured, can become object-targets, yet these same fgures are also the technologies through which the afective capacities of populations themselves are produced as object-targets (Anderson 2011, Dawney 2018). Tey emerge at both the centre and the limit of a historically specifc set of regimes that produce bodies and desires according to a particular normative order: as an object-target, particular fgures can embody abjects, deviants and villains, or heroes, aspirational fgures and leaders. Tey are propped up by the proliferating institutions and practices that modulate pleasures, afects and spatial confgurations. As fragile subject positions that both threaten and suture the present, Foucault's fgures not only personalise and give substance to these regimes but also provide ideal types of bodies that become the target of such regimes. On the surface, these fgures seem to operate in the service of dominant modes of *power over (potestas)*. Yet by virtue of their limit status, they also expose the contingency of such modes of organisation. Tis, of course, is Foucault's goal: it supports his lifelong project to undo—and to reveal as contingent—that which we universalise, naturalise and dehistoricise and, in doing so, to point to the possibility of an "otherwise." While Foucault, the critic and historian of the present, does not position as his task to think what such an otherwise might look like, he does end this frst volume with a hint at the possibility of a "diferent economy of bodies and pleasures" (Foucault 1978: 159), opening up a space to consider the transformative potential of fgures.

In summary, then, we can situate Foucault's contribution to thinking with fgures in terms of his acknowledgement of both how regimes of power can produce and personify fgures and, in turn, how they then can prop up these regimes by inhabiting a space at the limits of the normative. As part of his broader project to interrogate the workings of power relations at particular historical junctures, he demonstrates the way in which regimes of governance—such as criminal justice, welfare policy or healthcare—generate fgures as targets and objects and, in turn, how these give substance to ideas and values. Above all, Foucault brings the fgure to the social sciences, demonstrating its centrality to the analysis of power in modernity.

#### **Haraway's Figurations as Spaces to Inhabit**

Our fnal companion on this journey through conceptual accounts of fguration is Donna Haraway, who picks up and runs with Auerbach's insistence on the referentiality of the fgure, yet also draws heavily on Foucault's concern for the ways in which fgures are incorporated into regimes of power and play a part in the production of subjectivities. Like the previous two thinkers, Haraway's fgures are more-than-textual: they are material-semiotic, and as such they enact worlds through their material confgurations. While all three thinkers acknowledge the power of fgures to shape worlds, it is Haraway who explicitly adopts the fgure as counter-technology or critical device: she creates fgures that trouble binaries, draw on and play with master narratives, and ofer alternative stories. Within Haraway's fgurations, we see both echoes of Foucault's object-targets in terms of the binary fgurations that she deconstructs and also of the attention to their tropic capacities and role in storying the present that she takes from Auerbach.

Despite Haraway's mobilisation of fgures throughout her work and, indeed, her assertion that "I feel like I live with a menagerie of fgurations" (Haraway and Goodeve 2000: 135), her oeuvre contains very little direct discussion of fguration. Her 1997 book *Modest\_witness@second\_ millennium:femaleman\_meets\_oncomouse: feminism and technoscience* most explicitly lays out her understanding of fguration, and its emergence as a mode of critique and analysis can be seen in much "new materialist" cultural studies. Haraway's fgures are closely tied to relations of power, working to refect, difract and enact them otherwise (Haraway and Randolf 1997, Haraway and Goodeve 2000). Haraway's fgures have two main features. Firstly, as in Auerbach, they are tropic, referring outside of themselves in a way that troubles certainties and established binaries: "fgures do not have to be representational and mimetic, but they do have to be tropic; that is, they cannot be literal and self-identical. Figures must involve at least some kind of displacement that can trouble identifcations and certainties" (Haraway and Randolf 1997: 10). Secondly, fguration is understood as a mimetic practice that maps our world. It produces stories to which subjects can attach themselves or can gain purchase on life. Her mode of critique is to make a diference in these material-semiotic apparatuses, to unravel their telling and tell other stories in the process. For example, her fgure of the cyborg ofers a feminist vision of science and technology that operates against both masculinist appropriation of technology and forms of feminist subjectivity that revert to the natural (Haraway 1991).

For Haraway, drawing on Auerbach, fgures are "potent, embodied incarnated, if you will—fctions that collect up the people in a story that tends to fulflment, to an ending that redeems and restores meaning in a salvation history" (Haraway and Randolf 1997: 44). Her recognition of the vertical referential power of fgures in producing narratives of redemption emerges directly from Auerbach's work and echoes his teleological bent. Yet her own fgures are playful and subversive. She ofers them to the reader as sites to grasp onto the relations of gender, power and knowledge that produce them, but also as positions from which to enact diferent subjectivities: the cyborg, the oncomouse, the companion species and string fgures are all "performative images that can be inhabited" (Haraway and Randolf 1997: 11). Tey are images that do something, that actively participate in the making of worlds. Contemporary forms and logics of life are understood as an "implosion of bodies, texts and property" (Haraway and Randolf 1997: 7)—a menagerie where the literal and the fgurative, the factual and the narrative, the scientifc and the religious, and the literary are always drawn together (Haraway and Goodeve 2000: 141). Haraway ofers us a critical practice of fguration that involves paying attention to the production, appearance and work of fgures and fnding ways of detaching them from salvation narratives. Tese fgurations recognise the forms of domination they emerge from and the boundaries that they shore up. In exposing and disrupting these boundaries, such as the production of the unity of the self that relies on women's homogenisation and exclusion, her fgures provide the conditions of possibility for her ethics of coalition. In response, her fgurations are an experimental, playful and creative means for thinking outside of binaries and developing new forms of embodied subjectivity. In relation to technoscience, for example, she writes, "We inhabit and are inhabited by such fgures that map universes of knowledge, practice and power. To read such maps with mixed and diferential literacies and without the totality, appropriations, apocalyptic disasters, comedic resolutions, and salvation histories of secularised Christian realism is the task of the mutated modest witness" (Haraway and Randolf 1997: 11).

Drawing on Haraway, scholars in feminist science studies, environmental humanities and posthuman thought have developed a distinctly feminist practice of fguration. Like Auerbach's discussion of the Old Testament as stories that precede and give rise to a new world, feminist fgurations operate *pre-*fguratively: they enact possible futures. For example, Rosi Braidotti's fguration of the posthuman is imagined as a "conceptual persona, a navigational tool that helps us illuminate contemporary discursive and material power formations" (Braidotti 2019:22). Te posthuman works by creating minor knowledge systems and spaces of subjectivity and decentralising the fgure of the human. In doing so, fgurations like the posthuman embrace nomadic and fugitive subjectivities and question normative, exclusive and static modes of subjecthood. "Te posthuman as cartographic fguration is a branch of contemporary critical thought that allows us to think of what 'we' are ceasing to be—for instance, the Eurocentric category of universal 'Man'. It also sustains, however, the efort to account for what 'we' are in the process of becoming—the multitude of ways in which the human is currently being recomposed" (Braidotti 2019: 7).

### **An Agenda for the Study of Figuration: Figures and Affect**

Above, I have outlined a short and rather incomplete genealogy of critical work on fgures in order to set an agenda for their study. Tis agenda makes the following claims about the work that fgures do and about the contribution that a fgurative approach ofers to critical theory. Firstly, fgures are powerful world-making technologies: in Auerbach through the ordering and of time through Christian theology; in Foucault through their role in producing regimes of knowledge and biopower; and in Haraway through their potential to disrupt structuring orders. Tis means that we need to take them seriously as objects of analysis. Secondly, as critics we need to consider fgures within a broader architecture of fguration to make sense of the stories they tell about the world. Tirdly, there is political work to be done in amplifying minor fgures or producing and looking after the fgures that we feel can contribute to a better world. Finally, I suggest that analysing the work that fgures do and adopting critical and creative practices in relation to fgures reveals their vitality and afective force in sculpting worlds: how they lure us towards particular political architectures and provide substance for aleatory and minor ways of being and relating. It is for this reason that we need to pay attention to the afective capacities of fgures.

Te "afective turn" in the humanities and social sciences has drawn on a range of theoretical genealogies, including the psychiatric/psychoanalytical work of Tomkins, Freud and Sedgwick, and the Spinozist, materialist lineage of Deleuze and Whitehead. In the latter genealogy, afect refers to the capacity of bodies to be moved and to generate intensities in relation to other bodies, objects and ideas (Clough and Halley 2007). In this context, then, the afective capacity of fgures refers to their ability to generate feelings and embodied responses in those who encounter them: responses that are tied to emotions such as shame, fear, revulsion, love and joy. In mobilising afective responses, fgures draw us in to their narratives, tying us deeper to the stories that they personify. We can see this, for example, in the way that the fgure of the wounded soldier is mobilised in the UK to tether afects to discourses of nationalism and militarism (Dawney 2018), or how the fgure of the bereaved mother bolsters pressure groups like Mothers Against Violence (Dawney 2013). Figures personalise structuring myths and draw on them for their power: in both these cases, they harness afects through making visible corporeal vulnerability and sufering. Troughout the Christian tradition, religious authority becomes tangible and knowable through embodied pain and wounding, most specifcally through the body of Christ and the Virgin Mary. Te Christ and the Pietà, peppered as they are through the history of Western cultural forms, lie behind the contemporary staging and personalisation of sufering ofered through the fgures of the mother and the soldier, and it is through these tropisms that they are recognised, understood and their afective power amplifed.

If fgures operate at the level of afect, then the study of fguration needs to acknowledge the space of encounter: how fgures make experience intelligible to subjects in a particular way and how they form the material relations of bodies, texts, technologies and logics that constitute discursive formations. Figures are involved in afective relations with historied, situated publics. Tey work on our desires and generate intensities, moving bodies by tapping in to cultural myths and undercurrents and acting as loci for afective forces that coalesce around them. Tis "grip," or, as Claudia Castaneda puts it, ability to "captivate" (Castaneda 2002), is central to their power. Attending to the grip reveals their vitality and afective force in sculpting worlds: how they lure us towards particular political architectures. Adopting a critical and creative practice in relation to fgures in turn provides substance for alternative forms of life. It enables us to ask what desires, anxieties, material insecurities or existential fears are triggered through these fgurations, both augmenting what is there already and contributing to the ongoing formation of the social as fgures attract and seduce, or alienate, or horrify.

Lee Edelman's Lacanian polemic on reproductive futurity, *No Future*, positions the fgure of the child as a central organising point around which normative understandings of life and futurity are gathered, positioning the queer as its abject other (Edelman 2004). Trough his invocation and analysis of these fgures, Edelman efectively demonstrates the play of fgurations and the work that fgures do: their mutual structuring of normativity and its outside and the centrality of afect to their workings. Via Tiny Tim, the orphan Annie and Peter Pan, Edelman's fgure of the Child underpins the heteronormative order and requires the reproduction of sameness of identity. If the child stands for life of a particular structural order, the queer haunts its outside: the nonreproductive eroticised narcissism that can only be for death and the nonreproduction of the Same. Te queer can do no more than reject the child and embrace the death drive. As the part with no part, he (for he does seem to be a he) has no place within the reproductive futurity fgured by the child. He must stand with death, to queer the way that life is fgured. Whatever our take on Edelman's argument, what he successfully demonstrates is the power of the fgure to embody normative orders and to suture them through their appeal to already existing afective channels stirred by vulnerable embodiment. Te familial shaping of afective bodies produces subjects with the capacity to be afected by the Child according to the norms of reproductive futurity. We can see this too in Atwood's twisted near-future take on reproductive futurity, where it is Baby Nicole who adopts the childish innocence that stands for the future that must be protected, by both sides of the Canadian border and whose power to generate afective attachments is used for political gain.

For Edelman, the politics of queerness is a politics of refusal: to refuse to take part in the confation of life and familial orderings of reproduction, to refuse to take subject positions that respond to these orders and to refuse the progress-oriented drive for a better world. Te queer, as the constitutive outside of reproductive futurity, is defned by lack and by negativity, yet by actively inhabiting that space, by refusing to adopt normative myths and by permanently inhabiting the space of negativity, the queer remains as a spectre that can expose and dismantle the normative object. Te fgure of the queer in Edelman's work, as in Haraway's menagerie of fgurations, pushes against these orderings yet is constituted through them. Tese fgures work to produce other subject positions; not as an unspecifed otherwise, but to fabricate new architectures of subjectivity that are always a product of those boundaries and binaries that they interrupt. Haraway's fgures are, she claims, a kind of "gift"—she ofers them up as templates for ways of life that rely on and generate diferent narratives. Similarly, Braidotti, in her discussion of the fgure of the posthuman, argues that fgures ofer a "frame for the actualisation of many missing people, whose 'minor' or nomadic knowledge is the breeding ground for possible futures" (Braidotti 2019:23).

# **When Figures No Longer Hold**

Tese alternative fgurations, as with all fgures, are manifestly unstable. Tey rely upon, and prop up, particular structures of feeling and these structures move with the shifting ground of the material. Figures may no longer hold; they may be appropriated or subsumed within normative orderings that reduce their radical potential.

By virtue of their cultural and afective power, fgures can be radical. Yet, if they work through afective encounters with situated bodies, it follows that they can only resonate in the context of specifc historical circumstances. Figures "collect up and refect back the hopes of the people"—they provide a sense of the possibility of salvation or damnation or conclusion (Haraway 2000). Yet sometimes, we fnd that their grip no longer holds or that their potential is diminished through their resonation with other public myths. Tey may not always appeal as they once might: as Michelle Bastian points out, fgurations "need to be attractive, productive, and inviting. Tey need to be inhabitable and to resonate with already existing collective meanings—very difcult criteria to fll" (Bastian 2006: 1030). Similarly, Braidotti highlights the need for fgurations to resonate with contemporary bodies: "All fgurations are localized and hence immanent to specifc conditions; for example, the nomadic subjects, or the cyborg, are no mere metaphors, but material and semiotic signposts for specifc geopolitical and historical locations. As such, they express grounded complex singularities, not universal claims" (Braidotti 2018: 34). Tis is apparent in Bastian's discussion of how Haraway's cyborg fgure was appropriated through science fction, Silicon Valley and cyberpunk discourses, shifting away from its original fguration in the service of a pluralist, transversal feminist subject. As early as 2000, Haraway expressed concern that "cyborgs [can] no longer do the work of a proper herding dog to gather up the threads needed for critical inquiry" (Haraway 2003: 4). As Haraway has made clear, much of this appropriation confuses the fgure with the referent. In describing what she calls the "distressing half-life of the cyborg," she notes how the specifcity of the cyborg fguration has been diluted and dehistoricised and become a "maddening" way of describing any interface between humans and machines (Haraway 2000). Nevertheless, Bastian rescues the cyborg from technofascism, arguing that its undecidability and multiplicity allows for its "lost" aspects, such as those that highlight coalition, to resurface through counterhegemonic articulations, including US third-world feminism. Figures, as pathfnders for other ways of living and being, work diferently in diferent contexts. Tis becomes most clear at the between times—the junctures where historical moments give way to others, and the spaces of lag and emergence that appear at these times (Williams 1977). We might understand the contemporary juncture as such a time—sitting within neoliberal architectures, yet holding on to the certainties of the Fordist welfare state. It is at these times when reliedupon fgures become co-opted or no longer resonate with lived experience. Lauren Berlant names this an "impasse"—a temporal and spatial moment where incumbent fgures and desires no longer work for us, yet remain as impossible and damaging objects that keep us wedded to normative identity positions (Berlant 2011). Te anthropologist Anna Tsing has argued that the fgure of the abstract worker as a hinge for labour politics no longer resonates in the context of supply chain capitalism, where the decline of the white male union has heralded the rise of the entrepreneurial servant/manager as aspirational fgure (Tsing 2009). Te lack of grip that the fgure of labour holds is a challenge for a labour movement whose very being rests on the production of a particular fgure of labour—one that is no longer up to the job. Berlant argues we need new afective infrastructures and, I would add, new fgures (Berlant 2016). Yet the myths and narratives that might be better equipped for dealing with the precarious present perhaps need more work and encouragement. If democratic life is indeed subsumed under the metric of the market, then new fgures are needed to act as guides towards diferent forms of subjectivity and collective life.

# **Conclusion**

Tis chapter has outlined an agenda for thinking with fgures through a number of intersecting claims. Firstly, it has highlighted a series of key texts that inform study of the work that fgures do in cultural politics, demonstrating how they can be read together to both highlight the cultural power of fgures and work with counterhegemonic fgurations. Secondly, it has demonstrated the interrelationship of afect and fguration and the need to pay attention to how fgures lure and harness afects. Finally, it has shown fgures to be labile and vulnerable: their cultural specifcity and inherent instability mean that we cannot ever assume their stasis. In tying these claims together, this chapter is both an invitation to think with fgures and a suggestion as to how this might be done.

As I have argued throughout this chapter, fgures often work as a locus for afects, attachments and public feelings. It is the task of the critic to both identify and recognise how fgures shape political and cultural life and to analyse the mechanisms through which this takes place—how the "grip" of fgures is established, where their appeal lies, what their worldbuilding capacities are and how they act upon the world. Te afective approach to studying fgures outlined here has methodological implications too; bridging the space between body and text in this way is no mean methodological feat, nor is it a simple matter to trace fgures accurately across their many and varied cultural articulations. Nevertheless, the concept of the fgure invites an approach to cultural investigation which refuses to lie entirely within the text or the subject, instead focusing on their mutual composition in relation to wider political structures. A fgural critique thus needs to be experimental: it may attempt to inhabit the impossible space between representation and world, or tell stories, or connect seemingly disparate objects. It may work with others to generate new fgures or make new sense of existing ones. As critics, we must be aware of our role in the process of fguration and sensitive to the politics of our own intervention: Baby Nicole, like Edelman's Child and Foucault's hysterical woman, can be put to work in many diferent ways, and we would do well to heed Haraway's contention that "it matters which fgures fgure fgures" (Haraway, 2016:101). Figurations, like many aspects of disruptive, excessive life, run the risk of capture—of losing their radical potential through their incorporation into more dominant forms of imaginary. An approach that acknowledges their shifting relationships to power is essential. It takes work to hold on to the potential of the fgure and rearticulate its transgressive forms. In producing and articulating feminist fgurations, we need to be vigilant to the extent to which they still hold true and to how they interrupt themselves and each other: we need to take care of our fgures.

#### **References**

Anderson, Ben. 2011. Afect and biopower: towards a politics of life. *Transactions of the Institute of British Geographers* 37: 28–43.

Atwood, Margret. 2019. *Te Testaments*. London: Chatto & Windus.

Auerbach, Erich. 1984. *Scenes from the drama of European literature*, Minneapolis: University of Minnesota Press.


**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/ licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **3**

# **In "The Cloud": Figuring and Inhabiting Media Milieus**

**Scott Wark**

# **Introduction: Media Figure**

Our discussions of digital media are full of fgures. Most of our online interaction takes place on webpages that we visit by entering an address, or platforms that we access through portals. Tese services circulate data on networks and interact with one another using interfaces. Moreover, these services were once hosted on local machines known as servers; now, the data they process resides in a place we call the cloud. Te hierarchical information architectures that underpin these services are known as stacks. Sometimes, we think of the encompassing system of data, devices, interfaces, and services—"the internet," in other words, as it's actually lived—as an ecology or an ecosystem, which is to say, a self-governing, self-organising, holistic, and emergent system that we, in some sense, inhabit. Other times, we just call it life—work or leisure, productivity or entertainment, self-expression or data capture.

S. Wark (\*)

Centre for Interdisciplinary Methodologies, University of Warwick, Coventry, UK e-mail: s.wark@warwick.ac.uk

Tese terms, and others like them, make up the language that we use to denote, describe, and engage with contemporary media. Many of these terms have specifc meanings in technical felds, like computational science and human-computer interaction: interfaces allow users to communicate with soft- or hard-ware (Hookway 2014); platforms are a type of computational system organised around the decentralisation of content production and the recentralisation of data capture (Helmond 2015); cloud computing provides of-site computational resources, software services, or storage to customers on demand, obviating the need to maintain systems and services on their own premises (Hu 2015). But they are also woven through the vernacular language we use to help us make sense of the vast, complex, distributed, and encompassing media systems that underpin life today. Tese systems operate at speeds and extend across scales—both vast and microscopic—that exceed embodied human perceptual and cognitive capacities. Tey are made up of media technologies that are otherwise imperceptible to their users. Tis is where fgures come in to play. Interface, platform, cloud—each is a technical, computational term, but each is also a fgure. In this chapter, I want to formalise the role that these fgures—and others like them—have in our engagements with media. In brief, the proposition I want to develop in this chapter is this: fgures provide a means for making sense of how complex, distributed, and opaque media-technical systems inform, condition, and shape contemporary life.

Tis proposition relies on a somewhat-idiosyncratic understanding of its key terms: fgure, inhabitation, and media. Rather than conceiving of fgures as symbolic phenomena—as representations, metaphors, or fgures of speech—I want to argue that fgures make media inhabitable: that they are the means by which media can be lived with, lived through, and lived in today. Te point of departure for my conception of fgures is Donna J. Haraway's hugely infuential claim that we ought to understand fgures as "performative images that can be inhabited" (1997: 11). For Haraway, this claim sits within a theoretical framework designed to undermine distinctions between the world and our symbolic representations of it: understood as "material-semiotic processes," her fgures become a critical means for both understanding the world as it is, and for actively *making* new worlds that are inhabitable by—and amenable to—a variegated "menagerie" of human and non-human beings (see Giraud et al. 2018). Drawing on Haraway's work, I want to retain the idea that fgures can be "inhabited." But I want to revise this inhabitation's terms. I want to focus on "worlds" constituted—conditioned and shaped—by large-scale media-technical systems.

Haraway's vision of reality is messy, hybrid, and multiple. For her, there is not one world, but many; these worlds are not just populated by *us*—that is, by humans—but by collectives of beings that undermine distinctions between nature and culture, or humanity and its others (Tiele 2021). In invoking the category of "materiality" to ground these multiple messy worlds, Haraway nevertheless tacitly acknowledges that worlds are conditioned by a fundamental "ground": *the* world, our Earth, which is the ultimate source of, and limiting condition on, their potentiality. It follows that as the world changes, so too must our conception of fgures. Drawing on Benjamin H. Bratton's concept of "planetary-scale computation" (2015), I want to argue that ubiquitous media have changed the nature of the "worlds" that fgures can fgure by establishing a new, globalised ordering regime and by providing us with a new technical means to perceive the world itself as a large-scale system. Tese worlds are diferent in kind: we still require fgures to render them inhabitable, but the modes of inhabitation they engender are diferent.

Tis proposition relies on a broader and rather more ecumenical conception of media and technology than typically circulates in the humanities and social sciences. Drawing from recent work in media theory, I want to argue that media's networked distribution, their ubiquity, and their automated capacity to collect and process huge amounts of data mean that, in the aggregate, they also constitute *milieus*. In broad terms, a milieu is an environment, territory, or ecosystem—and, as media theorists like John Durham Peters (2015) and Antonio Somaini (2016) argue, it can also be proftably extended to media. Conceiving of media as milieus provides us with a conceptual means of recognising that, at scale, media constitute places that can be inhabited. Milieu literally means "middle place" (Peters 2016: 47): extending the concept of "milieu" to encompass media provides us with a conceptual language to articulate their capacity to not only organise work or leisure but engender the mediated environments in which contemporary life increasingly takes place. Conversely, though, I also want to emphasise the crucial role that fgures play in rendering such environments inhabitable at all.

After outlining Haraway's concept of fgures, this chapter will illustrate how they make media inhabitable by analysing one of contemporary media's key fgures: "the cloud." For marketers and computer engineers alike, the cloud refers to computational services that are accessed remotely using networked technologies rather than being run by a user, customer, or company on-site. Over the past few decades, though, this term has expanded into something much more encompassing. Depending on one's dispositions, attachments, and responsibilities, life is increasingly lived *in* "the cloud." Haraway's concept helps us to understand what this means. What links how "the cloud" is used now and how it was used by early systems administrators and engineers is its capacity to capture and articulate aspects of computation that are otherwise difcult to represent. Tis point is crucial: digital media are often characterised by their complexity, distribution, and opacity—above all, the imperceptibility of their operations to those who use them. Tough they might shape worlds, their operations are not straightforwardly commensurable with representational epistemologies. Te computational fgure of "the cloud" is therefore the latest in the line of cloud-fgures that mark out the limits of what can be represented (Damisch 2002). Te transition from *a* cloud to "the cloud" is one from a delimited and specifc symbol for a computational network to an articulation of distributed, complex, and encompassing technical condition of contemporary life that might only be accessed intermittently, but which nevertheless shapes what it means to live in the world today—a mediated milieu.

As fgure, "the cloud" transfgures what is heterogeneous, complex, and unrepresentable—media-technical systems that operate at speeds and scales beyond *human* perception (Mackenzie and Munster 2019) into what can be lived with and lived in. Tis, I want to argue, is precisely why we need fgures: to compass the gap between what is in excess of representation yet nevertheless conditions a life lived with, through, and in media.

### **Figure, Inhabitation**

Haraway's concept of fgures emerges from a mode of intellectual enquiry conducted as a *practice*: one that's dedicated to thinking through the divisions that pattern dominant—rationalised and masculinised—modes of knowledge production that are founded on the diminution of nature. Tough it has proven to be hugely infuential across the humanities and social sciences—and in particular for feminist science and technology scholars—it requires some explication, because much of its substance is articulated in her critical engagements with what she calls "technoscience"—the institutionalised and industrialised practice of conducting scientifc research and producing technological innovations for proft (1997). While we no doubt associate this practice and the fgure most closely with her most infuential piece of writing—"Te Cyborg Manifesto" (1985)—she comes closest to articulating what fgures are in later works.

In *When Species Meet* (2008), Haraway conceives of fgures as "materialsemiotic nodes or knots in which diverse bodies and meanings coshape one another" (4). Tis claim bears further unpacking, and it helps to read her statements at the start of this book with some at the start of another. In *Modest\_Witness@Second\_Millennium. FemaleMan\_Meets\_OncoMouse: Feminism and Technoscience*, Haraway characterises fgures as "tropes." Playing on this word's implicative richness and its capacity to evoke both fgurative use of language and, given its origin in the Greek word *tropos* that is, "to turn"—a sense of spatiality, movement, agency, and worldly instantiation, Haraway bends fgures into world-making contrivances (Phan 2019: 24). Her fgures are precisely *not* "representations or didactic illustrations," or semiotic phenomena that operate in a symbolic register, as do metaphors, analogies, or allegories, but conjunctive entities in which "the biological and literary or artistic come together with all of the force of lived reality" (2008: 4). It's hard to resist falling into a poetic register when trying to articulate what fgures are because they gain so much conceptual traction through this reactive meeting of modes. Teir tropic quality—their tendency *to fgure*, in the active sense—can only be understood conjunctively, as an "implosion of sign and substance, a literalness of metaphor, the materiality of trope, the tropic quality of materiality" (Haraway quoted in Hughes and Lury 2013: 795).

Grasping the tropic plenitudes contained within fgures helps us to understand Haraway's oft-quoted proposition, that fgures ought to be understood as "performative images that can be inhabited" (1997: 11). "Inhabitation," here, doesn't construe fgures as *containers* for other hybrid—entities. Tese fgures are real and actual entities (Hughes and Lury 2013: 795), but they don't exist outside of the tropic plenitudes that they gather. Tey shape and are shaped by these plenitudes, constantly turned by them even as they turn them otherwise—towards other fgures. Tis gathering—or fgur*ing*—is what invests fgures with their actuality and what takes them beyond being strictly semiotic entities. Haraway makes this clear in *Modest\_Witness@Second\_Millennium*:

[f]or example, think of a small set of objects into which lives and worlds are built—chip, gene, seed, fetus [*sic*], database, bomb, race, brain, ecosystem. Tis mantra-like list is made up of imploded atoms or dense-nodes that explode into entire worlds of practice. Te chip, seed, or gene is simultaneously literal and fgurative. (1997: 11)

Figures can be used as pivots that articulate worlds that are made in and through the messy and hybrid practice of doing and living. Tese worlds are entangled in relations that don't respect epistemological distinctions, like nature versus culture or, indeed, theory versus practice. So, for Haraway, to "inhabit"—understood, in the broadest sense, as *being*, played out by all kinds of entities through what they *do*—is to fgure and be fgured.

Herein lies the fgure's double function. When Haraway claims that "[w]e inhabit and are inhabited by such fgures that map universes of knowledge, practice and power" (1997: 11), she makes fgures diagnostic tools for understanding how worlds are put together. Te titular "modest witness," for instance, fgures a mode of scientifc knowing underpinned by a self-efacing—hence "modest"—scientifc subject who guarantees scientifc knowledge by witnessing its production through demonstrations (1997: 32). As Haraway notes, the right *to* witness belonged to those gendered male, raced as white, and of a moneyed elite. But what has made Haraway's fgures so infuential and suggestive for critical thinkers is that their world-making capacities can be used creatively as well as diagnostically (see Dawney, this volume; Bastian 2006). Because fgures make worlds, one can make alternate worlds by making alternate fgures. Tey aren't just maps *of* worlds; their tropic qualities mean that they actively *map* worlds, drawing them together in their wake.1 Figures promise their proponents not only a means of understanding how worlds cohere but also a means to conjure alternate worlds that might just harbour more equitable, more just, or more sustainable ways of being-together.

While acknowledging that the creative potential Haraway invests fgures with has been infuential, I want to emphasise her claim that they can be "inhabited." What makes Haraway's fgures—and, arguably, fgures *per se*—such useful and efcacious theoretical tools is their capacity to render complexity something that can be lived with, lived through, or lived in. Recall the "mantra-like list" of chip, gene, seed, foetus, database, bomb, race, brain, and ecosystem. Construed as fgures, these things become points of conjunction from which worlds emerge. To reduce them to representations not only re-introduces the separation between symbolic and material that fgures are designed to dissolve, it also eschews the basic theoretical insight of this conception of being. Inhabitation being and living—must necessarily be understood as being-*with*. So conceived, "inhabitation" is not a state that one simply chooses to adopt for a time before choosing another. Figures have more agency than this: they embroil *us*—conceived, broadly, against distinctions like those between humans and their others—in their worlds. Tis is where the claim that fgures are "actual" gains its force.

Here, though, we also butt up against the limits of Haraway's conception of fgures. We can explain how by asking a refexive, epistemological

<sup>1</sup>We need to attach a caveat to this invocatory power: fgures, as commentators like Astrida Neimanis have noted, are rife with "dangers" (2013: 26), because one doesn't always know whether the worlds actualised through the tropic fgure will be for the good. We see this most clearly in Haraway's most infuential fgure, the cyborg (1985). Tis fgure started out as a means of reclaiming technology for feminist ends and in opposition to "technoscience"; almost four decades later, though, it has arguably been recuperated by these very same forces to fgure hyper-commodifed, masculinist technological futures that reinstitute economic and racial hierarchies (Phan 2019; Cave and Dihal 2020).

question: what are fgures *for*? Tat is, what *problem* does Haraway's critical-theoretical practice respond to? Te force fgures contain has its own epistemological efcacy. Ultimately, thinking with fgures is counterposed to modes of thought premised on distinctions: nature and culture, male and female, human and non-human, and so on—right through to the ontological distinction that fgures themselves challenge, that between the material and the symbolic. Figures don't attempt to dissolve these basic categories, but rather demonstrate how holding them in tension can engender an endlessly productive practice. Te "string fgures" that recur again and again in her thinking fgure this aspect of fgures (Haraway 2016). Te gestures they invite—tying, folding, knotting, forming, and unravelling—dramatise a mode of being and doing in which distinctions-between—here, hand and string, but equally, material and symbolic—are stretched and tested, but never actually undone. Without these distinctions, fgures arguably lose their epistemological efcacy, that is, their capacity to make worlds appear.

In saying this, I don't mean to imply that Haraway's fgures are essentially idealist. Far from it. My claim is that they are designed to respond to a particular kind of (material-semiotic) problem: to show us messiness where we want to see distinctions; to, in other words, unspool relations from seemingly discrete objects. Te problem I want to use Haraway's conception of fgures to think through is simpler. Instead of using fgures to demonstrate the arbitrariness of inherited distinctions, I want to use them to explicate how otherwise-incompatible *things*—understood, broadly, to encompass not just discrete objects but also systems, processes, and confgurations—*become* inhabitable. What I want to propose is that we use fgures' capacity to make worlds in order to live in, through, with the complexity that characterises contemporary media.

I'm interested in a particular case: what I've been calling large-scale media-technical systems. Before turning to the example of "the cloud" to illustrate how fguring renders such systems inhabitable, though, I want to spend some time translating Haraway's concept into a media-theoretical register.

### **Mediated Worlds: Milieus and Non-representability**

Figures fnd a particular kind of efcacy in large-scale media-technical systems. Using Haraway's conception as a basis, the proposition I want to make is that fgures are necessary intermediaries between worlds made by media and their inhabitants. Te point of departure for this proposition is an intuition: digital media are full of fgures because fgures render otherwise-unrepresentable technical ensembles apprehensible and, therefore, inhabitable. Tis proposition relies on three interrelated lessons that I want to draw from media theory and related disciplines.

First, our contemporary situation invites us to expand what we mean by "media"—and consequently, how we understand their capacity to make worlds. Setting aside canonical debates about whether media theory ought to focus on technical devices themselves or on the people, practices, or societies involved in an instance of mediation (see Peters 2010), we can say that, in general, media are typically conceptualised as means of communication: as "middles" that join senders and receivers across time and space (see Guillory 2010). As scholars like John Durham Peters and Antonio Somaini have recently argued, however, this dominant conception of media has always been shadowed by another: the idea that media constitute environments. Peters and Somaini both note that the concept of "media," which comes to the English language via the Latin word *medium*, is the product of a bad translation of Aristotle's work from Ancient Greek. Te source of the word "medium" is a Greek word, *metaxy*, which is not only an intermediary substance or thing but an intermediary *place*: a "middle ground" (Peters 2015: 46; Somaini 2016: 30; see also Kittler 2009). For Peters, the word "medium" has always contained the potential to be understood in an expansive sense, encompassing not just the discrete device, the means, or the middle, but the "element, environment, or vehicle in the middle of things" (2015: 47). Peters recovers this alternate sense of media using the word "milieu," which means "middle-place." Tis concept gives us a way to understand how media can constitute worlds. In Peters' work, such worlds need not be digital; the spread of calendrical techniques, for instance, also makes worlds (2015: 176-184). But it does give us a particularly useful concept for understanding how ubiquitous digital media establish milieus that can be inhabited.

Second, I want to propose that we use the concept of the media milieu to signal a distinction between diferent kinds of worlds harboured by diferent kinds of fgures. Te worlds assembled by the "chip" or "database"—to recall two particular, media-technical fgures invoked by Haraway—are not necessarily of the same order as those assembled by "gene or seed." On the one hand, Haraway's conception of fgures is capacious enough to encompass some categorial splicing. It doesn't matter if the world unfurled from the "gene" intersects or overlaps with the world harboured by the "chip," because "worlds" arise in and through practices and modes of relational being that reorder the kinds of distinctions one might be tempted to make between, for instance, pre-industrial farming techniques and an industrialised agriculture that relies on computational infrastructures for its logistics. On the other hand, there's an argument to be made that large-scale media-technical systems engender a novel *kind* of world.

We can express this in concrete terms. Benjamin H. Bratton has recently proposed that computation has reached such a degree of complexity and distribution over the past few decades that it now operates at what he calls a "planetary scale" (2015). He explains this by pointing to the transformation of computation from a technical process—something conducted by specifc machines on specifc problems—into a "global infrastructure" that supports all kinds of operations in all kinds of spheres (14). Te overarching point he wants to make is that the emergence of planetary-scale computation challenges sovereignty: today, he argues, the global order is organised not only by interactions between nation-states or by the workings of globalised markets but by computational infrastructures—namely platforms—that now rival states and markets for power and infuence (see 327-31). But we can also translate his assessment of contemporary computation into the language of fgures.

Amongst its many uses, this global infrastructure provides new tools to model the world itself. Bratton notes, for example, that planetary-scale computation is a precondition of contemporary climate science, which uses world-wide data collection and huge collaborative modelling projects to understand climate change and to predict the future world it might engender (2015: 305-6). By giving us new tools to model the world, this infrastructure gives us a means of conceiving of the world in its entirety and as a—concrete and material, rather than ontological limit-condition for life itself. Planetary-scale computation institutes a historical break: all "worlds," to re-introduce Haraway's language, subsist in or on *this* one. Does this mean that all "worlds" supervene on computation? Not necessarily. Insofar as all "worlds" supervene on an actual and material *world*—which is how I understand the irreducible "material" part of Haraway's "material-semiotic" couplet—perhaps what it does mean is that those that do supervene on computation are no longer of the same order as the worlds fgured by Haraway even a few decades ago. As the world changes, so, too, does the material from which fgures can be made.

Bratton's concept of planetary-scale computation provides us with a way of qualifying what the claim that media are ubiquitous actually means: media are ubiquitous not only because they are everywhere or because they pervade daily life, but because they constitute new ways of conceiving, and so inhabiting, *the* material world. By referring to worlds fgured in, by, or as media as "milieus," I mean to signal this distinction in kinds of world. At frst blush, it might seem as though the revision of Haraway's concept of fgures I'm proposing amounts to an entirely diferent concept: if mediated worlds are made of distinct stuf, and if we interpose "milieus" for her concept of "worlds," are we not simply slipping her fgures into a diferent conceptual register? Just as Haraway's fgures can be revised using lessons drawn from media theory, though, media theory can also be revised using lessons drawn from Haraway.

In her explication of fgures at the beginning of *Modest\_Witness*, Haraway makes a claim that perhaps isn't always given full weight: "[a]ll language," she says, "including mathematics, is fgurative, that is, made of tropes, constituted by bumps that make us swerve from literalmindedness" (1997: 11). In her conception, fgures pervade the languages we—humans—use to make sense of the world around us. Haraway's claim that fgures are actual material-semiotic things that can be inhabited can be read as a limited claim, referring to particular instances of fguring. But it also contains the potential to be extend much more broadly. My claim is that that media theory actually *needs* fgures to make media both inhabitable and conceptualisable. Figures translate media's otherwise-incommensurable operations into a (conceptual) language that can be used to grasp the conditioning efects they have on our environments and, thereby, on contemporary life. Rather than demonstrating that (ontological) distinctions contain multitudes and messy relations, then, the particular, media-specifc instance of fguring that I'm indicating here does something else entirely: it draws heterogeneity—the objects, systems, processes, infrastructures, and confgurations that constitute planetary-scale computation—into worlds.

Tis mode of fguring is necessary for media theory, fnally, because and this is the third lesson I want to draw from media theory—contemporary digital media and the worlds they engender are incompatible with a particular epistemological operation: *representation*. Scholars who have been working on machine learning and artifcial intelligence, and the platforms that operationalise these techniques, have explained this incompatibility in a number of instructive ways. In many cases, it's impossible to reverse-engineer the automated processes these systems implement. Tis is not only because they operate at a scale that exceeds representation or that the algorithms they use are proprietary—though these barriers are real and difcult to surmount—but because they employ computational techniques that are often correlative and inductive. Once implemented, machine learning techniques of the kind that underpin computational processes—like sorting, ranking, categorising, recommending, and so on—incorporate recursive and self-optimising techniques that will produce diferent outcomes when trained on or applied to diferent sets of data (Mackenzie 2017). It is difcult—or often even impossible—to directly observe what these media do without tools to render them *observable* (Mackenzie 2018; Rieder and Hofmann 2020). For Louise Amoore, these media-technical systems represent a change in data processing's organising "paradigm" from "observation, representation, and classifcation" to what she calls "perception, recognition, and attribution" (2020: 41). Tat is, although the problem presented by media-technical systems premised on large-scale data processing is often presented in phenomenological terms—these systems are difcult to conceptualise because they exceed human representation (Mackenzie and Munster 2019)—it's better conceived of as a problem of un/nonrepresent*ability* or a problem engendered by such systems' complexity and the consequent challenge they pose to eforts to render them not only observable to non-machinic modes of perception but also, as Amoore points out, *actionable* by non-machinic entities (2020: 50; 55).

Tis is why we need fgures to be able to conceptualise media. Per Haraway, fgures are much more than "representations or didactic illustrations." As tropes, they tug at actual relations, demonstrating how they hold together and pulling them in to other and new confgurations. Te fgure of the "milieu" that's gained traction in recent media theory articulates mediated worlds that are not directly representable, and which emerge in the wake of changes to *the* world brought about by the emergence of planetary-scale computation, but which can nevertheless be inhabited. In other words, this means of fgur*ing* allows us to make sense of how complex, distributed, and opaque media-technical systems inform, condition, and shape contemporary life.

To illustrate how this works, I want to shift registers and turn now to a central fgure of contemporary media—"the cloud."

#### **In "The Cloud"**

In 2011, the National Institute of Standards and Technology—a laboratory that reports to the United States Department of Commerce—prepared a document outlining an ofcial defnition of "cloud computing":

Cloud computing is a model for enabling ubiquitous, convenient, ondemand network access to a shared pool of confgurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management efort or service provider interaction. (Mell and Grance 2011: 2)

Pared down to the minimum words needed to assure institutional uptake, this is a bare technical description of what's meant by the term "the cloud." But before it became a standard regulating government procurement, "cloud computing" began in marketing. Antonio Regalado traces the term to policy documents produced in 1997 by Compaq later acquired by Hewlett Packard—that pre-empted the shift that IT companies would eventually make a decade or more later from selling discrete products, like software programmes or computers, to selling distributed storage, computational capacity, or subscription-based software packages hosted on remote servers (Regalado 2011; see Kaldrack and Leeker 2015). According to the likely progenitors of the term who Regalado interviewed, it originally derived from the drawing of a cloud that engineers would use to represent networks between computers (2011). Whether or not these Dotcom-era tech workers are visionaries for coining the term "cloud computing" is a moot question. What makes this anecdote interesting is that it registers the epistemic shifts that this term has undergone between 1997 and 2011 and between 2011 and now.

Let's start by enumerating the diferent ways "the cloud" fgures and is fgured. First, there's the fgure as conceived by computer scientists and systems engineers. Tung-Hui Hu suggests that this particular fgure was frst used by those responsible for computer networks to locate the computers they had "direct knowledge of" in "the same epistemic space as something that constantly fuctuates and is impossible to know," which is to say, the entire system on which such networks rely, but over which they have no control: "the amorphous admixture of the telephone network, cable network, and the internet," amongst other things (2015: x). Second, there's the fgure as conceived by the marketers whose progenitors Regalado was so interested in tracking down. Tis cloud fgures a promise: computation recapitulated not as hard- and soft-ware that has to be administered, maintained, and confgured (see Spencer, this volume), but as a service that can be accessed on demand. We're used to the imagery of this particular fgure of "the cloud": airy, dreamlike, light-flled scenes in which computation is a breeze (see Cramer 2013).

Yet while "the cloud" fnds a certain degree of necessity in these fgures of, respectively, an unknowable epistemic space and a service that is accessible as and when one needs (or can aford), they aren't the primary fgure that I want to focus on here. Tere is a third fgure of the cloud that's of particular interest to us. Tis fgure renders "the cloud" inhabitable for a heterogeneity of users by exploiting clouds' capacity to fgure indeterminacy. Te institutionalisation of "the cloud" by the National Institute of Standards and Technology didn't stop cloud computing from being a marketing buzzword; rather, it gave it new legal and fscal purchase on procurement decisions made by the U.S. government and by companies infuenced by their standard-setting role (see Mosco 2015). When Tung-Hui Hu describes the cloud as "mute piece of infrastructure" that is "just there, atmospheric and part of the environment" (2015: ix), he captures the fgure's gradual transformation from vision statement to banal application with wide-reaching efects. Te proliferation of "cloud"-based computing services has turned an invocatory idea into the environment in which we conduct knowledge-based work.

