Maintenance and Philosophy

of Technology

What can we learn about the nature of technology by studying practices of

maintenance and repair? This volume addresses this question by bringing
together scholarship from philosophers of technology working at the fore-
front of this emerging and exciting topic.
The chapters in this volume explore how attending to maintenance and
repair can challenge and complement existing ways of thinking about tech-
nology focused on use and design and introduce new philosophical per-
spectives on the relationship between technology, time and human practice.
They examine the significance of maintenance and repair practices at dif-
ferent scales in relation to a diverse range of philosophical traditions and a
wide variety of technologies, from urban infrastructure such as bridges
and buildings to data technologies such as servers and software systems.
Together, the contributions highlight common themes in the philosophical
study of maintenance, including the role of skill, the significance of social
values and the potential of these practices to transform the technologies to
which they are applied. By reflecting on the different ways in which we
keep technologies going, from the devices we use in our homes to the large
technical systems which surround us, this volume reveals the philosophical
significance of practices of maintenance, not only as a source of new in-
sights but also as a resource for enriching our understanding of a variety of
existing topics in philosophy.
Maintenance and Philosophy of Technology will appeal to scholars and
advanced students working in philosophy of technology, philosophy of
engineering and science & technology studies.

Mark Thomas Young is a Marie Curie Postdoctoral Fellow in the

Philosophy Department at the University of Vienna. His research covers
two fields: the Philosophy of Technology, where he focuses on practices of
maintenance and the use of automating technologies, and the History and
Philosophy of Science, where he explores instruments, craft practices and
tacit knowledge in the early modern period.
Mark Coeckelbergh is a full Professor of Philosophy of Media and
Technology at the University of Vienna and ERA Chair at the Institute of
Philosophy of the Czech Academy of Sciences in Prague. He was President
of the Society for Philosophy and Technology and is the author of nu-
merous publications including Environmental Skill, AI Ethics, Self-
Improvement and The Political Philosophy of AI.
Routledge Studies in Contemporary Philosophy

Philosophy of Mental Disorder

An Ability-Based Approach
Sanja Dembić

Historical Explanation
An Anti-Causalist Approach
Gunnar Schumann

Exploring Extended Realities

Metaphysical, Psychological, and Ethical Challenges
Edited by Andrew Kissel and Erick José Ramirez

Expected Experiences
The Predictive Mind in an Uncertain World
Edited by Tony Cheng, Ryoji Sato, and Jakob Hohwy

Hybrid Societies
Living with Social Robots
Piercosma Bisconti

Philosophy, Expertise, and the Myth of Neutrality

Edited by Mirko Farina and Andrea Lavazza

Maintenance and Philosophy of Technology

Keeping Things Going
Edited by Mark Thomas Young and Mark Coeckelbergh

For more information about this series, please visit: www.routledge.com/Routledge-Studies-

in-Contemporary-Philosophy/book-series/SE0720
Maintenance and Philosophy
of Technology
Keeping Things Going

Edited by Mark Thomas Young

and Mark Coeckelbergh
First published 2024
by Routledge
605 Third Avenue, New York, NY 10158
and by Routledge
4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
Routledge is an imprint of the Taylor & Francis Group, an informa
business
© 2024 selection and editorial matter, Mark Thomas Young and
Mark Coeckelbergh; individual chapters, the contributors
The right of Mark Thomas Young and Mark Coeckelbergh to be
identified as the authors of the editorial material, and of the authors
for their individual chapters, has been asserted in accordance with
sections 77 and 78 of the Copyright, Designs and Patents Act 1988.
With the exception of Chapter 9, no part of this book may be
reprinted or reproduced or utilised in any form or by any
electronic, mechanical, or other means, now known or hereafter
invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in
writing from the publishers.
Chapter 9 of this book is freely available as a downloadable Open
Access PDF at www.taylorfrancis.com under a Creative Commons
Attribution-Non Commercial-No Derivatives (CC-BY-NC-ND)
4.0 license.
Funded by TU Delft.
Trademark notice: Product or corporate names may be trademarks
or registered trademarks, and are used only for identification and
explanation without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
ISBN: 978-1-032-32686-3 (hbk)
ISBN: 978-1-032-32687-0 (pbk)
ISBN: 978-1-003-31621-3 (ebk)
DOI: 10.4324/9781003316213
Typeset in Sabon
by SPi Technologies India Pvt Ltd (Straive)
Contents

List of Contributors ix
Acknowledgements xiii

1 Keeping Things Going: Maintenance and Philosophy

of Technology 1
MARK THOMAS YOUNG AND MARK COECKELBERGH

PART I
Metaphysics and Epistemology of Maintenance 37

2 Maintenance and the Humanness of Infrastructure 39

STEVEN VOGEL

3 Technology in Process: Maintenance and the Metaphysics

of Artefacts 58
MARK THOMAS YOUNG

4 There, I Fixed It! On the Status and Meaning of Repair 86

TIAGO MESQUITA CARVALHO

5 A Standpoint Epistemology of Repair? 113

CRISTINA BERNABÉU AND JESÚS VEGA-ENCABO

6 Sustainability as Planetary Maintenance 140

JOCHEM ZWIER
viii Contents

7 Towards a Realist Metaphysics of Software Maintenance 162

KEITH BEGLEY

8 Maintaining Perpetual Actuality in the Digital Age?

Simondon’s Conception of Maintenance and the
Networked Era 184
JOHANNES F.M. SCHICK

PART II
Ethics, Politics and Aesthetics of Maintenance 213

9 Maintenance of Value and the Value of Maintenance 215

STEFFEN STEINERT

10 An Eco-Ethics for the End of the Anthropocene: Finding

Ethical and Sustainable Paths through Consumerism,
Disposability and Planned Obsolescence 240
SIMON PENNY

11 Aesthetic Values in the Maintenance of Urban Technologies 264

SANNA LEHTINEN

12 Negotiating Visions of Waste: On the Ethics of

Maintaining Waste Infrastructures 279
JOOST ALLEBLAS AND BENJAMIN HOFBAUER

13 Repairing AI 306
TAYLOR STONE AND AIMEE VAN WYNSBERGHE

Index 326
Contributors

Joost Alleblas is a PhD Candidate at the Faculty of Technology, Policy, and

Management at Delft Technical University. Since 2019, he has been a
part of the ERC Project on Value Change in the Department of Values,
Technology, and Innovation. This project tries to understand how we
can design large technical systems so that they can accommodate chang-
ing values in societies. His research interests lie with energy transition,
nuclear energy and the role of visions in the promotion of both. Amongst
others, he studies the moralisation of socio-technical visions, especially
those visions that incorporate the realisation of moral ideals.
Keith Begley studied Mental and Moral Science at Trinity College Dublin,
Ireland, and holds an MA and a PhD in Philosophy from the University
of Dublin. He also studied Computer Science at the University College
Dublin and holds an HDip and an MSc in Computer Science from the
National University of Ireland. He has held positions as an Assistant
Lecturer in the Department of Philosophy, Maynooth University, a
Demonstrator in the School of Computer Science, UCD, and an Adjunct
Assistant Professor in the Department of Philosophy, TCD. He has pub-
lished on philosophy of language and linguistics, philosophy of artificial
intelligence, medical epistemology, computational philology, Heraclitus,
Jerrold J. Katz and Wittgenstein.
Cristina Bernabéu teaches Philosophy at Pedro Salinas high school in
Spain. She has written a dissertation about technical artefacts and gen-
der normativity and published on gender, feminist philosophy of tech-
nology and social epistemology. She is especially interested in material
and feminist ways of activism.
Tiago Mesquita Carvalho is a Researcher and Lecturer at the Philosophy
Institute (IF) in the Humanities Faculty in Porto working on a project
concerning technology, responsibility and catastrophes. He has degrees
in both Environmental Sciences and Environmental Philosophy and
x Contributors

Aesthetics. In 2019, he gained his PhD in FCUL with a thesis on how

the good life demands the deliberate use of artifacts, crossing both vir-
tue ethics, philosophy of technology and science and technology studies.
Before joining IF, he was a research assistant in the CoLABOR team,
aiming to build a technological impact assessment methodology in work
and employment. For 10 years, he has also been a volunteer at a DIY
bicycle workshop based in Lisbon.
Mark Coeckelbergh is a full Professor of Philosophy of Media and Tech­
nology at the University of Vienna and ERA Chair at the Institute of
Philosophy of the Czech Academy of Sciences in Prague. He was a presi-
dent of the Society for Philosophy and Technology and is the author of
numerous publications including environmental skill, AI ethics, self-
improvement, and the political philosophy of AI.
Benjamin Hofbauer is a PhD Candidate at the Faculty of Technology, Policy
and Management at Delft Technical University and a Research Fellow at
the Research Institute for Sustainability in Potsdam. In his work, he fo-
cuses on the moral implications of introducing and researching new and
emerging technologies as well as the ethical assessment of the research
process itself. While his thesis deals with solar climate engineering re-
search governance, his interests expand throughout the intersection of
climate change and technology, ranging from political philosophy and
environmental ethics to the ethics and philosophy of technology.
Sanna Lehtinen (PhD 2015, University of Helsinki) works as a Research
Fellow at the Aalto University School of Arts, Design and Architecture.
She is also a Docent in Aesthetics at the University of Helsinki. Sanna’s
research focuses on the intersection of urban aesthetics, philosophy of
the city and philosophy of technology. Sanna’s publications include jour-
nal articles published in Open Philosophy, Essays in Philosophy,
Philosophical Inquiries, Sustainability Science and Behaviour &
Information Technology, chapters in edited volumes by Routledge,
Bloomsbury, Springer and Oxford University Press as well as editing spe-
cial issues for Environmental Values, Contemporary Aesthetics, ESPES,
East Asian Journal of Philosophy and Open Philosophy. Sanna is a
Codirector on the Board of the Philosophy of the City Research Group
and an affiliate member of the 4TU Centre for Ethics and Technology.
Sanna is an assistant editor of Contemporary Aesthetics and founder and
Co-editor-in-chief of the new Philosophy of the City Journal. https://
www.sannalehtinen.com
Simon Penny is an artist, teacher and theorist with a long-standing focus
on emerging technologies, embodied and situated aspects of artistic
practices and critical analysis of computer culture. He published
 Contributors xi

Making Sense: Cognition, Computing, Art and Embodiment in 2017

and directed A Body of Knowledge: Embodied Cognition and the Arts
Conference UCI 2016. As a Professor of Art and Robotics at Carnegie
Mellon (1993–2000), he developed VR and robotics projects. He is
co-director and co-originator of the Industrial Crafts Research Network.
He is a lifelong maker and environmental activist with a focus on sus-
tainability, green technologies and regenerative agriculture. Penny is a
professor of Electronic Art and Design (Department of Art) at the
University of California, Irvine (with appointments in the depts of
Music and Informatics) where he teaches Art and Sustainability,
A Cultural History of the Anthropocene and How to be Clever with
Stuff, among other classes.
Johannes F.M. Schick has been Scientific Coordinator of the SFB 1187
Media of Cooperation since January 2022. He has been the head of the
research project “Action, Operation, Gesture: Technology as Inter­
disciplinary Anthropology” at the a.r.t.e.s. Graduate School for the
Humanities (University of Cologne) from October 2017 to February
2021. His research focuses on interdisciplinary (techno-)anthropology
(from Bergson, Espinas and Mauss to Simondon), Bergson’s philosophy
of life and the relation of anthropology to philosophy (and vice versa)
as well as the “Category Project of the Durkheim School”. He is a mem-
ber of the Editorial Board of the Durkheimian Studies (new series).
Furthermore, he is interested in phenomenological psychiatry, the phe-
nomenon of creativity, systems theory and practice theory. In his PhD
thesis, Schick focused on the relation of intuition and emotion in the
philosophy of Henri Bergson (published as Erlebte Wirklichkeit. Zum
Verhältnis von Intuition zu Emotion bei Henri Bergson, 2012).
Steffen Steinert is an Assistant Professor in the Ethics and Philosophy of
Technology section at Delft University of Technology. He has studied
philosophy, history and sociology at the Technical University of Dresden
and the New School for Social Research in New York. He completed his
PhD at the University of Munich before relocating to the Netherlands to
pursue an academic career in philosophy of technology. Steffen’s re-
search and teaching focus on philosophy and ethics of technology. He
has worked on brain–computer interfaces, social media and robotics.
Steffen’s main interest is in exploring the impact of technology on soci-
ety, specifically how it influences our values.
Taylor Stone is a Senior Researcher at the Institute for Science and Ethics,
University of Bonn, and a member of the Bonn Sustainable AI Lab. He
also is an associate editor of the Philosophy of the City journal. Taylor
received his PhD in Ethics of Technology from Delft University of
Technology and has held postdoctoral and lecturer positions in industrial
xii Contributors

design, ethics of technology and engineering ethics education. His re-

search focuses on the ethics of urban technologies and the incorporation
of environmental values into design processes and built artefacts.
Aimee van Wynsberghe is the Alexander von Humboldt Professor for
Applied Ethics of Artificial Intelligence at the University of Bonn in
Germany. Aimee is the director of the Institute for Science and Ethics
and the Bonn Sustainable AI lab. She is co-director of the Foundation
for Responsible Robotics and a member of the European Commission’s
High-Level Expert Group on AI. She is a founding editor for the inter-
national peer-reviewed journal AI & Ethics and a member of the World
Economic Forum’s Global Futures Council on Artificial Intelligence and
Humanity. In each of her roles, Aimee works to uncover the ethical
risks associated with emerging robotics and AI. Aimee’s current re-
search, funded by the Alexander von Humboldt Foundation, brings at-
tention to the sustainability of AI by studying the hidden environmental
costs of developing and using AI.
Jesús Vega-Encabo is a Professor of Logic and Philosophy of Science at the
Autonomous University of Madrid (Spain). He has published exten-
sively on epistemology, philosophy of mind and philosophy of science
and technology. His most recent research is focused on intellectual au-
tonomy, epistemic dependence, epistemic normativity and the nature of
artefacts.
Steven Vogel is a Professor Emeritus of Philosophy at Denison University
in Granville, Ohio USA. He is the author of Against Nature: The
Concept of Nature in Critical Theory (1996) and Thinking Like a Mall:
Environmental Philosophy After the End of Nature (2015).
Mark Thomas Young is a Marie Curie Postdoctoral Fellow in the Philos­
ophy Department at the University of Vienna. His research covers two
fields: the Philosophy of Technology, where he focuses on practices of
maintenance and the use of automating technologies, and the History
and Philosophy of Science, where he explores instruments, craft prac-
tices and tacit knowledge in the early modern period.
Jochem Zwier is a Researcher in Philosophy of Technology at Wageningen
University & Research and the Managing Director of the 4TU Centre
for Ethics and Technology. He studied Philosophy of Technology at
the University of Twente (MSc) and Radboud University Nijmegen
(PhD). His work focuses on the intersection of philosophy of technol-
ogy, ecology and environmental philosophy, particularly in the con-
text of the Anthropocene. His current work further revolves around
temporality and sustainability, the meaning of innovation and the con-
cept of anthropocentrism.
Acknowledgements

We would like to thank the Society for the Philosophy of Technology (SPT)
for their support of a special interest group on Maintenance and Philosophy
of Technology which was established in April 2022, alongside the many
participants of the group who provided feedback on many of the contribu-
tions that appear in this volume. In addition, we would like to thank the
following people for their helpful comments on earlier drafts of this manu-
script; Tobias Stadler, Nico Formanek, Mark Theunissen, Dan Weiskopf,
Maxine Morphis-Riesbeck, Jacqueline Bellon, Marcel Siegler, Frodo
Podscwadek, Tommy Lindgren, Luke Collison, Luca Possati, Scott Luan,
Elena Popa, Robin K. Hill, Dirk van Leemput, Marilena Pateraki, Andrea
Gammon, Madison Snider, Guru Madhaven, Bogdan Chernyakevich,
Zachary Pirtle, Sveinung Sivertsen, Katheleen Wallace, Kjetil Akø, Patrick
Zapata, Michel Bourban, Alexandria Poole, Jaana Parviainen, Céline
Kermish, Mateusz Salwa, David Jenkins and Sandra Zákutná. We’d also
like to thank the editorial team at Routledge for their help, support and
patience throughout the development of this volume. Finally, we would
like to thank the contributors of the volume itself for their enthusiasm and
support for this project, alongside their work responding to editorial com-
ments and reviewer feedback.
1 Keeping Things Going
Maintenance and Philosophy
of Technology

Mark Thomas Young and Mark Coeckelbergh

There is currently a growing recognition among philosophers of technol-

ogy, that despite a consistent effort to understand the nature of the rela-
tionship that people share with the technologies that surround them,
existing work within the field has nonetheless ignored certain forms of
technical practice that are very important. The practices we refer to here
are known collectively as maintenance and together comprise the different
activities that people must perform in order to ‘keep things going’. Until
now, philosophical reflections on maintenance have been few and far be-
tween. The most developed treatments of the topic within anglophone phi-
losophy, for example, can be found in Robert Pirsig’s Zen and the Art of
Motorcycle Maintenance (Pirsig 1974) and Elizabeth Spelman’s Repair:
The Impulse to Restore in a Fragile World (Spelman 2002), both of which
explored practices of maintaining technologies as a segue into the exami-
nation of wider issues relating to intellectual cultural division and social
justice, respectively. However, among the topics and themes addressed by
contemporary philosophy of technology, maintenance is almost nowhere
to be found. With the exception of a few recent articles, maintenance is a
topic which is often overlooked entirely.1
That maintenance has until now received so little attention from phi-
losophers of technology should strike us as surprising for a variety of rea-
sons. In the first place, the technological societies that we currently inhabit
literally depend on the constant performance of maintenance. Consider,
for example, the buildings which rise from our cities or the bridges, roads
and tunnels which enable the movement of people to and from work every
day, not to mention the amenities of modern urban living such as running
water, electricity and information and communications technology, all of
which depend on constant maintenance – not only of the material compo-
nents of such systems but also the institutions and technical cultures within
which they are located.2 Furthermore, the crucial importance of mainte-
nance in sustaining the function of technologies applies no less to tech-
nologies which purportedly operate without human involvement, such as

DOI: 10.4324/9781003316213-1
2 Mark Thomas Young and Mark Coeckelbergh

artificial intelligence and automation. The capacity of technologies such as

ChatGPT to generate humanlike conversational dialogues, for example,
depends on the ongoing maintenance of websites and code, telephone lines
and fibre optic cables which snake their way across continents and under-
neath oceans, data centres and servers, alongside the maintenance of elec-
tricity networks on which these rely. The sheer importance of maintenance
work is reflected in the relative proportions of tasks performed by those,
such as engineers and software developers, whose professions aim to pro-
vide us with these technologies. Regardless of whether we consider boats
or bridges or cloud servers or code, the amount of money and time devoted
to the initial development of technologies is often vastly outweighed by
what is required to keep them going.3 As a result, the majority of those
working in professions commonly thought to involve the design and pro-
duction of new things instead spend their time maintaining existing things
(Young 2021a). Whether we recognise it or not, maintenance and repair
are ‘in many ways, the engine room of modern economies and societies’
(Graham & Thrift 2007). Yet this simple fact appears to have gone unno-
ticed in contemporary philosophy of technology which has approached the
analysis of technical expertise by focusing exclusively on the knowledge
required to design and build new things. As a result, what is by far the larg-
est and most widespread form of technological expertise (Edgerton 2006,
p. 80), has remained unanalysed.
Yet the neglect of maintenance by philosophers is surprising not only
because of the ubiquity of this mode of human engagement with technol-
ogy but also due to its unparalleled intimacy. Compared to the lived rela-
tionships we establish with technologies through the activities of design
and use, maintenance very often requires forms of understanding which
combine aspects of both while also going beyond them in significant ways.
For example, while the designers of a technology may possess a privileged
understanding of how a particular technical design contributes to the
achievement of certain functional states, philosophers have long been cog-
nizant of the inherent limitations in their capacity to predict how a particu-
lar technology will affect and be affected by the natural and human
environments in which it comes to be employed.4 Designers of a technol-
ogy are very often aware that a deeper understanding of the nature of their
innovations can only be gained after they have become widely used.5 While
some aspects of this deeper understanding may stem from the users of a
technology, who in order to use it in the way they desire, discover things
the designers could not have predicted, this deeper understanding only
seems to be properly granted through the particular form of use estab-
lished through the practice of maintaining a technology over time.
When we seek to maintain, we must often establish a different relation-
ship with a technology than what is required when our goal is simply to
 Keeping Things Going 3

use it to achieve the purpose it was designed for. A common way of con-
ceptualising the nature of this relationship is through analogy to the con-
cept of care which emerged in feminist ethics in the late twentieth century.6
Whereas care ethics was originally conceived as applying to human rela-
tionships, recent work in science and technology studies (STS) has increas-
ingly explored its relevance as an ethical framework for analysing the
relationship between humans and the technological environments they in-
habit.7 The seeds for such approaches were sown already in early studies of
care in the feminist tradition.8 In Joan Tronto and Bernice Fisher’s well-
known definition of care, for example, the connection to maintenance is
already clearly visible:

On the most general level, we suggest that caring be viewed as a species

activity that includes everything we do to maintain, continue and repair
‘our world’ so that we can live in it as well as possible. That world in-
cludes our bodies, our selves, and our environment, all of which we seek
to interweave in a complex, life sustaining web.
(Fisher & Tronto 1991)

According to feminist theorists, one of the basic requirements for establish-

ing a relationship of care is forming an understanding of the needs of the
cared-for. Care theorists consider this to require a different moral attitude
to the introspective, disinterested stance typical to ethical analysis in the
tradition of liberal individualism, whereby we attempt to imagine what we
would need if we inhabited the situation of those for whom we care. In
contrast, feminist theorists argue that we must adopt a stance of attentive-
ness in which we look beyond ourselves and cultivate a receptiveness to
recognising the needs of those we intend to care for.9 This involves resisting
generalisations and paying adequate attention to the particular character-
istics of individuals and the circumstances they occupy. Applied to techno-
logical rather than social relations, the requirement of attentiveness
manifests itself in a receptiveness to what particular technologies require in
order for them to remain in a state we desire. In many cases, this requires
us to have a working understanding of how they are made and how differ-
ent aspects of their physical design contribute to their different functional
capacities. A poignant example of this kind of knowledge is found in
Working Knowledge: Skill and Community in a Small Shop (1987), in
which the sociologist Douglas Harper recounts his experiences document-
ing the practical expertise of Willie, a local repairman in North Country
New York. As Harper recalls:

Willie is alert to the machines he uses and the environment they service.
This might include something as simple as the proper use of a chainsaw
4 Mark Thomas Young and Mark Coeckelbergh

on a cold day, which I learned after I ruined a chain by not letting the
lubricating oil warm up enough before I began cutting. I later noticed
that after Willie started his saw he pointed it downward. Only when oil
began to spurt from the end of the blade, making a small black track in
the snow, did he begin cutting. He explained that when the temperature
is below zero, the engine must run long enough to heat the blade oil,
which is in a separate reservoir. It is a very simple idea, but it signals an
understanding of how the machine responds to the environment and
how its use must be modified when conditions change.
(Harper 1987, p. 161)

When our goal is merely to use a technology, we may not care whether
particular patterns of use lead to the premature degradation of certain
components. Maybe we intend to replace the technology when it can no
longer be used or maybe our intention was simply to use something for as
long as it lasted, much as one does with disposable paper cups. Attempting
to maintain a technology, on the other hand, signals a temporal commit-
ment which influences and shapes the relationships we maintain with
technologies through the act of using them. This commitment often mani-
fests itself in specific forms of practice, such as keeping things clean or
lubricated or ensuring that certain functions do not exceed particular
thresholds.
As Harper’s example illustrates, attempting to maintain a technology
very often requires us to deepen our understanding of how it works. In
addition to encouraging particular forms of use, maintaining a technology
also often requires us to open the black box, by popping the hood and
looking inside, unscrewing the casing in order to remove or clean compo-
nents or raising things up to see underneath. At the same time, the atten-
tiveness which many scholars understand to be central to the maintenance
attitude also requires us to go beyond the knowledge of designers in vari-
ous ways. Extended experience maintaining particular technologies may
grant insights unknown to those who designed them, such as how certain
components deteriorate earlier than expected under certain environmental
conditions or the effects of different patterns of use on the durability of the
technology as a whole.
As existing scholarship on maintenance demonstrates, while some forms
of maintenance practice may proceed according to predetermined plans,
such as when car manufacturers develop detailed maintenance plans for
the vehicles they produce, other forms are irreducibly situated insofar as
they involve regular inspection to monitor the effects of particular pat-
terns of use or the unique environment a technology inhabits as it devel-
ops over time.10 Such forms of attentiveness also encourage us to develop
an awareness of the unique characteristics possessed by individual tokens
 Keeping Things Going 5

of particular technologies as they are manifested within specific contexts of

use.11 A good example occurs in Zen and the Art of Motorcycle Maintenance,
in which Pirsig’s unnamed protagonist reflects on how the act of maintain-
ing his motorcycle required him to develop an understanding of the indi-
vidual personality of his machine:

a friend who owns a cycle of the same make, model and even same year
brought it over for repair, and when I test rode it afterward it was hard
to believe it had come from the same factory years ago. You could see
that long ago it had settled into its own kind of feel and ride and sound,
completely different from mine. No worse, but different. I suppose you
could call that a personality. Each machine has its own, unique person-
ality which probably could be defined as the intuitive sum total of every-
thing you know and feel about it. This personality constantly changes,
usually for the worse, but sometimes surprisingly for the better, and it is
this personality that is the real object of motorcycle maintenance.
(Pirsig 1974, p. 39)

Yet whereas Pirsig considered such attentiveness to the unique characteris-

tics and needs of different artefacts to form the basis of a desirable relation-
ship to technology, he also recognised the extent to which this attentiveness
stood in conflict with many aspects of the culture and society he inhabited.
For Zen and the Art of the Motorcycle Maintenance addressed a concern
that worried Pirsig’s contemporaries as much as it worries many today,
that the rise of consumerism, in which we purchase rather than make many
of the things we need, has prevented many from establishing such relation-
ships and as a result has caused people to become alienated from the material
objects which surround them. When things break or stop working, many
of us throw them away, buy new ones or return them to the manufacturer
for replacement or repair. Furthermore, the incentives against developing
the kind of intimate, engaged and active relationships with the technolo-
gies around us that Pirsig described are increasing. Artefacts are increas-
ingly designed deliberately in ways that prevent or discourage us from
repairing them when they break (Verbeek 2005, p. 226) or are constructed
from materials which are difficult, if not impossible to repair.12 Consider,
for instance, contemporary mobile phones or laptops which are often de-
signed in ways which discourage users from opening them to engage in
repair activities. Moreover, undertaking repair tasks ourselves often re-
quires specialised equipment, skills or replacement parts which can be
more expensive than simply replacing the item in question. Legal contracts
such as warranty agreements disincentivise repair activities insofar as per-
forming them may release the manufacturer from the obligation to repair
or replace defective products. Finally, manufacturers often attempt to
6 Mark Thomas Young and Mark Coeckelbergh

ensure the obsolescence of their products long before they reach a state
which we might consider unrepairable by encouraging us to view them as
outdated through the marketing of products with new features, for exam-
ple, or by discontinuing services such as updates that are geared towards
ensuring their products’ compatibility with the wider technological sys-
tems in which they are employed.
The romanticised vision of maintenance that Pirsig outlined stands in
stark contrast to the passive relationship consumers are often encouraged
to maintain towards their possessions in the context of industrial capital-
ism. Many modern commentators on the practice of maintenance have
followed his lead in understanding the activity of maintenance and repair
to present the possibility of establishing a more fulfilling and authentic re-
lationship with material objects. Steven Jackson, for example, identifies
repair as the activity by which ‘human value is preserved and extended’
(Jackson 2014). Matthew Crawford, on the other hand, understands prac-
tices of maintenance and repair to be intrinsically valuable and argues that
manual engagement with technology is necessary for human flourishing
(Crawford 2009, p. 64). Such claims reflect that despite its neglect in philo-
sophical literature, there is nonetheless an enduring conviction regarding
the desirability of engaging with artefacts through the practices of mainte-
nance and repair. We, the editors of this volume, for example, were both
fascinated by Pirsig’s account of technological maintenance when we read
Zen and the Art of Motorcycle Maintenance during our undergraduate
study. In fact, we both recall this being one of the inspirations that at-
tracted us specifically to the philosophy of technology and we know from
talking to other philosophers that we are not alone. Many of us who were
interested in learning more about the different kinds of active and engaged
relationships humans establish with technology drifted inevitably towards
early phenomenological accounts of technology as found in the work of
Heidegger and Merleau-Ponty, alongside the extension of these ideas by
later American philosophers such as Don Ihde and Hubert Dreyfus. But
the practice of maintenance and repair was not a topic that any of us en-
countered again. So why is it that despite the ubiquity and importance of
this form of technical activity, it has until now received only scant atten-
tion from philosophers of technology?
The reasons for this neglect most likely stem from a combination of
causes which complicates the task of providing a complete and definitive
answer to this question. However, in order to help facilitate this volume’s
goal of stimulating philosophical reflection on the practice of maintenance,
it is nonetheless worth drawing attention to some of the main obstacles
that have hindered the further development of philosophical reflection on
maintenance practices so that they may be more easily overcome. In the
first place, we might consider how recent trends and interests within the
 Keeping Things Going 7

philosophy of technology have helped to divert attention away from the

practice of maintenance. The later stages of the twentieth century wit-
nessed a transition in the philosophy of technology, away from transcen-
dentalist critiques of technology and towards more focused investigations
of specific technologies and technical practices. This shift was promoted by
its proponents under the banner of the ‘empirical turn’ in the philosophy
of technology.13 As Verbeek notes, the empirical turn remains clearly visi-
ble today in the two main orientations within contemporary philosophy of
technology.
The first aims to clarify the ontological and epistemological aspects of
engineering practice and employs a conceptual vocabulary and methodol-
ogy which owes much to analytic metaphysics and philosophy of science.
A common conviction among those writing within this orientation is to
consider the practice of designing as the essence of engineering itself. Not
surprisingly then, despite the ubiquity and importance of maintenance
within engineering practice, this orientation has nonetheless tended to con-
flate engineering with innovation: the practice of building something new.
This is reflected not only in the common tendency among philosophers to
analyse the practice and methods of engineering by focusing exclusively on
the activity of design but also in the dominant view that the production of
technical artefacts occurs exclusively through the material realisations of
their makers’ intentions (Young 2021a).
The second orientation within the empirical turn focuses on the social
implications of technologies and draws inspiration from phenomenology
and more recently, STS. This research line is more attuned to the use of
technologies and explores the relationships between the design of a tech-
nology and the way in which it enables, encourages or hinders different
forms of practice alongside the effects of the widespread use of particular
technologies within society. Yet, while work within this orientation recog-
nises that new forms of use or values may emerge around existing tech-
nologies, the fact that artefacts themselves may require constant work or
must often undergo customisation and modification in order to remain
functional is almost always overlooked. Instead, it is often implicitly as-
sumed that the nature and capacities of a technology are determined and
fixed once and for all through the initial processes of design and produc-
tion. It should come as no surprise then to find that the ethics of technol-
ogy which has emerged in the wake of the empirical turn has tended to
focus exclusively on practices of design and the ethical issues which arise
around the activity of making a new artefact or technology. Stemming
from the widespread conviction that the moral or political status of a tech-
nology depends on the way it was designed, philosophers are increasingly
looking towards the practice of design as both the source and solution of
ethical problems surrounding technology. This is reflected in the recent
8 Mark Thomas Young and Mark Coeckelbergh

emergence of a wide range of design for values approaches14 alongside the

common tendencies to focus exclusively on design in ethical discourses sur-
rounding specific technologies such as artificial intelligence. Dominant
trends and interests in contemporary philosophy of technology have there-
fore left few opportunities for the investigation of the different kinds of
ongoing work that is required after technologies are produced. Despite all
the work on human–technology relations, for example, one of the most
basic and pervasive ways in which humans relate to technologies, through
the activities of maintenance and repair, has been overlooked entirely.
Much existing work in the philosophy of technology seems to assume that
everything that’s worth exploring about technologies occurs in relation to
how they are made and consequently used.
This volume seeks to challenge this assumption by exploring a variety of
ontological, epistemological and ethical questions which emerge from at-
tending to the practice of maintenance. As such, it represents the first sys-
tematic investigation of the topic of maintenance within the philosophy of
technology. The chapters collected here consider maintenance in relation
to a wide variety of different technologies – from artificial intelligence to
socks (with many in between). Together, they not only demonstrate the
relevance and potential of maintenance for existing topics and discussions
in the philosophy of technology but also identify new areas for philosophi-
cal investigation which emerge when we shift our focus to the ongoing
work technologies require after production. As will become clear, unlike
many other topics in the philosophy of technology, maintenance does not
immediately align with a particular subfield, philosophical approach or
school of thought. The broad relevance of this emerging topic is reflected
in the diverse range of theoretical approaches and styles represented by
each of the contributions in this volume. By elaborating on how mainte-
nance matters for a variety of different topics and orientations in the con-
temporary philosophy of technology, we aim to clear a space for what we
hope will be an ongoing discussion on the topic.
There are reasons to think that the time is ripe for the emergence of
maintenance as a central topic in the philosophy of technology. A growing
awareness of our impact on the natural environment is leading to an
increasing interest in questioning and scrutinising the relationships we
maintain with the technologies that we create. The rising interest among
philosophers in topics such as the Anthropocene and sustainability reflects
how such concerns are increasingly perceived to be within the purview of
philosophical reflection itself.15 Luckily, however, the task of establishing a
philosophical discourse on maintenance doesn’t need to occur in a vac-
uum. Outside of philosophy, there is already a well-established body of
literature on the topic of maintenance and repair in disciplines such as STS,
sociology and geography. Over the past two decades, this interdisciplinary
 Keeping Things Going 9

field of maintenance and repair studies (MRS) has both generated a large
body of empirical accounts of maintenance practices and developed theo-
retical frameworks for their analysis. Yet, while it is not uncommon to find
MRS theorists drawing upon philosophical ideas or theories in their ac-
counts of maintenance and repair,16 up until now philosophers have re-
mained absent from such discussions. By collecting a range of contributions
which illustrate the relevance of maintenance to a variety of existing themes
in the philosophy of technology, this volume is intended as an invitation
for philosophers to join the table and contribute to developing our under-
standing of this important topic.
Our goals in the remainder of this introduction are as follows. First, we
will provide an overview of the development of MRS in Section 1.1 while
also highlighting dominant themes and theoretical approaches which char-
acterise the literature. Then we’ll turn our attention towards setting a
preliminary agenda and orientation for the development of a discourse
surrounding maintenance in the philosophy of technology by identifying
specific questions and areas of analysis surrounding maintenance that we
consider to be particularly suited to philosophical analysis. Finally, we’ll
provide a brief overview of the different chapters which are collected to-
gether in this volume.

1.1 Maintenance and Repair Studies

Aside from a few studies in the history of technology and anthropology,17
a focused body of literature on maintenance and repair didn’t really begin
developing until the 1990s. During this time, a selection of ethnographic
accounts appeared that focused on the maintenance of technologies in pro-
fessional contexts.18 These studies emerged alongside calls for a reorienta-
tion of scholarship in the history and sociology of technology away from
the previously dominant focus on processes of invention and towards the
active roles played by users in the development of technologies over time.19
One of the most tangible results of these calls to explore the active engage-
ments of users with technologies was the development of the field of the
social construction of technology (SCOT). However, despite the rapid in-
crease in popularity of such approaches, shaking the emphasis on innova-
tion turned out to be easier said than done. Critics pointed out how while
SCOT may indeed have acknowledged certain ways in which users play an
active role in the development of technologies over time, such approaches
still failed to look beyond periods of innovation (Mackay & Gillespie
1992; Edgerton 2010; Jackson 2014).
Shortly after the turn of the twenty-first century, the growth of the field
gained momentum from renewed calls to begin correcting for the dispro-
portionate scholarly focus on processes of invention, innovation and
10 Mark Thomas Young and Mark Coeckelbergh

production by examining practices of maintenance and repair specifically

(Edgerton 2006; Graham and Thrift 2007). Those promoting a turn to-
wards maintenance have often argued that the common tendency to ne-
glect consideration of the ongoing forms of activity required to sustain
technologies over time increasingly has produced a body of scholarship
that ultimately fails to explain how the objects, practices and institutions
which together comprise technology come into being. The reasons for this
neglect were sometimes identified in underlying assumptions and attitudes
surrounding the nature of technology itself – normal modes of picturing
and theorising technology that rendered maintenance invisible (Jackson
2014), for example, or cultural tendencies to glorify innovation which de-
valued maintenance work or cast it as uninteresting (Vinsel & Russell
2020). In other cases, scholars have examined how specific theoretical con-
cepts and approaches in individual disciplines, such as the traditional em-
phasis on processes of closure and stability in STS, have worked to divert
attention from the role maintenance plays in sustaining technologies over
time (Denis 2020; Denis, Mongili & Pontille 2015). This most recent phase
saw the popularity of the topic expand beyond the history and sociology
of technology to an eclectic range of disciplines and fields such as geogra-
phy, media studies, STS, art and craft studies, architecture and design the-
ory20 and coincided with the establishment by historians Lee Vinsel and
Andy Russel in 2015 of an interdisciplinary academic community devoted
to the topic called The Maintainers.21 There is now a growing body of lit-
erature on the topic of maintenance and repair which is expansive enough
to secure the status of maintenance and repair as an emerging field of in-
terdisciplinary study. The following provides a short overview of key
themes and common approaches within this literature.

1.1.1 Maintenance Beyond Materiality

The dominant strain of work in the field consists of ethnographic accounts

of maintenance and repair in different contexts which explore how order
is restored in cases of breakdown.22 Studies of this kind have long drawn
inspiration from a theoretical orientation in sociology known as ethno-
methodology, an approach to the study of everyday practices which under-
stands social order to arise as the result of constant, ubiquitous acts of
repair.23 It is long been common for studies in ethnomethodology to take
the form of conversation analysis (Sharrock 1989).24 Stemming from the
premise that normal conversation is rife with moments of breakdown in
which participants fail to understand each other, these studies explore how
successful communication is nonetheless achieved through numerous and
constant acts of repair, such as the addition of short clarifying dialogues or
hand motions, which are performed by participants during a conversation.
 Keeping Things Going 11

Through the analysis of conversations, proponents of ethnomethodology

produced what is perhaps the most extensive body of work on the phe-
nomena of repair within sociology (Henke & Sims 2020, p. 18).
During the 90s, these approaches came to be applied to the activity of
repair in technological contexts. Answering the growing calls to pay atten-
tion to the use of technology, sociologists began applying the methods and
insights of ethnomethodology to the analysis of work, focusing in particu-
lar on human interactions with technology. These studies prepared the
ground for a number of classic ethnographic accounts of maintenance and
repair within professional contexts which emerged shortly after.25 Today,
many proponents of MRS understand the field in terms of an extension of
ethnomethodological ideas and approaches to the analysis of technology,
exploring how socio-material order emerges not as a result of the way a
technology is designed, but rather of ubiquitous acts of repair performed
during use (Graham & Thrift 2007; Denis, Mongili & Pontille 2015).
The influence of ethnomethodology on current work in the field of MRS
is visible in various ways. For example, the emphasis in ethnomethodology
on the restoration of the social order of work as the goal of repair, along-
side the traditional methodological focus on the analysis of conversation,
has helped encourage a much wider conception of the activity of repair in
technological contexts than material engagement alone. Both Orr and
Henke’s seminal ethnographic accounts of repair work acknowledge the
importance of forms of repair in which a technology is ‘fixed’ by altering
the manner in which users perceive or engage with it. As Orr noted, ‘the
problems encountered by technicians are most fundamentally breakdowns
of the interaction between customers and their machines, which may or
may not include a malfunction or failure of some machine component’
(Orr 1996, p. 3). Drawing on examples from which users become con-
vinced that a technology has malfunctioned, these accounts illustrated
how technicians often seek to resolve the problem by engaging with the
user rather than the technology itself. This might involve offering explana-
tions for why a technology doesn’t operate as expected in certain contexts,
using measuring devices to provide demonstrations that it is actually
working properly or even providing instructions and training geared to-
wards enabling users to address minor problems themselves. Scholarship
on maintenance which takes its cues from ethnomethodology tends to ap-
proach the analysis of repair in ways which undermine traditional concep-
tions of breakdown as an objective state of an artefact. Instead, by
emphasising the way artefacts may appear as broken for some users but
not others, studies of this kind encourage an understanding of repair as a
relational activity in which the status of artefacts as broken or fixed de-
pends on the perspectives of the actors involved and is often subject to
negotiation (Henke 2019).
12 Mark Thomas Young and Mark Coeckelbergh

1.1.2 The Invisibility of Maintenance

Another way in which the influence of ethnomethodology is visible in cur-

rent MRS literature concerns the tendency for the activity of repair to re-
cede from view. As is well known, in developing ethnomethodology,
Harold Garfinkel drew inspiration from phenomenologists such as Martin
Heidegger and Alfred Schutz, alongside the later work of Ludwig
Wittgenstein in characterising central aspects of practice to be invisible to
those in the midst of performance (McHoul 1998; vom Len 2016 p. 86).
As is well known, Heidegger claimed that when we employ technologies
that are working correctly, we don’t experience them as visible isolated
objects, but rather as part of a background of undifferentiated equipment
upon which our activity depends (zuhanden). One of the ways in which
technologies become visible is when they prevent us from achieving our
goals because they are faulty or broken, in which case they appear to us as
individual objects (vorhanden).26 Wittgenstein, on the other hand, aimed
to undermine conceptions of skilled practice as rule-based by examining
how the creative activity necessary for enabling such activities remains in-
visible to practitioners themselves.27 These ideas can be seen to be reflected
in the basic tenets of ethnomethodology, which holds that while the skills
of repair are largely invisible to participants when immersed within the
flow of social activity, they can nonetheless be rendered visible and there-
fore accessible to analysis through disruption, a heuristic method famously
embodied in Garfinkel’s ‘breaching experiments’.
Today, (in)visibility continues to form a central theme within scholarship
on maintenance and repair. However, invisibility is a multifaceted theme
within MRS literature which not only reflects the influence of ethnomethod-
ology but also intersects with a variety of other themes and schools of thought
in sociology and closely related fields which touch upon social issues and is-
sues of power. For example, invisibility formed a central theme in the histori-
cal and sociological studies of housework which began to emerge during the
1980s. By exploring ways in which the gendered labour of housework failed
to gain recognition or valuation in different contexts, these studies gave rise
to what has become an enduring theme within feminist sociology – invisible
labour.28 While housework encompasses a variety of activities that include
the production of items such as food, some of the activities most readily iden-
tifiable as forms of housework, such as cleaning or clothing repair also rep-
resent forms of maintenance. For this reason, many well-known studies of
housework, such as Ruth Schwartz Cowan’s More Work for Mother (1983)
which focused specifically on the use of domestic technologies, are also con-
sidered to represent early works within the field of MRS (Russell & Vinsel
2018). By exploring how these forms of household maintenance have often
been performed outside the formal economy and have struggled to gain
 Keeping Things Going 13

adequate recognition, studies of housework contributed to establishing invis-

ibility as a dominant theme from the very origins of MRS.
The theme of the invisibility of maintenance in MRS was further rein-
forced through the influence of studies of ‘dirty work’, a research line pio-
neered by sociologist Everett C. Hughes that explores forms of labour
which are accorded a low social status or are considered to be degrading
or undignified.29 Invisibility forms a common theme in studies of dirty
work, many forms of which are deliberately arranged to be performed out
of sight, through the scheduling of cleaning activities ‘after hours’, for ex-
ample, or in spaces such as private homes or workshops. As a central topic
within the sociology of work, dirty work bears natural associations with
MRS literature insofar as the term is often taken to include forms of main-
tenance, such as janitor work or street cleaning.30 The influence of this line
of research is reflected in the common focus within MRS literature today
on the ways in which the visibility of maintenance practices is managed.31
As a result, literature in the field of MRS has developed increasingly nu-
anced conceptions of the relationship between the visibility and invisibility
of maintenance work alongside examining the different ways in which
maintenance and repair may be concealed or downplayed. Certain kinds of
artefacts, such as smartphones or laptops, are often designed in ways that
prevent the average user from being able to open them so that accessing
internal components for repair or maintenance is achieved primarily by
technicians in workshops with specialist equipment. In other cases, the
successful practice of maintenance is considered to depend on the erasure
of traces of the act of maintenance itself (Denis & Pontille 2018) Yet, while
invisibility remains a dominant concern within MRS literature, it is impor-
tant to note that maintenance practices are not always invisible. Some
forms of maintenance are displayed openly, such as cultural heritage con-
servation practices which often display or inform visitors of processes by
which artefacts are maintained over time (Jones & Yarrow 2013) or public
displays of infrastructure maintenance through which a government may
seek to reassert authority and control over structures that have been modi-
fied by citizens to meet their needs (Barnes 2017). Furthermore, insofar as
maintenance practices are often not distributed evenly through society or
the global political economy (Jackson 2014; Russell & Vinsel 2018;
Edgerton 2006, p. 80), the visibility of maintenance often depends on so-
cial or geopolitical location. For this reason, certain maintenance practices
may be visible to some but not to others.

1.1.3 Vulnerability, Fragility and Breakdown

Many MRS scholars understand the relative invisibility of maintenance

and repair to be at least partly responsible for cultivating widespread
14 Mark Thomas Young and Mark Coeckelbergh

misconceptions concerning the inherent stability of the different objects

that comprise our material culture. Within MRS literature, this insight of-
ten finds its most sustained elaboration in relation to infrastructure, which
as Steven Jackson notes is commonly experienced in many places of the
world to

…mostly work, for most people, most of the time. Because of this we
have tended to regard enduring function as a natural and more or less
permanent feature of systems, rather than as the ongoing, frequently
artful and often fraught accomplishment that it is.
(Jackson 2016)

A common goal in many MRS studies is therefore to move beyond the ap-
parent givenness of functional technology by restoring an appreciation of
the inherent fragility of the material world we have constructed around
ourselves. This often involves challenging the perceptions of what aspects
of technology should be considered typical, such as breakdown and failure
which are argued to be normal rather than anomalous occurrences in the
lifespan of technologies (Graham & Thrift 2007; Henke 2000; Young
2021b), or fragility or decay which is suggested to represent a normal
property of matter itself (Graham & Thrift 2007; Mattern 2018; Denis
2020). These approaches move us in the direction of what Steven Jackson
calls ‘Broken World Thinking’, an analytic orientation to the study of tech-
nology which

asks what happens when we take erosion, breakdown, and decay, rather
than novelty, growth, and progress, as our starting points in thinking
through the nature, use, and effects of information technology and new
media. Broken world thinking is both normative and ontological, in the
sense that it makes claims about the nature of technology and its rela-
tionship to broader social worlds, some of which may differ from deep-
rooted cultural assumptions. But it is also empirical and methodological,
an argument and provocation toward doing new and different kinds of
research, and new and different kinds of politics, in media and technol-
ogy studies today.
(Jackson 2014)

Focusing on the inherent vulnerability of our technological environments

inevitably propels maintenance to the centre of our thinking about tech-
nology insofar as it prevents us from taking the stability which character-
ises our experience of many technologies for granted. Instead, to account
for the apparent durability and permanence of those environments, we are
compelled to acknowledge the constant work of rebuilding, repairing and
 Keeping Things Going 15

adapting through which such stability arises. At the same time, focusing on
the inherent tendencies of artefacts towards decay or breakdown also rein-
forces connections with common theoretical approaches to the analysis of
maintenance such as care ethics, where an attentiveness to dependence and
vulnerability has long been considered a central component of relation-
ships of care by feminist theorists.32
Thinking through the consequences of fragility has proven to be a fruit-
ful analytic strategy among the variety of different subfields comprising
MRS and now forms a central theoretical orientation within contemporary
STS and studies of material culture.33 While such a shift has yet to occur
within the philosophy of technology, a focus on vulnerability nonetheless
bears important but as yet unexplored implications insofar as it encour-
ages us to invert the dominant line of questioning by resisting the common
tendency to focus primarily on how technology affects societies and exam-
ining rather than the different ways in which societies affect technology.

1.1.4 Maintenance and Repair as Transformative

An increasingly wide range of scholarship within MRS literature has

sought to undermine traditional dichotomies between innovation and
maintenance or repair work by emphasising the transformative potential
of the latter. This approach strikes a contrast both to traditional conceptu-
alisations of maintenance as consisting exclusively in the attempt to hinder
the changes artefacts undergo over time and common understandings of
repair as aiming only to restore an artefact to a previous state. While such
accounts acknowledge that some forms of maintenance or repair work
may indeed seek the preservation or restoration of artefacts in these ways,
they are at the same time concerned to illustrate how many of the goals
which motivate the practice of maintenance, such as prolonging the time
an object can stay in use (Weber & Krebs 2021, p. 11) or preserving physi-
cal or technical orders (Russell & Vinsel 2018), themselves often demand
the transformation of the underlying technologies. This is apparent in
studies of restoration practices which underscore the extent to which arte-
facts themselves must be changed in the process (Rubio 2016; Rubio
2020), alongside studies of infrastructure maintenance which draw atten-
tion to the practices of retrofitting and upgrading by which structures are
adapted to emerging interests or needs (Sims 2016; Young 2021a).34 By
acknowledging the creative potential of maintenance, these approaches
problematise intuitive distinctions between making and maintaining which
align creativity exclusively with the former and cast the latter as a deriva-
tive form of technical activity which consists of merely preserving the
products of prior creative activity. In doing so, they encourage us to under-
stand the activity of making not as confined to the period in which an
16 Mark Thomas Young and Mark Coeckelbergh

artefact is initially created but rather as extending throughout the histories

of artefacts. Such approaches also call for a reconceptualisation of the na-
ture of artefacts themselves, not as things with fixed and stable properties,
but rather as processes that are open-ended and dynamic, constantly re-
sponding to emergent features of the different environments in which they
are located.35
As a number of scholars have pointed out, acknowledging the creative
potential of maintenance and repair opens up novel lines of enquiry by
revealing new ways in which political processes come to manifest them-
selves in the artefacts which surround us. In many cases, changes to arte-
facts that occur under the guise of maintenance and repair serve the
interests of those in possession of political power, such as when infrastruc-
ture is modified to facilitate the growth of commercial activity to the detri-
ment of local communities. However, the transformative potential of
maintenance and repair also underscores the capacity of these forms of
activity to serve as forms of political action through which power may be
exercised by marginalised groups. In their study of the history of San
Diego’s Coronado Bridge, for example, Henke and Sims document how
activist work succeeded in enabling members of the local community to
influence decision-making processes surrounding seismic retrofits of the
bridge itself to ensure that modifications served their interests in addition
to those of the California Department of Transportation (Henke & Sims
2020). Henke and Sims’ example illustrates how ‘transformative repair
may also be championed by those who are excluded from existing power
structures or otherwise see repair as an opportunity to advance their inter-
ests’ (Henke & Sims 2020, p. 22).

1.2 Maintenance and Philosophy of Technology: Preliminary

Issues
In sketching the key themes and analytical orientations in MRS literature,
the foregoing sections illustrate the extent to which there is now an expan-
sive body of empirical research on the topic of maintenance and repair. As
we have seen, the analysis of much of this material has been achieved
through the employment of a variety of theoretical lenses common in soci-
ology and STS. In contrast, this volume aims to approach the topic of
maintenance and repair by employing a different set of theoretical re-
sources and approaches that can be found in contemporary philosophy of
technology while at the same time demonstrating how maintenance holds
the potential to advance a wide variety of ongoing discussions and themes
within the philosophy of technology. While it is common for MRS theo-
rists to draw on philosophical concepts in the elaboration of their argu-
ments,36 a sustained philosophical examination of many basic themes
 Keeping Things Going 17

regarding maintenance remains lacking. For this reason, a variety of basic

philosophical questions regarding maintenance remain unanswered. In this
section, we sketch a number of preliminary issues related to maintenance
which we regard to be particularly suited to philosophical analysis.

1.2.1 The Meaning(s) of Maintenance

In relation to technology, the term ‘maintenance’ is used to capture a wide

and diverse range of human practices. We maintain artefacts through the
act of using them in particular ways, cleaning components, applying pro-
tective coatings, replacing parts and fixing what is broken.
As highlighted in Section 1.1.1, maintenance is sometimes understood to
extend beyond material practice altogether, encompassing activities such
as the adjustment of user expectations or the provision of training. In light
of this heterogeneity, the question naturally arises concerning the extent to
which the term ‘maintenance’ can be understood to correspond to a coher-
ent and unitary form of technical practice. This is no easy question. For it
is not only the activities of maintenance that exhibit significant diversity
but also the goals motivating their performance. In some cases, mainte-
nance might be geared towards keeping something in existence, keeping
something in use or maybe instead keeping something operating optimally.
In other cases, maintenance may apply to the aesthetic features of objects,
preserving a particular look and condition. Sometimes maintenance aims
to change a state of events which has occurred, such as when we clean
away grime that has accumulated on the lens of a camera. Other times,
maintenance involves attempts to guard against things that may happen,
such as when we install protective barriers around windows in advance of
a coming storm. Given this diversity, it is not surprising to find a wide va-
riety of different ways of conceptualising maintenance within existing
MRS literature. This includes definitions of maintenance as ‘all the work
that goes into preserving technical and physical orders’ (Russell and Vinsel
2018) or as an activity which aims to ‘guarantee stability, to provide things
with a level of stability they do not possess “naturally”’(Denis and Pontille
2014). Other authors suggest we understand maintenance and repair as
the ‘myriad forms of activity by which the shape, standing and meaning of
objects in the world is produced and sustained’ (Jackson 2014, p. 234), or
as ‘represent(ing) interventions with the temporal aim of prolonging the
time an object can stay in use’ (Weber and Krebs 2021, p. 11).
One way to gain further insight into the nature of maintenance is by
examining its relationship to other forms of human engagement with tech-
nology, such as making or using. Maintenance bears deep connections
to use, insofar as the proper maintenance of many technologies is of-
ten considered to require using things in particular ways. For example,
18 Mark Thomas Young and Mark Coeckelbergh

maintaining a car engine may be understood to involve refraining from

activities which are known to damage engines over time, such as acceler-
ating in a high gear or shifting from drive to reverse without stopping. As
Carl Mitcham notes, maintaining represents a distinct form of using
which does more than merely ‘put an object to use in the straightforward
sense of engaging its technical or end uses’ (Mitcham 1994, p. 233).
Instead, this form of using is motivated also by the additional goal of pro-
longing the time a particular artefact may be used. For this reason, main-
taining can be understood as a form of using that looks beyond the
satisfaction of our interests and goals in the present and actively attempts
to influence the future states of a technology. While his account acknowl-
edges the creative significance of maintenance practices, Mitcham is at the
same time careful to distinguish maintenance from making, noting that
whereas maintenance

‘might even be said to extend or prolong making, since it protects or re-

tains the object made…. (maintenance) is subordinate not to making but
to using, and is a kind of using in that its aim is to keep something us-
able. Dusting the books and changing the oil in the car are repetitive
processes like making the bed or washing the dishes. Indeed, one is al-
most tempted to say that beds are not just made to be slept in but also
“made to be made”, that the second making is itself using, and that using
dishes entails not only eating off them but also washing and drying them.
In this sense maintaining can be seen as a natural prolongation of using,
and adjunct or secondary using, certainly a preparation for reusing’.
(Mitcham 1994, p. 233)

Yet, whereas Mitcham is concerned with establishing a distinction between

making and maintaining, more recent scholarship has sought to undermine
the very same distinction, arguing that maintenance should be considered
a form of making. Young (2021a), for example, examines how the distinc-
tion between making and using is premised upon the idea that the creation
of an artefact ceases at the end of production, a moment philosophers have
traditionally understood in terms of the material realisation of a designer(s)
intention. Yet by drawing attention to how the ongoing maintenance and
repair of artefacts often involve the attempt to realise intentions that differ
from those of the original designer(s), Young argues that rather than un-
derstanding making to be confined to the initial production of an artefact,
we should understand the activity of making to extend throughout the
lives of artefacts and to cease only when we no longer intend to continue
using them.
Another way of gaining insight into the nature of maintenance is by ex-
amining its relationship to repair. Despite the two terms often appearing in
 Keeping Things Going 19

tandem throughout MRS literature, questions concerning the similarities

and differences between maintenance and repair as forms of technical ac-
tivity have received surprisingly little attention. However, in a variety of
professional contexts, repair is often differentiated from maintenance by
emphasising its association with breakdown. According to this perspec-
tive, maintenance consists in monitoring and addressing developments in a
technology which may impact its ability to function such as the planned
replacement of parts or the renewal or replenishment of material such as
lubricants or protective coatings. Whereas this perspective casts mainte-
nance as routine work that addresses problems which typically occur
throughout the service life of an artefact, repair, on the other hand, is
aligned with unexpected developments which result in breakdown and loss
of function and is understood in terms of returning an artefact to some
prior condition (Perzanowski 2022, p. 15). Maintenance and repair are
therefore differentiated according to their temporal relation to breakdown:
maintenance is understood to take place before breakdown occurs, as op-
posed to repair which happens afterwards. Yet, while this way of distin-
guishing maintenance from repair may be common, as Aaron Perzanowksi
notes, drawing distinctions in this way leaves us unable to distinguish
maintenance or repair in a wide variety of cases:

Say your car drifts slightly from right to left on the freeway. Is a wheel
alignment from your local auto shop a repair or mere maintenance?
Functionality isn’t binary. Your car still works with misaligned wheels,
although not as well as you might prefer. The line between maintenance
and repair is malleable, but the distinction is still helpful.
(Perzanowski 2022, p. 16)

The problem with distinguishing maintenance from repair in this way, as

Perzanowksi correctly identifies, lies in the difficulty of drawing clear dis-
tinctions between cases in which a technology should be understood as
functional or non-functional. The challenges facing such a task are well
established within the philosophy of technology where the successful func-
tioning of an artefact is commonly considered to be a matter of degree or
perspective and something that depends on the skills and expectations of
the user.37 These difficulties underscore the limitations of conceiving the
difference between maintenance and repair as taking place before and after
failure, respectively.
A more nuanced conception of the relationship between maintenance
and repair can be found in common distinctions in engineering between
preventative and corrective maintenance. According to this approach, pre-
ventative maintenance encompasses activities performed in advance of the
failure of a system, artefact or component. Corrective maintenance, on the
20 Mark Thomas Young and Mark Coeckelbergh

other hand, addresses failures that have already occurred that threaten the
overall functionality or longevity of an artefact or system. According to
this perspective, examples such as wheel alignment, where attempts are
made to repair damage which does not prevent an artefact from fulfilling
its desired function entirely, are to be considered forms of corrective main-
tenance. Like maintenance, repair may also take place before breakdown.
However, rather than simply conflating maintenance with repair, the cor-
rective preventative distinction may be understood to preserve an exclu-
sive category of repair activities that correspond to cases that are not
commonly classified as maintenance.38 This applies to examples in which
technologies are restored to working order after a failure has led to a
break in their usage and includes cases such as the act of restoring an old
and abandoned car to working condition or rebuilding artefacts or sys-
tems after catastrophic failures. Yet while acknowledging the difference
between preventative and corrective maintenance may provide a more in-
tuitive way of distinguishing between maintenance and repair, this distinc-
tion is nonetheless susceptible to problems surrounding the ambiguity of
failure that we noted in Section 1.1.1. The slow, processual nature of
many forms of failure makes it difficult to pinpoint a precise moment
where failure occurs that could serve to distinguish preventative from cor-
rective interventions. As a result, the distinction between corrective and
preventative maintenance is widely acknowledged to be ambiguous in
many contexts.39
Finally, further insights concerning the nature of maintenance may be
revealed by exploring how it has been conceived in relation to practices of
retrofitting or rehabilitation in different contexts. A wide variety of differ-
ent kinds of artefacts undergo modification long after their production or
construction has ended in order to align them better with changing inter-
ests, needs or values. Examples of these kinds of modifications include the
widening of bridges or roads to facilitate greater levels of traffic or the
modification of vehicle engines to reduce carbon emissions. Among engi-
neers, these kinds of interventions are termed either retrofitting or rehabili-
tation and are often distinguished from maintenance and repair based on
the notion that the latter forms of activity are limited to restoring or pre-
serving the original designed form of an artefact. However, this distinction
does not always hold in practice. As Alampalli notes, in bridge engineering,
‘the line where maintenance ends and rehabilitation begins is very subjec-
tive. In most cases, the degree of repair, funding source, and agency policies
dictate if it is called maintenance activity or rehabilitation activity’ (Alampalli
2014). As we have already seen, by emphasising the transformative poten-
tial of maintenance, MRS scholarship has tended to downplay distinctions
between retrofitting and maintenance. In his study of seismic retrofits
of California Bridges, Benjamin Sims, for example, specifically identifies
 Keeping Things Going 21

retrofitting as a form of repair (Sims 2016). Akhil Gupta, on the other

hand, claims that ‘maintenance and repair often result in innovations that
enable infrastructures to function for periods far beyond their designed life
span and take on new functions that were not even envisioned when the
project was built’ (Gupta 2021).
The foregoing sections illustrate how the meaning of maintenance itself
is contested and inadequately conceptualised, leaving a number of basic
questions unanswered: What is maintenance? What are the motivations
and goals driving this form of technical activity? Does the term represent a
coherent category of technical practice, much like making? Or does the
term instead correspond to a heterogeneous collection of different forms of
activity? While existing literature on MRS has indeed taken steps towards
addressing these questions,40 a sustained discussion is yet to emerge.

1.2.2 Surveying the Landscape of Maintenance and Philosophy of

Technology

The range of philosophical issues related to maintenance and repair can be

understood to fall into three broad categories: ontological, epistemological
and normative. In Section 1.2.1, we already touched upon issues related to
the ontology of maintenance by exploring questions surrounding whether
maintenance represents a coherent and unified category of technical prac-
tice or whether the term instead corresponds to a heterogeneous collection
of different forms of activity. A different set of ontological issues relating
to maintenance concerns the objects of maintenance and repair itself.
Philosophers have long debated questions surrounding the ontological sta-
tus of artefacts and the kinds to which they are understood to belong. As
is well known, technical artefacts such as cars or screwdrivers straddle the
boundary between human culture and the material world insofar as they
cannot be adequately understood in terms of their physical characteristics
alone but must also be conceived in relation to the purposes they were cre-
ated or designed to achieve (Kroes & Meijers 2002). The nature of the re-
lationship between the teleological and physical character of artefacts is
itself the subject of debate among philosophers as is the extent to which
artefacts should be considered to depend on human minds for their exis-
tence. Existing literature on the philosophy of technology has tended to
explore these questions by focusing on the way in which artefacts are de-
signed, produced and used. What has received less attention however is the
way both the material and teleological characters of artefacts are sustained
and very often altered through the practice of maintenance and repair.
Exploring how artefacts are maintained over time can reveal new insights
into the role social phenomena play in the constitution of artefact kinds
and the nature of malfunction, and by doing so, raise a range of new
22 Mark Thomas Young and Mark Coeckelbergh

questions that the wide body of philosophical literature on technical func-

tions has yet to address in detail.41 Furthermore, by focusing attention on
the temporality of artefacts, the study of maintenance encourages us to
consider the role and significance accorded to change in the underlying
ontological frameworks we employ in our research.42
Many of the epistemological issues explored in the philosophy of tech-
nology have focused on the knowledge and processes by which new things
are created in engineering contexts. One common and longstanding line of
enquiry explores the extent to which knowledge involved in engineering
design can be understood to differ from scientific knowledge. Yet questions
remain concerning the nature of the engineering knowledge required to
maintain artefacts over time and the extent to which this differs from other
forms of technical knowledge. In what ways, for example, is the knowl-
edge required for maintenance and repair distinct from the knowledge em-
ployed in the design of artefacts? Despite common estimates which suggest
the majority of engineers engage in the practice of maintenance rather than
design,43 the knowledge engineers employ in tending to things over time
has received little attention. In the philosophy of engineering, epistemo-
logical issues pertaining to methodology, such as modelling and simula-
tions, experimentation, heuristics and processes of verification have been
examined primarily in relation to design. The same can be understood to
apply to the social epistemology of engineering which explores technical
norms and standards, collective decision-making and processes of justifica-
tion. Outside the context of engineering, questions remain concerning the
nature of maintenance knowledge and its employment or acquisition by
technicians or users. In what ways, for example, should we understand the
relationship between the situated character of maintenance and repair and
the transmission of repair knowledge through textbooks or online forums
and videos? Or to what extent does tacit knowledge, connoisseurship and
one’s own body play a role in understanding and diagnosing problems dur-
ing maintenance and repair?
The third category of philosophical issues relating to maintenance and
repair are normative. Existing literature on MRS has tended to approach
questions concerning the ethics of maintenance from the perspective of
care ethics or environmental ethics. However, work remains to be done
elaborating the connection between maintenance and ethics in these
fields, alongside situating ethical issues surrounding maintenance within
the landscape of contemporary philosophical ethics. While existential ap-
proaches to the topic have emphasised the value of establishing active rela-
tionships of engagement and care with technical artefacts, MRS theorists
have warned against the common tendency to romanticise maintenance,
drawing attention to the social conditions under which such work is often
performed (Mattern 2018). Similarly, whereas much scholarship in MRS
 Keeping Things Going 23

strives to counter the political and cultural neglect of maintenance by

drawing attention to the importance of the practices through which we
care for technologies over time, others warn us against uncritically valoris-
ing maintenance practices. Henke and Sims, for example, remind us that
‘repair can serve to prevent needed social change, exclude or remove cer-
tain groups of people from power and stabilise problematic infrastructures
and institutions, often at great social cost’ (Henke and Sims 2020, p. 15).
Exploring the values and interests motivating specific instances of mainte-
nance and repair in connection to social and environmental justice, there-
fore, represents an important opportunity, not only for deepening our
current understanding of the relation between the politics and ethics of
technology but also for contributing to a more nuanced discussion sur-
rounding the value of maintenance and repair.
The foregoing sections provide by no means an exhaustive list of issues;
however, they do give a sense of the breadth of possibilities that emerge
when philosophers of technology turn their attention to maintenance. By
representing the first foray into the topic of philosophy of technology, this
volume takes the first steps towards addressing these questions and pro-
vides the beginnings of what we hope will be an ongoing discussion.

1.3 Structure of the Volume

Unlike some themes in the philosophy of technology which are aligned
exclusively with certain traditions or schools of thought, the topic of main-
tenance cuts across such divisions. By showcasing a variety of different
philosophical styles, topics and approaches, the chapters in this volume
outline the breadth of the unrecognised significance of the topic of mainte-
nance to a diverse variety of subfields in the philosophy of technology,
from continental to anglophone traditions. In addition to illustrating the
potential of maintenance to advance thinking on a wide variety of philo-
sophical discussions, this volume therefore also functions to provide a
snapshot of the current diversity of methods, topics and approaches within
the field of philosophy of technology itself.
The volume is divided into two parts that cover different subfields in the
philosophy of technology. Part I, Metaphysics and Epistemology of
Maintenance begins with a chapter by Steven Vogel ‘Maintenance and the
Humanness of Infrastructure’ which explores the nature of the relationship
we share with the many different forms of infrastructure which frame ur-
ban life. Examining the Heideggerian notion that technologies in use recede
from view when functioning correctly, Vogel draws upon Marx in suggest-
ing that what becomes hidden is not so much their equipmentality but
rather their nature as products of human labour. However, against com-
mon tendencies to conceptualise this labour as something that occurred
24 Mark Thomas Young and Mark Coeckelbergh

in the past, as an historical fact, Vogel argues instead that we must recog-
nise infrastructure as continually produced through ongoing processes of
maintenance. By doing so, Vogel draws attention to inherent continuities
between the practices of making and maintaining.
In Chapter 3 ‘Technology in Process: Maintenance and the Metaphysics
of Artefacts’, Mark Thomas Young examines existing work on the meta-
physics of artefacts in order to demonstrate how the underlying framework
of substance ontology has hindered our understanding of the relationship
between technologies and time. Whereas the Western metaphysical tradi-
tion has traditionally approached questions concerning the persistence of
artefacts by attempting to demonstrate how they remain the same despite
undergoing change, Young examines how attending to maintenance re-
veals instead the way in which a wide range of artefacts persist only by
adapting to changing environments. In order to account for maintenance,
Young argues for the need to adopt a new underlying ontology centred
around processes rather than substances. In the final sections of his chap-
ter, Young sketches the outlines of such an approach and illustrates how a
process-based metaphysics of artefacts tends inevitably towards the em-
pirical analysis of social and political processes which determine the trajec-
tory of artefacts through time.
In Chapter 4, ‘There, I Fixed It! On the Status of Meaning and Repair’,
Tiago Mesquita Carvalho engages in a search for a philosophical theory of
the nature of technological repair. In doing so, he takes advantage of two
distinct sets of literature within the philosophy of technology: the meta-
physics of artefact functions and Ihde’s theory of postphenomenology, re-
vealing how important parallels between both approaches provide the
resources for a general theory of repair that recognises the capacity of ar-
tefacts to evolve over time in conjunction with the interests and desires of
their users. By examining how artefacts often derive their meaning from
contexts of use, Carvalho argues that repair may not necessarily seek to
restore proper function as intended by the designers of an artefact and may
instead seek new artefact forms and configurations. In the second section
of the chapter, Carvalho turns his attention to the epistemology of repair,
situating the knowledge and skills required to perform successful acts of
repair in relation to Aristotelian categories of techne and phronesis along-
side modern conceptions of tacit knowledge.
In ‘A Standpoint Epistemology of Repair?’, Cristina Bernabéu and Jesús
Vega-Encabo respond to Steven Jackson’s well-known calls for the devel-
opment of a standpoint epistemology of repair. After exploring the prom-
ise and potential of such an approach to yield a deeper understanding of
our wider technological environments, Bernabéu and Vega-Encabo turn
their attention to examining the central features of feminist standpoint
theories in order to ascertain the feasibility of developing such a theory in
 Keeping Things Going 25

relationship to maintenance and repair. While they identify several specific

difficulties facing the prospect of a standpoint epistemology of repair,
Bernabéu and Vega-Encabo nonetheless conclude their contribution by re-
affirming the broader epistemic potential of shifting our focus to the ongo-
ing activities of maintenance and repair through which our artefactual
worlds are sustained and reproduced.
Jochem Zwier’s chapter ‘Sustainability as Planetary Maintenance’ exam-
ines common analogies between maintaining technology and maintaining
the environment and illustrates how modern conceptions of sustainability
can be understood to arise from the application of conceptions of techno-
logical maintenance to the practices by which we care for the Earth. In
order to problematise this understanding of sustainability, Zwier traces the
metaphysical and ontological roots of the concept of technological mainte-
nance back to Aristotle, revealing essential hylomorphic features that fail
to be reflected in modern understandings of sustainability. According to
Zwier, these limitations ultimately undermine the meaning of the concept
itself, rendering sustainability a nihilistic form of striving that lacks an
objective. In the search for a more robust conceptualisation of sustainabil-
ity, Zwier explores the possibility of alternative understandings based not
on the conceptual resources of Aristotelianism, but rather on the work of
French philosopher of technology, Gilbert Simondon. Drawing upon
Simondon’s conception of technological individuation as an exploratory
and open-ended process of becoming which arises through establishing
new correspondences between different, previously unrelated aspects of
the world, Zwier examines how this theory holds the potential to inform
valuable new understandings not only of sustainability but also the nature
of maintenance itself.
In ‘Towards a Realist Metaphysics of Software Maintenance’, Keith
Begley examines the extent to which computational entities should be con-
sidered abstract objects which are therefore immaterial and unchanging
and whether this might complicate their status as potential objects of main-
tenance. Drawing on the account of abstract objects outlined by Jerrold
Katz, Begley argues that computational artefacts should be considered
composite objects comprised of both abstract and concrete constituents
and examines how this accounts for the widely discussed ‘dual nature’ of
such artefacts. In the second section of his chapter, Begley turns his atten-
tion to examining what the status of computational artefacts as realisations
of their creators’ intentions means for how we understand the nature of
their maintenance over time. In contrast to conservative conceptions which
understand maintenance to consist of the preservation of an original form
intended by a maker, Begley illustrates how software comprises a class of
artefacts which are explicitly intended by their creators to evolve over time,
providing a rich example of the transformative nature of maintenance.
26 Mark Thomas Young and Mark Coeckelbergh

In ‘Maintaining Perpetual Actuality in the Digital Age? Simondon’s

Conception of Maintenance and the Networked Era’, Johannes Schick ex-
amines the role maintenance plays in the account of technical objects de-
veloped by Gilbert Simondon and explores to what extent this account
remains relevant in our current digital age. As Schick demonstrates,
Simondon understood maintenance to be a mode of human engagement
with the internal structure of a technical object and its networked relations
to other objects through which we continually adjust it to the present mo-
ment. However, unlike the mechanical and electrical objects around which
Simondon developed his theories, technical objects in the digital age, such
as smartphones, appear to resist these forms of engagement in different
ways. In response to this concern, Schick illustrates how the use and devel-
opment of smart devices nonetheless conform to the processes of concreti-
sation and technicity which form the central elements of his account of
technological maintenance. In the final sections of his chapter, Schick ex-
amines how Simondon considered maintenance to operate in a temporal
space whereby the history of a technical object is related to a possible fu-
ture and draws from Simondon’s pedagogical theories a notion of story-
telling which he suggests may help us in overcoming the difficulties we face
surrounding the histories of technical objects in the digital age.
Part II of the volume Ethics, Politics and Aesthetics of Maintenance be-
gins with a chapter by Steffen Steinert titled ‘Maintenance of Value and the
Value of Maintenance’, which examines the relationship between technol-
ogy, value and maintenance practices by focusing specifically on socio-
technical systems and their lifecycles. Steinert’s contribution addresses the
lack of attention to the practice of maintenance in current discussions sur-
rounding the relationship between technology and values and expands be-
yond current literature in looking beyond how values come to be embedded
in technologies during the design phase to also include how values are
sustained and transformed through practices of maintenance. In elaborat-
ing what he terms the ‘maintenance perspective’, Steinert examines the
complex interplay between the preservation and transformation of values
in relation to three key aspects of socio-technical systems: agents, institu-
tions and hardware. In contrast to much current work on values and tech-
nology that often assumes the nature of technologies to remain fixed in the
face of evolving values, Steinert instead foregrounds how the stability and
change of values over time are reflected in the historical evolution of socio-
technical systems themselves.
In his chapter ‘An Eco-Ethics for the End of the Anthropocene: Finding
Ethical and Sustainable Paths through Consumerism, Disposability and
Planned Obsolescence’, Simon Penny examines broad questions concern-
ing the relationship between personal practices of consumption in con-
sumer societies and the impact of global cycles of production, consumption
 Keeping Things Going 27

and waste on the environment. By examining how planned obsolescence

manifests itself in the use of synthetic and composite materials that decom-
pose within planned timeframes into toxic residues, Penny draws attention
to the way consumer products often resist repair, recycling or reuse.
Reflecting on the accumulation of waste and environmental consequences
that result from such practices, Penny argues that each of us has a respon-
sibility to subvert the cycles of consumption and disposal which typically
characterise life in contemporary consumer societies. Drawing on his
broad experience upcycling materials as an artist and craftsman, he
sketches the outlines of a salvage ethics – a form of life centred around
repurposing rather than the disposal of artefacts which no longer serve
their purpose.
In ‘Aesthetic Values in the Maintenance of Urban Technologies’, Sanna
Lehtinen approaches the topic of maintenance from the perspective of
urban aesthetics. Departing from common conceptions of urban technol-
ogies in infrastructure studies as an invisible substratum upon which life
in the city plays out, Lehtinen instead examines how the way in which
different technologies make cities look and feel is a question of crucial
importance in determining how they should be maintained over time. By
examining cases of deliberation and disagreement concerning the evolu-
tion of urban technologies, Lehtinen’s chapter illustrates how aesthetic
values often intersect with social and ethical values and argues for the
importance of directly addressing aesthetic values in decision-making
processes regarding urban maintenance.
In ‘Negotiating Visions of Waste: On the Ethics of Maintaining Waste
Infrastructures’, Joost Alleblas and Ben Hofbauer explore the complex re-
lationships people share with waste in high-income consumer societies.
Their contribution illustrates how conceptions of waste as valueless are
reinforced by the design and operation of municipal waste systems which
render the flow and management of waste invisible in urban settings. By
exploring how values are embedded, afforded and resisted through the use
of socio-technical systems, Alleblas and Hofbauer illustrate how the pres-
ervation of these systems hinders the development of positive affirmations
of waste as a resource that is necessary for the establishment of more sus-
tainable practices of municipal waste management. The first step towards
achieving such a goal, they argue, lies in rethinking the nature of infra-
structure maintenance as a dynamic and reflexive process rather than the
activity of caring for a stable and rigid system.
In the final chapter ‘Repairing AI’, Taylor Stone and Aimee van
Wynsberghe investigate how attending to practices of maintenance and
repair holds the potential to enrich existing discourse on sustainable AI
in the field of AI ethics. Specifically, their chapter focuses on the way cur-
rent attempts to employ artificial intelligence to address environmental
28 Mark Thomas Young and Mark Coeckelbergh

problems are complicated by the underlying environmental problems

generated by the ongoing production of the technology itself. In order to
address this problem, Stone and van Wynsberghe argue for the need to
subvert the dominant tendency in AI Ethics to search for solutions at the
level of software design by reorienting our focus towards the global net-
works of material infrastructure and human labour from which the tech-
nology itself emerges. By approaching artificial intelligence as an evolving
and dynamic process rather than a fixed and stable entity, Stone and van
Wynsberghe elaborate on a conception of repair which extends beyond
merely restoring the functionality of a technology and captures the re-
flexive processes through which the development of technological sys-
tems are managed over time. From this perspective, they argue,
breakdown presents an opportunity not only to restore the status quo
but also to transform technologies in order to address emerging interests
and concerns and by doing so, attain more ethical and sustainable
configurations.
Together, these chapters signal the arrival of maintenance as an emerging
topic in the philosophy of technology. By shifting attention to what hap-
pens after technologies are produced, they represent the first steps within
the field towards a more complete understanding of the development of
technologies over time while at the same time addressing a range of con-
temporary concerns surrounding the technological cultures we currently
inhabit. It is our hope that this volume will contribute to and inspire fur-
ther philosophical and interdisciplinary work on a phenomenon and aspect
of technology and technological practice that is still not fully understood
or valued, both in academic literature and beyond.

Notes
1 The few exceptions include Young (2021a, 2021b); Karakas (2020). See also
the brief account of maintenance in Mitcham (1994, pp. 233–235).
2 For an account of how our cities would fare in the absence of constant mainte-
nance, see Weisman (2007, chap 3).
3 See, for example, Russell & Vinsel (2019); Mattern (2018); Edgerton (2006,
p. 100). For a discussion of this point in relation to software development see
Begley (this volume).
4 Examples abound of technologies that defy their producer’s expectations, see,
for example, Tenner (1996); Huesemann and Huesemann (2011).
5 See Rosenberg (1982; esp chap 6). For a discussion of the implications of this
on policy-making regarding technology development, see Collingridge (1980).
6 For an overview of the theme of care within maintenance and repair studies, see
Jackson (2016).
7 See, for example, Mattern (2018); Denis and Pontille (2015); de la Bellacasa
(2011); Jackson (2014).
 Keeping Things Going 29

8 Interestingly, Pirsig identified a relationship of care as the essence of mainte-

nance already in 1974, before the emergence of care ethics in feminist theory
during the 1980s. Reflecting on a mechanic’s failed attempt to repair his motor-
cycle, Pirsig noted that
The mechanics in their attitude toward the machine were really taking no
different attitude from the manual’s toward the machine or from the attitude
I had when I brought it in there. We were all spectators. And it occurred to
me that there is no manual that deals with the real business of motorcycle
maintenance, the most important aspect of all. Caring about what you are
doing is considered either unimportant or taken for granted.
(Pirsig 1974, p. 25)
9 See, for example, Tronto (1993, pp. 127–31; 1995).
10 Situatedness represents a key theme in maintenance and repair literature, where
many emphasise the situated character of repair work specifically. As Steven
Jackson notes, repair work is deeply embedded
in contexts that structure the need and possibility of repair, and furnish the
resources available for its accomplishment. In this view, repair work is deeply
contingent, dependent on social and material configurations grown and
evolved in particular times and places. Much of the diagnostic work embed-
ded in repair is in fact about calling these configurations to light, both to
understand the nature of breakdowns and problems encountered and to craft
effective and doable responses from the resources that context affords.
(Jackson 2019)
11 As stated by Willie in Douglas Harper’s Working Knowledge,
If you rush things, you can’t enjoy them. And it’s a challenge – no job’s the same.
If you had a thousand jobs in a year, not two of those thousand jobs would be
the same. Even the ones that are supposed to be the same aren’t. Things are
broken or worn in different ways – they each have their own characteristics.
(Harper 1987, p. 169)
12 See Penny (this volume).
13 See (Kroes & Meijers 2000; Achterhuis 2001).
14 For an overview of these approaches see (Van den Hoven, Vermaas & Van de
Poel 2015; Brey 2022).
15 See, for example, Stone & van Wynsberghe (this volume) and Zwier (this
volume).
16 See, for example, MRS studies which draw on the work of Dewey (Schubert
2019; Tironi 2019), Hans Jonas (Dant 2019) or Heidegger (Jackson 2014;
Graham and Thrift 2007).
17 Among the earliest studies to explore practices of maintenance were (Cowan
1983; Harper 1987).
18 See, for example, (Orr 1996; Henke 2000; Dant & Bowles 2003). For an influ-
ential work of architectural history from this period which focuses on mainte-
nance see (Brand 1994).
19 For criticisms of the dominant tendency within the history of technology to
focus on invention at the expense of use, see (Pursell 1995; Edgerton 1999;
30 Mark Thomas Young and Mark Coeckelbergh

Oldenziel 2001). For studies criticising the common tendency in social theory
to understand users as passive consumers of technology, see (de Certeau 1984;
Oudshoorn & Pinch 2003).
20 Among the most influential studies on the topic from this period are Jackson
(2014), Denis and Pontille (2015), Russell and Vinsel (2018).
21 For an account of the establishment of this community, see (Russell & Vinsel
2018; Vinsel & Russell 2020, p. 16).
22 In addition to the classic ethnographic accounts of maintenance and repair
(Orr 1996; Henke 2000; Harper 1987; Dant & Bowles 2003; Denis & Pontille
2015), see also the recent collection (Strebel, Bovet & Sormani 2019).
23 For an overview of ethnomethodology see (Sharrock 1989; vom Len 2016). For
studies of maintenance and repair which engage explicitly with ethnomethod-
ology, see (Suchman 1987; Henke 2000; Bovet & Strebel 2019).
24 For differences between conversation analysis and ethnomethodology, see
(Lynch 1993, chap 6).
25 See, for example, (Orr 1996; Downey 1998; Henke 2000).
26 For a thorough discussion of this idea in Heidegger’s work, see (Harman 2002).
27 For statements of this view in relation to the case of language use, see
(Wittgenstein 1980, §911, 161e; (Wittgenstein 2009, §122, 54e).
28 For an overview of the topic of invisible labour in feminist theory and sociol-
ogy, see (Crain, Winifred & Cherry 2016).
29 For overviews of the theme of dirty work, see (Hughes 1971; Simpson et al.
2012).
30 See (Dant & Bowles 2003) for an important study in MRS that draws explicitly
on this tradition.
31 See, for example, (Denis & Pontille 2017; Denis & Pontille 2018).
32 For an overview of the theme of vulnerability in feminist theory, see (Mackenzie,
Rogers & Dodds 2014).
33 See, for example, (Connolly 2013; Ramakrishnan, O’Reilly & Budds 2021) See
also discussions of network fragility in actor–network theory (Callon 1986;
Higgins & Kitto 2004; Lovell & Smith 2010).
34 See also Young (this volume).
35 This particular view finds its fullest articulation in infrastructure studies. See
also recent work in the philosophy of technology on the nature of artefacts as
processes (Young 2021a; Coeckelbergh 2023).
36 See above n.15.
37 See, for example, (Radder 2009; Karakas 2020).
38 Compare this interpretation to that of Jonathan Chapman, who notes that the
distinction between preventative and corrective maintenance ‘is important, as
notions of maintenance and repair are often incorrectly assigned as opposi-
tional ideas. Rather, maintenance and repair are aspects of the same thing, oc-
cupying different ends of a “spectrum of care”, with preventative maintenance
(e.g., washing and ironing a shirt) at one end, and corrective maintenance (e.g.,
replacing a broken button) at the other’ (Chapman 2021, p. 38).
39 Despite this conceptual ambiguity, distinctions between preventative and cor-
rective maintenance remain an important distinction in practice in many con-
texts, especially those where the costs of failure are high. For a brief historical
account of the emergence of conceptions of preventative maintenance, see
(Vinsel & Russell 2020, chap 3).
40 For an excellent recent study examining conceptualisations of maintenance see
(Denis & Pontille 2022).
41 For an overview of this literature, see (Preston 2009; Franssen 2013).
 Keeping Things Going 31

42 See Young (this volume).

43 See, for example, Edgerton (2006, p.xv); Russell and Vinsel (2019); Vinsel
(2019).

Bibliography
Achterhuis, Hans (ed) Robert P. Crease (trans) (2001) American Philosophy of
Technology: The Empirical Turn Bloomington: Indiana University Press.
Alampalli, Sreenivas. (2014) “Bridge Maintenance” in Wai-­Fah Chen & Lian Duan
(eds) Construction and Maintenance (Bridge Engineering Handbook) 2nd ed.
Boca Raton: CRC Press pp. 269–300.
Barnes, Jessica. (2017) “States of Maintenance: Power, politics, and Egypt’s irriga-
tion infrastructure” Environment and Planning D: Society and Space 35(1)
pp. 146–164.
Brand, Stewart. (1994) How Buildings Learn: What Happens After They’re Built.
New York: Penguin Books.
Brey, Philip. (2022) “Understanding Engineering Design and Its Social, Political,
and Moral Dimensions” in Shannon Vallor (ed) The Oxford Handbook of
Philosophy of Technology. New York: Oxford University Press pp. 395–416.
Bovet, Alain & Strebel, Ignaz (2019) “Job Done: What Repair Does to Caretakers,
Tenants and Their Flats” in Ignaz Strebel, Alain Bovet & Phillipe Sormani (eds)
Repair Work Ethnographies: Revisiting Breakdown, Relocating Materiality.
Singapore: Palgrave Macmillan pp. 89–125.
Callon, Michel. (1986) “Some Elements in a Sociology of Translation: Domestication
of the scallops and fisherman of St. Brieuc Bay” in John Law (ed) Power, Action
and Belief: A New Sociology of Knowledge? London: Routledge & Kegan Paul
pp. 196–233.
de Certeau, Michel. (1984) Steven Rendall (trans) The Practice of Everyday Life.
Berkely: University of California Press.
Chapman, Jonathan. (2021) Meaningful Stuff: Design That Lasts. Massachusetts:
MIT Press.
Coeckelbergh, Mark. (2023) “Technology as Process” in Glenn Miller, Helena M
Jerónimo & Qin Zhu (eds) Thinking Through Science and Technology:
Philosophy, Religion and Politics in an Engineered World. Maryland: Rowman
& Littlefield pp. 55–68.
Collingridge, David. (1980) The Social Control of Technology. London: Frances
Pinter.
Connolly, William E. (2013) “The ‘New Materialism’ and the Fragility of Things”
Millennium: Journal of International Studies 41(3) 399–412.
Cowan, Ruth Schwartz. (1983) More Work For Mother: The Ironies of Household
Technology from the Open Hearth to the Microwave. New York: Basic Books.
Crain, Marion G., Winifred, Poster R & Cherry, Miriam A. (eds) (2016) Invisible
Labor: Hidden Work in the Contemporary World. Oakland: University of
California Press.
Crawford, Matthew B. (2009) Shop Class as Soulcraft: An Inquiry into the Value
of Work. New York: Penguin Press.
Dant, Tim (1999) Material Culture In The Social World: Values, Activities,
Lifestyles. Buckingham: Open University Press.
32 Mark Thomas Young and Mark Coeckelbergh

Dant, Tim (2019) “Inside the Bicycle: Repair Knowledge for All” in Ignaz Strebel,
Alain Bovet & Phillipe Sormani (eds) Repair Work Ethnographies: Revisiting
Breakdown, Relocating Materiality. Singapore: Palgrave Macmillan pp. 283–312.
Dant, Tim & Bowles, David. (2003) “Dealing with Dirt: Servicing and Repairing
Cars” Sociological Research Online 8(2). Available at http://www.socresonline.
org.uk/8/2/dant.html (accessed January 20, 2021).
De la Bellacasa, Maria Puig. (2011) “Matters of Care in Technoscience: Assembling
Neglected Things” Social Studies of Science 41(1) pp. 85–106.
Denis, David J. (Denis, Jérôme & Pontille, David) (2020) “Why do Maintenance and
Repair Matter?” in Anders Blok, Ignacio Fárias & Celia Roberts (eds) The Routledge
Companion to Actor-Network Theory. New York: Routledge pp. 283–293.
Denis, Jérôme & Pontille, David. (2014) “Maintenance work and the performativ-
ity of urban inscriptions: the case of Paris Subway Signs” Environment and
Planning D: Society and Space 32(3) pp. 404–416.
Denis, Jérôme & Pontille, David. (2015) “Material Ordering and the Care of
Things” Science, Technology & Human Values 40(3) pp. 338–367.
Denis, Jérôme & Pontille, David. (2017) “Object Ontologies and the contrasted in/
visibilities of maintenance work” Maintainers II conference, April 2017,
Hoboken (New Jersey). Available at: http://themaintainers.wpengine.com/wp-
content/uploads/2021/04/DenisPontilleObjectOntologiesMaintenanceWork.pdf
(accessed July 22, 2023).
Denis, Jérôme & Pontille, David. (2018) “The Multiple Walls of Graffiti Removal.
Maintenance and Urban Assemblage in Paris” in A.M. Brighenti & Kärrolm, M.
(eds) Urban Walls: Political and Cultural Meanings of Vertical Structures
New York: Routledge pp. 215–235.
Denis, Jérôme & Pontille, David. (2022) Le soin de choses: Politiques de la main-
tenance. Paris: La Découverte.
Denis, Jérôme, Mongili, Allessandro & Pontille, David. (2015) “Maintenance and
Repair in Science and Technology Studies” Tecnoscienza: Italian Journal of
Science and Technology Studies 6(2) pp. 5–15.
Downey, Gary Lee. (1998) The Machine in Me: An Anthropologist Sits Among
Computer Engineers. New York: Routledge.
Edgerton, David. (1999) “From innovation to use: ten eclectic theses on the histo-
riography of technology” History and Technology: An International Journal
16(2) pp. 111–136.
Edgerton, David. (2006) The Shock of the Old: Technology and Global History
Since 1900. London: Profile Books.
Edgerton, David. (2010) Technology and Culture. 51(3) pp. 680–697.
Fisher, Berenice & Tronto, Joan C. (1991) “Towards a Feminist Theory of Care”
in Emily Abel & Margaret Nelson (eds) Circles of Care: Work and Identity in
Women’s Lives. Albany: State University of New York Press.
Franssen, Maarten. (2013) “Artifacts” in D. Pritchard (ed) Oxford Bibliographies
Online: Philosophy. Available at: https://www.oxfordbibliographies.com/display/
document/obo-9780195396577/obo-9780195396577-0145.xml (accessed June
10, 2021).
Graham, Stephen. (2010) “When Infrastructures Fail” in Stephen Graham (ed)
Disrupted Cities: When Infrastructure Fails. New York: Routledge pp. 1–26.
 Keeping Things Going 33

Graham, Stephen & Thrift, Nigel. (2007) “Out of Order: Understanding

Maintenance and Repair” Theory, Culture & Society 24(3) pp. 1–25.
Gupta, Akhil. (2021) “Infrastructure as Decay and the Decay of Infrastructure” in
Ghassan Hage (ed) Decay. Durham: Duke University Press pp. 37–46.
Harman, Graham. (2002) Tool Being: Heidegger and the Metaphysics of Objects.
Illinois: Open Court Publishing.
Harper, Douglas. (1987) Working Knowledge: Skill and Community in a Small
Shop. Chicago: Chicago University Press.
Henke, Christopher R. (2000) “The Mechanics of Workplace Order: Toward a
Sociology of Repair. Berkeley Journal of Sociology Vol. 44 pp. 55–81.
Henke, Christopher R. (2019) “Negotiating Repair: The Infrastructural Contexts
of Practice and Power” in Ignaz Strebel, Alain Bovet & Phillipe Sormani (eds)
Repair Work Ethnographies: Revisiting Breakdown, Relocating Materiality.
Singapore: Palgrave Macmillan pp. 255–282.
Henke, Christopher R & Sims, Benjamin (2020) Repairing Infrastructures: The
Maintenance of Materiality and Power Massachusetts: MIT Press.
Higgins, Vaughan & Kitto, Simon. (2004) “Mapping the Dynamics of New Forms
of Technological Governance in Agriculture: Methodological Considerations”
Environment and Planning A 36(8) pp. 1397–1410.
Van den Hoven, Jeroen, Vermaas, Pieter E. & Van de Poel, Ibo (eds) (2015)
Handbook of Ethics, Values and Technological Design: Sources, Theory, Values
and Application Domains. Dordrecht: Springer Netherlands.
Huesemann, Michael & Huesemann, Joyce. (2011) Techno-fix: Why Technology
Won’t Save Us or the Environment. Gabriola Island: New Society Publishers.
Hughes, Everett C. (1971) The Sociological Eye: Selected Papers. Chicago: Aldine
Atherton.
Jackson, Steven J. (2014) Rethinking Repair. In T. Gillespie, P. Boczkowski and
K. Foot (eds.), Media Technologies: Essays on Communication, Materiality and
Society. Cambridge: MIT Press, pp. 221–240.
Jackson, Steven J. (2016) “Speed, Time, Infrastructure: Temporalities of Breakdown,
Maintenance and Repair” in Judy Wajcman & Nigel Dodd (eds) The Sociology
of Speed: Digital, Organizational and Social Temporalities. New York: Oxford
University Press pp. 169–185.
Jackson, Steven J. (2019) “Repair as Transition: Time, Materiality and Hope” in Ignaz
Strebel, Alain Bovet & Phillipe Sormani (eds) Repair Work Ethno­graphies: Revisiting
Breakdown, Relocating Materiality. Singapore: Palgrave Macmillan pp. 337–347.
Jones, Siân & Yarrow, Thomas (2013) “Crafting Authenticity: An Ethnography of
Conservation Practice” Journal of Material Culture 18(1) pp. 3–26.
Karakas, Alexandra. (2019) “Maintenance, Function and Malfunction in Tech­
nology” Informácios Társadalom 19(4) pp. 40–49.
Kroes, Peter & Meijers, Anthonie (eds) (2000) The Empirical Turn in the Philosophy
of Technology. Amsterdam: JAI-Elsevier.
Kroes, Peter & Meijers, Anthonie. (2002) “The Dual Nature of Technical Artifacts –
presentation of a new research programme” Techné: Research in Philosophy and
Technology 6(2) pp. 4–8.
Lovell, Heather & Smith, Susan J. (2010) “Agencement in housing markets: The
case of the UK construction industry” Geoforum 41(3) pp. 457–468.
34 Mark Thomas Young and Mark Coeckelbergh

Lejeune, Christoph (2019) “Interruptions, Lunch Talks, and Support Circles: An

Ethnography of Collective Repair in Steam Locomotive Restoration” in Ignaz
Strebel, Alain Bovet & Phillipe Sormani (eds) Repair Work Ethnographies:
Revisiting Breakdown, Relocating Materiality. Singapore: Palgrave Macmillan
pp. 221–251.
Lynch, Michael. (1993) Scientific Practice and Ordinary Action: Ethnomethodology
and Social Studies of Science. Cambridge: Cambridge University Press.
Mackay, Hughie & Gillespie, Gareth. (1992) “Extending the Social Shaping of
Technology Approach: Ideology and Appropriation” Social Studies of Science
22(4) pp. 685–716.
Mackenzie, Catriona, Rogers, Wendy & Dodds, Susan (eds) (2014) Vulnerability:
New Essays in Ethics and Feminist Philosophy. New York: Oxford University
Press.
Mattern, Shannon. (2018) “Maintenance and Care” Places Journal. Available at
https://placesjournal.org/article/maintenance-and-care/ (accessed June 20, 2021).
McHoul, Alex. (1998) “Can Ethnomethodology be Heideggarian?” Human
Studies 21 pp. 13–26.
Mitcham, Carl. (1994) Thinking Through Technology: The path between Engi­
neering and Philosophy Chicago: University of Chicago Press.
Orr, Julian E. (1996) Talking about Machines: An Ethnography of a Modern Job.
New York: Cornell University Press.
Oudshoorn, Nelly & Pinch, Trevor. (eds) (2003) How Users Matter: The Co-
Construction of Users and Technology. Cambridge: MIT Press.
Oldenziel, Ruth. (2001) “Man the Maker, Woman the Consumer: The Consumption
Junction Revisited” in N. Creager, E. Lunbeck & L. Sheibinger (eds) Feminism
in Twentieth-Century Science, Technology and Medicine. Chicago: Chicago
University Press.
Perzanowski, Aaron. (2022) The Right to Repair: Reclaiming the Things We Own.
Cambridge: Cambridge University Press.
Pirsig, Robert M. (1974) Zen and the Art of Motorcycle Maintenance. New York:
Bantam Books.
Preston, Beth. (2009) “Philosophical Theories of Artifact Function” in Anthonie
Meijers (ed) Philosophy of Technology and Engineering Sciences (Handbook of
the Philosophy of Science Vol 9). Burlington: Elsevier pp. 213–233.
Pursell, Carroll (1995) “Seeing the Invisible: New Perspectives in the History of
Technology” Icon, 1, 9–15.
Radder, Hans. (2009) “Why Technologies are Inherently Normative” in Anthonie
Meijers (ed) Philosophy of Technology and Engineering Sciences (Handbook of
the Philosophy of Science Vol 9). Burlington: Elsevier pp. 887–921.
Rosenberg, Nathan. (1982) Inside the Black Box: Technology and Economics.
Cambridge: Cambridge University Press.
Ramakrishnan, Kavita, O’Reilly, Kathleen & Budds, Jessica (2021) “The Temporal
Fragility of Infrastructure: Theorizing decay, maintenance and repair” EPE:
Nature and Space 4(3) pp. 674–695.
Rubio, Fernando Domínguez. (2016) “On the discrepancy between objects and
things: An Ecological Approach” Journal of Material Culture 21(1) pp. 59–86.
 Keeping Things Going 35