Let's say contemporary life takes place in "the cloud." In my home, we pay for water, gas, electricity, and internet, but we also pay for access to software like *Creative Cloud*, Adobe's suite of image editing tools, and *Evernote*, a note-taking programme, whilst also getting access to Microsoft's *Ofce 365* suite through our respective employers. Tese services, which we would once have purchased and owned and run locally on our machines, have been transformed into subscription-based services that we pay a fee to access remotely. In economic terms, this transformation represents the extension of a "rentier" model (Christophers 2020) to software: access to computational processes is often no longer secured by ownership, but must be accessed intermittently. One consequence of this transformation is that software has been further "platformised" (Poell et al. 2019; Kaldrack and Leeker 2015). While "the cloud" captures this economic transformation, it also captures the efect that changing access to computation has on contemporary life.

To do work and to engage in leisure increasingly requires access to media. Te diferential nature of this access refgures our relationship *to* media and, by extension, to the means by which contemporary life is lived. Te relocation of software from local machine to "the cloud" has transformed not only our economic relationship to the means of work or leisure, but the "worlds" in which work and leisure can take place. Renting access to software—via subscription or, indeed, by allowing one's data to be collected and monetised—establishes specifc and limited relations between users and "the cloud," understood as distributed milieu. What "the cloud" therefore captures is the imbrication of everyday life *in* media, as modulated by access to systems and services that are not only out of our control but beyond our comprehension. But it also fgures the transformation of these systems' distributed operations from "mute infrastructure" into media that *are* inhabited through multiple quotidian acts of accessing: sending and receiving, requesting and resolving, loading and reloading, streaming and bufering, refreshing and exiting. Te accumulation of these (minor) fgures of access in users' everyday lives images a distributed milieu. Conversely, the unequal distribution of access—to bandwidth, data, particular information, or certain media—images a milieu that's not distributed equally.

What "the cloud" arguably fgures, then, is the capacity for media to constitute a milieu that can be inhabited *despite* being difcult to apprehend *as* media and as source of mediation. "Te cloud" often seems a condensate of nominally opposed qualities. An abstraction that transmutes a network of computational devices and their infrastructural supports into "logical objects" that can be apprehended and acted upon (Hu 2015: x). An energy-and-water-intensive, polluting, world-spanning material infrastructure that is computation's determinate site (Cubitt et al. 2011; Hogan 2015; Velkova 2021). A triumph of marketing, recapitulating computation, once something one owned and managed, as something one can outsource and hire in when needed. A means for turning real qualities into datafable quantities, conferring on us a "promise," as Louise Amoore puts it, that "everything can be rendered tractable, all political difculty and uncertainty nonetheless actionable" (2020: 55; see also Franklin 2012). "Te cloud" is able to articulate these nominally opposed qualities precisely because it's so all-encompassing.

Tis *prepositional* quality, or the capacity to fgure place or environment, is crucial to what "the cloud" is and does. Conceiving of media as encompassing milieus helps us understand what's at stake in fguring large-scale computational systems as akin to a natural phenomenon. In the fgure of "the cloud," old problems of representation merge with cutting-edge media technology. In his analysis of the use of clouds in Renaissance and Baroque paintings, Hubert Damisch uses the fgure of the cloud that recurs in so many paintings of mythological, divine, and secular scenes over these periods to propose an idiosyncratic theory of representation. Clouds are curious fgures precisely because they are an "unstable formation with no defnite outline or colour," but nevertheless possess "the powers of a material in which any kind of fgure may appear and then vanish" (2002: 31). Alongside a general point about the limits of linear perspective, Damisch's analyses conceive of the cloud as a fgure that "reveals only as it conceals" (61) and, in doing so, fgures "the limit of representation, of what is representable" (56).

Tis is what "the cloud" fgures today. Tis fgure doesn't undo a distinction or show us complex relations inhabited by heterogeneous things where we once saw discrete objects—replacing the fgure of the "chip," to recall Haraway's mantra one last time, with a fgure of distributed computation. Rather, fgures like this allow us to grasp how a distributed and heterogeneous process that is otherwise difcult to represent can nevertheless constitute one of contemporary life's integral sites. In lieu of revealing oppositions between form and matter or artifce and nature, "the cloud" condenses an-other place in which inhabitation becomes possible, turning heterogeneity into a diferentially accessed—and so always partially apprehensible—milieu. Access marks out the limits of understanding traced by limits of representation. We use media and are mediated by them, without necessarily being able to make sense of, experience, or apprehend them in their totality. Or: because we're in "the cloud," our diferential access to its particular services only ever gives us glimpses of it entire.

On one side, we have computational systems that shape contemporary life: platforms that are designed to deliver services and which incorporate recursive and self-optimising modes of organisation. On the other side, we have these systems as they produce efects in the world. For the users of these systems, this distinction collapses: there are services, and there is their source, "the cloud," which ultimately amount to the same thing. Rather than acting to "obfuscate" what really goes on behind our screens, Amoore argues that the genius of "the cloud" that it "render[s] perceptible and actionable (almost seeing) that which would otherwise be beyond the threshold of human vision" (41). Tis is Damisch's cloud logic in twenty-frst-century guise: instead of *putti*, the Virgin Mary, Christ, or pillars of cloud representing the divinity, our symbology represents where the work of mediation takes place.

In "the cloud," all we can see is that we're enveloped. Or all we're given to see is that which we can access, at this time and with a given set of resources. Pointing to the ground and declaiming that this, in fact, is where computation happens—indicating its material and/or infrastructural ground—misses the point: constituting a milieu, computation happens everywhere and nowhere. Its location is wherever it's needed to live.

# **Conclusion: Indeterminate Linings**

Figures run over. Commenting on the proliferation of "the cloud" in technology marketing, Peters notes that though this fgure may have originally been taken up "in engineering diagrams of networks," it "almost instantly took to the sky, taking selective advantage of the surplus and residue of the term" (2016: 61). "Te cloud" is, amongst other things, a marketing buzzword, a technical term for computer scientists and systems engineers, a promissory invocation of a technical utopia just around the next bend in the fbre-optic cable—and, woven through each of these, a fgure by which media become inhabitable in the present. As concept, "the cloud" might not hold together. But as fgure, "the cloud"—with its prepositional quality and its promissory lining—is able to articulate what it means to live in and through computation precisely *because* it is indeterminate. In this case, "the cloud" has visual connotations, but they're overwritten by an epistemic function: to make computational systems apprehensible as mediate technologies constitutive of milieus that can be inhabited, diferentially, as, how, when they're accessed (or rendered *accessible*).

Figures are not only inhabitable, per Haraway; they are also necessary for making sense of contemporary places of inhabitation that are shaped and conditioned by unrepresentable media-technical systems. Rather than thinking of "the cloud" as a technical conduit or a neutral container for a networked, technically mediated contemporary culture, we would do much better to think of it as more akin to what Peters calls "climate" specifc, localised, and subject, for each of us, to constant change (Peters 2015: 253-4; see also Horn 2018). In it, we have found a fgure that encompasses the unrepresentability of technical systems that, in their large-scale distribution and their platformised indeterminacy, are beyond us. Earlier, I outlined this problem as one of representability, but my argument has been that rather than operating as metaphors, fgures like "the cloud" respond to the problem of representing otherwiseunrepresentable media-technical systems as milieus.

It's important to note, by way of concluding, that the example of "the cloud" that I've used to illustrate the conception of fgures throughout this chapter could, equally, have been substituted for others, like portals, interfaces, platforms, networks, or data farms. As an example, questions of commensurability—that is, how unlike things are rendered comparable using metrics (Espeland and Stevens 1998; Van der Vlist 2016)—also operate by producing fgures. As Caroline Gerlitz and Bernhard Rieder argue, the interface used to access a computational platform "channel[s users'] activities into predefned forms and functions" (2018: 530). Tey explain this process by drawing on Phillip Agre's concept of "grammars of action," which decompose the uses of computational systems into discrete actions that can be logged and counted as they are undertaken. Te "grammar of action" is also a kind of fgure of speech and/or arithmetic that *re*composes discrete actions as numbers. Tis is how computational systems fgure the qualitative actions of *users* into themselves (Agre 1994). As "lively" metrics that have situated functions that are hard to extricate from their computational contexts (Gerlitz and Rieder 2018: 544), these operations also use fgures to reduce complexity—only, their fgures are of a numerical kind. Construed as means for making sense of how complex, distributed, and opaque media-technical systems inform, condition, and shape contemporary life, the fgure fnds methodological purchase in this media situation, too: we can use it to apprehend how computational systems construct a situated and contingent mode of inhabiting platforms by becoming habituated to their techniques of commensuration. Tis, I think, is the methodological imperative contained in fgures. Let's call this operation "fguring"—understood as a method for thinking media through the fgures that make their operations inhabitable.

But we end in the clouds. "Clouds," Peters says, "resist ontology" (2015: 260). Elsewhere, he also says that "[o]ntology, whatever else it is, is usually just forgotten infrastructure" (2015: 30). In the epistemological space traced by these two statements lies a conception of mediation and fguration for the present. In their complexity and their mutability, the media that constitute what we call "the cloud" aren't always amenable to the kinds of specifcation promised by contemporary theory and philosophy's taste for ontological modes of theorisation, or for (new) materialism. It matters little if we point to a data centre and say that the cloud is *there*. Between how we fgure it and how it fgures us, though, we fnd atmosphere, climate, milieu, *life*—work and leisure, productivity and entertainment, self-expression and data capture. Figures rendered habitable, in other words, as media—and media rendered not just liveable, but *thinkable*, in all their complexity, by fgures.

**Acknowledgements** My thanks to Will Viney, Tom Sutherland, and especially Celia Lury for their patient feedback on early drafts of this chapter—it's worlds better for their attention. Tanks also to colleagues in "People Like You" (https:// peoplelikeyou.ac.uk/) for their ongoing intellectual collaboration: Sophie Day, Helen Ward, and Roz Redd. Tis research was supported by the Wellcome Trust (205456/Z/16/Z).

# **References**


**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/ licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **4**

# **Figure to Ground: Felicity Allen Interviewed by Celia Lury**

**Felicity Allen and Celia Lury**

F. Allen Ramsgate, UK e-mail: fick@felicityallen.co.uk

Te artist Felicity Allen carried out a residency as part of the research project, People Like You: Contemporary Figures of Personalisation (https://peoplelikeyou.ac.uk/). As part of this residency, she developed a new series in her practice of Dialogic Portraits. Tis practice is a form that evolves in series. It generates a portraiture which recognises the labour and experience of the sitter as well as that of the artist, thinking together in the context of the painting that emerges from the sitting. Allen says,

In each series, I usually select and invite a number of people to come and sit for me, working a couple of days with each individual. As I paint them we talk and I make a minimum of two pictures. As people speak our faces constantly shift, and I often try to overlay hints of diferent expressions the pictures are therefore frequently about time spent together, and the relational exchange. At the end of the sitting I invite sitters to sign the work alongside my own signature, in token recognition of their labour, although the work remains mine. Following this, I usually make a recorded discussion with each sitter, and use both recordings and pictures to produce a flm or book.

For this residency, Allen invited her sitters to consider questions of traditional representation, including portraiture, as well as ideas of the digital self, and also made audio recordings with the sitters. The portraits and the recordings form the basis of a 12-minute film, *Figure to Ground—a Site Losing its System* (https://peoplelikeyou.ac.uk/portrait/ ). In this interview with Celia Lury, Allen addresses the significance of relations between figure and ground in her own and others' art practice.

C. Lury (\*)

Centre for Interdisciplinary Methodologies, University of Warwick, Coventry, UK e-mail: c.lury@warwick.ac.uk

**Celia Lury**: Te title of the flm has all sorts of associations, relating to the understanding of a person as a fgure, who stands out—or not—in relation to a ground, to who is visible and who is invisible. And, of course, it relates to perspective and projection in both visual arts and social sciences. How do you approach the notion of the fgure?

**Felicity Allen**: Figures start of as human for me. What to do with human fgures in pictures? As an 'emerging' artist in the 1980s there were considerable prohibitions on portraying fgures—portraits were conservative; narrative painting represented an 'English' failed engagement with modernism, a failure to understand 'painting'; expressionism was identifed with macho self-heroicising; women's bodies shouldn't be portrayed because they would always be subsumed into patriarchal consumerism.

In addition to these dictates, the 'new' media of video and photography were seen as a liberation from the reactionary representations of people associated with painting. Not only was painting necessarily reactionary, watercolour was the wrong medium for portraiture: for years the annual open exhibition at the National Portrait Gallery excluded watercolour while permitting the use of new technologies as soon as they entered the market. Photography took on the language of 'fgure to ground' as a composition technique in which a subject (a fgure) is positioned in relation to a ground by way of camera apertures, although I was frst introduced to the term when learning about Renaissance developments in perspective.

I now know that the ideas one engages with as a young person stay with one; however successfully they've been dispensed with, they're never quite despatched. Tey still play into my thinking about the fgure in relation to the ground—the ground as fgurative background, and the ground as medium. In painting one might make a representation which has a background and speak of fgure to ground, but 'ground' is also the sizing treatment and base colour on the canvas, for the picture itself. In this sense the picture itself is the fgure. So I have been interested in the friction of juxtaposing media apparently in argument with each other in the 1980s black and white photos beside oil paintings, now paintings as an integral picture-making part of a flm. In this sense I'm interrogating the possibilities of portraying fgures against or through diferent types of ground, unsettling the idea that a single ground might defne a fgure.

**CL**: I'm wondering about your mixed use of the mediums of watercolour and flm—including the aperture of perception each afords you as

the artist, as well as the way you mix up portrait and landscape 'page' orientations, a kind of layering of media, acknowledging their history and inter-dependence.

**FA**: I'd experimented with overlaying painted portraits over a flm sequence in *Te Disoeuvre no 1*, a flm fragment I made two years ago, and knew I wanted to explore this way of working in *Figure to Ground*. I started to relate this idea to Piero della Francesca's *Te Flagellation* (late 1450s), which has haunted me for decades. Trough extraordinarily detailed mathematical perspective, the picture is divided into two spaces which harmoniously hang together: to the right, the foreground, with three fgures in contemporary dress as if in conversation and, to the left, a much deeper perspective, showing Pontius Pilate witnessing Christ's fagellation, one and a half millennia earlier. Carlo Ginzburg's book about the painting (*Te Enigma of Piero*, 1985) shows how the architecture, including complex foor tiling, is portrayed with precise mathematical exactitude, securing the harmony.

#### **4 Figure to Ground: Felicity Allen Interviewed by Celia Lury**

Te pandemic, and the consequent gallery closures, has reinforced the way in which our engagement with pictures and image-making is increasingly mediated through several digital layers—photography or flm and then computers. In making *Figure to Ground*, I thought a lot about the way a painting is reproduced through digital media. I wanted to reveal the sense of the tactile and haptic one achieves through paint on paper, as well as the frustrations of the digital proxy: there are three digital processes between the viewer and the original picture (photography, flm editing software, and digital projection or display). It was a difcult decision to include any directly flmed images of people at all in the flm as I wanted to counter the assumption that flm (or photography) is the real, or the true, as opposed to the subjective nature of paint. If most of the representation of people is through digitised versions of paintings, while the background—traces of a hoverport that no longer exists—is made visible in digital flm, are the people real, or is the landscape? I complicate this question with the use of multiple voices along with the diegetic sounds of wind and sea as part of a composed soundscape.

I watched Black Audio Film Collective's 1986 *Handsworth Songs* again recently and was intrigued at how prescient it was in its mix of diferent types of image-making through flm: including flm of statues, a creepy model of a clown, archive documentary flm footage, and projections of black and white still photography as if hanging in a roomlike space which transforms into a series of fat images overlapping each other on the diagonal. Tat representation of photography was very deliberately questioning truth and presentation. In the period they were making that I made a deconstructive installation of the statues in the Victoria Memorial alongside colonial tea and cofee ephemera picturing black people as servants or agricultural labourers, and I think our generation was very conscious of arguments about picture-making, public space, and visual culture. Mostly these ideas were pursued by photographers and flm-makers rather than painters, but I just couldn't get along with the implements of the technology: I needed to work with the wetness of paint.

**CL**: Te flm *Figure to Ground—a Site Losing its System* seems to have a doubled approach to ground. It both locates the portraits in relation to a specifc place—Ramsgate, a port in Kent that has a long history as a signifcant point of entry and departure to England—and suggests that the ground for our lives is disappearing, maybe even being destroyed. Is this doubled approach also a way of thinking about the changing conditions for site specifcity?

**FA**: Tis is a great and perplexing question and a simple answer is, yes.

When I decided to pick up the US sculptor Robert Smithson's phrase 'a site losing its system' I was thinking of three types of sites: the site, as in a specifc site and their parallel 'non-sites' elucidated by Smithson (https://holtsmithsonfoundation.org/provisional-theory-nonsites); the site as the commonly foreshortened term for a website; and sight.

Te flm was made throughout the frst year of the Covid pandemic and, for me, virtually a year of lockdown. It was a year in which four global issues were highlighted in British politics: the treaty to take the UK out of the European Union, climate heating with the destruction of species, habitats and ecologies (implicitly connected with the global pandemic), Black Lives Matter as a global movement and the government's military response to people seeking UK asylum. For the flm I had thought I would be focussing on questions of the face, including facial recognition, but the Hong Kong demonstrations for democracy, and the use of face masks as resistance, followed by the use of face masks as protection against Covid, made this seem too topical to work with. I needed to pause in order to think, to fnd a way to introduce duration, if not narrative or history, into the flm. Te hoverport represented the fact that technologies come and go. Tings do change. Nature comes back. As well as the ground shown here being a site of entry and exit, there is potential for hope in the fgures the flm makes visible. Perhaps it's worth looking as precisely at individual sites as at individual faces, in order to think about what you've called the changing conditions of site specifcity.

All images in this chapter are flm stills from Felicity Allen, *Figure to Ground—A Site Losing Its System*, Digital Film, 2021.

Sitters from Felicity Allen's Dialogic Portraits included in her flm *Figure to Ground—a Site Losing its System* for People Like You: Contemporary Figures of Personalisation: Rashid Adam, Ego Ahaiwe Sowinski, Stanley Allen, Jemima Brown, Ayaan Bulale, Janice Cheddie, Abi Cooper, Luke Eastop, Jason Evans, Yael Gerson, Raga Gibreel, Althea Greenan, John Hall, Ollie Harrop, David Herd, Huang Jing-Yuan, Fiona Johnstone, Sue Jones, Sophia Lee, Zoe Lee, Lunatraktors (Clair le Couteur, Carli Jeferson), Antoine Marinot, Ruth Novaczek, Amarnah Osajivbe-Amuludun, Betsy Porritt, Kamsan Sivakumar, Salih Osman, Simon Smith, Dan Scott, Trish Scott, Gerrie van Noord, and Will Viney.

**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/ by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **5**

# **The Research Persona Method: Figuring and Reconfguring Personalised Information Flows**

**Liliana Bounegru, Melody Devries, and Esther Weltevrede**

We live in a time of intense political polarisation worldwide, fuelled by manipulated and manipulating personalised information fows. How are these troubling kinds of personalisation accomplished? How can the dynamics of personalisation—from algorithmic recommendations to targeted advertising—be studied up close, as users might experience them? What might personalisation tell us about how troubling content travels? Tis chapter examines the prospects of assembling research personas as a way to obtain "critical proximity" (Latour 2005; Birkbak et al. 2015) on

M. Devries

E. Weltevrede (\*) Department of Media Studies, University of Amsterdam, Amsterdam, Netherlands e-mail: e.j.t.weltevrede@uva.nl

L. Bounegru

Department of Digital Humanities, King's College London, London, UK e-mail: liliana.bounegru@kcl.ac.uk

Communication and Culture Department, Toronto Metropolitan University, Toronto, ON, Canada

how personalisation is produced, encountered and experienced online, drawing on previous and ongoing digital methods projects on misinformation, disinformation and authenticity. In complement to research approaches that undertake larger-scale studies of personalisation at a distance through statistical and computational techniques, we examine how persona methods may enable researchers to explore, study and *fgure*<sup>1</sup> personalisation up close through interplays and dynamics of algorithms and user features. Drawing from research that proposes that we look not only *inside* but also *across* algorithms (Ananny & Crawford 2016; see also Seaver 2017), and from digital methods approaches to repurposing "methods of the medium" (Rogers 2013), we discuss how confguring research personas can be used to study how personalisation is produced and accomplished through various actors, devices, interfaces, infrastructures, methods, techniques, user practices and data fows.

## **Challenges to Understanding Personalised Information Flows**

Researchers have built reliable tools and curated large amounts of data to track the spread of harmful information online.2 We can now map the spread of fake news and identify institutions, groups, individuals and bots that actively participate in mis- and disinformation on social media (Lim 2019). However, these more distant "big data" approaches tell us less about how users experience the highly personalised spaces and practices of current media environments, wherein problematic information forms and spreads. Personalised information fows aim to provoke systematised afective resonances between lived experiences, discourses, ideologies and networked logics. Tese resonances do not require truthfulness to work and thus render fact-checking less efective as an intervention (Devries and Brett 2021). Indeed, anyone's passionate adherence to the demonstrably false is not simply a matter of getting the wrong kind of

<sup>1</sup> See this volume's Introduction.

<sup>2</sup> See, for instance, https://digitalmethods.net/, https://publicdatalab.org/, https://digitaldemocracies.org/.

information. Instead, this devotion points to a coming together of antagonistic, highly charged responses and expressions that refect deep societal divisions. Personalised information fows channel complex processes of identity and group formation in search for connections with "likeminded people". Tis widespread search for familiarity constitutes the infrastructural principle of social media platforms. Indeed, Wendy Chun (in Apprich et al. 2018) has highlighted the homophilic (from homophily: love of the same) dynamics that sustain any social media and associated recommendation algorithms that group users according to similarity along some dimension, assuming their desire for such, and exacerbate diferences between groups. Tese dynamics are said to facilitate the emergence of "flter bubbles" (Pariser 2012), where users habitually interact primarily with users and content that share their values, likes, and preferences.3

However, the paths that lead to embracing mis- and disinformation via personalised information fows are not straightforward. For instance, a young, white, unemployed man engaged with white nationalist groups radically difers from a middle-class conservative woman focused on raising her children. Yet, these two individuals can end up in the same online spaces claiming that Covid-19 is a conspiracy. Te shaping of personalised information fows, in other words, depends on the particular and personal lived experiences of online users, yet is homogenising nonetheless.

Our schema of the complex and contradictory processes present in personalised information fows acknowledges three interlinked elements. Te frst is the *algorithmic infrastructure* of personalisation on social media platforms, specifcally the algorithmic curation of content for individual users based on a programmed, homophilic model that organises communities based on an assumed desire to see like-minded content. Te second is the shaping of manipulation, such as mis- and disinformation discourses, by corporate and institutional actors using *data fows* to afect user interactions. Cambridge Analytica famously explained how they grouped users according to psychographics and socio-economic data (Venturini and Rogers 2019). Tis allowed them to target each group

<sup>3</sup>Tis notion has been empirically interrogated and challenged by media researchers (e.g. Bruns 2019).

with specifc discourses concerning issues like national values and identity. Te third is the triggering of *cognitive and afective responses* that tap into lived experiences. Daily streams of targeted content cultivate afective responses (such as joy, fear, paranoia and rage) over time. In turn, these triggers concretise users' identifcation with the narratives and realities depicted by this content (Devries 2022).

Tere are several challenges to undertaking a multi-layered analysis of these elements, which led us to formulate the research persona method. First, it requires reconciling two radically diferent analytical approaches: the "view from above" via broad structural dynamics and the tracking of information fows, and the "view from within", that is, the afective, personal experience of users. Second is the problem of obtaining data in the frst place: most social media platforms are for-proft enterprises whose revenues are based on keeping and selling user data, and information about how such data is mobilised, unavailable for critical investigations.

Researchers have had to rely on reverse engineering methods, creating and repurposing data tracking tools just to glimpse automated personalisation processes. Another avenue for research into personalised information fows is through qualitative ethnographic methods such as participant observation (Hine 2008) and obtaining stream captures from real-life participants via internet panels or voluntary donations (Nechushtai and Lewis 2019; Bechmann and Nielbo 2018; Puschmann 2018). Yet, these more embedded approaches are not without issues. While they enable researchers to investigate the type of content recommended to users, they are less suited for capturing the interplay between user practices and algorithmic recommendations. And importantly, it is this interplay that provides curated content that in turn provokes actions that may shape, solidify and spark political views over time, as demonstrated in the storming of the US Capitol in January 2021, following months of intense mediated and networked propaganda centred on election fraud conspiracies.

In the context of studying personalised information fows, digital ethnographic approaches (as discussed, e.g. by Boellstorf et al. 2012) are particularly challenging not only because users inhabit multiple online cultural spaces at once, but also because it is difcult to gain trust and consent from users already distrustful of academic research (Phillips 2015). Online environments inundated by conservative or far-right talking points have historically attacked the academy and in particular have marked the social sciences as untrustworthy. Tis considered, research into manipulative information can be high risk for researchers. Relatedly, it can be challenging to gain the support of Research Ethics Boards for this type of research while ensuring researcher protection. Te research persona answers some of these concerns in a novel way: what if, instead of looking at others as the subjects of misinformation, researchers were to take themselves as the subjects of misinformation? Rather than examining "what do they see?" through interviews or ethnographies, one can explore "what would I see?" by exploring the interplay between user practices and algorithmic recommendations to assemble a relational perspective on the dynamics of personalisation.

#### **Situating the Persona as a Research Device**

Te research persona ofers a way for researchers to overcome these challenges and make visible and researchable the key moments of interplay between the three elements described above: *algorithmic infrastructure*, *data fows*, and *cognitive and afective responses*. We examine how personas may be confgured in digital methods research, including examples incorporating interface analysis, customised software and speculative methods. Te practice of the method produces research materials that, for the reasons discussed in the previous section, would otherwise not be available for critical inquiry. At the same time, the research persona method emerges from and challenges persona-based approaches in other felds, as we discuss below.

Te term "persona" has various defnitions and roots in diverse felds, including theatre, literature, anthropology, sociology, cultural and media studies, design, software development and marketing. Perhaps its earliest use is as "dramatis personae" and refers to the tradition of theatre actors wearing masks to signify character types or personas, a practice that continues today. Giles (2020) identifes a key tension in this use of persona, in that it entails both the performance of the self by *individuals*, including in online settings and a set of techniques used to perform a *group or category*, such as professions (e.g. academics).

Te fctitious persona is widespread in various forms of arts and entertainment (games, theatre, novels, etc.), marketing and HCI (humancomputer interaction). In software development and marketing, the persona has been used to represent practices, needs, motivations and behaviours of potential users and customers into archetypes to facilitate innovation and ideation. In these contexts, personas are understood as "fctitious, specifc and concrete representations of target users" (Pruitt and Adlin 2010, p. 5). UX and HCI design in particular (Tomlin 2018; Chang et al. 2008) use personas to represent *a type* of user: a singular entity that stands for a collective—that is, target audience and user groups. Similarly, marketing and advertising practices across diferent felds see the use of personas as tools for audience research to fgure out the needs, desires and wants of diferent populations. In these contexts, crafting a user persona follows a particular set of steps, including extensive research about users via qualitative interviews, existing data sources, analytics, and informal or anecdotal observations (Humphrey 2017; Ricci et al. 2018).

Te concept of a persona has also been a central object of study and analytical tool in new media and cultural studies. Beginning in celebrity studies (Marshall 2014), the concept of "persona" informs the study of the performance of the self by everyday social media users, microcelebrities (Marwick 2013, 2015) and infuencers (Abidin 2016), and has led to the emergence of a "persona studies" feld (Marshall and Barbour 2015 and Marshall et al. 2015). Here, the use of personas helps researchers understand how media afordances confgure behaviours shared by user categories. In other words, the analysis of personas and their formation involves the study not only of users but of digital objects as well: Marshall et al. argue that personas can be understood as "networks of digital objects with algorithmic components that have aesthetic and afective properties that enfold in a series of inter-objective and subjective felds of relations" (2019, p. 97).

As a method focused on processes of fguration, understood as "the activation of methodological potential in a process that is neither teleological nor mechanistic (…), but instead is a becoming-with" (Lury 2021, p. 40), the aim of the research persona is not to inform product design but to allow access to situations that enable the researcher to understand how digital infrastructures respond to user practices and how these responses are in turn experienced by the users. Te aim is to use the persona as a new media research device for studying platform personalisation and apps by locating the research amidst personalisation fows.

In the context of studying search engine personalisation, Feuz et al. (2011) have used fctitious Google accounts with carefully curated web histories to explore features of personalisation on Google Web Search. Relatedly, the walkthrough is a method for critical socio-cultural analysis of apps from a user-centred perspective (Light et al. 2018; Dieter and Tkacz 2020; Weltevrede and Jansen 2019). It invites the researcher to create a fctitious user persona to gain access to the platform features and interfaces to be studied. For example, Dieter et al. make a case for the research persona in the context of studying apps as a "methodological user surrogate, enabling access to app interfaces while facilitating heterogeneous research situations" (2019, p. 5). Notably, the notion of the "research situation" helps distinguish the research persona method from applications in other felds. For the persona to become a research device, it needs to be enlisted in and aligned with the purposes of research. How the persona is confgured, that is, what decisions are made about the sites, digital objects, activities and connections that make up the persona, depends on the research question and objectives. Te research persona can involve research techniques to obtain otherwise unavailable data and insights. For example, the use of personas in marketing and advertising research involves various methods to collect data about users and their characteristics, needs and situations (Caballero et al. 2014; Armstrong and Yu 1997). As well, research personas can be both automated (e.g. social bots) or directed by human actors. In social research, personas can be used to study spaces and processes that would otherwise be difcult to access. For example, ethnographers may adopt a persona when they undertake covert research (Hine 2004).

Personas are also sometimes used as a fctional device to get to the truth of a situation. Te use of fctive measures as opposed to relying on already-observed facts resembles a technique used by the New Journalism movement (Hollowell 2017; Pauly 2014) in the 1960s and 1970s, and more recently in journalistic investigations such as Roberto Saviano's book on the global trafc of cocaine (2016). Here the recourse to fctional elements and speculation inspired by imagining what must have happened aims to make a situation more real and authentic to the reader. In such deployments of the persona, imagined elements are not antithetical to understanding lived realities. Rather, as in fction writing, it can be a resource, a speculative point from which the process of fguration takes shape. Imagining other persons is key to developing empathy and understanding of the social, economic, ideological and cultural factors that deeply infuence people.

## **Confguring the Research Persona: Methodological Considerations**

With these points in mind, we situate the research persona as methodological experimentation alongside methods taken up by design, sociology, and media studies that engage speculation as part of the research apparatus (Dunne and Raby 2013; Wilkie et al. 2015, 2017; Benjamin 2016).

More specifcally, we draw on three methodological approaches. First, following digital methods principles, we acknowledge that researching digital objects, whether users, content or behaviours, requires mediumspecifc methods (Rogers 2013). Digital methods track the various forms digital objects take as they circulate from back-end to interface and from one platform to the next. Tis approach informs the creation and use of the research persona in such a way as to make possible the tracing of the various digital objects that confgure it: from profle information and images to status updates, likes, location and connections with other users. It also prompts us to attend to how the platform's personalisation algorithms respond to the persona's actions and features. In turn, the research persona also enriches the digital methods repertoire. Digital methods research often focuses on public platform spaces such as pages and groups and on research approaches that cultivate the "view from above" mentioned earlier (such as through the confguration of a "research browser" which seeks to disentangle the researcher from browser histories, preferences and personalised results, or through API (application programming interface)-based data collection). To complement these approaches, the persona ofers a way to examine private and personalised user spaces, such as the news feed and other personalised recommendation spaces, as sites of user fguration.

Secondly, we take the sensibility towards the inseparability of collective and individual experiences from digital ethnography and participatory design. Tis means understanding the becoming of users as connected not only to the technologies surrounding them but to the actions of other users with whom they are algorithmically afliated.

Tirdly, we draw on speculative methods (see, e.g. Wilkie et al. 2015, 2017) and performance studies (see, e.g. Madison & Hamera 2006). Te research persona is a speculative device; it involves producing an artefact to prompt an algorithmic personalisation situation that is inhabited over a period of time. Our collaborators, the scholars and artists Ioana Juncan and Alexandra Juhasz, pointed us towards character-building techniques from theatre practice to create such an artefact, such as those taught by Elmo Terry-Morgan at Brown University.4 Tese techniques, which encourage the research team to collectively imagine the persona's background and life story, signifcant life events and relations, as well as how they look, speak and think in their everyday life, evoke empathy within the researcher for an individual who might be situated amongst these fows in daily life, and are crucial to thinking through how a particular person would react to diferent types of content and afective charges.5 From this position of the speculative user experiencing personalisation, researchers can investigate and problematise the work that algorithmic recommendation systems do as part of manipulative information fows. Tis lends insight into the efectiveness (or lack thereof) of fows of misand disinformation. It is also an inventive device in its experimental, modifable and situation-specifc approach (Lury and Wakeford 2012) to making visible, researchable and accountable the social and technological processes that integrate manipulative personalised information fows.

<sup>4</sup>https://www.brown.edu/academics/theatre-arts-performance-studies/elmo-terry-morgan

<sup>5</sup>For more details on the fctional persona construction exercise run by Ioana Juncan and Alexandra Juhasz as part of the project Te Research Persona as Digital Method, see https://wiki.digitalmethods.net/Dmi/SummerSchool2019ResearchPersonaAsDigitalMethod.

#### **Three Examples of Research Personas**

To become a research device, the persona needs to be aligned with a research apparatus containing questions, objectives, research angles, analytical lenses and a particular narrative style (Marres and Gerlitz 2016). Furthermore, the confguration of the research persona takes on diferent forms depending on the research questions. In this section, we explore three ways of confguring research personas to address diferent lines of inquiry, as summarised in Table 5.1.

To further explain this, let's go back to how an online user is perceived by diferent layers of digital networked media. From the perspective of


**Table 5.1** Three examples of how research personas can be confgured

the infrastructure, which involves networks linking together platforms, apps, data trackers and so on, the user is a source of data. Tis data is provided by the user (e.g. login, likes, comments, posts or banking information), as well as derived from monitoring and stored information (e.g. device used, quality of connection, location of connection, credit rating information or connection time). Te infrastructure, in turn, can be used to aggregate data fows, not only to group users together but also to link diverse data about the same user on diferent platforms, apps and sites, so that, for example, buying patterns on one website become part of the personalisation process on a social media platform. At the back-end infrastructure that hosts these informational ecosystems, the kind of persona that needs to be developed is what we call an *infrastructure and algorithmic persona*. Tis particular type of research persona is designed to be recognised by a system (e.g. an app or a social media platform) that tracks the persona across platforms, websites, apps and so on.

In one of our research projects, we used research personas alongside software tracking tools to follow the circulation of user data on diferent networks and platforms via apps.6 We focused on dating and gaming apps to capture "app events", such as the processes by which two users are matched with each other or swiping an advertisement in dating apps to capture the data connections that are established in the back-end. In this particular research project, we paid attention to tracking the data emanating from the device (in this case, a smartphone) that our persona was using. We created two research personas and experimented with the level of profle detail necessary to orchestrate a match between our research personas. Sketching out broad partner interests and restricting the geographical proximity proved enough detail to facilitate the match. Furthermore, to ensure that our research personas are not accidentally confated with existing dating app users, we ensured that the personas were unique: we avoided generic names, as well as frst name and last name combinations that already existed. However, the socially meaningful event of the match is a data-poor moment in the back-end of dating apps, which becomes apparent when compared to the app event of swiping an advertisement (Fig. 5.1). To trigger advertisements on dating apps,

<sup>6</sup>https://wiki.digitalmethods.net/Dmi/SummerSchool2019ResearchPersonaAsDigitalMethod

**Fig. 5.1** The app event of swiping an ad: Data shared with ad networks using Tinder. Visualisation by Alice Ziantoni and Noemi Schiavi (Density Design), Digital Methods Summer School 2019

we experimented with how much profle detail, user activity and account connectivity is required. Installing a dating app and logging in with a new research persona account on one of our personal phones immediately triggered ads in the dating app. In contrast, none appeared on a clean research phone, signalling the relevance of the embeddedness of the research persona account in an active device and connected media ecosystem. Compared to the match, swiping an ad is a data-rich app event that allows us to further defne the roles of apps within the networked economy as brokers of user data (Weltevrede and Jansen 2019). Tese initial explorations suggest that app events are multifaceted and simultaneously become manifest and meaningful on the user frontend and the app/server back-end in diferent fgurations.

Te user takes on a diferent form from the perspective of the digital user interface. Here, instead of data, what is perceived is a series of media objects: profle and other pictures, for instance, name, content creation and reactions (via like buttons, for example) to other content and users. Te use of *the interface persona* brings to the fore the choices made in how a person presents themselves. Te aim is to better understand afordances and cultures of persona formation across diferent platforms. Tis includes the kinds of platforms and apps they typically would use and for what purposes, the types of vernacular modes of expression they would use depending on their particular ways of being, and the kinds of communities they would want to belong to. Here, it is the construction of the persona as an everyday user in the frontend interface that becomes the central focus of analysis. It requires that the researchers review several user profles to get a sense of how people act within specifc online communities, depending on their biographical details.

#### **Research Protocol on Interface Personas from Digital Methods Summer School 2019**

#### **What is the data space of persona-making according to diferent platforms, devices, infrastructures and media spaces?**

Considerations


Protocol

	- *Tis could be based on the qualitative analysis of platform interfaces (including the advertising interfaces), platform APIs and data, documentation, third-party platform features, etc.*

Te interface persona, which borrows heavily from the walkthrough method and UX design, does not involve the persona interacting with other (non-fctional) users. Rather, it is used as a device to attend to how platform features and cultural practices are involved in personalisation. We see this approach as particularly useful for setting up coding schemas to analyse, for instance, modes of engagement and the rise of new vernacular language and practices. Tis approach allows the researcher to focus on user engagement with content and the kinds of community building that emerges from such interaction. Critically as well, the interface persona pushes beyond questions of content in mis- and disinformation towards questions of community practices and values and how trust is built among users.

Te third kind of research persona is the *speculative persona*. Tis involves the creation of a complex character to attend to the diferent afective resonances that media objects, such as a fake news article or a politically biased meme, can have with users, and how this, in turn, cultivates and strengthens specifc modes of action, from voting choices to participating in demonstrations and illegal activities. Te speculative persona has a unique name and a face (e.g. generated via AI), and a rich and detailed biography.

Te speculative persona design is a collaborative process among researchers, with one of the purposes of the collaboration to bring to the fore and challenge researchers' preconceptions and assumptions. Our collaborators designed a persona on personalised misinformation fows in the Canadian context who was a young 22-year-old white man living in a post-industrial city in Ontario (see Neville and Langlois 2021). Te research group worked against creating a stereotype by including precise details about the character to develop a life story representing the potential features that might afect detailed personalisation fows that rely on emotions like aggrievement, entitlement or economic frustration. For example, they imagined the character as having grown up watching the status of one's middle-class parents disappear and frustrated by liberal politicians. Actively borrowing from theatre techniques, the researchers created a persona that they could relate to and empathise with rather than a device or stereotypical model to get at data. Inventing life details was key to identifying the persona's worldview regarding both interpersonal relationships and interest in specifc political and social issues. Deep consideration of how the persona thought of himself, what kind of hobbies he engaged in and the kinds of relationships he was pursuing further helped create a rich character. In the end, the character was derived from both sociological knowledge (e.g. available studies and media reports) and personal knowledge (e.g. friends and family members who had developed far-right tendencies or who had similar life experiences) on the part of the research group.

Once constructed, researchers then activated the persona on two platforms that corresponded to his profle—Facebook and Reddit—and subscribed the persona to a range of subreddits (from relationship advice to gaming to conservative and alt-right politics) as well as Facebook groups dealing with local politics. During weekly sessions, they recorded the recommended content for their persona on these two platforms, and as a group, discussed how the persona would react to diferent recommendations and why. Tis made it possible to understand the content, style and tone the persona would adopt in interactions with platform content. By liking some of the content, they were further able to see what kinds of recommendations were algorithmically provided. In so doing, they paid attention to how mis- and disinformation content can be tailored in many diferent ways, not only through diferent media forms but also through diferent rhetorical styles (e.g. passionate vs pseudo-scientifc).

In sum, the speculative persona enables the researcher to gain perspectives and insights into how algorithmically mediated content may be encountered and experienced. It serves as a space where researchers may feel and imagine how the persona is touched, provoked, angered or saddened by online content. It also enables speculative understanding about how manipulative personalised information fows cultivate (pre)conscious afective responses and, as much as possible, alternative ways of understanding the world.

#### **What the Research Persona Opens Up**

As discussed previously, the persona method allows researchers to capture fows of information that work to confgure the experience of individual users who interact with various content online. Unless we use autoethnographic techniques, these fows interacting with users whether algorithmic recommendations, media objects or images, or emotional comments—and users' afective responses to them are typically challenging to study. Using the research persona allows the researcher to enter a manufactured political positionality similar to (but not equivalent to) the situated experience of a user belonging to a particular community. Tis allows the researcher to experience similar fows of information and media objects as everyday users as they sit and interact amongst them. While this positionality does not enable us to address the question of how an actual user or certain demographic sees and responds to content, it does open up the possibility of exploring what platforms show in relation to what users do, that is, the dynamics that emerge between user practices and algorithmic recommendations in experimental settings. In this setting, the researcher's experience itself becomes the object of analysis, as opposed to the data or responses of others online. In this sense, the research persona can be considered a method that sources its fndings from embodied processes that happen at the level of the researcher themselves, as they fnd themselves encountering content, triggering data output and responding in various afective ways (confusion, intrigue, anger, excitement, etc.).

In other words, the research persona is not meant to further a researcher's interpretation of a subcultural space or represent that space through description. Tis is because a cultivated positionality cannot be said to *represent* the experiences of other users in hypothetically similar online/ political positions. While the persona builds on observed knowledge of online political trends (Haseman 2006; Snyder-Young 2010; Elliot 2017), its primary source of knowledge is the experience of the researcher themselves as they situate themselves within digital networks and record their embodied experience, from sights and semiotic interpretations to afective responses. Because of this, the research persona can be considered to follow other experimental and performative-ethnographic methods that change the relationship between the researcher and the researched (Pollock 2006). Here, the research persona no longer marks the researcher as "subject" and the interlocutor as "object of study." Instead of existing as separate entities, the research persona turns the researcher's embodied, interactive experience with information fows and other digital actors (homophilic networks, texts and images, and haptic platform features) into both the one who studies and the one who is studied. Trough this recursive approach, the research persona allows the researcher to encounter and thus map specifc interactive moments between themself and the digital fgures and entities surrounding them, efectively bringing to the fore the fguration work of automated data exchanges, taken-for-granted communicative afordances, and (pre)conscious habits and reactions that previously avoided critical analysis.

While the research persona borrows some techniques from covert research, it modifes these in signifcant ways to refect the diverging goals that it serves. Whereas covert research methods in sociological or anthropological studies have seen the researcher take on a fake identity to gain access to hard to reach or precarious social spaces (Calvey 2008), such as underground drug-distribution networks or Neo-Nazi organisations (Shoshan 2016), the persona method is not being used to study particular social groups. Rather, the persona method examines the dynamics that emerge between algorithms, platform infrastructures and users. Performing a fctitious identity in relation to the studied groups is essential in this process. In contrast to this, concealing the identity and aims of the account is not a necessary condition of research when using the persona method since the purpose is not to study particular groups of users (as is often the case with covert research). Instead, the interest is in observing the responses of the medium, both on the front end (e.g. in terms of content recommendations) and on the back-end (e.g. in terms of data exchanges). Diferent options are thus available to the researcher using this method. Te researcher may opt to make the persona account public, ensuring that its fctional character and its research purpose are clearly stated in the user profle, and perhaps including a webpage that provides more information about the project and a means by which the researchers can be contacted for more information by users. Te researcher may also opt for a private account whose profle and activities are not visible to other platform users and respond only to accounts that were not linked to specifc individuals. While the latter option may minimise the possibility of users initiating interactions with the research account, this decision may also shape information fows in ways that inspire further investigation. In either case, as Light et al. (2018) suggest, what is important is that the researcher devises strategies for responding to potential interaction attempts from other accounts to ensure not only that harm is avoided but also that relations cultivate care, empathy and mutuality as much as possible (see, e.g. Tiidenberg 2020).

While being refexive in their analysis, in covert research, the researcher does not include or feature themselves as the primary object of study. Rather, the majority of their data is sourced from the behaviours or conversations of their interlocutors. Covert research is classifed as such when there is information withheld from interlocutors to gain access to information or data provided by such interlocutors (Calvey 2008). Since the object of study in the research persona method is the dynamics of personalisation, this method is more appropriately qualifed as a performative or inventive approach to evoking fndings (e.g. Culhane 2017) than a covert study of given individuals or groups.

Regardless of these distinctions, the method raises important ethical and legal challenges which require thorough consideration. While the precise confguration of legal and ethical considerations to accompany the use of this method depends on the purpose of the research, and the legal frameworks in place, a set of concerns demand attention. First, while our application of the research persona avoids interactions with individual personal user accounts, the use of the persona nevertheless implies participation in the platform and particular forms of interaction. For example, to cultivate an algorithmically curated environment for the researcher to experience, the persona may interact with public platform content by, for example, clicking on news articles or blog posts and following public pages. Tese actions are recorded by the platform and made available via its various interfaces, including being visible to users. In designing research with personas, it is important to consider how these actions may be experienced as misleading or construed as deceptive and possible harms that could arise as a result. Ethics boards in diferent countries and research guidelines in diferent disciplines may have difering understandings of research involving these techniques. Some may see the use of some of the techniques described above as *de facto* leading to covert research and recommend that researchers carefully broach this question when designing their research projects. We recommend in particular that in approaching ethics boards, researchers give careful consideration to unintended consequences and harms.

Te process of generating and maintaining a research persona also raises other questions regarding ethical research, given its resemblance to tactics used by marketing research. Similar to the above examples of covert ethnographic work, marketing research tends to use the construction of a fake or covert profle to gain information about other users recording their likes, the nature of interactions or personal networks. In contrast, the research persona uses the embodied experience of the userresearcher to provide a nuanced examination of the particular networked positionality cultivated by the research persona's interactions. In this sense, the research persona also difers from previously discussed marketing tactics' use of personas to typify and gather information about other users. While the persona has previously been used as a device to *know persons*—whether through understanding customer needs, more precise segmentation for marketing, to ensure that users are better catered for by-products and services or to gain access to social spaces and communities—we are interested in how it may be repurposed as a device to *understand personalisation*, including the role that platforms, data and algorithms play in shaping contemporary sociality and in producing (sometimes troubling) associative spaces and dynamics.

While research personas may ofer fresh perspectives and promising lines of inquiry for new media research, they should be used carefully, not just in compliance with relevant legal and ethical rules, but with consideration of other persons, communities and users as a central concern in the research process. Creating one or two new profles on large platforms such as Facebook and Reddit—each with millions of users and flled with fake and empty accounts—to study algorithmic personalisation may be less troublesome than using them in smaller groups or more intimate settings. Given that fake and fabricated accounts are now widely studied, one would not want to inadvertently or unthinkingly contribute to proliferating the problem. In line with recent approaches to ethics in internet research, we advocate an ethic of care regarding the specifc circumstances and communities that may be afected by the use of such techniques (Tiidenberg 2020).

Apart from these concerns, the use of personas may not conform to some of the terms and conditions of platforms or apps like Facebook, which have attempted to mandate the use of online profles. Such platform practice itself (such as Facebook's 'real name' policy) is deemed unethical by many advocates (Haimson and Hofmann 2016). In practice, many users have multiple Facebook accounts and aliases for various legitimate reasons, such as further security from trolls, diferentiating between work and personal spheres, or building new family and friend networks. Not complying with platform terms of use is an object of ongoing debate in internet research (see, e.g. Tiidenberg 2020; Marres 2017). Te extent that this provides an ethical dilemma is currently the topic of public and academic debates that weigh the need for public scrutiny of large platforms like Facebook in the context of the problematic efects of personalised information fows and unbridled data-mining rights held by such platforms (see, e.g. Rieder and Hofmann 2020).7 Indeed, much essential research into platform and algorithmic bias requires methodological approaches that would be considered to not comply with legal terms of platforms, and yet may often be sanctioned by ethical review boards and peer review procedures due to its potential contributions to the public good (see, e.g. Eriksson et al. 2018; Sandvig et al. 2016). Tis point is also raised by Marres (2017), who argues that the preoccupation with compliance with terms of service of platforms whose own functioning often raises ethical problems, may distract from the imperative of interrogating the problematic role of these platforms in reconfguring collective life, which would bring additional ethical problems.

As an inventive approach (Lury and Wakeford 2012; Elliot & Culhane 2017), the research persona is a re-imagined way of using the researcher's

<sup>7</sup>For more detail, see in particular the Association of Internet Researchers Ethical Guidelines 3.0 that further delves into how platforms' terms and conditions are being fought against by the American Civil Liberties Union: https://aoir.org/reports/ethics3.pdf. Also see *Spotify Teardown* (2019) for an account of the ethical and legal struggles over corporate terms of use. Christian Sandvig in the US recently won a lawsuit that violating a website's terms of service does not violate the Computer Fraud and Abuse Act (https://www.ef.org/fr/deeplinks/2020/04/ federal-judge-rules-it-not-crime-violate-websites-terms-service).

positionality to open up space for conceptualising and mapping the processes of fguration and subjectivation that occur through digital fows of information and content across social media. Given its creative and performative features, we propose that the persona method be combined with participatory arts and design practices, such as material artefact and prototype design, co-design workshops, creative writing and situated performing arts. Deployed in these contexts, the research persona can act as an "elicitation device" (Marres 2017; see also Lezaun and Soneryd 2007) that prompts engagement, debate, collective learning and empathetic imagination.

An example of this can be found in the interactive online theatre show *Left and Right, or Being Who/Where You Are*, directed by Ioana Jucan (2021). Tis digital theatre performance developed an experimental and participatory process where actors situated themselves in the worldview of various politically charged characters. Tese characters were inspired from other types of qualitative online observation, but their development into unique characters emerged from the actors' use of the persona method; actors engaged in interactions with their online and ofine environment to cultivate an embodied experience that could inform their character development. Apart from insights provided to researcher/performers, uptake of the research persona in this setting also provokes the audience to think of political identities not as static or unalterable but rather as fgurations resulting from relational, embodied processes that occur over time and with technology through interactions. Notably, this kind of use of the researcher's positionality as both investigator and situated participant to produce theatrical insights is not without precedent and is part of new creative ways of presenting and thinking through embodied experience and performance (see *Performing,* Kazubowski-Houston 2017).

In the end, revealing the processes of confguration that shape users online via interactions with information fows is the key insight developed by the persona method; how these processes are explored or communicated either through writing, performance or other creative methods is a fexible enterprise.

# **Conclusion**

Te research persona is a means of fguring the co-construction of user experiences and behaviour at diferent levels of mediation, from back-end to interface to embodied experience. Tere is a defnite artifciality to the fguring process that the persona approach opens up as it creates experimental situations. Tese situations may be envisaged as experimental sites through which data fows, and where algorithmic processes and other personalisation systems and afective resonances can come to the fore. Tis artifciality thus opens modes of inquiry into spaces and dynamics that would not be possible to analyse otherwise. In that regard, the persona is both a research method—a way to set up situations where experiments and analysis can happen at all three levels of information infrastructure, interface and subjectivation—and an object of research. Its core feature is its refusal to set up critical distance, allowing refection on the relationality between researchers, those being researched and the platforms that participate in and enable these interactions. Only by constructing modes of encounter, even if starting with the imaginary as an entry point, can we start fguring out how to reinvest in digitally mediated social relationships and formulate principles for new types of personalised information fows.

**Acknowledgements** We would like to acknowledge Ganaele Langlois' essential role in the development of the ideas within this chapter. Tis chapter builds on collaborative work undertaken as part of the SSHRC-funded project Beyond Verifcation: Authenticity and the Spread of Mis/Disinformation, led by Wendy Chun at the Digital Democracies Institute at Simon Fraser University. It also builds on the project Te Research Persona as Digital Method (https://wiki.digitalmethods.net/Dmi/SummerSchool2019ResearchPersonaAsDigitalMethod), undertaken at the Digital Methods Summer School at the University of Amsterdam in 2019, and on research done by the Canadian Disinformation Network. We are grateful for the contributions of all participants in these projects. We are especially indebted to Wendy Chun, Alexandra Juhasz, Ioana Juncan, Ganaele Langlois, Jonathan Gray, Anthony Burton, Steve Neville and Greg Elmer for their invaluable input. We also greatly beneftted from presenting and discussing this research with panellists and participants at the Figurations: Persons in/out of Data conference in December 2019 at Goldsmiths. Parts of this work were also supported by the Dutch Research Council (NWO) grant number VI.Veni.191C.048.

#### **References**


**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/ by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **6**

# **Engines, Puppets, Promises: The Figurations of Confguration Management**

**Matt Spencer**

## **Introduction**

Te imagination of machines as human-like and of humans as machinelike has been a central facet of modern Western technical culture, most prominent in the kind of "thinking" computers do and the kind of "computing" that human brains do, but also spanning analogies between capacities for decisions, rationality and control or self-regulation. Arguing that metaphors, analogies and images, far from being mere poetic decoration, run deep in reasoning and in how the world thus becomes organised, one of the animating agendas of critical studies of technology has been a search for "a deeper, broader, and more open scientifc literacy" (Haraway 1997: 11), for a more metaphorically aware "critical technical practice" (Agre 1997), and for spaces to intervene in the cultural imaginaries of technoscience (Suchman 2007: 227).

M. Spencer (\*)

Centre for Interdisciplinary Methodologies, University of Warwick, Coventry, UK

e-mail: M.Spencer.1@Warwick.ac.uk

Te well-worn opposition of the "real" versus the "merely metaphorical" beckons, however. As does that of "fgural" and the "literal." For Agre, Haraway and Suchman, these oppositions did important work in challenging realisms, opening out technical culture to critique, and in Haraway's case, examining the terms in which Western culture imagines the real in the frst place. In this chapter, however, I want to explore the value of another contrast. If we understand "fgurations" in Haraway's words, as "performative images that can be inhabited" (Haraway 1997: 11), we might contrast fgurations with instrumental metaphors, the latter less a matter of habitation, and more one of associations available for use.

Writing from a feld almost entirely disconnected with that of Haraway, Daniel Dennett argued that the apprehension of something (be it person, animal or machine) as an agent with intentions and reasons of its own is itself a fnely tuned and adaptive trick, this "intentional stance" an acquired technique for living a life in common with others (Dennett 1989). Tis provides one reading of what Haraway refers to as the "tropic quality of all material-semiotic processes" (Haraway 1997: 11). Seeing others is always a "seeing as," then, and more direct than an intellectual equivalence; in Kukla's reading at least, a *stance* is to be understood practically, as "a way of readying your body for action and worldly engagement" (Kukla 2017: 4). Figurations understood as stances rather than associations, would be just such "ways of readying," constitutive of relations to others.

Dennett also described a "design stance," sitting somewhere between the intentional and the physical, a mode of interpretation that sees something as embodying a purpose, a normativity: that which it is there for, an embodied intention, its history coupled up to its behaviour according to what it is *supposed* to do (Millikan 2000). Crucially, for Dennett, these stances *are themselves designed*, indices however not of a master creator, but of the long and blind process of evolution. Te result: the installation of pragmatic instincts that enable animals like us *to both cope in and create* a world of social and technical complexity. Such entanglement of fgurations in species history further distinguishes them from the metaphors we use. Te emergence of tool use and complex sociality are part of the same story as the emergence of the forms of worldly engagement according to which some beings are approached as intentional or purposeful, setting others in relief as things.

#### **6 Engines, Puppets, Promises: The Figurations of Confguration…**

My interest in this chapter concerns the circumstances in which what counts as an intentional agent are reconfgured, the ongoing and unfnished history of technology-human entanglement. I look to a domain largely invisible to critical studies of technology: that of confguration management. In the 1990s, growing complexity of computing environments led to the relations between confguration managers and the systems under their stewardship being called into question. I examine "Promise Teory," a philosophy emerging from confguration management which treats computers as intentional agents. Promise Teory has been ignored by the social sciences, but its circumstances of origin are very familiar: the heterogeneous infrastructures of scientifc computing of the early 1990s, circumstances that also inspired Susan Leigh Star's widely infuential relational approach to infrastructure. Te common origins are revealing, as are common concerns with locality and distributedness. In reformulating relations with machines, metaphors of course abound, of puppets, engines, immune systems, orchestras. But we can also, I suggest, detect fgurational shifts, re(con)fgurations, shifts in forms of worldly engagement such that "things making promises" is more than a manner of speaking, a form of stewardship of distributed systems.

#### **Figuring Confguration Management**

Confguration management originated as a set of techniques developed in support of systems engineering, pioneered in the 1950s by NASA, in order to keep track, make manageable and make auditable the vast swathes of stipulations that accompany complex technical systems, asserting how they should be set up, the states their component parts ought to be in, dependencies that should be present, the versions that ought to be used, settings that should be set, switches ficked and plugs plugged (Watts 2011: 10). A vast exercise in paperwork (nowadays usually virtualised in confguration management databases) which is essential to the smooth running of innumerable infrastructures and platforms, yet which almost never surfaces into wider awareness. An infrastructure's infrastructure, if there were such a thing.

In the 1990s, confguration management in information technology made a subtle but signifcant departure from this broader tradition, associated with the transformative efects of automation. A new kind of confguration management tool appeared for managing the confguration of networked computer systems, which would systematically check whether systems under its stewardship conform to "policy" (the ofcially recorded confguration) and take remedial action to fx discrepancies. With automated apparatuses serving as their eyes and hands, systems administrators became the designers and operators of sophisticated automation infrastructures.

Te frst widely used automated confguration management tool was CFEngine. It was developed in the early 1990s by Mark Burgess, a theoretical physicist working at the University of Oslo. He released the software open source in 1993, and by the late 1990s it had become by far the most widely used tool of its kind. Over the next two decades CFEngine went on to serve as archetype for a class of tools that would redefne the nature of confguration management in information technology. Tese would later include Puppet (released in 2005), Chef (released in 2009) and Ansible (released in 2012). Te early success of CFEngine drew Burgess away from physics and into the world of systems administration, and he continued to develop the software alongside a theory of confguration management in the years that followed.

As a postdoctoral scientist, one of Burgess' duties had been the administration of his research group's network of computers. Tis was a classic situation of, to use the unwieldy catchphrase of the time, heterogeneous distributed computing: the kinds of machine used in research had a wide range of confgurations, versions of software, permissions associated with user accounts, fle system structures, scheduling of batch processing, and so on, often diferent models of device, typically running variants of UNIX.

Facing the challenge of managing such heterogeneity, Burgess looked to apply automation to the tasks of confguration management. Te standard approach of the time involved curating custom procedural scripts. Te activities a system administrator would otherwise carry out manually would be written as step-by-step algorithms that could be executed from a privileged centre of control. Tis "imperative approach to thinking" (Burgess 2015b: 2) turned out to be fragile. Te diversity of computing environments made these scripts complex in their own right, unwieldy to maintain and "brittle," tending to produce unpredictable results, especially when run against a machine in an unknown state (Spencer 2015). Luke Kanies, who would later author Puppet, dubbed the challenge of maintaining assumptions a problem of "software rot" (Kanies 2003: 119). Burgess, on the other hand, interpreted the problem as a physical one, the idea that in an unpredictable world, commanded systems will tend to diverge from a known starting point (Burgess 2015b: 4).

So CFEngine moved away from procedural scripting. It was based instead on a declarative approach. Te desired confguration is stated in a special syntax, without stipulating what to do, how to check, enforce, or make a repair in relation to it. A set of CFEngine's specialised "software agents" would then interpret and compare this "policy" against the observed state of various computers and generate contextually specifc steps for remedial action if necessary. Te activities of these autonomous agents were not intended to fx problems as a one-of complete repair. Rather, they were intended to run in the background in a decentralised fashion, producing over time a *convergence* between the actual and the proper state of afairs.

Tis combination of declarative policy, decentralised automation and convergent repair became the paradigm for IT confguration management and with huge impact beyond. Te automation of confguration is, for instance, the heart of cloud computing. Over the last two and a half decades, the paradigm expanded to provide comprehensive infrastructure automation, so much so that confguration management tooling exceeded its original purpose. Beyond just checking and repairing systems that already existed, tools like CFEngine also provide the means to spin up new infrastructure on demand, declaring it into existence, as it were. "With CFEngine," an Automation Engineer at LinkedIn is quoted as saying, "I can defne a new Software Defned Datacenter and ofer IAAS [infrastructure as a service] and PAAS [platform as a service] to my customers within 10 minutes" (CFEngine n.d.: 3). Te infrastructure of infrastructure indeed!

Te automation pioneered by the confguration managers paved the way for wider suites of tooling, which brought automation to code management, release pipelines, build processes, testing cycles and deployment. Together these tools became the technical foundation for agile and continuous delivery-based methodologies, which in turn transformed the manner in which functionality is delivered in digital environments, taking us from the "old approach" of delivering new versions of applications after long periods of stasis, to the current paradigm involving constant fows of small iterative changes (Humble and Farley's textbook *Continuous Delivery* being the enduring reference point for the new technical foundation; 2010). When Nef and Stark described the agile approach as "permanently beta" back in 2004, they saw it as an approach to design (2004). With the rise of automation and cloud infrastructure in the years that followed, however, we can now say it has expanded signifcantly in reach, an approach to the whole delivery lifecycle: not just design but also delivery and the operation and management of the systems that result.

It is testimony to the depth at which confguration management is embedded that it has escaped attention of critical studies of technology even in cases where they address cloud computing square on. Peters, for example, takes a user-centred view, associating cloud computing with cloud storage (Peters 2015). Hu and Bratton go in the opposite direction, with their focus on the evolution of physical infrastructures, a broad story of communication networks and datacentres, without addressing the question, missing in the middle, of the techniques by which it has become possible to tame that complexity (Hu 2015; Bratton 2016). Amoore similarly examines analytics and algorithms in light of material infrastructure, but leaves little space for understanding the "how" that makes these techniques possible (Amoore 2020). None of these thinkers appreciates the problematic of machinic autonomy that underlay the ability to craft self-regulating, automatically provisioning systems for computational infrastructures, and which is now intimately woven with digital culture.

# **Smart Intentional Infrastructure**

Automating confguration management was not simply a matter of fnding clever ways to script manual tasks. It required and fostered reinterpretation of the problem of confguration itself, which became a topic of debate and discussion among an emerging community of IT confguration managers, on mailing lists and at conferences. For Burgess, this intellectual project led towards the development of a theory of cooperation he called "Promise Teory." Promise Teory arose out of Burgess' attempts to formulate what it was he had been trying to do with CFEngine and became over the years a lot more than a theory of confguration.

Moving from the context of theoretical physics into the professional community of system administration, Burgess reports that he encountered a set of intuitions about computers that seemed to be aligned with the procedural scripting approach to confguration management. Tis idea, that computers were like obedient rule followers, rubbed awkwardly against his more physical, more cybernetically infected intuitions. Writing about his experience at the conference in Large Installation System Administration (LISA) in 1997, Burgess relates that

To me, the work I presented was just a small detail in a larger and more exciting discussion to make computer systems self-governing, as if they were as ordinary a part of our infrastructure as the self-regulating ventilation systems. Te trouble was, no one was having this discussion … In the world of computers, people still believed that you simply tell computers what to do, and, because they are just machines, they must obey. (Burgess 2015a: 4)

Tough frst by its name an "engine," when he returned for the following year's LISA, Burgess had armed himself with a new metaphor to cut through these preconceptions. His talk, published as "Computer Immunology," went on to be infuential in the feld. It used the metaphor of an organism and its immune system to perturb system administration thinking away from its familiar notion of the obedient computer. "CFEngine," he wrote,

fulflls two roles in the scheme of automation. On the one hand it is an immediate tool for building expert systems to deal with large scale confguration, steered and controlled by humans. It simplifes a very immediate problem, namely how to fx the confguration of large numbers of systems on a heterogeneous network with an arbitrary amount of variety in the confguration. On the other hand, cfengine is also a signifcant component in the proposed immunity scheme. It is a phagocyte which can perform garbage collection; it is a drone which can repair damage and build systematic structures. (1998: 287).

Te proper behaviour of cells which sustains the life of an organism is not enforced by their strict obedience to commands. While almost all cells do have a catalogue of instructions, some may have faulty DNA, something might go wrong in the reading, or a foreign agent may be interfering with the normal processes of interpretation. Te immune system is comprised of mechanisms capable of detecting and responding to aberrant behaviour, keeping things healthy at a higher level of organisation. Similarly, CFEngine is built on scepticism that commands can be sufcient to ensure convergence: machines may have missed instructions, might end up with multiple conficting instructions or might be missing dependencies for carrying them out. Instead, it implemented a set of autonomous agents capable of identifying problems, making remedial changes or bringing issues to the attention of the administrator.

Later refecting on this period of time, Burgess notes that he soon abandoned the immune metaphor in favour of a theory of promises based on the fguration of computers as intentional agents.

Te idea gelled in April of 2004 that autonomously specifed declarations of intent were simply promises--something conceptually opposite to obligations, or any other kind of declarative or imperative logic. Promises could be defned as a network that was not necessarily the physical network between computers, more like a network of self-imposed constraints that we call intentions … Emerging was a theoretical model for a kind of smart, intentional infrastructure based on graphs of autonomously made promises. Tis graph theoretical approach was an altogether more plausible and scalable approach to locality than deontic logic. (Burgess 2015a: 247)

Burgess maintained the contrast with imperative thinking within the theory of promises: "obligations" can be understood as a special case, as promises made for others, fragile in comparison with promises one makes for one's self. Tis lexicon was implemented in CFEngine's third version: policy was to be understood and encoded in terms of the promises that machines or systems make (to one another, and to users). And in addition to developing CFEngine, Burgess also built out this theory, developing it into a graph theoretical framework for the analysis and design of intentional relationships in distributed systems, in collaboration with Jan Bergstra, a Dutch computer scientist.

#### **6 Engines, Puppets, Promises: The Figurations of Confguration…**

To translate into more familiar sociological terms, promises formalise the normativities of technical systems: what it is that they are *supposed to do*. But instead of using the design stance, Promise Teory construes technical normativities via the intentional stance. Tey are treated as a thing's intentions rather than purposes that can be read into it. Tis shift between stances is strategic: the problem with addressing technical normativity via the design stance is that it is too easy to regard the purposes of designed things as residing in the mind of their creator, or else in some separate source of authority, such as the confguration management database. By locating the source of normativity outside of the technical thing, the problem of confguration management seems to be, indeed, one of imposing the correct behaviour on subservient infrastructure via obligations. Te language Burgess and Bergstra deploy, however, in formulating things' "autonomously made promises" afords things a depth of their own, enabling us to see their purposes as their own, as local to them. Tis localism is central. Promise Teory, they write, "is a relativistic theory of 'many worlds' belonging to its many agents" (Bergstra and Burgess 2019: 4). Where small and simple networks of computers could be controlled with impositions, the large-scale complex networks of contemporary information technology ought to be treated, in design and in maintenance, as "smart intentional infrastructure."

Te shift to the intentional stance of course raises questions of whether things "really" have intentions. Is this just a "fgure of speech"?

Perhaps this makes you think of promising something to a friend, but don't forget to think about the more mundane promises we take for granted: a cofee shop serves cofee (not bleach or poison). Te post ofce will deliver your packages. Te foor beneath you will support you, even on the twentythird level. If you could not rely on these things, life would be very hard. (Burgess 2015b: 39)

We are to think of promises, then, as environmental as well as explicitly designed (cofee is not designed as such), as embedded into surroundings by the particularities of cultural and technical histories. Te language of Gibson's ecological psychology would not be out of place here, for promises are relational to the kinds of bodies and purposes that come into articulation with them, and that may foster them over time (in the terminology of the theory, promises are made to specifc promisees, not to the world in general). Te post ofce's promise to deliver is addressed to *some* kinds of beings in its vicinity; for others, its promises may be quite diferent (its guttering promising perhaps a place to roost).

In certain places, however, the concept can seem rather thin: "an *intention* is nothing more than the selection of a *possible outcome* from a number of alternatives, based on an optimization of some criterion for success" (Bergstra and Burgess 2019: 7). Tey note that promises are intended to capture the sense in which inanimate objects "serve as proxies for human intent" (Bergstra and Burgess 2019: 14; also Burgess 2015b: 9). Are they in danger here of falling back into imposed intent, against their own notion of locality? If a promisor is a proxy, then we might indeed question whether there it has much real autonomy.

Bergstra and Burgess do fall back upon a "default" apportionment of intention as naturally human. We might, however, read this as a sign of context, an anticipation of aspersions of animism, the interpretive efects of being seen to confuse categories that have held fast in the West for millennia. Tere is of course a vigorous literature on the cross-cultural nature of apportionments of agency and perspectives between humans and nonhumans (notably, Viveiros de Castro 2012; Descola 2013): not, however, a literature with which their readership is likely to be familiar.

Instead of interpreting this *being a proxy* as implying that intent is derived from some particular humans that can be located and pointed to, we might more generously interpret it as referring to the embedding of technical things in the world, the historical sedimentation of selections that forms worlds. It is not as a physical object that a technical thing has its intent, but rather as an historical object. In cases where a designer can be pointed to, there is of course a particular human agent involved in this history, but even here, many elements combine which have unwritten and tacit pasts of imitation, inspiration and copying, as well as the historicity of cognitive capacities involved in the designing (Millikan 1984). A humble webserver would then bring together many threads of intent: that of its designer, the architect of its implementation, but also the history of imagining computers as network endpoints, the concept of service and the history of servitude.

Te appeal to a depth of intent in formulating the problem of confguration management is not, I suggest, a simple rhetorical move, making use of metaphor as a route to a better explanation. It draws on the deep history of relating to other beings as having intentions and purposes, the history of the fgurations we inhabit, of sharing space and co-habiting a world alongside others with intentions of their own. It is signifcant, I would suggest, that Promise Teory emerged from the relationship of confguration managers with the increasingly complex infrastructures under their care, rather than as an intellectual project driven by an abstract problem. Where confguration management had previously entailed a relationship of imposed order, the complex distributed systems of the 1990s (and since) entailed a subtle shift, in which those systems' contingencies, the fact that they may always be in a state other than the proper one, came to have new signifcance: no longer an invitation for a corrective intervention, they required stewardship and care (see also Kocksch et al. 2018). Te historicity of relations with technology is thus doubly entwined with the historicity of technical function (Spencer 2021): frstly in the fgurations of technical systems as purposeful and intentional, and secondly in the novel possibilities these shifts open up.

Like CFEngine, Promise Teory participated in the development of infrastructures more widely, most obviously where it is directly cited as inspiration or support. Te approach, for instance, is named as the basis for managing policy in the networking giant Cisco's "intelligent networks" (Cisco 2014). A second example comes from nearer to home. Adam Jacob, who had originally developed the "Chef" confguration management system, described his thought process in devising plans for a new application automation system he named "Habitat": "I think there is an application problem. I think we are thinking wrong about the shape of the application. And what if … applications could behave as wellbehaved actors in like a promise theory sense? And what would be the promises that those applications make to each other? And from there it led to Habitat" (Jacob 2018, np). Treating applications as actors rather than as software (which begs the question of whether they are running properly), Habitat provides an infrastructure for the mutual monitoring of applications and their ability to propagate information about each other through "gossip." Te intents embodied in these infrastructures have many sources, but it is not a stretch to include among their number a "promise theoretical" refguring of machinic intent.

## **Figuring Infrastructure**

While Promise Teory (and confguration management in general) may have escaped attention among critical scholars of technology, the circumstances in which automated confguration management appeared are rather familiar. In addition to being the site of emergence of CFEngine, scientifc computing environments of the late 1980s and early 1990s had a formative infuence on the development of social approaches to the analysis of infrastructure.

It is a truism of science studies that scientifc practice is not a singular phenomenon, but consists of a plurality of epistemic cultures (for instance, Galison 1996; Knorr-Cetina 1999). On a more mundane level, as Burgess puts it, in the sciences "every kid is special": because research agendas can point in their own unique directions, demands for specialist equipment and unique confgurations can readily override institutional pressures to adopt standardised technology. With tendencies towards computational heterogeneity built in, it is no surprise that CFEngine and modern IT confguration management emerged from a university research context rather than, for instance, the IT departments of commercial organisations, or Silicon Valley.

Another infrastructure very much of this moment was the Worm Community System (WCS). WCS was an information sharing tool, funded in the US by the National Science Foundation, and designed for the global research community studying C. Elegans, a nematode worm widely used as a model organism by molecular biologists. On the WCS team were two ethnographers, Susan Leigh Star and Karen Ruhleder, whose participation in and analysis of the process of implementation proved profoundly infuential in the development of sociocultural research into infrastructure, maintenance, computer-supported cooperative work, standardisation and communication (Star and Ruhleder 1996). What they called, with the gerund-ifed fourish of grounded theory, "infrastructur*ing,*" named that situated and practical process of *becoming infrastructure* which was both goal and problem for WCS.

Making sense of the WCS project, Star and Ruhleder argued, required a *relational* approach to infrastructure. In a departure from emphases on wide arcs of technological development, and the stabilisation of designs (e.g. Bijker et al. 1987), Star and Ruhleder foregrounded the embeddedness of infrastructure, its intrinsic relationality with and in sites of practice. Technologies become infrastructures as a contextual achievement, in which their use becomes integrated into routines of practical activity, so much that they are no longer explicitly put to use, becoming "sunk into" the background of practice. Because of this relationality, established infrastructure resurfaces once more in special moments, those of "infrastructural inversion," either through the methods of a social scientist or as a result of a fault or breakdown which disturbs practice (Star 1999).

Te WCS aspired to infrastructure. It aspired to become the basis for collaboration across a wide community of researchers. But becoming infrastructure is not straightforward and certainly hard to impose. Despite the best eforts of the team, WCS ended up being little used (Star 1999: 380). Te problems it faced could not be traced to a single root cause; the ethnographers encountered diverse resistances cropping up across diverse sites. To make sense of these challenges, Star and Ruhleder turned to the cybernetician Gregory Bateson's theory of communication, arguing that contextuality itself had frequently become the source of problems for the project. For instance, technical instructions that were seen as simple information about what to do by their originators, were for some recipients complex signs that diferentiated kinds of persons: those for whom the instructions appeared straightforward, and those for whom they were anything but.

As with Bateson's levels of communication or learning, the issues become less straightforward as contexts change. Tis is not an idealization process (i.e., they are not less material and more "mental"), nor even essentially one of scope (some widespread issues may be frst order), but rather questions of context. Level one statements appear in our study: "Unix may be used to run WCS." Tese statements are of a diferent character than a level two statement such as "A system developer may say Unix can be used here, but they don't understand our support situation." At the third level, the context widens to include theories of technical culture: "Unix users are evil—we are Mac people." As these levels appear in developer-user communication, the nature of the gulfs between levels is important. (Star 1996: 117)

Successful infrastructures bridge local contexts. Te process of *becoming infrastructure* is thus liable to act as an irritant for all kinds of contextual particularities. Te connection I have in mind with confguration management, with Burgess and Promise Teory, is not about how they imagine infrastructure, but rather, this: the way in which, in circumstances of confguring complex distributed IT systems, locality becomes ontologically foregrounded.

To make the resonance clearer, the connection might be made with Star's earlier and similarly infuential research into cooperation. In her 1989 account of Berkeley's Museum of Vertebrate Zoology, written with James Griesemer, she argued that consensus is not required for cooperation or for the emergence of common understandings. Cooperation, in short, must emerge *across* boundaries of intelligibility, rather than breaking those boundaries down. Tey argued, indeed, that "*all* science requires intersectional work" (Star and Griesemer 1989:392 emphasis added). Teir concept of "boundary objects" denoted those interactive forms that mediate and coordinate across divides of intelligibility. Tese included *repositories* such as libraries or museums, *ideal types* that "delete local contingencies from the common object" (Star 2015, p254) such as representations found in scientifc atlases, *terrain with coincident boundaries* such as the common referents on maps (while the professional biologists and amateur collectors Star and Griesemer studied produced very diferent kinds of maps, the referents they had in common facilitated their collaboration), and administrative *forms and labels* which use semiotic constraints to standardise information (Star and Griesemer 1989).

Star later complained that in its scholarly reception, the concept of boundary objects was most frmly associated with interpretive fexibility (Star 2010). Any such object would indeed need to be meaningful across diferent local contexts. But she was also concerned with the composition of open systems, something that received less attention. In an early formulation presented to an audience of artifcial intelligence researchers, Star suggested that boundary objects are "simultaneously metaphor, model, and high level requirement for a distributed artifcial intelligence system" (Star 2015 [1988]: 249). Such a system would need to be characterised by processes that mediate between their local particularities and higher-order coherence. For boundary objects, that meant a "tack[ing] back and forth" between the kind of vagueness which renders things capable of being held in common between heterogeneous viewpoints and their local manifestations specifc to just one (Star 2010: 604-605).

It would not be hard to read Promise Teory in these terms. Unlike deterministically imagined obedient computer networks, the coherence of smart intentional infrastructure emerges from just this kind of "tacking," sailing against a prevailing wind (whether fgured as "rot," "drift" or "divergence") by means of a series of zig-zagging trajectories. In common is the proper policy, the ofcial record; in contrast its local manifestation is the contingent promise of the thing itself, which always of course may be otherwise than what it is supposed to be. Te functional coherence of the whole is an achievement neither of the ideal or the contingent, but of the processes by which they are brought into interaction.

Star and Burgess, in other words, developed theories and pragmatics for building systems of distributed coordination. In both cases the promise of understanding emergent coordination across many, locally heterogeneous, nodes, required suppressing the intuition that this would be done, or explained by, the imposition of a single master ontology across the network, and in both cases the authors were engaged in projects for confguring systems, rather than acting as disengaged observers.

## **Mythologies, or Why Figure Confgurations?**

Promise Teory stands out among technical philosophy for its refusal to give priority to abstract, formal understandings of computation, typifed for instance in the mathematical theory of algorithms, and instead emphasises the empirical materialisation of computing infrastructure, in specifc networks and functional distributions, entailing contingent and localised embodiments of intent. Te lack of a privileged centre was likewise a starting point for Star. She cites David and Smith: "When control is decentralized, no one node has a global view of all activities in the system; each node has a local view that includes information about only a subset of the tasks" (quoted in Star 2015 [1988], p246). A distributed system is brought together by means of boundary objects, not by obedience to a master command.

Te tension between a formal "algorithmic" interpretation of computing and a more materially grounded approach is familiar and lively, perhaps never more so than in recent debates in critical studies of technology. For Ian Bogost, "the algorithm has taken on a particularly mythical role in our technology-obsessed era, one that has allowed it to wear the garb of divinity" (Bogost 2015: np). "In its ideological, mythic incarnation," he argues, "the ideal algorithm is thought to be some fawless little trife of lithe computer code, processing data into tapestry like a robotic silkworm. A perfect fower, elegant and pristine, simple and singular" (Bogost 2015: np). In the lexicon of media theory, as well as in general parlance, the concept of "the algorithm" has become a handle with which to grasp the implications of computing in society. Divine, we might surmise, because such an abstract understanding leaves little room for appreciating the locality of intent. A god has surely little need for boundary objects.

Allowing the formal abstractions of algorithms to stand as synecdoche for the material complexities of computational infrastructures is irresponsible (a point also argued by Chun 2011; Dourish 2016: 2). "Concepts like 'algorithm' have become sloppy shorthands, slang terms for the act of mistaking multipart complex systems for simple, singular ones" (Bogost 2015: np). Much the same sentiment is voiced by Burgess, who opens his book *In Search of Certainty* with the proclamation that "[t]he myth of the machine, that does exactly what we tell it, has come to an end" (Burgess 2015a: 1).

Bogost, in an echo of Marx's analysis of "commodity fetishism," suggests that the divine algorithm is falsely animated by a trick of the eye, by which we overlook the material conditions of production, the real work that goes in to creating these efects. Burgess suggests that our problem is our lack of a sufcient vocabulary to address the empirical locality associated with contingent technical systems. Te mobile associations produced in the wake of the automation of confguration management do both jobs: the "engine," the "phagocytes," and the "promises" evoke the missing agency of technical systems, while the expansion of automation tools populates the world of IT confguration management with metaphors that give fresh form to the subject positions of the otherwise overlooked maintainers and repairers, whose hidden work, in the background, enabled computational systems to look like "robotic silkworms" in the frst place. "Puppet" dressed up the system administrator as the puppeteer in control, and "Chef" as one engaged in a fnessed art of high esteem, both in stark contrast to the beleaguered service personnel in stereotypically subterranean ofces, grappling with unwieldy infrastructures, mundane problems and an onslaught of helpdesk requests.

Metaphors stick and slip. Just as a controlling puppeteer hardly captures the autonomy that Promise Teory attributes to technical things, so too do well-worn metaphors come to stand for the opposite of their original intention. "Orchestration," for instance, was one of the earliest metaphors for the automated management of networks. It suggests many autonomous parts moving in concert, but like the puppeteer it also implies a centre of control. Commentators were already using this connotation to tease out signifcant diferences in the early 2000s. "Orchestration always represents control from one party's perspective. Tis difers from choreography, which is more collaborative and allows each involved party to describe its part in the interaction" (Peltz 2003, p46). Kubernetes, probably today's most well-known distributed computing system, is widely referred to as an *orchestration* system, yet refects on the ambivalence of the metaphor within its own documentation: "Te technical defnition of orchestration is execution of a defned workfow: frst do A, then B, then C. In contrast, Kubernetes comprises a set of independent, composable control processes that continuously drive the current state towards the provided desired state" (Kubernetes n.d.).

Metaphors of puppets and orchestras may appear blunt instruments to humanistic critics accustomed to searching for subtler layers of signifcance. But as fguration, as re(con)fguration, the working over of even these blunt instruments is an index of a question: what is (a)kin to the systems we are building and looking after? If fgurations are inhabited, in the sense of being stances, ways of readying oneself in relation to another, attending to fgurations might attune us to the ways that the evolving nature of technical stewardship has re(con)fgured relations, and not just the words we use to talk about them. Te stance we take towards distributed intentionality, towards our "smart intentional infrastructure," is not just how we choose to represent it, but the kind of readiness entailed in relations to the local and contingent in a complex system.

Te work of fguration nevertheless stirs up in its wake a detritus of metaphorical imprecision, and with it an enduring salience of "myth," a treasured metaphor for metaphors gone stale: mistaken, naïve, of their moment. For what we do with "myth" is exactly what Star elicited through Bateson. By referencing Western stereotypes of "primitive" thought, myth takes our communication "up" a level or two. It draws attention from the content to the context of the "mythical" belief, as some Other's belief, which sorts out kinds of persons, those provincial people, from elsewhere, or back then, who would take it at ground level. Te kind of person who would be taken in, and who they are like. Applied across contexts, it dredges up infrastructures both entrenched and would-be: the perfect fower algorithms and the obedient computer.

Te urgency of debunking myths in the critical studies of technology is a legacy of its obsession with the politics of representation, with the metaphors in and of technology, with "whose metaphor brings worlds together, and holds them there" (Star 1990: 52). If a newer wave of scholarship might be identifed, around what Amoore calls a "*cloud ethics* … concerned with the political formation of relations to oneself and to others" (Amoore 2020: 7) "sustained by conditions of partiality and opacity" (Amoore 2020: 8), would it be a surprise if it were precisely practical relations with heterogeneous distributed computing systems that prompted the most radical re(con)fgurations of the technical beings we live among and through?

**Acknowledgements** Many thanks to Scott Wark, Celia Lury and Will Viney for their comments and feedback and for their eforts in bringing this collection together. Tis research was supported by the UKRI Future Leaders Fellowship Scaling Trust: An Anthropology of Cyber Security (MR/S037373/1). No new data were created in this study.

# **References**


Stefan Timmermans, Adele Clarke and Ellen Balka 243-259. Cambridge: MIT Press.


**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/ licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **7**

# **Figuring Molecular Relapse in Breast Cancer Medicine**

**William Viney and Sophie Day**

# **Introduction**

Medical practice and research in oncology increasingly involve and respond to highly various and heterogeneous disease classifcations, which evolve with time and in response to emerging programmes of treatment and research. Tese refect novel practices that produce diferent forms of risk prediction and risk analysis. Supported and facilitated by translational research programmes, trials networks, and interoperable

W. Viney (\*) Department of Anthropology, Goldsmiths, University of London, London, UK e-mail: w.viney@gold.ac.uk

S. Day Department of Anthropology, Goldsmiths, University of London, London, UK

Patient Experience Research Centre, Imperial College London, London, UK e-mail: s.day@gold.ac.uk

data sharing platforms maintained by diverse collaborative groupings, there has been a proliferation of breast cancer subtypes and treatments (Day et al. 2016; Cambrosio et al. 2018; Bourret et al. 2021; Day et al. 2021). During an ethnographic study of personalised breast cancer medicine and healthcare in a London hospital (2018-2021), members of a research group of which we are members learned that research in tumour biology is highly complex (Day et al. 2021; Day, Smith and Ward, chapter 8, this volume; Viney et al. 2022). And while treatment for primary disease is increasingly efective compared to other tumour types, especially when treated at an early stage (Cancer Research UK 2021), the return of cancerous cells after the treatment of a primary tumour known as a 'recurrence' or 'relapse'—can be difcult to predict for individual cases. It is cancer's uncertain return, its temporality, and the fgures its times generate that is the subject of this chapter.

Cancer cells have been observed in human blood since at least the later nineteenth century (Ashworth 1869). Fragments of cell-free nucleic acids in human blood were frst described in the 1940s (Mandel and Metais 1948), and cfDNA was identifed and associated with cancer in the mid to late 1970s (Leon et al. 1977). Te serial measurement of these and other biomarkers has long provided hope: "sequential measurements of DNA concentration may be a useful tool for monitoring the efects of therapy" (Leon et al. 1977: 650). Te promise contained in this 'may' is just one temporal efect of liquid biopsies, which have been made to occupy the future anterior tense,1 which we argue anticipates fgures and generates fgures to be retrospected (Brown and Michael 2003). Considering these fgures allows a greater appreciation of how cancer research places and transfgures disease, placing it in time, and making it 'historical'. Te study we focus on here fgures the relapse of disease for patients treated for breast cancer and who are classifed as being at high risk of developing metastatic recurrence. Our engagement with this study occurred when the scientifc research group published early fndings, and, fortuitously, we were able to observe how scientifc fndings were being

<sup>1</sup> Some liquid biopsy studies now track cell-free and circulating tumour DNA in time-based practices: 'early' and 'earlier' and 'just in time'—so resonating with precision practices in other domains. As a consequence, liquid biopsies represent enduring promissory fgures—'soon' and 'not yet' that herald improved patient outcomes.

made public. We noted how concepts of disease and the people associated with disease were being fgured by these fgures. We wanted to learn how people and groups of people are incorporated and at times excluded by numbers, images, and texts. If fguration is a methodological practice, this chapter reports on fgures as objects of discovery and description and, concurrently and in combination, as epistemic and symbolic matters of concern.

During our research, in formal and informal interactions with clinical researchers, we learned that, at molecular levels now explored via contemporary diagnostic and treatment technologies, tumour biology does not usually conform to easily predicted norms and averages. As one oncologist explained as we began our interviews, "we're in this era where it's expanding the varieties and number and types of cancer due to the molecular characterisation of each patient's cancer ... virtually every patient has a diferent type of cancer due to the genetic changes that occur in the cancer compared with normal tissue." Teir colleague in oncology summarised: "there are no more averages. I can quote averages from clinical trials, but we don't know."

One promise of a more 'personalised' or 'precision' medicine is that recommendations will provide 'the right treatment, to the right person, at the right time' (Keogh 2015; Scoltz 2015). Despite this promise, tools to enhance precision and prediction in oncology move by contingent increments, where defnitions of 'right' change gradually. Diferent scales and speeds of implementation and impact can be narrated from diferent situated perspectives. Narrating these diverse perspectives on 'targeted' approaches to oncology is not easy with the views of patients and staf varying widely. One claim of this chapter is that understanding developments in translational medicine might be aided by documenting and discussing how fgures are made and maintained by diferent actors involved in this work of targeting. Patients, clinicians, biomedical, and other kinds of researchers make and interact with these fgures, and they comment, evaluate, and form expectations about fgures. In this chapter, fgures are an empirical proof and promise in ways that are at once confrmatory and confounding. In precision oncology as in other felds, fgures are contested and can attract conficting values based on diferent interpretations of their performance.