Rubio, Fernando Domínguez. (2020) Ecologies of the Modern Imagination at the

Art Museum. Chicago: University of Chicago Press.
Russell, Andrew L. & Vinsel, Lee. (2018) “After Innovation, Turn to Maintenance”
Technology and Culture 59(1) pp. 1–25.
Russell, Andrew L. & Vinsel, Lee. (2019) “Make Maintainers: Engineering
Education and an Ethics of Care” in M. Wisnioski, E. Hintz & M. Kleine (eds)
Does America Need More Innovators? Cambridge: MIT Press pp. 249–269.
Schubert, Cornelius. (2019) “Repair Work as Inquiry and Improvisation: The
Curious Case of Medical Practice” in Ignaz Strebel, Alain Bovet & Phillipe
Sormani (eds) Repair Work Ethnographies: Revisiting Breakdown, Relocating
Materiality. Singapore: Palgrave Macmillan pp. 31–60.
Sharrock, Wes. (1989) “Ethnomethodology” Berkeley Journal of Sociology 40(4)
pp. 657–677.
Simpson, Ruth, Slutskaya, Natasha, Lewis, Patricia & Höpfl, Heather (eds) (2012)
Dirty Work: Concepts and Identities. New York: Palgrave Macmillan.
Sims, Benjamin. (2016) “Seismic shifts and retrofits: scale and complexity in the
seismic retrofit of California bridges” in Mike Hodson & Simon Marvin (eds)
Retrofitting Cities: Priorities, Governance and Experimentation New York:
Routledge, pp. 13–33.
Spelman, Elizabeth V. (2002) Repair: The Impulse to Restore in a Fragile World.
Boston: Beacon Press.
Staudenmaier, John M. (1989) Technology’s Storytellers: Reweaving the Human
Fabric. Baskerville: MIT Press.
Strebel, Ignaz, Bovet, Alain & Sormani, Philippe (eds) (2019) Repair Work Ethno­
graphies: Revisiting Breakdown, Relocating Materiality. Singapore: Palgrave
Macmillan
Suchman, Lucy A. (1987) Plans and Situated Actions: The Problem of Human-
Machine Communication. Cambridge: Cambridge University Press.
Tenner, Edward. (1996) Why Things Bite Back: Technology and the Revenge of
Unintended Consequences. New York: Vintage Books.
Tironi, Martin. (2019) “Inquiring and Experimenting with Urban Ecology:
Pragmatist Lessons from Public Bicycle Repair” in Ignaz Strebel, Alain Bovet &
Phillipe Sormani (eds) Repair Work Ethnographies: Revisiting Breakdown,
Relocating Materiality. Singapore: Palgrave Macmillan pp. 187–219.
Tronto, Joan C. (1993) Moral Boundaries: A Political Argument for an Ethic of
Care New York: Routledge.
Tronto, Joan C. (1995) “Women and Caring: What Can Feminists Learn About
Morality from Caring?” in Virginia Held (ed) Justice and Care: Essential
Readings in Feminist Ethics New York: Routledge.
Weisman, Alan. (2007) The World Without Us. New York: Thomas Dunne Books.
Verbeek, Peter-Paul (2005) What Things Do: Philosophical Reflections on
Technology, Agency and Design. Robert P. Crease (trans). University Park:
Pennsylvania State University Press.
Verbeek, Peter-Paul (2022) “The Empirical Turn” in Shannon Vallor (ed) The
Oxford Handbook of Philosophy of Technology. New York: Oxford University
Press pp. 35–54.
36 Mark Thomas Young and Mark Coeckelbergh

Vinsel, Lee & Russell, Andrew L. (2020) The Innovation Delusion: How Our
Obsession With The New Has Disrupted The Work That Matters Most.
New York: Currency.
Vinsel, Lee (2019) Researching “What Do Engineers Do All Day?”: Innovation,
Maintenance and Everyday Engineering. The Maintainers.Org. Available at:
https://themaintainers.org/researching-what-do-engineers-do-all-day-innovation-
maintenance-and-everyday-engineering/. Accessed August 10, 2023.
vom Len, Dirk. (2016) The Creation and Development of Ethnomethodology.
New York: Routledge.
Weber, Heike & Krebs, Stefan (2021) “The Persistence of Technology: From
Maintenance and Repair to Reuse and Disposal” in Stefan Krebs & Heike Weber
(eds) The Persistence of Technology: Histories of Repair, Reuse and Disposal.
Bielefeld: Transcript Verlag pp. 9–26.
Wittgenstein, Ludwig. (1980) Remarks on the Philosophy of Psychology, vol.1.
Oxford: Basil Blackwell.
Wittgenstein, Ludwig. (2009) Philosophical Investigations, rev. 4th ed. Oxford:
Wiley Blackwell.
Young, Mark Thomas. (2021a) “Maintenance” in N. Doorn & D. Michelfelder
(eds) The Routledge Handbook of the Philosophy of Engineering. New York:
Routledge.
Young, Mark Thomas. (2021b) “Now You See it (Now You Don’t): Users,
Maintainers and the Invisibility of Infrastructure” in M. Nagenborg, T. Stone,
M. Woge & P. Vermaas (eds) Technology and the City: Towards a Philosophy of
Urban Technologies. Springer, pp. 101–120.
Part I

Metaphysics and Epistemology

of Maintenance
2 Maintenance and the
Humanness of Infrastructure
Steven Vogel

I
An important characteristic of urban infrastructure seems to be its invisi-
bility. It’s infra, ‘below’, very often literally, and therefore hidden: our life
in cities depends in great measure on the water mains, electrical conduits,
fibre optic cables, steam tunnels and so forth that are constantly there
working beneath our feet but which we rarely consider. Even when ele-
ments of infrastructure are not so literally buried below us (e.g., roads,
bridges, sidewalks, cell phone towers, train tracks), we still hardly ever
notice them, and instead simply use them, in a way that takes them for
granted as a part of urban life that seems to take place beyond (or ‘above’)
them. Our life takes place in a certain sense with them without as it were
ever being about them.
They are tools, that is to say: they make urban life possible (it certainly
would grind to a halt without them) but are not themselves what that life
focuses on. In that sense, we could call them ‘invisible’, although not with-
out some hesitation: we can and do of course see them – even the under-
ground ones, sometimes – but in an odd sort of unseeing sight, a vague
awareness (like our awareness of our clothes, say, there when we think
about them and, of course, there even when we don’t – and yet when we
don’t nonetheless somehow invisible).1
I would like in this essay to investigate the nature of this invisibility or
quasi-invisibility of infrastructure, this curious hiddenness of that upon
which we rely for our lives in cities (which means, for most of us: our
lives), asking about what exactly is hidden about it and in what that hid-
denness consists. In doing so, I want to raise in particular the questions of
what infrastructure actually is and what its relationship is to the human
beings without whom it would not exist at all: the ones who use it, first of
all, but also the ones who build it and finally and most importantly – and
here is the connection to the broader concerns of this volume – those who
so crucially and necessarily maintain it.

DOI: 10.4324/9781003316213-3
40 Steven Vogel

The alert reader with some familiarity with continental philosophy will
already have thought of Heidegger and the well-known account he pro-
vides in Being and Time of what he calls Zuhandenheit. These important
passages point out that the ‘things’ with which we most directly have to do
in our everyday lives are tools, equipment, Zeug: the chairs, screens, lights,
etc., that are all being employed right now as I write, as you read, but all
with the same character just mentioned of not being noticed, of being as it
were taken for granted. It is only when such objects break, Heidegger ar-
gues (or get in our way or turn out to be missing), that we suddenly notice
them and bring them – typically in frustration or even in panic – into the
centre of our attention. Their mode of Being, as he puts it, switches from
Zuhandenheit to Vorhandenheit: from being (as one translation puts it)
‘ready-to-hand’ to being ‘present-at-hand’ (Heidegger 2008 ss. 15–6).
It is tempting to use this analysis to talk about the sort of infrastructure
that I have been discussing.2 Infrastructure, one might say, is precisely the
sort of Zeug that Heidegger has in mind in these passages: reliable, con-
stantly there, but never thematised or examined – the lights that are on that
we don’t notice, the heat dependably emitted from the radiators, the Wi-Fi
we use to get our email, etc. We count on it working without thinking
about it, and yet of course, we know it sometimes also fails: the power goes
out, the heating system runs out of fuel or the Internet goes down. And
everyone knows what happens at such moments: a sharp and sudden
switch of attention as we recognise first the very existence of the systems
that have failed but, second, also both their vulnerability and indispens-
ability, followed by a concern, indeed a demand, that they be repaired as
soon as possible. Now the systems are not hidden or invisible at all: they
jump to the centre of our notice, as we (often impotently) stare at them and
try to figure out what’s wrong. This is clearest of all at moments of gener-
alised failure, as occurs, for example, in large-scale blackouts, where it is
not simply the electric lines in one’s house or under one’s feet that one
comes to notice but rather the massive grid upon which large regions so
systematically depend. Like the Zuhanden in general, one might want to
say, infrastructure is constantly there, taken for granted and reliable, yet at
the same time is invisible to us – and only comes to visibility when it fails.
Heidegger acknowledges as well that ‘invisibility’ is not quite the right
way to describe the way such infrastructure appears or rather fails to ap-
pear: the activity of using and manipulating equipment, he says, is ‘not a
blind one’ but rather ‘it has its own kind of sight, by which our manipula-
tion is guided’; the name he gives this sight is Umsicht, translated (rather
misleadingly) into English as ‘circumspection’ (Heidegger 2008, 98). The
objects with which we interact in our everyday use of them are not strictly
speaking absent, but rather are simply there for us in that use without our
remarking on them. Umsicht is a special sort of ‘sight’, associated with the
 Maintenance and the Humanness of Infrastructure 41

kind of skilful coping or unthematised know-how a trained artisan has

with respect to her craft – the kind of coping we all exhibit in our abilities
to walk, to write with a pen and to use a mouse without having to think at
all about what we are doing.
Heidegger’s purpose isn’t simply to distinguish between two different
sorts of sight: the direct staring-at associated with the present-at-hand and
the circumspection with which we use everyday tools, but also, impor-
tantly, to assert the priority of the latter over the former. That’s the real
point of his discussion of failure or breakdown: the direct visibility of the
present-at-hand only arises when circumspection fails, which is to say that
it always and only appears after and on the basis of prior circumspection.
Indeed, even as we carefully examine the broken tool by focusing our vi-
sion upon it, circumspection remains operative in our use of the other tools
that must be employed for that focusing to be possible – the flashlight we
use to peer into the stubbornly silent car engine, the clothes we wear as we
do so or the garage floor on which we stand. The pure theoretical looking
associated with situations of breakdown depends upon circumspection,
not the other way around – and the objects with which that looking is
concerned are, by definition, broken and therefore deficient objects, useless
objects, which is to say they are not the sorts of objects with which we
normally have to do. Thus, the moment of breakdown does not provide an
opportunity to see the tool as it ‘really is’, shifting it from an unfortunate
invisibility into the shining light of visibility; rather in such a moment, we
lose our grasp of it, failing to understand it as the very tool it is. Indeed,
Heidegger writes,

The less we just stare at the [tool], and the more we seize hold of it and
use it, the more primordial does our relationship to it become, and the
more unveiledly is it encountered as that which it is.
(Heidegger 2008, p. 98 emphasis added)

The broken object that we stare at is ‘veiled’; only in our use of it do we see
it – circumspectively – for what it actually is.
These last considerations suggest that it would be a misunderstanding to
think of situations of breakdown – the blackout, the broken water main,
the Internet outage – as somehow providing special insights into what ur-
ban infrastructure really is.3 Rather, in such moments, we see not what it is
but rather precisely what it isn’t: a complex collection of present-at-hand
objects connected neither to each other nor to us and doing nothing at all.
This Heideggerian insight is an important and valuable one, suggesting as
it does that what’s ‘invisible’ about infrastructure (or rather, what’s ‘visible’
only circumspectively) is not the mere fact that it exists (it’s not ‘hidden’,
i.e., in the way that buried treasure might be, or for that matter a clog in a
42 Steven Vogel

pipe underground that prevents water from getting to a faucet) but rather
what it is – and in particular its status as something that human beings use.
Yet, there’s something that this Heideggerian discussion misses, something
hidden I would suggest from that account as well.
There’s a passage in Marx’s Capital that’s striking both for its similarity
to the Heideggerian account of breakdown and for its important difference
from that account. Marx notes in it that the objects built by workers in a
factory are never built ex nihilo but rather always out of other objects
(Marx 1977, p. 289). In order to produce new things, labourers must work
both on other things (the materials they transform with their labour) and
with other things (the tools they use to effect those transformations). Those
materials and tools are themselves the product of earlier labour (this is so
even of what we call ‘raw material’, Marx points out: coal must be mined
before it is burned to provide energy, cotton or corn must be grown before
being turned into fabric or ground into cornmeal and even the horses and
cattle employed in farm labour are themselves the result of prior processes
of domestication). The labourer’s activity, that is, depends upon the activ-
ity of previous labourers. But in the passage I’m pointing to, Marx adds
that this dependence on previous labour is ‘irrelevant’ to the labour pro-
cess itself, just as the baker’s labour in producing the bread we eat is irrel-
evant to the process of digestion. It only becomes relevant, he writes, when
something about the labour process goes awry: it is only

by their imperfections that the means of production in any process bring

to our attention their character of being the products of past labor.
A knife which fails to cut, a piece of thread which keeps on snapping,
forcibly remind us of Mr. A, the cutler, of Mr. B, the spinner. In a suc-
cessful product the role played by past labor in mediating its useful
properties has been extinguished.
(Marx 1977, p. 289)

The resemblance of this last idea to the Heideggerian one about break-
down shattering the ‘invisibility’ of the Zuhanden is clear. But what’s
revealed in the moment of breakdown, according to Marx, isn’t the
vorhandene object at which we simply stare, but rather the worker whose
prior acts of production show themselves as having been incompetent.
I want to focus on this idea – that what’s ‘hidden’ about our tools is not so
much their status as ready-to-hand as the fact that they are built by human
labour. I’d suggest that this is true not simply when the tools are them-
selves being used in the sorts of commodity-building labour processes
Marx is concerned with in Capital, but rather (nowadays) in all our use of
tools, which is to say – and here Heidegger is perhaps more helpful than
Marx – in all our activities in general. As I raise the hammer or wear the
 Maintenance and the Humanness of Infrastructure 43

clothes or turn on the faucet, I am depending not simply on the reliability

of the tool I employ but at what is really a deeper level on those other hu-
mans whose labour made it possible for that tool – and therefore that
reliability – to come to be. In the city, in particular, it’s not simply the spe-
cific tool I might be using but rather an entire system of infrastructure
upon which I am depending: the wires, the pipes, the roadways, the build-
ings and the sidewalks that form the taken-for-granted environment within
which my life takes place and which indeed makes it possible: all of these
are built by human labour.
Yet, there’s something wrong with Marx’s account too. The idea that a
knife that fails to cut reminds us of ‘Mr. A, the cutler’ seems to refer back
anachronistically to a world of individual craft production, where indi-
vidual cutlers (with various degrees of skill) sell their wares to neighbour-
hood buyers. Nowadays, the labour process that produces a knife is almost
unimaginably more complex, and the labourers involved in that process
are too multifarious to even start to consider: those in the factory who
produced the blade, others who built the handle, others who assembled the
final knife, others who packaged it (and those who made the packaging)
not to speak of those who transported it over what are likely vast distances
to the store at which it’s purchased or the warehouse from which it is de-
livered to the user’s home. We’re not reminded of all these people, under
current conditions of production and distribution, and so under these con-
ditions, Heidegger is right, although for reasons that might be more his-
torically contingent than ontological: when it fails we think instead of the
thing in front of us, the useless knife, that frustrates us and leaves us an-
grily looking about for another one (or ordering it from Amazon). We
blame, if one could put it this way, the thing itself and not the human
process by which it was (incompetently) produced.4
The point I’m trying to make about Heidegger and Marx, it will be clear,
is a complicated one. Heidegger is right, I want to say, to criticise the tra-
ditional view whereby the world is made of present-at-hand objects (let’s
call them Things) that exist as static objects independent of us at which we
stare, but he’s wrong – or rather he does not go far enough – when he says
that what we mostly have to do with in our everyday experience (but which
are hidden in that experience or seen only circumspectively) are ready-to-
hand ‘tools’ because he fails to notice that those tools themselves are al-
most always built ones. What’s hidden isn’t their equipmentality but their
builtness – or rather, the fact that their equipmentality is made possible by
their builtness. To put it another way, he fails to notice the role of concrete
physical activity – in Marx’s term, labour – in the everyday experience of
Dasein that he describes in Sections 15 and 16 of Being and Time. Tools
aren’t just things we ‘encounter’ (begegnen) but rather things we use, and
that use itself is a physical action. Furthermore, and equally importantly,
44 Steven Vogel

the tool itself is the product of prior physical action, which action in turn
made use of prior tools. What we fail to notice in our use of tools – or what
we notice only circumspectively – is the role of human labour in their com-
ing to be. It is this replacement of concern for and recognition of human
labour that builds the world around us with a view of that world as con-
sisting merely of Things, I am trying to suggest, that explains what is hid-
den in infrastructure.

II
Infrastructure, I’m arguing, is best thought of not as a Thing, not some
present-at-hand object or group of objects or system of objects that exist
somehow independently of human activity. Rather it is something that we
use and that we have built. Not a Thing nor the motion of a Thing nor a
tool-Thing that humans manipulate to achieve various ends, rather it is
itself a human doing, a form of human practice. And if this is so, then its
hiddenness will not be the hiddenness of a Thing either. A Thing is hidden
when it is occluded by another one and is thereby prevented from being
(directly) seen. But what’s hidden about infrastructure is exactly its non-
Thingness, its status as a human doing – as something humans use and
have built. (In this, literal, sense – and only this one – we could say that
infrastructure is a ‘social construct’.) To overcome this hiddenness would
be to understand that status, to see infrastructure in its humanness and its
builtness and to stop seeing it as an environment that, although in principle
man-made, nonetheless appears as something separate from and indepen-
dent of us and our use of it.
If what’s hidden in infrastructure is its humanness, its role as a human
doing, it follows that it need not be ‘invisible’ in the ordinary sense – need
not be, for example, literally underground. The sort of hiddenness I mean
can be operative, for instance, in elements of infrastructure that seem quite
visible, even monumental. The George Washington Bridge that spans the
Hudson River and connects Manhattan with New Jersey is in no ordinary
sense of the word something hidden: rather it is a visible marvel, a piece of
sublime architecture that serves as an exemplar of human imagination and
ingenuity and the human ability to use reason to increase human mobility
in new and extraordinary ways. But whether we stare at it (as present-at-
hand) in admiration from afar or simply drive over it (as ready-to-hand)
during our regular commute, we rarely if ever notice or consider those who
actually built it – those thousands of people without whose hard and dan-
gerous work it would never have come to be. We know the bridge to be
built, of course – we don’t think of it as a piece of ‘nature’ – but neither its
visibility (when we stare at it) nor its reliability (when we drive across it) in
 Maintenance and the Humanness of Infrastructure 45

any way reminds us of its builtness. It is in this kind of fact, I think, that
the real hiddenness of infrastructure consists (Figure 2.1).
When it opened in 1931, the George Washington Bridge was the longest
bridge in the world. Construction had taken four years and had been as-
tonishingly complex.5 Foundations had to be built into the Hudson River
on each side, which required (on the New Jersey side) two steel cofferdams
to be constructed and then positioned underwater. When one of the cof-
ferdams failed, drowning several workers, a third had to be built to replace
it. A total of 32 tower footings were bolted to the foundations and con-
crete columns more than 600 feet high were floated down the river on
barges and set on top of each footing. Anchorages were built on land to
anchor the ends of the suspending cables; the cables themselves had to be
manufactured, strung over the river in a difficult and dangerous set of op-
erations and then fed over the towers and attached to gigantic eyebars in
the anchorages. Suspender cables were then hung down vertically from the
main cables. Only after all this was the roadway itself added to the bridge.
It had been built offsite, and like the foundation columns, had to be brought
downriver on barges to a position just under the bridge, at which point
cranes were used to lift it into place (Rockland 2008, pp. 53–62).
I have been able to find no record of how many people worked on the
bridge to build it, although we do know that at least 12 people died during

Figure 2.1 The George Washington Bridge.

46 Steven Vogel

the process. Still one might simply try to imagine how many people must
have been involved: those needed to construct the cofferdams, to build the
foundations, to fabricate the tower columns, to place them on the barges,
to pilot those barges, to construct the anchorages, to manufacture the ca-
bles, to build and then operate the machines that strung them, to check
their placement, to attach the suspender cables, to assemble the roadway –
and, of course, one could go on, thinking about the painters, the electri-
cians, the security guards, the safety inspectors, plus those who built the
barges, built the eyebars, built the hardhats the workers wore to protect
themselves, cooked the meals they ate at the end of the day and sewed and
repaired the clothing they wore. Then there were the lawyers, the accoun-
tants, the architects and furthermore those who cleaned their offices, who
typed their contracts, who filed their documents, who answered their tele-
phones, and so forth. The numbers must surely have run into the hundreds
of thousands. But although the labour of those myriads of workers is in the
bridge – one might even say in a way it is the bridge – it remains, and they
remain, hidden, as we look at it or drive across it: unseen, unthought about
and invisible (Figure 2.2).
Of course that the George Washington Bridge was built, and was built
by humans, is in one sense obvious, as I have already said – no one would
deny it. Yet, in our everyday experience of it, we rarely give any thought at
all to the actual people who built it and the actual labour in which they
engaged to do so. Those whose work caused the bridge to exist remain in

Figure 2.2 George Washington Bridge Construction Workers.

 Maintenance and the Humanness of Infrastructure 47

this sense hidden to us. In addition, there is an important temporal charac-

ter to this hiddenness: when we experience the bridge (again, either the
present-at-hand bridge at which we gaze or the ready-to-hand one upon
which we drive), we experience it as a contemporary fact, something in
front of us now, while on the rare occasions when we consider its builtness
the labour involved appears, if it appears at all, as a historical fact, memo-
rialised occasionally on anniversaries or with plaques, mentioned briefly
on the bridge website and written about in books by those with a profes-
sional or an enthusiast’s interest in the bridge’s story (but the plaques, actu-
ally, are few; the website is more focused on current traffic conditions than
on bridge history and the books seem much more interested in the bridge’s
physical characteristics – the massiveness of its columns, the strength and
number of its cables – than in those who actually did the work). Yet, to
treat the building of the bridge as something that happened, something
that took place once in the past but is now finished, is still to treat the
bridge itself as a Thing – not a Thing of nature, to be sure (because it was
made by humans), but an artificial Thing and so a Thing nonetheless. To
treat it this way, I want to suggest, is still to misunderstand what building
is and what the role of human beings and human labour is in producing
and making possible (‘socially constructing’) the world in which we live.
Infrastructure is not built once and then left alone. Its character as a hu-
man doing is not a fact simply of its past but also, crucially, of its present.6
The labour involved in constructing the George Washington Bridge is not
simply something that took place in a heroic past of bridge-building, but
rather is ongoing. Just like every piece of urban infrastructure, whether
aboveground or below, the bridge requires constant maintenance. Dozens
if not hundreds of people work on the bridge daily.7 A team of painters is
always at work somewhere on the span, engaging in work that is as impor-
tant as it is dangerous: rust and corrosion are a constant threat and paint-
ing helps to prevent them (Posse & DiLella 2018).8 Structural maintenance
workers are constantly inspecting elements of the bridge for signs of fail-
ure, using various specialised trucks and other machinery (Ascher 2005,
p. 50). Rivets need to be replaced regularly, concrete needs to be repaired,
expansion joints need to be emptied of debris and lane markers need to be
repainted (Weisman 2007, p. 35). Linseed oil must be periodically injected
into the cables to keep them pliable and resist rust (Rockland 2008, p. 22).
In what follows, I want to focus on ongoing maintenance as a key and
often overlooked element in infrastructure.9 Ongoing maintenance and re-
pair are a crucial part of all urban infrastructure; without such activities,
the infrastructure would quickly collapse.10 The water mains underneath
New York must be regularly inspected for leaks, clogs, faulty valves and
bad gaskets and when found each must be repaired or replaced. New
York’s subway system, which is actually located beneath the water table,
48 Steven Vogel

is constantly threatened with flooding (as happened catastrophically dur-

ing Superstorm Sandy in 2012). More than 700 pumps operate night and
day to keep it dry, removing 13 million gallons of water a day (Weisman
2007, p. 24), while hundreds of workers are constantly inspecting for leaks
and plugging those that are found, with more than 150 more employed
simply unclogging drains in the system – and, of course, the pumps them-
selves require regular maintenance, oiling and repair. A ‘geometry car’
travels daily around the tracks employing complex instrumentation and
numerous analysts to look for dangerous changes on the tracks (Estes
2014). Weeds such as the ailanthus tree that sprout in subway tunnel walls
and ceilings have to be noticed and removed before they grow so large as
to threaten those tunnels’ structural integrity (Weisman 2007, p. 26). The
city’s underground electrical infrastructure requires constant inspection
and frequent low-level repairs; since both inspection and repair typically
mean some segment of the network must be taken offline, the problem of
scheduling such work and using redundancies in the system to avoid local
loss of service turns out to be both an important and computationally quite
difficult one. Under New York is also a complex system of tunnels for the
distribution of steam, used for heat in many buildings but also for other
services as well, and it too must be monitored on a 24-hour basis; valves in
the steam tunnels that filter out condensed water (which can cause explo-
sions when it comes into contact with the hot steam) have to be inspected
regularly and replaced annually (Brown 2018).11 The manholes workers
use to perform these tasks themselves have to be kept clear of debris and
water (Ascher 2005, p. 186). The need for constant monitoring and main-
tenance of the various nodes of the Internet is, of course, well known – and
not always very well carried through.
It’s important to resist thinking of ‘building’ (in the past) and ‘mainte-
nance’ (in the present) as two distinct sorts of activities. Rather they are
continuous: building takes account of future maintenance and mainte-
nance itself involves building. The George Washington Bridge’s design
from the beginning included catwalks whose only purpose was to allow
workers to access various parts of the structure so as to make future in-
spection, maintenance and repair possible. And when such activities take
place today, they cannot help but change and therefore (re)build the bridge:
a massive repair project currently under way, for instance, will involve
among other things replacing all of the suspender cables.12 The bridge was
originally built with the thought that someday a second, lower level would
be added, and was designed to allow such an addition, which finally took
place in the early 1960s: did that addition count as building, or as repair
(of something missing) or as necessary maintenance to prevent the legend-
ary traffic jams the upper level was experiencing? No piece of infrastruc-
ture is ever complete; it’s always ‘under construction’. Design, building,
maintenance and repair are all parts of a single process – a process in
 Maintenance and the Humanness of Infrastructure 49

which human beings work on the world around them to attempt to shape
elements of that world into forms that then make up the environment that
they inhabit and that help to structure their lives. This is the sense in which
I want to call urban infrastructure a (ongoing) human doing, rather than
treating it as a once-built but now independent Thing.
It is a human doing in another sense too – one that again is superficially
obvious but has implications that are rarely acknowledged. For if we say that
infrastructure (and tools in general) comes to be what it is in its use, we need
to notice that ‘use’ is always active, always a physical doing. Even we ‘end-
users’ of infrastructure, the ones who take it for granted, treat it as simply
and reliably there without noting it, fail to recognise its complexity and its
builtness and the ongoing maintenance needed to keep it functioning – even
we have to act, to do: the infrastructure we use does not function entirely on
its own. In order for water to flow from the faucet, I have to turn the handle;
in order to get light in the room, I have to flip the switch; in order to change
the temperature of my home, I have to adjust the thermostat. These actions
are small ones – and, of course, a significant part of the benefit infrastructure
brings is the way it gears down, so to speak, the amount of work the user
must perform to achieve a task, turning the handle on a sink being much less
difficult than going down to a stream to fill up a jug and then bringing it back
to one’s home – but they are actions, doings, all the same. We users engage in
maintenance and repair as well: recharging our phones, washing clothes, re-
placing lightbulbs and unclogging drains.
Thus building, maintenance, repair and (end-) use, although analytically
distinguishable, blur into each other in practice. The crucial point is that
every step in the functioning of infrastructure involves human activity: this
is what I mean by calling it a ‘human doing’. I’ve emphasised the impor-
tance of recognising the builtness of infrastructure, but because building
turns out to be continuous with maintenance and repair and even with use
we can say that building actually never ends, but is rather ongoing. Indeed,
‘building’ might just be a word for human activity in the world – I am
tempted to say here simply that building is what we do. Humans are active
(‘doing’) creatures, who transform the world around us at every moment
whether this be through constructing new items in that world, working to
change the items that exist or using those items to make or change other
ones. So the things we build – our infrastructure or one might say our en-
vironment in general – are always being built, always under construction.

III
Maintenance as an ongoing activity (and one often carried out by someone
other than the end-user) fits only uneasily into the Heideggerian schema of
readiness-to-hand and presence-at-hand. This is true first of all because as
I drive across the bridge its readiness-to-hand, its ‘withdrawal’ from my
50 Steven Vogel

attention, turns out to depend upon the bridge maintainers’ direct atten-
tion to it as the present-at-hand object of their work.13 They have to work
on it in order for it to be able to withdraw from me. But, second, even that
latter presence-at-hand is only imprecisely described as the result of failure
or breakdown because of course it’s exactly in order to prevent breakdown
that the work of maintenance takes place. Even what could be called
repair – the removal of debris from the expansion joints on the bridge
roadway, say – may be so common as to be planned for and scheduled,
suggesting that it occurs not in response to failure but rather as something
expected, and so merely an element of regular maintenance. And of course,
if the failure is expected it starts to be unclear in what sense it can be said
to be ‘failure’ at all. One might ask whether the rusting of the steel, against
which so much of the activities of bridge maintenance are directed, is cor-
rectly described as a failure of the steel – or is it simply part of what steel
is, part of what it is for steel to function as a tool? (What would steel that
did not rust be like? It surely wouldn’t be steel.)
In truth, the necessity of ongoing maintenance suggests that ‘failure’ is
not an occasional event interrupting the otherwise successful functioning
of the equipmental totality within which our day-to-day life takes place,
but rather is in a certain way constant, indeed is part of what it is to be
equipment. It is a fact of the world – one that engineers and architects are
fully aware of but that often escapes the notice of philosophers – that
things break, and are always breaking. Pipes leak, walls crack, fasteners
crumble, adhesive bonds weaken and joints split apart. Wires overheat.
Weeds grow. Electrical circuits short out. Software faces situations its de-
signers never anticipated. These are not unfortunate and surprising acci-
dents but are rather intrinsic characteristics of the real world (arguably
consequences of the second law of thermodynamics). To build something
in the world is to create something that from the very start is already col-
lapsing, crumbling, falling apart, succumbing to entropy – already, that is,
‘failing’. Again, that’s part of what building is: it’s not something that
building ‘confronts’ or is ‘constrained’ by. The necessity for constant main-
tenance of the things that we build follows directly from this fact and
means as I have suggested that building and maintaining are not distinct
activities.14
When we think of a piece of technology as a ‘tool’ that we ‘use’, we tend
to imagine it as an independent item in the world that the user manipulates
to achieve some particular goals and one that in the ordinary case does so
smoothly, flawlessly and invisibly. But what I have been arguing is that this
apparent smoothness – the ‘withdrawal’ of the ready-to-hand – is itself a
difficult (human) achievement and one the production of which requires an
enormous amount of work: both that needed to construct the tool in the
first place but also crucially the ongoing work of maintenance required to
 Maintenance and the Humanness of Infrastructure 51

keep it functioning normally. I drive over the bridge without incident; but
that drive would not be possible were it not for the work of those who
paint the bridge, inspect and repair the cables, inject the linseed oil, etc.
Indeed, as already mentioned, my own relationship with the tool is not free
of work either, and so not entirely smooth: as I cross the bridge, I have to
engage in an effort too, even if it’s merely a matter of pressing my foot
down on the accelerator or maintaining my grip on the steering wheel.
These too are forms of maintenance, responding to the tendency of the
automobile on its own to ‘fail’ by moving in directions other than I intend
or by succumbing to friction by stopping entirely.15
But at this point ‘failure’ no longer seems like the correct word. To build
an object in the world, whether it be a bridge or an automobile or a sewage
system, is to produce something subject to the laws of physics, including
those having to do with friction, momentum, entropy and so on. That such
objects obey those laws and thus are inevitably and from the very begin-
ning always breaking, crumbling and collapsing is simply a fact about
what it is to be an object in the world: why would one call this a ‘failure’?
(What would it be to be a ‘success’? Not to be subject to the laws of phys-
ics? But that would mean: not to be real!). It is a fact of building that the
objects we build never fully do what we expect them to – they never look
nor function nor age the way we intended; they always have surprising
qualities and always produce results we did not anticipate. ‘Failure’ in this
sense does not occur occasionally or only when our building somehow
goes awry: these rather are simply unavoidable facts about what it is to
build and about what it is to be a real object in the world.16
That the objects we deal with are real means that they never fully do
what we expect, but it also means that our dealings with them always in-
volve activity, ‘doing’, work on our part. The world where effort is unnec-
essary, where what we build is exactly what we want and where failure
never occurs and so maintenance is never needed is the world of fantasy; in
the world of reality, our doings are always difficult, inevitably incomplete
and never-ending. Building cannot help but require maintenance and re-
pair; maintenance and repair are themselves kinds of building; and even
‘use’ requires physical action, which is to say effort, labour.
To recognise the builtness of everything around us in the urban world –
where building is understood to involve maintenance and use as well – is
to recognise, perhaps paradoxically, that everything around us escapes our
control and that our labour does too. To say we build the world around us
is not to say that we control or ‘dominate’ it, although surely we try to do
the best we can to produce the results we wish our actions to have. That
lack of control is evidenced, of course, when tsunamis destroy nuclear
power plants or blackouts bring down large cities or regions, but it is evi-
denced too by the fact that we cannot get water from our faucets without
52 Steven Vogel

turning the knobs and cannot magically produce light from our lamps just
by imagining it.17 We have to act in the world to do anything, and to act in
the world is immediately, and constantly, to be subject to forces beyond
our control.
One might be inclined to find in these last considerations the justification
for the idea that there is in the world something beyond or beneath or be-
fore human action whose ‘otherness’ from that action (and from us) we
need to recognise and before which a certain humility is required: some-
thing that produces the difficulty of work, the inevitability of collapse and
the constancy of ‘failure’ in our buildings. I have been arguing that what’s
hidden in infrastructure – and indeed what is hidden in our relationship
with our environment in general – is its humanness, which is to say its
character as the product of human labour. But perhaps something else
needs to be acknowledged as well: the moment of resistance that our la-
bour faces, the entropic and unexpected elements that I’m insisting arise in
all our buildings and that explain the need for constant maintenance.
Perhaps, it could be argued, that moment deserves a name like ‘nature’,
meaning something independent of the human, a fundamental otherness to
the world.
Yet, this would be mistaken, I think, and would evince a misunderstand-
ing of what labour is. For the concept of labour contains within itself this
moment of ‘resistance’: the latter is not something that labour faces or re-
sponds to but rather it is part of what labour is – which means in turn that
that moment is not correctly described as a moment of ‘otherness’ at all nor
as something beyond the human. To work is to do something in the (real)
world, as opposed to thinking about it or planning it or desiring it. Those
latter things might be said to ‘face’ resistance or otherness, but work does
not. To ask about the ‘source’ of the difficulty humans encounter when
they interact with their environment, and then to identify that source with
nature or with otherness, is to view humans as at first somehow distinct
from that environment and then only subsequently ‘encountering’ it – but
to do this is once again to divide humans from the rest of the world and to
imagine them as (at first) outside it. Such a view subtly betrays the sort of
anthropocentrism that it thinks of itself as leaving behind. Rather we are
in the world, always already, and so our labour in the world does not ‘en-
counter’ resistance or difficulty but rather simply is difficult. The infra-
structure we build ‘fails’, and so requires ongoing maintenance, not because
a nature separate from us sadly always thwarts us, but because our build-
ing of infrastructure – our labour – is real and takes place within the real
world. That there is no confrontation with ‘otherness’ here becomes clear
if one starts to think about what labour would be without this moment of
so-called resistance – how could there be labour without effort, without
doing something in the world? When I turn the knob on the faucet to get
 Maintenance and the Humanness of Infrastructure 53

the water, just as when I place the jug in the stream for the same purpose,
I don’t face anything ‘other’: doing things like that is all I do in the world –
it’s the most familiar, non-other, sort of thing there is. And so the claim that
infrastructure is a human doing, precisely because the ‘doing’ here is meant
as real doing, as labour, does not require supplementation by the notion of
a ‘nature’ independent of or standing below or prior to the humanly built
world. Dualism needs to be rejected all the way down, including the dual-
ism between human activity and an ‘otherness’ such activity is said to
‘encounter’.

IV
I want to conclude with some brief remarks about those who engage most
directly with the infrastructural labour I have been discussing: the workers
themselves, that is, who are constantly building, rebuilding, maintaining
and repairing infrastructure. Among the benefits infrastructure provides,
of course, is a reduction in the effort involved in various daily tasks – I’ve
used the example above of getting water from a faucet rather than from a
stream. But if one counts the labour involved in constructing the water
system itself – the aqueducts, the pipes, the pumps, etc. – and adds to it the
maintenance and repair work constantly required to keep all the elements
of that system running as well, one realises that part of the labour saved
has simply been shifted elsewhere, away from the end-user to others. The
building of infrastructure does not only add efficiency or save labour with
respect to achieving various human goals but it also serves as a means of
organising and distributing the labour needed to achieve those goals –
which is why the question of what infrastructure to build (and how to
maintain it) is a political one. Surely, furthermore, the hiddenness of the
human labour involved in the building, maintenance and repair of infra-
structure that I have been discussing is not unconnected to the relative in-
visibility in our society of those who actually engage in that labour – labour
that is generally poorly paid, not well respected, associated with marginal
populations, etc.18 The hiddenness of infrastructure and the hiddenness of
those workers are two sides of the same phenomenon.
But there is no necessity to either side, no intrinsic reason why infra-
structure’s builtness or why those whose labour in fact builds it must inevi-
tably be hidden. Rather it follows from certain facts – political ones – about
the social arrangements under which we live. One could also imagine alter-
native systems where those whose ongoing work makes it possible for the
rest of us to enjoy heat, water and electricity and travel over bridges with
relatively low expenditures of labour would be highly honoured by society,
or one where that work was shared among all capable citizens. Such sys-
tems would be ones where infrastructure would not be so hidden (in the
54 Steven Vogel

sense I have been discussing), because the fact that the building and main-
tenance of our environment is always a social effort, requiring the work of
many different members of the community, would be a matter of everyday
awareness and acknowledgement. The system we actually live under, of
course, is not like this, but rather one where the built environment we in-
habit appears as the result of a series of individual projects, each of which
is engaged in for private monetary gain: there is little occasion under that
system for a sense of social solidarity to develop that might make it possi-
ble for our deep reliance on each other’s labour to become something we
daily notice and appreciate, and that might therefore also make possible an
overcoming of what I have been describing as infrastructure’s hiddenness.
Still, there is a sense in which the very workers I have just mentioned –
the ones who build the infrastructure we depend upon and whose ongoing
work to maintain and repair that infrastructure is crucial to its continued
functioning – are in what might be called a privileged position with respect
to the possibility of such an overcoming of infrastructure’s hiddenness.
Whereas for the rest of us infrastructure does function as something simply
there, simply reliable and barely noticed – the taken-for-granted back-
ground condition that makes possible the ordinary activities of our own
daily lives (which is to say, of our own labour) – for those who actually
work directly on that infrastructure itself, that simplicity can likely never
seem so simple. For them, the fact that infrastructure is humanly built and
humanly maintained, and furthermore is constantly subject to ‘failure’ and
in danger of full-scale breakdown, is quite obvious: infrastructure is not
hidden from them. Nor too can they fail to notice that their fellow citizens,
as they drive across bridges or turn on lights or enjoy the heat in their
homes, depend upon the labour of others in order for their own lives and
work within the city to be possible. One might therefore find in those infra-
structural workers themselves – poorly paid and marginalised as they often
are, in many cases literally hidden underground in their work, and rarely
noticed by those whose lives they so significantly support – a group with a
kind of special knowledge of the real character of the world we all inhabit,
a knowledge to which we might be well advised to pay attention.19

Notes
1 And this is so even for that important element of urban infrastructure that does
require citizens to venture – eyes open – underground, the subway (sometimes
named the Underground itself), seen and heard and smelled for sure and yet
somehow rarely itself the object or goal of their activities. They are focused,
one might say, on what will happen when they reach their station and emerge
above the Earth, but hardly ever on the fact that right now they are travelling
under that ‘ground’ on top of which their ordinary lives take place and which
they normally take to be solid.
 Maintenance and the Humanness of Infrastructure 55

2 See, for example, Star (1999). For this citation and an introduction to some of
the issues connected with the maintenance of infrastructure, I’m indebted to
Mark Thomas Young and his paper ‘Now You See It (Now You Don’t): Users,
Maintainers and the Invisibility of Infrastructure’ (Young 2021).
3 There is a tendency to do this, for instance, in various studies of infrastructure
in the philosophy of technology and urban studies. Young makes this point in
Young (2021, pp. 103–5). Thus, Stephen Graham writes in the introduction to
his edited volume Disrupted Cities: When Infrastructure Fails that ‘studying
moments when infrastructures cease to work as they normally do is perhaps the
most powerful way of really penetrating and problematising those very nor-
malities of flow and circulation to an extent where they can be subjected to
critical scrutiny’ (Graham 2010, p. 3). The 2003 blackout in the Eastern United
States and Canada is often used as an example – see, for instance, Timothy
Luke’s essay ‘Power Loss or Blackout’ in the same volume (Luke 2010) or Jane
Bennett’s Vibrant Matter (Bennett 2010, esp. chap. 2).
4 There is a name for this process, in the history of Marxist thought: the fetishism
of commodities. Georg Lukács called it reification (Verdinglichung), the pro-
cess whereby human labour appears in the form of apparently independent
Things. See his History and Class Consciousness (Lukács 1971).
5 See ‘George Washington Bridge Across the Hudson River at New York, NY’
reprinted from the Transactions of the American Society of Civil Engineers
vol 97 (New York: The Port of New York Authority, 1933), as well as Michael
Aaron Rockland, The George Washington Bridge: Poetry in Steel (New
Brunswick, NJ: Rivergate Books, 2008).
6 In The Poverty of Philosophy, Marx writes that for bourgeois economics ‘there
has been history but there is no longer any’ (Marx 1973, p. 105).
7 I have been unable to find anything like an exact number which in itself indi-
cates how little awareness or interest this sort of labour excites.
8 It is however an urban myth – which is also apparently told about the Golden
Gate Bridge and the bridge over the Firth of Forth in Britain – that painting
begins at one end of the bridge and then when completed at the other simply
starts all over again.
9 See Young (2021, pp. 109–110). Stephen Graham and Nigel Thrift in their es-
say ‘Out of Order: Understanding Repair and Maintenance’ (2007, p. 10)
speak of ‘the remarkable neglect’ in contemporary discussion of urban infra-
structure ‘of the massive and continuous work that is necessary to sustain’ it.
The current volume is a welcome attempt to overcome this neglect.
10 Alan Weisman’s strange and moving book The World Without Us (Weisman
2007) describes in startling detail what would happen if human beings sud-
denly one day simply disappeared, which means among other things the end of
that ongoing maintenance: the results would not be pretty.
11 In the middle of the last century, a series of billboards in Times Square advertis-
ing cigarettes used the steam from this system to provide realistic-looking puffs
from the cigarettes of larger-than-life smokers.
12 See ‘Restoring the George: 10 Year Capital Plan Construction Program’, Port
Authority NY NJ. Available at: https://www.panynj.gov/bridges-tunnels/restoring-
the-george.html (accessed June 15, 2022).
13 And note that Heidegger actually never points out that work always does have
such an object – because Heidegger, unlike Marx, never really focuses on work
itself. Despite all that hammering in Being and Time, it’s never really made
clear what is being built.
56 Steven Vogel

14 See, along these lines, Steven J. Jackson’s essay ‘Rethinking Repair’ (Jackson
2014) which argues for what he calls ‘broken world thinking’, a thinking that
he describes as what happens ‘when we take erosion, breakdown, and decay …
as our starting points in thinking through the nature, use, and effects’ of things
like infrastructure (Jackson 2014, p. 221). He suggests, as I would too, that
such an approach might lead us to ‘a deep wonder and appreciation for the
ongoing activities by which stability … is maintained’, leading to a kind of
‘admiration and even reverence [for] the sheer magnitude of the work repre-
sented in the ongoing maintenance and reproduction of the established order’
(Jackson 2014, p. 222). See also Graham & Thrift (2007, pp. 5–6).
15 And this is true of ‘self-driving’ cars as well, which continue to require human
drivers prepared to take over when (inevitably) ‘failure’ threatens.
16 This holds for so-called ‘digital’ objects too, which are often mistakenly dis-
cussed as if they were non-physical. ‘Software’ is only relatively soft: it takes
the form of a physical pattern of transistors and resistors conducting currents
in various complex ways, and it only functions as it does (and ‘fails’ when it
does) because of something about that pattern.
17 Yes, ‘smart’ homes are making it possible to turn lights on remotely, or to pro-
gram them to turn on ‘by themselves’: but still work has to be done to make
those things happen. Nothing happens in the world that doesn’t involve – well,
something happening in the world.
18 See Young (2021, p. 12) and Graham and Thrift (2007, p. 18).
19 Jackson, for instance, talks of the idea of something like a ‘standpoint episte-
mology of repair’. See Jackson (2014, p. 10).