# **Background to 'An Exploratory Breast Lead Interval Study' (EBLIS)**

One approach gaining popularity among biomedical researchers in oncology involves monitoring the presence of tiny fragments of cell-free DNA (cfDNA), some of which can be associated with tumour cells known as circulating tumour DNA (ctDNA). Te practice of detecting and analysing ctDNA in fuids for diagnostic purposes is called 'liquid biopsy'. Liquid biopsies are created to detect, measure, and analyse disease in a minimal, residual state. Researchers use genomic and other omic sequencing techniques to identify these molecular signs of disease and, in oncology, disease recurrence. While temporal horizons of risk are currently formed for individual patients based on data using population averages or norms that contribute to the formation of groups and subgroups of patients, research studies tracking ctDNA adjust the practice, scale, and the horizon of relapse using molecular data from individuals. Tis approach detects and tracks micrometastatic cells in ways that are molecularly specifc to individual patients, illuminating *when* disease relapses by reducing the question of *who* is afected to a single person, and opening new opportunities for learning *why* cancer returns.

As part of our work in the hospital service, our research group followed the progress of one observational study tracking ctDNA in a group of breast cancer patients: EBLIS, an Exploratory Breast Lead Interval Study. We were told by a clinical researcher involved in EBLIS that the study represented the only example of personalised breast cancer medicine in this London breast cancer service. Previous work by members of our research group have documented changes to how clinical practice and research is conducted in this service (McGrath-Lone et al. 2015; Day et al. 2016; Day et al. 2021; Viney et al. 2022). As a translational research study that follows a cohort of patients whose primary treatment had concluded, the status of EBLIS as a unique example of personalised medicine intrigued us. Motivated to learn more about this work, we followed study participants, clinicians, researchers, administrative as well as patient samples, and data over time and across diferent sites, following a process of data making and circulation, analysis, and communication.

Our observational and interview research followed EBLIS towards the end of its initial four years of activity. In total, 194 breast cancer patients had been recruited from 3 UK clinical sites (1 October 2013—8 July 2016). Out of which, 188 were tracked for the frst 4 years of the study. While this group of patients constituted a group or cohort they were not recruited based on shared biological or therapeutic characteristics. And patient-participants were varied in terms of age at diagnosis, histology, hormone receptor status, and treatment. But all patient-participants shared in risk categories—numeric values that related them as a group to a near but imperfectly known future. Tey were at high risk of clinical relapse based on a digital risk prediction tool called Adjuvant! Online (see de Glas et al. 2014; Lambertini et al. 2016). Here 'high risk' was defned in terms of risk of mortality equal or greater than 50% at 10 years without therapy, or corresponding to a relapse rate of 65% at 10 years without treatment. While EBLIS generated novel numbers for clinical researchers, this was possible because it was embedded in existing practices of clinical fguration. Tis wider scheme of fgure-making between existing, emerging, novel and established material occurs at many scales, and the dynamic temporalities of making ctDNA a clinically actionable biomarker will become clearer by describing the circulation and combination of diferent fgures of disease.

At the beginning of their care in the hospital, patient-participants provided a sample of their tumour. Tis sample was used by researchers to specify the molecular profle of their primary disease, using a range of next generation—omics sequencing technologies. Researchers leading EBLIS collaborated with a US biotechnology company to create a "bespoke amplicon design pipeline" that nominated PCR primer pairs for a given set of genetic variants. In brief, somatic variants were identifed at a patient level, pairing the primary tumour sample with matched white blood cells using whole-exome sequencing (WES). Each patient's tumour was attributed a signature, composed of 16 highly ranked pieces of genetic code selected to create a "custom patient-specifc panel" of mutations. Subsequently, the clinical research group collected blood samples at six-month intervals and checked this against the patient-specifc panel, to track and measure changes in the volume and characteristics of ctDNA over time. Te study therefore developed a 'personalised' analysis insofar as it tracks disease at molecular, patient-level and patient-specifc conceptions of disease progress. Each patient participant in EBLIS has had their tumour sequenced in a way that is specifc to their tumour characteristics. Tey are then followed on the basis that they are unique, each patient participant providing their own baseline for an evolving sequence of testing and retesting.

Approximately 6 months into our feldwork EBLIS reported its design and results, providing information for the frst 49 patients who had relapsed since consenting to be part of this research study. Appearing in *Clinical Cancer Research*, these results were published in April 2019 (Coombes et al. 2019). EBLIS had set out—among other things—to bring more predictive certainty to recurrence, and this publication uses diagrammatic reasoning to present its experimental fndings. It did so with a set of contrasting fgures, expressed in diagrams, plotting somatic changes of emerging experimental interest against common clinical biomarkers and screening technologies used in the NHS. It visualised a new temporal event or horizon in order to present and illuminate an otherwise obscured object of clinical interest: a threshold number of ctDNA that declares cancer's return. By tracking ctDNA, the group suggest the clinical utility of "molecular relapse", which they separate from "clinical relapse", a term used to denote recurrences discovered through existing screening systems and technologies.

Te fndings summarised and plotted in diagrams suggest an ability to detect a recurrence up to 2 years ahead of existing screening technologies (median=8.9 months; range=0.5-24.0 months). Te diagrams (see Fig. 7.1) are works of fguration that gather, propose, and represent this novel event as a temporal interval by contrasting molecular and clinical relapse, demonstrating the potential usefulness of one fgure in contrast to another. More generally, the study uses, enacts, and contrasts many fgures: its work of fguration has many dimensions and horizons, with these two-dimensional diagrams used to picture cancer's progression central among them.

One fgure is a grouping of 6 panels, where A-E visualise an increase in plasma levels of ctDNA in 5 individual patients, 1 patient per panel, as they were collected over time. Te *y* axis shows variant allele frequency (VAF) of gene mutations that compose each "custom patient-specifc

**Fig. 7.1** "Personalized profling detects rising ctDNA ahead of clinical relapse" diagram published in Coombes et al. (2019)

panel", and these lines plot the number of variant reads divided by the number of total reads and reported as a percentage. Te *x*-axis shows days after surgery. Te lead time—the interval between ctDNA increasing in plasma samples to indicate recurrence and when the patient was diagnosed via the hospital's existing screening system—is shown at the top of each panel by a red triangle and blue triangle connected by a dotted line and expressed in days. Tis is the time interval between clinical relapse (red triangle) and molecular relapse (blue triangle). In contemporary breast cancer care CA 15-3 biomarkers are routinely used to monitor for cancer relapse. Tis biomarker is also graphed here (teal circle), with baseline levels (32 U/mL) marked in light blue.

Tese diagrams are used as visual evidence and proof of a wider set of practices that call on fgures in diferent ways—methods of analysing and valuing fgures in the interests of accurately predicting change. Tey show how EBLIS fgured cancer in time: according to the durational intervals signifcant to tracking ctDNA and to a process of fguration that makes this time known as an experimental object. As we observed the study, we also became aware of the diferent horizons of hope and expectation being tied to the study's progress. As results emerged, patients were able to picture themselves in novel ways. However, while the research was being conducted in clinical contexts, the application of study outcomes were diferently distributed. Interviews revealed the many interpersonal and institutional contexts by which practices of material and symbolic fguration were taking place, as well as the wider, consequential phases of clinical research that diferent stakeholders expect EBLIS to follow.

# **Inhabiting Cancer's Figures**

Diagrams establish and maintain internal and external relations, drawing connections between multiple objects, practices, and persons. Tey contain common elements, according to what Sybille Krämer and Christina Ljunberg call "diagrammatic scenarios"—a synoptically fattened order or form spread across a two-dimensional plane, involving the interplay of points, lines, or surface points that articulate a set of homogenised relations. Tese elements ensure the diagram's reproducibility as a schema. Rather than genres of image that are self-referential, Krämer and Ljunberg ask that we consider the "alloreferential" capacity of diagrams, which signify many and multiple concepts and objects of knowledge external to their form (Krämer and Ljungberg 2016: 10-11).

Te word 'diagram' has this connectivity stowed in its etymological history, deriving from the Greek *diagraphein*—*dia-*, 'across, through', *graphein*, to 'scratch,' 'carve,' 'write'; 'to mark out by lines'. Modern diagrams function as "icons" that, according to Pierce, are used to gather "a set of rationally related objects" amenable to experimental practice and reason (Peirce 1976, 4: 316). And yet diagrammatic icons, for Pierce, do not relate objects passively or by neutral reason but take "the middle part of our reasonings", as mediation (Peirce 1998, 5: 163; Bender and Marrinan 2010: 23-56). EBLIS represents an approach to determining cancer's presence that is fgural, insofar as it involves the presentation of knowledge mediated in diagrams. It invokes visual forms that picture relapse as a temporal phenomenon for individual patients. But as an experimental approach not yet fully tested and adopted into clinical practice, it is also temporary, provisional, and promissory, with regard to its truth and its future use. So, although the basis of likeness and comparison for current prognostic tools can draw frustrated confusion among users, the signifcance of EBLIS is that likeness and comparison are personalised, serialised as n=1. Te objects they relate, the lines marked out mean they give a capacity to be inhabited and embodied (Haraway 1997: 11). Tis capacity, however, remains a promise at this stage of the research, it is prefgured, and these promises difer for patient-participants involved in the study as they do the clinicians and scientists leading the research, and the organisations, funders, and companies that participate in the study's progress.

Figural representation in the tradition considered by Erich Auerbach (1938/1959) involves a frst event or person signifying both itself and the second that it involves or fulfls. Serial and recursive forms of historical representation structure and authorise the anticipation of prospective potential and retrospective analyses. Past and future phenomena are at once articulated and entwine. As a historical and aesthetic framework, Auerbach's theory of fguration illuminates the present as it is known and experienced as fulflling a past, with each event or type of the past having the potential to join another in the future. It is in this sense these fgures depend on time lived conditionally, partially, or in state of waiting and abeyance. As historian Hayden White noted of Auerbach's fgures, their fow and sense of trajectory is not known according to a linear sequence: "the making of a promise," White writes, "can be deduced retrospectively from a fulflment, but a fulflment cannot be inferred prospectively from the making of a promise" (White 1999: 89).

While Auerbach's historical theory concerned a combination of theological and literary examples, combining the material-symbolic fulflment of parabolic narratives, signs, and portents. He theorised the distributed efects of fgures as epistemic patterns and as ways of materialising the abstract, making the symbolic concrete in a moving present. Te making of molecular relapse as a form of historical knowledge cannot (yet) be determined prospectively; it needs the retrospective relief of clinical relapse to have epistemic value, in the same way that biblical testaments are marked old and new in a fgural relation to one another. While personalised blood monitoring of ctDNA presents a novel technological intervention based on high-throughput, next-generation genetic sequencing it also follows an archaic analytic structure, insofar as it relies on serial fgurations of something unresolved: a (yet to be defned) disease-in-progress. Figures of this kind follow one and another, as fgures prefgured, as each test depends and has value according to its place in a sequence.

As a study whose aim is to bring greater certainty over future events, the outcomes of EBLIS were uncertain when we were conducting our feldwork. Given this uncertainty its fgures needed to be managed for patient-participants, and they, in turn, needed to learn how to manage fgures of promise. When enrolled into the study patient-participants were told that clinicians and researchers "will not be performing any tests that have an infuence on your care. It is therefore unlikely that the study will yield any new information that will afect you personally."2 As a consequence, study feedback to patient-participants was limited to communications concerning their continued participation and enrolment. Because clinical researchers were unsure of the outcomes of the study and

<sup>2</sup>Patient Information Sheet (13/LO/1152), Version 4, 10/10/2018.

because there exists no data to show that treating patients based on ctDNA has better clinical outcomes, patient-participants in EBLIS could not follow ctDNA levels as they were being tracked by this research group. Tey occupied and were preoccupied by the fgures but did not (yet) inhabit them.

When participants were recruited and when they were later consented to extend the study in 2018–2019 for a further 4 years, they were told "there is no beneft to you personally from taking part in this study."3 But visits to the clinic formed part of the research and gave patients access to an oncologist: a skilled specialist that understood and appreciated the potential long-term side efects of hormone medication that many participants were taking. Te clinician who met them every six months felt the care she gave was minimal: "they are so stable, it's quite a steady thing in the trial clinic just doing the same thing on them every six months. You're not giving them anything, especially in the EBLIS trial, I'm not treating them, I'm not giving them any medicine". But patients we spoke to perceived (and some reaped) the benefts of a greater level of care than if they were not research participants.

When we interviewed them almost all the women we met reported benefts that were psychological or material, personal, interpersonal, or social. Jill told us that she liked "the possibility to come here [to the hospital] every half year, even though my treatment is fnished already".4 Tis gave her "peace of mind" and made her feel "more relaxed" about her cancer and its uncertain future. Likewise, Margret appreciated the sense that she was being monitored, not by the new technologies being pioneered by EBLIS researchers but by simply feeling that "someone was keeping an eye on me". Gaining peace of mind in the present and near future, through the continuity of care that research participation ofered was seen as one beneft. Te benefts of liquid biopsies in the future were less clear.

Tese contrasts over the valuation of care in the research may indicate the difering interpretations of what being 'stable' means according to patients and clinicians, at diferent stages of cancer care, particularly in

<sup>3</sup>Patient Information Sheet (13/LO/1152), Version 4, 10/10/2018.

<sup>4</sup>All names used are pseudonyms.

the precarious period after treatment when cancer may or may not recur. Tey are diferences of perspective that pivot with the availability and distribution of clinical and experimental fgures, that also mark a diference between 'standardised' provisions of care and those associated with research. Although participants were not able to inhabit the personalised fgures that tracked ctDNA, they valued being monitored in trial clinics and believed that they would help bring improvements to clinical practice in years to come. Participating in this research study helped them manage the uncertain relations with their own health, while contributing to the transformation of relations to risk for others.

Te promise-fulflment structure characteristic of Auerbach's description of fguration illuminates these shadows these developments in EBLIS and in the larger feld of molecular genetics. Of the 13 women we interviewed in 2018–2019, 4 understood that EBLIS was tracking their ctDNA in a way that they understood to be 'personalised' or 'individualised', and in this sense they understood the broad intention of the specifc research study. Many spoke in broader terms of tracking biomarkers in the blood, and this would, they assumed, help bring about a more timely diagnosis for other cancer patients. Comprehension of when this might be possible and how tended to be vague. One participant explained that she simply wasn't sure whether the study was looking at "bits of DNA or bits of protein. I don't know what they were looking for".

How the information EBLIS generated about patients and patient groups, and how study information may infuence the course treatment, was also unclear to others. One woman was confused as to whether or not EBLIS would help her understand her risk of relapse. As the conversation developed she said she wanted to learn more about what was being discovered: "what are they fnding out from it, because that's interesting in itself" she said. "Even if they were fnding out nothing, that would be quite interesting too, wouldn't it?" Another woman expressed her frustration at not being told more about the research study she was involved in. Susan told me that she is eager to participate in research provided that participation meant the outcomes were transparent. She felt that an opportunity was being missed to think diferently about the individual as a research participant: "you're dealing with people and you want to make it individualised medicine," she said, "so if you want to individualise it, you can't ignore the individual." And yet, as we found when we interviewed clinicians and researchers, returning experimental results with no proven or straightforward programme of treatment could risk a duty of care involving clinical, ethical, and legal promises that cannot currently be fulflled.

Wider contributions from social science and humanities scholars have noted the harms of elevated or unrealistic promises and expectations associated with more personalised or precise medical approaches (see Feiler et al. 2017; Erikainen and Chan 2019). Tese scholars stress the shortfalls between promise and reality and the personal and public losses that follow (see Dickenson 2011; Prasad 2016; Maughan 2017; Rushford and Greenhalgh 2020). Others have documented how hype, promise, and expectation play a constitutive role in biotechnological innovation, with discursive speculation infuencing the material shape, quality, and extent of collaborations, resource allocation, and markets (e.g., Brown and Michael 2003; Brown and Michael 2003; Martin et al. 2008; Adams, Murphy, and Clarke 2009; Tutton 2012; Haase et al. 2015). Tese sociologies of expectation, hope, and anticipation document the work of discursive prospecting that accompanies biotechnological innovation. Research in liquid biopsies coordinates and manages resources via an iterative, test-retest logic of embedded promises and expectations. And studies like EBLIS, with its graphic recomposition of text, image, and number, and its provisional stratifcation of persons according to molecular progression, do not pose one possible future but many.

In addition to the views of patient-participants, whose hopes and expectations we found to be managed within a framework of rolling consent common in translational research, our interviews with clinical and laboratory researchers managing EBLIS saw its potential in diferent ways; they recognised its accomplishments, uncertainties, and possibilities, fguring diferent durations via given modes of participation. Currently, patients receiving cytotoxic therapies for overt metastatic disease rarely see curative outcomes. In the more distant future, with an expansion of trials and studies that can investigate using ctDNA levels to guide clinical decision-making, it might be possible to "salvage patients who are ctDNA-positive with second-line therapies" (Coombes et al. 2019). Here there is a desire for a just-in-time change to future outcomes, based on better predictions of what is likely to pass. One clinical researcher, interviewed at the time the article in *Clinical Cancer Research* went to press, described EBLIS as being able to "open the door to potentially an entirely new paradigm" for diagnosis, treatment, and monitoring.

Tis clinical interest in treating patients earlier and providing them with better clinical outcomes was contrasted to the potential problems this technology might cause in the more immediate future:

we've developed a test which is in advance of having any treatment for the patients, which has been proven to beneft them. So in a way, from the patients' point of view, I think it's a bit of a disaster, because now it's going to be wheeled in, these results are all going to be given to all the patients. Tey're going to have their results that show they've got some problem in the blood, and then they're going to come back three months later and it's going to be even higher. Tere's going to be no scan evidence of any disease, and the doctors won't know what to do.

At a stage when the patient using liquid biopsies in this breast cancer clinic remains a fgure to be realised—at least in the NHS—the treatment of patients remains dependent on future programmes of research that enter further, iterative patterns of promise and fulflment.5 In this clinician's view EBLIS indicates the need to follow stratifed sub-groups of relapsed patients whose earlier treatment could show the beneft of treating at the point of molecular rather than clinical relapse:

at the point of molecular relapse, you could have a total of perhaps as many as 10 to 20 phenotypes of patient. So you're going to have to design multiple trials for each of those subcategories of molecular relapse, each of which will involve as many as 1,000 patients, and long follow up, and survival analysis.

Previous patient cohorts that were divided and treated by broad molecular and histopathological groups may now face further subdivision, according to when and how they relapse via threshold numbers of ctDNA

<sup>5</sup>At the time of writing this chapter, liquid biopsies were being trialled in diferent NHS sites (see NHS England 2020).

detected. What is distinctive for the personalised tracking of cancer when thought in terms of its promissory structure then is not that it is subject to a see-saw motion of hype and disappointment, or that it makes a single promise for a group of susceptible patients and investors, but that the system of analysis redistributes the basis of clinical groupings and the temporal grounds that once grounded predictions over long- and shortterm durations. Historical time for individuals and groups is reconfgured with molecular evolution. Trough a logic of serial testing and retesting, EBLIS marks out in lines the course of disease for individual patients, while recalibrating how cancer patients compare (or no longer compare) to others.

Te graphic compositions noted above relate and visualise movement in forking, braiding deltas, where lines are read as fgures not only of changing biomarkers that indicate somatic change but also interpersonal and comparative fgures of analysis and feeling, which mark out links within and between individual patient-participants, as well as the wider ecologies of contemporary translational research in the biosciences (see Crabu 2018; Rajan and Leonelli 2013). Figuration links and combines via various scales, sources, and kinds of data, at once intimately personal and radically impersonal in terms of both duration and bureaucracy.

Studies such as EBLIS fgure cancer's progression in experimental periods or intervals. Tese are signifcant for individuals in that they may infuence the categories of disease status that help guide clinical decisionmaking. Tey may afect more generalisable defnitions of precision by adjusting the 'right person' at the 'right time' with the 'right treatment'. In this sense, rather than simply fulflling existing hopes and expectations attached to fxed ideas of 'precision' or 'personalised medicine', EBLIS refgures the temporality of the personal and the precise. However, even within the relatively short time period that EBLIS has been active, a study punctuated by documenting the utility of ctDNA for tracking breast cancer patients at high levels of patient specifcity, it does not maintain a fxed understanding of the right person or the right time.

Te frst phase of research showed the non-invasive detection of preclinical metastases using a personalised ctDNA analysis. Researchers used tumour exome data to design patient-specifc 16-plex assays and deep sequencing of plasma cfDNA at an average depth of 100,000 reads per target, a sensitivity to the level of a single, mutant molecule. Te next phase of research involved extracting the whole-exome sequencing (WES) data from serial plasma samples to fnd novel mutations and new copynumber events that evolved from the primary tumour (see Hastings et al. 2021). "You can also track the evolution of the tumour," explained one bioinformaticist, "and you can also see if a patient is responding to treatment or not." While the clinical researcher focussed on the validation of liquid biopsies in diferent patients, their colleague highlighted a potential for further research to understand the specifc molecular characteristics of each relapse. Since blood samples taken from EBLIS participants were relatively large in volume, researchers explained that this next phase would use the same samples and occupy the same time points. In this sense the frst phase of EBLIS could serve fgures that then contrast to experimental fgures of the future, where the frst iteration informs the next. As serial test EBLIS has a serial, test-retest relation to its own progress. Tis additional layer of potential in the study data, working in parallel to patient-level tracking in follow up, and WES tracking of progression and monitoring among metastatic patients, promises a prognostic tool. Te excitement that accompanies developments in molecular oncology may not necessarily lead to an infnitely granular segregation of cancer categories, treatments, and predicted outcomes, but to their recombination, albeit made up of diferent fgures and a diferent idea of portraiture.

In short, EBLIS has multiple horizons composed of overlapping parts. Tese compositions and contrasts are a work of fgures used to present its data; a broader, interlocking programme that progresses by incremental phases. Indeed, in combination with EBLIS phases one and two is another; there is a further phase of research that is broadly biostatistical and predictive in nature, aiming to take clinical and genomic data from other EBLIS sub-projects to build a prediction model using machine learning techniques. Tis model would "apply to anybody who's coming into the study, or any cancer that gets sequenced. You could run it against this model and see if they ft the criteria of a patient who might relapse". At the time of our interviews, using the data of relapsed patients to create a predictive model was in an early phase of planning and development. But even as a hope or possibility it tells us how a personalised, n=1 tracking study could be used to build diferent kinds of prospective cohorts—data to build a prognostic tool to determine another set of standardised outcomes.

#### **Conclusion**

In her 2019 memoir *Te Undying*, Anne Boyer describes her diagnosis and treatment history. "My tumour," she writes, "started on a screen, and I returned it there. I entered its precise qualities into the prognostic calculator that promised to display the future in a pictograph. Te dead women were represented by forty-eight dark pink frowning faces, the living ones by ffty-two smiling green ones. All of these faces were supposed to, like me, be forty-one years old and with exactly the same version of my disease, but none of these faces, living or dead, said why or when or who" (Boyer 2019: 41-42). Boyer explains that her disease is known to her as a screen image and her tumour's mediated qualities are not exclusively biological, they extended across a vast and comparative network. Te sensory status of cancer as a 'silent killer' has long been linked to its malignant and unpredictable danger, giving further reason to picture it in diferent numbers, images, and texts. Personifying cancer's evasiveness—as a fgure difcult to see, hear, or touch—is closely linked to cancer's exposure, capture, and control (Sontag 1978; Bowker & Star 1999; Lochlann Jain 2013; Semino et al. 2018).

Tis chapter has explored how novel molecular fgurations of breast cancer challenge established methods of picturing its future course, both by breaking up sub-groups and by allowing the rapid introduction of targeted therapies. Meanwhile, research in the feld of 'liquid biopsies' generates ways of fguring disease recurrence that tracks changes in disease for individuals, as a line or path determined by combinations of data. What this chapter has been keen to stress is how the workings of EBLIS for individual patients involve a *layering* of fgures, emerging and residual, novel and archaic in pattern, that parallel trajectories of development and progression integral to wider infrastructures of translational research. To what extent such fgures can be symbolically and materially inhabited is a question of time. Or rather, a question of how time is questioned, coded and tracked, transformed into protocol and standardised.

**Acknowledgements** Te authors would like to thank their many collaborators. In particular, the staf and patients who participated in our interviews. We are grateful to Charles Coombes and his colleagues, whose involvement and feedback to our studies have been invaluable. We would like to acknowledge the close support of Kelly Gleason, Helen Kicono, Sanela Andrijac, Charlotte Ion, and Molly Gray. Tis research was supported by the Wellcome Trust (205456/Z/16/Z).

# **References**


**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/ licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **The Gardener and the Walled Garden**

## **Sophie Day, Jayne Smith, and Helen Ward**

# **Introduction**

Tis chapter explores the close connections between health care and research in a London hospital through Jayne's—one of the authors1 —experiences. We are an anthropologist (Sophie Day), a patient with breast cancer (Jayne Smith) and a clinical epidemiologist (Helen Ward) with diferent positions in this research hospital and diferent perspectives on experimental cancer

S. Day (\*)

Department of Anthropology, Goldsmiths, University of London, London, UK

Patient Experience Research Centre, Imperial College London, London, UK e-mail: s.day@gold.ac.uk

J. Smith London, UK

H. Ward Patient Experience Research Centre, Imperial College London, London, UK e-mail: h.ward@imperial.co.uk

<sup>1</sup>We use frst names in the text when referring to each other.

care (Day et al. 2021). Te frst-person plural that we adopt therefore shifts in its referents. Te letter from Jayne (below) shows that she wanted to know whether the samples she contributed to several medical research studies were useful and what had come of, and from, them. Strict governance of health data precluded Jayne from fnding out herself, but Sophie and Helen had university positions that allowed them to cross garden walls into what are sometimes called Trusted Research Environments.2

Jayne is an absent presence in the 'detective work' we describe, marked by a moniker, 'the gardener'. Tis fgure organised information fows among staf around hospital and research sites, many of whom had had never met Jayne and never knew her history. Because of this trafc, it organised our collaboration initially, confguring an inclusive 'we' that refers to our explorations of the history and implications of data-intensive health research and care as well as an exclusive 'we' that refers to the eforts that Sophie and Helen made to fgure out what had happened to Jayne's samples and data. Combining insights as a patient and as staf, we show how this 'name'—referring to Jayne's occupation—fortuitously ofered a conduit into a landscape of research and care, and the connections and gaps between areas of work as they changed over a period of six years.3 We then turn to what the gardener was cultivating, namely 'Grumpa', Jayne's name for her tumour. If Jayne considered Grumpa was hers and indeed part of her, she was happy to share her tumour and Grumpa was detached repeatedly from Jayne in the form of 'golden' or 'precious' tissue samples and data. Tese 'cuttings' or 'seeds' elicited further work as clinical and laboratory researchers cultivated diferent forms of Grumpa in a series of walled gardens. We therefore understood that there were several gardeners in several gardens, all cultivating aspects of Grumpa and sensing the tumour diferently through work practices which themselves changed in response to varied developments including eforts to realise the values of health data more efectively. We recognise a

<sup>2</sup>Walled gardens describe protected data enclaves where information from health services can be accessed by researchers. Platforms such as Facebook and Google popularised the concept of walled gardens as a way of storing and protecting data they collected on people's browsing histories or preferences (Plantin et al. 2018). Walls were designed to exclude competitors from access to valuable assets. Health regulators also developed practices of walling gardens or Trusted Research Environments to protect patient confdentiality.

<sup>3</sup> Sophie and Helen have shown the importance of diferent perspectives on health services through collaborative work among staf, patients and researchers (Ward and Day 1997; Day et al. 2017).

series of fgure/ground reversals that shift the relations between gardener, plant and garden. Grumpa too can be fgured as a gardener, cultivating us all—the three authors as well as clinical and research staf—insofar as it motivated sustained exploration into its mutable materiality and the conditions in which it diminished or thrived.

#### **A Letter, Jayne Smith (2019)**

After Sophie and Helen had conducted interviews and attended relevant meetings, Jayne put her thoughts into a 2019 letter for the three of us.

"After 2 years of living in fear and denial, I was diagnosed with bilateral metastasised breast cancer in early 2013. … Just by looking at my breasts it was obvious that the disease was advanced, … but the clinical staf who treated me showed me the utmost kindness…. In fact, I got the impression that they saw me as an extreme case, if not a curiosity, hence the heightened interest in me.

From almost the beginning of my treatment I became involved in some kind of research. Tat, in itself, gave me some purpose in dealing with my disease, with a hope that my misfortune could eventually be benefcial to other breast cancer suferers, and it therefore put a positive spin on my condition. I was frst involved in some research with Helen about patient experience, which also helped me clarify things in my own mind.

Te frst two years of my treatment consisted of hormone medication, which seemed to work for about 18 months, but then fungation4 set in, and I had to accept surgery. At the same time, I was ofered the opportunity to take part in the RADICAL trial, which was testing a drug which would boost my existing hormone medication. Te registrar and trial coordinator seemed very keen that I should do it, so I agreed - if it could be benefcial to me and also help others, why not?

A few weeks after the trial started, I had a mastectomy and lumpectomy, and the tissue removed was given to the RADICAL research team. I was on the trial for three and a half years, and it seemed to work by keeping my disease stable without my sufering extreme side efects.

<sup>4</sup>Fungation occurs when a breast tumour involves the local skin causing a wound which can ulcerate and become infected.

Every four weeks blood samples were taken and sent of for research. As time went on the clinicians caring for me became more and more amazed that I was tolerating it so well. When the trial had to end in June 2018 because my cancer had progressed, the tissue from my second mastectomy also went to research.

Up until now, I just thought that all my cancerous boobs and bodily fuids had disappeared anonymously into an abyss of data, together with those of millions of other cancer patients - just a drop in the ocean. However, I did hear unofcially that my 'bits' were viewed as coming from a 'gold' patient, and that there were only 2 other gold patients in this lab. Given the opportunity, I would love to reveal myself as that 'gold' patient and fnd out how my samples were used and whether they were instrumental, even in a tiny way, in any breakthrough in the treatment of breast cancer. I know patient confdentiality is of paramount importance, but there must be a way round it for consenting patients.

In the 'Garden' analogy, to me my breast cancer is a unique hybrid plant I have grown, which has been taken for propagation into a walled garden to which I have no access. I would like to see what has happened to it. Did it end up on the bonfre? In the compost? Were seeds/cuttings taken? etc.

Is there a shortage of patients willing to allow their tissue etc. to be used in research, and if so, would the ability to know the outcome increase patients' willingness to participate? Te fact that I am still involved in some kind of research such as this continues to put a positive spin on my condition."5

# **"I'd like to know what they've done with my stuff" (Jayne, 2018 interview)**

We met in 2013 as Jayne became a patient. She presented relatively late with advanced disease and wanted to avoid surgery and chemotherapy. Following her initial diagnosis and treatment preferences, as Jayne writes

<sup>5</sup>We use double quotation marks for verbatim citations and single quotation marks for records from our feld notes.

in her letter, she began hormonal treatment—with an aromatase inhibitor called letrozole. Her tumours shrunk and she remained relatively well for more than a year. In 2014, the tumour on her left side started fungating and, in early 2015, she had a mastectomy and lumpectomy. She also joined a clinical trial, the RADICAL drug treatment trial (Seckl et al. 2017), for three and a half years until further symptoms meant she had to stop the trial drug. As far as Jayne was concerned, the treatment 'which was to boost up the letrozole' had worked and perhaps saved her life. She subsequently had a second mastectomy and changed her aromatase inhibitor. We had heard about some of these developments from colleagues, for example, when Jayne featured in a newspaper article about a local gardener on a cancer trial (Rivers 2016) and when she gave a talk to an experimental medicine conference. Her 'case' interested staf in the service and beyond and was attached to the label of gardener as it was discussed, with her consent, at internal and external clinical meetings.

Jayne has contributed to Imperial College Tissue Bank, RADICAL trial samples and data, and routine health records but she has access only to her own clinical records. Healthcare staf can retrieve material they need for their job, and some staf have research roles giving them access to datasets related to the institutional tissue bank or to clinical trials, which also sit independently. Governance of research data requires that every tissue sample and related data can be tracked in both directions back to the patient and forward to the analysis—to ensure research integrity. Being trackable does not mean that data remain attached to their source, and indeed materials are de-identifed and stripped of personal markers before use. Tracking is achieved through an allocated identifer which circulates inside a research setting without enabling individuals to be identifed. However, researchers often want further samples from or information about their donor for which they rely on intermediaries who can re-identify and re-attach patient samples to the identifer. Where relevant, researchers also feed their results back to senior clinicians who will re-identify individuals if they consider fndings clinically relevant. Tis 'airlock' process enables only a few people with specifc job roles to 'unlock' pseudonymisation and transfer data into and out of research environments.

# **Walled Gardens**

Jayne's data and samples reside in three repositories—the Imperial College Tissue Bank, RADICAL trial samples and data, and health records which are walled gardens, albeit of very diferent dimensions, and they are insulated from each other by formal techniques of governance and access.

Jayne had little interest in remaining anonymous, protected by walls that also excluded her. Her questions about what happened to her data might provide a way, she said, of "turning my misfortune into a positive", that is, generating research fndings that would help future patients. She wondered about the value of her monthly blood donations and multiple scans during more than three years on the RADICAL trial: "It would be wrong to expect a cure to come out of my samples, but something…" because, in her view, the trial drug had worked. Jayne was most interested in her tumour samples. She described the removal of a fungating tumour in her frst operation and explained with pride how the research technicians waited for a blood sample so it could be couriered together with the tumour to the laboratory. In an interview just before we visited that laboratory, Helen asked, "Have you any idea what they've done with your tissue?" Jayne replied, "… As I mentioned in my speech to the people at the ECMC6 or whatever it was... this fungating monstrosity, we nicknamed it Grumpa-Loompa.7 We've always referred to is as Grumpa. I said to her (my sister), 'I'm going to the lab today.' She said, 'I hope Grumpa is not there looking at you in his jar.'" It was through Grumpa that Jayne fgured herself as a gardener who had cultivated this tumour unwittingly alongside her everyday occupation. After contributing to various walled gardens in the hospital and university, she thought that her cancer and the tissue samples it provided for other gardeners constituted

<sup>6</sup>ECMC: Experimental Cancer Medicine Centre, a network of cancer research centres in the UK. Jayne had given a talk to one of their meetings about her experience.

<sup>7</sup>Based on the Oompa-Loompas from Roald Dahl's Charlie and the Chocolate Factory. https:// en.wikipedia.org/wiki/List\_of\_Charlie\_and\_the\_Chocolate\_Factory\_characters#Te\_Oompa-Loompas. Jayne and her sister seem to have associated these fgures with their small size, incessant factory work, and mutable, mischievous, improvisational qualities rather than the imperial and racist tones that many have perceived. Tese qualities resonated with their perceptions of embodied breast cancer.

a unique learning opportunity. As she suggested during the lab tour described below, 'I don't want to be big-headed about it. I think I was a bit special when I started because it was so advanced when I presented myself… I think there was a lot of interest in my tumours and me I suppose because of that. [My friends with cancer] haven't had anywhere near as much interest in them as I have, they've felt a bit factory, conveyor belt type thing'.

#### **Walled Garden 1: The Tissue Bank**

From 2013 to 2018, Jayne provided samples to the Imperial College Tissue Bank, which is licensed by the 2004 Human Tissue Act8 to collect samples with permission for research. When patients donate to the tissue bank, they consent to participate in unspecifed research rather than particular studies, and today, they generally provide enduring consent for research use of surplus samples from continuing health care investigations. Samples sit within a walled garden and can only move outside the institution through a material transfer agreement or an existing site license for collaborative research with appropriate data sharing agreements. Researchers apply to the tissue bank to use samples in specifed studies within a given time frame—usually for the exploration of emerging questions in basic laboratory science but also in research training or for testing equipment. As far as laboratory researchers are concerned, tissue banking governance provides the fexibility to ask and explore preliminary questions.

Research technicians provided integral, albeit informal, support during Jayne's many hospital appointments from 2013 to 2018. Tey also constitute an interface between the service and research but, before 2018, it was not considered appropriate to open this conduit to Jayne herself or indeed to Sophie and Helen except in very partial ways. It was after Jayne

<sup>8</sup>Te Human Tissue Act (2004) came into force on 1 September 2006 and legislates on the use of human tissue samples. It established the Human Tissue Authority (HTA) to regulate activities concerning the removal, storage, use and disposal of human tissue samples for defned Scheduled Purposes, including 'research in connection with disorders, or the functioning of, the human body'.

stopped contributing regular samples that Kelly Gleason, Senior CRUK9 Research Nurse at Imperial College London, organised a visit to a laboratory that had worked with Jayne's tissue bank samples; she invited the three of us to join the tour.

Te laboratory group were studying the epigenetics of evolution in hormone-positive breast cancer. Teir work relied on repeated samples from the same individuals who had received neither surgery nor chemotherapy. As the head of the laboratory confrmed, these series of samples were 'as rare as white fies', and therefore 'golden' or 'precious'. Since Jayne had initially declined surgery and never undergone chemotherapy, hers were among the small number with which this laboratory group obtained the DNA fngerprint of tumours over a period of one to two years—before, during and after endocrine treatment. Tey tried to establish what counted as the same or diferent types of tumour by assessing genetic heterogeneity and asked what made some tumours start to grow again.

During our visit, we learned how tumour samples arrived in dry ice by courier. Close liaison between laboratory staf and clinical research technicians was essential because the samples had to be used immediately in the research. We were shown some of their techniques and tools, including live cell lines of breast cancer from Sister Catherine Frances, a Catholic nun who developed metastatic disease in the chest wall and pleura in 1971. Cells from her pleural efusion were the frst to be successfully cultured, and her MCF-7 cell line has led to over 25,000 published reports (Lee et al. 2015). Sister Frances' cells were oestrogen-receptor (ER) positive like Jayne's, and subsequent research using this cell line led to major advances in therapy including tamoxifen and aromatase inhibitors.

Material from serial biopsies has improved understanding of the mechanisms of tumour evolution in ER-positive cancers under selective pressure from aromatase inhibitors (Patten et al. 2018; Rosano et al. 2021). Related studies (see Viney and Day, this volume) have explored cell-free circulating tumour DNA in blood samples for biomarkers that may improve prognosis and suggest earlier interventions (Magnani et al. 2017;

<sup>9</sup>Cancer Research UK is the world's largest independent cancer research charity, funded almost entirely by public donations.

Hong et al. 2019; Coombes et al. 2019). Such 'liquid biopsies' ofer huge advantages over solid tumour biopsies for the monitoring of disease since they are relatively easy to give as well as to receive, process and store (Hastings et al. 2021). At the end of the visit, Jayne was in conversation with the head of the laboratory who said that he 'did not have green fngers' and wasn't a gardener. She replied that the work he had shown us in the laboratory suggested that he had all the skills and could also be a gardener, if he put his mind to it.

Jayne was much more interested in the uses and values of her tumour samples than the 22 blood donations we found that she had also made available for research through the tissue bank. An audit in 2019–2020 showed that Jayne had provided an unusually large number of samples. Four hundred and seven people each provided between one and twentysix samples with an average of between two and three; only nineteen people provided ten or more samples. Te audit showed that these samples were explored in collaborative research with Sweden and the USA, for example, as well as in the UK.

#### **Walled Garden 2: RADICAL Trial**

Exploratory studies using tissue bank samples sometimes lead to proposals for clinical trials. Trials require specifc approvals and consent from participants since they involve 'investigational medicinal products' such as drugs or devices.10 Tey require meticulous record-keeping including the validation of all samples and results in protected databases.

Research technicians were responsible for recruitment and follow up to the RADICAL trial under the institution's Cancer Clinical Trials Unit. Jayne's monthly blood samples were spun and stored in a RADICAL freezer. Te samples were managed thereafter by the Clinical Trials Unit at a site nearby. Te technician responsible for RADICAL from 2017 to 2018 explained how she entered results and data onto an *InForm* ITM (Integrated Trial Management) System, which is used widely in the pharmaceutical industry and charity sector. Tis 'walled garden' includes data

<sup>10</sup>Medicines for Human Use (Clinical Trials) Regulations (2004).

imported from several hospital systems. A technician manually extracted material from the hospital service data system to combine with reports from trial participants and results from separate imaging and neurology systems before collecting signatures from clinicians for the site fle. Tese data were audited on conclusion of the study and archived. Only then, in 2018, did Jayne's blood become accessible outside RADICAL and, as far as we could discover, samples stored at the drug company labs11 were returned to the local centre to be either destroyed or repurposed for other studies.

Interviewing the principal investigator (PI) of this study, Sophie learned that cancer prognosis was worse when fbroblast growth factors, particularly FGF2 (fbroblast growth factor number 2), become elevated. His group investigated molecular mechanisms *in vitro*, then in animals and eventually in people afected by a range of cancers who had become resistant to treatment with letrozole or anastrozole. Te group developed a blocker to FGF2 called AZD4547, which they hoped would overcome resistance to treatment. After a pilot study, they trialled the compound in combination with letrozole or anastrozole and reported subsequently that about one-third of participants benefted (Seckl et al. 2017). Te research programme then stalled because the group were unable to stratify participants ahead of treatment: 'we need to know how to select those patients [who will beneft] and not the ones for whom it doesn't work, and currently we can't do that. Tere is a test which gives you results before imaging can, within a few weeks of starting treatment, but it would be better to know before you start treatment. Tat is tough. … If we could select patients properly, we could do a bigger trial and properly answer whether this inhibitor works or not' (feld notes, 2019). Tis next step of stratifying patients and selecting only those who might beneft from the treatment required either serial biopsies, which they did not have, or appropriate surrogate markers.

Financial issues may also have contributed to the hiatus in this research programme. Interviewing the frst research technician responsible for the study, Sophie heard that the trial drug was 'on the shelf' until researchers made diferent combinations available for trial across a greater range of

<sup>11</sup>Te RADICAL trial involved the company AstraZeneca https://tinyurl.com/y54z34gz

cancers, thus defning a larger potential market for anything that might be licensed. A colleague also suggested that participants sufered too many side efects for the company to adopt the treatment in early (as opposed to late) breast cancer, which was their only fnancially viable option because it would include a larger number of people.

When Jayne heard this news, she was unsure whether her donations had been useful but remained convinced that she benefted personally from her 46 cycles of treatment. In addition, she felt she had profted from the close monitoring and incidental fndings that were shared. At her frst diagnostic appointment in 2013, possible signs of cancer were mentioned in Jayne's lungs, liver and pelvis. Eventually, a consensus developed that there were four small cancerous nodules in Jayne's lungs while RADICAL trial monitoring suggested that there was no cancer in her liver, just fatty cysts. A torn retina was also found and repaired 'then and there'; subsequently, an issue about drainage in her eyes was treated in the hospital, which Jayne understood might have caused glaucoma if left untreated. Jayne also felt that she would not have been recommended her second mastectomy in 2018 had her clinicians not been involved in research, since the tumour was so small—only 15 mm—when it was discerned.

Sophie and Helen learned that Jayne would receive formal notifcation of trial results when they became available if she had requested them in her original consent form. Te trials unit told us that the results were still being analysed at the beginning of 2020 and referred us to a key summary on the CRUK website. Here, the investigators report that the trial showed that AZD4547 combined with one of two aromatase inhibitors appeared to be safe and showed anti-tumour activity in some people.12 Trials are underway to explore whether results can be improved by selecting patients with specifc biomarkers who may beneft most from the drug combination (Tarantino et al. 2020). Jayne hopes that the work will continue.

<sup>12</sup> https://www.cancerresearchuk.org/about-cancer/find-a-clinical-trial/a-trial-of-azd4547-forbreast-cancer-that-is-oestrogen-receptor-positive-got-worse-despite-having-anastrozole-orletrozole-radical#undefned

### **Walled Garden 3: Patient Records**

Jayne's patient records contain traces and links to most of the research activity described above. Clinical consultants recorded decisions in her notes after reviewing the results of tests through which her health and response to the RADICAL trial treatment were monitored. Paper patient records have long been used in hospitals and other clinical settings, and, in recent decades, test results that were stored in electronic form were also printed to add to a patient fle. Initially, Jayne had a paper record which contained copies of letters, results, procedures, treatments and clinical notes. She said that her fle became so large and heavy that staf would have to use a bag to carry it. Although her paper records contained an enormous amount of detailed data, they were not shared outside the hospital and so were largely inaccessible for research, audit or to Jayne herself. In 2016, the hospital introduced an electronic health record (EHR) system hosted on a platform run by the company, Cerner. Clinicians involved in patient care can view these records in the same way as previous paper records. Te system links to other local health records (see below), and Jayne now has some access to these through a patient platform called the Care Information Exchange; Jayne can look at her recent results, add comments and upload data from health trackers. She explained, however, when hospital care was radically curtailed in 2020 that she did not want to receive any results by phone or electronically, only in person.

## **A Changing Landscape: from Walled Gardens to Data Flows**

Te fgures of the gardener and of Grumpa have evolved in relation to their grounds, the walled gardens. Rapid developments in data collection and the increasing interoperability of data systems mean that traces of Jayne in her data and materials are now embedded in much larger warehouses. Both data and samples may appear to have "disappeared anonymously into an abyss of data" but they are also contributing to the creation of value in the UK's life sciences strategy to build assets from unique NHS data sets. Te gardener and Grumpa are valued as "pluripotent" elements for future research with these datasets.

Although Jayne's materials sit in three and no doubt further walled gardens, some people can travel between them, including research technicians. Given appropriate consent, excess samples can also be repurposed for subsequent research, and clinically relevant information shared. Jayne, for example, consented to the collection of 'archival tissue samples' in RADICAL for exploratory work via tissue banking to look for markers that might infuence the development of breast cancers or help explore patient responses to treatment. From 2013 to 2020, the ways that data are collected, stored and used were transformed in health services and research. EHRs such as Cerner enable easier reporting and sharing of data, and NHS investment in these EHRs "supports our wider interoperability strategy and avoids the 'walled garden' legacy of trapping data in institutions" (Swindells and Smart 2017), simultaneously contributing to core UK government strategies aligning health, life science and economic policies (Department for Business, Energy and Industrial Strategy, 2017).

In the local NHS Trust where Jayne is a patient, a Whole Systems Integrated Care (WSIC) database is now extending this infrastructure (Bottle et al. 2020). A researcher who has been closely involved in its development explained: "[it] is currently used for direct patient care, service evaluation, commissioning and for research [through the system known] as 'Discover'. For direct patient care, the WSIC team developed disease-specifc dashboards, which can be accessed by healthcare professionals with a legitimate relationship with WSIC. For other uses, the database is de-identifed" (interview, 2020). Tis single integrated care system in North West London contains data on 2.4 million people and can be used by clinicians to support the provision of care, by managers and auditors to review activity as well as generating statutory reporting, for example on cancer waiting times. A pseudonymised form of the database (Discover) can also be used for research, and patients who have consented to be contacted for further research can be re-identifed if they meet a study's inclusion criteria (Fig. 8.1).

Since 2020, developments in the collection, storage and use of health data have further intensifed in response to the COVID-19 pandemic.

**Fig. 8.1** Walled gardens of data: links across the local health sector used to create the Whole Systems Integrated Care database; Figure reproduced from Bottle et al. (2020), under Creative Commons Attribution 4.0 International License

A researcher we interviewed explained how the use of individual and group-level patient data "has been even further facilitated due to COVID, in a way, in that we've accelerated development of a virtual platform that our researchers can access, and we'll have access to anonymised EHR data from Imperial College Healthcare Trust" (interview 2020). Te WSIC platform has also been used to track COVID-19: people who use the Care Information Exchange are invited to provide weekly updates on whether they have experienced symptoms, and they can respond to other surveys about their care and preferences, for example, relating to a contact tracing app (Bachtiger et al. 2020). Jayne has participated and found it interesting, indeed unusual, to be invited to provide a written ('free text') account of the impact of COVID-19 on her experience of cancer services.

Tese larger data warehouses are not alternatives to the walled gardens described but rather a larger garden: "Te technical solution comprises a 'walled garden' approach, which uses secured virtual sessions run from within a secure infrastructure. … All projects are logically segregated from each other within the safe haven, and access is controlled and permitted only to those users who have been registered and attended information governance awareness training courses, as well as completed online information governance tests annually for their reaccreditation" (Lea et al. 2016).

Data developments associated with EHR, WSIC and Discover mean that Jayne's data can be aggregated with millions of other patient records in a way that was not possible fve years ago. Data produced from her care—the details and dates of her diagnosis, test results, treatments, visits, etc.—also link the hospital she attends and primary care (UK general practice). A clinical researcher explained how this infrastructure enabled approaches other than traditional clinical trials, "(we) have moved on, beginning to see the utility and using e-health data and electronic health record data, rather than collecting vast amounts of information on patients that we recruit to studies. And how we can really make the best use of that information, to do almost quasi-experimental or natural experimental designs, and improve patient outcomes" (interview 2020). Tey provided examples showing that this approach can occur in near real time with the introduction of alerts, for example, to a patient who may have sepsis. Tey can then assess whether alerts led to any improvement in outcomes (Honeyford et al. 2020).

As Jayne cautioned, however, data that is readily available in large quantities is not necessarily any more reliable. She said that letters to her general practitioner (GP) in her medical records had multiple errors including incorrect dates for her scans and her most recent treatment. Te very size of these linked data sets "does not eliminate and may even amplify systematic error" (Ehrenstein et al. 2017), which can undermine their usefulness even if the greater scrutiny may also reduce errors.

In sum, our investigations found traces of Jayne's history of treatment and research participation in clinical records in both identifed and deidentifed form, in parafn blocks and serum samples in banks that are kept for 20 years, in DNA sequences and in research results and papers. Sophie and Helen were able to explore three 'gardens' in depth and found that Jayne and other patients have provided materials for local research in surgery, a spectroscopy study associated with cell biology and drug delivery systems, other types of cancer including metastatic cancers, a xenografting study with doubled systems of consent because it involved animal work, PhD projects through specifc consents and through the tissue bank. Along with samples from other patients, Jayne's contributions have informed several research papers as well as our own research on the impact of developments in cancer medicine (Day et al. 2017, 2021; McGrath-Lone et al. 2015). We did not fnd out about derivative uses in further studies such as those repurposing clinical trial bloods.

## **Grumpa**

Jayne thought her involvement in research and care was "all of a piece really" because of the collaborative focus on cultivating Grumpa, whether attached to or detached from its host. As Jayne wrote in her letter (above), "In the 'Garden' analogy, to me my breast cancer is a unique hybrid plant I have grown, which has been taken for propagation into a walled garden to which I have no access. … Did it end up on the bonfre? In the compost? Were seeds/cuttings taken? etc".

Tis fgure, Grumpa, was delineated collaboratively over a period of six years by several other gardeners as well as Jayne. Staf in the hospital and university sensed the cancer diferently in the clinical trial, the laboratory research programme and during Jayne's continuing care. 'Cuttings' were taken for research from Jayne's initial diagnostic biopsy in 2013 and shared. In 2015, Grumpa was distributed again following a lumpectomy and a mastectomy, and once more after another mastectomy in 2018. Relational, comparative and perspectival glimpses (Gal 2016) across at least some of these walled gardens constituted scaling devices which put together a history to Grumpa—and care plans and prognoses for Jayne. But, as the gardener, the one who produced and grew Grumpa and made all the collaborative work across cancer care and research possible, Jayne felt that she had been excluded from the results of this work and their potential relevance for breast cancer care more generally. Te history to her 'cuttings' was outside Jayne's control and practices of governance also made it very difcult to efect a comparative history across 'trusted research environments'.

RADICAL trial results from patients with a range of cancers were analysed as a combined set after the trial closed. Jayne was one of many contributors and her Grumpa samples seemed to have "disappeared anonymously into an abyss of data…". Investigators were also frustrated that they could not diferentiate between participants and select only those likely to beneft from the trial drug. Since cancers afecting diferent individuals respond to treatment and other evolutionary pressures in different ways, it is difcult to conduct clinical trials as though the indexicality of data is uniform and stable. By comparison, the exploratory epigenetic research conducted by another research team was more of a 'natural' experiment, rather like the new uses of linked data made possible by the WSIC database. Some materials remained indexed to Jayne over time even though they appeared to have been detached from her continuing care. Sophie and Helen's detective work showed that information travelled between the clinic and this laboratory group and that developments in one environment were understood in relation to the other—her clinicians were also active research investigators. A natural history of tumour evolution was constructed by integrating the results of clinical observations with laboratory and data research to track the evolution of cancers.

We were all struck by the 'immortal' cell line from Sister Frances and Jayne explained how she would love to fnd that her samples had been similarly important: "Given the opportunity, I would love to reveal myself as that 'gold' patient and fnd out how my samples were used and whether they were instrumental, even in a tiny way, in any breakthrough in the treatment of breast cancer." Tis cell line evokes a traceable continuity from donation to discovery that is rare, but recognisable. It reminded Sophie and Helen of research using HeLa cell lines, developed from a sample taken and used without consent or knowledge from Henrietta Lacks. Tis history is extensively documented as a history of racial and economic abuse that has become well known through the book and flm *Te Immortal Life of Henrietta Lacks* (Skloot 2010), which depicts the extraction of value without compensation. Jayne considers the (con)fguration of her samples in more positive terms. Her materials have not been used for *in vitro* cell lines (for which specifc permission would be required), but the laboratory team clarifed that rare, repeated samples such as hers were of substantial value to research into the evolution of hormone-positive cancers exposed to treatments *in vivo*.

Te contrast between the two types of cancer research we have described, a clinical trial and a laboratory programme informed by clinical observations also indicate multiple ways of being cut out of or included in prognoses. Jayne considered that her care benefted directly from research involvement. Like many other people, she hoped to improve the lives of future generations just as previous generations had contributed to her own wellbeing: 'if I've got to have this awful disease, at least it can do somebody else some good. It's made me feel better about it.' Benefts of building on historical legacies from generations of people afected by and working with cancer13 are commonly indexed to a distant collective future. But Jayne found that her research involvement was continuous with the ongoing care, personal and "near futures", what Jane Guyer calls a sedimented, cumulative sense and experience (Guyer 2007).

Describing the research uses of Jayne's samples is "to speak of a distributed, heterogenous thing" (Landecker 2000) which will likely continue to change. It was Grumpa, we suggest, that constituted the key fgure driving liaison between Jayne, cancer services and research to explore and respond to its evolution. Staf were aware of what is called clonal evolution, describing distinct subpopulations of cells that emerge.14 Most models consider that driver mutations and medical therapies represent important triggers in the environment that prompt adaptations. Te Grumpa fgure from which cuttings were taken enabled inferences to be made about developments in this adaptive landscape and enrolled the labour of clinical and research staf as well as Jayne herself. Preliminary fndings raise the possibility that the "metastatic cascade" in hormonedependent breast cancers is associated with chance epigenetic events rather than the clonal evolution characterising these cancers at an earlier stage before treatment (Rosano et al. 2021).

<sup>13</sup> See Guyer's (2007) reconsideration of the gifts described by Marcel Mauss that can only be returned indirectly across generations.

<sup>14</sup>Davis et al. (2017) note that in a cohort of 104 triple-negative breast-cancer (TNBC) patients, resolving subclones with deep sequencing identifed 1 to 19 subclones per patient (Shah et al. 2012). Another study used multi-region sequencing of 50 breast cancers and identifed only 1–4 major clonal subpopulations in each patient (Yates et al. 2015).

## **Conclusion**

Classifcations, treatments and knowledge change at diferent rates as they index possible futures in care and research. Jayne's questions about her data and samples led us to ask how her materials shaped several, more or less heterogeneous but interconnected forms of person and cancer, care and research. Jayne saw her stuf '*disappearing anonymously into an abyss of data'* in a study that in her view also saved her life while developing a 'unique profle' in a study of cancer evolution where her *golden samples* might also inform continuing care. Fortuitously, the fgure of the gardener, as a moniker for a person whose identity could not be shared across settings, allowed us to begin to 'fgure out' processes that were connected in some ways and separated in others. However, it is the second fgure of Grumpa, the cancer that lived with Jayne and yielded cuttings and seeds, that elicited collaboration among the authors as well as healthcare staf and researchers. Grumpa, distributed to various walled gardens, brings together the experimental and observational, care and research, the personal and impersonal, and the singular and plural as it changes in response to its surroundings, which are also changing.

Helen Verran (2010) describes two forms of generalising, where a onemany relation embeds or abstracts a 'case' such as ours as an example of something in general while a whole-part relation makes the history an emergent entity in a vague whole, whose parts will never add up to a complete picture (Verran 2010; Winthereik and Verran 2012). In Verran's view, there is an irreconcilable tension between these forms of generalising that demands a double vision. Sophie and Helen did not trace clear outcomes from Jayne's participation in research, nor any typical trajectory for those involved in an experimental cancer care combining dataintensive, laboratory and clinical research with health care. We (three) did not fnd how Jayne's data—stored, sometimes aggregated with others, and analysed—were applied in care settings or further scientifc studies. However, describing this collaboration from 2019 to 2020 in terms of fgures produces aspects of one-many *and* whole-part generalisations within a constitutionally incomplete picture of many moving parts.

Despite what was in Jayne's view a disappointing lack of closure, that is, the lack of a 'eureka' moment to our investigations, she concluded after discussing a draft of this chapter that her story and our combined fguring might encourage discussion between staf and patients about research that would "turn" what it fgured (Haraway 2008:159). Te small audit conducted by research technicians was conceived in similar terms: when results were shared, might they promote discussion and engagement with research and tissue banking, as suggested in published studies? (Bryant et al. 2015). Jayne wrote, "Te process of contributing to research is a positive incentive, and makes you feel a bit more special and supported. However, don't be under any illusions that your contribution will, on its own, be responsible for any 'Eureka' moment - it is still an unidentifable drop in the ocean. But without all the drops there would be no ocean".

**Acknowledgements** We would like to thank staf for sharing their insights and participating in interviews. Charles Coombes, Kelly Gleason and her team facilitated the work in cancer services and cancer research. Tanks also to colleagues in 'People Like You' (https://peoplelikeyou.ac.uk/): Will Viney, Roz Redd, Scott Wark and Celia Lury. Tis research was supported by the Wellcome Trust (205456/Z/16/Z). Helen Ward also acknowledges support from the NIHR Applied Research Collaborative North West London and the Imperial NIHR Biomedical Research Centre.

# **References**


**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/ by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **9**

# **Data Through Time: Figuring Out the Narrative Self in Longitudinal Research**

**Jane Elliott**

## **Introduction**

How might we describe and make sense of an individual's life? Is it best understood with reference to their accomplishments, family life, voluntary work and career—elements that might be narrated in a eulogy at their funeral? Or would this account miss the texture of their daily experience, the habits and routines that form the constant backdrop to these events? We now have substantial data resources from longitudinal studies that have tracked large samples of individuals over many decades. We also have myriad and increasing opportunities for tracking and recording our own daily lives and the lives of others. How might we extract and combine this information to understand, and potentially improve, individual lives?

Tis chapter has two parts. Te frst briefy explores the ways in which individuals have fgured within longitudinal research in the social

J. Elliott (\*)

Department of Sociology, Philosophy and Anthropology, University of Exeter, Exeter, UK

e-mail: Jane.Elliott@exeter.ac.uk

sciences and highlights an emerging set of methods focused on reconstructing individual cases within quantitative longitudinal research. Te second is partly inspired by recent literature that emphasises the importance of attending to the mundane, the routine and the everyday (Highmore 2004, 2011; Pink 2012; Back 2015; Neal and Murji 2015). Specifcally, I raise questions about the implications of the digital revolution (and in particular the self-tracking movement or 'personal informatics'), for future research practices within longitudinal studies. It is now possible for detailed information to be collected in real time on individuals' habits, behaviours and vital signs (Lupton 2016; Nef and Nafus 2016). Tis potentially provides researchers, and individuals themselves, with material that can be used to develop a diferent type of understanding of a life—one that focuses more on routine, lived experience and the practices and habits of daily life.

Te thread that binds these two halves is the suggestion that it is instructive to think through what can be considered as 'fgure' and 'ground' in our research, and in our representations of individuals' lives. Te metaphor is apt, partly because of the long sociological tradition of trying to understand the individual (or fgure) in social context (the ground) without unduly privileging either the agency of the individual actor or societal structures and constraints (Mills 1959). It may also have renewed utility as we try to 'fgure out' what it means to be an individual, and to lead a worthwhile and fulflling life in today's digital society. In particular, we could more easily conceive of fgure and ground in temporal terms. Perhaps the 'ground' are the routines of daily life which, by defnition, pass by almost unnoticed, while the fgures are the events and experiences that loom large in our memories and our narratives about ourselves. Indeed, could a better understanding of the 'ground' of the habits and rhythms of our quotidian existence provide the key to understanding how to lead 'better', more fulflling, lives?

In order to explore these questions I draw primarily on exemplars from Britain's portfolio of national longitudinal studies of individuals' lives. Tese include the long-running household panel study, known as "Understanding Society" (Buck and McFall 2011), together with the set of world-renowned cohort studies that have followed thousands of individuals from their birth (in the spring of 1946, 1958 and 1970 respectively) through childhood, adulthood and middle age (Power and Elliott 2006; Welshman 2012; Pearson 2016). As will be discussed below, these studies provide an instructive case study because they have been used by a wide range of researchers from diferent disciplines. Tis includes novel uses of the data to reconstruct or "refgure" individuals (Sharland et al. 2017, Tinkler et al. 2021; Carpentieri et al. 2022; Waller et al. 2020).

### **Section 1: Longitudinal Studies and Quantitative Representations of Individuals' Lives**

Some of the earliest longitudinal studies were carried out in the United States in the early twentieth century and focused on understanding children's development (Phelps and Colby 2002). However, Britain is unique in the world in having a portfolio of four national birth cohort studies that have followed individuals, born in a specifc year, through childhood, and into adolescence (Pearson 2016),1 and adult life (Wadsworth et al. 2006; Power and Elliott 2006; Elliott and Shepherd 2006). A key feature of longitudinal research is that by maintaining contact with a large sample of individuals, and re-surveying them, typically every fve to ten years, throughout their lives, it is possible to build up a rich and detailed record about the experiences of each member of the study. Tis is a type of quantitative life story, addressing many diferent aspects of each cohort member's life. Tese include their education, childhood experiences, employment, housing, relationships, fertility, social participation and physical and mental health (Ferri et al. 2003; Wadsworth et al. 2006; Power and Elliott 2006; Elliott and Shepherd 2006).

<sup>1</sup> Indeed, the UK also has a number of cohort studies based in specifc areas of the country, for example, the Avon Longitudinal Study of Parents and Children which started in the early 1990s and Born in Bradford. In addition, the UK has one of the largest household panel studies in the world, the UK Household Longitudinal Study. Te arguments made in this chapter apply equally to any longitudinal study that focuses on individual lives through time but the focus will be to use the 1958 cohort study as an exemplar.

Te ability to follow the development of individuals throughout their lives has an inherently appealing, narrative quality (Elliott 2008). Indeed, parallels can be drawn between the 1958 cohort study and Michael Apted's popular long-running documentary 'Seven Up!'. Tis has followed a much smaller sample of just 14 individuals from when they were 7 years old in 1964 (Burawoy 2009; Torne 2009).2 Te original premise for the series of documentaries was the Jesuit adage: 'give me a child until he is seven years old and I will show you the man'. Apted deliberately chose children from contrasting social class backgrounds in order to see how material circumstances impact on individuals' aspirations and life chances (Willis 2009). Te British Birth cohort studies have also focused on inequality and on understanding the extent to which poverty and deprivation prevent individuals from realising their potential (Wedge and Prosser 1973; Wedge and Essen 1982).

However, in contrast to Michael Apted's documentary approach, the majority of information collected in the British Longitudinal Studies is highly structured or quantitative. Terefore, its analysis typically involves the estimation of multivariate and longitudinal models. Tese focus on associations between diferent variables, and identifying which *factors* have the greatest impact on an outcome of interest later in life. Te models produced, typically populated by columns of coefcients and standard errors, can seem a far cry from stories, or narratives, about real human beings (Elliott 2005).

Tere is therefore a sense in which the cohort members themselves are obscured in the quantitative analyses that characterise the majority of work carried out using data from these studies (Elliott 2005, 2008). As Armstrong (2019) has argued, "Ironically just as these data points could claim to reveal a new numerical description of the individual, their combination and comparison … involved choreography of data points quite separate from the individual" (p. 110). In other words, we risk losing touch with the uniqueness and complexity of individual lives as these are represented as sets of summary variables that can be manipulated by the epidemiologists, economists, psychologists and sociologists who use the

<sup>2</sup>Apted (who sadly died early in 2021) was sometimes called a "Longitudinal Documentarian" (Torne 2009).

datasets. And we lose touch too with the ability of individuals to refect on their own lives and, perhaps, to compare them with those of others. It is rare for cohort members to be given a voice and enabled to refect on their own experiences.

One exception to this is a qualitative study conducted with a subsample of 220 members of the 1958 cohort between 2009 and 2010 (Elliott et al. 2010). Individuals were asked about their communities, social participation and weekly routines as well as being given an opportunity to tell their own life story. At the end of the interviews, cohort members were also asked about their experiences of being in the study throughout their lives (Parsons 2010). Many had positive memories of how it had made them feel 'special' in early life to be part of a study that would be useful to wider society. However, there was also a desire among some study members to receive more feedback about the study in the form of case studies and stories about other cohort members. As one cohort member said:

*I think most of the feedback that comes back is very, very generic which--, I tend to get bored halfway through reading so I don't bother…maybe some examples, some, I don't know, common case studies, stories, that sort of stuf would make it more interesting and I'd read it then.* [Interview 239] (From Parsons 2010, p. 15)

It is perhaps too strong to claim that individual cohort members actually disappear in the multivariate statistical analysis of their data. Rather they provide an essential background, contributing to the mass of data points from which statistical models are estimated. Whether we are researchers or readers of research fndings, we know that the individuals are there. It is the representative nature and large size of the sample that ensures the statistical models are credible representations of underlying processes in society (Hawkes and Plewis 2006; Mostafa et al. 2020). Even so, the intense focus on variables in multivariate analyses means that the agency and refexivity of individuals are likely to be obscured (Abbott 1992). In the quantitative, multivariate, longitudinal models that capitalise on the detailed prospective information in the cohort studies, it is the coefcients that populate the models which *fgure,* while the cohort members themselves provide the 'ground'*.*

Set against this, the relative invisibility of individual study members has the advantage of protecting the anonymity of those who have contributed a great deal of very personal, and sometimes sensitive, data throughout their lives. In contrast, in Apted's '*Seven Up!'* series the individuals are the key fgures in the documentaries. Indeed these individuals have taken on an almost celebrity status.3 However, this level of visibility has led some participants to opt-out. Five of the fourteen participants have declined to participate in at least some of the updates over the years. For example, Charles, recruited for the documentary from an elite public school, dropped out after 21 Up and has never returned; whereas Peter dropped out of the series after *28 Up*, following a campaign against him in the tabloid press due to his criticism of the Conservative government during his TV interview. He returned to the series for *56 Up* in order to publicise his band.

#### **Reconstructing the Individual Within Longitudinal Cohort Studies**

Despite the tendency of longitudinal studies to obscure the individuals who take part in them, there are a few examples of research which do take a more individual case-based approach. Tese studies recognise that the detailed and temporal nature of the studies, and the location of cohort members in a specifc historical context, mean that the studies have considerable narrative potential (Elliott 2005; Elliott et al. 2010; Waller et al. 2020).

Indeed, a number of researchers have adopted imaginative methods which in some senses reconstruct the individuals who have been fragmented into a set of variables so that those who were in the background come to fgure. For example, Singer et al. (1998) use the Wisconsin

<sup>3</sup>When Tony Walker (one of the participants) was interviewed about the death of Michael Apted on the BBC Radio 4 Today programme on 9 January 2021 (8:48 a.m.), he was treated as a celebrity by the interviewer, who asked questions about what the study had meant to him, thanked him for his contribution and said, 'Many of us feel we know you'—Tony replied by simply remarking 'I thought it was about Michael'.

longitudinal study to understand more about the factors that can lead to depression for some women. Tey use diferent waves of the Wisconsin study (many years apart) to piece together individual life stories for a small sub-sample of individuals. Singer et al. argue that "new insights are obtained as detailed information about real people are brought into focus" (Singer et al. 1998). Tese insights can then be used to generate hypotheses, which can in turn be tested using statistical models.

A recent paper, drawing on this approach, has used data from the longrunning British Household Panel Study to construct case studies of families who have been supported by social workers (Sharland et al. 2017). A key aim was to explore whether this more narrative methodology, focused on the lives of individual families, would provide insights into a counterintuitive fnding emerging from statistical analysis. Namely that families who have contact with social services have *poorer* outcomes than families in similar circumstances without support. In the authors' words "In the absence of complementary qualitative material, quantitative life histories seemed worth trying, to catch a glimpse of the stories beneath the aggregates" (Sharland et al. 2017 p. 670). Sharland et al. are understandably tentative in their advocacy for this method, based on its limited use to date. However, they conclude by arguing that given the impressive array of quantitative longitudinal studies in the UK, the USA and Europe, researchers might make better use of the "largely untapped narrative potential that may enrich our understanding of how lives unfold. Te quantitative life history narrative method ofers a chance to realise this potential" (Sharland et al. 2017).

Very recently, in the UK, two separate historical studies have adopted similar techniques and risen to this challenge. Peter Mandler's study on the history of secondary education since 1945 includes the creation of 150 pen portraits of cohort members from the 1946, and the 1958 cohort studies in order to understand more about the family backgrounds, educational and occupational trajectories of two separate generations (Carpentieri et al., 2022). Te *Girlhood and Later Life project* led by Penny Tinkler focuses on girls growing up in Britain in the 1950s to 1970s. Te team uses materials from the 1946 British Birth Cohort study (known as the National Survey of Health and Development (NSHD)) to reconstruct biographies of women from diferent education and class backgrounds, in order to understand more about their opportunities and life courses. As they write: "we can do more than generate statistics from birth cohort studies such as the NSHD; we can also recompose persons. Te crux is how we understand data and persons. Recomposition entails scavenging for various (including unrecognised) data, and combining them to generate biographical collages" (Tinkler et al. 2021).

Tese studies provide examples of ways in which individuals, who are usually expected to fade into the background within large-scale studies can be re-confgured or re*fgured* by researchers who have an interest in documenting the experiences of individual cases. Indeed, what makes the cohort studies a compelling resource for this kind of work is that the large sample size makes it possible to select very specifc cases for analysis and to understand those individual lives in context of the much broader sample.

It is noteworthy that historians are prominent in the cadre of researchers who have started to use the cohort studies in this new way. Case studies of individual cohort members provide insights into the past, and their prospective nature means that, in contrast to the use of oral histories, there are fragments of detailed information collected contemporaneously. In these uses, while it is individuals who fgure they are primarily of interest for the insights they provide into the broader historical picture, the experiences of going to a Grammar School or a Secondary Modern School in Post-war Britain, or the diferent opportunities perceived as available for boys and girls. As Tinkler et al. refect, "Recomposition is … interested in the singularity of individuals, it attends too to the historical and relational embeddedness of personhood" (Tinkler et al. 2021). Te particular appeal of these case studies is perhaps that we can gain some sense of the 'big stories' of individual lives. We can look for continuity and change in circumstances over many decades, and we can gain insights into the childhoods and young lives experienced half a century ago.

#### **Big Stories and Small Stories**

Te focus in both conventional multivariate analysis and the relatively recent work on the re-composition of individuals within longitudinal studies leaves us with another question or conundrum—namely what are the best ways of documenting and understanding individuals' more quotidian experiences? Tis question highlights an interest in understanding fgure and ground in a more temporal sense. When we recount our own life stories or compile a CV we focus on key *events*, experiences or transitions—the dates of birth of children, when we changed job, or moved house. Indeed these are also the key pieces of information documented in many longitudinal studies about peoples' lives. It is these events that 'fgure' in our lives against a backdrop or 'ground' of quotidian routine. "Almost by defnition, the quotidian can be overlooked, not actually noticed for much other than for its sameness and its continuities" (Neal and Murji 2015, p. 812). In the second half of this chapter I want to focus on how, and why, we might rehabilitate these daily experiences and place them centre stage, to make them fgure. Habit and routine are central features of our everyday lives, and yet the every day has been largely ignored by the cohort studies.4 Te metaphor of fgure and ground can therefore be applied not just to the contrast between the individual case study and the large sample, but also to our temporal focus. In the analyses of the cohort studies it is life events and key transitions that fgure against the taken for granted ground of everyday experience.

Tere are a few examples of the cohort studies trying to collect some of this mundane and everyday information in the past. For example, journalist David Ward reports that the 1946 Birth Cohort Study recorded that he had '*meat (unspecifed), peas and potatoes (and blancmange for pud) for dinner on 15 June 1950'* (when he was aged 4) (Ward and Payne-Humphries 2013). Indeed, there have been a few isolated and relatively unsuccessful attempts in the cohort studies to collect and analyse a few days of dietary diaries and activity diaries (Crawley and While 1996). However, the burden that this puts on respondents, and the difculty of collecting data in a consistent manner, has led the studies to focus on recording more major life events such as house moves, job changes and births, marriages and deaths. Where there is interest in more regular activities such as exercise, and other forms of leisure or social

<sup>4</sup>As Back (2015) has highlighted, Gofman's thought and empirical work is key to an ethnographic tradition in sociology of attention to everyday life. My focus here is therefore more specifcally on the quantifcation of the quotidian.

participation, the cohort studies have typically relied on standard selfreport retrospective survey techniques (Sacker and Cable 2006).

Tere are some parallels here with the distinction made between big, medium and small stories in work on diferent levels of narrative in the social sciences (Phoenix and Sparkes 2009; Grifn and Phoenix 2016; Back 2015). Te big and medium stories are the accounts that individuals give about aspects of the long durée of their lives, often in response to interview elicitation, whereas the small stories are only likely to occur in conversation and correspond to refections on the everyday and routine aspects of life.

## **Section 2: Opportunities and Challenges for Longitudinal Research Provided by Self-tracking**

Te emergence of new technologies for monitoring and recording daily life at an individual level provides both opportunities and threats to wellestablished longitudinal studies. Wearable devices such as Fitbits, and an increase in techniques and tools for 'self-tracking' or 'personal informatics', now make it more possible to understand, or at least to record, life as it is lived at the quotidian level. Digital self-tracking "has become a mass phenomenon through omnipresent smart phones" (Heyen 2020 p. 124). Self-tracking technologies are marketed as providing insights for the individual user, but they could also be adopted for use in large-scale studies. Digital wearables and associated apps could provide new methods for collecting and recording data that would correspond to some of the small stories of daily life. As will be discussed below, these methods would need to be acceptable to participants to avoid jeopardising continued involvement in longitudinal research. Before exploring the potential use of new technologies for collecting data in the major longitudinal studies, it is worth briefy discussing the growing literature on self-tracking and the 'quantifed self'.

#### **Self-tracking and the 'Quantifed Self'**

Te proportion of those using a smartphone in the UK has risen very rapidly from around 17% to 87% between 2008 and 2020. And it is those in the youngest age groups who are most likely to use a smartphone (99% of those aged 16–24) (Statista 2021). Using data collected in 2016 it was estimated that around a third of internet-connected people worldwide track their health and ftness via an online or mobile app or a wearable device (Herder 2016). In 2017 there were reported to be as many as 325,000 health apps (Research2Guidance 2017). Now that technology to facilitate constant monitoring of all sorts of diferent types of behaviour is so available to individuals—what is the potential for longitudinal research to incorporate this type of information?

Te pace of change makes it difcult to know with any accuracy how many people are engaged with some form of purposeful or 'active' selftracking. Tere will of course also be a spectrum of engagement. While some individuals may occasionally use a form of self-tracking (e.g. a steps counter), others are much more deeply engaged in projects to observe, analyse and change daily habits and behaviour. One manifestation of this is the 'quantifed-self movement', started in California in 2008 by Gary Wolf, which now includes conferences and meet-ups around the world. Te quantifed-self website has the strap line '*self-knowledge through numbers',* and provides numerous resources designed to help individuals understand themselves better, and make changes to their habits and routines.

A clear theme of the quantifed-self movement is that by observing, recording and then analysing their data over time, an individual can gain greater insights, greater control over their life and the ability to improve outcomes. As Heyen (2020) has discussed, using examples from his ethnographic work on self-tracking, "self-related insights are taken into account by the self-tracker in his daily routines … and they contribute, according to his own perception, to his improved well-being" (Heyen 2020, p. 129). Arguably, the individuals who engage in self-tracking are also seeking to distinguish between fgure and ground. Te process of collecting and recording data using wearables and apps helps to discern the aspects of daily life which are most salient for infuencing an outcome of interest. Frequently, the emphasis is on being able to visualise patterns in the data so that the important fgures emerge from the background "noise" of irrelevant measurements (Ruckenstein 2014; Kristensen and Ruckenstein 2018). Tis individual approach to gaining insights typically does not make use of the same principles of statistical inference used in large-scale longitudinal studies. Here the sample size is a single individual (i.e. an n of one), and the logic is that by collecting multiple data points over time and varying diferent factors (usually individual behaviour) clear patterns will emerge from the data. However, while both selftracking practices and longitudinal studies both rely on time, this is framed in very diferent ways. While self-tracking practices rely on a cyclical and repetitive conception of time in order to observe, record and modify behaviour on a daily basis, longitudinal studies in the social sciences rely much more on a linear conception of time. Time, therefore, fgures in rather diferent ways in these two approaches.

Within the growing body of literature on the practicalities, advantages and experiences of self-tracking, questions have been raised about the type of self that is promoted and constituted by these practices. For some, there is potential for these digital practices to constitute a new kind of surveillance, building in normative expectations about appropriate behaviours, sleep patterns, body size, etc. (Lupton 2012; Ruckenstein 2014). Tere is also concern that 'self-knowledge through numbers' as supported by the QS community promotes the model of the ideal neoliberal citizen, that is, the self-monitoring and self-optimising individual who voluntarily aims to control and discipline their everyday behaviour (Lupton 2012; Depper and Howe 2017; Sanders 2017).

In a more optimistic vein, Kristensen and Ruckenstein (2018) use longitudinal engagement with a group of Danish self-trackers to explore the concept of the 'laboratory of the self'. Tey suggest that "Self-trackers use technologies to take the self apart, to highlight certain 'authentic' aspects of it or to intensify human agencies or senses. Tey try out applications and devices: starting of somewhere, learning about themselves and coming out of the experience in another place" (p. 3635). Tis leads to the argument that self-trackers are not necessarily dupes skilfully cajoled into digital consumption and constant utilitarian self-improvement. Rather Kristensen and Ruckenstein provide evidence of refexive individuals whose engagement with personal informatics makes them more attuned to the emergent properties of the self and enables them to be more conscious of their "agentic aims and powers" (p. 3631).

Tis explicit examination of the nature of the self that is promoted via self-tracking is echoed in the works of Rapp and Tirassa (2017). Teir focus is on how we might try to improve the technologies that enable personal informatics in order to go beyond the rather 'utilitarian self' of the quantifed-self movement. Contrary to Kristensen and Ruckenstein, Rapp and Tirassa argue that the self currently implicit here is the self of behaviourist psychology: a self that is ultimately unknowable and therefore under-theorised. Tis can result in a self that appears to consist only of a set of observable behaviours reduced to data points (Armstrong 2019). Tis perspective on self-tracking suggests that what actually changes is not the self but the behaviour or indeed the visible (or measureable) body.

By invoking the phenomenological, subjective self as a far more interesting object for study, Rapp and Tirassa prompt an exploration of how personal informatics could be developed to allow individuals to engage much more fully with this subjective, experiencing self. Using the framework of four aspects of the phenomenological self (the past, present, future and interrelated self), they proceed to sketch a research agenda and set of guidelines. Key to their argument is that technologies should be developed in a way that transcends the focus on behaviour change and allows for a more thoroughgoing refection on the self, one that foregrounds the importance of both context and environment. Teir work, therefore, resists the pressure for us to become neoliberal subjects who "are constantly encouraged to change their habits – rather than society and institutions – in order to become happier more productive people" (Chun 2016). It is also noteworthy that their four aspects of the phenomenological self move the focus from the cyclical time of habit and routine and place the individual more clearly in linear time.

Rapp and Tirassa (2017) and Kristensen and Ruckenstein (2018) clearly start from divergent perspectives on the conceptualisations of the self promoted by personal informatics practices. However, both sets of scholars provide us with the promise of re-fguring the self from being constituted only by behaviour to being fully subjective, refexive and focused on interaction with the environment. Tis suggests that personal informatics could have a "grander ambition" and develop "the capability of revealing something of the individual's self" (Rapp and Tirassa 2017 p. 340).

#### **The Potential Use of Self-tracking in Longitudinal Studies (Figure and Ground)**

For any long-term longitudinal study, there is likely to be a tension between making maximum use of innovative data collection techniques and maintaining consistency to ensure that the longitudinal design of the studies can be exploited to the full using appropriate statistical analyses.

Arguably, if there were enough resources, then it would be possible both to preserve and to add in new data collection strategies that make use of emerging digital technology. However, those running the studies also need to be cognisant of the burden that data collection places on cohort members. Te longitudinal studies have been able to maintain extremely high response rates due to the loyalty of cohort members, who have participated since childhood (Mostafa et al. 2020). Tere is understandably a fear that introducing new forms of data collection may alienate long-term respondents and compromise the quality of the studies for future researchers.

Part of the problem here is the relative lack of research to date on the ways that individuals routinely use digital devices in their daily lives. Tere are the beginnings of a body of research on individuals' self-tracking (Nafus 2014; Ruckenstein 2014; Ajana 2020; Heyen 2020; Lupton 2020). However, the more active and engaged individuals who constitute the Quantifed-Self movement are still only a tiny percentage of the population.5 Tis means that it is difcult to assess the potential for using

<sup>5</sup> In a personal communication via the Quantifed-Self forum, Gary Wolf reported in January 2021 that 'We have about 6500 users with accounts. I estimate that we've had between 20,000 - 40,000 individuals interacting in person or online in all our formats over the last decade. (Tat means posting, registering, or attending in person, not merely viewing a web page.)' https://forum.quantifedself.com/t/forum-stats/8448/7.

digital recording and tracking methods in a representative sample of British cohort members.

Recently, to address this issue, the directors of the cohort studies commissioned qualitative research to assess the acceptability to cohort members of using innovative methods to collect new types of data (Ipsos 2019). During 2019, interviews were conducted with samples of 28 individuals from each of 4 cohort studies (i.e. a total of 112 interviews), complemented by a focus group discussion from each cohort. Key questions included how cohort members would feel about providing access to their social media activity, their travel (as automatically recorded via travel cards) and their fnancial transactions (using a specially designed app). Interviews and focus groups also covered the more general use of new technologies such as apps to actively or passively collect detailed data including screen time, GPS and activity tracking.

Despite the strong loyalty of cohort members to the longitudinal studies, it was striking that across the interviews and focus groups, study members *consistently* reported that novel data collection felt like a form of surveillance and therefore regarded it with unease. Some of the comments included:

*Te more that the study moves towards big brother tracking, I would struggle with it and may withdraw from the study. (BCS70, telephone interview, 83, did not take part at age 46) (p. 87 Ipsos Mori, 2019)*

*I wouldn't like to do any of it it's too personal, too private that feels like big brother is watching me. (BCS70, telephone interview, 103, took part at age 46) (p. 85, Ipsos Mori 2019)*

*No, I wouldn't agree to any of that… I know they always say: 'big brother knows where you are' and I'm sure somebody does but I don't want to have all these apps and things to make it even more. I'm not interested in any of that, no. (NCDS, telephone interview, 10, took part at age 55) (p. 85, Ipsos Mori 2019)*

Tere was a more positive reaction to the idea of collecting exercise data using a Fitbit, or similar wearable device, as this was seen by cohort members to be directly linked to health research and therefore an acceptable part of the study.

*I feel that is a diference as it can show the study how many steps I have taken and how many calories I have burned then yes as it was just health focus which is important rather than how long I have spent checking the weather on my phone. (BCS70, telephone interview, 103, took part at age 46) (p. 86, Ipsos Mori 2019)*

It is interesting that in the frst three of these quotations the cohort members each invoke the fctional 'Big Brother', originally conceptualised within Orwell's dystopian novel *1984* (and then popularised by the reality TV series). Tis is a ready shorthand for surveillance that covers the most private and seemingly inconsequential activities of life. Here then we see a contrast between individuals being uncomfortable with tracking of everyday habits and experiences that seem to have no readily understandable beneft for research, while there is an acceptance that monitoring the body—calories input and expended—can have a value for understanding and improving health.

Tere are aspects of the major longitudinal studies that now capitalise on the use of the web and personal computers to simplify data collection (e.g. the age 62 sweep of the 1958 cohort is collecting a dietary diary using the web). However, no extensive use is being made as yet of wearable devices or the ability of smartphones to prompt the user to report on activities over the course of a day. Tis means that consistency is maintained. However, what remains missing from the detailed quantitative linear chronicles of longitudinal studies is a feel for the daily lives and everyday practices of cohort members—how much time they spend commuting, working, watching television, out with friends or asleep; how many steps they take; how their heart rate varies over the course of a day, whether they eat three meals or multiple snacks; etc. Tis description of what is missing is not to diminish the value of the rich data of the cohort studies but rather to serve as a reminder that they provide only a partial picture of individuals' lives. Tey foreground linear time, and it is this which fgures against assumed, but invisible, daily experiences. As Back has argued, "the everyday matters because it ofers the ability to link the smallest story to the largest social transformation" (Back 2015 p. 834)

In an article for the *NY Times* magazine, Gary Wolf, a co-founder of the Quantifed-Self movement, wrote that:

We track ourselves all the time, but something changes when we digitize this self-monitoring … when the familiar pen-and-paper methods of selfanalysis are enhanced by sensors that monitor our behaviour automatically, the process of self-tracking becomes both more alluring and more meaningful. Automated sensors do more than give us facts; they also remind us that our ordinary behaviour contains obscure quantitative signals that can be used to inform our behaviour, once we learn to read them. (Wolf 2010)

Tis desire to adopt methods which allow patterns to surface from the background noise of data and to fgure out what is meaningful, once again returns us to the metaphor of fgure and ground. We want to believe that there is more to life than random noise and that meaningful patterns will emerge if we only have the tools and patience to be able to observe what is really there.

Tere are also some interesting parallels here between the promise or 'allure' for individuals that once we fully understand ourselves we will be able to improve our lives and our well-being, and the promise of the cohort studies whose overriding aim has always been to provide policy insights that will improve the lives in the aggregate, especially for disadvantaged groups within society. As Ferri et al. wrote in conclusion to their 2003 book on the cohort studies:

To gain a proper understanding of what (policies are) likely to be most efective, when and with whom, we need much more research on the mechanisms and processes of success and failure in an increasingly complex changing world. Investigation of the interactions of the efects of social change with the development of individual lives will continue to drive research using the cohort study data in the years to come. (Ferri et al. 2003: p. 312)

Whereas the power of the cohort studies lies in the large sample size as well as the length of observation, for individuals using digital methods to track and record their behaviour the sample size is an n of one. Both approaches hope to be able to discern meaningful 'patterns' from among the background noise of a superfuity of data points. Both, therefore, use methods of analysis that will enable the fgure to be distinguished from the ground. What is also shared here is the possibility of collecting data *over time* and observing how change in one domain impacts on outcomes in another. However, as discussed above, there is a sense in which the conception of time is subtly diferent in the two approaches. In largescale longitudinal studies, the emphasis is on linear time with cyclical time assumed, but relatively obscured in the background. In contrast, the process of self-tracking has tended to interrogate habits embedded in cyclical time, "practices acquired through time that are seemingly forgotten as they move from the voluntary to the involuntary, the conscious to the automatic" (Chun 2016, p. 6).