References
Ascher, Kate (2005) The Works: Anatomy of a City. New York: Penguin Press.
Bennett, Jane. (2010) Vibrant Matter: A Political Ecology of Things. Durham:
Duke University Press.
Brown, Nicole. (2018) ‘NYC steam system: Why we have it and how Con Ed main-
tains it’ AMNY. Available at: https://www.amny.com/news/nyc-steam-system-
1.20146953/ (accessed on April 27, 2022).
Estes, Adam Clark. (2014) ‘This Superheroic Train Keeps New York City’s Subway
Safe’ Gizmodo. Available at: https://gizmodo.com/this-superheroic-train-keeps-
new-york-citys-subway-safe-1571987376 (accessed June 20, 2022).
Graham, Stephen. (2010) ‘When Infrastructures Fail’ in Graham, S (ed) Disrupted
Cities: When Infrastructure Fails. New York: Routledge. pp. 1–26.
Graham, Stephen & Thrift, Nigel. (2007) ‘Out of Order: Understanding Repair
and Maintenance’, Theory, Culture and Society 24(3) pp. 1–25.
Heidegger, Martin. (2008) Being and Time, translated by J. Macquarrie and
E. Robinson. New York: Harper Perennial.
Jackson, Steven J. (2014) ‘Rethinking Repair’ in Gillespie, T., Boczkowski, P. &
Foot, K. (eds) Media Technologies: Essays on Communication, Materiality and
Society, Cambridge, Mass.: MIT Press. pp. 221–240.
Lukács, Georg. (1971) History and Class Consciousness, translated by Livingstone,
R. Cambridge, Mass.: MIT Press.
 Maintenance and the Humanness of Infrastructure 57

Luke, Timothy W. (2010) ‘Power Loss of Blackout: The Electricity Network

Collapse of August 2003 in North America’ in Graham, S (ed) Disrupted Cities:
When Infrastructure Fails. New York: Routledge. pp. 55–68.
Marx, Karl. (1973) The Poverty of Philosophy. Moscow: Progress Publishers.
Marx, Karl. (1977) Capital Volume 1, translated by Fowkes, B. New York: Vintage
Books.
Posse, Erica & Dilella, Chris. (2018) ‘These workers have one of the scariest jobs
in the world’ CNBC Make it. Available at: https://www.cnbc.com/2018/06/20/
this-crew-of-bridge-painters-has-one-of-the-most-dangerous-jobs-in-the-world.
html (accessed on 28th April 2022).
Rockland, Michael A. (2008) The George Washington Bridge: Poetry in Steel.
New Brunswick: Rutgers University Press.
Star, Susan Leigh. (1999) ‘The Ethnography of Infrastructure’, American Behavioral
Scientist 43(3) November/December pp. 377–391.
Weisman, Alan (2007) The World Without Us. New York: St. Martin’s Press.
Young, Mark Thomas. (2021) ‘Now You See It (Now You Don’t): Users,
Maintainers and the Invisibility of Infrastructure’, in Nagenborg, M., Taylor, S.,
Woge, M., & Vermaas, P. (eds.), Technology and the City: Towards a Philosophy
of Urban Technologies. Cham: Springer Nature Switzerland. pp. 101–119.
3 Technology in Process
Maintenance and the
Metaphysics of Artefacts

Mark Thomas Young

In this chapter, I aim to demonstrate that despite the breadth of existing

work on the metaphysics of persistence, we’re yet to properly understand
the relationship between technology and time. In order to address this
problem, I’ll argue for the need to begin exploring a topic which to date
has attracted little attention from metaphysicians: maintenance. However,
as I’ll aim to show, doing so requires us to flip the script in the metaphysics
of artefacts, by taking leave of the substance ontology which has tradition-
ally provided the framework for western metaphysics.
Before we begin, it is worth clarifying what I mean by substance ontol-
ogy because there are at least two senses in which philosophers use the
term. The first is used to refer to a class of approaches within ontology that
employ philosophical conceptions of substance. According to this defini-
tion, substance ontology was bequeathed to the western philosophical tra-
dition by Aristotle and has remained dominant to the present day where it
currently exists in different variations.1 The second sense in which philoso-
phers use the term substance ontology is broader and applies less to spe-
cific theoretical commitments than it does to a set of implicit presuppositions
guiding ontological analysis. According to this second sense, substance
ontology is more properly understood as a research paradigm, one which
includes theories that not only predate Aristotle but which often don’t in-
volve philosophical conceptions of substance at all (Seibt 1997; Seibt
2004). It is this latter conception of substance ontology in particular, which
is often contrasted with rival ontologies which understand change and pro-
cess to be fundamental. Central among the assumptions which unite the
different theories that are considered to fall under the research paradigm of
substance ontology is the idea that at the most fundamental level, reality
consists of static, enduring entities (Rescher 1996 p. 29; Dupré 2018).
Increasingly, philosophers have been drawing attention to the way in
which the paradigm of substance philosophy forms an implicit framework
that lurks beneath the apparent diversity of philosophical thinking in the
western tradition. According to Johanna Seibt, for example, because the

DOI: 10.4324/9781003316213-4
 Technology in Process 59

principles of substance ontology are located within the ‘presuppositional

depth structure’ of debates in metaphysics,

they have hardly received any attention so far. Even those contemporary
ontologists who drop the traditional talk about substances still import
the presuppositions of the traditional paradigm; even in ontologies
whose basic entities are events, tropes, or state of affairs, the presup-
positions of substance ontology are still operative.
(Seibt 1997)

As Anne Sophie Meincke points out, theories which are often viewed as
standing in opposition to one another, such as Democritian Atomism and
Aristotle’s account of substance, for example, or even the more contempo-
rary metaphysical debate between Perdurantism or Endurantism, never-
theless all share a commitment to the central insight of the substance
philosophy paradigm – that reality is fundamentally composed of static
enduring entities (Meincke 2019, 2021). For philosophers concerned with
uncovering the research paradigm of substance ontology, the goal is very
often to show how, by forming an implicit framework within which discus-
sions themselves take place, it has constrained the development of alterna-
tive approaches which understand change or process, rather than static
entities as the basic building blocks of reality.
While a number of recent studies have attempted to illustrate the poten-
tial of such process-based approaches to enable progress in areas as diverse
as the philosophy of biology (Nicholson & Dupré 2018) and philosophy
of psychology (van Geert & de Ruiter 2022), the benefits of adopting a
process-based approach to questions surrounding the passage of artefacts
through time has yet to be elaborated. My goal in the current chapter is to
do precisely that: by outlining how the paradigm of substance philosophy
has constrained philosophical thinking surrounding questions concerning
the relationship between technology and time. As I’ll attempt to show, ar-
tefacts represent a particularly unique area of research for which process
perspectives hold the potential not only to solve existing problems but also
to reveal new avenues for research.
The first section of this chapter provides a brief outline of the various
ways in which the western metaphysical tradition has approached ques-
tions concerning the persistence of artefacts. Here I aim to show that un-
derlying the diversity of approaches in this tradition, are a number of
shared assumptions which align closely with the research paradigm of
substance philosophy. In the second section, I draw upon existing litera-
ture in maintenance and repair studies to illustrate how a wide variety of
artefacts can be understood to undergo continuous change after produc-
tion through the act of maintenance. My goal here is to show that insofar
60 Mark Thomas Young

as the underlying framework of substance ontology faces a variety of limi-

tations in accounting for these forms of change in the metaphysical analysis
of artefacts, the study of maintenance highlights the need for an alterna-
tive ontological framework. As I argue that this alternative can be found
in process philosophy, the final section of the chapter explores the form a
process-based metaphysics of artefacts might take by drawing on resources
central to the recent ‘process turn’ in the philosophy of biology. In con-
trast to existing methods and approaches in metaphysics, I demonstrate
how a process philosophy of artefact maintenance tends inevitably to-
wards the empirical analysis of social and political processes which deter-
mine the trajectory of artefacts through time.

3.1 The Problematic Status of Change in Western Metaphysics

While it is not often recognised as such, maintenance forms the topic of
one of the earliest and most enduring metaphysical puzzles in the western
philosophical tradition: The Ship of Theseus. Consider, for example,
how the original formulation of the problem by Plutarch depicts the re-
placement of materials as occurring specifically within a context of
maintenance.

The ship wherein Theseus and the youth of Athens returned from Crete
had thirty oars, and was preserved by the Athenians down even to the
time of Demetrius Phalereus for they took away the old planks as they
decayed, putting in new and stronger timber in their places, insomuch
that this ship became a standing example among the philosophers, for
the logical question of things that grow; one side holding that the ship
remained the same, and the other contending that it was not the same.
(Plutarch 1859, p. 21 emphasis mine)

Over time, however, the practice of maintenance has come to be excluded

from discussions surrounding the Ship of Theseus. In modern philosophy,
the version of this paradox which attracts most discussion stems from the
work of Thomas Hobbes. In Hobbes’ modified retelling, instead of being
simply discarded, the planks composing the original ship (which we
must assume have not suffered decay), are reassembled into another ship
(Hobbes 1656, p. 100). Hobbes’ version introduces a case of fission,
whereby one artefact appears to have evolved into two: the ship consisting
of replaced but new planks and the ship consisting of the reassembled
original planks, prompting the question of which, if any, should be consid-
ered identical with the original ship. As a result, most philosophers today
will recognise the Ship of Theseus paradox more immediately as a puzzle
concerning identity rather than maintenance. Yet, this downplays the deep
 Technology in Process 61

connections which exist between conceptions of identity and maintenance

which the original formulation of the paradox succeeded in highlighting.
After all, to ask whether the ship remains the same throughout the replace-
ment of materials is at the same time to ask what it means to maintain a
technology. For this reason, conceptions of how artefacts persist through
change always imply particular ways of conceiving maintenance. In the
latter sections of this chapter, we’ll return to the question of the relation-
ship between identity and maintenance, but for now, I want to dig deeper
into questions surrounding the persistence of artefacts.
Regardless of the formulation we consider, the Ship of Theseus paradox
can be understood to represent a version of a perennial metaphysical di-
lemma known as the problem of change. Put simply, the problem is this: if
an object can be said to undergo change, then it must become different
from what it was before. Yet, if it is not the same both before and after
change, on what basis can we say that one object has actually undergone
change? (Wasserman 2006). Among classical philosophers, the problem of
change was often viewed as the central difficulty underlying metaphysical
theories that posited change as the fundamental basis of reality. The prob-
lem stemmed from the concern that the presence of change meant that
things possessed contradictory properties – that the ripening of a banana
meant that it could be both green and unripe and yellow and overripe, for
example. Insofar as this was widely understood among classical philoso-
phers as threatening the possibility of knowledge, language and even
thought itself (Lee 2010), resolving the problem of change has been consid-
ered a task of crucial philosophical importance since antiquity. In fact, this
problem continues to draw the attention of philosophers today. There is
now a wide variety of responses to the problem of change, among which it
is possible to distinguish two main orientations:

Milder approaches seek to resolve the problem of change by maintaining

that not all aspects of an object undergo change. Instead, these ap-
proaches seek to demonstrate how certain features, specifically those
that grant identity to an object, remain the same over time. The most
well-known example of such an approach appears in the work of
Aristotle, who drew a distinction between two different kinds of change
things could undergo: accidental change, such as the painting of a house,
during which the identity of the object is preserved, and essential change,
such as the bulldozing of a house, during which the thing in question
ceases to exist. Coupling the identity of things tightly to their essential
properties, Aristotle sought to segregate those parts of an object that
underwent change from those which determined the identity of the ob-
ject, making it possible to say that an object was changing while still at
the same time remaining the same. Modern variants of this approach
62 Mark Thomas Young

employ the notion of identity criteria which are understood to deter-

mine the kinds of changes a particular object can tolerate or not. For
artefacts, identity criteria have been suggested to stem from a variety of
sources including collective beliefs about kind membership (Bloom
1996; Pearce 2016), sortal concepts (Hilpinen 1993) or even stipula-
tions from the makers of artefacts themselves (Thomasson 2003).
Stronger Approaches seek to provide accounts of identity over time which
satisfy the requirements set out by Leibniz’s law of the indiscernibility of
identicals. According to such approaches, in order to justifiably assert
that an object has undergone change, it must be demonstrated that the
object both before and after the change are identical, understood in
terms of sharing all their properties. Among responses that satisfy this
more stringent requirement, the most successful arguably stem from the
perspective of four dimensionalism which considers artefacts to be ex-
tended not only in space but also through time. Repudiating the com-
mon and intuitive notion that an object wholly exists at any one moment
in time, four-dimensionalism instead considers an artefact to exist as a
series of temporal parts extended throughout time. What this means is
that an object is not wholly present before or after any change, instead
what we find at each moment are simply temporal parts of a larger pro-
cess which is always identical to itself.

The foregoing passages highlighted how existing responses to the problem

of change within the western philosophical tradition exhibit significant
variation. However, what is striking is that despite their variety, these dif-
ferent responses nonetheless share at least two assumptions concerning the
nature of change and identity. Perhaps, not surprisingly, these shared fea-
tures, which following Johanna Seibt can be understood to be located in
the ‘presuppositional depth structure’ of the different theories, point to-
wards a particular kind of ontology – one based around things rather than
processes. The first assumption concerns the ontological priority of sub-
stance over change. In neither of the two main approaches to resolving the
problem of change that we considered above, does change occupy a posi-
tion of ontological priority. While milder approaches recognise the exis-
tence of change, by aligning the identity of objects with those aspects
which remain the same, they also tend to shuffle it into the corner. As Anne
Sophie Meincke notes, this tendency to deny ontological significance to
change represents a fundamental assumption underlying substance ontol-
ogy which

crucially rests on the idea that change – if there is any – must have a
non-changing basis. Whatever appears to change is believed to have an
unchanging core. Change, if it exists at all, is taken to be parasitic upon
 Technology in Process 63

stasis; it does not have an independent ontological status. It resembles

the fluctuating images on a mirror surface, leaving the mirror itself the
same whether it shows microbes on a petri dish, a murder, or passing
clouds.
(Meincke 2019)

According to this perspective, change is understood as something that hap-

pens to things and for this reason, it depends on the existence of stable,
enduring entities. Stemming from the intuition that for change to be pos-
sible there must be some thing that changes, these theories typically under-
stand change as the ‘doings’ of stable entities (Rescher 2000, p. 6) or as
something that happens to them (Dupré & Nicholson 2018, p. 12) and for
this reason, accord it a derivative and secondary ontological status. The
denial of the ontological significance of change emerges even more force-
fully from the stronger approaches considered above. One of the most
common criticisms of four-dimensionalist solutions to the problem of
change, for example, is that they succeed in resolving the problem only by
denying the existence of change altogether.2 In order to explain how the
same object can instantiate contradictory properties, such as the banana
which is green and unripe at a particular time and then yellow and overripe
at a later time, four dimensionalism encourages us to view the green and
yellow bananas as different temporal parts of a larger, timeless whole. By
explaining the development of the banana in terms of a succession of un-
changing temporal parts, this approach eschews the possibility of becom-
ing itself. As Simons notes, four dimensionalism offers ‘not an explanation
of change, but an elimination of it, since nothing survives the change which
has contrary properties’ (Simons 2000).
The second assumption underlying the two approaches concerns the
separation of identity from change. As Johanna Seibt has noted, the ten-
dency to conceive of identity as static is a shared feature of substance-
based approaches to questions of persistence (Seibt 1997). The idea that
only that which does not undergo change can be a candidate for identity
has informed discussions of persistence in western metaphysics since antiq-
uity. Aristotle’s distinction between essential and accidental properties, for
example, posited a division at the heart of objects which demarcated those
aspects which undergo change from those which confer identity. A similar
opposition between identity and change continues to inform modern ap-
proaches within the metaphysics of identity. As Peter Simon has noted, the
ship of Theseus paradox draws out two opposing tendencies within the
sortal concepts we employ to track the identity of objects such as ‘ships’.
On the one hand, there is a tendency to link the identity of an artefact with
the matter composing it. On the other hand, an opposing tendency consid-
ers identity to reside in an artefact’s form. Yet, regardless of the conception
64 Mark Thomas Young

we employ, however, the identity of the ship is nonetheless conceived as

being unable to tolerate even the smallest amount of change.

Form-constant objects cannot survive a disruption of continuity of their

form: even the shortest intervention which interferes with the capacity
to function counts as destroying the artifact. Matter-constant continu-
ants cannot survive the slightest change in their constituent matter. On
the other hand, a matter-constant continuant does not require a con-
tinuously maintained form. It could survive dismantling and reassem-
bly, provided it is put back together without any matter being lost or
gained.
(Simons 1987, p. 200)

In discussions of persistence then, an identity changed is very often an

identity lost. This common tendency to strictly separate change from iden-
tity in Western metaphysics emerges as a necessary consequence of the fact
that for the vast majority of theorists, change is understood to corrupt
identity (Meincke 2019). This is reflected in the nature of the dilemma
which these approaches are intended to address, which understands change
to be the problem that must be solved, in order to rescue the possibility of
a world which can be understood, thought and communicated of. The
underlying strategy western metaphysics has hitherto employed to address
the challenge of explaining persistence therefore consists of demonstrating
how a changing thing has remained the same despite undergoing change.
Despite the diversity of views concerning exactly how objects should be
understood to remain the same, this is one assumption which is rarely sub-
jected to critical scrutiny or even considered a topic for discussion at all. In
the following sections, we will turn our attention to examining the pro-
cesses by which artefacts develop through time, in order to begin exploring
reasons why this apparently self-evident way of approaching questions
surrounding the persistence of artefacts may not be so obvious after all.

3.2 Maintenance and Modification

In relation to technology, maintenance encompasses an extremely wide
variety of different kinds of activities. We replace parts, apply lubricant or
protective coatings, clean away grime or salt, polish, sharpen and shine, all
in the name of maintenance. Insofar, as we perform these activities deliber-
ately in order to achieve specific goals, maintenance can be understood to
be inherently normative and inextricably linked to specific expectations
concerning how particular artefacts should be. The particular activities we
perform under the guise of maintenance therefore reflect our underlying
intuitions that our knives should be sharp and our skillets seasoned, our
 Technology in Process 65

cars free from rust and our bike chains well oiled. Despite the diversity in
the kinds of activities we regard as maintenance, focusing on the goals we
aim to achieve by maintaining allows us to see how, at the broadest level,
maintenance may be characterised as an activity which aims to ensure co-
herence between our understanding of the state of an artefact and a par-
ticular set of normative expectations.
In philosophy, the normative expectations we seek to realise through the
maintenance of artefacts are often identified with the intentions of those
responsible for producing the artefacts in the first place. This view corre-
sponds to a particular conception of how artefacts are made which has
remained dominant in the Western philosophical tradition since antiquity.
According to this view, which I call the ‘intentional conception of making’
(Young 2021), artefacts are produced through the material realisation of
the intentions of one or more human agents. These intentions, which may
include specifications regarding function, form or material composition,
are understood as determining the identity of the artefact that is created,
alongside establishing persistence conditions which determine the kinds of
change the artefact in question can tolerate and those which result in its
destruction.3 An important corollary of this view, and one which again
highlights the pervasiveness of the research paradigm of substance ontol-
ogy, is that the intentions which determine the identity of artefacts are not
themselves considered to undergo change over time.4 For this reason, the
proponents of the intentional conception of making tend to understand the
goals of maintenance to consist of the preservation or restoration of the
original intentions an artefact’s creator sought to realise in the act of pro-
duction. A prominent example of this approach is found in the work of
Simon Evnine, who considers various maintenance activities, such as the
replacement of parts, as a form of restoration:

a restorer of a sculpture may replace its lost nose; the restorer of a car
may replace a failed battery.…The process of replacement mimics the
original making: it involves an agent who supplies the matter to do
what the original matter did, acting with an intention to repair an object
of the kind originally made…Let us call this kind of replacement
restoration.
(Evnine 2016, p. 76)

Insofar as this perspective understands the persistence of artefacts to de-

pend on the continued possession of identity-granting features intended by
their designers, maintenance has an important role to play. By ensuring the
coherence between the state of an artefact and the intentions of its creators
through activities such as the replacement of parts or the protection of cer-
tain features from damage, maintenance can be understood to be crucial in
66 Mark Thomas Young

enabling the persistence of artefacts. At the same time, this account also
determines what operations count as maintenance for any particular arte-
fact because it draws an implicit distinction between acts of maintenance
and modification. Consider again the Ship of Theseus. According to this
view, certain operations performed upon the boat, such as the replacement
of planks, may count as maintenance only to the extent that they succeed
in restoring the original intentions of the maker (assuming of course that
the maker intended to create an artefact which could tolerate the replace-
ment of the material constituting it). Other forms of activity, such as con-
verting the ship to a DUKW by adding wheels and a chassis to the hull,
would represent a form of modification that would count as a new act of
making because the intentions we aim to realise in performing these opera-
tions differs from what we can assume the maker of the ship originally
intended.
Despite forming the standard view on the processes by which artefacts
are made and maintained, this view encounters various difficulties when
we attempt to apply it to actual histories of artefacts in use. As philoso-
phers have long recognised, the normative expectations users hold about
how their artefacts should be do not always match the intentions of those
who produced them. There are two main reasons for this: the first is that
normative expectations differ among users depending on their individual
interests and expertise. Whereas one user may regard a heat pump as per-
forming adequately, for example, another user who is a heating, ventila-
tion, and air conditioning (HVAC) engineer may consider the same heat
pump to require modification in order to match their expectations for op-
timal function (see Figure 3.1). Still other users may seek to modify the
same heat pump in order for it to be operated in an environment for which
the artefact was not optimised during the design process. The second rea-
son concerns the capacity for our expectations regarding artefacts to
change over time. We may be satisfied with our phone, for example, until
we learn that a similarly priced model takes far better pictures or is able to
run much larger programs smoothly. The capacity of normative expecta-
tions for particular artefacts to evolve over time is seen not only in relation
to personal devices but also to artefacts that we own or use collectively.
Consider, for example, safety codes for engineered structures. Many struc-
tures which were originally regarded as safe when they were constructed in
the early twentieth century eventually came to be viewed as unsafe as our
understanding of structural mechanics and the risk of collapse during seis-
mic events increased.5 The evolution of engineering standards provide a
clear indication of the capacity for our normative expectations to change
over time.
Yet while philosophers have long recognised that users come to hold
ideas about how particular artefacts should be that differ from what was
 Technology in Process 67

Figure 3.1 Heat pump modified by the author’s father-in-law to direct airflow and
monitor temperature.

intended by the producers of those artefacts, what has not received ade-
quate attention is the extent to which this diversity leads artefacts them-
selves to change over time. Examining the histories of artefacts in use
reveals how a wide range of different kinds of artefacts undergo change
through the activities of retrofitting, repair and modification. These pro-
cesses of transformation occur for different reasons. Artefacts are often
transformed, for example, when they change hands between owners who
possess different normative expectations. When buying a house, for exam-
ple, we are often interested in how it can be altered in order to better align
with our interests: whether an office can be converted to a kid’s room, for
example, or whether there is enough space to build a patio in order to host
barbeque parties in the summer. In other cases, artefacts transform when
our collective normative expectations evolve over time. This is particularly
apparent for long-lived engineered structures such as bridges, whose histo-
ries often reveal a wide range of changes that were undertaken in order to
adapt the structures to evolving standards for safety. Consider the Golden
Gate Bridge in San Francisco as an example. During the 1950s, the bracing
on the underside of the bridge deck was modified to prevent torsional in-
stability during high winds which had recently led to the collapse of the
Tacoma Narrows Bridge. After the collapse of the Cypress Viaduct during
the Loma Prieta Earthquake in 1989, the Golden Gate Bridge District de-
cided to fund a series of seismic retrofits which have been ongoing for the
last two decades and involved the modification of almost all parts of the
68 Mark Thomas Young

structure, including the approaches, pylons and bridge deck (GGBHTD

n.d.). Further changes made to bring the structure in line with current
safety standards include the installation of pedestrian barriers (Bateson
2012, p. 70) and concrete median barriers to prevent cars from crossing
into the path of oncoming traffic (Josephson 2017, p. 76).
Until only recently, the processes by which artefacts continue to undergo
development after their production had received only scant attention in
scholarship on technology.6 The recent emergence of the interdisciplinary
field of maintenance and repair studies, however, reflects an increasing in-
terest in exploring these processes and there is now a growing body of lit-
erature examining how technologies undergo adaptation after production.
However, exactly how many artefacts undergo processes of adaptation
throughout their histories still remains an open question and one which
can only be addressed through further empirical research. It is nonetheless
clear already that for some artefact kinds, such as houses and boats, trans-
formation represents a normal feature of their histories. For other artefact
kinds, however, transformation depends more on the particular context in
which they will be employed. Consider cars as an example. While a small
minority of car owners in the Global North may be interested in modifying
their vehicles, most cars undergo very little transformation throughout
their histories insofar as maintenance routines typically consist of the re-
placement of parts ordered directly from the manufacturer. Outside this
context, however, transformation is much more common. In many African
countries, for example, the lack of access to cheap genuine replacement
parts means that vehicles are often modified through the activity of repair.
In their biography of a 504 Peugeot that had been imported from the
Netherlands to Ghana in the early 90s, anthropologists Verrips and Meyer
recount the modifications made to the vehicle by its owner Kwaku:

he replaced the two-chamber carburettor of his old Peugeot by a one

chamber one in order to save petrol. The air inlet was not original any
longer, but consisted of a piece of pierced tin. A lot of gaskets and rub-
ber parts, such as bushings were ‘indigenous’, that is, cut out of old
tubes and tyres…at several places nails were used a lock pins. Some
rubber tubes were fixed with iron wire, whereas others which evidently
were out of use were closed with old spark-plugs, butterfly nuts or even
pieces of wood. In order to prevent the knocking of some worn-out
main bearings in the engine block, Kwaku had put pieces of greasy pa-
per between them and the crank-shaft. Instead of using a special spring,
his distribution chain was held in its proper place with a piece of copper
pipe. He had raised the oil level in his shock-absorbers beyond normal
so that he could drive more comfortably on roads full of potholes.
(Verrips & Meyer 2001)
 Technology in Process 69

Transformation is also common for artefacts that move between contexts.

This might include cases where objects trade hands between owners who
intend to employ them in different ways or for different purposes, such as
when a new owner converts a fishing vessel to a recreational vessel by re-
moving a crane and winch assembly in order to make space for relaxing on
the back deck. Or it might include cases of artefacts travelling between
different geographical areas, such as when cars imported from Europe to
Eritrea required specific modifications in order to better handle the local
terrain and high altitudes (Belluci & Zaccaria 2012). Transformation
which occurs as a result of shifting contexts also includes cases where arte-
facts exist over long time periods through which new patterns of use or
social values emerge. Examples of such cases include the modification
of buildings to include wheelchair ramps or retrofitting cargo ships with
wingsails.
A common claim in the literature on maintenance and repair is that
modifying artefacts in order to adapt them better to the contexts in which
they will be employed itself constitutes a form of maintenance. In their
seminal paper on the topic, Stephen Graham and Nigel Thrift, for exam-
ple, emphasise the transformative potential of maintenance practices by
claiming that:

maintenance and repair can itself be a vital source of variation, impro-

visation and innovation. Repair and maintenance does not have to
mean exact restoration. Think only of the bodged job, which still allows
something to continue functioning but probably at a lower level; the
upgrade, which allows something to take on new features which keep it
contemporary; the cannibalization and recycling of materials, which al-
lows at least one recombined object to carry on, formed from the bones
of its fellows; or the complete rebuild, which allows something to con-
tinue in near pristine condition. And what starts out as repair may soon
become improvement, innovation, even growth.
(Graham & Thrift 2007)

Despite forming a common theme in the literature on maintenance and

repair,7 the idea that maintenance can be transformative in this way ap-
pears to conflict with dominant metaphysical approaches to the analysis of
artefacts which recognise sharp distinctions between the practices of main-
tenance and modification. As we noted above, such distinctions are under-
stood to stem from the nature of the intentions guiding each form of
activity; only when an operator’s intentions are identical to those of an
artefact’s creator does their activity count as a form of maintenance. In
contrast, many scholars of maintenance and repair studies appear content
to recognise cases in which artefacts are deliberately modified as examples
70 Mark Thomas Young

of maintenance. In the remaining part of this section, I want to outline

some reasons why I think we should be inclined to relax distinctions be-
tween maintenance and modification and follow suit in considering opera-
tions performed to realise intentions that are distinct from those of the
original producers of an artefact, as forms of maintenance.
The first reason emerges from a closer examination of the kinds of prac-
tices, such as replacement, that we often identify with maintenance. In
philosophical discussions surrounding the Ship of Theseus paradox, it is
often assumed that the new planks involved in the replacement of the origi-
nal material differ only to the extent that they are not composed of the
same wood as the original planks. Yet, replacement of this kind, where
replacement occurs with the same yet numerically different parts, consti-
tutes the exception rather than the rule when it comes to replacing parts on
a boat. It is when parts are due for replacement, that owners often take the
opportunity to upgrade their vessels by replacing the part with one which
expands the capacity of the boat in different ways. When a motor is due for
replacement, for example, an owner may decide to replace it with one
which includes the possibility for autopilot functions, leading them to
modify the steering mechanism and install new components at the helm.
Often, the service lives of different parts of a boat make replacement with
the same part either impossible or extremely costly. After all, it is often
unlikely that owners will be able to find the exact same engine model that
was produced 40 years before. Exploring the histories of actual boats pro-
vides further insights into the way in which the replacement of parts is very
often undertaken with a view towards modifying vessels for new purposes.
This may be best illustrated by the case of battleships. In his study of the
maintenance of twentieth-century battleships, historian David Edgerton,
for example, notes how

A ship was not necessarily a stable entity. Ships were often radically
changed, often more than once, in the course of their lives…. What a
Queen Elizabeth battleship was and looked like in 1918 was pretty
clear, but by 1939 they were different ships. Between 1924 and 1934
they were given substantial refits, a process which included trunking the
funnels into one, and fitting huge anti-torpedo bulges on their sides, as
well as changes in the smaller armament. Then, in the 1930s….they
were given new engines, major changes were made to the guns and
mountings, and much of the superstructure was rebuilt.
(Edgerton 2006, p. 93)

As the central example through which philosophers have approached the

question of maintenance, the Ship of Theseus paradox therefore exhibits
serious limitations insofar as it vastly understates the capacity of boats to
 Technology in Process 71

change over time. And it’s only boats that undergo transformation through
the replacement of parts. As most people who own cars or bikes will im-
mediately realise, routine maintenance activities very often involve replac-
ing components such as tires or seatposts with versions better suited to the
conditions in which these artefacts will be employed and the specific goals
and interests of their owners. The customisation that occurs through such
routine maintenance practices reflects how the replacement of parts is of-
ten undertaken with a view towards upgrading artefacts.
The second reason for embracing a conception of maintenance as trans-
formative connects with broader meanings of maintenance as enabling
something to continue existing over time. Put simply, the modification of
artefacts in ways that were not intended by their designers is very often
what enables them to avoid destruction. To see why, consider again the
case of the Golden Gate Bridge. After reviewing vulnerability studies for
the bridge in the wake of the Loma Prieta Earthquake, the Bridge District
judged the risk of collapse in a future earthquake to be unacceptably high.
In order to align the structure with current expectations surrounding seis-
mic integrity, the district was faced with two options: either replacing the
entire structure with a new bridge or initiating a major retrofitting project
which would lead to the modification of nearly all parts of the existing
bridge over a period of three decades (GGBHTD n.d.). As we have seen
already, the district chose the latter option and began the first of three
phases of seismic retrofits in 1997. What these deliberations illustrate,
however, is something that existing discussions on the persistence of arte-
facts appear to have missed entirely. That is, for many artefacts, it is only
because they have been able to change and adapt to emerging interests that
they continue to exist today. When we examine the histories of a variety of
different kinds of artefacts, this simple fact is found to apply widely.
Houses, for example, often face demolition when they can no longer be
changed to match our interests. It is when our phones can no longer sup-
port updates that we begin to consider replacing them. For many artefacts
then, it is the capacity to evolve and adapt which determines how long they
remain in use before they are disposed, decommissioned or destroyed. Yet,
while this simple insight most likely will not strike us as particularly sur-
prising, it nonetheless represents a challenge to deeply held assumptions in
the existing discourse on the metaphysics of artefacts. In stark contrast to
dominant approaches that understand artefacts to persist by remaining the
same, what we find when examining the histories of artefacts in use is pre-
cisely the opposite conclusion – that the continued existence of a wide va-
riety of different kinds of artefacts is premised on their capacity to change.
In Section 3.3, we will begin exploring what it might mean for the meta-
physics of identity over time, if we take seriously the insight that artefacts
persist by changing. In order to do so, we’ll turn our attention to the recent
72 Mark Thomas Young

‘process turn’ in the philosophy of biology, where similar insights concern-

ing the dynamic nature of biological entities have motivated a shift to-
wards an ontology based on processes rather than substances. In seeking
the outlines of a process philosophy of technology, our starting point will
be an examination of the extent to which some of the basic concepts and
approaches central to this growing body of literature might also be appli-
cable to the analysis of the passage of artefacts through time.

3.3 From Biology to Technology: Towards a Process

Philosophy of Artefacts
A growing number of philosophers of biology now suggest that the nature
of the processes by which biological entities move through time demands a
different kind of ontology to that which has remained dominant in western
metaphysics since antiquity. Instead of an ontology based on substances,
they argue that the capacities of organisms to grow and adapt to their en-
vironments necessitates a shift towards an alternative perspective which
takes processes as the fundamental constituents of reality.8 This perspec-
tive, known broadly as process philosophy, inverts the traditional ontologi-
cal hierarchy of western metaphysics by taking change rather than stasis to
be primary. Process philosophy is a broad category used to capture a di-
verse range of philosophical theories, spanning antiquity to the present,
which together share a number of basic tenets. These shared features in-
clude the idea that time and change are the basic categories of metaphysical
understanding, that processes should be regarded as the fundamental com-
ponents of an ontological theory and accordingly that entities which we
experience as static and enduring are best understood in terms of processes
and change (Rescher 2000, p. 5). Contemporary proponents of process
philosophy often consider this broad theoretical orientation to represent a
return to the philosophical perspectives of antiquity which characterised
the nature of reality in terms of constant change and against which classical
philosophers reacted through the development of substance ontology.9
Among the features of organisms that philosophers have identified as
being particularly suited to a process-based metaphysics are their capacity
to ‘develop progressively over time, acquiring certain properties and ca-
pacities and losing others along the way’ alongside their possession of a
metabolism through which they ‘constantly exchange energy and matter
with their surroundings’ (Dupré & Nicholson 2018). In previous sections
of this chapter, we explored how the histories of a wide range of artefacts
reveal similar processes at play. As the Ship of Theseus paradox so aptly
illustrates, artefacts also exchange matter with their surroundings, through
processes which challenge the dominant ways of understanding the rela-
tionship between technology and time in metaphysics. In his discussion of
 Technology in Process 73

the material composition of artefacts, for example, metaphysician Simon

Evnine suggests that the capacity of many artefacts to exchange their mat-
ter over time warrants the ascription of a ‘metabolism’ to artefacts in a
metaphorical sense (Evnine 2016, p. 16). Similar observations for large
socio-technical systems can be found in the burgeoning literature on
‘urban metabolism’. Yet, despite broad similarities between the processes
by which organisms and artefacts persist through time, a process philoso-
phy of artefacts nonetheless has yet to be developed. One of the main rea-
sons why relates to a widespread and longstanding resistance among
philosophers to recognise the extent to which these characteristics can also
be understood to apply to technologies. Since antiquity, metaphysicians
within the Aristotelian tradition have been more interested in exploring the
differences between natural objects and artefacts than they have been in
investigating their similarities (Preston 2022).10 However, this tendency to
focus on their dissimilarities has led many to overlook the various ways in
which the passage of artefacts and organisms through time share a variety
of important similarities, and by doing so, has helped to hinder the devel-
opment of a deeper appreciation of the relationship between technology
and time in western metaphysics. In arguing for the relevance of a process-
based metaphysics for biological entities, Daniel J. Nicholsen, for example,
draws a comparison between the life cycle developments of machines and
organisms and concludes that:

the form of an organism does not stay fixed during its lifetime. Organisms
grow and develop. Some undergo major morphological changes, such
as metamorphosis. Nothing comparable happens to a machine…given
that the fixity of its internal structure is precisely what allows it to per-
form the function it was designed for.
(Nicholson 2018)

But machines also undergo growth and development. While it may be true
that the internal structures of many machines remain fixed throughout
their service lives, there are nonetheless a wide range of machines which
undergo major changes while continuing to perform the functions that
were intended by their designers. Consider, for example, the common prac-
tice of modifying car engines with turbochargers, aftermarket air filters
or newer performance chips in order to improve their performance.
Furthermore, in some cases, machines must undergo change in order to
continue performing the functions they were designed for. Examples such
as this include industrial machinery, which avoids replacement by being
upgraded to match the operational capacity of newer models, or the retro-
fitting of vehicle engines in order to remain compliant with evolving legal
requirements regarding emissions.
74 Mark Thomas Young

In earlier sections of this chapter, we examined how a wide range of ar-

tefacts develop and change throughout their histories in order to match
evolving expectations concerning how they should be. Furthermore, we
noted how for many artefacts, this capacity to undergo adaptation is cru-
cial for avoiding replacement and consequent disposal or destruction.
Examining the histories of artefacts in use therefore allows us to see how
H.G Wells’ famous dictum ‘adapt or perish, now as ever, is nature’s inexo-
rable imperative’ can be understood to apply as much to artefacts as it does
to natural organisms. Given such broad similarities between the temporal
character of organisms and artefacts, we should be forgiven for suspecting
whether a process-based metaphysics might also present a number of ben-
efits for the analysis of technology. Our suspicions may be enhanced fur-
ther when we recognise the extent to which the basic analytical approaches
currently applied in the study of biological entities as processes, bear an
immediate relevance and clear applicability to the examination of the his-
tories of artefacts. Consider, for example, how the adoption of process
perspectives generates a new range of explanatory tasks that differ from
those which have traditionally occupied the focus of metaphysics. As John
Dupré notes:

One way of seeing the difference between a process and a substance

ontology is in terms of the explanatory problems they face. Whereas a
substance ontology has to provide explanations of change, a process
ontology instead faces the problem of explaining stability or the appear-
ance of stability…This perspective profoundly transforms questions
about identity over time. First, the persisting individual is not a meta-
physical given, in which the individual is identified and investigated to
detect the conditions under which it might persist. Rather, it is some-
thing the individual does: persisting is, indeed, hard work.
(Dupré 2018)

Rather than attempting to deny the presence of stability altogether, process

philosophers seek to show how the semblance of enduring things emerges
through the stabilisation of processes. In recent work in the philosophy of
biology, homeostasis, metabolism and evolution have all been suggested to
represent processes of stabilisation from which the stability of species and
organisms over time can be understood to derive.11 Yet, stabilisation also
represents a particularly promising way of conceptualising the temporal-
ity of artefacts insofar as it provides an analytical framework through
which we can explore the different ways in which artefacts continue to
change after production. The fact that, unlike organisms, artefacts do not
stabilise themselves and instead require this work to be performed by ex-
ternal agents reflects the importance of the practice of maintenance in
 Technology in Process 75

enabling the persistence of artefacts over time and helps shed light on what
maintenance might be understood to mean in a metaphysical context. For
artefacts, persistence is also hard work. And that work is maintenance.
Yet, exactly what maintenance represents in metaphysical terms requires
further elaboration. Before a process philosophy of artefacts can develop
an explanatory framework through which the stability of artefacts can be
accounted for, we must first determine what the stabilisation of processes
should be understood to mean in relation to technologies over time. The
answer to this question is anything but obvious. The literature on process
philosophy of biology, for example, offers at least two distinct suggestions
for how stability should be understood in relation to biological entities: the
first is the stability of properties, such as an organism’s achievement of a
constant temperature through homeostasis. The second is the stability of
directional processes, such as the sequences of life cycle development indi-
vidual organisms undergo (Dupré 2018, n.7). In the following, we will
consider each conception of stability: of properties and developmental pro-
cesses, to see whether they might be fruitfully applied within a process
philosophy of artefacts.
The first candidate, the stability of properties, represents perhaps the
most intuitive option insofar as it reflects common conceptions of mainte-
nance as hindering processes of change which the materials comprising
artefacts undergo. Consider, for example, the iron which makes up much
of the steel from which engineered structures such as bridges or building
frames are constructed. Iron begins as iron ore – rocks and minerals rich in
oxides. In order to produce metallic iron which can be used to make steel,
iron ore must be refined through a process which requires great amounts
of heat to drive off the oxygen, converting the oxide to a metal. The result-
ing product however is inherently unstable, and if left unprotected, it im-
mediately begins corroding – reacting to water and oxygen and slowly
reverting back to an oxide in the form of rust, a similar compound to the
ore from which it began. Here, we see how the materials from which engi-
neered structures are composed exist as processes that must be stabilised in
different ways, through the application of protective coverings such as
paint and vinyl coatings, for example, or through the replacement of cor-
roded elements and materials, in order for the bridge to appear as a stable
entity. Yet, while it might be tempting to understand stabilisation in terms
of the attempt to hold the material properties of artefacts in stasis, we saw
in the last section a variety of reasons why this would not provide a suit-
able basis for a general process philosophy of artefacts based around the
practice of maintenance.
In the first place, conceptualising stability in terms of the stasis of prop-
erties would severely restrict the range of practices that such an account
could consider. After all, very few activities that we identify as forms of
76 Mark Thomas Young

maintenance manage to prevent the change of materials from which arte-

facts are composed without also modifying them in the process.12 Even
activities which explicitly aim at delaying or hindering the transformation
of materials often involve adding other materials: protective coatings or
weather barriers, for example. Furthermore, a wide range of activities we
identify as maintenance often deliberately facilitate the transformation of
materials composing artefacts over time. For example, proper maintenance
of a cast iron skillet allows the build-up of material on the cooking surface,
facilitating the development of a superior cooking surface, while at the
same time preventing the development of rust. Second, understanding
maintenance in terms of preventing the change of materials would sever
the relation between maintenance and persistence for a wide variety of
artefacts. As we noted in the previous section, the continued existence of
many artefacts depends crucially on their capacity to adapt and change.
For such artefacts, understanding the stabilisation of processes as consist-
ing of preventing change would leave us unable to explain how it is they
persist through time.
What we require then is a broader understanding of the stabilisation of
processes which is wide enough to encompass activities which both pre-
vent and enable the change of materials. Viewed in this light, directional
processes yield a more promising option for an object of stabilisation. As
we saw in the last section, a closer examination of the range of mainte-
nance practices that we perform on artefacts reveals their shared charac-
teristic to lie in the attempt to guide change in the direction of certain
normative expectations. Here we find a conception of the stabilisation of
processes that can serve as a basis for a general account of technological
maintenance. According to this account, instead of understanding artefacts
to be comprised of processes that demand stabilisation, artefacts them-
selves are understood as processes that can develop in different ways over
time. Stabilisation consists of actively intervening in the processual devel-
opment of an artefact in order to guide it in ways which ensure coherence
between our understanding of how the object is and our normative expec-
tations for it (i.e., how the artefact should be). The central benefit of inter-
preting stabilisation in this way is that it enables us to explain how artefacts
persist through time not only by resisting change but also by undergoing it
too. It applies as much to cases in which our normative expectations do
not undergo change, as it does to cases in which we seek to adapt artefacts
to emergent interests and uses. Maintaining an artefact therefore means
different things; in some cases, it may involve seeking to prevent the mate-
rial change of an artefact over time while in others it may involve their
deliberate modification through practices of redesign, retrofitting and
customisation.
Understanding stability to pertain to the relation between artefacts and
normative expectations in this way also enables us to see how problems
 Technology in Process 77

maintenance can be understood to address, alongside the ways in which it

does so, may extend beyond materiality altogether. For example, by con-
ceiving instability in terms of a divergence between our understanding of
the state of an artefact and our normative expectations, this account makes
room for the emergence of a instability even in cases where artefacts do not
undergo change. In his study of seismic retrofits of Californian bridges,
Benjamin Sims examines the problems addressed by practices of repair as
forms of ‘slippage’ which occur between our normative expectations and
the material conditions of artefacts (Sims 2016).13 In addition to identify-
ing slippage in cases where artefacts undergo material change over time
through degradation, Sims also highlights how slippage can occur through
obsolescence.

when objective conditions are perceived as stable, but normative expec-

tations have changed: for example, engineering standards and building
codes have been updated, there is a movement to add bike lanes to exist-
ing bridges, broadcasters want to transmit high definition television sig-
nals. Existing technology comes to be seen as outdated or not responsive
to social needs.
(Sims 2016, p. 16)

Obsolescence, like all forms of instability which may develop throughout

the history of an artefact, requires attention. Left unchecked, obsolescence
threatens the persistence of artefacts, drastically increasing their risk of
replacement, disposal or destruction. However, obsolescence, like all forms
of slippage, can be resolved through stabilisation in which we perform
work to restore coherence between understandings of the state of an arte-
fact and certain relevant normative expectations. One of the most impor-
tant activities through which such coherence is restored is practices of
repair and modification which bring the state of an artefact ‘back in line’
with how it is expected to be. Yet, examining the different ways in which
stability can be restored leads inevitably towards a broader conception of
maintenance than that can be found in the existing literature on the meta-
physics of artefacts. Artefacts can be stabilised not only through material
practices but also through discursive work. If you become convinced that
the state of an artefact is not as it should be, for example, I can attempt to
convince you either that you are mistaken about the state of the artefact or
that your normative expectations should be altered.14 In either case, if I am
successful in my attempts to convince you, then coherence will have been
restored and the artefact stabilised.
Yet, despite the various options at our disposal, not all cases of slippage
are resolved through stabilisation. Failing to stabilise artefacts can bring
their processes of development to an end. It is when we fail to see how our
bike can be changed to match our expectations that we may decide that it
78 Mark Thomas Young

needs replacement or should be discarded, for example, or when a govern-

ing body decides that it is too costly to modify a bridge that no longer ad-
equately meets the needs of society that it ends up being decommissioned
and replaced. This latter example highlights the extent to which the stabil-
ity and instability of a wide variety of artefacts often possess irreducible
social and political dimensions. Not only does the question of whose nor-
mative expectations matter but also the wider political structures that de-
termine the power that different actors and groups within society may
exercise. Insofar as instability represents the potential for change, cases in
which it is primarily actors and groups with little political power who feel
that an artefact does not match their expectations, most likely will not re-
sult in destabilisation. As an example, consider again the case of the Golden
Gate Bridge. Since its construction in 1937, the number of people who
have jumped to their deaths from the bridge has steadily increased. Today,
the bridge remains the deadliest structure in the world for suicides: to date
over 1500 people have lost their lives in this way. It has long been thought
that this unusually high number of deaths is at least partly a result of the
great height of the bridge above the water (75 metres) combined with a
relatively low guard railing (just over one metre). Already by the 1960s,
activist groups began campaigning in order to convince the Bridge District
to establish safety barriers such as higher guard railings in order to help
prevent further deaths. However, 30 years of lobbying the board directly
ultimately failed to destabilise the artefact insofar as members of the Bridge
District remained convinced that the bridge matched their normative ex-
pectations. The cause of the high rate of suicides, they argued, lay not in
the design of the bridge but rather in wider political developments such as
the defunding of mental health services in California. After years of failed
attempts, activist groups realised the need to switch tactics and began at-
tempting to cultivate broader support for change among members of
the public. This new strategy eventually proved successful in convincing
enough members of the board to change their normative expectations re-
garding the safety features of the bridge.15 This slippage led to attempts to
restore coherence, resulting in the installation of safety nets which now
span almost the entire length of the bridge.

3.4 Conclusion
Throughout this chapter, I’ve attempted to outline the rudiments of a wider
research program that approaches the metaphysics of maintenance by ana-
lysing the nature of artefacts as processes. Such a program would involve,
in the first place, empirical work to chart the contours of such processes,
not only in the case of individual artefact tokens such as the Golden Gate
Bridge but also within and between different artefact kinds. Second, it
would involve exploring the structure and nature of those processes by
 Technology in Process 79

investigating how they are influenced by the range of different environ-

ments within which artefacts are located alongside identifying the condi-
tions that enable or constrain material change.
By now, however, the attentive reader will have noted a shift towards
concepts and questions which seem more naturally associated with science
and technology studies (STS) than metaphysics. Yet, while exploring social
and political processes which drive technological change might seem meth-
odologically far from home for metaphysicians, we shouldn’t be too sur-
prised to find such tasks at the centre of a metaphysical research program
for the study of artefact maintenance. After all, artefacts have long been
understood to straddle the boundary between the human and material
worlds. As is well known, artefacts cannot be apprehended entirely in
terms of their physical shape and dimensions but must also be understood
in relation to human purposes, to what they are considered to be for. As
noted above, existing studies in the metaphysics of artefacts have tended to
overlook how this teleological component changes over time.16 However,
attending to the maintenance of artefacts renders such tendencies infeasi-
ble. As I have argued in this chapter, the changes that many artefacts un-
dergo through practices of maintenance can only be explained through
recourse to changing perceptions regarding what they are for and conse-
quently, how they should be. Examining such changes leads inevitably to a
process-based account that takes change to be primary.
Yet, there is a final concern about this account that I have yet to address,
one which metaphysicians in the Western tradition have long considered to
be of crucial importance. After all, if we cannot specify those aspects of an
entity which remain the same after undergoing change, does not such an
account fall foul of the same concerns regarding the possibility of meaning-
ful communication about the world which motivated the development of
substance ontology in the first place? Not at all – eschewing the idea that
reality, at the most fundamental level, consists primarily of static enduring
entities does not mean simply accepting the problem of change. Instead, by
operating with a different set of basic ontological assumptions, proponents
of process philosophy are able to address the problem in ways that are
unavailable to those operating within the paradigm of substance ontology.
Johanna Seibt, for example, advances a process account which recom-
mends framing questions concerning the ontology of persistence in terms
of processes rather than things. Her ‘recurrence account of persistence’
addresses the problem of change by taking statements about change and
identity over time to apply to processes at different levels of specificity.
According to Seibt, statements about identity over time, such as the ‘the
Golden Gate Bridge today is the same bridge that was built in 1937’ apply
to general dynamics: processes understood as dynamic subjectless activi-
ties. Such processes are both ontologically fundamental and homoeomer-
ous. What this means is that while these processes are dynamic, they do
80 Mark Thomas Young

not themselves undergo change and so are not analysable in terms of tem-
poral parts or phases. Statements about change, on the other hand, such as
‘the bridge used to have a concrete deck, now it has a steel orthotropic
deck’ apply to specific dynamics: processes which, because they do possess
different parts, exhibit the presence of change from one time to another
(Seibt 2008).
Another approach, proposed by Robert Koons and Timothy Pickavance,
employs a notion of processes in terms of human practices to explain
how artefacts can be understood to persist through time despite undergo-
ing change. What they call the ‘continuous history theory’ of artefactual
persistence

exploits certain practices of use and maintenance. For example, if a

watch persists, it is because there is a certain ongoing history of use of
the watch as a watch and of maintenance of the watch as a watch. Let’s
suppose that these social processes or practices have a kind of unity
through time, that the process as a whole is metaphysically fundamen-
tal, not the various instantaneous events that make up the process…. If
that is so, then the metaphysical unity of the process over time can be
used to ground the persistence of the artifact.
(Koons & Pickavance 2017, p. 561)

While the full potential and limitations of process-based accounts remain

to be fully elaborated, the approaches above illustrate how a transition to
an ontology based on processes rather than substances does not necessarily
preclude us from responding to the problem of change. At the same time,
the shift towards an ontological framework which understands change to
be primary holds important consequences for the topic of maintenance
within metaphysical discussions of artefactual persistence. Rather than
being a derivative activity tending to the preservation of that which has
already been created, process philosophical perspectives reveal how main-
tenance can be understood as constitutive of the nature of artefacts them-
selves, as the glue which binds their identity through time. As calls for a
process philosophy of technology grow louder,17 maintenance begins shift-
ing from the periphery to the spotlight, and in doing so, moves closer to
emerging as a central topic in the metaphysics of artefacts.