# **Conclusions**

Tis chapter has explored the ways in which individuals can be made to appear, or disappear, in longitudinal research, whether that is in largescale cohort studies or in recent work on personal informatics. Invoking the metaphor of fgure and ground raises the question of what counts as the 'ground' that is, what is the backdrop or context against which the subjects of research (i.e. the fgures) can be made to appear or disappear, and to what extent does that context actually serve to constitute the fgure itself. When focusing on large-scale, quantitative and structured cohort studies, the backdrop or ground can be understood to be both the large representative sample that frames and makes sense of each individual's set of unique data points, and the historical and geographic context. Indeed this methodological approach to understanding individual lives is already well-rehearsed within the literature on the Life Course (Giele and Elder 1998). And this literature draws attention to the way in which historical events, such as the Great Depression, not only provide a backdrop to a life but actively constitute the experience and subjectivities of each individual.

What is key in large-scale quantitative research is that, paradoxically, in order to focus on understanding the factors that may infuence *individual* outcomes the individual research subjects are efectively removed from sight. Although each individual contributes myriad data points, their data is deliberately anonymised. It is the researcher and not the research subject who crafts causal narratives. Using multivariate, and sometimes multilevel, statistical modelling techniques, variables and coefcients appear to have agency, that is, these are the fgures of interest here. Even innovative case-study approaches that have sought to refocus attention on individuals rarely seek ultimately to foreground the individual but rather to develop deeper understandings of causal process or historical context and change.

Te second half of this chapter shifted attention to the implications for longitudinal research of the increase in self-tracking practices and personal informatics. While these activities, with an n of 1, appear to put the individual centre stage it is still instructive to consider what constitutes fgure and ground in this novel approach to 'personal science' (Heyen 2020). For an individual self-tracker looking for patterns in their data over time, the ground is perhaps those aspects of individual experience and behaviour found *not* to be relevant for achieving the outcome of interest; whether this is improved ftness, attention, sleep patterns, or wellbeing. Te practice of self-tracking is motivated by a belief that with the right tools and techniques it will be possible to discern the meaningful patterns in the data, to fgure out what matters and to adjust behaviour accordingly.

What the burgeoning literature on personal informatics often neglects however is a deeper or more explicit theory of what constitutes the self (Rapp and Tirassa 2017; Kristensen and Ruckenstein 2018). Arguably if the data points, collected by and on an individual, are no more than representations of behaviour, then the self becomes no more than the coordinator of that behaviour. Such a self would arguably be completely uninteresting and one dimensional if it were not for two narrative elements, the ability to infer causal links from the quotidian data observed, recorded and visualised in cyclical time, but also the possibility for change over linear time. In this context, narrative serves to vivify data points and constitute a self that is traceable over time *and* can change over time in a way that can be meaningfully understood. In seeking to fgure out the individual in longitudinal research, we therefore need to attend to more than the contrast between (or mutual constitution of) fgure and ground, but their mutual constitution in cyclical and linear time. Perhaps the greatest challenge for the future is how to make best use of new technologies for data collection while also considering how to place a thoroughgoing subjective, or phenomenological, self at the centre of our research narratives.

**Acknowledgements** Many thanks to JD Carpentieri and Jon Lawrence for their helpful comments on earlier versions of this chapter. Special thanks to the editors of this collection for hosting the conference that initially sparked my interest in fgures and fguring, and specifcally to Celia Lury for very helpful comments on my various drafts.

# **References**


*Epidemiology & Community Health* 50 (3): 306–12. https://doi.org/10.1136/ jech.50.3.306.


Health and Development)'. *International Journal of Epidemiology* 35 (1): 49–54. https://doi.org/10.1093/ije/dyi201.


**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/ licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **10**

# **Figuring Out Exposure: Exploring Computational Environments and Personalisation in Interdisciplinary Air Pollution Research**

**Emma Garnett and Srishti Bhatnagar**

# **Introduction**

Te epidemic of asthma and other non-communicable diseases triggered by air pollution has fnally placed the environment frmly under the purview of global public health (WHO 2018). Clare Herrick (2020) argues this is an opportunity to shift the optics of health away from behavioural patterns of consumption and towards the complex causalities of toxic exposure. Yet, seductive ideas of technical fxes, awareness-raising and reifed models of behaviour change continue to persist in policy and public debates. Tis is a conceptual and methodological challenge that requires fnding new ways of rendering sensible the variegated, interconnected

E. Garnett (\*)

Social Science and Urban Public Health Institute, King's College London, London, UK e-mail: emma.garnett@kcl.ac.uk

S. Bhatnagar Delhi, India

and inequitable geographies of environmental exposures and harm (Fortun 2012; Sultana 2021). Rather than producing more data, calls for 'smarter' sensing focus pollution monitoring in places of concern, hotspots and in relation to everyday practices like walking or cooking (Reis et al. 2015), often through participatory and experimental approaches (Lezaun et al. 2017). In this chapter, we discuss an interdisciplinary project that is taking up this task by combining computational methods and embodied data to simultaneously map, know and respond to air pollution. In the project, air pollution is fgured through a composition of data practices that include the generation of exposure data by people with asthma. We use the concept and method of fgure to explore the tensions that emerge when individuals are both objects and subjects of research. By involving the experience of people afected by air pollution, the project engages with a key concern in public health research and practice of how to best identify exposure risks and generate knowledge that can efectively inform action.

Te efects of processes of computation and data science on daily life are being critically engaged with by researchers working in and across the domains of health and medicine (e.g., Day et al. 2017; Prainsack 2017; Radhakrishnan 2021; Ruckenstein et al. 2017), government and welfare services (e.g., Eubanks 2018: Khera 2019), cities and urban planning (e.g., Duarte and Priyanka deSouza 2020; Mattern 2017; Tironi and Sánchez Criado 2015), among many others. Sensing technologies in particular have been described as "a new extension of social control" and "a site where alternative modalities of power are being forged" (Nafus 2016: xiii). In their ethnographic work of a national programme for smart homes in Chile, Tironi and Valderrama (2021) detail the various explanatory logics of domestic sensors provided by government ofcials, technicians and the residents invited to monitor and quantify their energy use as part of a national programme. According to the authors, a central aim of the initiative is to improve the environmental performance of homes by no longer relying on what people say they do, but on what they actually do: participatory and live information is produced through continuous and recursive feedback from sensors installed in people's homes (2021: 194). In this example of "sensing governmentality" (Tironi and Valderrama 2021), *personalisation* is not understood "in the sense of users shaping technologies within their own practice but as technologies that recognise their users and shape themselves accordingly" (Suchman 2012: 222). It is this 'intelligent' or 'aware' feature of sensing that concerns us in this chapter because it invites new interdisciplinary ways of doing environmental health research. To engage with these developments, we draw on social science and humanities studies of digital technologies that aim to extend the human body's 'innate' capacities through informational means (Creager 2018; Viseu 2003; Viseu and Suchman 2010). We are going to focus specifcally on the use of wearable technologies in the interdisciplinary feld of air quality and exposure science, a research practice and approach that is increasingly common. By encouraging personal exposure monitoring, new and diferent relations between people and air are made available for computational calculation.

Tis chapter is based on ethnographic research of a UK-India funded research project using wearable sensing technologies to generate individualised data on air pollution in New Delhi, India (2018–2020).1 We were both researchers on this project, Emma Garnett from the UK and Srishti Bhatnagar from India, and involved in some aspects of the feldwork led by a multi-sited, interdisciplinary team of senior co-investigators in computer science, public health, digital design and sociology (based in a variety of institutional locations across the UK and India).2 For Emma Garnett, the collaboration formed a component of a separate postdoctoral project examining air pollution sensing technologies in biomedical and public health research.3 Te project in Delhi is an interesting interdisciplinary case study because of its inclusion of social science research as a central work package, which is rather novel in air quality and exposure science. As an international research project, it also represents an aspiration in science, and often research led by the Global North, to expand data coverage of air pollution by monitoring 'hyper-local' or 'micro-environments' in cities in the Global South, and so encapsulates the geopolitical relations and historical legacies of public health, science, and biomedicine.

In the sensing project we discuss here, the project team also sought to include the experiences of people who bear the brunt of the health costs

<sup>1</sup>Funded by a Global Challenges Research Fund grant.

<sup>2</sup>Tese institutional locations will remain anonymous in this chapter.

<sup>3</sup>Funded by an ESRC New Investigator Award: ES/R008612/1. Te study sought ethical approval from King's College London Ethics Ref: LRS-18/19-10426

of air pollution and/or who might not be in a position to necessarily "*claim* clean air" (Negi and Srigyan 2021: 63, emphasis included): children with asthma. We conceptualise 'the child with asthma' as a key fgure in contemporary air quality science and public health. It is a fgure that brings together a concern to improve public health interventions through more granular measurements of air pollution exposure and its efects on the body by focussing on the people and communities who bear the greatest health burden. As a method, we use the fgure of 'the child with asthma' to explore the tensions that emerge when involving people understood to be impacted by air pollution as both objects (sensing bodies that measure air pollution) and subjects (knowing bodies that experience and respond to environmental exposures) of research.

### **'Person-Centred Environments'**

Te concept of 'person-centred environments' (PCEs) was coined and presented by computer scientists as a novel methodology that incorporates clinical knowledge, big data practices and analytics to track and quantify the embodied efects of air pollution. It can be understood as a practice of personalisation, in which the individual person is the object of study: the person wearing the sensor produces a radius of data points that quantifes exposure 'inside the body' through the linking of air pollution measurements, environmental variables, breathing rate and other vital responses. Tey are also subjects of research because self-produced geo-time stamped data and the self-reporting of daily trajectories are also considered important for understanding exposure. People participating in this study help provide information about social, environmental and genetic contexts from which multiple data points can be generated about air pollution's health efects (e.g., a measure of air pollution and a quantifed bodily response). Our ethnographic research studied this methodological approach in action and analysed how it infuences the way in which the problem of air pollution is known and understood.

Recent evidence shows there is no safe level of exposure to many pollutants (Kelly and Fussel 2015). Tis challenge has ethical implications because if the spatial demarcations between healthy and unhealthy air are unclear, then eforts to improve health in an equitable manner are constrained. Te practice of personalisation in scientifc research is a response to this challenge because it enables new lines of diference to be stratifed along social and biological lines. In our case of PCEs, new groupings of people in relation to air pollution are based on molecular profles of disease and knowledge of the biological relationship between air pollution and asthma. If the 'sensitivity' of children with asthma to air pollution is known, then how they respond to diferent environmental settings and circumstances is turned into something to be further studied and quantifed. In terms of public health eforts to improve health, personalisation articulates and locates exposure and risk in ways that have consequences on how the actions and responsibility to improve it are framed. Asthma and the anticipated diferences in the air that it reveals in human bodies is one way in which new normativities and mechanisms of public health are being sought. As Lury and Day (2019) argue, personalisation is always a process of generalisation because entities (such as cancer, but also, as in this case, air pollution's efects) are specifed through the sorting and arranging (drawing lines of inclusion and exclusion) of classes of persons (e.g., people with asthma and children).

Te body of research on asthma is huge. It remains a major research topic because it is among the most prevalent non-communicable diseases in childhood, and early life exposures to air pollution play a role in determining the disease in later life (Te Global Asthma Report 2018). In our research, we found that focusing on the living environments of people with asthma was used as a way of starting to explain uncertainties in knowledge and the complex and non-linear causalities of air pollution and health (Fortun et al. 2014; Kenner 2018). Te conceptual starting point of PCEs is that bodies are permeable, particularly those with asthma, and that living in polluted environments, like areas of urban Delhi, produces measurable perturbations on individuals' internal physiologies. Te hypothesis that follows is that these perturbations are also likely to be modifed by other contextual factors, such as the built environment or variables related to socio-economic status. Tis way of accounting for the environment envisions a neat and all-encompassing way of translating incommensurable phenomena into actionable knowledge. Identifying these kinds of explanatory logics provides a lens from which the politics of environmental health shaping the interdisciplinary intersections of a 'global' air quality science and computational public health can be investigated (Landecker 2011).

Linking bodies to environments in this way turns 'vulnerable bodies' into objects of knowledge through which signifcant infuences of 'bad' environments can be measured in relation to a normative 'good' (Choksey 2021). In the case of this project, analytical algorithms are being developed to extract relations of interest from the material contexts of exposure (Amoore and Piotukh 2015). However, by trying to establish more precise measurements of air pollution, computational methods also establish which individual, environmental, biological and social pathways are relevant (Lury and Day 2019: 19). Te process of fguring out exposure in this way relies on setting aside some aspects of a person's surroundings. As we will go on to show, the fgure of the child with asthma was put to work in order to distinguish the environmental relations that matter for public health universally. In doing so, we show how personalisation and the 'optimisation' it promises can serve to limit the possibilities of situated actions and solutions. Advice to reduce pollution and exposure remains central to public health eforts, and more accurate data is often imagined as a tool to aid everyday decision-making. Yet the options available for people to avoid exposure or clean their breathing spaces are not equally shared in and across societies. We found that this conundrum is also embodied in the fgure of 'the child with asthma,' which we use as a method and concept for unpacking how PCEs confgure sociality and health.

Having introduced PCEs, in what follows we provide a brief overview of the wearable sensors involved in the study and how they work. We then describe the project's interdisciplinary methods and the ways in which they provided a research context for our own ethnographic study of embodied data practices of personalisation. Following this, we detail two occasions in the early stages of the data collection in Delhi in 2019. Trough the fgure of 'the child with asthma,' we explore the tensions between participants as objects and subjects of research and how this relates to the methods of interdisciplinary research.

## **'Experimental Entanglements': The Wearable Sensors Study**

Te concept of PCE is integral to the interdisciplinary design of the project in Delhi, which was led by computer scientists in the UK. Te wearable sensors used were developed in the UK and deployed in India to explore their potential as an alternative method of monitoring air pollution for public health. Te intended participants were children medically recognised to be living with asthma. Te protocol for using wearable sensors was adapted from another project led by one of the teams in the UK and which, until COVID-19, continued to be used and modifed for related air pollution studies globally. Tis kind of fexibility in the conduct of monitoring projects is characteristic of global air pollution experiments, in which sensing and monitoring infrastructures are imagined as replicable in diferent places. It is an approach that reinforces a model of research shaped by a colonial legacy of technoscientifc solutions being led by institutions in the Global North and applied to Global South settings. Tis is particularly signifcant to note given there are numerous air quality projects in Delhi that are led by Indian scholars and institutions (including projects involving our colleagues based in India and the UK) and are also exploring alternative systems of sensing, monitoring and managing the air (Gani et al. 2022). Indeed, sensors have helped facilitate new and innovative formats for doing science, open data formats and participation in cities globally, and are therefore objects that highlight (and challenge) the power relations and systems of privilege that structure the fow of knowledge, people and resources in a global air quality science (Negi and Srigyan 2021). Indeed, Emma Garnett was the only core team member who had not conducted research in India before but took on a signifcant project role.

Our main focus in this paper is on the practical work of generating data, because it allows us to better understand the end goal of research and how this might difer in an interdisciplinary research team. Te wearable sensor technologies designed by computer scientists record personal exposure to outdoor and indoor air pollution. It was a clinical study design. Each participant in the study was provided with a set of air quality monitors that measure PM2.5, a heterogeneous pollutant made up of microscopic solid or liquid airborne matter. It is also a pollutant that is commonly used in air quality and exposure science as a proxy measure for human health risk. Te wearable sensors consist of a GPSenabled wearable sensor designed to record physiological changes (for instance, the participant's breathing rate) and track movements. As well as mapping data at a more granular scale than is common in studies of air pollution, like other lower-cost sensing studies, the aim was to also pick up local sources of pollution and aspects of urban environments that mediate exposure (e.g., Hagan et al. 2019). Te devices were packed in small sling-bags to carry the power adapter and phone that provide access to an app interface sharing the sensor readings. Te breathing sensors were worn directly on the body. A clinical protocol using the molecular profles of asthma was adapted to monitor the infuence of exposures on children with asthma who are known to be susceptible to air pollution's toxic efects (see, e.g., Vardoulakis and Osborne 2018). Participants were instructed to wear these sensors for a 48-hour period and to continue with their normal daily routines.

In addition to the sensor data and qualitative interviews, the project's data collection strategy included a survey covering demographic details of each individual (age, educational status, class, caste, religion), the composition of family and household, medical history and asthma management. Other sets of questions sought to capture relevant features of the locality of the participant, such as fuel use and housing quality (e.g., ventilation). A participatory workshop led by digital designers formed the fnal project work package and was explicitly tasked with translating the qualitative and quantitative data of exposure into impactful knowledge and outcomes. Te aim of the workshop was to develop personal stories about air pollution with the same participants who wore the sensors and to then develop these into a series of public-facing animated memes. Although air pollution is routinely fgured through numbers, other ways of linking these fgures and quantifed accounts of air pollution through artistic and creative methods are increasingly recognised as important by researchers and policy makers. Te interdisciplinary design of this project was therefore arranged in such a way that alternative, personal narratives of air pollution might be told.

We will now examine and unpick two moments of friction in the interdisciplinary team that relate to the air pollution sensing data practices. Tese tensions emerged in part because of the refexive social science involvement in the study. However, they were also practical problems that are generative to think with because they articulate the specifc occlusions of personalising air pollution—particularly in relation to understanding systems of marginalisation. Te research in this chapter is based on ethnography of the research process, including our involvement in fortnightly team meetings, the conduct of interviews (by Srishti Bhatnagar) and facilitation of a workshop. By tracing how the interdisciplinary methods unfolded and played out in practice, we show that the combination of methods in the project generated very diferent kinds of data about air pollution that led to epistemological and ontological frictions which support refexivity (Garnett 2017). Trough a discussion of the experimental entanglements (Fitzgerald and Callard 2015) animated in PCEs, we delineate some of the fraught intentions and ambitions the fgure of 'the child with asthma' produced (cf. Murphy 2017: 82).

#### **Confguring Environmental Health 'Pathways'**

Te design and implementation of wearable sensors that materialise a PCE and personalised model of exposure require two key features: (i) the participation of patients with asthma willing to measure their exposure as part of the research study and (ii) developing interpretive frameworks from which to determine signifcant environmental infuences and thereby predictors of health. At the time of writing this chapter, the computer scientists on the project are sorting through the large amounts of data generated by the study to identify signifcant patterns and relationships. Here, however, we focus on the practical work involved in confguring a computational research environment from which data practices of personalisation in air pollution research could be conducted. We highlight how personalisation in PCEs relies on the embodied practices of sensing, in which wearing the sensors continually enfolds possible environmental (contextual) infuences through real-time measurements. It was anticipated by the various researchers who contributed to the design of the research of PCEs (this research design is part of a longer-term vision and approach) that the monitored body contains measurable responses to air pollution. Tis starting point is based on evidence of the genetic and environmental factors that modulate susceptibility and response to air pollution. We draw on Lury and Day's (2019) notion of "pathways" to account for how the increased sensitivity of people with asthma presumed an individualised response is measurable. By measuring exposure inside and outside the body, these already well-evidenced exposure pathways served as a background from which individualised health efects of air pollution could be studied in relation to public health. Put another way, the specifc gene-environment pathways were not the object of study, rather they facilitated claims to personalisation because the individualised data could simultaneously refer to a generic classifcation, in this case of vulnerable groups or 'at risk' groups.

### **Negotiating Participation in Research**

A core feature of the project was the recruitment of children with asthma to participate in monitoring their exposure. We were variously involved in project discussions of the recruitment process, working closely with colleagues in public health, liaising with doctors and health professionals providing care for asthma out-patients and speaking to parents and children about their involvement in the research. Early on in the recruitment process, several senior investigators in the team expressed concern that the individuals and communities most likely to sufer high levels of air pollution and associated health burdens could be missed because of the study's protocol. It was argued that there are many young people who have asthma or asthma symptoms but are living without an asthma diagnosis. Te implication was that by only recruiting via government and private hospital out-patient lists the project would likely fail to record a range of social and health experiences of air pollution. Following this discussion, one of the co-investigators contacted an environmental NGO working closely with people living in areas of Delhi that experience high levels of air pollution because of well-known emission sites nearby. Te aim of making initial contact was to explore the possibility of the project potentially extending the recruitment process beyond formal medical centres in order to cover a range of urban areas. Te NGO agreed to help and a couple of weeks later the research team working in Delhi (including the authors) was introduced to around 60 children attending an afterschool club. During this meeting, the sociologists and public health researchers encouraged introductions, provided an overview of the project and initiated a lively conversation about air pollution and its health impacts.

Unsurprisingly to some of the team, no one claimed to be experiencing breathing difculties and only a few reported occasional symptoms of asthma-like conditions. But this fnding troubled others in the team. Te feld note extract below recounts discussions in the weekly project meeting following the visit, in which some of the difculties involved in 'fnding people with asthma' outside of biomedical health settings were identifed:

Te team discusses the previous week's feld visit to an after-school club in North Delhi run by [the NGO] and acknowledges how supportive they have been to the project. Tose in the meeting who joined the visit assured the rest of the team that the children who attended the discussion were from a 'low-income group' with limited access to medical care [a criterion previously agreed for the three diferent socio-economic classifcations prescribed by the project protocol]. Te 'happy news' is that very few children attending the after-school club reported experiencing breathing difculties. Despite living in an area often considered as sufering high air pollution (near waste sites with open burning, close to construction sites and vehicle pollution) the surrounding environmental conditions did not correlate with the children's accounts of their own health. (Fieldnotes 19 February 2019)

Te meeting discussion threw up two rudimentary fndings that were hard to reconcile with the project's protocol. First, air pollution and asthma are experienced as diferent phenomena with diferent social meanings and cultural associations. Second, the causes and experiences of health were not the same as biomedical interpretations of risk and harm in global health (Das 2015). Te fgure of 'the child with asthma' brought into focus an interdisciplinary tension concerning the best way to bring the problem of air pollution in relation to the problem of asthma so as not to occlude social marginalisation.

For the computer scientists, 'the child with asthma' was primarily a context from which air pollution could be studied anew. Participation was exclusionary, involving only people with asthma because diagnosis is a requirement for identifying 'a personal pathway': the strong evidence base of air pollution's impacts on asthma meant 'relations' were understood to be largely imperceptible but 'out there' and therefore ready to be measured. For them, the relations opened up the possibility for air pollution to be understood in 'novel' ways. Te identifcation of children with asthma was thus a necessary part of the project. In a conversation with one of the computer scientists, we were told that the breathing sensors do not have the sensitivity to pick up physiological responses to air pollution in a person without asthma. A person could only be fgured as an object of research in PCEs through their capacity to reveal infuences and contain potential correlations. In order to make air pollution perceptible in the big data sets of computational air quality science the person has to be 'cut out' (cf Amoore and Piotukh 2015) or provisionally produced. Tat is, personalising air pollution relies on the overdetermination of asthma which is then able to assume a causal relationship rather than a 'merely' contextual one (Sunder Rajan 2012: 6). Tis determination helps the computer scientists manage the sheer amount and complexity of environmental exposures, yet it also sets the stage within which only some forms of health and sociality can take shape.

As was recognised by the project team, identifying the environmental factors that trigger asthma or exacerbate symptoms by relying exclusively on biomedical categorisation resulted in overlooking some of the socalled upstream determinants of health. By upstream determinants we are pointing to what Nancy Krieger calls "the causes of causes," in which the conjoining of "power over" and "power to do" structures people's exposure to material and environmental health hazards (2008: 223). Setting criteria and exclusions is part of knowledge-making, but it is nonetheless signifcant that in this situation the criteria of asthma created a contradiction because it excluded the possibility of including the experiences of those often excluded from discussions about what to do about air pollution (children, marginalised social groups). Friction in the interdisciplinary team meant that the status of asthma as a biomedical category (a ground or context) for fguring out air pollution's efects was contested, and the uneven geographies and unequal impacts of exposure difcult to approach as a problem of social justice *and* public health in the research.

#### **(Personal) Data Frictions**

In a second example of friction, in a rather tense team meeting during the frst data campaign in summer 2019, the descriptions "add[itional]" and, later, "complement[ary]" were used by one of the senior scientifc investigators to describe the role of the social research data in the project. In this framing, the narratives generated by the interviews were understood as providing supplementary variables of use only for interpreting or validating the quantitative sensor data. Tis description of the potential value of the data was refuted by several of the team involved in the qualitative research. It was argued that the narrative histories of asthma, practices of care and "the intimate knowledges of air through their bodies and their bodies through air" (Negi 2020: 20) generated in the interviews should also count as knowledge and evidence about air pollution.

Tis diference in valuation was in part structured by a temporal out of jointness (Fitzgerald and Callard 2015) between the "in real time" data of PCEs and the non-linear temporalities that characterised the qualitative assessments of living with asthma. In the quantitative data, the person is fxed in space and time, and the person only comes into view through analytical processes where the co-occurrence of particular data elements gives rise to them (Amoore and Piotukh 2015: 354). However, these data cannot necessarily account for other temporalities such as, for example, cycles of stress (although this is a known trigger for asthma) nor contingencies in everyday forms of decision-making. Tensions concerning the meaning of data are the result of disciplinary and epistemic hierarchies. Tey also resonate with what has been written about the practical challenges of developing explanatory and interpretive frameworks for making sense of environmental data in the felds of postgenomic science and informatics (Richardson and Hallam 2015; Prainsack 2017). By reducing the environment to a series of compartmentalised variables, the 'sensing body' (in our case study) supports a puzzle-solving process in which social and material relations of exposure are assembled into something comprehensible. Tis is aided by gene-environment links to established biological mechanisms of asthma, which seems to be setting the parameters for exploring new ways of thinking, working and explaining air pollution (cf Richardson and Hallam 2015: 234).

Ethnographic insights from other research emphasise the importance of the patient's work of participation in scientifc eforts to fll "data gaps" (Prainsack 2017: 24). In our case, the ways in which sensors were worn and mobilised, or not, also frequently undermined assumptions about the person as a unit of analysis in PCEs. Together these diferences concerning data raised questions about who benefts from an expansive and "inclusive" approach to air pollution monitoring? When discussing their experiences of wearing the sensors and participating in the research, participants very often reported not wearing the devices as self-tracking devices. Many children told us they were wary of taking the sensor outside in case it was misplaced or broken. Te parents of others explained that they were concerned about what some people might say about their child if they wore the sensors outside the home. Tese issues either came as a surprise for some of the computer scientists in the team or were framed as something that needed to be overcome. Tese interactions with the sensor are also a way to understand the diferentiated experience of air pollution and disease, as we will continue to explore.

Eforts to evoke an 'in real time' response were hindered because it was not always possible, for those participants who were interested, to read the aggregated data on the app interface (although fnal reports were provided for each family by the team). Yet it was precisely this data, generated through sensing in everyday practices, like socialising outdoors or playing with friends in neighbouring areas, that the computer scientists wanted to understand to specify individual exposure pathways and, indeed, the kinds of data we thought might be interesting to explore with our social science colleagues further. In these moments the interviews were practically if not epistemologically valuable to the computer scientists because they aided the preliminary screening and interpretation of sensor data. For instance, the interviews and feldwork helped to provide an explanation for static rather than mobile data collection, why there were long disruptions in data collection (as when families faced connection issues) or why there were sometimes two data sets for one individual (when a participant wanted to try the sensor out again). Although these errors could be managed through statistical technique, they also provide a juncture for attending to the interpretive processes of data practices of personalisation.

Indeed, the challenges of generating personal data are illustrative of the ways in which persons in PCEs always also stand in relation to other personal and nonpersonal environments of exposure (Lamoreaux 2016). Tese tensions encourage a shift in analytical focus, away from what makes environments 'bad' to how they are made 'good' by participants themselves. Te interview transcripts collated a number of creative ways young people dealt with events that aggravated their asthma: not running around by playing batsman rather than a feld position in games of cricket; working extra hard to overcome missing days from school; by trying and testing medicines to see what works in diferent situations; and tracking causes of symptoms and what improves them. As demonstrated in the reference to the team meeting discussing the role of qualitative data, the PCEs gestured towards other possible experimental entanglements where the object of research and its aims could be questioned by its subjects.

To further explore this point of unsettling the relations of subjects and objects we also want to consider a children's workshop that was hosted by one of the collaborating institutions in North Delhi on a Saturday in early May 2019. It was originally intended as a knowledge translation activity, however, due to various logistical challenges, the workshop ended up taking place as the frst data campaign was just getting started and therefore intervened in the conduct of the research rather than simply facilitating dissemination. It brought together the research team, parents and children, including the children who we met at the after-school club. Te clinical protocol was put to one side and therefore who could participate was not predetermined beyond their being a child or young person.

Te workshop sought to visualise personal, embodied narratives by exploring experiences of air pollution collectively, as this description of the process explains:

I mean that's how it was also planned [to use diferent artistic mediums to develop the narratives] but giving it fexibility, that is also how it worked, that you start frst with just playing with the idea of pollution. How can we experience it? What are the words to be able to experience it? Ten what are the colours to be able to experience it? And then we moved to this idea of, like, a character who is experiencing pollution; then that character's setting, which is kind of taken from their experience of their environment, and then it comes to be personal, drawing from personal experience. (Group Interview, 11 May 2019)

Te mixed media approach proposed provides a diferent form of participation in which the purpose is to encourage descriptions and depictions of it and to go beyond, as one facilitator put it, 'generic accounts' of the problem. By characterising a person with asthma which is relatable to the participants, and potentially wider publics, the 'person with asthma' becomes a subjective storytelling device in which the objects of air pollution, asthma, and inequality are held together without one displacing the other. Tis not only afected the objects of research but the power dynamics assumed in the clinical research because participants were not there to understand their own risks but rather to consider what actions people and communities can take to ensure more breathable air for children with asthma. In one instance, some of the children shared a specifc verb to describe a feeling that breathing in pollution can create, like a trapped cough or having something stuck in your throat. Te word was not known to any of the adults in the room, and it shifted the tone and dynamic of the dialogue because the children participating momentarily became the experts in how to describe and account for air pollution's efects.

Following the workshop the visualisation created by the children was further developed in dialogue with the interview transcripts to create a series of animated memes (to be shared publicly) that linked the problems people face to calls for collective action (with the tagline "let me breathe"). PCEs were temporarily denaturalised, and the question of the future shown to be neither inevitable nor obvious; interventions by children with asthma as subjects of the research, not only as objects, were introduced. In working in the disconnections between the individual accounts of air pollution produced in the interviews and sensor data collection, how the fgure of 'the child with asthma' would lead to the public health promises that underpinned its emergence was destabilised. Tis is not a gap in the data but a destabilisation which reveals the dynamics of inclusion and exclusion that characterise personalisation and which predetermine the parameters for knowing air pollution exposure.

### **Conclusions**

Global inequalities in air pollution and uncertainties in health evidence have led to new methods, technologies and collaborations for studying the variegated environmental contexts of exposure in diferent places. Drawing on our participation in a computer science-public health-social science collaboration measuring embodied air pollution exposure in Delhi, we use the fgure of "the child with asthma" as a concept and method for studying the interdisciplinary methods involved in making environments personal. Te computational framework of PCEs meant research participants were approached as both research objects and subjects: measuring air pollution inside bodies stabilised the relationship between air pollution and health as an object, whilst the person wearing the sensor mediated air pollution through situated negotiations and engagement with it as a subject. However, this dynamic relationship caused tensions between the project's team members during the setting up of the research process and when trying to determine the role and value of people's involvement and the diferent data sets they produce. First, we showed how clinical representations of asthma result in personalised monitoring that infuence which 'environments' are included in research. By invoking and requiring an asthma diagnosis, the places and contexts of exposure to pollution that the research also sought to get a handle on were excluded. Second, we showed how the diferent methods continually reconfgure the subjects and objects of research in ways that destabilise what the problem is (asthma or air pollution, public health or inequality) and how to approach it analytically (instrumentalising qualitative data or engaging with it in a critical manner as at the workshop). By tracing a series of subject-object relations we argue that practices of fguring air pollution as personal in interdisciplinary research create new biomedical categories—vulnerable or 'at risk' groups. However, the environmental and social "pathways" embodied sensing also materialise is limited by an associated narrow defnition of exposure, despite the possibilities for its empirical expansion.

We therefore use the fgure of 'the child with asthma' to investigate the possibilities and implications of computational methods (sensors, algorithms) for understanding the dynamic relations between embodiment (e.g., the fgure) and context or practice (e.g., ground or background) in air pollution exposure. By fguring out exposure in this way it is anticipated that public health interventions can be specifed and improved. We found the interdisciplinary tensions that emerged in the project generative to think with because of the ways personalisation indicates the possibility or necessity of action beyond the generation of scientifc knowledge and data (Rajan 2006: 179)—although the sensors were ultimately only a tool of foresight, a way to manage probable futures (the environmental conditions that increase the likelihood of worsening asthma symptoms) rather than target the causes of exposure and delineate actions. Nonetheless, the introduction of personal air pollution monitoring is promising to capture societal and geographic diferences in exposure that could make perceptible new causal pathways that are not only biological in nature. In our case, it is hoped by some of the computer scientists involved that even if not calculable individual knowledge and experiences can ultimately enlarge computational analysis. What stymies these eforts are pre-determined ideas of symptoms and disease that are stable across environmental and cultural contexts (even if their physical and experiential manifestation are recognised as situated). Te fgure of 'the child with asthma' has helped us elaborate some of these operations of inclusion/exclusion in data and algorithmic practices, that at once expand and contract ways of accounting for the embodied efects of air pollution and harm.

Supporting a more granular and targeted approach to data generation does not necessarily challenge the socio-political conditions and inequalities that allow environmental hazards to happen in the frst place (Senier et al. 2017). As Michelle Murphy argues, granular studies of embodied health (e.g., air pollution personalisation) need to be interlinked with contestations over the physical production and distribution of chemical harm (2013: 698), including how prevalent public health and global health approaches might obfuscate these dynamics. Although the dynamic subject-object refguring of embodied air pollution sensing in our case study did attempt such interlinking to address these complexities, the challenges involved in making commensurate the various roles and possibilities of personal data means how it actually infuences approaches to exposure and risk is difcult to discern. Te interdisciplinary methods that PCEs facilitate did help to disrupt the dominant idea that personal data or information about air pollution is necessary for public health knowledge and action, however. Te actions carried out by young people and their families to manage their exposure highlight some of the ways in which 'good' environments are (as shown in the interviews) or could be (as explored in the workshops) achieved in diferent situations. By studying wearable sensors in research practices, we demonstrate the value of recognising these more ambivalent "interconnections between bodies and data" (Radhakrishnan 2021) as occasions when air pollution is fgured diferently. Tis research hopefully opens up further points of inquiry, in terms of who benefts from personal data and how future fgurations of air pollution might incorporate the social and environmental causes of exposure that are not currently available to computation.

**Acknowledgements** Te authors would like to thank the wider project team in the UK and India for their support and the young people and their parents in Delhi who participated in the research. An early presentation of this chapter at the Innovating Air Pollution Governance: Collaborations, Interruptions and Regenerations panel organised by Maka Suarez, Katie Cox and Rohit Negi during the 2019 4S conference in New Orleans helped develop our thinking. Many thanks to Celia Lury, Scott Wark and William Viney for their insightful and formative feedback on previous versions of this chapter.

### **References**


**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/ by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **11**

# **Figures of Speech: Stuck in the Middle with 'People Like You'**

**Celia Lury**

## **Introduction**

A while ago, my niece posted a photo of a takeaway cofee cup with her name on it on Instagram. Her name was misspelt. Her followers were asked to vote as to whether she should adopt the new spelling. When I spoke to her shortly afterwards, she observed that she has a friend who has a 'Starbucks name', that is, the name the friend gives when asked for her name in Starbucks' franchises. Another of her friends apparently says his name is 'My drink', which my niece described as 'a bit mean'. I used to give my proper name, spelling it out from a vague sense of sympathy with the person requesting the name but, on refection, just introducing an unwanted teacherly or surveillance dynamic to the interaction. My niece said that she enjoys seeing the misspellings of her name. A friend of mine says that she gives the name of the person asking her—the name, that is, that the employee is required to display as part of their uniform.

C. Lury (\*)

Centre for Interdisciplinary Methodologies, University of Warwick, Coventry, UK e-mail: c.lury@warwick.ac.uk

I just managed to stop myself doing this recently when I was asked for my name: the woman who asked wasn't wearing a name badge but an apron with the restaurant chain name on it. In a diferent food chain, my daughter says that customers are required to give orders via an iPad and are automatically given the name of a celebrity. She says she would prefer to be given a number.

I start with this everyday example to foreground what is common knowledge: that naming is socially signifcant as a mechanism for the identifcation of persons, and as such is routinely subverted in (sometimes mean, sometimes mundane, sometimes humorous) practices of misrepresentation, anonymization, subterfuge and impersonation. Tere are many points of interest in this example—the use of frst names alone, the asymmetries involved in the use of names by employee and customer, the subsuming of personal names by company or celebrity names, the limited opportunities for expressions of recognition and solidarity and so on. In what follows I want to consider some examples in which persons are constituted in the use of names, pronouns and numbers: the organization Not In Our Name (NION) and the hashtags #JeSuisCharlie and #MeToo. All three examples, I propose, may be understood as fgures of speech, understood to mean a word or phrase that entails an intentional deviation from everyday language use—spoken or written—in order to produce a rhetorical efect.

As part of a collaborative project on personalization (https://peoplelikeyou.ac.uk/),1 I focus on four inter-related dimensions by which these fgures of speech constitute persons. First, I am concerned to show how each fgure constitutes a simultaneously singular and plural person. Second, I am interested in the role of participation, making a distinction between 'participating in' and 'being part of'. Te questions I want to ask in this regard are: who is included, who is excluded in these simultaneously singular and plural persons when they are constituted by fgures of speech? Does anyone belong? A third concern is whether how and how the fgures might be understood to constitute proper or improper

<sup>1</sup> 'People Like You': Contemporary Figures of Personalisation is a Collaborative Award in the Medical Humanities and Social Sciences funded by the Wellcome Trust Foundation (205456/Z/16/Z).

persons, where 'proper' is used to reference property (including in the self), propriety, appropriation and appropriateness. In doing so I draw on understandings of the proper as the ability to circumscribe place (Bourdieu 1980; de Certeau 1980). My exploration will address how place or territory is circumscribed in these examples by linking the making of place or territory to recursive mechanisms of participation. Te fourth and fnal concern is epistemological, that is, the concern is whether and how the persons constituted in the fgures of speech I describe can speak the truth. Te conclusion brings these concerns together in a proposal that what distinguishes the persons of #JeSuisCharlie and #MeToo is that they are 'stuck in the middle' with 'people like you'.

#### **Pronominalism**

While I will go on to consider the role of names and numbers in constituting these examples of persons as part of fgures of speech, I start by introducing a variety of ways of thinking about pronouns. I begin with Emile Benveniste (1971 [1956]) since he explicitly addresses their role in relation to the category of person, arguing that they enable persons to be established in a specifc relation to the act of speaking. Importantly, the referential relationship the personal pronoun creates is described as circular: it refers to something when it is used, and what it refers to is this use itself, that is, the speaker's self-reference and the referentiality of the message are co-constitutive.2 Over and beyond this, Benveniste further proposes that some pronouns, specifcally frst- and second-person pronouns ('I' and 'you' in English), are distinguished as the only 'personal' pronouns on the grounds that they alone call into existence an unrepeatable object:

"I" designates the one who speaks and at the same time implies an utterance about "I"; in saying "I," I cannot *not* be speaking of myself. In the

<sup>2</sup>Relatedly, Jacques Lacan and other psychoanalysts and political theorists identify what they describe as a splitting or doubling between the subject of the enunciation and the subject of the statement, exploring the implications of this splitting or doubling in language for subjectivity and political agency.

second person, "you" is necessarily designated by "I" and cannot be thought of outside a situation set up by starting with "I"; and at the same time, "I" states something as the predicate of "you." But in the third person a predicate is really stated, only it is outside "I-you"; this form is thus an exception to the relationship by which "I" and "you" are specifed. Consequently, the legitimacy of this form as a "person" is to be questioned. (1971: 197)

For Benveniste, the frst and second personal pronouns or person forms are further uniquely characterized by their 'oneness': 'the "I" who states, the "you" to whom "I" addresses himself are unique each time. But "he" can be an infnite number of subjects—or none' (1971: 199). For Benveniste, the uniqueness inherent in the 'I' as a fgure of speech means there can be no genuinely plural form of the frst person. Instead, he describes 'we' as a 'junction between "I" and the "non-I"' in which 'I' is dominant or transcendent:

Tis junction [of 'I' and 'non-I'] forms a new totality which is of a very special type whose components are not equivalent: in "we" it is always "I" which predominates since there cannot be "we" except by starting with "I," and this "I" dominates the "non-I" element by means of its transcendent quality. Te presence of "I" is constitutive of "we." (1971: 202)

On this basis, he distinguishes between what he calls the undiferentiated 'we' of Indo-European languages and the two distinct forms present in some Amerindian, Australian and other languages—commonly described as the 'inclusive' and 'exclusive' forms 'I + you' and 'I + they'.

In a related strand of thinking, Roman Jakobson (1971 [1956]) describes personal pronouns as shifters, that is, as words whose reference shifts in each use. In his analysis of pronouns, Jakobson is especially concerned with the frst personal plural ('we' in English),3 the collective subject of which he also recognizes to be uncertain. To formalize this uncertainty, he develops a distinctive approach that makes use of cybernetics and information theory to emphasize the importance of

<sup>3</sup> Jakobsen's emphasis on the distribution of 'we', so Julia Kursell argues, stems from his wish to act as a kind of linguistic diplomat in the Cold War, both 'present in and absent from the Soviet Union' (2010: 221), writing about Russian verbs from the US, having left Russia in 1920.

relations between a message and its underlying code. Tis enables him to introduce two general distinctions: 'one between language and that which it narrates, and one between an event and its participants'. 'Four items', he says, 'are to be distinguished, a narrated event, a speech event, a participant of the narrated event, and a participant of the speech event, whether addresser or addressee' (1971: 133).

In a discussion of pronouns in English literature, John Frow introduces the claim made by some linguists:

that all the selves performed or implied by language are fgures of speech, fgures of the self which may in turn be embedded in multiple refexive layerings, not only in the direct and indirect quotation of others or ourselves but in mockery, in "taking of" another person, in acting, in the citation of adages or scraps of unattributed speech, in innuendo, and in all the "keyings" by which we shift from one register to another; from one fguring of the self to another. (Frow 2016: 164-5)

However, he stresses that while deictic markers such as pronouns involve the establishment of a reference point in both speech and writing, they are like and unlike in that 'both work with a temporal reference point, but in writing it is not necessarily equivalent to the time of enunciation' (Frow 2016: 168).

Frow is specially interested in 'free indirect discourse', which he describes as combining personal and impersonal narrative modes. He draws on literary examples to develop this analysis, using them to argue both that Benveniste is mistaken in proposing that only the frst and second person can point to subjectivity, and to show that the use of pronouns may involve shifts in the speaker/addressee relationship. In describing these shifts he draws on Brian McHale's notion of integrational reading and the 'vertical' or 'mimetic context', what McHale calls the text's 'reconstructed level':

Among the things readers know how to do with texts is to reconstruct, taking their cue from the actual sentences of the text, entities not actually given by those sentences "in so many words." Such entities include characters' psychologies, relationships among characters, fctional worlds, and even attitudes (e.g., irony), themes, and "ideas" in the largest sense—as well as … voices and speaking positions. (McHale 1983: 34 in Frow 2016: 172)

But Frow further adds that such shifts may also include shifts in the speaker-addressee pair (i.e., the speaker/addressee pair may change as well as the relationship between any individual speaker/addressee pair). In the movement of these shifts, he suggests, it is possible to identify the shifting perspective of a 'narrating instance', and it is this (both impersonal and personal) perspective that constitutes free indirect discourse. Concluding with Agamben, he suggests that entry into discourse

takes place in the endless occupation of the deictic shifters which at once situate me and render me discontinuous with myself, or rather constitute my self as a site of shifting reference. Tat passage through the empty places of the pronouns and the persons of the verb is something like a journey through nonbeing, a constitution of the subject in the experience of absence. Te pronoun system, like the characters who occupy it, guarantees identity and the dispersal of identity in the same articulation. (Frow 2016: 180)

Tese approaches to pronouns are introduced here to show the complexity of the ways in which their use may constitute persons. In the analysis of fgures of speech that follows it will be supplemented by a consideration of numbers as well as names.