Notes
1 For an overview of these variations, see Dumsday (2019, pp. 69–72).
2 For the original statement of this objection to four dimensionalism, see
McTaggert (1927, p. 14).
3 The role played by artefact kinds in helping to determine persistence condi-
tions, alongside their relation to the intentions of the producer, is a matter of
debate. See, for example, Bloom (1996); Evnine (2016, p. 72); Koslicki (2018,
 Technology in Process 81

p. 221) for different views on the relationship between intentions, artefact

kinds and the modal flexibility of artefacts.
4 This is a standard feature of intentionalist theories from antiquity to the pres-
ent, see, for example, Young (2021); Koslicki (2018, p. 222).
5 For an account of such processes in relation to bridges in California, see Sims
(2016).
6 This has often been considered a consequence of the dominant focus on inven-
tion and innovation in a variety of fields attending to the historical study of
technology, see, for example, Edgerton (2006, p. 81); Russell & Vinsel (2018).
For a notable exception, see Brand (1994).
7 See, for example, Edwards et al. (2011); Schaffer (2011); Henke & Sims (2020,
p. 21); Gupta (2021); Harper (1987, p. 19); Lejeune (2019); Denis, Mongili &
Pontille (2015).
8 See, for example, the contributions collected in Nicholson & Dupré (2018),
alongside Dupré (2017, 2018).
9 See, for example, discussions of Heraclitus in Dupré and Nicholson (2018);
Winters (2017, chap 1); van Geert and de Ruiter (2022, chap 1).
10 For an overview of the different ways in which this tradition has upheld opposi-
tions between artefacts and natural objects, see Newman (2004).
11 See, for example, Dupré (2017); Nicholson (2018); Meincke (2018).
12 The removal of salt residue from boats or the underside of cars with the use of
water or cleaning lenses with fluids are among the few examples.
13 By choosing the term “slippage” for such cases, Sims encourages a shift in dis-
cussions of notions of stability in science and technology studies, away from
the traditional focus on how stability is established and secured through pro-
cesses of closure and towards an examination of how stability is lost, often re-
peatedly throughout the lifetime of an individual artefact. See also Denis and
Pontille who encourage a similar shift in noting how “even at rest, artefacts are
not as sealed and as stable as they may appear, maintenance and repair studies
show. The ability for technologies to remain the same and to be taken for
granted by most of their users requires a constant work that such terms as
‘black box’ or ‘immutable mobiles’ seriously understate’. (2020).
14 Examples of cases of ‘discursive repair’ such as this are a common feature of
ethnographic studies of maintenance and repair. See, in particular, Henke
(2000); Orr (1996).
15 For a more detailed recounting of this story, see Bateson (2012).
16 Among the few exceptions are studies exploring cases of functional exaptation,
see for example Thomasson (2014); Preston (1998).
17 See, for example, Coeckelbergh (2023); Young (2021); Brunet (2021).

Bibliography
Bateson, John. (2012) The Final Leap: Suicide on the Golden Gate Bridge. Berkely:
University of California Press.
Brand, Stewart. (1994) How Buildings Learn: What Happens After They’re Built.
New York: Penguin Books.
Belluci, Stefano & Zaccaria, Massimo. (2012) “Engine of Change: A Social History
of the Car-mechanics Sector in the Horn of Africa” in Jan-Bart Gewald, André
Leliveld & Ive Peša (eds) Transforming Innovations in Africa: Explorative
Studies on Appropriation in African Societies. Leiden: Koninklijke Brill NV
pp. 237–256.
82 Mark Thomas Young

Bloom, Paul. (1996) “Intention, History and Artifact Concepts” Cognition 60

pp. 1–29.
Dumsday, Travis. (2019) Dispositionalism and the Metaphysics of Science.
Cambridge: Cambridge University Press.
Brunet, Tyler David Price. (2021) ‘The Machine-Organism Distinction’, PHD
Thesis, University of Cambridge. Available at: https://www.repository.cam.
ac.uk/items/83ecf29a-d462-45e8-88d0-1e4d3729375c (accessed June 10, 2023).
Coeckelbergh, Mark. (2023) “Technology as Process” in Glenn Miller, Helena M.
Jerónimo & Qin Zhu (eds) Thinking Through Science and Technology:
Philosophy, Religion and Politics in an Engineered World. Maryland: Rowman
& Littlefield pp. 55–68.
Denis, David J. (Denis, Jérôme & Pontille, David) (2020) “Why do Maintenance
and Repair Matter?” in Anders Blok, Ignacio Fárias & Celia Roberts (eds) The
Routledge Companion to Actor-Network Theory. New York: Routledge pp.
283–293.
Denis, Jérôme, Mongili, Allessandro & Pontille, David. (2015) “Maintenance and
Repair in Science and Technology Studies” Tecnoscienza: Italian Journal of
Science and Technology Studies 6(2) pp. 5–15.
Dupré, John & Nicholson, Daniel J. (2018) “A Manifesto for a Processual
Philosophy of Biology” in Daniel J. Nichlson & John Dupré (eds) Everything
Flows: Towards a Processual Philosophy of Biology Oxford: Oxford University
Press pp. 3–45.
Dupré, John. (2017) “The Metaphysics of Evolution” Interface Focus 7: 20160148.
Available at: http://dx.doi.org/10.1098/rsfs.2016.0148 (Accessed May 20,
2023).
Dupré, John. (2018) “Processes, Organisms, Kinds and the Inevitability of
Pluralism” in O. Bueno, R. Chen & M.B. Fagan (eds) Individuation, Process and
Scientific Practices. New York: Oxford University Press pp. 21–38.
Edgerton, David. (2006) The Shock of the Old: Technology and Global History
since 1900. London: Profile Books.
Edwards, Paul N. et al. (2011) “AHR Converation: Historical Perspectives on the
Circultation of Information” American Historical Review 116(5) pp. 1393–
1435.
Evnine, Simon J. (2016) Making Objects and Events: A Hylomorphic Theory of
Artifacts, Actions and Organisms Oxford: Oxford University Press.
van Geert, Paul and de Ruiter, Naomi. (2022) Toward a Process Approach in
Psychology: Steeping into Heraclitus’ River. Cambridge: Cambridge University
Press.
GGBHTD (n.d.) Seismic Retrofit Construction Project [Online]. Available at
https://www.goldengate.org/district/district-projects/seismic-retrofit/ (accessed
June 15, 2022).
Graham, Stephen & Thrift, Nigel. (2007) “Out of Order: Understanding Main­
tenance and Repair” Theory, Culture & Society 24(3) pp. 1–25.
Gupta, Akhil. (2021) “Infrastructure as Decay and the Decay of Infrastructure” in
Ghassan Hage (ed) Decay. Durham: Duke University Press pp. 37–46.
 Technology in Process 83

Harper, Douglas. (1987) Working Knowledge: Skill and Community in a Small

Shop. Chicago: Chicago University Press.
Henke, Christopher R. (2000) “The Mechanics of Workplace Order: Toward a
Sociology of Repair. Berkeley Journal of Sociology Vol. 44 pp. 55–81.
Henke, Christopher R. & Sims, Benjamin (2020) Repairing Infrastructures: The
Maintenance of Materiality and Power. Massachusetts: MIT Press.
Hilpinen, Risto. (1993) “Authors and Artifacts” Proceedings of the Aristotelian
Society, 93, pp. 155–178.
Hobbes, Thomas. (1656) Elements of Philosophy, the first section, concerning the
body: Written in Latine by Thomas Hobbes of Malmesbury and now translated
into English. London: Andrew Crooke.
Josephson, Paul R. (2017) Traffic. New York: Bloomsbury Academic.
Koons, Robert C. & Pickavance, Timothy (2017) The Atlas of Reality: A Com­
prehensive Guide to Metaphysics. West Sussex: Wiley Blackwell.
Koslicki, Kathrin. (2018) Form, Matter, Substance. Oxford: Oxford University
Press.
Lee, Mi-Kyoung. (2010) “Antecedents in early Greek philosophy” in Richard Bett
(ed) The Cambridge Companion to Ancient Scepticism. Cambridge: Cambridge
University Press pp. 13–35.
Lejeune, Christoph (2019) “Interruptions, Lunch Talks, and Support Circles: An
Ethnography of Collective Repair in Steam Locomotive Restoration” in Ignaz
Strebel, Alain Bovet & Phillippe Sormani (eds) Repair Work Ethnographies:
Revisiting Breakdown, Relocating Materiality. Singapore: Palgrave Macmillan
pp. 221–251.
McTaggert, John M.E. (1927) The Nature of Existence Vol 2 C.D. Broad (ed)
Cambridge: Cambridge University Press.
Meincke, Anne Sophie. (2018) “Persons as Biological Processes: A Bio-Processual
Way Out of the Personal Identity Dilemma” in Daniel J. Nicholson & John
Dupré (eds) Everything Flows: Towards a Processual Philosophy of Biology.
Oxford: Oxford University Press pp. 357–378.
Meincke, Anne Sophie. (2019) “The Disappearance of Change: Towards a Process
Account of Persistence” International Journal of Philosophical Studies 27(1)
pp. 1–19.
Meincke, Anne Sophie. (2021) “Processual Animalism: Towards a Scientifically
Informed Theory of Personal Identity” in Anne Sophie Meincke & John Dupré
(eds) Biological Identity: Perspectives from Metaphysics and Philosophy of
Biology. New York: Routledge pp. 251.
Newman, William R. (2004) Promethean Ambitions: Alchemy and the Quest to
Perfect Nature. Chicago: Chicago University Press.
Nicholson, Daniel J. (2018) “Reconceptualizing the Organism: From Complex
Machine to Flowing Stream” in Daniel J. Nicholson & John Dupré (eds)
Everything Flows: Towards a Processual Philosophy of Biology. Oxford: Oxford
University Press pp. 139–166.
Nicholson, Daniel J. and Dupré, John (eds) (2018) Everything Flows: Towards a
Processual Philosophy of Biology. Oxford: Oxford University Press.
84 Mark Thomas Young

Orr, Julian E. (1996) Talking about Machines: An Ethnography of a Modern Job.

New York: Cornell University Press.
Pearce, David. (2016) “Collective Intentionality and the Social Status of Artifact
Kinds” Design Science 2(3) pp. 1–28.
Plutarch. (1859) John Dryden (trans) Plutarch’s Lives Vol 1. Boston: Little Brown
and Company.
Preston, Beth. (1998) “Why Is a Wing Like a Spoon? A Pluralist Theory of
Function” The Journal of Philosophy 95(5) pp. 215–254.
Preston, Beth. (2013) A Philosophy of Material Culture: Action, Function and
Mind. New York: Routledge.
Preston, Beth. (2022) “Artifact” in Edward N. Zalta & Uri Nodelman (eds) The
Stanford Encyclopedia of Philosophy (Winter 2022 Edition). Available at:
https://plato.stanford.edu/archives/win2022/entries/artifact/ (accessed August 23,
2022).
Rescher, Nicholas. (1996) Process Metaphysics: an Introduction to Process Phi­
losophy. Albany: State University of New York Press.
Rescher, Nicholas. (2000) Process Philosophy: A Survey of Basic Issues. Pittsburgh:
University of Pittsburgh Press.
Russell, Andrew L. & Vinsel, Lee. (2018) “After Innovation, Turn to Maintenance”
Technology and Culture 59(1) pp. 1–25.
Schaffer, Simon. (2011) “Easily Cracked: Scientific Instruments in States of
Disrepair” Isis (102(4) pp. 706–717.
Seibt, Johanna. (1997) “Existence in Time: From Substance to Process” in J. Faye,
U. Scheffler & M. Urchs (eds) Perspectives on Time (Boston Studies in the
Philosophy of Science Vol. 189). Dordrecht: Springer pp. 143–182.
Seibt, Johanna. (2004) “Process and Particulars” in Michel Weber (ed) After
Whitehead: Rescher on Process Metaphysics. Frankfurt: Ontos Verlag pp.
113–134.
Seibt, Johanna. (2008) “Beyond Endurance and Perdurance: Recurrent Dynamics”
in Christian Kanzian (ed) Persistence. Frankfurt: Ontos Verlag pp. 133–164.
Simons, Peter. (1987) Parts: A Study in Ontology. Oxford: Oxford University
Press.
Simons, Peter. (2000) “Continuants and occurrents” Proceedings of the Aristotelian
Society Supplementary Volume 74(1) pp. 59–74.
Sims, Benjamin. (2016) “Seismic Shifts and Retrofits: Scale and Complexity in the
Seismic Retrofit of California Bridges” in Mike Hodson & Simon Marvin (eds)
Retrofitting Cities: Priorities, Governance and Experimentation. New York:
Routledge pp. 13–33.
Thomasson, Amie. (2003) “Realism and Human Kinds” Philosophy and Phenom­
enological Research 67, pp. 580–609.
Thomasson, Amie. (2014) “Public Artifacts, Intentions, and Norms” in
M. Franssen, P. Kroes, T.A.C. Reydon & P. Vermaas (eds) Artefact Kinds:
Ontology and the Human-Made World. Cham: Springer pp. 45–62.
Wasserman, Ryan. (2006) “The Problem of Change” Philosophy Compass 1(1)
pp. 48–57.
Winters, Andrew M. (2017) Natural Processes: Understanding Metaphysics
Without Substance. Cham: Palgrave Macmillan.
 Technology in Process 85

Verrips, Jojada & Meyer, Birgit. (2001) “Kwaku’s Car: The Struggles and Stories
of a Ghanian Long-Distance Taxi Driver” in Daniel Miller (ed) Car Cultures.
Oxford: Berg pp. 153–184.
Young, Mark Thomas. (2021) “Maintenance” in N. Doorn & D. Michelfelder
(eds) The Routledge Handbook of the Philosophy of Engineering. New York:
Routledge.
4 There, I Fixed It! On the
Status and Meaning of Repair
Tiago Mesquita Carvalho

4.1 Introduction
What happens during repair? How does one learn to repair an artefact?
and what is the relationship between repair and function? And if repair is
a skill that requires knowledge, how can it be described? Can there be a
general theory of repair capable of reducing knowledge to steps and proce-
dures? To address these matters, we will use the theoretical underpinnings
of post-phenomenology as formulated by Don Ihde, as well as other ana-
lytical concepts from authors who have put forward a metaphysical theory
of artefact function. The similarities with other forms of knowledge in
science, technology, design and ethics will also be explored to clarify sev-
eral issues around repair, mainly its relation with function and the design-
er’s intention, its meaning and the nature of the knowledge it uses.
In Section 4.2, we start by examining the definition of repair as a
means to restore the proper artefact function, bridging a discontinuity
between its initial state before the breakdown and its final state after the
repair. This definition will be explored to determine how one can tell
whether a repair has taken place since a good test often means reusing
the artefact. Nevertheless, there are artefacts that cannot be tested with
use. This leads us to question the nature of the knowledge required to
repair an artefact and its relation to the knowledge used to produce and
design said artefact.
In Section 4.3, we move forward with Ihde’s understanding of the place
of technology in the lifeworld. We draw some implications about the na-
ture and meaning of their function and breakdowns. The lifeworld is filled
with users’ intentions and these interplay with the ambiguity of artefacts.
An analysis of repair has to take this into account. Proper function and the
designer’s intention are not enough to appreciate their selection and varia-
tion in reproduction, as well as the dynamics of the transfer between cul-
tures. Scrutinising examples of how the same artefacts are used and
repaired in different cultures, rather than a mere comparative exercise, has

DOI: 10.4324/9781003316213-5
 There, I Fixed It! On the Status and Meaning of Repair 87

analytical value for the concepts surrounding repair and maintenance.

Furthermore, an analogy between artefacts and texts is drawn, highlight-
ing relevant affinities. Section 4.4 continues this exploration by framing
repair in context and exploring how a technical syllogism regarding repair
must be context-specific. The technical syllogism of repair serves to qualify
what comprises a good or bad repair and which judgements and actions
are valid to achieve the functionality of an artefact. However, the syllogism
is highly dependent on the context and its available parts and tools.
Section 4.5 discusses the relationship between repair knowledge and sci-
entific and technological knowledge in the design of artefacts. The similari-
ties with other forms of knowledge are pointed out and their role in the
skill of repair is discussed. Here we distinguish between repair knowledge
that can be formalised in rules and procedures and that which represents
the cultural consequences of the accumulation and transmission of person-
alised forms of that knowledge. Finally, Sections 4.6 and 4.7 discuss a
phenomenology of the repair attitude and its structure.

4.2 About Repair
Preston (2009, p. 217) draws attention to how phenomena like mainte-
nance, repair and rebuilding should be integral to a theory of artefact func-
tion. Although there are significant differences between individual and
collective repair concerning the division of labour and the organisation of
work, at first sight, it does not seem essential for the purpose of this analy-
sis to distinguish types of repairs based on artefact kinds. Repair is either
done individually or collectively, and there must be a vision of all the steps
required for an artefact to achieve its function. The complexity of the arte-
fact, as well as time constraints and lack of personnel and resources, have
a bearing on whether a repair should be accomplished individually or
collectively.
The object of repair seems to be artefacts. To cultivate and trim a garden
or to heal a patient is to cooperate with an already active process in such
entities. It seems inappropriate to say that we repair a garden or a person
except perhaps when used as a figure of speech. There is something distinc-
tive about the object of repair that means it is more of a machine than part
of life or nature. The much-vaunted term ‘technological fix’ for solving is-
sues denotes a cultural propensity to turn social or political issues into
technical ones. Some authors, such as Spelman (2002, pp. 4–6), seem to
identify repair on a continuum with other terms like restoration, preserva-
tion or even healing, which encompass a recovery of some earlier state
before a breakdown, whether this is understood to apply to living entities
such as ecosystems, geopolitical entities like nations or things such as
relationships.
88 Tiago Mesquita Carvalho

Following Baker (2009, p. 82), we will consider repair to only be con-

cerned with technical artefacts. However, we are not saying that repair
does not play a role in the nexus between material and social realities. As
Henke and Sims (2020, p. 19) note, repairing a bridge, for instance, is not
only about fixing a structure but also a whole system of artefacts, institu-
tions and organisations. There is a continuum across the social and tech-
nological orders, implying that repair is not only about changes in artefacts
but also social systems. Furthermore, repair sometimes depends on the best
available knowledge. Repair usually occurs after something breaks down,
but the causes for the failure may remain hidden while a makeshift solu-
tion is adopted. Knowledge, standards and conceptions of maintenance
and repair coevolve.
Besides, when we say that repair concerns artefacts, these include arte-
facts generated by activities such as art, craft, invention or engineering.
Repair may apply to artistic, cultural or decorative artefacts. One can
repair a painting or a frame, an activity which is often related to the con-
servation and restoration of cultural property. The renovation and refur-
bishment of technologies such as infrastructures, buildings and houses are
also a type of repair. In addition to being a collective endeavour, they must
be carefully planned with tasks assigned and deadlines met. Activities such
as amending software code, adjusting the brightness of videos and images
or subtitling films are also examples. These cases are not really an excep-
tion to the definition so much as a narrowing and specialisation of the
meaning and judgements involved.
At the same time, it does not matter whether we are discussing the repair
of mobile phones, automobiles or bridges. Each repair skill seems distinct
and has its own language shaped by the particularities of the materials and
components that usually break down, but there should be something com-
mon to the whole structure of the repair skill, that is, the need to restore
their functional state. In the first instance, the repair is then defined as re-
storing an artefact to working order after a breakdown. It appears to be a
return to the past, a way of reestablishing continuity.
Repair is therefore teleological, an end-directed activity, towards a final
state of the artefact, such that it works and delivers a function. For now,
we will assume that such a final state should resemble the initial state, as
accomplished by the artefact’s proper function (Vermaas & Houkes, 2003,
p. 265).1 In other words, repair must assume, from the outset, the existence
of a solution. There is no repair without belief in a solution and the ability
to find it; one moves towards a problem not as an end in itself but on the
assumption that it is a means to an end, towards the meaning of ‘it works!’
or ‘there, I fixed it’, whatever that may be. Repair is thus guided by the aim
of providing the artefact with a working state, imagined as reliable, and
acting as the regulator of all steps and actions to be taken.
 There, I Fixed It! On the Status and Meaning of Repair 89

In addition, a more demanding definition of repair requires that the final

state is defined as one where the fulfilment of the artefact function can be
accomplished without any mishaps, provided that the user is aware of how
to use it. A test is needed and use is often what suffices provided one knows
what to do. A car out of fuel is not broken but it cannot be driven.
Moreover, we do not need to look at a more complex artefact with several
subsidiary functions to understand how repair can be about correcting
secondary mishaps and not just restoring the proper function. A dull knife
still cuts and a wobbly chair can still be sat on. Repair may well contribute
to restoring the artefact’s proper function, but often it also means perfect-
ing a secondary part of the artefact so the performance of that function can
be perfectly accomplished.
In addition, although repair is concerned with the artefact’s future use,
one who repairs does not necessarily know how to use it. The repaired
artefact can be used as a subsidiary or as a part of another one. In collec-
tive repair, the purpose of functionality is usually achieved without it being
necessary for everyone involved to know how to use the artefact. For ex-
ample, a team of mechanics can repair a car engine without it being neces-
sary for each of them to know how to drive an F1 car. On the other hand,
to accomplish the repair of a wide range of artefacts, the repairer has to
know whether they have reached their final state by performing a test to
check whether the repair was successful. While it may be tempting to as-
sume that testing requires use, this is not always the case. If the pillars of a
metal road bridge start to become corroded by the salt and moisture of an
aquatic environment, they have to be repaired, although crossing the
bridge is not considered a reliable test of a well-performed repair. The use
of the repaired part of the bridge is not sufficient evidence of the appropri-
ateness of the repair. Here, to be validated, repair perhaps requires an in-
strument to show that the repaired structure now accords with safety
guidelines. This then implies that various types of knowledge are required
to validate a successful repair.
In the above examples, repairing seems to involve a type of knowledge
distinct from the specialised knowledge required to produce the artefact.
Repairing requires knowledge, although it does not have to be the one used
in production, but rather knowing about the artefact’s inner workings. In
the bridge example, theoretical and conceptual knowledge about the
bridge’s corrosion performance in the surrounding aquatic environment is
essential to understand how it should be both monitored and repaired.
Without this knowledge, it would be impossible to repair the bridge and
ensure its safety and structural soundness. For our purposes, specialised
knowledge can be considered either scientific or technological. Knowledge
about the chemical and mechanical corrosion of the bridge’s materials
and structures has a bearing on the need for repair and what to repair.
90 Tiago Mesquita Carvalho

The question arises as to the relationship between the knowledge needed

to repair the bridge and the knowledge needed to make it.

4.3 Defining Repair
Building on Ihde’s extension of Heidegger’s thinking, it is assumed that
many, if not all of the artefacts we possess, are not isolated but instead
interwoven with other systems and infrastructures (Ihde, 1990, p. 32). In
repair, there is often a ceteris paribus principle, that is, the assumption that
the context in which artefacts work is consistent. Repairing rests on the
assumption that the artefact’s technological context remains subsidiary to
its use. To succeed, it needs stability in the world. Our analysis of repair is
based on the assumption that there are no artefacts per se in the lifeworld.
The artefacts and infrastructures that surround us belong to an ‘in-order-
to’ context where they fulfil their function and provide meaning.
Repair is concerned with how to make an artefact work again. If repair
implies knowing how an artefact works, it also implies that the repairer
understands the significance of the artefact and how it fits into the interests
of that everyday user. The point is that an artefact being broken can only
be understood in relation to a lifeworld where artefacts already possess a
normative function which determines how they are supposed to perform
(Houkes & Vermaas, 2010, p. 111). The lifeworld proper to artefact de-
signers, repairers and users is one where intentions, assumptions and skills
are built around what artefacts ought to accomplish. An artefact can exist
in a world lacking minds with propositional attitudes, but the same arte-
fact is only broken by virtue of being made by human beings. Hence, any
given artefact is more than its chemical and physical components arranged
in a given structure. Malfunction, proper function and repair are then con-
ceptually linked as they are dependent on human intentions. One cannot
conceive an ontology of artefacts’ function or malfunction while maintain-
ing that they are mind-independent or that our concepts do not have a
bearing on what they are (Baker, 2009, p. 89).
Furthermore, in line with Ihde, the ambiguity of each artefact is re-
vealed according to their contexts of use and the spatial and temporal
constraints of each lifeworld (1990, p. 69). This means that the ambiguity
of any artefact stems from the way its current proper function is depen-
dent on its social and cultural history.2 This history hinges on several con-
tingencies that blur the division between proper and non-standard
accidental functions, while simultaneously providing them with a mean-
ing. Every artefact is subject to a hermeneutic process that secures, extends
or modifies the designers’ original intentions. And this, as Ihde notes, is
because a technological object, whatever else it is, becomes what it ‘is’
through its uses by being brought into the range of praxis (Ihde, 1990,
 There, I Fixed It! On the Status and Meaning of Repair 91

pp. 70–71). A philosophical treatment of repair must therefore recognise

how it plays out with the ambiguity of technology in general and how
malfunction is the nexus where accidental functions can acquire relevance.
Repair is never about the object as an object, but about what it is used for
and its practical purposes in the user’s lifeworld. An instrumental and
metaphysical view of artefacts which understands the use, behaviour and
repair of an artefact to be deduced from the physical and material proper-
ties of its components does not suffice. Every artefact has a social and cul-
tural element, and without considering such elements, artefact function
cannot acquire the usual everyday meaning in daily practices. The artefacts
of the built environment are therefore not just matter arranged according
to scientific principles and technological procedures. As Ihde points out,
they enter the everyday lives of users through immediate, sensuous and
perceptual contacts of the body that foster or curb disclosing possibilities.
The artefacts in the lifeworld become interpretable while manifesting their
multistability3 (Ihde, 1990, p. 144), subject to various interests.4 It is
through them that, as users or consumers, we can identify malfunctions.
It is also useful to distinguish two senses of perception, a sensory one
and a cultural or hermeneutical one. Imagine two subjects A and B, who
while returning to their respective parked bicycles, find that each has a flat
tire. While subject A considers repairing the puncture with a patch, subject
B reads the situation as requiring a trip to a bicycle shop. The micropercep-
tions are the same, but the macroperceptions are different. These two
senses of perception both belong to the lifeworld (1990, pp. 29–30). A sen-
sory embodied impression about what malfunctions in an artefact is inter-
woven and linked with a hermeneutic and cultural perception of its
meaning. The relationship between these two perceptual features is not
one of derivation, but one of figure-to-ground. Any microperception needs
a hermeneutic-cultural macroperception field to occur and there is no mac-
roperception without its microperceptual foci. The interaction between
these two perceptual features makes it possible to understand issues linked
to the transfer of artefacts from one culture to another, as well as to under-
stand why the repair of any artefact depends on the meaning it acquires in
a certain culture.
This distinction allows one to see why an artefact’s proper function is
necessary but not sufficient to understand the reproduction of artefacts
inside a culture and their various transformations. One needs to recognise
that the designer’s intentions regarding artefact function are only part of
what defines them. There is a whole field of artefact design that is usually
unintended, as well as interactions of the intended proper function with
social and cultural realities. The purposes of artefacts are often found by
trial and error and cannot be removed from human intentions. In the same
way, assuming that the meaning of a literary work lies in its author’s
92 Tiago Mesquita Carvalho

intentions is a fallacy (Ihde, 2008, p. 51), the idea that one or more design-
ers are responsible for a given artefact’s future uses or proper function is
also a fallacy. This would be analogous to the idea that readers must sub-
scribe to a single interpretation of a given literary work. Such a hermeneu-
tic game does not ignore that there are intentions underlying texts and
artefact design, but when the texts and the artefact move on to practical
contexts, these intentions end up interplaying with users’ intentions.
Against the deistic and individualistic claims of the designer who creates
a form with a single purpose, the material plasticity of artefacts subjects
them to a lifeworld constituted by macroperceptual intentions of users
with their own cultural history. Artefacts should not be seen as conditional
on transparency and control (Ihde, 2008, p. 58). Design, and repair for
that matter, should be regarded more as experimental and material inter-
rogations of possibilities that may result in what is desired and expected.
Examples in the history of technology abound. The phonograph, the email
(Nye, 2006, p. 44) or the Minitel (Feenberg, 1999, p. 121) have all slipped
from their authors’ original intentions and were then reconfigured by both
users and designers. An accidental function became a proper function once
it merged into social reality.
These examples reveal how multistability is the result of several human
and social processes and their interaction with changing interests and con-
texts. Artefacts evolve because of unplanned contingencies of use, driving
the complexity of relationships between designers and their own cultural
and geographical contexts. Consequently, the original proper function is
often subverted, adapted, diminished or even ignored. An artefact is never
completely transparent to its designers, as it contains an indeterminate set
of unintentional possibilities inscribed in its physical structure. More gener-
ally, the relativisation of the designer’s intentions vis-à-vis how a given ar-
tefact should work highlights the heuristic value of flaws, failures (Petroski,
1996, p. 89) and users’ interpretations, ultimately promoting new trajecto-
ries and practices of material culture.
To understand both repair and design, one must then account for a the-
ory of culture that recognises how artefact functions are subject to the in-
terplay between practices, materiality and the intentions of both users and
repairers. Authors with an analytic background, such as Vermaas and
Houkes (2003) and Preston (2003, 2009), have examined non-intentional-
ist etiological theories about the history of selection and reproduction of
artefact functions and their similarities with biological functions. Ihde’s
viewpoint seems to be in line with Preston’s (2003, p. 603) critique of
Vermaas and Houkes (2003) concerning the insufficient role played by de-
signers’ deliberations and how proper functions often start out as acciden-
tal or system functions. Intentionalist theories of function downplay the
history of an artefact’s uses and contexts and place the onus of function on
 There, I Fixed It! On the Status and Meaning of Repair 93

designers. Like Preston’s example of marigold beer, the designer’s hopes,

expectations and intentions seem to be the criteria that qualify a repair as
acceptable or deviant if it restores proper function or introduces an acci-
dental function. On the other hand, users too have a deliberative history
(Preston, 2003, p. 606) that repair often helps to realise, given that use it-
self is creative. We argue that it is the suitability of users’ intentions in their
lifeworld that ultimately grounds the adequacy of repair and what counts
as a function. Users have desires, wishes and expectations concerning what
the artefact should accomplish. The reproductive and selective processes of
artefact evolution depend not only on designers and their intentions but
also on organised patterns of macroperceptual practices.
Artefacts, by belonging to a culture and the users’ lifeworld, then be-
come alive. They are quasi-living entities, focal points of cultural and
technological networks in which function and meaning are reconfigured
through breakdowns, repairs, design, collaborative participation and im-
provements that may stimulate the development of communities and skills.
Many times, an accidental function is instituted through a hermeneutic or
material reappropriation of users that will have a bearing on the designer’s
intentions. This allows repair to be conceived as exploring and establishing
new uses and stimulating new practices in which individual creativity may
make an artefact functional according to the meaning it acquires in any
given culture.
Repair must then restore a function to artefacts and while such a func-
tion may be the proper one, this is not necessary. It can be about assigning
a function that the artefact already allows or that some minor alteration
permits. Repair discloses the artefact and lingers in the oscillation between
restoring its proper function or modifying it to another one. Hence, com-
parative exercises about the uses of an artefact and how it is repaired in
distinct cultures or socio-economic contexts are more than an exercise of
curiosity. It has an analytical scope for the concepts used in the philosophy
of technology and engineering. The history of selection and reproduction
of a given artefact shows the instability of its proper function and the lim-
ited role of the designer’s intentions. But it also displays the limits of the
idea that there can be a pure or neutral artefact, that is, an artefact that
does not carry the designers’ intentions and the macroperception about the
meaning of its function. The adoption of a particular transferred artefact
depends mainly on its multidimensional ability to fit into an existing prac-
tice (Ihde, 1990, p. 126).
Most often, an artefact is transferred without the technology, that is, the
macroperception that assigns meaning to the artefact, just as when a text
is transferred between cultures and ages its reading becomes different.
Thus, the transferred artefact engages with its context of use, becoming
‘what it is’ by regularly fulfilling a function and acquiring its meaning in
94 Tiago Mesquita Carvalho

relation to such a new context. A ready-to-be-scrapped automobile in af-

fluent societies is, in other cultural contexts, a source of materials or may
even be considered repairable, for example, if converted to an animal
trailer (Houkes & Vermaas, 2010, p. 110). A commonplace glass Coca-
Cola bottle, as portrayed in Jamie Uy’s 1980 film, The Gods Must Be
Crazy, appears as a single, detached artefact for an isolated tribe in the
Kalahari Desert. But soon, notwithstanding its weirdness, it starts per-
forming functions already of existing artefacts.
We can now broaden our earlier preliminary definition of repair. What
we have seen so far has revealed how repair should not be conceived as
restoring the proper function of an artefact, but instead about restoring the
meaning a particular function acquires in the user’s lifeworld. Mechanics
like Willie (Harper, 1987, pp. 19–21), for example, practice repair not only
by restoring artefacts to their initial states but also by making modifica-
tions that go far beyond the designer’s intentions and fit his client’s stan-
dard of living. The tradition of repairing Japanese porcelain used in tea
ceremonies with golden lacquer, known as Kintsugi, is another example.
Such porcelains are valued for having traces of their fragility and display-
ing a history of use. The goal of repair is bound to show imperfections
while ensuring functionality. The golden lacquer mends, but also becomes
an integral part of the bowl. Thus, within the tea ceremony macropercep-
tion, the artefacts acquire an exquisite aesthetic value that ordinary pottery
cannot. On the other hand, as Spelman details (2002, pp. 17–20), a paint-
ing restoration is a repair that aims not to remove but to reverse the effects
of time. That is, repair does not only have to restore a function but also to
safeguard the integrity of the artwork. In the case of a painting, repair is
personalised but it should be minimal. The steps taken matter and may,
according to the context, be considered as damage or not, as having an
aesthetic value or harming the appreciation of the painting.

4.4 Repair and Context

In Section 4.3, we saw how two factors should be considered when defin-
ing repair: the artefact to be repaired and its context, both as a place where
micro- and macroperceptions are joined. Repair and malfunction (Houkes
& Vermaas, 2010, p. 108) are highly context-specific and, hence, the
knowledge of how to repair something is mostly acquired in each context.
Existing tools, parts and materials are needed that assist, hinder or enable
the success of the repair. The judgement about what is exactly malfunc-
tioning and what is to be done with a broken artefact token cannot be
abstracted from the in situ context in which it ought to work again.
Repairing an artefact token is neither pre-given nor identical to a universal
end state of a given type of artefact.
 There, I Fixed It! On the Status and Meaning of Repair 95

Repair takes place when the contextual conditions surrounding an arte-

fact token provide the background context in which perceptions, judge-
ments and actions become appropriate. This can be represented by a
technical syllogism that qualifies what is a good repair and depends on the
knowledge exacted in a given context.5 Abstracting the actions and judge-
ments that compose each context as if they were rules that apply across all
repair activities of a given artefact type may lead to flawed approaches.
Gerrike et al. (2017, p. 107) have advised that applying methods success-
fully depends not only on the formal adequacy of the method but also on
the skills and judgements utilised by users in applying it to a particular
context. However, the conditions of an artefact’s context do not always
reveal from the outset what can be repaired and what counts as repaired.
Often, there are no appropriate tools or materials, which can lead to inge-
nious solutions. In such cases, a depleted context does not hinder repair,
but rather fosters a more hands-on creative activity closer to tinkering and
invention: informal repair. This is how we often encounter a broken arte-
fact in everyday life. Its analysis is also phenomenologically rewarding.
Furthermore, we often observe, especially in more marginal areas of the
planet, customised and patched artefacts that reflect their context (Smith,
2010, p. 1). These make-do repairs are not temporary solutions but rather
reflect the very way in which repair knowledge is manifested and how its
efficiency relates to its context.
Other contextual features surrounding some artefacts are related to their
importance for everyday social reproduction, such as in the case of routine,
preventive or life-cycle maintenance. Repair can be an ongoing process
throughout the life span of an artefact or technological system. Sometimes
repair work never really restores a system to the much-envisioned state of
its proper function. In the rural countryside of Portugal, for instance,
houses are seldom finished. They are often inhabited while being repaired
or upgraded existing next to the building materials in a state of stasis, de-
pending on the next capital input. A final solution may not even be avail-
able and, while a temporary one can be achieved, it is dependent on
variable circumstances.
It may also be the case that successive repairs lead to the replacement of
an artefact’s parts, in what could be another instance of the Ship of Theseus.
Rather than exploring the metaphysical question, however,6 let us look at
examples of how the same broken artefact will be repaired and restored in
different ways in different cultural and economic geographies. Different
macroperceptions give rise to different interpretations of the artefact’s
proper function that go beyond designer intentions. In societies where re-
pair is primarily a professional activity and there is an abundance of tools,
parts and components, it is easy to overlook how the context of repair
dictates what is or is not possible to do. Repair may seem merely a return
96 Tiago Mesquita Carvalho

to the initial proper function of an artefact that appears more or less bound
to an already fixed meaning. In other contexts of greater economic scar-
city, however, repair appears as a more comprehensive approach that ex-
plores the possibility of using different materials and tools as replacements
for defective parts.
Ernesto Oroza, a designer, has examined how artefacts in Cuba are
prone to fluctuate beyond their proper function. He calls this patched work
character of artefacts ‘technological disobedience’ (Oroza, 2009, p. 20).
Since Cubans are subjected to a political regime and an embargo that limits
their access to essential commodities, many have become maker-repairers.
The Rikimbili (Figure 4.1), for example, are two-wheeled adapted vehicles
that utilise chainsaw engines or plastic bottles as fuel tanks. Some are used
as movable, pedal-activated sharpeners or grinders.
In keeping with the analogy between technology and text, we can say
that informal repair often means a partial rewriting of the original mate-
rial inscription given by engineers, designers and past users (Latour, 1992
p. 236). As Bruno Latour notes, the very description of what artefacts do
is brought to light once they are removed from their everyday lifeworld

Figure 4.1 A Rikimbili.

Photo by Ernest Oroza. Oroza, E. (2021) Rikimbili. Available at: http://www.technologic
aldisobedience.com/wp-content/uploads/2021/07/Rikimbili-55.jpg (Accessed: 19 June 2022).
 There, I Fixed It! On the Status and Meaning of Repair 97

through an extraordinary event such as a crisis (Latour, 1992, p. 260)

that shatters, breaks or interrupts the usual context of use. Disruptive
possibilities appear that illustrate what Feenberg (1999, p. 221) called
secondary instrumentalisation in which users can assign technology other
meanings.
The point is that examining actual cases of repair illustrates how the
proper function of artefacts is not fixed, but frequently interpretable.
Similarly, after breaking down, the postponement of repair can reveal all
the social relations implicit in that artefact as a way of life. This is precisely
Winner’s argument (1977, p. 331). Breakdowns and dismantling could fos-
ter an intended research programme, an ‘epistemological Luddism’ that
can have high educational value by illuminating the ‘buried substance’ of
our civilisation: how artefacts are more than a means to an end and how
human relations are transformed by their presence. According to Winner,
repair is a confession of our technological addiction that causes severe
withdrawal symptoms when consumption is interrupted. The symptoms
could, however, provide an explanation for how the structure of human
relations is determined by the technology in question, thus making it pos-
sible to justify technological choices.
Returning to the analogy between technology and text, informal repair
is in many ways similar to the practice of a medieval copyist monk tasked
with transcribing an ancient yet incomplete manuscript of a work they had
once read. Because the text has gaps, the monk has to ‘repair’ it in the act
of copying. Likewise, any artefact may contain scripts of action and per-
ception (Verbeek, 2006, p. 1113) put forward by engineers and designers
and interpreted by its users. The artefact may present defects relative to
how it usually performs, just as the medieval monks’ attempt to be as faith-
ful as possible to the original meaning encountered through previous con-
tact with the text. He cannot, however, avoid introducing certain expected
words or terms where they are absent. The final meaning of the manuscript
may or may not be preserved, but the text has undergone some changes.
While copying, however, he notices, like the translator of José Saramago’s
novel ‘The History of the Siege of Lisbon’ (1998), that his practice allowed
him to rewrite a manuscript, allegedly changing its meaning by introducing
a simple ‘not’.7

4.5 Repair Knowledge
How does someone who repairs proceed to be successful? We have al-
ready noted that repair is a knowledge-driven activity, but what kind of
knowledge is employed and how does it relate to the activity? In many
cases, knowledge of repair seems to be acquired in the field, through pro-
longed contact with the artefact and its context. Someone who repairs
98 Tiago Mesquita Carvalho

seems to possess a knowledge distinct from others, a skill by which they

know exactly how to apply that knowledge through a technical syllo-
gism. Nevertheless, this seems to have been changing for several decades.
Informal repair linked to the ‘shade-tree mechanic’ has faded. A case in
point is modern bicycles.8 With the advancement in technology, bicycle
maintenance and repair have become increasingly more complex and are
now rigidly controlled by manufacturers in an attempt to constrain inter-
changeability between parts from the same artefact type. The solution to
a repair problem is becoming predetermined from the outset.
Many new artefacts contain either scientific or technological state-of-
the-art knowledge. However, that knowledge does not necessarily include
the knowledge required to repair the artefacts themselves. The scientific
and technological knowledge guiding the design of many artefacts does not
seem to be of the same nature as the practical knowledge required to repair
them. A more general epistemological question that repair raises is whether
certain human activities can be defined by another kind of knowledge than
the propositional knowledge based on laws, rules and procedures. We will
now explore this point.
Informal repair implies a form of perception that does not resort to
rules, procedures, concepts and their formal sequencing but is nevertheless
operative and highly effective. This idea can be traced back to Aristotle
regarding the relationship between phronesis and techne, both intellectual
virtues that although related to what might be otherwise, are also distin-
guished from each other (Aristotle, 2008, NE 1140a 24–30). For Aristotle,
the image of the good is not given beforehand but depends on the circum-
stances surrounding the agent (Broadie, 1991, p. 198). The right action
resists a universal prescription precise enough to be valid in all cases: it is
up to the agent to judge and feel what is right in any particular context.
Although techne requires a universal logos for all bodies – the image of the
final, functional state of the artefact – informal repair concerns knowledge
appropriate to a specific form of making. Whoever makes or repairs some-
thing proceeds with an idea of the final state to be achieved. However, this
knowledge is not enough to know how to achieve that state or how to give
form to the matter or artefact. Experience is needed. There are many simi-
larities between doing and making, but as Mitcham (1994, pp. 131–4)
emphasises, there is a distinction between techne and modern technology,
in that the former is always driven towards particular goods, while the lat-
ter, given the rationalisation of the productive process, allows for mass
production and procedural-like rules. The major difference between re-
pairing and science is akin to the contextualism of phronesis and techne. It
lies in how the repair skill based on the knowledge required to fix certain
artefact tokens does not imply universal knowledge about all artefacts of
that type which hold regardless of time and place. The erosion of time,
 There, I Fixed It! On the Status and Meaning of Repair 99

wear and tear from use, and the flexibility with which users can interpret
the artefact, alters both its meaning and what counts as its ideal function.
Whereas the natural sciences apprehend a body and its behaviour as an
example is generally framed in a lawful pattern or rule valid for all times
and places. Although there are analogous cases, each informal repair deals
with singular individuals not with examples of artefact kinds.
Thus, put briefly, if techne is analogous to art and the personal element
in making something is lost in mass production, informal repair still seems
to retain an element of being in touch with concrete materials and singu-
larities and with the contingencies of each poietic process. Processes of re-
pair, like craft, are never identical but involve the discovery of what is
appropriate. The craftsman and the repairer have to let the artefact speak
to their senses in order to work with it. Similarly, to become good practical
agents, we have to live with others to ‘pick up’ and emulate how to act
properly while understanding why we do it (MacIntyre, 1999, p. 69). The
conclusion is that there appears to be a non-conceptual element (Dreyfus,
2002, p. 398) in both techne and phronesis since concrete circumstances
are what matter in determining the criteria for what to do or what to
make. In its similarity to the contextual qualities of phronesis, repair is
thus able to take on the features of a ‘practical labour’ that pays attention
to the local features where it takes place, with the ‘concrete and actual re-
alities of its natural setting’ (Harper, 1987, p. 20).
Despite the attraction of rigorous formalisation based on mathematical
and scientific knowledge, the above considerations seem to prevent repair,
much like techne and phronesis, from being formalised in a decision theory
capable of identifying optimal operative actions. A step-by-step procedure
cannot be determined in advance as it depends heavily on the particulars
of each item, just as attempting to create a science of ethics or a science of
repair seems futile, although the general structure of each can be identified
(Aristotle, 2008, NE 1094b11). The skill of repair appears to grow on a
‘knack for’ or a ‘feeling’ which employs knowledge that is only revealed in
the action itself, that is, in the practice of repair. It is therefore an example
of both tacit knowledge and connoisseurship (Polanyi, 1962, p. 54). Each
repairer knows more than they can say and tends to express it inaccurately.
Harper defines Willie’s intuitive knowledge along similar lines to Polanyi’s:
an embodied, imaginative and explorative action that is attentive to sound
and tactile clues and signals and that is familiar with the nature and behav-
iour of the shape, hardness, plasticity and resistance of the materials that
make up the type of artefacts he repairs. To some extent, this is a more
refined, objective knowledge. The repair person must have a deep under-
standing of the materials of the artefact, as well as an operative knack for
the laws of physics and engineering that govern its operation, an old ‘mé-
lange of hand and theoretical knowledge’ (Harper, 1987, p. 20).
100 Tiago Mesquita Carvalho

Science, guided by truth and ideals of value-neutrality, has a fundamen-

tal criterion for judging the adequacy and justification of explanatory theo-
ries. However, the criterion of efficiency that guides the production and
design of technological objects allows for a whole variety of cases that
satisfy the criterion of functionality, of what really works. What makes a
technological proposition successful is not its truth value, but its adequacy.
The same problem admits different interpretations of what counts as ‘be-
ing solved’. The multistability of artefacts is demonstrated in repair and
the fact that its procedure cannot be put in binding formulas and rules.
Moreover, context dependency hinders the possibility of constituting a
heritage of repair practices.
The methodical observation and idealisation of reality in mathematics
and science are codified in laws and formulas. They foster the historical
construction of a cultural and scientific heritage with technical potential
that grows from generation to generation and becomes continuously acces-
sible as both truthful and operational. In mathematics and science, what is
new is built upon old premises and discoveries which do not lose their
relevance and application. On the other hand, the skill of informal repair
implies an original personal contact with the meaning of artefacts and a
heuristic understanding of how they work without necessarily appealing to
first principles. This has many similarities with how actual design proceeds
according to the description offered by Bucciarelli (1994, p. 18). Design and
engineering are not just applied science (Boon, 2006, pp. 29–30, Kant &
Kerr, 2019, p. 687). By exploring the question ‘What does this artefact do?’,
repair becomes akin to a way of thinking and acting in engineering design
by which the artefact is a living element of a cultural matrix.
In informal repair, artefact and mind come into play through a herme-
neutic game of responses and counter-responses, projections and explana-
tions that form an arc of intentionality about what remains to be done in
the technical syllogism. Through the repair attitude, the artefact is made
personal, even if it has been designed according to mathematical formal-
ism. A set of models, concepts and principles of scientific or technological
knowledge can nevertheless be helpful. They can be included in a herme-
neutic game through the tokens they instantiate and the way they are em-
bedded in particular components. It is moreover to be expected that the
precision of concepts or instrumental measurements should assist the re-
pair of complex mechanisms, as in the case of the bridge discussed in
Section 4.2.
However, in order to be applied, these concepts must be translated into
concrete particulars. Formal knowledge at the level of concepts, laws and
principles has to be converted into the technical syllogism of repair and by
which the particulars, whether they are screws, bolts, metal plates, batter-
ies or electrical circuits, become contextualised and operative. That is, the
 There, I Fixed It! On the Status and Meaning of Repair 101

in situ personalisation intrinsic to tacit knowledge comes into play if it is

to be effective and acquire a vital form. This is how the gap between ab-
stract knowledge and personal, tacit and concrete knowledge is bridged.
Repair knowledge is therefore personalised knowledge. It is built on the
basis of stories, mistakes, attempts, experimentation and readings about
what is going on. This accords with the traditional epistemological distinc-
tion between the propositional knowledge of ‘knowing that’ and the praxi-
cal knowledge of ‘knowing how’ (Vincenti, 1990, p. 3; Ryle, 1949, p. 441).
Repair, as mentioned previously, blurs such distinctions because knowing
that an artefact obeys certain principles and laws is sometimes needed to
account for use (Houkes, 2006, pp. 103–4) or to test whether an artefact
is working. Propositional knowledge can also point to clues and hints use-
ful for tracing a repair hypothesis. Knowing how thus involves the means
by which truth is made operational and even useful.
Nevertheless, artefacts today are more likely to become obsolete due to
the increased applicability of natural scientific knowledge affecting innova-
tions that crawl into the built environment of everyday life. As old arte-
facts become replaceable and are less likely to be used and repaired, they
become relics of past forms of life. The car mechanic today cannot fathom
how the same cars were repaired in the past while Newton’s laws of me-
chanics have not yet lost their relevance.9 The conclusion is that for societ-
ies bent on reaping the fruits of scientific and technological knowledge,
each age has its own forms of repair, as artefacts are either replaced or
improved, while craft, maintenance and repair practices are obliterated,
precisely because repair knowledge cannot be accumulated or gathered in
the same way or pace as scientific and technological knowledge. A scien-
tific or mathematical discovery, on the other hand, does not remain bound
to its historical origin, but rather projects forcibly beyond it and is diffused
into the artefacts of the world of repair. Repair always remains attached to
its age, but on the condition that the renewal of repair skills keeps up with
the transformation of the artefacts to be repaired.
There are then two distinct procedures for the accumulation and trans-
mission of knowledge based on the distinct features of theoretical and
practical knowledge. Informal repair practices have a testimonial charac-
ter. Like phronesis, practical reasons are context-specific and depend on
personal transmissibility due to their having to be learned in concrete and
particular situations. Since they are prone not to be formalised, they are
more fragile and vulnerable. The transformation of the traditional work-
shop of old and the practices it harboured was precisely one of the points
in Harper’s study (1987, pp. 20–24). The separation between the workers
and the knowledge surrounding their work is increasingly giving way to
formal repair. The division of labour and its rational organisation also
achieve the unified and continuous character that unites former crafts and
102 Tiago Mesquita Carvalho

informal repair and in which, as we have seen, a living, personal intelli-

gence is attuned to circumstances, materials and tools. The knowledge of
informal repair becomes more and more external to the worker, and the
condition for this externalisation is its conservation through procedures,
rules and formulas.
With the rationalisation of repair, there is however an epistemic trans-
formation. It ceases to be a hands-on experience and becomes more theo-
retical and detached. As a result, certain tasks become routine, specific and
predictable and, therefore, less dependent on the lengthy acquisition of
practical knowledge. The solutions for repairing something also become
more piecemeal and specific as these practices go through the same process
of rationalisation typical to production. As Harper notes, the solutions of
rationalised repair become more objective, limited to the replacement of
broken parts with functional ones and less ‘intuitive’ and less dependent on
understanding the whole that makes each artefact. Repair becomes a ‘by
the book’ procedure, losing in the process the faculty of the repairer to
refer each fault and each part to the artefact’s unity: ‘the process of repair
has become simple, but the knowledge of the repairman in relation to the
system he or she repairs is very limited. Repair has become deskilled in
much the same way as the work of assembly’ (Harper, 1987, p. 23). The
repairman loses authority over what is repaired because, in a way, he does
not know what he has repaired. Repairing is increasingly seen by consum-
ers as the replacement of parts, granted by procedures, as the authority of
the tacit knowledge of repair loses power. Formal or rationalised repair
and informal or non-rationalised repair then imply different kinds of skills
and competence learning (Harper, 1987, p. 26).
One of Husserl’s points (2008, pp. 64–7) in his Krisis was to show how
the legacy of cultural achievements, such as theorems, concepts and tools,
while becoming widely available for use by everyone, conceal the original
meaning that once led to their creation, severing the living evidence of their
source in the pre-scientific lifeworld. This fosters an obliviousness of the
original meaning, as though the condition for formally transmitting this
objectified heritage over generations is the forgetfulness of the original
evidence. Husserl’s quest thus consists of retrospectively questioning the
meaning of the geometric and formal methodologies each scientist receives,
making them personal and akin to the original meaning. Otherwise, the
meaning of the original evidence is emptied by a formal, methodological
and operative game where only rules must be followed, creating a scientific
life entirely devoted to logical activities, crowned by the success of its prac-
tical application and without consideration for the good life.
Repairing, at least in its informal features, is thus a skill that often
involves a continual search for meaning to be successful. It proceeds by
exploring the purpose of artefact components before any rules and
 There, I Fixed It! On the Status and Meaning of Repair 103

methodological procedures. The dependence of repair on labour is then

not merely an economic question but an epistemic condition for the thriv-
ing of this type of knowledge. This evidence and its comprehension allow
one to demonstrate and communicate to others the ‘how it is done’ of re-
pair through intersubjectivity. This is why repair cannot be taught but only
understood by people who already share the same context.
Following Husserl’s critique, we argue that the practice of repair is not
as prone as mathematics and science to forget the original grounding in the
lifeworld from which all theoretical and practical life proceeds. Like practi-
cal reason, informal repair deals with particular entities in different con-
texts and, thus, can hardly become systematic. Harper’s observations about
how repair is being transformed highlight how formal rationalised repair
inevitably tends toward a procedural theory of replaceability. Although
repair practices are subject to recurrences that obscure and pale in com-
parison to a discovery, they retain the property of never becoming a formal
game. Indeed, every ‘there, I fixed it!’ suggests that the artefact provides
evidence of the technical syllogism, an ‘it works!’ that validates the percep-
tions, judgements and maxims that mean a repair skill has been successful.
Repair may and does in fact work with inherited tools and instruments,
but it cannot inherit abstract knowledge and concepts without translating
them into embodied practical knowledge.
This is one of Polanyi’s most important points (1962, p. 53) made in his
defence of the value of epistemic traditions. Motor or cognitive practices
based on tacit knowledge, such as informal repair skills, resist being sum-
marised in rules and procedures not only because knowing them as rules
does not ensure successful performance but also because these procedures
are always insufficient. The possibility of transmitting and accumulating
knowledge internal to these practices becomes restricted since this knowl-
edge can only be transmitted by example, between master and apprentice
or in convivial settings (Illich, 2009, p. 11). The apprentice only becomes
capable when they begin to make their own judgements. Both master and
apprentice will then be aware of what is correct or incorrect in their activ-
ity without being able to say precisely why. There are striking similarities
with the development of practical moral reasoning as a skill (Vallor, 2015,
p. 109; Coeckelbergh, 2019, p. 271). The drawback is that much of the
tacit knowledge linked to skills, such as those found in arts and crafts,
against a backdrop of accelerated changes in the material environment, is
prone to being lost.
Following Kant and Kerr (2019, p. 699), informal repair knowledge is
then essentially a know-how that goes beyond the know-that about the
artefact’s physical features. It includes: 1) procedural knowledge about a
sequence of actions and 2) operational knowledge to go through the par-
ticulars, an embodied practice that involves interventions aiming for an
104 Tiago Mesquita Carvalho

end state. Following such observations and based on the knowledge type
proper to repair, some conclusions can be drawn. A first, somewhat
common-sense assertion is that, although an artefact can be made either
through craft or industrial mass production, there can be no industrial
mass repair. Moreover, whether artefacts are handmade or industrially fab-
ricated, repair can be expected. However, in the former, repair knowledge
remains generically in the makers’ skill set, whereas, at the other end of the
spectrum, there are differences between producing and repairing in terms
of the intensity and the type of knowledge required. Formal repair, never-
theless, can be gradually expected because advanced everyday artefacts
that compose our modern lifeworld – cars, aeroplanes and network
systems – experience recurrent malfunctions requiring troubleshooting
manuals with rules and procedures to identify breakdowns and swiftly in-
dicate replacement of malfunctioning parts.
Regarding the technical knowledge used in production, it is perhaps use-
ful to distinguish general artefact types, such as commodities, and unique
artefacts, such as prototypes, industrial machines, infrastructures or tech-
nologies used for scientific instrumentation and calibration. The former
are mass-produced, fostering a labour market for informal repair and skills
that can, as a rule, ignore the scientific and technological knowledge used
in production. In the case of the latter, a failure might be a flaw in the de-
sign or production process and thus may require the designers themselves,
as possessors of the technological knowledge that supervised their making,
to return to the worktable or to actively collaborate with repairers.
The repair of unique artefacts is then extremely restricted. In these tech-
nologies, there is usually just one token per type and hence their function
must still be restored without a labour market for repair. In other words,
because in such cases repair depends on restoring the proper function, it
may hinge on scientific and technological knowledge and careful method-
ological procedures. Nevertheless, as we have noted previously, repair only
happens when scientific and technological knowledge are realised in action
and disposition. Here repair remains simultaneously linked to the major
premise of an abstract understanding, even perhaps to a set of procedures
and steps, but as well as to an empirical application that requires the use
of tacit knowledge guided by theoretical clues. In many cases, this leads to
a division of cognitive labour in which the knowledge required for repair
is accessible only to specialists.
Herein lies much of Winner’s (1977, pp. 330–2) social critique of arte-
facts that increasingly demand ongoing, specialised training to be repair-
able, severing users from engaging with them. This process echoes the
rationalisation of the social order through labour division and the very
design of mass-produced artefacts, making the appearance of informal re-
pair more unlikely and reducing it to formal repair. As a set of knowledge,
 There, I Fixed It! On the Status and Meaning of Repair 105

formulas, rules and procedures, technology has progressively crept directly

into human praxis. However, these examples show that this process is lim-
ited because it is anchored to informal, tacit repair skills supported by an
ongoing reflective conversation between individuals, artefacts and their
context (Schön, 1992, p. 142). Repair may thus be an example of how
some skills resist formalisation and calculation. Their epistemic relevance
appears ironically within the very process of expanding scientific and tech-
nological knowledge to various fields of human activity.

4.6 The Phenomenology of Informal Repair

We now proceed to inspect the basic structure of the informal repair skills
that precede the rationalisation of formal repair. Such a basic structure
may be informed by scientific and technological principles that underscore
judgement and troubleshooting. These skills are exercised in the pre-scien-
tific, intuitable lifeworld in which all life is played out. In such an everyday
world, one does not experience artefacts as if they were geometrical or
ideal figures. The artefacts of our surroundings are experienced in an ac-
tual, usual way, without any need to add the formal possibilities of these
subjects to them. They remain equal to themselves, connected to each other
and far from having an arbitrary behaviour, exhibiting characteristic and
recurrent modes that foster an empirical intuition about their future be-
haviour. The natural attitude takes the surrounding artefacts for granted,
constituting the background context of the lifeworld. An accident or fail-
ure can remove them from this everyday character, inviting users to under-
stand how artefacts work. The surprise that artefacts can break down
triggers an inquiry, marking the beginning of the repair attitude.
If the theoretical or scientific attitude stems from an epoche of the pre-
scientific world, the repair attitude is also a modification of praxis, a ‘going
towards’ that signals the artefact as ‘repairable’, but without abandoning
the lifeworld structure. The world of human purposes linked to praxis re-
mains the same in the repair attitude. It changes how some artefacts ap-
pear, making them thematic and not just damaged or broken, but also
having ‘possibilities for’ as objects that can be combined, assembled and
disassembled. For the repair attitude, there is no point in objectively know-
ing the artefacts unequivocally and universally as in the way mathematical
or physics practices approach their objects. It is enough to deal with the
artefacts in how they appear to experience in a regular and empirical style
of a general causality with its own essential structure based on the familiar-
ity of the surrounding pre-scientific lifeworld (Umwelt). This general cau-
sality of the lived, somatic body has its usual truths that suffice for repair
and for which the ‘true being-in-itself’ of the natural sciences don’t add
much. As Husserl says, our kind of practical knowledge of things as they
106 Tiago Mesquita Carvalho

are for us is perfectly adequate for human purposes and is open to other
transformations that possess unity in themselves (Husserl, 2008, p. 352).
The first point to note is the way in which the repair attitude regards
artefacts, that is, seeing them according to a problem interpreted in light of
the artefact’s intended meaning. One should not lose sight of the fact that
every artefact, broken or not, may hold other meanings. Artefact function
is inextricably linked to meaning, but they are to be considered different.
A broken chair may become a source of firewood and a broken stereo may
become a bookend. Meaning is related to the users’ lifeworld and how the
material environment grants a temporary or permanent function to attain
that meaning. The interpretation of an artefact as being ‘broken’ is also the
moment when meaning swings and can be pushed to other final states. The
temporary suspension of the usual function of things draws limits to tech-
nological determinism (Feenberg, 1999, p. 78).
Through the repair attitude, everyday artefacts, despite having their
proper function, now appear with various functional shades and various
‘possibilities for’ that can be explored. Adopting Uexküll’s terminology
(1988, p. 113), the funktionale Tönung, the shading of the aspects and
qualities of artefacts takes on a character of ‘manipulable by tool’ or ‘re-
placeable by another part’ that other attitudes do not grasp. Each part or
tool now has the shade of ‘serving for something’ subsidiary to the desired
function. Those who repair do not shade the parts of the artefact as objects
of scientific knowledge or aesthetic appreciation but rather as objects that
may or may not be manipulated by certain tools and have parts that can be
removed, replaced or adjusted.
The connection between the artefact and its context, that is, between the
artefact and the available tools, parts and the macroperception, forms the
action circle of the repair attitude, composed of perception marks and ac-
tion marks. For example, the availability of a spare part reveals that repair
by replacement is achievable. Similarly, the availability of a tool makes it
possible to perceive the manipulable character of other aspects of the arte-
fact. Perception and action are not only pre-existing logical or physical
possibilities but also possibilities that stem from the world specific to the
repair attitude and the repair skill.
Tools and spare parts are also subject to interpretation according to the
repair attitude for a given class of artefacts. Spare parts have a specific use
content, that they ‘work like’ the component to be replaced, and tools have
a ‘coupling content’ regarding the artefact. For example, when we want to
sit down, we do not have to look for a chair but for a surface that has a
‘seating content’. These shading possibilities contribute to establishing the
difference between what is possible vis-à-vis what is feasible,
We can now see how the repair attitude seeks replacements for an
‘intended shading’ that can fulfil the same function of a broken part or an
 There, I Fixed It! On the Status and Meaning of Repair 107

unavailable tool. Consequently, the difference between repair stricto sensu

and repair as recreation lies in how much of the artefact’s initial state
should be aimed for and what counts or not as a repair solution. A patch,
a make-do or a temporary arrangement restores the artefact to its initial
state, recovering its functionality, but does not restore it equally. The in-
formal repair attitude opens semantic possibilities: the artefact can be re-
paired but it has shifted to another function or even to an object of
appreciation. According to its intended meaning and contextual possibili-
ties, an artefact undergoes an intervention that can range from stricto
sensu repair, improvised repair, creative repair or recreation. Repair can be
creative. The repair attitude thus reveals how an artefact’s meaning is
available to be fixed. A damaged mountain bike can be repaired to make it
work again as a city bike, but it can also be transformed into a motor ve-
hicle or even a coffee bean mill. In the latter case, the meaning of the arte-
fact has changed.
We can now try to summarise how repair knowledge is acquired and
practised. What distinguishes an amateur repair from a more advanced
one is the former’s inability to limit a large number of hypotheses to an
operative number. In the repair attitude, the learner goes through a period
of discomfort and exploration by trial and error. The problem is identified
and assessed before they can devote their attention to it. Relevant clues,
aspects and patterns are followed until the source of the problem can be
pinpointed with relative certainty. A trial use of the broken artefact is then
conducted so that it can reveal the origin of the malfunction. Hypotheses
and verification tests are put forward to restrict the scope of explanations
and repair steps.
In the early stages of learning repair skills, repairers try to detail each
component’s role in relation to the whole. Each change is followed by ob-
servation of the artefact’s behaviour, paying attention to the way slight
modifications bring about change that is not random, but causally linked
to it. After observing the results, each further change may or not continue
to cause a more satisfactory result. Error, therefore, has a fundamental
heuristic role in improving skill and establishing a good or bad judgement
or action. Only those who are capable of perceptions and judgements
proper to the repair attitude can say why there is an error. To repair is also
to accept the possibility of going back to make amends, in the constant
readiness to discover other issues, in an oscillation between ‘seeing the
whole’ and ‘seeing the detail’. This process manifests itself as a ‘focusing-
defocusing’ structure. With time and practice, for the same artefact type,
repair becomes embodied and a state is reached where ‘one knows what
one is doing’. It means a tenure, a familiarity with the nature of the recur-
ring problems usually affecting the artefact. It marks a state of readiness
to perceive what needs to be done. A skilful repairer has perfected the
108 Tiago Mesquita Carvalho

focusing–defocusing structure to the point where all steps appear to be

predetermined, as though they were just following rules. In reality, a re-
pairer follows perceptions and judgements recognisable as significant. It is
this personal character of the repair skill that appears not only to be a
procedural compliance but potentially describable by an algorithm (Young,
2018, pp. 64–5). As pointed out previously, there is nevertheless a differ-
ence between understanding that repair can be described and explaining it
as if the repairer was following explicit rules (Dreyfus, 1997, pp. 256–7),
mainly because interpretation is involved in the repair attitude. There is no
knowledge of repair without a skill whose mastering is fraught with
doubts, hesitations and dead ends. It is a lived skill marked by an emo-
tional history bound with frustration, persistence and resolution that en-
ables its embodiment. Its reluctance to be spelt out in rules is the counterpart
to not being easily forgotten, as it seems always available to be reinstated.
Moreover, it is possible to assess retrospectively how appropriate a re-
pair may be, not based on the ideal solution for achieving the final state,
but on how that repair has successfully enabled certain options in the
search for a functional solution. An illustrative scene can be seen in the
docudrama Apollo 13. After the moon-bound spacecraft suffered an un-
foreseen explosion, all that mattered was bringing the crew back to Earth.
One scene raised the question of how a carbon dioxide leak could poison
the astronauts. The mission of the NASA ground team was to show the
astronauts on board the Apollo 13 spacecraft how to repair a filter, so they
did not asphyxiate. To do this, the ground team had to reproduce materials
they knew existed on the space shuttle and interpret them as appropriate
for the problem at hand, eventually using duct tape to arrive at the ‘filter
purifier’ solution. As flight director Gene Kranz, played by actor Ed Harris,
previously states: ‘I don’t care what anything was designed to do. I care
what it can do’ (Apollo 13, 1995). The technical syllogism of the repair
enacted on Earth was then described in a set of schematic procedures and
transmitted to the astronauts by radio. They reproduced the instructions
without ever understanding the meaning of what they were doing. There
was never any question of whether another solution was possible, as
though the ideal solution could be different from the only solution the
ground team was able to find under those circumstances. In short, repair,
like any practice, presents context-specific criteria to define the solution
that works but which, nevertheless, can define what counts as a good or
bad repair.

4.7 Conclusion
This chapter advanced a post-phenomenological framework inspired by
the work of Don Ihde in order to understand what repair is about. In the
first part, this framework was used to question some assumptions in the
 There, I Fixed It! On the Status and Meaning of Repair 109

metaphysics of artefact function by underlining how artefacts are part of

the user’s lifeworld. Designers and repairers must acknowledge how arte-
facts, especially when they break down, are prone to become interpretable.
Artefact function theories should then take into account that what an arte-
fact does hinges on how it is used, according to the meaning it may acquire
in a lifeworld. The difference between the proper function assigned by the
designer’s intentions and the accidental function is blurred as the user’s
intentions and the repairer’s context-dependent character play a role: re-
pair should be understood as an activity that restores a function, but not
necessarily as restoring the proper function. Social and cultural processes
thus have a bearing on the selection and reproduction of artefact function.
Consequently, cultural differences and technology transfers gain philo-
sophical relevance: other social relationships, interests and institutions in-
teract to define what exactly repair restores.
Afterwards, an epistemology of repair skill was advanced in order to
outline its relationship to different types of knowledge. The skill of infor-
mal repair was analysed and contrasted with the emerging formal repair in
the context of progressive automation and rationalisation of productive
processes. Some conclusions were drawn from the likelihood of informal
repair turning into a formal, procedural-like knowledge given the advance-
ment of labour division and work organisation. Additionally, a phenome-
nology of the structure of informal repair was sketched: it highlighted how
its knowledge is akin to current practices of design and engineering, allow-
ing one to see how scientific principles and laws are not necessary for every
repair practice but are nonetheless apt to complement it. Informal repair
rests on a translation of scientific and technological propositional knowl-
edge into practical, useful knowledge.

Funding Information
This publication is funded with National Funds through the FCT/MCTES –
Fundação para a Ciência e a Tecnologia/Ministério da Ciência, Tecnologia
e Ensino Superior (Foundation for Science and Technology/Ministry for
Science, Technology and Higher Education – Portugal), in the framework
of the Project of the Institute of Philosophy – University of Porto with the
reference UIDB/00502/2020.

Notes
1 For Vermaas & Houkes, any theory of artefact function should be able to dis-
tinguish between its proper and accidental function: ‘Proper functions can typi-
cally be understood as functions ascribed standardly to the artefact, whereas
accidental functions are ascribed only occasionally’.
2 This does not excuse the fact that prototypes or highly specialised artefacts
are more rigid or more stable regarding their capacity to acquire accidental
110 Tiago Mesquita Carvalho

functions. But as Robert Smithson’s ‘A Tour of the Monuments of Passaic’

shows, any artefact can become something else under an aesthetic ‘seeing how’.
3 Ihde (1990, pp. 144–6) puts forward this term as both a way to question a
single or unified trajectory of Technology and its autonomy. Technologies are
ambiguous and have a variety of embeddedness across cultures. The polymor-
phism that visual perception has regarding some objects like the duck-rabbit or
the Necker cube is expanded by phenomenological analysis to highlight how
objects are also multistable in the relationships that humans establish with
them.
4 In an overview of philosophical theories of artefact function, Preston (2009,
pp. 214) notes how such theories should account for how artefacts are multiply
realisable and multiply utilisable and that such notions are two sides of the
same coin.
5 A technical syllogism is the description of a practical judgement about what is
at stake in a certain artefact and what needs to be done and can be done in that
context. It’s important to note that a technical syllogism is not a claim about
how in fact human beings think and proceed when dealing with artefacts (for
another view, cf. Dreyfus, 1997, p. 233). Its aim is to provide an explanation. It
consists of a major and a minor premise followed by an appropriate action in
case both premises hold true and possible, respectively. In the case of repair, the
major premise contains the general end of restoring a function to the artefact or
a step towards it. The minor premise is causally linked to the major premise as
a particular means and takes into account the circumstances of the context in
which the agent finds herself. The skill of a repairer shows itself upon identify-
ing the relevant features of the minor premise. Deliberation about how to pro-
ceed thus ceases when the principle of action is integrated into the agent. The
conclusion of the syllogism is the principle of the next step of reparation.
6 For further discussion on the metaphysics of artefact persistence, see Young
(this volume).
7 This ‘not’ completely shifted the historical meaning of the siege of Lisbon by
stating that the foreign crusaders have ‘not’ helped the Portuguese in conquer-
ing the city to the moors after all. Saramago’s point was to illustrate how his-
tory’s meaning is open to be interpreted through the very same way used to
settle what were the facts.
8 Lindsey, J. (2022). ‘Are modern bicycles the end of DIY maintenance?’ Outside
Online, 8 November. Available at: https://velo.outsideonline.com/gear/modern-
bicycles-the-end-of-diy-mechanic/ (Accessed: 9 November 2022).
9 What we mean is that for a mesological, everyday use, Newton’s laws continue
to be useful for mechanics and kinematics without the need to add more nu-
merical precision.

References
Aristotle (2008) Ética a Nicómaco A. C. Caeiro (trans.) Lisboa: Quetzal Editores.
Baker, L. (2009) ‘The Metaphysics of Malfunction’, Techné, 13(2), pp. 82–92.
Boon, M. (2006) ‘How science is applied in technology’, International Studies in
the Philosophy of Science, 20(1), pp. 27–48, https://doi.org/10.1080/026985
90600640992
Broadie, S. (1991) Ethics with Aristotle. Oxford: Oxford University Press.
Bucciarelli, L. (1994) Designing Engineers. Cambridge: The MIT Press.
 There, I Fixed It! On the Status and Meaning of Repair 111

Coeckelbergh, M. (2019) ‘Skillful coping with and through technologies’, AI &

Society, 34, pp. 269–287, https://doi.org/10.1007/s00146-018-0810-3
Dreyfus, H. L. (1997) What Computers Still Can’t do. A Critique of Artificial
Reason. Cambridge: The MIT Press.
——— (2002) ‘Samuel Todes’s account of non-conceptual perceptual knowledge
and its relation to thought’, Ratio, 15(4), pp. 392–409.
Feenberg, A. (1999) Questioning Technology. London: Routledge.
Gerrike, K., C. Eckert and M. Stacey (2017) ‘What do we need to say about a de-
sign method?’ In: 21th International Conference on Engineering Design (ICED
2017), 21–25 August 2017, Vancouver, Canada.
Harper, D. (1987) Working Knowledge. Skill and Community in Small Shop.
Chicago and London: The University of Chicago Press.
Henke, C. and Sims, B. (2020) Repairing Infrastructures. The Maintenance of
Materiality and Power. Massachusetts: The MIT Press.
Houkes, W. (2006) ‘Knowledge of artifact functions’, Studies in History and
Philosophy of Science Part A, 37(1), pp. 102–113.
Houkes, W. and Vermaas, P.E. (2010) ‘Malfunctioning’ in Technical Functions.
Philosophy of Engineering and Technology, vol 1. Dordrecht: Springer, https://
doi.org/10.1007/978-90-481-3900-2_5
Husserl, E. (2008) A crise das Ciências europeias e a fenomenologia transcenden-
tal: uma introdução à Filosofia fenomenológica. Ferrer, D (trans). Lisboa:
CFUL.
Ihde, D. (1990) Technology and the Lifeworld: From Garden to Earth. Indiana:
Indiana University Press.
——— (2008) ‘The Designer Fallacy and Technological Imagination’ in Philosophy
and Design. From Engineering to Architecture. P. Vermaas, P. Kroes, A. Light,
S. Moore, (eds.) Dordrecht: Springer, https://doi.org/10.1007/978-­1-­4020-
6591-­0_4
Illich, I. (2009) Tools for Conviviality. New York: Marion Boyars.
Kant, V. and Kerr, E. (2019) ‘Taking Stock of Engineering Epistemology:
Multidisciplinary Perspectives’, Philosophy & Technology, 32, pp. 685–726.
Latour, B. (1992) ‘Where Are the Missing Masses? The Sociology of a Few
Mundane Artifacts’ in W. Bijker and J. Law (eds). Shaping Technology/ Building
Society. Studies in Sociotechnical Change. Cambridge: The MIT Press, pp.
225–258.
MacIntyre, A. (1999) Dependent Rational Animals - Why Human Beings Need the
Virtues. Chicago: Carus Publishing.
Mitcham, C. (1994) Thinking Through Technology. The path between Engineering
and Philosophy. Chicago: University of Chicago Press.
Nye, D. (2006) Technology Matters. Cambridge: MIT Press Book.
Oroza, E. (2009) RIKIMBILI. Une étude sur la désobéissance technologique et
quelques formes de réinvention. PU Saint-Etienne.
Petroski, H. (1996) Invention by design: how engineers get from thought to thing.
Cambridge: Harvard University Press.
Polanyi, M. (1962) Personal Knowledge. London: Routledge.
Preston, B. (2003) ‘Of Marigold Beer: A Reply to Vermaas and Houkes’, The
British Journal for the Philosophy of Science, 54(4), pp. 601–612.
112 Tiago Mesquita Carvalho

——— (2009) ‘Philosophical Theories of Artifact Function’ in A. Meijers, (ed.)

Handbook of the Philosophy of Science, Philosophy of Technology and Engi­
neering Sciences, North-­Holland, pp. 213–233, https://doi.org/10.1016/B978-
0-­444-­51667-­1.50013-­6
Ryle, G. (1949) The concept of mind. Chicago: Chicago University Press.
Saramago, J. (1998) The History of the Siege of Lisbon: A Novel. Mariner Books.
Schön, D (1992) ‘Designing as reflective conversation with the materials of a design
situation’, Knowledge-Based Systems, 5(1), pp. 3–14.
Smith, C. (2010) Design for the other 90%. Cooper-Hewitt.
Spelman, E. (2002) Repair. The Impulse to Restore in a Fragile World. Beacon
Press. Boston.
Uexküll, J. (1988) Dos animais e dos homens: digressões pelos seus mundos
próprios A. Candeias (trans.) Lisboa: Livros do Brasil.
Vallor, S. (2015) ‘Moral Deskilling and Upskilling in a New Machine Age:
Reflections on the Ambiguous Future of Character’, Philosophy & Technology,
28, pp. 107–124.
Vermaas, P. and Houkes, W. (2003) ‘Ascribing Functions to Technical Artifacts:
A Challenge to Etiological Accounts of Functions’, The British Journal for the
Philosophy of Science, 54(2), pp. 261–289.
Verbeek, P.-P. (2006) ‘Tecnopólis: a vida pública dos artefactos tecnológicos,
Análise Social, XLI(181), pp. 1105–1125.
Vincenti, W. G. (1990) What engineers know and how they know it: analytical
studies from aeronautical history. Baltimore: Johns Hopkins University Press.
Winner, L. (1977) Autonomous Technology. Technics-out-of-Control as a Theme
in Political Thought. Cambridge: MIT Press.
Young, M.T. (2018) ‘Intuition and Ineffability: Tacit Knowledge and Engineering
Design’ in A. Fritzsche and S. Oks, (eds.) The Future of Engineering. Philosophy
of Engineering and Technology, vol 31. Cham: Springer.
5 A Standpoint Epistemology
of Repair?
Cristina Bernabéu and Jesús Vega-Encabo

5.1 Introduction
In ‘Rethinking Repair’, Steven J. Jackson proposes an exercise of what he
calls ‘broken world thinking’ (2014, 221). This notion reveals the fragility
of the world we inhabit, which is shaped by a complex set of relationships
between natural beings, social structures, technical infrastructures and ev-
eryday objects. Phenomena such as erosion, breakdown, decay or dissolu-
tion come to the fore, as opposed to the idealistic insistence on innovation
and design. However, rather than passively dwelling on the fate and mem-
ory of ruin and debris, this exercise is grounded in a more optimistic no-
tion. It recognises the potential of fragility to rethink how we collectively
achieve order in the face of erosion and decay. To put it in more cosmic
terms, it is as if we restore order and stability over and over again, conjur-
ing up cosmic forces of destruction with our rituals.1
But fortunately, the discovery of fragility does not push us into this eter-
nal confrontation between the forces of the cosmos. On the contrary, it
leads us back to a forgotten realm, where the silent activities of mainte-
nance, repair and care of those environments take place and where life and
the world in Hannah Arendt’s sense are reproduced (Arendt, 1958). Jackson
integrates all of these activities into a broad concept of ‘repair’, that includes
‘the subtle acts of care by which order and meaning in complex socio-tech-
nical systems are maintained and transformed, human value is preserved
and extended, and the complicated work of fitting to the varied circum-
stances of organizations, systems, and lives is accomplished’ (2014, 222).
The fulcrum of a world that is maintained in its ‘natural’ order, full of mean-
ing for those who inhabit it, seems to be found in the activities of repair.
Jackson also reminds us that behind the neglect of repair and mainte-
nance activities lie issues related to the distribution of knowledge and
power. We have become increasingly aware of how the invisibility of some
activities in society often reveals profound inequalities, particularly in
terms of the value we attach to the potential epistemic claims on the part

DOI: 10.4324/9781003316213-6
114 Cristina Bernabéu and Jesús Vega-Encabo

of those involved in those activities. Thus, Jackson suggests and briefly

outlines the premises of a potential ‘standpoint epistemology of repair’.
This epistemology would first and foremost prioritise breakdown, allow-
ing us to adopt a perspective that reveals what remains invisible in our
everyday existence. In addition, it should endow repair itself with the
potential to unearth those possibilities that have been crushed by the domi-
nance of the forms of fixed objects and artefacts or the stability of socio-
technical systems (Jackson 2014, 230–31).
Repair and maintenance activities are present in almost all of our inter-
actions with the material world in which we live. Those who engage in
such activities participate in shaping our technological worlds. The ques-
tion we will explore in the following pages is the extent to which repairers
can constitute a standpoint on which to ground certain epistemic claims
about significant aspects of our socio-technical reality. Is a standpoint epis-
temology of repair possible? The answer to this question cannot be defini-
tive, and we will point out some of the difficulties we face in attempting to
delineate the collective of fixers from which a standpoint might emerge,
along with the supposed epistemic advantages derived from their social
location and the collective process of mutual recognition. The core of the
argument rests on a broad characterisation of the fixer, which we introduce
in Section 5.2: we are all repairers insofar as we participate in the (re)pro-
duction of artefactual environments. If we identify fixers with a group of
technicians specialised in repair, we are just narrowing a category that en-
compasses multiple agents and multiple aspects. It is then worth consider-
ing how a standpoint can emerge from such a diffused and heterogeneous
figure. We will show that there is no expeditious way for establishing a
specific epistemic privilege of the fixer because there is no way for the col-
lective to be unified by their common needs and interests formed in a situ-
ation of oppression or subordination as standpoint epistemologies often
require. Nevertheless, in the course of this chapter, we will highlight several
aspects that make it still attractive to think about the role of the fixer in
terms of critical epistemological conceptions. First, we will highlight the
invisibility of repair and maintenance activities in our technological cul-
ture, as well as their hierarchical subordination to more visible figures. This
hierarchy makes it difficult to recognise their knowledge claims (Section 5.3).
Second, we will consider how our theories of the technological world
can reintegrate the fixer into conceptions that align more closely with the
technological process and even our everyday experience with artefacts
(Section 5.4). Third, we will suggest that the category of the fixer is associ-
ated with specific forms of experience and practices in the (re)production
of socio-technical environments, in line with the basic tenets of standpoint
epistemologies (Sections 5.5 and 5.6). Fourth, we will recognise that repair
activities can become sites of social critique with transformative potential,
 A Standpoint Epistemology of Repair? 115

as Jackson suggests. However, repair as critique does not necessarily rely

on a mechanism of reversing locations to gain access to a less distorted
view of socio-technical reality, as standpoint epistemologies suggest.
Instead, it stems from an awareness of the fragility and vulnerability of the
material world. Anyone who engages critically as a repairer and main-
tainer, embodying an attitude of care and respect for things, can contribute
to this consciousness. It is a form of consciousness that does not refer spe-
cifically to the social and epistemic conditions of subordination or oppres-
sion, as required by standpoint epistemologies in their more classical
versions, but rather refers to the material world we inhabit (Section 5.7).

5.2 Repair
First, what should we understand by ‘repair’? An appealingly broad notion
of what repair is emerges from Jackson’s previous quote. In our everyday
environments, many artefacts and objects suddenly stop working, and we
tend to attribute this failure to a misalignment of structural elements that
affect their intended functionality (given what they are supposed to do). In
most cases, repair consists of putting ‘things’ back in place, if possible.
However, in many cases, the damage is irreversible and becomes an inexo-
rable and definitive waste of components and materials, and it is advisable
to discard the artefact. Repair can be done by users themselves, who adjust
or replace components, or by technicians who know the mechanisms in-
volved, the expected functioning of the device and the appropriate condi-
tions for its use. In this sense, repair operates within a framework of shared
expectations about the artefacts, objects and technical systems that sur-
round us. It is not crucial for our purposes to delve into how these expecta-
tions are established, whether by designers and manufacturers or through
open and diverse trajectories of use and exchange. Repair is an activity
intimately connected to how we navigate our collective expectations and
encompass various dimensions of our social and cultural life.
Take one of our most common household appliances, the refrigerator.
We have established numerous cultural expectations around it, not only
regarding its efficient operation but also regarding how it should be main-
tained, repaired or replaced. When one or more of these expectations are
violated, we sense that something is wrong with the appliance and that
intervention is necessary. However, we are often mistaken about the sig-
nificance of this breach of expectation. Recently, one of the authors of this
chapter experienced such a breach of expectation when their refrigerator
stopped cooling during a summer heat wave. It appeared to be broken or
at least not functioning as expected. However, as the technician promptly
pointed out, restoring expectations did not require much effort, certainly
not replacing the appliance. We had misinterpreted what was wrong
116 Cristina Bernabéu and Jesús Vega-Encabo

with it: the appliance was certainly not in the best condition, but there
were ways to restore its efficiency with more careful use. Repairing con-
sisted of readjusting the actions and expectations we had for the device
(Henke 2000). Therefore, it is essential to understand what we mean by
shared expectations: not every artefact or technical system entails a fixed
and non-negotiable set of expectations. On the contrary, they depend not
only on the demands of various actors, including designers, manufacturers,
users and especially repairers but also on the intervention capacities avail-
able to us. These expectations are shared in the sense that the reproduction
of our artefactual and technical worlds necessitates – even if only locally
and temporarily – adjustments among some of these expectations. In any
repair context, negotiation occurs regarding the interpretation and signifi-
cance of different expectations.
Once we understand repair from the perspective of the expectations, we
build around objects (artefacts, systems and infrastructures) and their use
(of which the failure of a particular component is only one aspect), there
are many activities that can be considered part of ‘repair’ in this broad
sense. In fact, fixing a broken object is only one possible intervention that
seeks to maintain the expectations we share about the elements of our
material culture and the socio-technical environments we build and in-
habit. There are numerous routine interventions we make to keep things in
a good disposition. We avoid neglecting them and give them the necessary
attention: we clean them, perform routine checks, replace components and
so on.2 A similar experience can be observed with cars. Even if the techni-
cal checks are left to specialised technicians, we become aware of how
frequently we need to intervene to keep the car in order, which is insepa-
rable from our cultural expectations of cars. In most cases, this means an
understanding (however minimally) of their own materiality, that is, of
how their material condition (their mundane existence) embodies their cul-
tural significance. Intervening in them also means arranging them in an
orderly manner, avoiding certain mismatches with other objects and arte-
facts that are necessary for our practices of use.
What we might call maintenance and repair practices emerge from the
flows of sustained activity in the cultural environments in which we live. In
this continuous flow (what phenomenologists understand as being-in-the-
world), there are constant breakdowns and small or large misalignments
that we inadvertently repair. Only in some cases will it take a special effort
to restore existing expectations in the flow of activity with things and thus
rebuild the bonds that bind us to artefacts and things.
What else is there to say about these expectations we build around our
material culture? As is the case in our cultural practices in general, we are
dealing with a conglomerate of predictive and normative aspects.3 We ex-
pect objects to behave in a regular way, according to certain purposes and
 A Standpoint Epistemology of Repair? 117

modes of operation. But we also generate expectations about how other

agents are supposed to use, manipulate and intervene in them. Many of
these expectations have to do with what different actors should do or
should have done. When there is a need for intervention, either to maintain
regular behaviour or to reinforce compliance with certain norms, we con-
tribute to repair in the sense of restoring broken expectations. We rely on
this being borne out in our activities without the need to make specific in-
quiries. We intervene, or require others to intervene, when predictive or
normative expectations break down. No doubt, certain standards have
been affected: in our case, those related to the adaptation of the internal
temperature regulator to the environmental conditions and to the needs
(which could also be adapted) of intensive use during a heat wave, given
the age and condition of the appliance itself.
The examples could be multiplied. They all capture aspects such as the
following: (1) the interventions that I can or should make as a user/repairer
are in principle culturally regulated; there are predictive and, above all,
normative expectations about the nature, type, intensity and frequency of
the interventions I am allowed or obliged to make (some of these are legal
or contractual constraints); (2) so are the circumstances in which it is nec-
essary to consult or engage a specialised technician, which points to a so-
cial division of repair/maintenance labour and thus highlights the figure of
the repair technician or fixer, as we will call it in the rest of this chapter4;
(3) furthermore, it should not be forgotten that these expectations (and
how they are culturally regulated) change over time, but also with the so-
cial and economic situations of the actors and in the ways in which ‘techni-
cal’ knowledge is distributed in society.
This view of repair contrasts sharply with a purely functional approach.
It does not focus only on specific interventions by a specialised technician
who solves the problem caused by the failure. This is what we might call
a narrow view of repair. In contrast, our multivariate model allows and
requires a variety of interventions to maintain the good disposition of ar-
tefacts, systems and infrastructures. In maintenance and repair activities,
it is not easy to isolate what could be considered the purely technical as-
pect of breakdown. This is crucial for the purposes of this chapter because
it clearly shows that knowledge is necessarily distributed among different
actors and cannot be attributed to a single, specific group of authorised
and recognised experts. For this reason, we consider the fixer to be an
ambiguous and diffuse figure. To reiterate, our approach does not focus
only on technicians who repair devices or infrastructures but rather in-
sists on how repair permeates all stages and activities involved in the
(re)production of our material and artefactual environments. It is impor-
tant to keep in mind that living in these environments means that we are
engaged – to a greater or lesser extent, and with a more or less distinctive
118 Cristina Bernabéu and Jesús Vega-Encabo

awareness of doing so – in tasks of maintenance and repair. By ‘repairing’

in this broad sense, we as actors contribute to reproducing both the condi-
tions of existence/persistence of artefacts and the bonds we create through
them with other actors. Repair, in this broad sense, encompasses a con-
tinuum of activities ranging from seemingly minor interventions in our
everyday environments (such as cleaning or rearranging) to more elabo-
rate restoration exercises that can improve the functioning of devices.
Accordingly, repair, as we have said, consists of collaborating through our
interventions in the persistence and reproduction of our environments.5

5.3 Knowledge Subordination
We have outlined a broad view of repair in which each of us is responsi-
ble, with a greater or lesser degree of involvement, for specific repair in-
terventions that keep our expectations about artefacts, technical systems
and infrastructures in place. It is also important to note that these differ-
ences in involvement are also reflected in the distribution of resources and
epistemic competencies among the actors involved in the (re)production
of artefactual environments. Even during the design phase, we are ac-
tively shaping the conditions for repair and influencing cultural expecta-
tions regarding what should be maintained and how maintenance and
repair should be carried out. This implies that we assume the availability
of knowledge and skills to be displayed when needed in order to keep
things working. We sustain our technical and artefactual worlds with
small gestures that bring specific knowledge and varying degrees of exper-
tise to bear.
It is well known that societies exhibit a cognitive division of technical
labour, which implies a differential distribution of cognitive resources and
skills. But it is equally significant that the mechanisms of knowledge distri-
bution are invariably accompanied by marks of visibility, recognition and
acknowledgement. Some knowledge claims are more visible and more
readily accepted than others. Let us take a simplified model of technical
actors that distinguishes between what we will call makers, including de-
signers and producers (product developers, manufacturers, etc.), and users
or, more generally those people who inhabit, and take advantage of, tech-
nical environments. To these traditional figures (maker and user), we have
added the fixer.6 Now we think that one of the most deeply rooted assump-
tions regarding artefactual and technical knowledge is that makers (design-
ers and manufacturers) are the bearers of the knowledge that is deemed
necessary (even essential) to create and maintain these environments.7
They also regulate the value of other possible knowledge claims. Consider
the case of users who insist that this or that artefact must work in this or
that way. Suppose they succeed in making it work. Are they entitled to
claim knowledge about what the artefact does and how it does it? In a
 A Standpoint Epistemology of Repair? 119

sense, they are not, as their knowledge about the artefact is measured
against the expectations set by the maker. A similar argument can be made
for fixers and their knowledge claims. The knowledge of fixers, developed
in local contexts of repair, is subordinate to the normative requirements
that define the expectations for keeping things in good order. Although we
will return to this issue later in this chapter, in defining the possible epis-
temic privilege of the maker, it is important to remember that this epis-
temic subordination merely translates a model of agency into epistemic
relations. Several philosophers and anthropologists have recently criticised
(Ingold 2011; Preston 2013) ideational and hierarchical models of agency,
according to which all production emanates from and conforms to an idea
that hierarchically takes control of it.
We have seen that one of the main characteristics of artefactual environ-
ments is how their creation and maintenance involve multiple agents that
deploy specific kinds of knowledge and skills. From ordinary users to ex-
pert technicians, they all have to acquire a particular experience with the
objects and technical systems they install, use, maintain and repair. It is
within these environments that artefacts gain special significance as they
‘coexist’ and ‘interact’ with numerous other artefacts and systems. The
sustainability of artefactual and technical environments relies on complex
interdependent relationships between objects and actors. Together with
our experience and knowledge about the functioning, development and life
of instruments, artefacts and technical systems, we build our expectations
and manifest our dependencies. Indeed, we rely on artefacts for the most
part of the activities we carry out and, in doing so, we exhibit a capacity to
form certain predictive expectations about their behaviour. In addition, we
need to do something else: we need to weave trustworthy networks that
help shape our expectations about, for instance, the fact that they have
been designed according to certain standards, that they will be properly
maintained and repaired, that the skills and capabilities required for de-
sign, repair and use will be developed and so on. These are normative ex-
pectations about the behaviour of other agents, their knowledge and skills
and the reliability of these artefacts and systems.
As it should be, all of these expectations that underlie our dependency
networks, both predictive and normative, have a distinct epistemic di-
mension. Dependency networks only work if we collectively recognise the
value and importance of certain forms of knowledge. This recognition is
particularly crucial in the context of repair activities, where we rely on
the specific knowledge and expertise of those who restore our expecta-
tions. However, it is often the case that this knowledge remains invisible
and unrecognised. The local and distributed nature of this knowledge
contributes to its lack of collective recognition.8 As a result, the potential
contributions of these repair experts to our understanding of technical
environments are not fully realised.
120 Cristina Bernabéu and Jesús Vega-Encabo

It is an essential fact that much of this experience with artefacts and

technical systems is considered expendable or incidental even in the repro-
duction of artefactual environments. It is hard to see how this experience
could claim epistemic recognition and constitute an authoritative voice
concerning the persistence and maintenance of our artefactual environ-
ments and the socio-technical systems that are sufficiently stable to respond
to our needs and purposes. At least in our cultural imaginaries, and par-
ticularly in our theoretical views (as if they were ideological construc-
tions),9 the knowledge and experience of those who contribute to the
maintenance and repair of our artefactual world play a secondary role,
hardly capable of explaining the complexity of technical environments and
their innovative and transformative potential. This image is at odds with
how the reproduction of life in everyday environments depends not only
on the incorporation of technical innovations but also on a complex and
sometimes invisible social and technical division of labour in maintenance
and repair contexts. Given this, it becomes relevant to consider a stand-
point epistemology of repair that would succeed in inverting certain power
relations in order to authorise subordinated forms of experience and
knowledge within the cognitive division of socio-technical labour.