# **Figures of Speech**

Not In Our Name (NION) was the name of a US organization founded in 2002 to protest the US government's response to the events of 11 September 2001. Its Statement of Conscience called on the people of the US 'to resist the policies and overall political direction that have emerged since September 11, 2001, and which pose grave dangers to the people of the world'. Te organization was disbanded in 2008. A version of the organization's name—'Not in my name'—was adopted as a slogan as part of public demonstrations in cities across the UK to protest the involvement of the UK government in the war against Iraq in 2003.

In this example,4 the capacity for the (possessive) personal pronoun 'our' to act as a shifter is muted by its containment as part of the name of a collectively constituted organization (that has a Conscience). And the propriety of this collective name is itself secured through the ways in which individual membership is accomplished by signature, since signatures are a way to indicate a unique individual whose persistent existence—continuous across time and space, independent of context—is legally recognized (Frow 2002). Te fgure of speech that is composed in this example of the use of a pronoun is thus a properly circumscribed collective entity comprised of many unique 'I's, ones or proper individuals. Te existence and identity of the collective entity and the singular individuals who comprise the entity are understood to be independent of each other and to be context-independent, only temporarily sutured by a signature and a (capitalized) name. Nevertheless, while its capacity to be a shifter is restricted, attention to the use of the pronoun 'our' in the name Not In Our Name encourages us to see that the circular logic of identifcation of persons as individuals that is involved here relies upon a short-circuiting. It is the political and legal authority of the state—including the maintenance of an apparatus of naming, including registers of birth, marriage and death, as well as laws of forgery and impersonation that allows signatures to be used as legitimate identifers of both individuals and organizations. In short, it is the state apparatus that gives legitimacy and political efcacy to the address made by this simultaneously singular and plural fgure of speech to government.5

<sup>4</sup>Amongst other uses of this phrase revealed by a Google search at the time of writing, I found that 'Not in my name' has been adopted by a British Muslim organization, Te Active Change Foundation, whose website states, 'As British Muslims we utterly condemn ISIS who are abusing the name of Islam with their acts of terrorism'. And the same slogan is also being used in religious and political protests relating to the slaughter of cows in India.

<sup>5</sup> 'By becoming legal, the proper name enters a whole network of apparatuses (demographic records, criminal records, fscal records, voting records, immunization and health records) through which the state can both identify an individual and efect calculations and operations whose domain is the population. From the state's standpoint, fxing a reference—that is, ensuring that a legal name identifes one and only one subject—is … an essential precondition of modern politics' (Deseriis 2015: 23).

However, when the words 'Not in our name' or 'Not in my name' are not used as names but are statements of conscience appearing on a placard carried by someone at a demonstration, the specifc person or persons to which 'my' or 'our' refers shifts. While relying on the symbolic convention that the person holding a placard intends or motivates the meaning of the statement on the placard, the fgure who speaks is not given a unique, fxed identity and is not accorded a prior or future existence. Tat is, when 'Not in my name' is inscribed as a statement on a placard that moves from one individual to another, and no name is given, the personal pronoun references a transient individual-among-otherindividuals, a not-quite-proper person. Te individuals if or when they carry the placard are individuated, but they are identifed not in their uniqueness but in equivalence or sameness (although they may of course be identifed as unique individuals in practices of surveillance, including automated facial recognition, a technique in which a part of the body is constituted as an involuntary signature by the state or some other surveillant entity).

As the placard moves from person to person, as there is a fguring of one self to another, *my* 'Not in my name' has the same standing as *your* 'Not in my name'. In short, the meaning of the words on the placard is not tied to the identifcation of 'you' or 'me' as unique (contextindependent) individuals but to 'our' indication—a pointing out—as an individual member of a (context-specifc) 'we', of one-among-other-ones. While it is possible to work out that one of the intended addressees is the government (in part, perhaps, because of some other aspects of the context of use, such as the route of the march which the placard holders follow), we can also infer that the individual indicated is also addressing other individuals, who by their co-presence, can point to and be pointed out to each other. Making a 'we' that is an 'I + you' is one of the ways in which the solidarity of this fgure of speech is given substance even if it loses form, as when it becomes a crowd in which the '+ you' might become a '+ they'.

Te hashtag #JeSuisCharlie emerged on Twitter in 2015, following an attack by gunmen at the ofces of the French satirical magazine *Charlie Hebdo*. Two days after the event, the hashtag had been used over fve million times on Twitter, making it one of the most popular topics in the platform's history. It was soon joined by #NousSommesTousCharlie. Most uses of these French-language hashtags were not from French accounts. Immediately following their appearance, another appeared— #JeNeSuisPasCharlie, although in much smaller numbers (just over 74,000 in the frst few days). Other hashtags included #JesuisAhmed ('I am Ahmed', in reference to Ahmed Merabet, the police ofcer who was shot outside the Charlie Hebdo ofces by the same gunmen who killed members of the magazine's staf). Later that year, Willem, one of the cartoonists employed at the magazine declared, 'We vomit on those who suddenly declared that they were our friends' (http://www.lepoint.fr/ societe/willem-vomit-sur-ceux-qui-subitement-disent-etre-nosamis-10-01-2015-1895408\_23.php).

In the use of the hashtag #JeSuisCharlie, the use of a name alongside the pronoun 'Je' does not restrict its capacity to act as a shifter. While Charlie is capitalized as a proper name, it is not functioning properly.6 Te 'Charlie' of 'I am Charlie', we may infer, if we are aware of the events of the attack on the magazine *Charlie Hebdo*, is a shortening of the magazine name, the frst part of which is said to refer to both the cartoon character Charlie Brown and Charles de Gaulle, a former French President, with the second part, Hebdo, short for *hebdomadaire*— 'weekly'. However, as the hashtag was used again and again, as the number of speech events aggregated, as one self was fgured to another, it was also possible to infer that the 'I' in this fgure of speech was not an employee of the magazine. Indeed, as the number of hashtags associated with this phrase grew in size, as it appeared many, many times a day not just once a week, it became clear that the fact that the hashtag user who said, 'I am Charlie' was *not* a member of *Charlie Hebdo*, was the point.

However, although I earlier gave numbers that attest to the widespread use of the hashtag, there was uncertainty as to how the collective nature of the subject of this fgure of speech was to be understood. While Robert Payne argues that 'the contagious complexity of the slogan cannot be captured by quantitative measurement of tweet volume and frequency',

<sup>6</sup>Te propriety of speech is regulated in a range of ways across platforms: a Facebook page called 'We Are All Khaled Said', which was a focal point in Egypt's revolutionary movement, was deactivated by Facebook on the grounds that the account holder, Wael Ghonim, had used a pseudonym (Tufekci 2017).

he also suggests that the hashtag was a 'mass demonstration of individualised solidarity' (2018: 279):

the individual Charlie *is* the collective, united in defence against attack upon any one of its members and the values that each embodies. … the performative function of mass repetition of the speech act '*Je suis Charlie*' serves to inaugurate a new subject who seeks recognition within a restricted feld of norms. Te paradox of this subject is that its singularity emerges only through multiple acts of individuality, none of which is fully autonomous. No single Charlie originates the subject position from which all claim to speak. (2018: 281)

In contrast, Inka Salovaara-Moring (2015) argues that '*Je suis Charlie* functions as a non- or post-human agent that tells a story on the behalf of *w*e, *the assemblage*. Using Charlie as "*we*" does not only defne the action and experience, but the narration itself'. However, this assemblage is also understood to approximate a subject: she says,

In the digital media ecosystem, the implied 'agent' is almost irreplaceable. Te narrative structure including 'we' ('black people', 'freedom fghters', 'Muslims') provides a narrative with an ideological trajectory. Te connection between time, space, narrative and history becomes clear as the group achieves a refexive self-awareness as a 'subject' that is analogous of the individual. (2015)

Te suggestion here is that this fgure of speech is not composed as a collective entity of unique, independent ones or even of one-amongother-ones, but as a more-and-less-than-one.

As such, it is perhaps not surprising that as the fgure of speech came into existence, the impropriety of the speech act elicited a critical response from others. Payne observes that

one user is irritated by and cynical about the Charlie movement: '*comment ça m'énerve ceux qui mettent "je suis Charlie" et qui n'ont jamais entendu parler du journal Charlie hebdo, tt ça pr follow le mouv*' ('It really annoys me how people post "Je suis Charlie" and have never heard of the paper Charlie Hebdo, all just to follow the movement'). … Another user labels the opportunists more bluntly: '*Et y'a les moutons qui te mettent "je suis Charlie"*  *partout*' ('And here are the sheep posting "Je suis Charlie" everywhere'), followed by a 'suspicious face' emoticon. (2018: 284–5)

He concludes with the suggestion that for some users, 'Charlie supporters lack sincerity, individuality and knowledge of context' (2018: 284). Te contexts most commonly invoked as lacking—by participants and analysts—were those of nation and race.

Camille Robcis (2015) notes that some 'commentators seem particularly upset by the British and American insinuation that the content of *Charlie Hebdo* might indeed be read as racist and, consequently, that one may condemn the murders without embracing the identifcatory universalism that [French Prime Minister] Valls and others have called for' (Robcis 2015). Some users, she notes, interpreted the use by other users of #JeNeSuisPasCharlie as the 'hypocrisy and shared misunderstanding of what were sometimes called Anglo-Saxons'. In another commentary, Alana Lentin (2018) describes the event of Charlie Hebdo as an example of a 'white context' requiring 'black analytics'.7 Her argument is that the appeal for the need to understand the 'French context relied on a "white analytics" that opposes the centrality of race as an interpretive framework that a "black analytics" foregrounds'.8 In making this argument, she unpicks the politics of 'the urge to contextualise' (2018: 52):

the proposition that providing French context could overcome what were portrayed as misreadings of the reasons for and backdrop to the attacks on Charlie Hebdo did not complexify these rationales. Rather, because the context provided did not engage with an epistemology of black analytics, it could not ultimately dispel the tendency to present a Janus-faced picture of French society—those who identifed with a personifed "Charlie" (republican, secular, white) and those who did not (communitarian, religious/ fundamentalist, racialized). (2018: 49–50)

<sup>7</sup> She writes, 'I propose that the aftermath of "Charlie Hebdo" is a ftting example of the need for more attention to be paid to the lessons of Du Boisian double consciousness, an acute awareness of the other's world that endows Black and otherwise racialized people with an insight into the white structures in which they live' (2018: 47).

<sup>8</sup> See also Sanjay Sharma's analysis of 'Blacktags', the vernacular term for popular racialized hashtags, which, he argues, 'reveal the contagious efects of networked relations in producing emergent racial aggregations, rather than simply representing the behaviour of an intentionally acting group of Black Twitter users' (2012: 48).

Te variety of these lay and academic analyses speaks to the complexity of the co-constitutive inter-relationship of a speaker's self-reference and the referentiality of the message. As previously mentioned, Willem, who was a contributor to the magazine, explicitly rejected the identifcation with 'Charlie' made by others. We can see this response as an assertion by Willem, in the shadow of his experience of an attack on his life and that of his colleagues, of the impropriety of anyone else asserting an existential link. Or, if we follow Jakobson, we can see his response as a denial of the right of others to participate in the narrated event through participation in the speech event. In contrast, those who responded by using #JeNeSuisPasCharlie opposed the sentiment of those speakers who said #JeSuisCharlie but accepted their right to participate in the narrative event by exercising their own ability to do so. In other words, one aspect of what is contested in this fgure of speech is whether you can participate *in* an event without being a part *of* it, whether and how speech events and narrative events are articulated together and what kind of subjective or other agency, authority or credibility this layered articulation—or contextual integration—might aford.

Combining Jakobson with Lentin's analysis of the politics of contexting raises the issue of whether and how participating in and being part of come to be associated with belonging, place and territory. She writes,

Te suggestion that all speech is free belies the facts that speech uttered and heard is deeply unequal and that the diferent actors within it have varying degrees of freedom. An Australian example indicates this: Uthman Badar, the President of the Muslim group *Hizb-ut Tahrir*, was asked to address the Festival of Dangerous Ideas, an event with the tagline, "a series of talks that bring contentious ideas to the fore and challenge mainstream thought and opinion." However, Badar's presentation at the event … was canceled due to outrage over the topic of "honor killings," the subject the organizers asked him to address. As Randa Abdel-Fattah notes, as a Muslim, Badar was not allowed to have a "dangerous idea," because to do so would imply "that he is a Muslim of Australia, not a Muslim in Australia". (2018: 54-55)

Both the hashtag #JeSuisCharlie and the hashtag #JeNeSuisPasCharlie circulate in an algorithmic infrastructure in which context awareness is difcult to say the least: across the platforms of digital media there is considerable potential for multiple possible confusions, tangles and confrontations in the circular self-(p)referencing and platform-induced dynamism of trending. Indeed, the event of #JeSuisCharlie can be seen as an instance of what Elena Esposito (2004) calls 'second-order blindness' and what Gregory Bateson (1972) calls schizmogenesis: the continual reproduction, confrmation and intensifcation of diference.9

Bateson developed the term schizmogenesis in an analysis of doublebinds or double-takes, which he describes as examples of transcontextualism, a genus of syndromes or cognitive tangles associated with the 'more than circular' that arise when individuals learn—or fail to learn—how to deal with uncertainty in relation to context. At the heart of this genus, says Bateson, is the human capacity to deal with the 'weaving of contexts and of messages which propose context—but which, like all messages, whatsoever, have "meaning" only by virtue of [the] context' in which they are received (1972: 275-6). Contexts may set the stage for a 'certain class of response', but learning what changes and what stays the same across contexts is challenging, and 'breaches in the weave of contextual structure' are common. Certainly, in the case of fgures of speech that emerge in social media, the layering that McHale identifes as part of a contextual or integrational reading seems likely to lead to multiple breaches in context, perhaps even making ground-truthing or providing a (common) ground for readings impossible to establish (Day and Lury 2017).

Te MeToo Movement is a movement against sexual harassment and assault. Tarana Burke, a US social activist and community organizer, began using the phrase 'Me Too' to refer to sexual harassment in 2006. Te hashtag #MeToo spread online in October 2017 following sexual misconduct allegations against Harvey Weinstein. Te hashtag was frst tweeted by the US actor Alyssa Milano around noon on 15 October 2017 and had been used more than 200,000 times by the end of the day

<sup>9</sup> In his work on populist reason, Ernesto Laclau speaks of how foating signifers generate an equivalential chain, which 'has an anti-institutional character: it subverts the particularistic, diferential character of the demands' (2005: 38). Te 'internal frontier' of populism reproduces an us-versusthem discourse, a model of constant expansion. Tis model has a broad explanatory sweep which Ravi Sundaram (2015) refnes by providing an account of how contemporary populist mobilization in India is embedded in the specifc informational strategies associated with digital media.

and tweeted more than 500,000 times by the next day. On Facebook, the hashtag was used by more than 4.7 million people in 12 million posts during the frst 24 hours. Te platform reported that 45% of users in the US had a friend who had posted using the term.

In this example, as with #JeSuisCharlie, the relation between the singular and the plural, the one and the many, is continually being remade by the platforms in which the fgure of speech moves. Also as in the case of #JeSuisCharlie, this example demonstrates that disbelief, doubt and speculation are the unavoidable outcomes of the serial calibration of signal and noise, in(ter)ference and (un)certainty across contexts. However, as well as triggering other tangential or derivative claims (#NotMeToo, #HimToo, #NotAllMen, #YesAllWomen, #BelieveWomen, #BelieveAllWomen),10 some of the speakers in the movement are engaged in a series of trials legal and otherwise, in which the relation of personal pronouns to individuals who can be and sometimes are named is being put to the test. In these contexts, the question becomes, is 'I' the subject of 'Me(Too)'? And if so, is the 'I' of 'Me(Too)' telling the truth?

Here I want to suggest that it's not necessarily helpful to say that this fgure of speech participates in an era of post-truth. Or to say that truth is now 'after the fact'. According to Benveniste, personal pronouns—as self-referential signs—cannot be used incorrectly; as they do not state anything, 'they are not subject to the conditions of truth and escape denial' (1971: 220).11 However, this is not Jakobson's view as becomes apparent in his discussion of the third person plural, 'we', which he describes as both a shifter and a non-shifter. For Jakobson, 'we' is a nonshifter insofar as, in many languages,12 it conveys at least some

<sup>10</sup>Relatedly, see Emily Rosamond's discussion (2020) of how social impact bonds (SIB) operate as 'derivative character investments'. She writes, 'By depicting benefciaries as better able to morally direct their lives, [SIB promotional videos] represent SIBs as path-changing devices, threading more fulflling life paths through society. Tey encourage derivative character investments in bundles of bettered behavior, narratively linked to changed life paths at scale'.

<sup>11</sup>His explanation as to how this is so relies upon his understanding of correct or proper use: 'it is a fact both original and fundamental that these "pronominal forms" do not refer to "reality" or to "objective" positions in space or time but to the utterance, unique each time, that contains them, and thus they refect their proper use' (1971: 219).

<sup>12</sup>Amia Srinivasan writes, 'In many … languages—including Malay, Finnish, Hungarian, Estonian, Armenian, Bengali, Persian, Ewe and Swahili—the problem of the gender-neutral third person pronoun doesn't arise, because of the absence or near absence of grammatical gender. In these lan-

information as to category of person, specifcally information as to category of person (gender) and category of number (more than one) (Kursell 2010).

As is well known, the information conveyed as to gender varies signifcantly across languages. In current uses of the English language, however, the category of person—gender—is being supplemented by a category of number through the advocacy of the use of the third person plural— 'they'—to indicate a (single) person of fuid, non-binary or trans-gender. Tere is a lot to be said about this but here I focus on the use of 'they' as a category of number to suggest that the kind of number called into existence is a distributive number,13 a statistical number supported by, but not exclusive to, the calculation of digital data.

In ordinary English language use, a distributive number is a word that answers 'how many times each?' or 'how many at a time?', while the distributive property law in mathematics concerns the ordering or sequencing of arithmetical operations. Adrian Mackenzie (2016) draws on both these understandings when he suggests that distributive numbers should be the name for those numbers that emerge from the sequencing of the arithmetical operations of conjoining (adding, subtracting, multiplying and dividing) of probability distributions in complex statistical techniques such as Markov Chain Monte Carlo simulation (MCMC). He develops his analysis by suggesting that while MCMC is designed to individualize entities, the aim is to describe relations between individual entities or events that are neither identical to nor independent of each other. MCMC does this, he says, by individuating an entity by calculating how more or less similar the entity is to many others in probabilistic terms, specifcally by identifying an individual entity or event as a *joint probability distribution* within diferent intersecting populations. Te probability

guages, the same word is used for "he" and "she", and sometimes for "it" as well. In Ojibwe, an indigenous North American language whose nouns are not classifed by gender but according to whether they are considered animate or not, the singular third person pronoun *wiin* is used for both "she" and "he". In Turkish, the equivalent of 'he', "she" and "it" is simply *o*, which seems to me unimprovable' (2020). Srinivasan also notes that while the American Dialect Society chose 'they' as its word of the decade in 2020, it has been used as a singular pronoun for over 600 years. 13Te sufxes -some (as in twosome, threesome), -ly (weekly, annually, regularly) and -fold (as in two-fold, three-fold) are sometimes used. A conspicuous contemporary use of distributive numbers is in -arity or -adicity, to indicate how many parameters a function takes.

of any individual entity is thus understood as always distributed—added to, subtracted from, multiplied and divided—in relation to many populations in many times (in ordered sequence)14 in time. Mackenzie writes,

In post-demographic understandings of data, individuals appear not simply as members of a population (although they certainly do that), but themselves as a kind of joint probability distribution at the conjunction of many diferent numbering practices. If individuals were once collected, grouped, ranked, and trained in populations characterised by disparate attributes (life expectancies, socio-economic variables, educational development, and so on), today we might say that they are distributed across populations of diferent kinds that intersect through them. Individuals become more like populations or crowds. (2016: 116)

It is as an example of a personal pronoun that is also a distributive number, I suggest, that the #MeToo has the capacity to refute Benveniste's claim that 'I cannot *not* be speaking of myself' as a 'one'. Tat is, #MeToo is a platform-enabled participatory intersection of populations, in which as Mackenzie notes, 'the lines between objective and subjective, or aleatory and epistemic probability, begin to shift not towards some total computer simulation of reality but towards a refolding of probability through world and experience' (2016: 126). 'Me' is conjoined with 'Too' in a way that makes visible the short-circuiting that is made invisible in the use of signature and the popular social media phrase 'You do you', which both short-circuit recognition of the mutual constitution of self and speech.

As a fgure of speech, #MeToo is simultaneously a person that is one and many, not as the addition of either unique independent ones or of equivalent ones but as a conjoint person or condividual (a dividing with) (Deseriis 2015). As such, it is an instance of free, indirect discourse, the (im)personal perspective of a constative (probabilistic) and performative

<sup>14</sup>Tis ordering of time in time is an important characteristic of written narrative as described by Frow. He writes, 'Impersonal narration shifts its deictic centre from the situation of utterance which is the norm for spoken language, to the spatio-temporal coordinates corresponding to the central or focalized character, or rather to whichever character is central or focalized at that point in the narrative' (2016: 168). Te suggestion here is that statistical techniques provide opportunities for (parametric) focalization.

(participatory) practice of conjoining. It involves a pronominalism that involves a constantly shifting de- and re-aggregation of participation such that the 'we' that is 'MeToo' is simultaneously inside and outside both the relations 'I + you' and 'I + they'.15 Indeed, it is perhaps because such fgures of speech comprise wholes that are vague in the sense that they are simultaneously inclusively exclusive and exclusively inclusive (Agamben 1998; Verran 2007; Guyer 2014) that #MeToo provokes the emergence of competing 'totals' or 'wholes', including not only #NotMeToo, and #HimToo but also #NotAllMen, #YesAllWomen, #BelieveWomen, #BelieveAllWomen. Tis is also perhaps what makes the fgure of speech that is #BlackLivesMatter a total or whole that is simultaneously 'bigger' and 'smaller', both 'more' and 'less' than #AllLivesMatter.

# **Stuck in the Middle**

To conclude, I want to suggest that the analyses of these examples as fgures of speech illustrate some of the limits and the possibilities of a nonrepresentational politics: the challenges such a politics pose for understandings of persons, and the kinds of relations that can exist between the singular and the plural, the one and the many, the proper and the improper as well as raising questions about whether and how truth may be established. Can the performative 'I' of 'I *promise*' (to tell the truth, the whole truth and nothing but the truth) be rendered equivalent to the self-(p)referencing 'I' of 'I *like* this' or the 'I' that is more *and* less likely to be 'you' at this (and that) time? What kinds of causal relation between past acts and present responsibility—if any—can be established for these conjoint fgures of speech? In what ways, if at all, can relations between '#MeToo' and 'you' be rendered equivalent to the relations that Benveniste identifes as existing in 'we' between the personal pronouns 'I + you', or even 'I + they'? In what sense can or should such claims be described as free speech?

<sup>15</sup>For Sanjay Sharma (2012), Blacktags 'are instrumental in producing networked subjects which have the capacity to multiply the possibilities of being raced online. … ethno-racial collective behaviours on the Twitter social media platform are grasped as emergent aggregations, materialized through the contagious social relations produced by the networked propagation of Blacktags'.

In a discussion of improper names (2015), Marco Deseriis introduces Slavoj Zizek's claim that the guarantee of 'the identity of an object in all counterfactual situations—through a change of all its descriptive features—is the retroactive efect of naming itself. It is the name itself, the signifer, which supports the identity of the object' (Zizek 1989: 94-95 in Deseriis 2015: 24). However, he also cites Ernesto Laclau who says 'the reverse movement also operates: [empty signifers] can never fully control which demands they embody or represent' (Laclau 2005: 108 in Deseriis 2015: 25). Deseriis' own position—or wager as he puts it—is that 'rather than expressing heterogeneity or homogeneity, diference or totality, the improper is a mode of mediation between these two poles. Tis mediation is evident not only in the passage from the one to the many (and vice versa) … but also in the relation between signifying and a-signifying practices within the assemblage' (Deseriis 2015: 25).

Walter Benjamin proposes that a 'pure middle' is one whose middleness is not defned with respect to determinable end-points, but is, rather, an infnite and infnitely divisible space. Of this space, the critic Peter Fenves writes:

Nothing can withstand this space intact: infnite divisibility is the 'law' of this space, which, however, cannot be posited as a law, since this division is never governed by an identifable rule. Te 'law' of this space, the rule by which its infnite divisibility is articulated, must likewise be infnitely codivisible: in German, *mitteilbar*, which is to say, 'communicable'. (Fenves 2001: 255)

Clearly, the pure middle is an abstraction, but the implication of the introduction of abstraction into social life by way of media-specifc operations of communicability is precisely what has been considered here. What has emerged across the analysis of these examples is the importance of considering media-specifcity for the ways in which a speaker's selfreference and the referentiality of language are co-constitutive. In the last two cases, for example, the elements are data-points, and the connections, couplings or conjoints are the hashtags, 16 likes, shares, retweets

<sup>16</sup>A hashtag is a form of punctuation that both connects and divides, as well as being the symbol to indicate a number, or a unit of currency as well as a bone fracture and, in conjunction with @, an address or place.

and so on, which are brought into multiple and dynamic relations with each other in the distributive operations of platforms of all kinds. If we consider these couplings as the introduction of abstraction of a mediaspecifc kind into social life, we can see we are not simply witnessing a proliferation of (im)proper persons, but are 'stuck in the middle' with 'People (more and less, sometimes and sometimes not) Like You', participating in and/or being part of a totality that is a vague whole (Guyer 2014). Both #JeSuisCharlie and #MeToo are fgures of speech that cannot be summed up by the addition of either individual 'I's, unitary 1s or 'ayes', but rather constitute more or less proper persons, existing simultaneously inside and outside 'I + you' and 'I + they', for whom the circumscription of ground, place or territory cannot be fnally determined.

**Acknowledgements** Te author would like to thank members of the People Like You project, specially Sophie Day, Will Viney and Scott Wark, as well as John Frow for their comments on this paper. Te research was supported by the Wellcome Trust (205456/Z/16/Z).

#### **References**


**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/ by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **12**

# **Ubiquitous Surveillance and Data Selves**

**John Frow**

*Are we made up entirely and without residue of the data that defne us, or is there a disjunction between our data shadows and our embodied selves? How do we come to recognize ourselves, our selves, in the pronouns that interpellate us online, and what is it exactly that we recognize? What does it mean to occupy the semantic and positional space of the pronoun 'you'? And is there a continuity or a discontinuity between the systems of surveillance and data aggregation that address us and the systems that refer to us? Te markers of identity generated by such systems work by both individuating and classifying us; this paper seeks to think about the range of possible relations between that generality and that particularity.*

Paper presented at the conference *Figurations: Persons In/Out of Data*, Goldsmiths College, University of London, December 2019. Warm thanks to Celia Lury and Scott Wark for their work in organizing the conference and to Nathaniel Tkacz for his very helpful comments on an earlier draft.

J. Frow (\*)

Department of English, Te University of Sydney, Sydney, NSW, Australia e-mail: john.frow@sydney.edu.au

Last year my (very latent) Facebook account was hacked and a fgure bearing my name and my visual icon joined me up to a number of plausible sites, and a couple of somewhat less plausible ones, on which it then made pitches announcing special deals on gambling sites and promoting casinos. I've now withdrawn completely from Facebook so I can't tell whether this doppelgänger still represents me there, although given its apparent autonomy it may well have taken on a vampiric life of its own.

Tat's a frst example of identity theft, one in which it's still quite clear which is the real and which the false John Frow: my doppelgänger 'represents' me and for some people may 'be' me, but for people who know me it will, I hope, be obvious that its words in support of gambling are unlikely to be mine. A second, equally banal example: a couple of years ago my wallet was stolen in Barcelona and my bank later traced the trail of my credit card up toll roads to Lyon, where it stopped, the card having been cancelled. My bank believed me, and someone's representation of themselves as me was labelled a misrepresentation. But consider a third possibility, that of a complete and successful theft of identity. Koopman (2019: 4-5) summarizes the resulting 'permanent and irreversible erasure of the entirety of [the victim's] personal information and therefore their entire informational identity' as follows:

No driver's license, no passport, no bank account number, no credit report, no college transcripts, no employment contract, no medical insurance card, no health records, and, at the bottom of them all, no registered certifcate of birth. Te scenario is chilling: everyone around you well attached to their data while you are dataless, informationless, and as a result truly helpless. What would you make of yourself? What could others make of you? What would the bureaucracy be able to do when you petition it with your plight, given the fact that no bureaucracy can address a subject as other than their information? … Tey would have no way of addressing you from one day to the next, of recording you in their databases, of numbering or naming you, and so no way at all to deal with you on anything approaching a consistent basis. You could not even receive special support through special court orders because, completely unrepresentable as information, you would have no way of being registered into a court, for that would require rendering you into the data from which you have been detached.

Te subsumption of personhood into documentary information that is evidenced in the case of identity theft has its beginnings with the governmental systems of early modernity, when a proliferating apparatus of writing begins to integrate verifed identities in cadastral and demographic records and to generate both a systematic scrutiny and the rights and recognition accorded by the state to those verifed identities. Te paradox at the heart of this process is that, while an identity document purports to be 'a record of uniqueness', it must at the same time be 'an element in a classifying series' (Caplan and Torpey 2001: 8). And the converse of that dichotomy is the further paradox that these mechanisms of construction, verifcation, scrutiny and control are at the same time lived by me as the confrmation of my uniqueness, my being as a person. Personalization in this sense is a function of generality, 'a mode of individuation in which entities are precisely specifed by way of recursive inclusion in types or classes' (Lury and Day 2019: 18). My documents identify me not only as 'the' person that I am but as 'a' person, one of *those* persons, one of those entities that are classed as persons; and as a member of all those other classes of human being that count and that make me count for this act of classing.

Individual identity is legally underpinned by the documents of state, but, as Ruppert (2011: 218) writes, 'people are not governed in relation to their individuality, but as members of populations. Te embodied individual is of interest to governments in so far as the individual can be identifed, categorized and recognized as a member of a population. Tis is the general problematic of governing, which is to know the nature and then govern and regulate the forces of the collective body'. Populations are known and made legible—indeed, are constructed as populations by means of devices and practices of identifcation of individuals as categorized subjects, 'an element in a classifying series'. Te basic identifers for governmental scrutiny are the biographical data that register a name, a set of kinship relations, a gender, an ethnicity, an address; in addition, biometric data record certain unique physical attributes or habits of bodies (facial structure, fngerprints, iris geometry, height, gait). Te biographical data move outwards to the set of social relations within which they are meaningful; the biometric data move centripetally to identify a singular body diferentiated from all other bodies. What is of interest to the state may be either the general category (Which classes of people have need of particular social services? What is the desirable distribution of resources across a particular population?) or hypothetical individuals apprehended as members of a category (Which particular neo-Nazis or radicalized Muslims are likely to espouse violence? Which former chemistry teachers are likely to be cooking ice?). In each case the state will aim to build up a coherent picture through the use of statistical evidence, or through a cumulative record of transactions between an individual and the various branches of the state or commercial institutions (driving license, criminal record, medical records, property holdings, and so on). It will identify patterns of group circumstance or patterns of individual conduct, and it will seek to amass and correlate patterns of information across whatever databases are at its disposal.

Tose databases are now for the most part digital, and thus susceptible to algorithmic operations of search and analysis which build 'what is uncertain and unknown into the identity calculation itself' (Amoore 2008: 25). In the case of national security systems, such speculative uses of data, monitoring whole populations in quest of individual anomalies, are intended to generate as-yet unsuspected patterns rather than to fnd evidence to support known possibilities (Raley 2014: 123). Tey bring into being what Amoore (2011: 27) calls a form of data derivative, meaning 'a specifc form of abstraction that distinctively correlates more conventional state collection of data with emergent and unfolding futures'; the data derivative comes into being from 'an amalgam of disaggregated data', sorted by way of recursively refned algorithmic association rules and given visual form as 'risk map, score or colour-coded fag' (Amoore 2011: 27). It might, for example, be derived from an associative matrix connecting a fight destination, fare payment by a third party at short notice, a dietary choice, and a history of attendance at a religious institution. Te knowledge formed here is 'actuarial' (Andrejevic 2012: 95), converting 'the databased residue of daily life' (Amoore 2009: 52) into an actionable probability. A risk value is assigned to an individual, and this pre-emptive identifcation allows the analyst either to read outwards to the 'nodes of connections between data' (Bauman et al. 2014: 125) (i.e., the network of a suspect's personal connections) or to act to avert an immediate threat. Te data derivative is 'indiferent to the contingent biographies that actually make up the underlying data'; it 'is not centred on who we are, nor even on what our data says about us, but on what can be imagined and inferred about who we might be – on our very proclivities and potentialities' (Amoore 2011: 28). To my singular body it attaches a virtual state, my data shadow, which then defnes me.

Edward Snowden's revelations about the US National Security Agency identifed a range of forms of Internet surveillance, among the most prominent of which are, on the one hand, the PRISM programme, which, through its XKEYSCORE software, allows analysts to read from the servers of Internet service providers every keystroke of every person's online activity; and, on the other, the direct upstream harvesting of data from private-sector Internet infrastructure—'the switches and routers that shunt Internet trafc worldwide, via the satellites in orbit and the high-capacity fber-optic cables that run under the ocean' (Snowden 2019: 122). Tis is genuinely ubiquitous surveillance. Yet note that it's difcult to draw a clean line between such state surveillance and the dataharvesting capabilities of corporate information harvesters and brokers, with some of which—Snowden mentions 'Microsoft, Yahoo!, Google, Facebook, Paltalk, YouTube, Skype, AOL, and Apple' (Snowden 2019: 122)—the NSA has a closely symbiotic connection.

A number of diferent kinds of commercial corporation harvest data on Internet users: retail corporations, search engine operators, social media companies, data brokers, data analytics providers, and so on. Apart from data-broking and data-analysis frms, which sell information directly to their clients, the business model on which these corporations run is personalized advertising based in the interactive capacity of digital media, with harvested data either used directly, in the case of retail frms, or auctioned of to advertisers in the case of search engines and social media companies. Te pioneers in the feld were probably the giant consumer credit bureaus like Experian, Equifax, and TransUnion (cf. Lauer 2017), and retail corporations like Walmart and Amazon in the US and Tesco and Marks and Spencer in the UK, which hold and monitor massive amounts of data on the contact details, purchasing history, and lifestyle preferences of their customers, along with all the ancillary information that fows from it (fnancial status, sexuality, mobility, physical ftness, cultural tastes, dietary and pharmaceutical choices, and so on). But the feld of personalized advertising is now dominated by the two biggest players, Google and Facebook, with the other three technology giants, Microsoft, Apple, and Amazon, integrated into the feld in somewhat diferent ways.

Let me take Google as the key example here. In Shoshana Zubof's comprehensive account, the dot.com crisis of 2000 provided the occasion for Google's founders to abandon their initial strong opposition to advertising. Two conditions allowed the company to transform online advertising, particularly after its purchase of DoubleClick in 2008 (Cheney-Lippold 2017: 20) and as it came to realize that it was indexing not only, at an aggregate, topological level, the entire network but also a second world, that of individual users, which it then overlaid on the frst in order 'to deliver relevant search results to the users, and to deliver relevant users to advertisers' (Stalder 2010: np). Te frst condition it satisfed was that its computational tools and infrastructure enabled it to create user profle information from analysis of search patterns, keystroke by keystroke, and to match advertisements to the user's interests as they are deduced from these traces of online behaviour. Te second condition was its development of an options-based pay-per-click revenue system which it carried to the contextual advertising system it established for its search engine and for Gmail, such that when an advertiser bids for a keyword the system tracks those of the user's searches that are contextually relevant for it (e.g., a search for online clothing), matches the user to a product range, serves the advertisement, and, if the visitor clicks on it, invoices the advertiser for the price negotiated for that particular user profle—all in real time (Turow and Draper 2012: 135).

Unlike older business models targeted only to keywords or content, Google was thus able to tailor advertisements to the interests, the social connections, and the physical and online locations of a particular user, and it did so by collecting stores of what Zubof calls 'behavioral surplus', which embrace 'everything in the online milieu: searches, e-mails, texts, photos, songs, messages, videos, locations, communication patterns, attitudes, preferences, interests, faces, emotions, illnesses, social networks, purchases, and so on' (Zubof 2019: 128). Hence the expansion of Google, Facebook, Amazon, and Apple into the Internet of things: a world of information-gathering devices, each of which is 'a slightly diferent confguration of hardware, software, algorithms, sensors, and connectivity designed to mimic a car, shirt, cell phone, book, video, robot, chip, drone, camera, cornea, tree, television, watch, nanobot, intestinal fora, or any online service' (Zubof 2019: 129).

We all have a sense of the sheer scale of the enterprise: in 2020 there were 6.9 billion Google searches a day and the company generated \$116 billion, 97% of its total revenue, from advertising sales.1 Its subsidiary company YouTube had 2 billion monthly users in February 2019, Chrome had 62% of the browser market globally, Android had 2 billion users in mid-2017, and Google Maps and Gmail each had well over a billion users. Te services that Google and its parent company Alphabet ofer are multifarious, but their aim is singular: to collect behavioural data about individuals that can be monetized as advertising revenue. As Zubof puts it: 'With click-through rates as the measure of relevance accomplished, behavioral surplus was institutionalized as the cornerstone of a new kind of commerce that depended upon online surveillance at scale' (Zubof 2019: 83).

Te collection processes employed by Google and other commercial entities are structurally homologous with those of state intelligence services. 'Online surveillance at scale' harvests information that had never previously been captured at scale—'about people's time-space paths through the course of the day, the details of when and where they chat with friends, even the random queries that drift through their minds (to the extent that these are transformed into Google searches)' (Andrejevic 2012: 93)—and it does so by making use of algorithmic procedures, such as mathematical association rules, which move between the commercial sphere where they were initially developed and that of national security apparatuses (Amoore 2008: 26).

But the ubiquity of surveillance doesn't mean that we live in a world of totalized panoptic control. Mark Andrejevic has suggested the alternative metaphor of a world made up of a series of distinct but sometimes overlapping digital enclosures, meaning the coverage range created by the interactive and data storage capabilities of any digital surveillance technology—a world characterized, then, 'by a proliferation of diferent

<sup>1</sup>https://techjury.net/stats-about/google/, drawing on statistics from TechCrunch and Statista.

monitoring networks with varying capabilities for information capture under the control of diferent entities' (Andrejevic 2012: 93). Under certain conditions (e.g., a totalitarian government such as that of China with a tight hand on the public domain) data from a number of diferent enclosures might be aggregated; and security agencies such as the NSA do in practice make use of commercially gathered data, either by stealing it or by exploiting software vulnerabilities or merely by requesting access to it. Tis is an area in which tech companies in the West are, or want to be seen to be, pushing back, but probably the most we can say about this is that the balance between privacy and omnivorous data collection is precarious and in a state of considerable fux. Further, the coexistence of digital enclosures within an overarching assemblage means that information collected for one purpose—the mapping of the built environment by Google Earth, for example, or the monitoring of the fow of water or electricity or trafc—might be migrated 'across a range of other, sometimes unanticipated functions' (Andrejevic 2012: 93; cf. Lyon 2014: 5-6, 8). Te trade goes both ways, with technologies and software developed for military or security purposes fnding their way into the surveillance activities of business—or, more precisely perhaps, with an increasing lack of diferentiation between these spheres.

Te identifying and personalizing data that we yield through digital interactions may be given voluntarily or involuntarily. Involuntary generation of data takes place by means of cookies or other tracking devices which introduce memory, or statefulness, into a stateless system such as the basic Internet protocols (Sipior et al. 2011: 3), and they are thus deictically charged: localized in time and space to a particular Internet subject. Alternatively, the involuntary generation of data takes place by way of the network of automated sensors (facial recognition systems, RFID tags, location tracking, the plethora of sensors on any smartphone, and so on) that cover our world, directly registering traces of our bodily presence in space and time (Kang and Cuf 2005: 94). In some instances we may give permission for our data to be collected; but since the alternative is not to use the interface at all, and since privacy agreements tend to be unreadably lengthy and legalistic, the permission can only technically be said to be voluntary: it is in efect a function of 'a regime of compulsory self-disclosure' (Andrejevic and Gates 2014: 191).

Te prevalence, on the other hand, of *voluntary* disclosure of personal data seems, given its value for commercial exploitation or for scrutiny by the state, to require some explanation. 'We need to understand', writes Koopman (2019: viii), 'why we do not question, and why we even eagerly participate in, projects of government data harvesting and corporate data collection, and a raft of programs designed to store and analyze every fake of data dandruf we cannot help but leave behind in nearly everything we do'. Te explanation—beyond the sheer usefulness of centralized medical or administrative databases—surely has to do with what it is that digital interaction ofers: a mode of sociality, the afrmation of a sovereign self, pleasure in the construction and display of a public self, and the promise of '*genuine* individuation', such that 'disidentifcation will no longer be necessary as a way of maintaining individuality in a scene of falsely personalized address' (Cohen 2019: 174). In a context of unfathomably complex communications, the 'practical consciousness' of digital subjects works as though communication were unproblematically immediate and intimate and is built on an imaginary of 'sovereign control, a sovereignty of self-hood' manifested through willing personal disclosure (Bauman et al. 2014: 138). Such disclosure is part of 'a sharing practice involving mutuality and reciprocity rather than a one-way fow of information' (Raley 2014: 133); the gift of free labour to websites forms a community, a set of social relations, a commons. Constructing a profle and engaging in Facebook's 'Like' economy, for example, 'transforms users' afective, positive, spontaneous responses to web content into connections between users and web objects and quanta of numbers on the Like counter' (Gerlitz and Helmond 2013: 1358).

Digital disclosure, in generating value, is formative of social relations; this is that double movement by which the Internet takes the form of being 'always and simultaneously a gift economy and an advanced capitalist economy' (Terranova 2000: 51). Te social media profle and timeline and accumulated posts and the acts of friending and liking and rating make up 'a presentation of persons' (Koopman 2019: 7): a composition of the self that persists across time and across digital space. Made entirely out of data—out of stories, images, afects, arguments, observations—it corresponds to that other presentation of persons that is formed *without* my willing it from the algorithmic compilation and analysis of tracked online data and that may convert it into value. My profle, freely ofered to the world, is one of the dual sources of the profling, the 'reputation', that sells me to advertisers or that defnes me for the state.2

Taken together, the regimes of voluntary and involuntary disclosure thus construct what Goriunova (2019: 126) calls the *digital subject*, a concept that includes 'a subject of a data profle or of a Facebook stream, a history of browsing or search engine queries, mobile phone positioning records, bank transactions, sensor data, facial recognition data, biometric movement recognition data, or email inboxes, among other things. Te digital subject thus moves between captured, unique, and persistent biological characteristics and premeditated forms of symbolic expression, judicially inferred subjects of actions, and performed identities'.

One important way in which commercial difers from state surveillance is that in most instances the state works in the third person—it talks to itself *about* its subjects—whereas commercial surveillance converts its descriptive data into second-person address. Online advertising forms a vocative self: a self substantiated by the nameless and invisible voice that addresses me. Here's how it speaks to me—a few sentences taken at random from websites I've recently visited:

*Discover the 7 steps to harness your ambition and rescue your dormant business Lego Marvel Avengers: Create the Ultimate Quinjet. Shop Now Try Prime Video Free Go Now, Go Tere, Go Anywhere*

In each of these examples the pronoun 'you' is silently embedded in an imperative that works ambiguously as both an order and an invitation: an anonymous speaker addresses me as a subject who is invited or ordered to attend to an injunction. Te speech lacks authority, since I don't know who is speaking, and although these sentences are in the imperative mode they have no power to compel other than by awakening my interest—a remote chance, since I don't run a business, dormant or otherwise, or want to know what a Quinjet is or to accept an ofer that I understand is meant to hook me into a subscription I don't need, or to buy a new car.