5.4 First Steps Towards a Theoretically Inspired Standpoint

Epistemology of Repair
Jackson (2014) introduces the idea of a possible standpoint epistemology
of repair as follows:

The question is this: can repair sites and repair actors claim special in-
sight or knowledge, by virtue of their positioning vis-à-vis the worlds of
technology they engage? Can breakdown, maintenance, and repair con-
fer a special epistemic advantage in our thinking about technology? Can
the fixer know and see different things – indeed, different worlds – than
the better-known figures of “designer” or “user”? Following on the
claims of Hegelian, Marxian, and feminist theorists, can we identify
anything like a standpoint epistemology of repair?
(p. 229)

This series of rather rhetorical questions paints a fluctuating picture of

what Jackson might be aiming at with his proposal for developing a stand-
point epistemology of repair. We identify at least four possible issues here:

a) The first asks whether fixers can make particular knowledge claims
about the technological world.
 A Standpoint Epistemology of Repair? 121

b) The second one subtly shifts the focus and asks whether our thinking
and reflection about technology might gain some epistemic advantage
by adopting the perspective of maintenance and repair. For now, let’s
understand the epistemic advantage in very neutral terms, such as ob-
taining more accurate or reliable representations, acquiring more knowl-
edge or gaining a better understanding.
c) The third simply contrasts the epistemic positions of designers and us-
ers with those of fixers by pointing out the obvious fact that they know
(or see) different things.
d) The fourth delves more deeply into traditional (Hegelian, Marxian and
feminist) motivations for a standpoint epistemology, suggesting a po-
tential inversion thesis: those who have been forgotten and ‘unjustly
oppressed’ in the practices of socio-technical (re)production (the fixers)
could claim an epistemic advantage over the makers (namely the de-
signers) on certain issues.

It is easy to identify double ambiguity in this motivation for a standpoint

epistemology of repair: the first is relative to those whose knowledge is at
stake; the second concerns the nature and ground of the epistemic advan-
tage. We will address the latter in the next sections, so let’s focus on the
first ambiguity now.
For “who is in a position” to claim knowledge in the face of the com-
plexity of our technological world? The first question seems to point to the
actors involved in the real practices of repair and maintenance, whose epis-
temic ‘rights’ have not been sufficiently recognised. The second question
pertains to theorists who have the ability to navigate between the perspec-
tives of different actors, such as designers, users and fixers. These theorists
can shift their focus to a series of aspects that have been forgotten or re-
mained invisible in our theories of technology.
True enough. Shifting the focus to the perspective of the ubiquitous phe-
nomenon of repair and maintenance seems to confer the theorist a certain
epistemic, even special, advantage. But this does not mean that it is the
fixer (and the collective of actors involved in repair and maintenance ac-
tivities) who is entitled to claim an epistemic privilege. Theorists could, in
a sense, take ownership of their experience and knowledge by changing
their approach and recognising that there are aspects of this experience of
breakdown and decay that call for more appropriate attention to repair
activities.
One might even suggest that what is being pursued is what we might call
a purely theoretical standpoint epistemology of repair. Those theoretical
approaches that seek to adopt the fixer’s own perspective gain an epistemic
advantage in terms of understanding how artefactual environments are (re)
122 Cristina Bernabéu and Jesús Vega-Encabo

produced over those that continue to privilege solely (and mainly) the fig-
ures of the designer or the user.
So formulated, this approach only vaguely reflects the tenets of a stand-
point epistemology. However, it helps to refresh some interesting points
characteristic of these epistemological views. It can be said that this ap-
proach responds optimistically to the question of whether and how the
theorist can access the point of view of those who are actively engaged in
repair activities. Indeed, it does not assume that it is necessary to be located
in the perspective of the oppressed or subordinated for this move to be
epistemically fruitful.10 If an epistemically advantaged standpoint is not
given, but must be achieved, as we will show in section V.5, perhaps it is
the theorist who carefully attends to the needs and interests of the fixers
that might have access to an advantaged standpoint. As some feminist
standpoint epistemologies have pointed out, it is not obvious that it is
women as such who have epistemic privilege, but those groups with social
and political consciousness who, through their theorising, respond objec-
tively to the interests of the collective. It is thus the objective facts related
to the role of repair activities in our technical and artefactual reality that
ultimately animate more faithful and inclusive theories.
To some extent, this move toward a theoretically inspired standpoint of
repair has been accomplished by more recent literature on maintenance
and repair.11 Nevertheless, how to access the specific point of view of the
fixer is a question which is rarely addressed. New aspects of our techno-
logical world are revealed when we turn our attention to those objects and
infrastructures that are broken or malfunctioning: their role in a complex
network of other objects and actors becomes visible (Martínez and
Laviolette 2019). The disruption of the continuity of our activity with ob-
jects when a breakdown occurs produces an experience of (thematic)
awareness of what has remained in the background. Nevertheless, it is not
enough to refer to this experience to bring technologies and infrastructures
out of invisibility, let alone repair and maintenance activities. Mark Young,
in a recent article, insists on the importance of analysing how we concep-
tualise the phenomenon of breakdown. It needs to be understood not as a
simple anomaly or as a deviation from a form-norm that repair could (and
should) bring back to its ‘ideal’ as fixed in design and manufacture (Young
2021). This understanding views artefacts and technology as permanent
and stable (in their formal and functional sense), perpetuating a certain
form of commodity fetishism that makes everything involved in making
our artefactual worlds possible, such as the endless process of maintenance
and repair, disappear from view.
Analytical clarity and explanatory power must be gained in order to
get ‘a deeper understanding of the aspects of everyday life which often
reside in the background’ (Young 2021, 116). But what connects – if
 A Standpoint Epistemology of Repair? 123

necessary – a different conceptualisation of breakdown and decay in the

artefactual world with the particular ways in which the fixers conceptu-
alise their activity and become aware of their activity? One way to bridge
this gap is by examining the growing ethnographic literature on mainte-
nance and repair activities (Orr 1996; Henke 2000, Graham and Thrift
2007; Denis and Pontille 2015; Strebel et al. 2019; Denis 2020; Henke
and Sims 2020), because it directly engages with the forms of knowledge
and value that shape the fixer’s point of view. There we ‘discover’ that
breakdown is not only a functional mismatch connected to the form and
matter of objects and infrastructures but above all a recurring aspect of
our existence with artefacts, of the ongoing relationship between things
and people, not a punctual failure. It is necessary to ‘fix’ people and
things in normal contexts of use and function, and not to ‘think’ of break-
down and repair as (abnormal) events. In a way, this corrects a ‘misrep-
resentation’ of the technological process and of everyday experience with
artefacts (as well as of the nature and role of infrastructures) that depend
on a ‘dominant’ point of view in which the ‘maker’ ideally (and norma-
tively) determines the mode of relations and experience with the techno-
logical realm.
However, the scope of this shift in perspective remains limited because it
neglects to account for how it responds to specific concepts, experiences,
values or practices of the particular standpoint of the fixer. Most impor-
tantly, it fails to explain how the scope and nature of a possible epistemic
advantage (or privilege) can be grounded in the practice of the participants
in repair activities. Is it possible to formulate a standpoint epistemology of
repair that focuses on the knowledge claims of the fixers and how they are
grounded? To address this question, it will be helpful to revisit some of the
familiar claims of standpoint epistemologies, which we will explore in the
next section.

5.5 Standpoint Epistemologies: Main Tenets

The very possibility of a standpoint epistemology rests on its ability to
justify the epistemic privilege it claims for the dispossessed. It assumes
that disadvantaged groups, by virtue of their oppression, have an epis-
temically privileged view of the social world, at least compared to the
view of dominant groups, who have a more limited and perhaps distorted
understanding of social life. However, this assumption goes beyond the
existence of different perspectives on a given issue and does not simply
emphasise the epistemic advantages derived from the fact of occupying a
certain position. This is an uncontroversial thesis as it is a ‘natural’ condi-
tion for the existence of epistemic positions in general or a result of ac-
cumulated (social) experience (and the acquisition and deployment of
124 Cristina Bernabéu and Jesús Vega-Encabo

expertise). Standpoint epistemologies hold – controversially – that certain

knowledge claims, at least on a certain set of issues and after a process of
consciousness-raising, are authorised by the fact that certain groups oc-
cupy certain positions of subordination.
The idea that epistemic privilege is conferred by occupying a subordi-
nate position within the social structure has Marxist roots and is based, in
particular, on Georgy Lukács’s (1971) explanation of the epistemic grounds
of the proletariat’s consciousness-raising. However, feminist epistemology
is one of the fields in which it has been most extensively developed in re-
cent decades (Smith 1974; Collins 1990; Hartsock 1983; Harding 1991).
It is not our purpose here to revisit the myriad of debates that standpoint
epistemologies have generated. Therefore, we will summarise the main te-
nets of any standpoint epistemology in three characteristic claims: (i) the
situated knowledge thesis, (ii) the epistemic privilege thesis and (iii) the
inversion thesis.12

(i) The situated knowledge thesis. The clearest formulation of this first
thesis appeals to an understanding of the nature of knowers. In con-
trast to traditional approaches to epistemology, which assume a tran-
shistorical, universal and disembodied knowing subject, it is argued
that we are all situated knowers, shaped by our social experience and
location (Haraway 1988). In short, it states that ‘[k]nowledge claims
are always socially situated’ (Harding 1993, 54), according to social
categories such as gender, ethnicity, class and so on. Knowers (and
their knowledge) are shaped by their social and historical locations
and differences in social factors create significant epistemic differences;
they affect what one can know and also the epistemic claims to which
one is entitled (Alcoff 2007).13
(ii) The epistemic privilege thesis. Standpoint epistemologies go beyond
the descriptive claim that location shapes the epistemic capacities of
the knowers. They argue, controversially, that some epistemic benefits
derive from attaining a particular standpoint. Social location confers
certain epistemic advantages on some groups, always relative to the
perspectives of other groups. How to articulate these advantages is a
matter of controversy within the field of standpoint epistemologies. To
the extent that they manage to occupy such a standpoint, epistemic
subjects are able to correct distorted views, at least of their own social
situation, and to open paths to the discovery of new truths. They don’t
assume that all perspectives are equally valid in epistemic terms, but at
the same time – otherwise, they would risk undermining the situated
knowledge thesis – they deny the possibility of formulating an objec-
tive standard against which different perspectives can be measured.14
 A Standpoint Epistemology of Repair? 125

(iii) The inversion thesis. As Tanesini (2018) shows us, this thesis embod-
ies the most characteristic emphasis of standpoint epistemologies. It is
based on the idea that the standpoint of the socially oppressed is epis-
temically advantageous, compared to the perspectives of the oppres-
sors. Those who occupy a socially (and in a sense, epistemically)
subordinate role reverse their position with the socially (and in a
sense also epistemically) privileged. Contrary to what common sense
suggests (i.e., that those with lower incomes, fewer resources and
lower levels of education are less likely to understand the social
world), it is argued that ‘the perspective of the downtrodden is epis-
temically privileged’ (Tanesini 2018, 337). That is, the economically
dispossessed and the politically oppressed, are able to see things that
the privileged usually ignore (Wylie 2003). In this sense, some authors
have spoken of a ‘bifurcated consciousness’ (Collins 1990); a con-
sciousness that knows both the dominant group’s vision, because it is
the dominant one, and the oppressed group’s vision, because it is the
oppressed one. In this sense, the standpoint of the minority groups
would have a higher probability of acquiring a more accurate view of
the social world, since it would be able to describe it in a less distorted
way; at least in the sense that it could be more faithful to the social
relations which are produced by material conditions. For this reason,
the inversion thesis can only be understood as being part of the politi-
cal commitment of those who claim to be located in their particular
standpoint.

To avoid misunderstandings, it is useful to add some comments to these

three theses:
First, as several authors have pointed out (e.g., Wylie 2003; Tanesini
2018; Ashton 2020; Fricker 1999), the mere fact of being oppressed does
not automatically make one’s perspective a standpoint. Being oppressed is
not a sufficient condition for epistemic privilege. This is crucial because it
is what makes it an achievement, not something that is automatically ac-
quired. Rather, the transition from a mere perspective to a standpoint can
only be achieved through a process of critical reflection, that is, a ‘con-
scious reflection on how power structures and resulting social locations
influence knowledge production’ (Intermann 2010: 785). This situates
standpoint epistemologies as part of critical theory (Anderson 2020;
Intermann 2010; Wylie 2003; Crary 2018). One acquires a certain epis-
temic advantage in the very process of becoming aware of one’s social loca-
tion, and this is the result of a collective process: knowledge is elaborated
from the perspective of those who share collectively recognised experi-
ences, needs or interests.
126 Cristina Bernabéu and Jesús Vega-Encabo

The second remark is not problem-free. How to make theses (i) and (ii)
compatible has been the subject of intense debate within feminist episte-
mologies. For our purposes, it is sufficient to recall that claims to a less-
distorted and even objective view of reality and a richer understanding of
the social world do not imply a notion of objectivity as disinterestedness
and as lacking any anchor in a particular social location. The debate has
raised some troublesome points regarding broader epistemological posi-
tions on justification and knowledge, such as objectivism, contextualism
or relativism. Whichever way out of these philosophical difficulties one
is willing to accept, it must be agreed that adopting a particular stand-
point must ensure epistemic improvement without abandoning the idea
of non-neutrality.15
The third remark is widely accepted and has to do with the risk of es-
sentialising the standpoint. If the experience of oppression due to belong-
ing to a certain social group is a part of what explains epistemic privilege,
there is a risk of thinking that all members of that group have the same
experience. Indeed, when we assert that women share a common experi-
ence, we run the risk of assuming that all women suffer gender oppression
equally. Alessandra Tanesini offers a good solution to this risk. In her prag-
matist approach, the knowledge that women may have about their social
situation is determined by shared interests and utility (which enable them
to gain a better understanding of their situation), rather than by member-
ship or subjective identification.16 In this understanding, the standpoint is
formulated through the identification of common interests rather than
through an a priori demarcation of the group.
This clarifies the ambiguities of Jackson’s proposal for a possible stand-
point epistemology of repair. On the one hand, it cannot be reduced to a
mere theoretical view of the main actors of the technological world (even if
it reflects their interests and needs), but must emerge from the experiences
and practices of the repairers themselves and of the collectives involved in
the (re)production of the artefactual world; on the other hand, as can easily
be seen from the brief reconstruction we have made of some of the central
tenets of standpoint epistemologies, the nature and grounds of epistemic
privilege relate to how relations of subordination (social and epistemic) are
inverted in a collective process of identification or, at least, of adoption of
shared (social and political) interests. In this sense, it is relevant to ask
whether or not a standpoint epistemology of repair is possible.

5.6 A Standpoint Epistemology of Repair?

Can the fixer then achieve a standpoint? The answer to this question is nei-
ther clear nor straightforward. However, let us examine each of the three
central tenets of standpoint epistemologies to explore some favourable
 A Standpoint Epistemology of Repair? 127

considerations. Justifying the situated knowledge thesis should not pose

much of a problem. As knowers, fixers seem to be located in a situation
determined by their material and social conditions. Within this circum-
stance, they articulate their specific experience of the artefactual and socio-
technical world. Therefore, there are no hindering circumstances that
prevent this position from becoming an epistemic locus from which to ex-
plore its conditions concerning the technological world and, more specifi-
cally, its (invisible) contribution to its (re)production. Likewise, it does not
seem problematic to accept that the category of fixer itself is a significant
situating factor, as is the case with the category of gender concerning the
(re)production of the social conditions of existence. We believe we have
provided sufficient grounds to consider the category of fixers as delineating
particular forms of experience in dealing with our shared expectations in
socio-technical and artefactual environments. Potentially, these can serve
as a starting point for constituting other forms of knowledge and for open-
ing up new trajectories of socio-technical transformation.
The two normative theses (epistemic privilege and inversion) obviously
pose more serious challenges, as the possibility of the fixer’s perspective
emerging as a genuine standpoint rests upon them. First, how can the fixer
become a site of epistemic privilege? An answer to this question is neces-
sary in order to satisfactorily explain how epistemic advantages are ac-
quired. Second, how might we account for such privilege in terms of its
subordinate location with respect to the epistemic practices of the technical
world? Both aspects must necessarily be intertwined: this depends on the
epistemic location that emerges as a standpoint. This is crucial because the
standpoint is a collective achievement that should, for example, correct the
conditions of invisibility of maintenance and repair practices or bring to
light assumptions that distort our accounts of how our material and socio-
technical environments are (re)produced. In this sense, the process that
leads to achieving a standpoint is also a process of forming a knower who
is able to make certain epistemic claims and who demands recognition, for
they have become masters of their own voice, so to speak. Because the
standpoint is rooted in the conditions of their lived experience as fixers,
they are in a position to make visible aspects that would be denied to the
dominant perspective. Therein, lies the potential epistemic advantage.
But this is one of the first problematic aspects: in what sense are relations
of subordination so pronounced in the field of the distribution of knowl-
edge and socio-technical skills that a reversal of roles is necessary to bring
this transformative potential to light? Is there something in the disregard
and invisibility of repair and maintenance activities, on which numerous
theorists have insisted as we have seen, that makes it possible to speak of
social subordination in the case of the reproduction of the technical condi-
tions of reality?
128 Cristina Bernabéu and Jesús Vega-Encabo

Although the answer to these questions is not obvious, at least one point
is clear: a form of subordination is reflected in hierarchical models that
characterise our collective imagination and are projected into recognisable
philosophical positions. Creators and inventors stand out as emblematic
figures in the production of the technical world. In the background, the
rest of the figures always appear, from the manufacturer to the user, but the
fixer remains completely forgotten. This implies that each of these figures
(which are, in principle, roles) occupies different (social and socio-technical)
locations from which they can make their knowledge claims. A hierarchi-
cal space is created. In it, the makers dominate over the rest of the ‘figures’:
they do so not only by virtue of the hierarchies that articulate social and
political (and economic) forms of recognition but also in metaphysical
terms. It is they who determine the metaphysical fate of the objects they
create, who ‘define’ their nature.17 All other figures must conform to the
norm established by the makers’ interventions in the world. But more im-
portantly, this explains a certain epistemic privilege of the makers (the
special authority on what an artefact really is): they possess practical
knowledge as makers, as opposed to the mixture of observational, inferen-
tial and testimonial knowledge, certainly incomplete, of the users and of
other figures who participate in the (re)production of the socio-technical
world. Of course, users can either intervene and reshape the intended uses
of an artefact or idiosyncratically adapt artefacts to their uses, but they can
hardly claim a hierarchical inversion, since the prevailing paradigm is a
unidirectional communication model in which messages go from makers
to users. Moreover, it is not easy to see how the inversion could affect epis-
temic privilege since it is ostensibly based on a specific form of authority
that only the maker can exhibit.18
It is an obvious fact that repair and maintenance practices have been
overlooked in our understanding of how we collectively shape our artefac-
tual and technical worlds. In this sense, they share with other collectives a
certain subordination to ‘nobler’ tasks, as well as to those social groups
that are more decisive when it comes to shaping socio-technical systems.
The question, then, is whether the particular location of fixers can become
a legitimate framework for epistemic claims about what the technical world
is and how it is created and maintained. However, if we examine the most
relevant claims of standpoint theory, such a framework should not only be
legitimate but also superior to other locations, such as those of designers or
users. It is important to clarify that, within this framework, the location of
the fixers should be understood as superior in the sense that they are able
to access more significant truths about the technical world and thus can
gain a better understanding of it by denouncing the distorted visions that
‘dominant groups’ have naturalised and protected from criticism.
Accordingly, we can point out two major difficulties in elaborating a
standpoint epistemology of repair. The first is to identify the collective that
 A Standpoint Epistemology of Repair? 129

embodies the figure of the fixer. Who are the fixers? The activities of main-
tenance and repair, as we have seen, permeate the entire socio-technical
system; moreover, they are part of the essential activities that sustain our
material culture. In what sense, then, can we identify a group or a collec-
tive of fixers? In what sense is it a collective that can make specific epis-
temic claims? It is not clear whether ‘fixers’ constitute a collective with
distinct social group identities or established social roles that share the
same set of cognitive and practical skills. Furthermore, it is uncertain
whether there exists a fixer’s standpoint either in terms of identification
with a social group or in terms of having ‘shared objective interests’ that
would motivate them to adopt that standpoint in a way that contributes to
their interests and goals. It is worth remembering once again that repair, in
our sense, implies a wide variety of tasks, ranging from infrastructure
maintenance, cleaning, care of objects, minor repairs, replacements, etc.
Additionally, the ‘cultural’ diversity in our understanding of repair should
be considered: there are different historical–cultural trajectories in how we
understand and engage in repair, and not all of them have ‘subordinated’
repair to nobler projects of endless innovation and change.
Second, even if it were possible to identify a group with its own experience
and knowledge, we lack an explanation for the potential epistemic privilege
they would seek. No one denies that they might be able to focus on specific
knowledge related to their activities (practical knowledge, tacit knowledge,
etc.) and even forms of (thematic) understanding of the technical world in
which they intervene. One might think that it is precisely this fact that is
often forgotten, that it represents unacknowledged knowledge and unheeded
claims to be part of the game of mutual recognition. But this does not give
us the rationale for the superiority of their own perspective, the sense in
which it would be less distorted. And it does not even allow us to assume
that fixers as a collective possess the resources to elaborate a particular epis-
temic standpoint. We cannot adopt a standpoint epistemology in which
mere positioning confers epistemic privilege. If they are capable of articulat-
ing their experience with their specific cognitive resources and by doing so
reach an enhanced understanding, it must be an achievement that requires
the collective itself to engage in the development of tools that strengthen
their epistemic claims and transform the practices involved in the (re)pro-
duction of artefactual worlds. They must have the ability to intervene in the
transformation of these practices themselves.

5.7 Repair as Critique19
Section 5.6 highlighted two challenges in applying standpoint epistemolo-
gies to repair: first, it is difficult to define fixers as an epistemic subject
based on shared interests or needs; second, it seems impossible to apply
an inversion thesis to collectives involved in repair and maintenance
130 Cristina Bernabéu and Jesús Vega-Encabo

compared to other collectives engaged in technical (re)production. Indeed,

as long as the criteria used by standpoint epistemologies to identify and
delimit oppressed social groups (workers, women, people of colour, etc.)
susceptible to developing a standpoint remain unchanged, these difficul-
ties appear insurmountable. This does not mean that the insights provided
by critical epistemologies like standpoint theory do not help recognise the
transformative role that can arise from considering the experiences of re-
pair and maintenance. In fact, limits and perspectives illuminate each
other in an analysis of the paths that collectives take to arrive at a
standpoint.
Within traditional standpoint epistemologies, there are usually two
paths that serve to demarcate social collectives that are capable of consti-
tuting genuine standpoints. We will refer to them as the identity path and
the material path. Once again, drawing a comparison with gender will help
us illustrate the challenges. On the identity path, women are considered as
such when they subjectively identify with the norms of femininity. In the
case of the material path, women are defined as such based on the position
of subordination they occupy in the sexual division of labour (reproductive
sphere).
If the fixer were to follow the path of identity, becoming a fixer so as to
be part of the collective would consist of self-consciously assuming such a
role, making it a defining aspect of one’s own identity. At first glance, this
path does not appear to justify the claim that people involved in mainte-
nance and repair activities occupy a particular standpoint. Since anyone
who subjectively identifies with such a figure can become a repairer, this
path results in a category of potential knowers that is too diluted and dif-
fuse. This category includes those who engage in Do-It-Yourself (DIY)
practices, which Wolf and McQuitty define as a ‘specific form of prosump-
tion (…) in which individuals use raw and semi-raw materials and compo-
nent parts to produce, transform, or reconstruct material possessions,
including those extracted from the natural environment’ (2011: 154). As is
well known, DIY practices apply to all areas of our lives, from hacking
with Ikea furniture to DIY urbanism to the transformation of automobiles.
In the context of maintenance and repair, there are some well-known refer-
ence sites, such as iFixit (Fixing the World 2023), a website that offers re-
pair guides for various devices and forum support for users. However, it is
far from clear whether fixers are framed by a set of norms with which they
could identify in order to participate in the collective. Norms of femininity,
for example, are shaped by structural conditions that dictate social and
cultural expectations for women. Nothing similar can be seen in the vari-
ous movements that are involved in the ‘visibilisation’ of repair and main-
tenance tasks or even the production and manufacture of products. It
should be noted, however, that DIY practices have recently adopted a more
 A Standpoint Epistemology of Repair? 131

collective approach, as can be seen in the shift towards DIT (Do It Together)
strategies, the proliferation of repair cafés (Repaircafé 2022) or the grow-
ing trend of the libraries of things (Library of Things n.d.).
The material path seems more promising. Reproductive and care work
is, within this path, the material and social place that collectively defines
women or that makes them an oppressed subject, as opposed to the collec-
tive of men, who perform the productive tasks. It is the tasks of care and
reproduction that can lead some women to adopt a specific point of view
about themselves as a collective, as well as about the hierarchical structur-
ing of social reality. The epistemic privilege of those who become feminists
lies in the fact that they are in a position to unmask the fantasy of patriar-
chal reality, which is built on a fantasy of autonomy, individuality and
productivity. It reveals the fact that reproductive labour, despite being
treated as dispensable, is an essential activity of the social world.
Similarly, in the material and cognitive divisions of the technical world,
the fixers should be demarcated by their subordinate position. However,
the question arises: subordinate in relation to whom? We have previously
insisted on the hierarchical model we project in our images and ideas of
technical (re)production, which seems to subordinate all the figures and
actors to the makers. But this, as such, does not delineate the collectives
according to the social, economic or material conditions of work. The eco-
nomic and material conditions under which groups in the Global North
perform their technical maintenance tasks may differ significantly from
those in the Global South, where individuals engage in demanding repair
and maintenance work under poorly paid labour contracts. Nevertheless,
it could be argued that just as women are identified by the assigned tasks
in the sexual division of labour, the collective of fixers is identified by their
particular care and treatment of artefacts and technical environments –
their ‘reproductive’ tasks, which may endow them with a particular vision
for unmasking the fantasy of domination and technical control of environ-
ments. This does not mean that the collective of fixers as such acquires a
privileged standpoint on the technological world by virtue of being in con-
ditions of subordination, but it does mean that the experiences of fixers
and users reveal specific ways of understanding the conditions of reproduc-
tion of the technical world. The awareness stemming from this knowledge
can undoubtedly contribute to correcting distorted perceptions of the na-
ture of technology itself.
It is necessary to mention something relevant here: the hierarchical
structure of the technical model is not entirely similar to the hierarchical
structure that defines the social world. Mainly because the positions, roles
or figures that define it (designers, users and fixers) are transversal. Thus,
the superiority that designers have over users and fixers in this model is
always related to a specific aspect that only a careful theoretical work of
132 Cristina Bernabéu and Jesús Vega-Encabo

contextualisation can adequately identify. Thus, IT staff fixes artefacts in

obedience to what the designs of the device establish as functionally ap-
propriate. However, they do not do so in the same sense as maintainers in
the Global South countries, such as workers in India, do so, nor in the
same sense as iPhone employees or users who repair their devices at home
following tutorials on YouTube. To be clear, the oppression or subordina-
tion of fixers in the tech world is intersectional. As far as this question is
concerned, suffice it to say that the intersectional nature of their subordi-
nation prevents the inversion thesis from being completely fulfilled or at
least prevents it from being fulfilled in the same way as it is fulfilled in the
standpoint model. Its intersectional and delocalised character means that
this inversion can only be claimed in relation to those specific aspects that
can be analysed from the specific contexts in which the work of fixers is
inscribed.20
Should we then abandon the promises of such critical epistemologies in
the case of repair? Perhaps it would be more feasible to approach repair
and maintenance activities through the lens of the ethics of care. Yes, there
is an urgency to take care of the world we inhabit together, so the ethics
of care is essential to understanding many of the normative cues at play in
repair and maintenance. However, this ethical commitment is inseparable
from an epistemology that emphasises the transformative potential that
derives from a process of becoming aware of the meaning of these activi-
ties, the experiences gained and their social, cultural, and, of course, tech-
nical insertion and impact (de la Bellacasa and Puig 2011; de Wilde 2021;
Martin et al. 2015; Liao and Huebner 2021). The experience of fixers, as
well as their skills and the possibility of bringing to light our expectations
about artefactual environments, should inform our knowledge about
what needs to be transformed if we are to take care of the world (Houston
et al. 2016). Although fixers as a group may lack a clearly defined epis-
temically privileged standpoint, there is a growing social demand, particu-
larly from certain collectives, for paying attention to the conditions in
which repair and maintenance activities occur. Movements advocating for
the rights of repairers have highlighted the need to unify demands and
recognise shared interests. Consider, for example, the way some commu-
nities are drawing our attention to the need to create ‘a more caring and
well-maintained world’ (themaintainers.org) by defending certain values,
under-recognised embodied experiences of maintenance and disrupting
ties to material things.
The current social and political reality has led certain groups of citizens
and fixers to claim themselves as a potentially transformative subject linked
to their perspective as fixers and maintainers. In a way, they embody a
 A Standpoint Epistemology of Repair? 133

certain social awareness of the cultural and technical centrality of these

activities and advocate for their rights as a collective. This suggests that the
social movement for the Right to Repair (2023) (which has been gaining
strength) is not only focused on individual rights but also encompasses a
broader, collective understanding of the technical and social worlds. At its
core, this movement seeks to create a better world through political
engagement.
Let us take stock. Fixers do not fit neatly into the framework of classic
oppressed collectives that have been the focus of standpoint epistemolo-
gies. Their position does not arise from the same material conditions or
group identification as traditionally marginalised groups. It is a collective
in which consciousness-raising does not pass through the first-person expe-
rience of oppression. In this sense, it has nothing to do with the conscious-
ness-raising characteristic of identity politics. However, it is still possible to
preserve the critical and transformative potential of adopting the stance of
repair and maintenance.
In her analysis of the standpoint as critical consciousness, Crary (2018)
suggests that what is important to engage in the development of a radical
political perspective is the acquisition of a certain sensitivity to harm.
However, we need to remember that the harm here pertains more to the
material world itself, with its fragility and vulnerability, rather than to
specific groups or collectives and their material or identity conditions. This
form of consciousness-raising emphasises our responsibility to care for and
protect the material world we inhabit, rather than focusing solely on self-
identity. From a philosophical perspective, this approach aligns more natu-
rally with proposals of an ecological nature, such as Lorraine Code’s
epistemology (2006) or interactionist approaches like José Medina’s epis-
temology of resistance (2013). The former involves a contextualist locali-
sation of epistemic practices, while the latter relies on forms of epistemic
friction between collectives. Both frameworks allow us to understand,
without relinquishing the situated knowledge thesis, how to address the
contextually shaped claims to recognition of experiences and practices re-
lated to the (re)production of our artefactual world, particularly regarding
our ways of inhabiting it. What we are witnessing, therefore, is a shift in
orientation, where consciousness-raising no longer stems from a dialectic
of opposition and inversion, but from an awareness of the fragility of exis-
tence and the constant negotiation that is required to inhabit our environ-
ment and engage with the material world. By embracing this perspective,
we can acknowledge the significance of repair and maintenance as prac-
tices that contribute to our understanding of, and our responsibility, to-
wards the world we inhabit.
134 Cristina Bernabéu and Jesús Vega-Encabo

5.8 Conclusion
Is a standpoint epistemology of repair possible? The question remains
open. We have argued that several aspects point to the impracticality of
using standpoint theories to ground clear epistemic advantages to fixers
over other socio-technical actors. We have introduced a broad notion of
repair and maintenance that we believe more adequately describes the way
we collectively sustain our artefactual environments. We all play this role
in one way or another. This is not to say that there are no groups that en-
gage in specific repair activities, with their own knowledge and skills. What
is difficult is to delineate the collective that specifically adopts a standpoint
from which to claim some epistemic advantage over the intimate workings
of our technological world or its social or political conditions. Thus, it may
not be easy to see how to substantiate the theses of epistemic privilege and
inversion that would define the normative framework of standpoint episte-
mologies in the case of repair.
On a more optimistic note, however, shifting the focus to the complex and
heterogeneous experiences of ourselves as fixers does indeed contribute to
highlighting its critical potential. Three brief observations will suffice to re-
mind us of its epistemic and transformative nature. First, the fixer’s perspec-
tive helps us to demystify certain fantasies of domination and control over
technical environments; the fixer reminds us that the effective conditions of
habitability of our technological world are always underpinned by constant
attention and care. Second, without necessarily emerging as a specific stand-
point, there are collectives that contribute to the awareness of these condi-
tions, which in turn has increased the demand to transform the social and
cultural conditions in which we develop our repair activities. Third, the
shared experience of repair and maintenance – at its various levels – gives
rise to the articulation of a collective awareness of the fragility of things. In
a sense, the authority of the fixer’s voice functions less as an inversion of
epistemic positions with other socio-technical actors than as a way of re-
specting the voice of our complex material and technical worlds. Perhaps
attending to the intricate and heterogeneous activity of repair will help us
better understand the ties that bind us to things and the sometimes illusory
expectations we project onto the artefactual environments we inhabit.21

Notes
1 This is why Jackson can claim that repair is the fulcrum of two worlds that
confront (Jackson 2014, 222).
2 Steven Vogel in a chapter included in this volume (‘Maintenance and the
Humanness of Infrastructure’) insists on this idea of maintenance as an ongo-
ing activity, but also in the continuity between the activities of building and
maintenance (Vogel, 2024).
 A Standpoint Epistemology of Repair? 135

3 Technological systems are also subject to stochastic processes that do not gen-
erate precise predictive expectations, but that are part now of how we need to
take care of our artificial worlds. Theoretically, we can build, so to say, sto-
chastic expectations about how these systems and environment could behave
and ‘evolve’.
4 In fact, in our highly differentiated technological environments, at least for
some devices and certainly for complex technical systems and infrastructures,
the expectation by default is that others – specialised technicians – will do the
repair and maintenance.
5 This insistence in the conditions of persistence (and reproduction) of the envi-
ronments should not be viewed from a purely static dimension. We assume that
our artefactual worlds change constantly and with them our expectations. In
repair and maintenance, we care about their stability, which does not have to
be mistaken with the mere preservation of unchanged environments. An earlier
chapter in the volume explores this idea in further detail (Young, 2024).
6 The figures are idealised; they reflect roles and not specific actors, although it is
obvious that one can later identify each role more or less closely with specific
groups. This is true of both designers and fixers, as it is of any other possible
technical roles.
7 In ancient philosophy, Plato was critical of this idea and considered that it is
users who are entitled to judge how artefacts should be produced by artisans
(see Republic 601c–602a and Cratylus 390b–d). But since at least the Middle
Ages, when Maimon introduced the idea of maker’s knowledge in his Guide for
the Perplexed, a particular image of the epistemic privilege of makers in rela-
tion to their products has been consolidating. See also Nicolas of Cusa in his
dialogues De Idiota, or Vico in his Scienza Nuova. For recent reconstructions
of the tradition of maker’s knowledge, see Pérez-Ramos (1988) or Floridi
(2011). See also comments on Section 5.6. We thank the editors of this volume
for the suggestion to note this broad historical change in the consideration of
the figure of the maker.
8 This has been an insistence in STS studies since the 1980s. The localised and
distributed dimension of knowledge related to repair and maintenance is much
more evident, as well as its invisibility. For more on the (in)visibility of repair
practices, see Young and Coeckelbergh (2024).
9 This can be detected both in philosophical discourses on technology and in the
socio-technical imaginaries we project in our cultural works. Technical worlds
seem to be suspended in the air, as if they were worlds of ideas. From its
Platonic matrix, thinking about technical production only recognises the role
of makers and users. Or since when in our collective imagination about tech-
nology do repairers or maintainers take centre stage? Consider, for example,
the imaginary projection of our digital future: where and how are the work or
skills of those who maintain and repair the endless series of software failures,
the network crashes, the machines, the huge data factories, represented?
Needless to say, that this does not correspond to the very experience of techni-
cal workers.
10 Social location is crucial in delineating the epistemic advantage but only to the
extent that we gain access to the standpoint through critical reflection on the
position of a group within a framework of unequal distribution of knowledge.
So, it cannot be denied that we may gain access to the relevant standpoint
through theorising, insofar as there are objective facts that are publicly acces-
sible. See the comments by E. Anderson 2020 on this possibility.
136 Cristina Bernabéu and Jesús Vega-Encabo

11 This is certainly true of Jackson’s pioneering paper, but also of a recent contri-
bution by Young (2021).
12 These designations respond to a reconstruction recently proposed by A.
Tanesini (2018). But theses defined in similar terms can be found in Bowell
(2023) (Internet Encyclopedia of Philosophy) or Ashton (2020). The core ideas
are related to how social location has epistemic consequences captured in the
theses of situated knowledge and the claims of epistemic privilege or advantage.
More controversial, but as we will see not less essential, is the inversion thesis.
13 The core insight of standpoint theory, that is, the idea that the epistemic
perspective (and thus the justification of beliefs) depends on the experience
that subjects have in relation to the specific social position they occupy
(whether privileged or oppressive) is built on the situated knowledge thesis
(Ashton 2020).
14 The very fact that different experiences, needs and interests shape different epis-
temic practices adds a potential epistemic improvement to the extent that it pro-
motes inquiring from a perspective in which bias and distortion are at least
counterbalanced. Rather, the controversial issue lies in how such a non-biased and
non-distorted understanding can be achieved if it is anchored to a particular stand-
point. That is, it is not enough to argue that the standpoint adds a new perspective
on a subject: it must seek to reinforce the epistemic virtues of good research.
15 For a discussion of the different positions, see Harding (2004), Rolin (2006)
and Ashton (2020).
16 ‘Putative knowledge is thus said to be socially situated in the sense that, given
their shared interests, some knowledge claims are more useful than others for
collectives and their members’ (Tanesini 2018, 336). In our opinion, her pro-
posal is more faithful to the distinctive political character of standpoint episte-
mology (which has to do with gaining a better understanding of the social
world in order to be able to transform it), rooted in feminism as an emancipa-
tory movement.
17 The model has been current in western metaphysics at least since the work of
Nicholas of Cusa. It is recognisable in those views on the metaphysics of arte-
facts that delineate the artefactual kinds on the basis of the intentions of the
makers, who in turn have privileged knowledge (in a very strong sense implying
the impossibility of error and ignorance) about their nature. See, for example,
Thomasson (2007).
18 What we mean by this is that given the way in which privilege is understood,
based on a form of knowledge inaccessible as such to the other figures, a genu-
ine inversion between the figures of the maker and the user is not possible
(obviously, neither is it possible for the fixer). Hence, the viability of a stand-
point epistemology of repair that emerges from the effective conditions of ex-
perience and knowledge of the fixer would depend on a correct understanding
of the thesis of epistemic privilege.
19 This section has benefitted greatly from very insightful comments from the edi-
tors of this volume and from the reviewers of the chapter.
20 Note that the requirement to analyse oppression through the filter of intersec-
tionality is not exclusive to the technical model. It is also the case of contempo-
rary feminist theories, including standpoint epistemologies, which seek to
account for the fact that gender oppression affects us in complex, non-univocal
ways and through diverse axes of oppression that intersect with each other,
giving shape to the specific oppression in each case or context.
 A Standpoint Epistemology of Repair? 137

21 We would like to thank Diego Lawler and Fernando Broncano for many dis-
cussions on the nature of the artefacts and the reproduction of the technical
world. This chapter has been enriched by the observations and comments of
the reviewers and, particularly, of the editors of this book. The first version of
this chapter was presented at a seminar on Maintenance and Philosophy of
Technology organised by Mark T. Young, which was followed by a very fruitful
discussion that helped us to polish our ideas.

References
Alcoff, L.M. (2007) ‘Epistemologies of Ignorance: Three Types.’, in S. Sullivan and
N. Tuana (eds), Race and Epistemologies of Ignorance, pp. 39–58. New York,
NY: State University of New York Press.
Anderson, Elizabeth (2020) ‘Feminist Epistemology and Philosophy of Science’,
The Stanford Encyclopedia of Philosophy (Spring 2020 Edition), Edward
N. Zalta (ed.), https://plato.stanford.edu/archives/spr2020/entries/feminism-
epistemology/
Arendt, H. (1958). The human condition. Chicago University Press.
Ashton, Natalie Alana (2020) ‘Relativising Epistemic Advantage’, in Martin K
(ed.) The Routledge Handbook of Philosophy of Relativism. New York:
Routledge.
Bowell, T. ‘Feminist Standpoint Theory’, Internet Encyclopedia of Philosophy, on-
line: https://iep.utm.edu/fem-stan/, accessed April 11 2023.
Code, L. (2006) Ecological Thinking: The Politics of Epistemic Location. Oxford:
Oxford University Press.
Collins, P. (1990) Black Feminist Thought: Knowledge, Consciousness, and the
Politics of Empowerment. London: Routledge.
Crary, A. (2018) ‘The methodological is political: What’s the matter with ‘analytic
feminism’. Radical Philosophy, 202, pp. 47–60.
de la Bellacasa and Maria Puig (2011) ‘Matters of care in technoscience: Assembling
neglected things’. Social Studies of Science, 41(1): 85–106. https://doi.org/10.
1177/0306312710380301
de Wilde, M. (2021) ‘A Heat Pump Needs a Bit of Care’: On Maintainability
and Repairing Gender–Technology Relations. Science, Technology, & Human
Values, 46(6): 1261–1285. https://doi.org/10.1177/0162243920978301
Denis, D.J. (2020) ‘Why do maintenance and repair matter’, in Anders B. et al
(eds.) The Routledge Companion to Actor-Network Theory. New York:
Routledge, pp. 283–293. https://doi.org/10.4324/9781315111667-31
Denis, J. and Pontille, D. (2015) ‘Material Ordering and the Care of Things’,
Science, Technology, & Human Values, 40(3), pp. 338–367. https://doi.org/
10.1177/0162243914553129
Fixing the world, one gizmo at a time. | iFixit News (2023) Available at: https://
www.ifixit.com/
Floridi, L. (2011) ‘A Defence of Constructionism: Philosophy as Conceptual
Engineering’, Metaphilosophy, 42(3): 282–304.
138 Cristina Bernabéu and Jesús Vega-Encabo

Fricker, M. (1999) ‘Epistemic Oppression and Epistemic Privilege’, Canadian

Journal of Philosophy 29(Suppl.), pp. 191–210.
Graham, S. and Thrift, N.J. (2007) ‘Out of order: understanding repair and main-
tenance’, Theory, pp. 1–25. https://doi.org/10.1177/0263276407075954
Haraway, D. (1988) ‘Situated Knowledges: The Science Question in Feminism and
the Privilege of Partial Perspective’, Feminist Studies, 14(3), pp. 575–599. https://
doi.org/10.2307/3178066
Harding, S.G. (1991) Whose science? Whose knowledge? Ithaca, NY: Cornell
University Press.
Harding, S. G. (1993) ‘Rethinking Standpoint Epistemology: What is Strong
Objectivity?’, in L. Alcoff and E. Potter (eds.), Feminist Epistemologies,
New York; London: Routledge, pp. 49–82.
Harding, S. G. ed. (2004) The Feminist Standpoint Theory Reader. New York and
London: Routledge.
Hartsock, N. (1983) ‘The Feminist Standpoint: Developing the Ground for a
Specifically Feminist Historical Materialism’, in Sandra H. and Merrill H. (eds.)
Discovering Reality: Feminist Perspectives on Epistemology, Metaphysics,
Methodology, and the Philosophy of Science. Dordrecht: D. Reidel, pp. 283–310.
Henke, C.R. (2000) ‘The mechanics of workplace order: Towards a sociology of
repair’, Berkeley Journal of Sociology, 44, pp. 55–81.
Henke, C.R. and Sims, B. (2020) Repairing Infrastructures. Cambridge,
Massachusetts, London, England: The MIT Press.
Houston, L. et al. (2016) ‘Values in Repair’. CHI ‘16: Proceedings of the 2016 CHI
Conference on Human Factors in Computing Systems May 2016, pp. 1403–
1414. https://doi.org/10.1145/2858036.2858470
Ingold, T. (2011) Being alive: essays on movement, knowledge, and description.
Abingdon, Oxon; New York, NY: Routledge.
Intermann, K. (2010) ‘25 Years of Feminist Empiricism and Standpoint Theory:
Where Are We Now?’, Hypatia, 25(4), pp. 778–796. https://doi.org/10.1111/
j.1527-2001.2010.01138.x
Jackson, S.J. (2014) ‘Rethinking Repair’, in Taleton, J. et al. (eds.) Media
Technologies: Essays on Communication, Materiality, and Society. The MIT
Press. https://doi.org/10.7551/mitpress/9780262525374.003.0011
Liao, S. and Huebner, B. (2021) ‘Oppressive Things’. Philosophy and Phenome­
nological Research, 103(1): 92–113. https://doi.org/10.1111/phpr.12701
Library of Things. (n.d.) Available at: https://www.libraryofthings.co.uk/
Lukács, G. (1971) History and Class Consciousness; Studies in Marxist Dialectics.
Translated by Rodney Livingstone. Cambridge, MA: MIT Press.
Martin, A., Myers, N. and Viseu, A. (2015) ‘The politics of care in technoscience’.
Social Studies of Science, 45(5): 625–641. https://doi.org/10.1177/0306312
715602073
Martínez, F. and Laviolette, P. (2019) Repair, brokenness, breakthrough. 1st ed.
New York: Berghahn Books.
Medina, J. (2013) The Epistemology of Resistance: Gender and Racial Oppression,
Epistemic, Injustice, and Resistant Imaginations. Nueva York: Oxford University
Press.
 A Standpoint Epistemology of Repair? 139

Orr, J. E. (1996) Talking about machines: Ethnography of a modern job. New York:
ILR Press.
Pérez-Ramos, A. (1988) Francis Bacon’s Idea of Science and the Maker’s Knowledge
Tradition, Oxford: Clarendon.
Preston, B. (2013) A Philosophy of Material Culture. London: Routledge.
Repaircafe (2022) About – Repaircafe. Available at: https://www.repaircafe.org/en/
about/
Right to Repair Europe (2023) Home – Right to Repair Europe. Available at:
https://repair.eu/
Rolin, K. (2006) ‘The bias paradox in feminist standpoint epistemology’, Episteme
3(1–2), pp. 125–136.
Smith, D.E. (1974) ‘Women’s Perspective as a Radical Critique of Sociology’,
Sociological Inquiry, 44(1), pp. 7–13. https://doi.org/10.1111/j.1475-682x
Strebel, I., Bovet, A. and Sormani, P. (2019) Repair work ethnographies. Singapore:
Palgrave Macmillan.
Tanesini, A. (2018) ‘Standpoint Then and Now’, in Miranda F. et al. (eds.) The
Routledge Handbook of Social Epistemology. 1st ed. New York: Routledge,
pp. 335–343. https://doi.org/10.4324/9781315717937-33
Thomasson, A.L. (2007) ‘Artifacts and human concepts’, in Eric Margolis &
Stephen Laurence (eds.), Creations of the Mind: Theories of Artifacts and Their
Representation. Oxford University Press. pp. 52–73.
Vogel, S. (2024) ‘Maintenance and the Humanness of Infrastructure’, in Mark T.
Young and Mark Coeckelbergh (eds) Maintenance and Philosophy of
Technology. Keeping Things Going. New York, NY; London: Routledge,
pp. 1–36.
Wolf, M. and McQuitty S. (2011) ‘Understanding the Do-it-yourself Consumer:
DIY Motivations and Outcomes’, AMS Review 1, no. 3–4, pp. 154–170.
Wylie, A. (2003) ‘Why Standpoint Matters’, in Robert F. and Sandra H. (eds)
Science and Other Cultures: Issues in Philosophies of Science and Technology.
New York, NY; London: Routledge, pp. 26–48.
Young, Mark T. (2021) ‘Now You See It (Now You Don’t): Users, Maintainers and
the Invisibility of Infrastructure’, in Michael N. et al. (eds.) Technology and the
City. Dordrecht: Springer, pp. 101–119.
Young, Mark T. (2024) ‘Technology in Process: Maintenance and the Metaphysics
of Artifacts’, in Mark T. Young and Mark Coeckelbergh (eds) Maintenance and
Philosophy of Technology. Keeping Things Going. New York, NY; London:
Routledge, pp. 1–36.
Young, Mark T. & Coeckelbergh, M. (2024) ‘Keeping Things Going: Maintenance
and Philosophy of Technology’, in Mark T. Young and Mark Coeckelbergh (eds)
Maintenance and Philosophy of Technology. Keeping Things Going. New York,
NY; London: Routledge, pp. 1–36.
6 Sustainability as Planetary
Maintenance
Jochem Zwier

6.1 Introduction
The following investigation asks in which sense sustainability is to be un-
derstood in terms of maintenance and what the limits and implications of
such an understanding are. While it may be clear that sustainability and
maintenance are closely related since sustainability seeks to maintain habi-
tats and habitability, it is less clear what this relation ultimately entails.
Sustainability remains a contested concept that has spawned a growing
discourse about which techno-scientific interventions are necessary and de-
sirable, which ethical and political implications are justified and how sus-
tainable solutions are to be governed (Salas-Zapata and Ortiz-Muñoz,
2019; Marshall and Toffel, 2005). For instance, questions have been raised
about whether and why sustaining food supplies trumps sustaining the
character of the land or ways of farming (Mariola, 2005). Although the
importance of such debates can hardly be overstated in light of the loom-
ing climate catastrophe, it is worth noting that as a byproduct of the in-
creasingly resounding call for more effective action and concrete sustainable
initiatives, the ontological implications of the concept of sustainability are
often left by the wayside. This is unsurprising since it is usually sufficiently
clear that whatever sustainability or ecological maintenance ultimately
means, sustainable and renewable developments appear to be superior to
fossil-fuelled developments. Be that as it may, this chapter departs from the
supposition that a philosophical inquiry into sustainability is nonetheless
worthwhile since it helps understand and critically examine the horizon
towards which sustainable developments are oriented. Accordingly, the
purpose here is to try and clarify the metaphysical and ontological implica-
tions of the concept of sustainability, which can be done by explicating its
relation to maintenance.
To this end, Section 6.2 attempts to clarify the metaphysical and onto-
logical roots of the concept of maintenance. It aims to demonstrate that
these roots lead to Aristotle. As a result, the concept of maintenance turns

DOI: 10.4324/9781003316213-7
 Sustainability as Planetary Maintenance 141

out to correspond to a hylomorphic ontology and correlated technical–

antagonistic apprehension of reality. The purpose of the section is to expli-
cate this correspondence and show how the concept of maintenance lies at
the heart of technics. This prepares the question regarding how it informs
and maps onto the concept of sustainability.
Section 6.3 develops this question by discussing the Anthropocene. It
demonstrates how the technical–antagonistic aspect of maintenance re-
turns in sustainability, now conceptualised as the striving to maintain the
habitability of an increasingly unruly Earth. The question then becomes
whether the hylomorphic ontology of maintenance returns similarly.
Section 6.3 develops a negative answer to this question by contrasting the
verticality and transcendence of Aristotle’s hylomorphism with the hori-
zontality and nihilism of sustainability. This problematises the concept of
sustainability by showing how the question regarding what or who is to be
sustained necessarily remains unanswered. This helps articulate the limits
of understanding sustainability as technical maintenance, as it demon-
strates that while maintenance and sustainability both involve a kind of
movement, the movement of sustainability must be negatively character-
ised as a fleeing movement: while it may be clear what to move away from
(unsustainable fossil-fuelled civilisation), a positive determination is kept
in abeyance.
Section 6.4 argues that as long as sustainability mirrors the schema of
technical maintenance, it must either ignore its nihilistic horizon or affirm
its fleeing movement. In raising doubts regarding the sustainability of flee-
ing itself, the work of Simondon is discussed to explore the possibilities of
planetary maintenance beyond hylomorphism. It concludes with an evalu-
ation of these possibilities and their implications.