<sup>2</sup>On what she calls 'ubiquitous online reputation calculation', cf. Rosamond (2019).

And this 'you' that is addressed to me is at once specifc and indistinct, neither singular nor plural but somehow both at once, a generalized addressee who is nevertheless me alone, the sole receiver of these words.

When the crooner asks, 'Who … stole my heart away?' and answers, 'No one but you', we don't know who this 'you' is, other than that he or she has the attribute of having stolen the singer's heart away; we don't know which person might fll this empty slot. We do know that it's not us, the persons listening to the song and in some, perhaps indirect sense being addressed by it: this 'you' passes to the side of us. But the identity of the 'you' can be subsequently specifed, either within the song—by being given a name, for example—or by the adducing of external information, perhaps about the singer's or songwriter's biography. In conversation between two people the contextual specifcation of the pronominal shifter is in the frst instance total: 'you' is the other partner to the dialogue. Given the citational capacity of all speech, however—our tendency to weave the speech of others into our own—the specifcation may be more complex. When more than two people are present in a conversation the reference of the second-person pronoun may require disambiguation; in this case the referent of the pronoun will be the most likely or most salient candidate in the contextual feld. In the case of written texts, the reference of second-person pronouns is always in a sense an act of simulation, a pretence that the openness of reference has always already been flled, that the nameless reader was always the one intended to receive this word. Seeming to single me out, personalized online address has the ofkey familiarity of intimate words spoken by a stranger.

Te uncertainty of deictic reference is at the heart of the process of contextual specifcation that we know as the interpellation efect: a process of conversion of a non-specifc into a specifc but uncertain designation of the pronoun 'you' (Chun 2017: 3 and passim). In Charles Fillmore's example, you are a young woman who has been wolf-whistled in the street. You want to reprimand the whistler but it's not clear whether you are the intended target (it might have been some other young woman in the street), and 'to turn around and scowl is to acknowledge that you believe the message was intended for you, and that may be taken as presumptuous' (Fillmore 1997: 59). Te uncertainty of reference applies both to the person who whistled and to the young woman, and in both cases, we know the *kind* of person who fts the description, but not the particular instances that would fll those generic slots. In Althusser's version of this, a policeman calls out 'Hey, you there' in the street and I turn around, assuming he means me (Althusser 2001: 118). Both cases represent a situation of ambiguity in which I respond to the message by appropriating it to myself: I fll it with my desire to be the one hailed or whistled at, (mis)recognizing myself in the pronoun uttered by the other as though it were personally addressed to me. Although Althusser's account is problematic, based as it is in a model of subjection to and by a sovereign power enacted through my response to the Absolute Subject—at once the State, the Father, and God—it nevertheless gets nicely at the mechanisms of imaginary singularization and personalization through which recommendation systems and targeted online advertising operate.

In algorithmic recommendation systems such as those used on music streaming platforms, the particularized musical identity of the addressee is constructed from the continuous collection and aggregation of contextual data points. Content fltering systems like Pandora organize music by analysis of its structural features and continuously revise their weightings as they match them with feedback from listeners, without regard to genre labels, cultural mapping, or demographic position. A collaborative fltering system like Spotify, by way of its Te Echo Nest subsidiary, takes this a step further by combining the outcomes generated by acoustic analysis software with 'semantic analysis of online conversations about music that take place every day, all over the world—millions of blog posts, music reviews, tweets and social media discussions' (Prey 2018: 1090-1). Overlaying on these analyses a preference analytics that captures3 and records in real time a listener's musical behaviour and preferences,4 Spotify treats the cultural mapping of music as a further insight, diferentiating music that is structurally similar in accordance with the highly diferentiated taste cultures of the digital world. Te 'paradigmatic claim' of such algorithms is 'to specify the individual in the

<sup>3</sup> Seaver (2019) theorizes recommendation systems precisely as traps.

<sup>4</sup>Cohen (2019: 173-4) notes that 'Preference isn't about identity. It's about ranking, which is why it's so useful to the personalization industry – it tends not toward identifcation, disidentifcation, or judgment but toward measurement and quantifcation, which then come to infect subsequent scenes of (dis)identifcation and judgment'.

complex conjugated personalized address: 'People like you like things like this' (Lury and Day 2019: 24). What these and other recommendations systems, like those of Netfix or Amazon, have in common is their generation of a 'you' that is not based on fxed markers of identity, either demographic (class, gender, age, ethnicity, and so on) or generic (jazz fan, Christian rock fan), the properties of which are presumed to be known in advance. Rather, they generate personalized recommendations from categories that emerge from a process of recursive revision.

If in a formal sense the profles and 'reputations' constructed for us and addressed to us by advertising and recommendation engines have no content other than the acts of recognition or misrecognition—of imaginary personhood—that transiently fll them, it is nevertheless the case that these shifters are constantly being specifed contextually through acts of rigid designation that seek to tie them to a name and a legally established identity (and that are just as constantly resisted by acts of counter-naming or heteronymy or masking). In practice this means the construction of data-shaped personal selves—data shadows or data doubles—across online databases, where information freely ofered on social media as a referential truth (self-expression, life-writing, autobiographical timelines, and so on, however fctive these truths might be), or volunteered to state or corporate databases, or captured from phone usage and location tracking or from facial recognition systems, is fastened to a persistent identity by a kind of *point de capiton* pinning my online transactions and pathways to the ofcial documents that are the baseline of my composite existence. Te development of cookies and of even more persistent tracking IDs such as Flash cookies or web beacons has been at the heart of the ability of the state and corporations to silently monitor my activity in this more or less integrated way across convergent sites and devices. Although for many purposes on the Web it doesn't matter whether anyone knows you're a dog, and much information harvesting and analysis is concerned not with 'the personal identity of the embodied individual but rather the actuarial or categorical profle of the collective' (Hier 2003, cited in Cheney-Lippold 2011: 177), these tracking and fastening devices nevertheless allow in principle for the attachment of a corpus of data to my name, and of my name, together with the descriptive attributes it brings with it, to my embodied self. Stalder (2010: np) distinguishes three types of profles, which together create a comprehensive profle of each user: 'First, users are tracked as "knowledge beings" and, in a second step, as "social beings," exploiting their real identities, contacts, and interaction with those contacts. A fnal data set captures the users as physical beings in real space'. We might think by analogy of the legal identifcation of criminal culpability by means of documentary evidence such as eyewitness testimony or forensic analysis, where the indexical tie to the body of the culprit is established by a witness's sworn statement that they have seen this person's body or by the traces the body leaves at the scene of the crime: here too there is nothing but data, nothing that we can call simply a truth; but the verdict, the truth-saying, of culpability is given by the accumulation of those traces of information.

Te person addressed by the second-person pronoun or implied by deictic markers or captured by stateful trackers is not a substantial particular, a self-identical presence, but the occupant of a semantic place; the space and time for which he or she serves as a reference point are constructed in dense networks of metaphor, and the body that orients that person in space and time is imagined and positioned through these networks. Since the place of the shifter may be occupied by anyone who is addressed by it, the 'you' is structurally riven, positional and alienable yet embodied, a reference point in time and space and yet movable from discursive point to point, a fgure in a statement (Frow 2014: 164). Yet is this shifting 'you' not grounded in a material and experiential reality that *occupies* the pronoun and that we experience as the solidity of a selfhood? In one sense, it quite clearly is. My body can be arrested and thrown in prison; it can be tortured or killed. I can be enticed to spend money that I have earned by virtue of real physical or intellectual labour; I can go into debt and undergo material hardship when I lose my credit rating or my right to work. We could, therefore, posit the self of experience, as William James or Alfred Schutz do, as taking place in a feld of deictic reference encircled by the body and mobilized in my face-to-face interactions with others, and then grant that everyday experiential self priority over the more remote modes of selfhood engaged in the worlds of secondary representations. We could, that is to say, posit a necessary gap between my embodied everyday self and my data self, my data double, my data shadow; between the 'I' that I live from the inside and the 'you' that is directed to me from without or the 'he' or 'she' that describes me in a database.

Yet thinking about the relation between real and algorithmic personhood in this dichotomous way is, I think, both conceptually and politically unhelpful. It posits, in the frst instance, a body that is distinct from the information that shapes it. As Irma van der Ploeg has argued across a series of papers, the increasingly prevalent translation of aspects of our bodies into digital code is not a matter of changed representations, with 'the thing itself' remaining the same (van der Ploeg 2012: 178), but a fundamental change at the level of ontology, since 'there is no clear point where bodily matter frst becomes information' (van der Ploeg 2003: 70). Te body through which we apprehend ourselves and others may look like a ground truth, but that body is not a pre-discursive matter. It is information in its substance and its processes—in the DNA that composes it, in its homeostatic regulation of the endocrine, immune, and autonomic nervous systems—and in all of the systems (the regime of state security, the state welfare and schooling apparatus, the insurance industry, the taxation system, medical databases, regimes of visual representation) in which it is inscribed. Likewise, for the person who is and who understands themselves through their body, this is nonetheless a body experienced through a bodily imaginary, the efect of a 'system of exchange, identifcation and mimesis' in which I shape my sense of myself by way of a recognition and incorporation of the bodies of others (Gatens 1996: 31). It is through this imaginary body that my fundamental fantasies about who I am and how I engage with others are shaped.

In the same way, to think in terms of a dichotomy of digital and real personhood is to posit too stark a disjunction between representations of the self and an ofine actual self. Te digital subject—the vocative self of advertising, the data double of surveillance systems—is not an external representation but the constantly mutating efect of 'the practices through which one *becomes* data through interactions with numerous other actors and actants' (Ruppert 2011: 225). Taking issue with the concept of the data double, Koopman (2019: 170) argues that 'data has become a crucial part of the very terms by which we can conduct ourselves. We are our data. Terefore we are precisely not doubled by it'. Interaction with data, whether voluntary or involuntary, witting or unwitting, is integral to the actuality of our selfhood. Tis is a matter of a pragmatic and contingent formation of digital personhood; the algorithmic subject is sustained by the interplay of systems of ubiquitous surveillance but also by the 'unique combinations of distributed transactional metrics that reveal who they are' (Ruppert 2012: 124). I become who I am through my engagements with the real and the digital worlds, and the diference between those realms is increasingly tenuous. Tis data self that I 'am', however, is never singular: we can't speak of digital selfhood as a consistent aggregated identity formed across databases, since diferent databases measure and construct diferent realities: a reality of consumer desire, a reality of cultural preferences, a reality of political convictions and actions, a reality of economic capacity, and so on. Against the impetus of the state, and perhaps our own habitual impulse, to reduce identity to a single point of reference, we can at most speak of a plurality of more or less convergent, more or less heterogeneous forms of personhood, none of which is an absolute ground.

Tese distinctions may look scholastic, but I think there are strong reasons for teasing them out. In the case of my initial example of identity theft I tried to demonstrate the futility of any appeal to a real and embodied self. What counts in practical—that is, in administrative—terms is not the body that I am but the forms of documentation that make me up and the way they ft together. Similarly, resistance to state or corporate surveillance can't be grounded in appeal to the fundamental and singular reality of my person but only in alternative ways of fguring myself and of challenging or ignoring the specifc forms of fguring and naming that construct and address me. 'Figuring' here means both calculating and performing the form of the person, and it comprises the acts of recognition that construct me as 'an element in a classifying series' and thus as a governable subject. If we are to understand the new modes of personhood of an evolving world of information technologies and self-educating machines, we must understand the complexities of new systems of construction and extraction of value, the extending universe of ubiquitous surveillance, and the changing forms of address that situate me in this universe.

'We', I said: 'We must understand'. But what 'we' must understand is the slipperiness of these pronominal shifters and the way they construct communities of understanding which are far from self-evident: who is this 'we', who is the 'me', and what's the status of the slide between them that I performed a few sentences ago; who is the 'you' I'm addressing in the now of this room, and in the non-time of this writing: you present and absent, 'you' singular and 'you' plural, and what kind of plurality does that singular become? 'We', whoever we are, must above all learn to be distrustful of the communities we invoke and of the 'you' that invokes us and with which, in this time of speaking and this non-time of writing, I address and invoke you now.

#### **References**


Snowden, Edward. 2019. *Permanent Record.* London: Macmillan.


**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/ by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **13**

# **Figuring Accompaniment: The Creation of Urban Spaciousness**

**AbdouMaliq Simone**

# **Something Else Besides**

At the height of urban India's initial stage of the pandemic, there was much attention to the fgure of the desperate migrant setting of by foot, often thousands of kilometers from their home villages. Faced with the shutdown of factories operating on non-existent margins and the loss of accommodation as well as livelihood, there was little choice but to head toward settings where they could, as was the common refrain, "fgure things out." Usually caught within the repetitive rhythms of 14-hour shifts with little disposable income to circulate beyond the itineraries of work to hostel to work, there were few opportunities for these workers to grasp the larger setting in which their labor was situated, even when they had resided in a particular city for years. Income was to be remitted elsewhere, as was the directionality of everyday communication, with distant parents, children, and lovers. Te artefacts of cheap consumption were usually available at the factory gates.

A. Simone (\*)

Urban Institute, Te University of Shefeld, Shefeld, UK

Tere were those who, for whatever reason, amid many, did not venture "home," who remained in the city, even as they were compelled to set of in random circuits in the search for temporary food and shelter. Unable to be stably absorbed into any economic function or provisioning mechanism, these workers wandered across a landscape of details—a sudden burst of arguments between street sellers desperate for a sale, the glittering of rust on an abandoned car in the late afternoon sun, the lines of fowing water from a cracked pipe. All of these details became roadmaps, pointers to take particular directions, and then the discovery of empty schoolyards where one could sleep or the back stairwell where things were half-discarded.

For those who had rarely exercised voice, this landscape of rampant detachment, of all of the ways in which people, buildings, and materials were not connected to each other, provided a tentative platform for their growing confdence to strike up conversations with various passers-by, ofer outlandish propositions about money-making schemes, or most usually to comment upon some element of the other person that had imbued them with great signifcance. Without apparent commitments or attachments, without the luxury to compare themselves to others and refusing the obligation to always think of the others to whom they had been attached, these excursions revealed all varieties of spaces that didn't seem to be committed or attached to the uses they would seem to purport.

Tis was not only a matter of schools being turned into vast communal houses, markets into mathematics classrooms for orphaned teenagers, shrines into all night popular political assemblies, or government tax ofces into repair shops for homemade inventions. Rather, they were all those spaces in the midst of things, within crowded thoroughfares, jetties, underpasses, hallways, and arenas that didn't quick ft with what was happening around them. Spaces ever so slightly out of joint, where the anomalous, the marginal, while clearly visible remained ever so slightly indetectable, enough for moments of rest, the rehearsing of some kind of weirdness, a base to build a modicum of confdence to venture forward or back. Instead of being preoccupied with fguring out what would become of them, what would be their fnal destination, these workers saw themselves like these spaces ever so slightly out of joint; saw themselves as *accompaniments* to the "normal" goings-on; saw themselves as neither adding nor subtracting, rectifying, or disturbing, but as something else *besides* what was taking place.

What would happen if we viewed fguring as involving accompaniment, or as always also accompaniment; something that does not discernibly alter the visual and sensual dimensions of an event or entity, that remains apparently aloof from its confguration, but which nevertheless prompts a reorientation of view and engagement; which at least raises a degree of uncertainty about what it is we are confronting in an appearance that otherwise has all the hallmarks of an integrity and coherence. For all of the anxiety often demonstrated about securing defnitive boundaries for self and other, for collective lives sufused with sufcient commonality to ensure reciprocal recognition and mutual obligations, accompaniment is permanently out of place, disinterested in whether that which is accompanied assumes a particular fgure or not.

For, accompaniment means something that operates aside, on the side, that does not entail obligation, nor a manifestation of mutual desire. If I accompany someone, it does not mean that the person accompanied could not accomplish the task on their own. Someone can still perform the "solo" without missing something essential. Accompaniment is a supplement, that shows up, now and then, goes along for the ride. It is not unafected by the going along, but it is not essentially invested in the outcome of the task at hand; it does not constitute a debt to which the recipient is owed, even though such a debt economy might ensue from a particular accompaniment.

We are accompanied by an array of "companions" throughout the urban environment. Rather than seeing the built environment as the stage through which to exercise our privileges, or as the concretization of aspirations, needs, and accomplishments, the built environment acts as an accompaniment to whatever we do. It pays attention to our practices; it bears witness to our travails and attainments. Tere is always something not used or only partially used, something that remains just out of reach, something barely noticeable or deemed irrelevant that accompanies all that is standard operating procedure, all that are demarcated, sectored, and zoned spatial arrangements. Accompaniment is a submergent infrastructure that suggests something else than what is recognized.

Figuring is often construed as a process of things "closing in on themselves," of accentuating the distinction between fgure and ground through which the outlines of a stable entity might be construed; stable in the sense that we might come to know what to expect of it, that is rendered in a form recognizable through multiple iterations, that holds "its own" amid conficting expectations. Figuring closing in on itself suggests the constitution of a target, something to take aim at, as it also embodies its own aims, and thus further suggests the primacy of straight lines, grids, and probabilities. Te generalized conversion of the world into multiple targets, where specifc populations are targeted for specifc policies and probabilistic action, clearly has intensifed the compression of space-time, as everything is reachable through less dense and circuitous mediations (Bishop 2018; Valayden 2016). Te generalization of the commodity form and its elicitations and compressions of singular afect turn individuals into entire worlds and, at the same time, fracture them into infnitesimal pieces of codes, biomes, body parts, behaviors, and inclinations (Law 2015; MacKenzie 2016) . Just as the urban, for example, is populated by an exponential explosion of objects, data points, and niche markets, spaciousness is reduced, space closes in on itself.

Acknowledging the accompaniment of fguring is then a restitution of spaciousness, a sense of individuation, of contributing a sense of the improbable to worlds fading in their distinctiveness. As Stiegler (Stiegler 2018) reminds us, the continuous updating of the fgure through its subjection to repeated runs of relational calculation, of fguring its constant re-positioning in terms of shifting relations to a constantly expanding set of "neighbors"—of things and events that might have relevance to its operations—produces a generalized blurring. Tis is a sense where it is not only difcult to fgure things out by getting a hold of the fgures that might be involved, but where fguration itself dissipates in the profusion of the technical. Rather than zeroing in with the precision of determining the defnitive coding and composition of fgures, accompaniment potentiates the spaciousness required for fguring to endure its availability to digital architectures and the concomitant simulations and fabrications that can ensue from algorithmic re-composition.

As Denise Ferreira da Silva (Ferreira da Silva 2018) demonstrates in her play of fguring, of subjecting the fgures of patriarchy, femininity, and racial identity to their own algebraic inversions, these fgures can be decomposed in ways that enable a sense of incalculable spaciousness to emerge. If the fguring of specifc bodies is based on contractual, juridical arrangements—the terms of recognition, rights, responsibilities, and value—then potentialities of what exists are appropriable only through extraction, and thus the exercise of violence. Te fguring of the juridicalethical edifce of the properly human fgure takes place as propriety through property; in other words, through the proper management of property that is fundamentally unruly, and in need of management. Black life as property, thus, has existed as that fundamentally unruly, chaotic potentiality that needs to be properly managed so that its resource can be extracted and deployed. Tis potentiality is subject to contract, to a particular set of equations where specifc fgures embody specifc rights of management and sets of obligations. Te ability to manage property becomes then the exercise of liberty. What da Silva attempts is a mathematics extricated from contractual relation that frees the fguration of the calculable to incomputable potentialities, imbuing a sense of spaciousness to the process of fguring.

Even within the confnes of the contractual relationships which defned the black body as a captured object on which the infiction of violence was necessary in order to bring its potentialities to life, the subsequent deformation of the fguration of gender, its reduction to the amorphousness of fesh, posited possibilities of extensionality of and among bodies that both portends the fungibility of human life yet, at the same time, potentiates other more uncertain formulations of bodies not easily decipherable according to the convention terms of liveliness and scale. For across traditions of American black thought, the sense of accompaniment has been an omnipresent characteristic of everyday life. Rather than viewing the relationships of bodies, land, animals, plants, and other materials as part of an integral ecology or integrated metabolism, there was rather the sense of these things accompanying each other, passing through and among each other; where each could be "called upon" as some available exteriority to lend a hand, to get through a particular conundrum (King 2019; McKittrick 2006; Wynter 2003). Each had its autonomy, its situatedness in other worlds, but at the same time was always on hand, even though what exactly was called upon was often never clear, rarely assumed the fguration of a divine force or a specifc identity, but even so, whose presence could be recognized. Across the manifestations of "long walks to freedom," to cellphone recordings of police violence, to the multitude of minor refusals to being pinned down in the incarceration to proper relations, black lives have accompanied each other in the ebbs and fows of a collective fguration that does not consent to any particular set of recognitions.

In a not dissimilar fashion, Dhanveer Brar (Brar 2021) encourages thought on an ecology of black generativity and constraint far beyond the "caricatured landscapes of post-apocalyptic urbanism." Grasping the mineral interiors of three instantiations of black electronic music and their embodied sensory intelligence and antagonisms, he focuses on the way in which music operationalizes specifc capacities to navigate and refgure the confuence of racialized precarity and enduring capacities to create life beyond the normative vernaculars. Chicago, Detroit, and London are rendered as transversal, oscillating planes of urbanity curating and dispersing sounds, propositions and maneuvers, which not only take blackness to the world, but create an experimental world from blackness that reverberates within the midst of unequivocal oppression, providing an enhanced spaciousness. He calls attention to the ways then that urban residents are living in contexts that both exceed the fgures through which we understand them yet which penetrate our very core as metabolism, infection, and vibration. Not simply reducible to spirits of resistance, resilience, or reserves of long-honed creative practice, the navigational instruments operationalized through black music are instead the concerted deployment of machines, reinventions of sonic architectures, as well as the deliberate and systematic workings of bodily and cognitive capacities, whose fgurings are less the production of image but force.

# **Urban Gathering and Arrangements**

In *Surat l-isra*, Allah says, *"If mankind and the djinn gathered in order to produce the like of this Qur'an, they could not produce the like of it, even if they were to each other assistants."* While pointing to a gathering beyond any specifc gathering as that which produces the Quran, implicitly this surah raises the question of what kind of gathering could that be among humans and djinns1 , with humans representing a particular consciousness and djinns perhaps embodying what could be called a "worldly sensibility" excessive of the registers of that consciousness.

In this accompaniment of djinn to human, gathering here is a conjunction of actions rather than the cultivation of a higher, interior consciousness. In bringing together the manifestations of distinct forms of sensing, not readily compatible, not readily translatable each to the other, there is a gathering that does not merge, does not integrate. It draws our attention to the ways in which complex urban environments are sensate agencies in and of themselves, even if such environments may not be clearly organized into a series of readily identifable entities, machines, or agencies. It draws our attention then to the kinds of capacities and sensibilities embodied by particular environments and the use they make of and draw from particular forms of inhabitation and populations of inhabitants—both human and djinn.

Tis points to a critical question today: How do residents largely unanchored from their long-honed stabilities observe the situations that require from them new ways of seeing, new dispositions of circumnavigation? How do they attempt to reconcile their "blind spots," to read themselves into the surrounds, into a background that senses their existence in ways inaccessible to them? How do they compose points of views, angles onto things capable of observing prospective trajectories across time and space—where observation is a process of composition that requires gathering up potential collaborators diferentially distributed across personal networks and territories of operation?

To use Day and Lury's (Day and Lury 2014) term, how do residents *render*, gather up, and turn over or turn out particular visualizations of their urban contexts and urban problems? Particularly in ways that publicize knowing when they do not know, and not knowing when they do, so that their liveliness is not fully captured by the particular and

<sup>1</sup>A species of invisible beings created by fre frequently referred to in the Quran that proceed the existence of human beings. Tey are viewed as an integral accompaniment to human existence, a supplement at times generative or destructive. Djinns are frequently viewed as the essential nonhuman guarantor of human creativity and fallibility.

oscillating expectations of capitalist surveillance. With so many factors at diferent scales at work in rendering places knowable and subject to multiple interventions; with so many alternating interplays of shadow and light, how do residents of volatile, ever-shifting urban terrains garner a sense about how to move, what to do next?

So when I say that *djinn* might represent the accompaniment of a worldly sensibility to human will, I point to a sensibility that implicates a body or agent (as a unity of experience) in ways that exceed capture of, to or by any defnitive institution. Mediated social-technical circuits are not subsumed to the intentionality of any one agent. Our agency, rather, is implicated through and by these circuits. Tis is an enactment of agency not bifurcated in terms of self and other, human and more-thanhuman. It is a mode of engagement with an environment not directed by conscious intention, but rather an intersecting of multiple operations. A body recursively incorporates the feelings generated by immersion in crisscrossing data streams into specifc embodiments of observation, of attending to the surrounds that exceed the conventional vehicles of sensing or its distribution into the perceptible and imperceptible (Massumi 2017). Tis is what happens when human and djinn work together.

For our shaping, an enduring performance for the world always must operate through that which cannot be seen, and increasingly the deployments of information environments for purposes of surveillance and domestication make it urgent for there to be bodily operations that might remain imperceptible, under the radar (Citton 2017). As such, the rhythms of endurance are not about the resilience of human life, about the never-ending resourcefulness of a subaltern imagination. It is not about a virtuous general ecology that, in the end, works out a functional recalibration of elements each diminished in their own terms, each insuffcient to the replenishment of the other. Rather, endurance also entails the actions of bodies indiferent to their own coherence, where bodies churn and stave of death in their extension toward a liveliness of things in general, and where bodies become a transversal technology, as gesture, sex, gathering, and circulation operate as techniques of prolonging (Wills 2016).

What is it that a person pays attention to in a world where so many things can be attended to; how does one create a spaciousness of operation when life can theoretically be moved and oriented in so many diferent directions. Here the notion of *arrangements* becomes critically important. Unlike the predominant use of *assemblage* to connote the intricate compositions of materials, events, forces, and entities that constitute the salient fgure of subjectivity, particularly for urban operations, *arrangement* seems to decline the primacy of attention to composition and rather amplify acts of *accompaniment.* As normative protocols of relations remain largely tied to the contractual, to familial forms of obligation, or to the afective intensities of libidinal attachments and the reciprocities of friendships, arrangements, while folding in bits and pieces of conventional contextual categories and relational processes are not subsumed by any of them. In a refection on the ways in which symbiosis and parasitism characterize relationalities of the biome and viral as metaphorical substrates of the human, Chun-Mei Chung (Chung 2020) points to arrangements as those "intelligence operations and complex linkages that are concealed, dark, secret, and challenging to see, rewriting internal and external boundaries." While not precisely mapping onto the kinds of urban arrangements I invoke here, Chung, nevertheless, points to the ways in which arrangements straddle uncertain lines between the generative and debilitative, that do not announce themselves transparently as having particular objectives in mind, and which are not clearly mappable in terms of their reach or even in terms of what they do.

Tis is not to say that genealogies of arrangements are impossible to conduct. For years I have attempted to examine the ways in which production systems that entail vast subcontracting across hundreds of homebased workshops generate large volumes of clothing and hardware. Tese are certainly arrangements structured by brokerage, monopolistic control of supply and distribution chains, and exploitative pricing systems abetted by skewed regulatory frameworks. But the process of securing the stability of these arrangements, making sure every actor and component adheres to the specifed transactions, also seems to entail a process of sporadic and largely indecipherable lateral relations, where workshops elaborate their own largely provisional distribution networks for of-thegrid productions, and where it is not clear who is in charge, nor how the operation in its entirety actually works. For here, little is consistent; sometimes things work, sometimes they don't. Tis temporality of the *maybe* suggests not necessarily a defcit of integration, but a stuttering, a process amenable to interruption and detour, where the diferent workshops, their workers, the brokers, the stalls in the market where the "illicit" output is eventually sold, and the wildly fuctuating tastes of the usual consumers *accompany* each other with a looseness that suggest fortuity and chance rather than strategic planning.

In a large night market across multiple streets in Kebayoran Lama (Jakarta) that supplies meat and fresh produce to the hundreds of small local markets across the region, the daily functioning requires a wellorchestrated choreography of directing supply trucks into the market area, unloading the trucks and distributing the goods, of retrieving and setting up tables on which the produce will be displayed, of shifting goods around to meet the needs of regular customers, of extracting various market fees in which to pay of various local ofcials and police as the operations of the market are ofcially unsanctioned, and then of dismantling everything and cleaning a large swathe of otherwise public space for by 8 a.m. there is absolutely no trace that this market even existed. Clearly all of the functions have to be interconnected, arranged, but on any given night those who were porters become sellers, drivers become cleaners, sellers become fee collectors, and so forth in a system that remains unspecifed, reliant upon unforeseen initiatives that shift the division of labor around. Te plausibility of these shifts are of course anchored in the eventual capacity of everyone involved to do every conceivable job, but there is no underlying reason why this has to be the case, or when it becomes the case. Each actor and function simply appears to accompany the other, both symbiotically but also with a fundamental looseness that detracts from the need for a coherent fguring of the organization of the market. Rather, the market seems held together on the basis of how easily it could fall apart.

Indeed, arrangements sometimes become visible only as they shift over time, when their *stagecraft* becomes apparent, in the taking to the stage of specifc constellations of assessment, brokerage, mutual attentiveness, provisional rules, and collaborative practice. For example, when the implicit governance systems of migrant hostels, no longer articulated to the subsidies and salaries paid out by formal employers, have to shift to new arrangements of resources, social connections, livelihoods, and social identities. Place-based arrangements may be converted to more transversal collaborations; the application of specifc lenses of institutional readings of given contexts may necessitate readjustments in the ways in which resources are allocated and adjudicated. Household functions may be redistributed across multiple locations, where a single address serves more the pragmatic function of having an address than representing a coherent household unit. Each re-arrangement has its own subjective condition, which lasts as long as it works, and where how long it works depends upon who is paying attention, in what ways, and for what reasons. Particularly during the pandemic year of 2020, when a slew of restrictions were placed on public mobility and the pursuit of everyday livelihoods, it became evident the extent to which the appearance of predominant forms of social organization were simply a veneer of rationalities that had little traction in the actual ways in which residents organized places to sleep, procedures for attaining and allocating available resources, and tacit rules for the use and distribution of space in ways that continuously contested clear divisions between the proprietary and non-proprietary.

Ofcially localities might be the conjunctions of formal households, property, zones, discrete institutional competencies, and authority relations, but when faced with exceptional conditions, they were then visualized not just as compensating for these conditions with special arrangements of all kinds, but simply extending, recalibrating, or improvising upon a social economy of of-the-grid arrangements that had already existed over time. While it is certainly possible to elaborate a symbiotic relationship between that ofcial veneer and the plurality of makeshift arrangements that could be seen as underwriting them, it is more a matter of viewing them as mutual accompaniments, only loosely sutured in ways difcult to calculate. For in most respects, drawing upon Harootunian (Harootunian 2000), such arrangements are unremarkable, embodying little subjective depth and rather existing as a series of horizontal displacements, where one concrete manifestation does not clearly translate into any other.

Additionally, a sense of spaciousness can be produced through acts of refusal, of keeping things apart rather than putting things together. Instead of querying puzzles as pieces that must inevitably ft, there are relationalities that may maintain a sense of proximity among processes and problematics that are not necessarily resolvable through integration or synthesis. For example, if specifc libidinal attachments remain salient in relationships to family, household, community, and state, what happens to libidinal economies as particular roles and responsibilities are refused. Rather than indicating social breakdown or disintegration, what happens when detachment simply signals a tacit refusal to "tie up" one's desires into the expected formats. For example, what happens to the desires for familial afection when fathers refuse to attach themselves to the purportedly normative behaviors that constitute fatherhood; what happens when women concretely manifest indiference to the presence of men as an integral aspect of household composition? What happens when the role of "worker" no longer provides a valorized basis through which individuals identify themselves and their worth? What happens when localities refuse to refect compliance to the prevailing standards of viability? How and to what extent are familiar obligations recalibrated in new terms or simply refused altogether? What kinds of arrangements mark a detachment from the normative tropes of socialization? Here arrangements as matrices of accompaniments do not so much constitute new coherent fgures as they work through and around the vestiges of those no longer operative in the ways originally intended. Rather they function as ciphers marking a series of detachments that keep each element in view but without settling into stable forms of clearly delineated features or responsibilities.

# **Figuring the Extensions**

Te urban has long ceased coinciding with the fgure of the city. If by the city we mean a densifcation of agglomerations among materials and bodies expressing distinct locational advantages, territorialization of accumulation, and the rise of intensely individuated performances of citizenship and self-fashioning. Te city-form is a mode of refexivity capable of territorially binding the intensities of relations to a coherent synthetic operation based on the integration of diferentiation—that is, work detached from the primacy of household-centered economies, re-socialized as industrial labor, continuously elaborated through networks of servicing and management, abstracted as elements of fnancialized calculation, and availed increasingly individuated opportunities of consumption and social identifcation.

If the fgure of the city disappears, or at least is partially dissipated in the extension of urbanization processes across more pluralized dispositions, through peripheries, hinterlands, corridors, conurbations, and regions, are there appropriate fgures capable of marking, cohering, or imagining this process. Te typical conventions have been those of volume, such as the "megacity" indicating exponential expansion, or that of a supplement, such as "urban region" as the city plus something else. "Urban periphery" has long been invoked as a way of marking a transitional or liminal space, as that which awaits the city's "arrival," or as a modality which transfers the city across space, the very means of its extensionality. Most contemporary theorization of these extensions, however, marks a disjuncture with the fgure of the city and views their spatial histories as refecting more dispersed, erratic, and polyvalent articulations (Keil 2018; Lefebvre 2014; Monte-Mór 2014; Schmid et al. 2018).

It is not a matter of extending the city into new territory but in the simultaneous intercalibration of very diferent logics of settlement and production, an interweaving of divergent tenure regimes, land uses, and modes of inhabitation that instead of settling into distinct patterns of agglomeration and inter-connectedness are continuously disturbed and re-oriented through additional spatial products and development initiatives. Here, the intensive contiguities among industrial estates, peasant farming, upscale mega-housing developments, the voluminous rollout of cheap pavilion housing, expansions of informal settlement, premium logistical infrastructure, and feral landscapes resist any clear governmental integration despite intricate spatial planning. Projects come and go, often with wildly diverse temporalities and efcacy; ongoing development is not a matter of whether past projects have proved viable or not as they are informed primarily by a sense of *eventuality—*that is, eventually whatever is developed will acquire some proftability even if the terms for that are not presently available. As such, it doesn't matter the extent to which industrial zones, housing developments, and commercial estates may remain half-empty, for the addition of more spatial products is seen as engendering a new context for what already exists, in a constant repositioning of the built environment into something else that may then prove key to a renewed sense of viability.

In Jakarta's massively expanding urban region, emerging metropolitan areas such as Cikarang embody intensive mixtures of logistical apparatuses—internal ports, high-speed and light rail systems, new freeways and air cargo ports—new town developments, such as the present construction of 250,000 apartment units, six universities, ten hospitals at Meikarta, and an array of tens of thousands of migrant hostels, low-end housing and mid-level commercial zones spread across the metro. All of this is set as yet another iteration of residential and industrial development that has been underway for the past four decades. Te subsequent diversity of layering and sedimentation that takes place intersects obdurate economic functions that co-exist with land uses and projects that have changed repeatedly over short time spans. While developments may be spearheaded by the combination of major landholders transitioning into major regional politicians, aided and abetted by the profusion of a new generation of small-time brokers, and Indonesia's major real estate developments and fnancial institutions, the political and technical power brought to bear is unable to cohere these extensions within the conventional planning tools or protocols of speculation (Firman and Zul 2017; Herlambang et al. 2019; Shatkin 2019). Rather than work as a series of coherent synergistic or multiplier efects, the discrepant elements seem to simply accompany each other, all exerting some kind of infuence in a semi-detached state, but without a clear sense of proportions involved.

For what might be considered subaltern actors in this region, there seems to be a strong reluctance to contribute to any coherent fguring of what is taking place, and rather an investment in capacitating the very looseness of relations among the discrepant elements of the built environment and the diferent logics of accumulation at work. Along the raised embankment of an irrigation canal that now separates corporately held agricultural land from the almost magical appearance of Meikarta, residents originally from the island of Madura, across from Surabaya, have long operated from hundreds of makeshift compounds, with their various assortments of junk, found and stolen items, including steel beams, bags of concrete, broken door frames, thousands of bolts and screws dismantled from who knows how many infrastructure projects. Renowned as artisans of the "useless" and providers of what anyone needs for almost any kind of project, the Madurese are the consummate archivists, rarely discarding anything, and talking about and arranging their "wares" in such a way as to propose interconnections among things that might often seem outlandish and impossible but nevertheless of potential value to an audience that seems to take many of these propositions sufciently seriously to maintain these archivists in business. A row of cheap migrant hostels, for example, abandoned because of internecine confict or simply bad positioning in face of food drainage, can be completely dismantled in a matter of hours and the components reinserted in wide range of repairs, house extensions, junk markets, and small factories before the day is over.

Te Madurese are not only collectors of materiality but also cheap jobs as well. Tey won't usually do the jobs themselves because it impinges upon their sense of freedom, but collect them to be distributed to others—for example, particularly porters, janitors, cleaner, and security guards. Te objective is not so much job-*placement* per se, but brokering connections among diferent jobs as part of an expansive information network, which circulates updates about what is taking place across different factories, construction jobs, internal customs ports, and service centers. Such a network not only facilitates the "just of the truck" acquisitions of materials or the ability to ofer "quick solutions"—material inputs—to projects or operations facing unanticipated problems, but concretizes "of grid" relations among places and functions, that is, those that do not ft into any of the prevailing conceptions about how things and places are to be connected to each other.

Tis positing and materializing of "of grid" relations is not conducted within the register of realizing unexpected potentialities. It doesn't concern itself with developing alternative worlds or inventive usages. Rather, it functions as intensive artifciality, even noise; a means of interrelating things not informed by a specifc vision or even objective. It concerns an infusion of incomputable instrumentality in the intersection of the quotidian experiences of hundreds of "service" workers and laborers across a landscape characterized by moving things around, constantly improvising where they might ft, disrupt, and supplement operations of almost any kind.

Te Madurese are constantly on the run. Even within their internal dealings along this irrigation canal, it is unclear what relationships one makeshift compound has with any other, or whether any "project" is simply the result of individual brokerage or some kind of intricate collective choreography among them. It is not clear whether or not a tacit moral economy of sharing markets, a complementing of distinct networks, or a fortuitous interweaving of competition is at work. What is evident is a very loose sense of any afliation with *property.* Madurese are stereotypically known as thieves with almost extraordinary powers and agility, as well as being indiscriminate in terms of weighing the relative value of whatever they can get their hands on. Nothing is deemed either waste or luxury, even as they are known for driving a hard bargain around anything they attempt to get rid of. Tey certainly know the market price and how to set it. But any sense of *propriety with property* is far removed from daily operations that attempt to draw lines across the "backdoors" of nearly everything that exists in this area.

In amplifying the essential brokenness of the world, of things out of their *proper* place, no matter where they end up or how they are used, this economy goes beyond reparation to highlight how that brokenness suggests its own propositions devoid of the will to restore functionality. Te Madurese, known for breaking the integrity of projects, repurpose elements from that brokenness to dispositions that they have little interest in defning, but rather seek to perpetuate a state of brokenness as generative of a continuous circulation of materials across diferent hands, diferent sites, and diferent uses.

Here, relations are proposed that are detached from obvious genealogy, that compress things conventionally viewed as impossible to be together, and that have no way of knowing whether they will endure or not. Tis techno-poetics of relationality implicitly addresses the fundamentals of urbanization itself, that is, as a process simultaneously human and inhuman; that does not proceed simply as an artifce of human will, but as a *techne* both with and *without* its own registers and afects (Simondon 2009; Simondon 2017). In other words, the *technical* dimensions of the relationalities of urbanization come from all over the place, and work in diferent degrees, proportions, and manifestations that come to be associated with it but also do not intrinsically belong to it. Tis is because there is no essential overarching fguration attributable to urbanization outside of its profusion of technical relationalities—its capacity to continuously repeat everything we might know about it, and upend itself at the same time.

### **Concluding the Surrounds**

In the clamor of countervailing projects and logics at work in generating contemporary urban inhabitation and operation, what constitutes viable modes of fguring able to navigate the intricate physical and social landscapes of discrepant times and strange spatial juxtapositions? Instead of envisioning processes of urbanization as the unfolding of defnitive forces of value capture, asset creation, and resource extraction, how are these albeit salient categorizations of spatial production accompanied by a growing multiplicity of entities and their exertions? Particularly at the extensions, just beyond what has customarily been purported to be "the real city," it is increasingly evident that a continuous recalibration of "projects," material inputs and residues, and altered ecologies of reciprocal causation are generating landscapes that exceed the salience of available vernaculars of analysis and intervention. Here, intricate landscapes of provisional sutures, half-lives, difractions, disjuncture, compensation, and transience create unsettled urbanities and populations.

Here there is a play on the interrelationship between fgure and (back) ground, even as this couplet is incessantly reproduced. On the one hand, it is clear what is taking place in these extensions of the urban, replete as they are with now easily recognized spatial products. We seem to know where they are going, even when it is likely they may never reach their "destination." Tey hold forth a seemingly contradictory *promise*—that of a capacity to encompass greater numbers of persons into the predominant tropes of urban productivity; providing assets and opportunities, and enhanced logistical profciencies; and a capacity for any particular instantiation of the built environment to be more than it appears to be. To be constantly fungible, re-doable, and where everything is eventually useful to someone. In this mix of standardization and singularity, fgure and ground are constantly being reversed in order to accommodate the duality of this promise. To *stand by this promise* is not to adhere to its specifc fguration but rather to the possibility of a fguring that eventually emerges from a background that cannot be mapped but to which one might be exposed.

For as many of the inhabitants of these extended urban regions I have talked with now frequently point out, it is important to pay attention to the background. For them, the background combines a willingness to suspend the judgment that what you see is what things are, an acknowledgment that beyond the immediacy of a person's context that there is a feld of vision that can be grasped and composed in excess of what is presented, and a belief that this willingness to see in a diferent way, a way that does not tie everything together into a coherent image, will enable the person to better navigate the ins and outs of everyday urban life. Tese processes of willingness, acknowledgment, and belief are then often crystalized into a particular working image and constitute a promise.

Tus, while dedicated genealogies may be capable of grasping how particular built environments, spatial dispositions, and fabric got to be the way they manifest themselves, there is something that eludes coherent narratives of development and prospective futures. Tese are spaces of intensive contiguity of the disparate—disparate forms, functions, and ways of doing things. Tey are replete with gaps, interstices, breakdowns, contested territories, and sediments of dissonant tenure regimes, fnancing, legalities, and use. Instead of being able to discern legible articulations among the details of composition, the proliferation of housing, commercial, industrial, logistical, recreational, entrepreneurial, and governmental projects are less subsumed into overarching logics of capital accumulation or neoliberal rationalities as they are "strange accompaniments" to each other, where nothing quite fts according to design, where things dissipate or endure without obvious reason, and where improvised alliances of use and rule continuously reshape what it is possible for any particular individual or institutional actor to do.

I call this mode of accompaniment, of not clearly discernible or translatable territories of operation, *the surrounds.* Te surrounds constitute neither an explanatory context, nor relations of interdependency. Tey are not strictly geographical phenomenon nor temporal, but can alternate to varying degrees. Te surrounds do not surround a given space, project, environment, or ecology as a boundary-limit or some constitutive outside. Tey are not some alternate reality just over there, just beyond the tracks, or the near horizon. Sometimes they are heterotopic, exceptional, intensely specifc, hidden in plain sight, prefgurative, or dissolute. In all instances the surrounds are infrastructural in that they entail the possibilities within any event, situation, setting or project for something incomputable, unanticipated to take (its) place. While such surrounds have always existed within cities, the urban extensions amplify the ways in which they exert both a structuring efect in the rapid coverage of land with multiple projects *and* a by-product of the tensions and countervailing logics at work in the very construction and composition of these extensions.

#### **References**


**Open Access** Tis chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/ by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

Te images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.