6.2 Maintenance as Technics, Technics as Maintenance

This section inquires about the metaphysical characteristics of the concept
of maintenance. A glance at the concept suggests that it necessarily con-
cerns individuals and antagonism. An individual being and associated
functioning must be maintained to antagonistically ‘strife against’ (anti-
agon)1 its degradation and corruption. One gives the woodwork of a house
the occasional fresh coat of paint to keep it from weathering and withering
away, thus maintaining habitability. One changes the oils and filters of a
car to prevent a clogged engine, thus keeping it operational. One cultivates
the garden to subdue the proliferation of weeds and encourage the prolif-
eration of crops, thus keeping a garden rather than an overgrowth. Such
examples all involve some individual (house, car and garden) and func-
tions that are prone to wear and tear. Accordingly, to maintain is to coun-
teract degeneration and corruption, thereby literally ‘keeping in hand’
142 Jochem Zwier

(manu-tenere) the original identity and ideal functioning of things. Such

‘keeping in hand’ is characterised by a double keeping: a keeping-under2
(the elements, the ravages of time) and a keeping-up (things: identity and
functioning).
However elementary, these observations bear witness to nothing less
than an apprehension of reality or an ontology: there are individuals, there
is their potential decay and there is the care for their upkeep. Rather than
simply given, such an apprehension grows from philosophical roots, which
are in this case to be found in the fertile ground of Aristotle’s metaphysics.
Two of his metaphysical considerations particularly resonate with the
above outline of maintenance: the first is hylomorphism, according to
which things or individuals are considered in terms of form (morphe) and
matter (hyle) and the second concerns the relation between nature (physis)
and technics (technè).
In Aristotle’s hylomorphism, any fabricated individual3 is constituted
because a technician (technitès) brings matter in form. Following the ca-
nonical example, the smith knows how to manipulate the silver matter in
such a way that a chalice is formed to serve its sacrificial purpose. The
name for the smith’s knowledge is technics (technè) (Aristotle, Physics
2.1.193a 31–2). Given the theme of this volume, it is worth observing that
this can be literally understood in terms of maintenance, meaning that a
conception of maintenance lies at the heart of technè. Just as carpenters do
not by themselves create wood, smiths do not create the silver matter, since
both originate in nature as trees and ores. They are given to technics rather
than being its product.4 With reference to the etymology of maintenance
(manu-tenere), smiths take and keep this given matter into their hands to
mould it. Such ‘handling’ is done by observing the outlook (eidos) of the
fully formed and functional individual (telos), where observing both means
‘seeing’ and ‘complying with’: the smith sees and complies with the out-
look of the form of the chalice and correspondingly handles matter to
bring it in the form of the completed chalice (cf. Glazebrook 2000, p. 104).
Accordingly, while belonging ‘to bringing-forth, to poiesis’ (Heidegger
1977, p. 13), technics is less a matter of pure creation or absolute imposi-
tion than of ‘taking in hand’, that is, maintenance and manipulation of
what is given. By implication, when defined as the archè tès kinèseoos
(Visser, 2014, pp. 68–73), this cannot mean that the technician is archè in
the sense of sole principle and origin. Rather, reading the genitive tès the
other way around, archè tès kinéseos means to take command5 of a move-
ment, much akin to how in commandeering a ship, one takes possession,
one handles a given. As Heidegger (1977, p. 8) explains: ‘The silversmith
considers carefully and gathers together’ rather than creating from scratch.
And does this not point to both the heart of the smithy and the archetype
of all technics, namely fire? Mythologically speaking, it is by making up for
 Sustainability as Planetary Maintenance 143

his forgetful brother’s mistake (cf. Stiegler, 1998) that Prometheus offers
fire and technics to otherwise ill-equipped humankind, placing it into their
hands to maintain and manipulate. As pyrotechnicians, humans do not
create fire from scratch, but handle and manipulate a natural given to their
ends, be it in the form of torches, ovens or eventually internal combustion
engines. As the latter example suggests, this understanding of technics as
maintenance does not only concern so-called (and arguably mislabelled)
primitive technics like torches and fireplaces (cf. Clark and Yusoff, 2014)
but equally holds for modern technics like electronics, where rather than
creating it ex nihilo, engineers handle, manipulate and modulate a natu-
rally given movement (of electrons, or, put archaically, lightning) to achieve
their ends. Such manipulation demonstrates how a conception of mainte-
nance understood as ‘taking-in-hand’ lies at the heart of technics itself.
Considering technics in this way further points to the antagonistic as-
pect of maintenance. If not maintained properly, the fire dies or over-
grows into a wild conflagration. The technics of fire accordingly consist of
maintaining a rather precarious equilibrium, which Aristotle’s thought
generally articulates as the middle. This middle is most often associated
with Aristotle’s ethics, where it marks the virtuous position between two
extremes, for example, courage between cowardice and recklessness,
(Aristotle, Nicomachean Ethics, 1115b 17–9). Yet, the same figure re-
turns in technics as maintenance. To maintain the middle is to domesti-
cate the movements of nature (physis, which is always understood in
terms of movement or kinesis, cf. Glazebrook 2000, p. 103)6 by way of
striving against two extremes that it harbours, namely corruption or go-
ing into itself on the one hand (phftoras, the fire dying), and unrestrained
going outward on the other (genesis, wildfire). Such middle maintenance
can be observed in modern electronics as well, where circuits have an
equilibrium ‘operational window’ between inactivity (phftoras, e.g., a
dead battery) and hyperactivity (genesis, a short circuit).
Interpreted along these lines, technics handles and maintains naturally
generated movements (fire; lightning, silver and wood) and temporarily
stabilises them into a form. Maintaining stabilised informed matter such as
a particular wooden bed is antagonistic since it implies striving against the
very natural movements that gave rise to the bed, but which also risks up-
setting its equilibrium or middle. To clarify, consider how Aristotle fa-
mously states that ‘if you planted a bed and the rotting wood acquired the
power of sending up a shoot, it would not be a bed that would come up,
but wood’ (Aristotle, Physics II,1, 193a). Aristotle’s example expresses the
precarious equilibrium of the wooden bed: if not maintained, for instance,
by just leaving it outside, exposed to the natural elements, it de-forms and
recedes into the movement of physis again, either by sprouting (genesis) or
rotting away (phftorás). Aristotle famously says that technics ‘completes
144 Jochem Zwier

what nature cannot bring to a finish’ (Aristotle, Physics II,1, 199a) to ex-
press how the bed as informed matter owes its completion to nature as well
as to technics. Yet, this always comes with the potential of its reverse,
where nature consumes what technics completed: the bed is consumed by
worms, the iron by rust, etc. This antagonism concerns, in Arendt’s words,
the ‘unending fight against the processes of growth and decay through
which nature forever invades the human artifice, threatening the durability
of the world and its fitness for human use’ (1958, p. 100).
This then lays out the metaphysical and ontological roots of the concept
of maintenance, which corresponds to a hylomorphic ontology and cor-
related technical–antagonistic apprehension of reality. Maintenance un-
derstood as ‘taking-in-hand’ or ‘handling’ lies at the heart of technics
itself. If sustainability at the very least resonates with maintenance inas-
much as sustainability generally concerns the maintenance of habitats and
habitation, the ensuing question is whether and how the mentioned meta-
physical characteristics of maintenance inform or disinform the concept of
sustainability.

6.3 Sustainability as Maintenance
Sustainability closely resembles the above interpretation of maintenance.
To sustain is to maintain, here not so much a specific bed or chalice, but
the environment, an ecosystem, and one’s oikos or habitat. As such, sus-
tainability echoes the antagonism found in maintenance. As Horatius fa-
mously noted (Epist. I, 10,24) that ‘you may drive nature out with a
pitchfork, but she will keep coming back’, the concept of sustainability
necessarily expresses this strive against the return of nature.
After all, it now seems abundantly clear that we either find ways of be-
coming sustainable or degenerate into the physis whence we came. The
discourse of the Anthropocene expresses this antagonism particularly well.
On the one hand, it stages a highly technologised humanity, the anthropoi,
as the dominant Earth-shaping force. On the other hand, it stages the phy-
sis, now considered beyond local natural environments, ecosystems and
habitats as the planetary physis named Earth, and articulates how it has
awoken from its slumber (Hamilton 2017, p. 45). While this slumber,
which might be referred to as the Holocene, may have given the impression
of a stable physis that sustained ever-increasing extraction of (fossil) re-
sources, the Anthropocene orchestrates a new agon between technologised
geo-humanity and the Earthly, geo-physis (cf. Zwier and de Boer 2023).
Among others, this strife can be witnessed in extreme weather events,
global warming, desertification, droughts, crop failures, dwindling biodi-
versity, etc. The general response to such extremes and loss of a Holocenic
stable equilibrium or middle appears to consist in the maintenance of both
 Sustainability as Planetary Maintenance 145

protagonists of the Anthropocene: the anthropoi and Earth qua physis. If,
as suggested, maintenance involves the domestication and handling of phy-
sis, it is now up to geo-engineers and sustainable developers to tame the
unruly Earth system and abate its threat of becoming uninhabitable (cf.
Baskin 2015; Hamilton 2013). Conversely, it appears up to technology
designers, educators, ethicists and policymakers to help (re)domesticate
the anthropoi who have lost their way in excessive fossil-fuelled consumer-
ism and who must become responsible and responsive to the Earth and its
new climatic regime.
This general response serves to indicate how the antagonistic metaphys-
ics inscribed in technical maintenance returns in sustainability, as emphati-
cally expressed in the Anthropocene’s strive between geo-technè and
geo-physis. However, examination of the other aforementioned metaphysi-
cal dimension of maintenance, namely hylomorphism, will reveal a signifi-
cant difference.

6.3.1 Hylomorphism and Verticality

In the Aristotelean conceptualisation, what is to be maintained or kept in

equilibrium is guided by the outlook of the form. By complying with the
observed form of the bed, the technician seizes the material movements of
nature (trees growing and decaying) to maintain them in the form of a bed.
Technical things thus ‘stand upright’ over against the ‘growth and decay
through which nature forever invades the human artifice’ that Arendt men-
tioned (1958, p. 100). The vertical character of such ‘standing upright’,
which also resounds in maintenance as ‘upkeep’ is critical to hylomor-
phism and deserves further attention.
To clarify this verticality: consider how the bed that rots away and thus
re-enters the movements of nature (as worm food, then as bird food, etc.)
only concerns the particular artefact. The general form remains untouched.
It is for this reason that Aristotle ties the notion of form to that of entele-
cheia, which Heidegger renders as en-telos-echein, ‘something’s holding
(or maintaining) itself-in-its-completion-(or limit)’ (Heidegger 2000, p. 63;
cf. Glazebrook 2000, p. 99–100). The form denotes a completion, an op-
timum, the ideal thing in optima forma. In a wordplay that references his
own name, Aristotle also refers to it as the ‘best goal’ or ‘optimal goal’, the
telos ariston (Aristotle, Nichomachean Ethics, 1099b.17). Put simply, it is
only by observation of its complete and optimal form that one can recog-
nise a broken bed and construct a new one (cf. Section 6.2) – how else
could one recognise a thing as being broken or deficient? With regards to
the verticality of ‘standing upright’ and ‘upkeep’, this means the following:
if technics domesticate the material movements of nature to maintain them
in form (e.g., of a bed), the complete and optimal form that guides this
146 Jochem Zwier

process itself stands out from these movements of corruption and genera-
tion: it stands out and is beyond the growth and decay of physis, and is
therefore metaphysical.7
Notwithstanding their other disputes, Aristotle remains beholden to his
teacher Plato on this point. Plato’s Cratylus demonstrates a similar meta-
physical verticality of form, where in the discussion of a broken shuttle of
a loom, the craftsman tasked with making a new shuttle observes not the
broken artefact, but ‘looks up to’ its ideal, general and incorruptible form
(Plato, Cratylus, 1997a, 389b; cf. Aydin 2021, p. 19). In Platonic doctrine,
the particular, material, and therefore transitory shuttle participates in this
form. Similarly, in Socrates’ famous metaphor of the three beds in The
Republic X, the carpenter imitates the ideal form of a bed (and the painter
imitates that imitation) (Plato, Republic, 1997b, 598). Further, Plato’s
speculation on a demiurge (literally a technician or artifex) as a world-
maker does not create ex nihilo but observes and is guided by outstanding,
metaphysical ideas as blueprints (Plato, Timaeus, 1997c, 28a6).8
What follows is that the verticality of the hylomorphic schema is best
understood as an interrupted verticality that transcends to the metaphysi-
cal. The outlook (eidos) or idea concerns a higher kind of being, because,
unlike particular shuttles, beds and chalices, Plato’s ideas and Aristotle’s
forms are themselves ever-present, untouched by vicissitude and self-
sustaining, and thus require no maintenance. These general forms are ob-
served (in the double sense of seen and complied with) in the sublunary,
material and particular dealings of mortals.9
This may then give rise to the following question: how does the hylo-
morphic verticality found in technics as maintenance relate to sustainabil-
ity? Does sustainability similarly look up to and observe a transcendent
telos ariston or optima forma? What follows attempts to demonstrate that
the answer to this question must be negative. The hypothesis to be devel-
oped is that rather than vertical-metaphysical, the figure of thought at
stake in sustainability is horizontal-nihilistic.

6.3.2 Sustainability and Horizontality

To clarify this point, consider how sustainability expresses a kind of

movement. This does not primarily mean the scientific, technological and
political-economic progressive movement required to achieve sustainabil-
ity. Rather, in the more fundamental terms of the hylomorphic scheme,
sustainability expresses movement insofar as it seeks to maintain, domesti-
cate or handle the movement of unruly matter (hyle) such as CO2 back into
formation (morphe). Although CO2 may be the most telling example of a
material movement of physis in disequilibrium (consider its resonance with
 Sustainability as Planetary Maintenance 147

the previous example of wildfire as extreme physical genesis), one can also
think of animals as matter to be maintained (sustainability as maintenance
of biodiversity), of ecosystems (sustainability as maintenance of the envi-
ronment) or even the Earth (sustainability as maintenance of the dynamics
of the Earth system). All these examples concern a natural material move-
ment in disequilibrium, either on the side of growth (CO2) or on the side of
decay (biodiversity). Sustainability is the attempt to maintain and to han-
dle this movement, similar to how the upkeep of a bed or shed handles the
natural movement of decay.
However, where the hylomorphic scheme in technical maintenance is
guided by a positively determined metaphysical form such as the complete
and optimal bed or chalice, movement in sustainability is primarily deter-
mined negatively: away from pollution, no more externalities and no lon-
ger depleting fossil resources (and, on the associated political spectrum: no
more inequality, no poverty, no exclusion, etc.).10 In short, a movement
away from a fossil-fuelled Holocenic mode of existence that has become
untenable in the Anthropocene. But if the ideal of sustainability seeks to
depart from the fossil-fuelled era, which destination does this departure
seek? Or, put in the terms used in the above interpretation, if sustainability
is a kind of maintenance, and if technical maintenance is guided by an
optimal form that can be observed, what would be the corresponding
optima forma guiding sustainability?
As it appears, the negatively determined ideal of sustainability hardly
resembles the positively determined Platonic idea and Aristotelean eidos.
It is here that sustainability meets nihilism, which in Nietzsche’s formula-
tion means ‘[t]hat the highest values devaluate themselves. The aim is
lacking; “why?” finds no answer’ (1986, p. 9). Plato’s ideas and Aristotle’s
forms can be understood as such ‘highest values’, but they have lost their
fecundity. Leaving aside the genealogy of this loss, the aforementioned
antagonism between geo-technics and geo-physis in the Anthropocene
makes it manifest. The Aristotelean conception ultimately implied cosmic
stability with an enduring presence of optimal, outstanding forms. This
was paradigmatically expressed in the unchanging and perfect circular
movement of the superlunary sphere (moved by a still more perfect un-
moved mover) but similarly engages the optimal forms of technics such as
the ideal, unchanging and stable form of the chalice, bed or Plato’s shuttle.
By contrast, today’s geophysics, geology and Earth systems science attest
to a volatile cosmos, in general, and Earth, in particular. As a result, it
becomes impossible to speak of the Earth’s optimal form. If still meaning-
ful at all, one must face the metaphysical nihil of this optimum and admit
its contingency. Although the Earth can indeed be said to have gone
through an optimum, where relative stability and climatic conditions
148 Jochem Zwier

allowed for civilisation to flourish, this optimum now appears as a hap-

penstance exception rather than the rule, the rule being an unruly planet
where contingency rather than perfect cosmic stability reigns (cf. Clark
2011; Hamilton 2017).
Responding to being deprived of an ultimate positive idea or eidos, sus-
tainability formulates its ideals negatively: no more externalities, no more
pollution, zero waste, etc. If there are ‘sustainable development goals’, the
goals in question primarily indicate reparations and corrections of waste-
ful ways that now appear erroneous.
A symptom of this situation is that all ultimate positive determinations
now appear deeply suspect: is there an optimal form of the relation be-
tween technics and nature? Offering a formulation of this optimum would
be a hazardous undertaking. At best, it would be unmasked as some ro-
mantic and harmonious projection. At worst, it would be ridiculed as some
green fantasy. Furthermore, political suspicions are bound to arise: for
whom exactly would it be an optimum? For humans? Which ones? For
Global Northmen who wish to prolong their security, rule and comfort? In
the words of Stacy Alaimo (2012, p. 562):

Scholars in the humanities, or, more aptly, the posthumanities, may well
ask what it is that sustainability seeks to sustain and for whom.
Questions of social justice, global capitalist rapacity, and unequal rela-
tions between the global North and the global South are invaluable for
developing models of sustainability that do more than try to maintain
the current, brutally unjust status quo.

Timothy Morton (2018, p. 77) similarly asks:

What exactly are we sustaining, if not the one-size-fits-all agricultural

temporality pipe that has sucked all lifeforms into it like a vacuum
cleaner, pretty much, over its 12,500-year run?

Irrespective of whether such suspicions are justified, the relative ease by

which they now emerge11 serves to indicate, in Nietzsche’s terms, the de-
valuation of the ideal optimum, which thereby no longer serves as a bea-
con to orient the movement of sustainability.
In this qualified sense, sustainability can be said to be nihilistic. Rather
than a pejorative moral judgement, this names a formal characteristic, or
rather an in-formal characteristic, since the outlook of an optimal form is
absent. This does not necessarily mean that sustainability lacks any ideals
whatsoever but articulates how the status and topology of the ideal have
changed. To clarify, consider the potential objection that in the final analysis,
sustainability is not only negatively determined (no more externalities, etc.)
 Sustainability as Planetary Maintenance 149

but also positively oriented towards an optimal outlook or ultimate idea.

This idea could then be formulated as something like ‘enduring sufficient
quality of life for everyone’. Yet on the one hand, such a formulation paints
such a vague vista that in comparison to the idea of eidos of Aristotle’s opti-
mal bed, it rather appears as an idyl. On the other hand, as remarked previ-
ously, a closer examination of such an ideal demonstrates its deeply contested
character: sufficient quality for whom? Which quality exactly? Who is ex-
cluded from ‘everyone’ (humans? animals? other life forms, etc.)?
Accordingly, the ideal no longer follows the verticality of a metaphysi-
cal optimum that speaks for itself from on high. Instead, the ideal of sus-
tainability corresponds to a horizontal topology. The mentioned suspicions
(sustainability for whom? which quality of which lives? etc.) demonstrate
that rather than speaking from transcendent heights, the ideal of sustain-
ability becomes contested on a horizontal terrain.12 Accordingly, it be-
comes less a question of looking up to an ideal than of constructing one
from the ground up, a process inescapably accompanied by political con-
testations of who and what is included and excluded. The observation of
optimal forms makes place for their generation. In Latour’s words, the
transcendent notion of ‘Nature’ makes place for ‘world’ as ‘that which
opens to the multiplicity of existents, on the one hand, and to the multi-
plicity of ways they have of existing, on the other’ (Latour 2017, p. 35).
The political contestation regarding sustainability ultimately comes down
to ‘what existents have been chosen [to exist], and what forms of existence
have been preferred?’ (ibid. p. 37; cf. Zwier and de Boer 2023, pp. 76–
81). The question concerning what exactly is to be constructed thereby
necessarily recurs. However, since the blueprints from Plato’s Timaeus
have gone missing, their negatives must suffice: whatever is to be con-
structed, it must be less polluting, less exclusivist, less anthropocentric,
etc. The ideal guiding sustainability thus appears negatively determined as
‘less suboptimal’, while a positive determination recedes like a horizon
when approaching it.
Yet, if sustainability faces a receding horizon, if what ultimately is to be
sustained remains elusive, this destitution hardly constitutes a hindrance to
its movement. This is unsurprising when considering that not every move-
ment seeks a destination. In fleeing a crime scene or the site of a disaster,
one is not concerned with arriving anywhere in particular. For sustainabil-
ity, the designated crime scene and the site of disaster is the globally warm-
ing wasteland of Holocenic fossil-fuelled civilisation. Rather than requiring
a clear destination, the necessity of fleeing the scene suffices to instigate its
movement. Today’s virtually ubiquitous recital of sustainability as ‘sustain-
able development’, ‘sustainable entrepreneurship’ and ‘sustainable growth’
(e.g., Ploum et al. 2018) attests to this imperative of movement, which
neither has nor requires any ideal goal or optimal form.
150 Jochem Zwier

To recapitulate, the initial observation was that sustainability mirrors

technical maintenance, given how it concerns a metaphysical antagonism
between (geo)technè and (geo)physis. The question was whether the hylo-
morphic metaphysics that characterises technical maintenance correspond
similarly. The answer is, on the one hand, sustainability follows hylomor-
phism insofar as it anxiously seeks to maintain and contain matter (hyle)
such as CO2 in formation (morphe). On the other hand, due to its nihilis-
tic character, sustainability parts with hylomorphism and is no longer
guided by a transcendent optimal form. Since sustainability moves away
from something determinate (unsustainable fossil-fuelled civilisation) to-
wards nothing determinate, its movement is best characterised as flight.
Maintenance in the sense of moving matter (hyle such as CO2) away from
disequilibrium remains central, but the morphe found in the hylomorphic
scheme no longer guides such movement. As a result, not some optimal
form, but the fleeing movement itself is to be maintained.

6.4 Technical Maintenance of Actuality or Potentiality?

The above diagnosis of sustainability’s nihilism and flight of course does
not imply that it should be altogether abandoned. It could be argued that
the widely advocated idea of ‘becoming more sustainable’ can simply do
without a positively determined optimum since determinations of the sub-
optimum from which to flee are sufficiently clear. Habitability simply
trumps the alternative. Be that as it may, this raises the question of whether
such flight is itself sustainable or is itself to be sustained. Does moving to-
wards sustainable technical maintenance of the Earth ultimately imply pe-
rennially fleeing from sub-optima? It appears that since it mirrors technical
maintenance, sustainability must either ignore this question and prioritise
actual evasive maneuvers or affirm a sustained fleeing movement. Perhaps,
both options offer a sufficient response to the situation of the Anthropocene.
Still, and this is the wager of this final section, it may be worthwhile to
further explore the limits of this situation. Although such an exploration
cannot be expected to formulate a full-fledged alternative and some post-
nihilistic conceptualisation of sustainability, it can shed light on the pre-
suppositions that undergird current conceptualisations to open alternative
avenues. One way in which this can be done is by once more returning to
the technical character of sustainable planetary maintenance.
The technical maintenance mirrored by sustainability corresponds to
the cybernetic figure of the homeostat in its specific guise as a thermostat.
The now popular image of the Earth-as-thermostat shows this: relatively
modest technologies like solar panels and the more ambitious technologies
like solar radiation management are alike in aiming to dial back the run-
away thermostat of global warming (Zwier and Blok, 2019, pp. 622–7).
 Sustainability as Planetary Maintenance 151

Such dialling back corresponds to the negative feedback in cybernetic ho-

meostats, where the system’s output (classical example: observed steam
engine speed) feeds back into the system’s input (the steam inlet, by way of
a centrifugal governor), to minimise the difference between the intended
and observed behaviour and maintain an operational equilibrium condi-
tion. This equilibrium echoes the aforementioned Aristotelean equilibrium
as optimum (telos ariston) between genesis (e.g., an overdriven steam en-
gine) and corruption (e.g., a halting engine). Dialling back the Earth-as-
thermostat similarly feeds back observed outputs such as ‘5°C too high’ to
inputs such as ‘produced emissions’. Yet, because of sustainability’s nihil-
istic character, the difference is that the optimum setting of the thermostat
remains unresolved at worst and a perennial political contestation at best
(cf. Zwier and de Boer, 2023).
If the thermostat thus portrays the technical character of sustainability,
this raises the question of whether technics primarily consists of maintain-
ing homeostasis and equilibrium. Aristotle and sustainability answer posi-
tively: technics maintains the material movement of physis in equilibrium.
Conversely, Gilbert Simondon answers negatively. His work is therefore
instructive for exploring the limits of sustainability.

6.4.1 Simondon versus Hylomorphism

Simondon criticises the hylomorphic scheme for being an abstracted, dead-

locked and ultimately misguided way of metaphysically considering tech-
nics and ontology in general. It is abstract because it corresponds to a
specifically aristocratic and detached way of perceiving reality. Seeing the
technical process in terms of pre-given matter to be ‘handled’ (i.e., main-
tained) into pre-established forms ‘corresponds to the knowledge of some-
one who remains outside the workshop and considers nothing but what
enters and exits it’ (Simondon 2020, p. 30). For Simondon, such a perspec-
tive not only concerns workshops, artisans and technics but also fundamen-
tally characterises metaphysical thinking and the tradition of philosophy
(cf. Combes 2012, p. 1). This tradition privileges static individual beings
(cf. Section 6.2) such as ‘final optimal form’ and ‘pre-given matter’, and in
so doing wrongfully overlooks the process of their individuation.
Conversely, in attempting to ‘grasp ontogenesis in the whole unfolding of
its reality’, Simondon’s project consists of seeking ‘to know the individual
through individuation rather than individuation through the individual’
(2020, p. 30).
Thinking ontogenetically requires a transfiguration of the hylomorphic
scheme. Among many other examples (cf. Voss 2018, p. 110), Simondon
concretely develops such thinking in his analysis of brick moulding, where
he replaces stability with the concept of metastability. Rejecting the
152 Jochem Zwier

hylomorphic differentiation between pure matter and form, Simondon

(2020, p. 22) explains:

To give a form to the clay is not to impose the parallelepiped form onto
raw clay: it is to pack the prepared clay into a fabricated mold. If we
start from the two ends of the technological chain, the parallelepiped
and the clay in the quarry, then we can experience the impression of
realizing in the technical operation an encounter between two realities
of heterogeneous domains and of instituting a mediation through
communication

A brick cannot be made from wet sand, which would simply disintegrate
upon drying. Hence, rather than passive matter waiting to be actively in-
formed, the clay matter is positively determined as a potential for deforma-
tion (Simondon, 2020, p. 25; cf. Combes 2012, p. 5). On a molecular level,
clay is already in-formed in the swampy soil in such a way that the other
material form of the mold:

limits and stabilizes rather than imposing a form: it provides the goal of
deformation and achieves it by interrupting it according to a definite
contour […] the mold plays the role of a fixed set of modeling hands,
acting like halted kneading hands.
(Simondon 2020, p. 25)

The individuation of a brick establishes communication between hitherto

heterogeneous domains of becoming. In this sense, it is similar to how the
individuation or becoming of plants first established communication be-
tween a cosmic order (light energy) and an intramolecular order (mineral
salts, oxygen, etc., cf. Combes 2012, p. 4). Similar to how a plant is the
communicative medium connecting these orders, brick moulding estab-
lishes communication between the clay and the brick mould. Since it is it-
self technically prepared, ‘the pure form [of the mold] already contains
actions, and the raw material is the capacity of becoming; the actions con-
tained in the form encounter the becoming of the matter and modulate it’
(ibid., p. 25). For Simondon (ibid., p. 41), this marks the insufficiency of
the hylomorphic schema, which:

does not account for implicit forms, since it distinguishes between the
pure form (which is called form) and the implicit form, which is con-
flated with other features of matter under the name quality.

The reference to ‘kneading hands’ and ‘modulation’ recalls the interpre-

tation of technical maintenance as taking in hand or ‘handling’ of physis
 Sustainability as Planetary Maintenance 153

(Section 6.2). When Simondon criticises the hylomorphic schema of the

metaphysical tradition, it accordingly remains questionable whether this
critique can be fully directed at Aristotle himself. After all, as the exam-
ple of the bed made clear, the wooden matter out of which the bed is
made is, in Simondon’s terms, ‘implicitly formed’.13 Still, Simondon’s on-
togenetic consideration of communication between two chains of becom-
ing (the clay with potential for deformation and the mould as a potential
limit to deformation) marks a significant departure from hylomorphism.
When these chains enter into communication in the technical operation
of brick moulding, something new enters into reality. This is to say that
in contra hylomorphism, there is no pre-given, transcendent and com-
plete optima forma that serves as the ultimate orientation point (telos
ariston) for a technician. Instead, the form is an outcome or ‘endpoint’
(Combes 2012, p. 5) of the technical operation: ‘The technical operation
integrates implicit forms rather than imposing a totally new and foreign
form on a matter that would remain passive vis-a-vis this form’ (Simondon
2020, p. 42).
Instead of unchanging, stable beings like Aristotelean forms or Platonic
ideals, Simondon privileges the metastable: ‘The original state of being is a
state that goes beyond coherence with itself, that exceeds its proper limits:
original being is not stable, it is metastable’ (Ibid. p. 369).14 As Yuk Hui
(2016, p. 84) explains:

‘metastable’ describes a transitional state through which the individual

as product is given to us, but further individuation can take place when
the individual is stimulated by external information or by energetic ex-
citation from within.

Simondon’s paradigmatic example is crystallisation. It cannot be under-

stood as the imposition of form, but only as communication. When a su-
persaturated solution enters into communication with heat energy, it starts
to crystallise, which in turn triggers more crystallisations. The condition
for this occurrence is that the amorphous solution is neither stable nor
unstable, but metastable: the solution is charged with potential and when
a threshold is passed, the amorphous state transitions to a crystalline struc-
ture, without ‘the contribution of some foreign form’ (Simondon 2020,
p. xxiii). Ascribing the process of becoming found in crystallisation to real-
ity as such, and referring to the potential charge of a supersaturated solu-
tion, Simondon (2020, p. 352) points out that

This charge of the undetermined can be called nature (…) a veritable

reality charged with potentials actually existing as potentials, i.e. as an
energy of a metastable system.
154 Jochem Zwier

If hylomorphic ontology ultimately understands reality as self-maintaining

actuality in terms of transcendent forms (see Section 6.3.1 and footnote 9),
Simondon’s ontogenesis understands reality as self-maintaining potential-
ity, always maintaining ‘a reserve of becoming’ (2020, p. 358) that gives
rise to new communications and individuations.

6.4.2 Metastability or Elusive Homeostasis

Simondon’s conceptualisation of technics and metastability raises the fol-

lowing question: if, as argued (Section 6.3.2), sustainability qua planetary
technical maintenance is characterised by flight, what does Simondon’s re-
vision of technics and hylomorphism imply for this characterisation?
The philosophy of individuation does not counter nihilistic devaluation by
formulating a new optimal form to guide sustainable planetary maintenance.
It does not indicate the optimum setting to which the Earth’s runaway ther-
mostat should be dialled back. The possibility of such static anchoring points
is principally refused. Instead, Simondon sidesteps this metaphysical trajec-
tory. Rather than maintenance geared towards some actuality, which is to
say, recalling Heidegger’s characterisation of metaphysics, some ‘enduringly
present’ optimal form (cf. Section 6.3.1), Simondon’s thought considers the
question of sustainability as one of maintaining potentiality. Not the homeo-
stat, thermostat or any other stable state, but becoming is to be taken in-
hand. Such taking in-hand or maintenance would then not consist of the
conservation of form against physical corruption and generation, but in es-
tablishing novel communications.
Within the scope of sustainability and the theme of the Anthropocene,
this primarily indicates communication between technologised humanity
and Earth. Whereas sustainability here sees antagonism between the geo-
technè of humanity and the geo-physis (Section 6.3.1), Simondon’s thought
allows for seeing novel communications and individuations. This not only
means that an otherwise statically determined humanity now wields newly
individuated technologies such as solar panels and geo-engineering facili-
ties but also indicates the ongoing individuation of humanity and the
Earth. A significant aspect of this individuation is that humanity no longer
merely lives on Earth but must live with the Earth. This can be regarded as
a novel communication, similar to how the living individuation of plants
first established communication between a cosmic order and an intramo-
lecular order.
While investigating the implications of this novel communication re-
mains an ongoing task (cf. Hamilton 2017; Zwier and de Boer 2023), it is
clear that it cannot be constrained to an Anthropocentric perspective ac-
cording to which the situation is imposed by humanity. Just like the clay
and the mould are considered as two chains of becoming, technologised
 Sustainability as Planetary Maintenance 155

humanity and the Earth likewise find themselves communicating anew.

Regarding technics, specifically, Simondon holds that its evolution is not
imposed by human forms, but it individuates by itself as a quasi-living be-
ing (cf. Lindberg 2022, p. 54). His interpretation of the Guimbal turbine
shows that although human anticipation and invention essentially belong
to the process, the turbine shapes its own conditions of operation and thus
‘renders itself possible’ by constituting a ‘techno-geographic milieu’
(Simondon 2017, pp. 57–8). The turbine only works when inserted into
the water, only when it establishes novel communications that condition
the water to function both as the energy source and the cooling mecha-
nism. Whereas Heidegger still prioritised Aristotle’s silversmith as ‘care-
fully gathering together’ the material movements of nature into an observed
form (cf. Section 6.2), for Simondon such gathering or self-conditioning is
non-anthropocentrically interpreted as the outcome of several individuat-
ing chains: the technical individuations internal to the turbine which con-
tain ‘an entire electrical factory’ (2017, p. 58); the water becoming milieu
for its functioning and the social individuations resulting from its gener-
ated energy, etc.
Viewed from this perspective, the question comes to concern the new
techno-geographic milieu of the Anthropocene. Rather than sustaining
some ineffable optimal form, the task then consists of understanding what
living with the Earth entails. Simondon’s ontogenetic reworking of hylo-
morphism, technics and ontology opens avenues for investigating this
question beyond the limits of sustainability understood as planetary tech-
nical maintenance. Although such an investigation remains to be carried
out further, two presumably informative considerations can be offered in
closing.
First, a critical dialogue with Simondon appears in order. It could be
argued that the Earth becoming uninhabitable of itself nothing else than a
physical individuation on the level of the Earth System, where a situation
crystallises that no longer happens to support the individuations that are
associated with civilisation. This raises the critical question of how
Simondonean thinking can accommodate this kind of ‘negative individua-
tion’ or what Bernard Stiegler refers to as ‘disindividuation’ (2006).15 As
noted, Simondon abandons metaphysical dualistic oppositions of form
versus matter and (human) technics versus nature, to instead privilege
compositions of physical, living, psychical and technical chains or ‘regimes
of individuation’ (Simondon 2020, p. 12). Yet, this seems to imply a cer-
tain unidirectional or even upward development moving from the physical
(e.g., crystals) to the living (e.g., organisms), to the technical and to the
psychical (human) and social. What, however, if the physical individuation
of the planetary habitat that renders it uninhabitable, thus comes to under-
mine rather than support ‘higher level’ individuations like psychical and
156 Jochem Zwier

social individuations? If such a question is on point, it could be argued that

in stressing individuation as becoming or Aristotelian genesis, its compan-
ion term of corruption now comes to haunt Simondon’s philosophy of po-
tentiality and becoming.16 In any case, Simondon’s attempt to move beyond
a rigid opposition between matter and form, as well as between nature and
technics, allows for addressing such critical ecological questions anew (cf.
Lindberg 2022, p. 62).
Second, it may be clear that abandoning maintenance as conservation of
an actuality in favour of maintenance as communication with a potential-
ity comes at a cost: it becomes experimental rather than cybernetically
controlled (cf. Bardin 2021, p. 38). Experimental does thereby not neces-
sarily refer to a strictly controlled scientific experiment but involves recep-
tivity and generosity to what remains beyond control and what remains
potentially new. This calls for a different comportment: rather than either
seeking to conserve some optimal form or flee from the sub-optimum
(cf. Section 6.3.2), Simondon’s thinking invites a progressive comportment
that welcomes new individuations. Instead of the homeostatic control ac-
cording to some existing yet elusive optimum, it opens towards mediation
with what remains hidden, with what is always ‘richer than its self-
coherence’ (Simondon 2020, p. 369). It is here that Simondon (2020,
p. 411) equates the philosopher and the technician:

Thales, Anaximander, and Anaximenes are technicians above all else …

the veritable technician is the one who is a mediator between the com-
munity and the hidden or inaccessible object.

Concerning the abovementioned question of whether sustainable planetary

maintenance is finally characterised as flight, it remains to be decided
whether this experimental cost is too high. If the currently accepted view
understands sustainability as the maintenance of either some actual plane-
tary optimum, or at least a habitable minimum, then rethinking mainte-
nance in terms of potentiality may appear too naively optimistic about
individuations and its ‘potential becomings’, too audacious or even reckless
in light of what needs actual attention. On the other hand, the alternative
appears to be the affirmation of the sustained fleeing movement of sustain-
ability. What is more, even when (rather favourably) assuming that the
optimal thermostat setting is sufficiently clear and uncontested, past at-
tempts to dial it in can be evaluated by their fruits. It then appears that to-
day’s rather blatant failures to reset the thermostat point to more than a few
rotten apples (cf. Malm, 2018). As the 2023 IPCC Synthesis Report states:

In 2018, IPCC highlighted the unprecedented scale of the challenge re-

quired to keep warming to 1.5°C. Five years later, that challenge has
become even greater due to a continued increase in greenhouse gas
 Sustainability as Planetary Maintenance 157

emissions. The pace and scale of what has been done so far, and current
plans, are insufficient to tackle climate change.
(IPCC 2023)

Seen in this light, even if Simondon’s privileging of potentially novel com-

munications and progressive comportment appear preposterous in the
light of looming catastrophe, the failures of sustainability’s planetary tech-
nical maintenance might be read as a justification to nonetheless venture in
this direction.

6.5 Conclusion
This chapter asked in which sense sustainability is to be understood as
maintenance and what the limits and implications of such understanding
are. It argued that sustainability follows the metaphysical schema of tech-
nical maintenance, but runs up against a limit of form. Whereas technical
maintenance knows the optimal form of what to maintain (a bed, a house
and a bicycle), sustainability only knows the suboptimal form that should
not be maintained. As a result, it engages in a fleeing movement. Although
understandable and even justifiable in light of the present climatic situa-
tion, the question remains whether such flight determines the limit of what
planetary maintenance means. By way of its critique of hylomorphism and
emphasis on individuation and potentiality, Simondon’s work opens an
alternative avenue. Exploring this avenue suggests that sustainability and
planetary maintenance need not mirror the schema of technical mainte-
nance, but might investigate newly established communication between
technics, humanity and the Earth. If Simondon is thereby read as a critique
of maintenance and its adherence to hylomorphism, a critique cannot be
understood as rejection or refusal, but as delimitation. Accordingly, while
technical maintenance of individuals such as chalices, beds or even planets
is necessary, Simondon suggests that it is not sufficient. It neither exhausts
the potential of technics nor how it is maintained or ‘taken in hand’ nor
how it enters into individuating communication with the Earth.
Whether one opts for sustainability’s flight or Simondonean experi-
ments, it appears that the conditions for considering these options are
given for now. Perhaps, the response to this gift must be to cultivate it
further, to ask which potential Earth is to be maintained, for which indi-
viduals or individuations and to which potential ends.

Notes
1 Throughout this chapter, etymological references are used, not to prove or fix
their final meaning, but simply to suggest that the genealogy of the words we
use is not inconsequential and can inform and inspire reflection on what we –
often implicitly and unreflectively – mean by them.
158 Jochem Zwier

2 The Dutch word for maintenance is ‘onderhoud’, which literally translates as

‘under-keep’, thus expressing this characterisation of maintenance.
3 Fabricated individuals (e.g., a chair) differ from natural or physical individuals
(e.g., a tree) insofar as the latter carry their movement towards from within
themselves (the tree becoming full-grown), whereas fabricated individuals in-
volve an external movement in which the technician or craftsperson plays a role
(a tree never grows into a chair). Given the present focus on maintenance and
sustainability, this chapter focuses on fabricated individuals and leaves the ques-
tion of natural individuals (and whether they ultimately derive from a technical
paradigm) open. See Heidegger (1998, pp. 202–4); Glazebrook (2000, p. 99).
4 The required intermediary technical steps of turning trees into lumber or ores
into silver do not contradict their physical givenness.
5 Next to ‘origin’ and ‘principle’, archè also means ‘command’, as Agamben
(2019) makes clear in ‘What is a command?’
6 The word ‘nature’ similarly expresses movement, as it derives from ‘nāscī’ or
‘being born, being generated’ or simply ‘growing’.
7 Meta means after, beyond, above or transcendent, which is how metaphysics is
traditionally understood: after and above the physical.
8 In modern philosophical thought on technology, the clearest echoes of this
vertical metaphysical thinking can be found in the work of Friedrich Dessauer
(cf. Dessauer, 1956; cf. Mitcham, 1994, pp. 29–33).
9 The metaphysics of interrupted verticality thereby resonates with Heidegger’s
definitions of metaphysics as ‘enduring presence’ (Beständige Anwesenheit)
and ‘onto-theology’ since observed forms or ideas are enduring and constitute
the highest (theos) instance of being (cf. Heidegger 2002, pp. 42–75).
10 This can be observed in the ‘sustainable development goals’ such as ‘no pov-
erty’, ‘zero hunger’ and ‘reduced inequalities’, combined with industrial devel-
opment with ‘no externalities’ like pollution, use of fossil resources, etc.
11 The very phenomenon of this ‘relative ease’ may of course well be interpreted
as a laborious and hard-fought achievement by critical scholars, activists, etc.
12 In Morton’s work, ‘ontology […] is a vital and contested political terrain’
(2013, p. 20). One could find the devaluation of the highest goals anticipated
in Descartes’ mechanistic universe, where true goals are restricted to the do-
main of human morality, yet only as ‘conjectures’, whereas real optima forma
are inaccessible: ‘for it does not appear to me that I can without temerity seek
to investigate the [inscrutable] ends of God [les fins impenetrables de Dieu]’
(Descartes, 4th meditation (1911, p. 20).
13 Simondon acknowledges this in his discussion of Aristotle’s bed: ‘Here, the
technical operation accommodates the living form and partially diverts the lat-
ter for its own benefit’ (2020, p. 42).
14 According to Anne Sauvagnargues, Simondon’s philosophical extension of the
thermodynamic notion of metastability is crucial for his account of reality:
‘Metastable being, in disequilibrium, involves this state of asymmetrical disequi-
librium which accounts for tension and the production of the new’ (2013, p. 58).
15 Where it could be argued, as Stiegler does, that the becoming uninhabitable of
the Earth primarily points to the bankruptcy of psychical and social-collective
(dis)individuation since the ‘physical individuation’ of ecological collapse is
rooted in unfettered consumerism (cf. Stiegler et al. 2021).
16 In ‘The Limits of Human Progress’, Simondon touches on this problem but is
mostly concerned with the ‘saturation’ of potentials, rather than corruption
(cf. 2010).
 Sustainability as Planetary Maintenance 159

References
Agamben, G. (2019) Creation and anarchy: The work of art and the religion of
capitalism. Stanford University Press.
Aydin, C. (2021) Extimate technology: Self-formation in a technological world.
Taylor & Francis.
Baskin, J. (2015) Paradigm dressed as epoch: The ideology of the Anthropocene.
Environmental Values, 24, 9–29.
Bardin, A. (2021) Simondon Contra New Materialism: Political anthropology re-
loaded. Theory, Culture, and Society, 38(5), 25–44.
Clark, N. (2011) Inhuman nature. Sociable life on a dynamic planet. Los Angeles:
Sage.
Clark, N., & Yusoff, K. (2014) Combustion and society: A fire-centred history of
energy use. Theory, Culture & Society, 31(5), 203–226.
Combes, M. (2012) Gilbert Simondon and the Philosophy of the Transindividual
(T. Lamarre, Trans.). Cambridge: The MIT Press.
Descartes, R. (1911) The Philosophical Works of Descartes (E. S. Haldane, Trans.).
Cambridge: Cambridge University Press.
Dessauer, F. (1956) Streit um die Technik. Frankfurt: J. Knecht.
Glazebrook, T. (2000) From “physis” [Greek] to Nature, “techne” [Greek] to
Technology: Heidegger on Aristotle, Galileo, and Newton. The Southern Journal
of Philosophy, 38(1), 95–118.
Hamilton, C. (2013) Earthmasters: The Dawn of the Age of Climate Engineering.
New Haven: Yale University Press.
Hamilton, C. (2017) Defiant Earth: The Fate of Humans in the Anthropocene.
Cambridge: Polity Press.
Heidegger, M. (1977) The question concerning technology and other essays
(W. Lovitt, Trans.). New York: Harper and Row.
Heidegger, M. (1998) On the essence and concept of physis (T. Sheehan, Trans.) In
W. NcNeill (Ed.) Pathmarks (pp. 183–232) Cambridge: Cambridge University
Press.
Heidegger, M. (2000) Introduction to Metaphysics (G. Fried & R. Polt, Trans.).
New Haven & London: Yale University Press.
Heidegger, M. (2002) Identity and difference. (J. Stambaugh, Trans.) Chicago:
University of Chicago Press.
Hui, Y. (2016) The parallax of individuation. Angelaki. Journal of the Theoretical
Humanities, 21(4), 77–89.
IPCC (2023) Urgent climate action can secure a liveable future for all. Press Release
20 March 2023. IPCC_AR6_SYR_PressRelease_en.pdf [accessed: 20 March
2023].
Latour, B. (2017) Facing Gaia: Eight Lectures on the New Climatic Regime.
(C. Porter, Trans.) London: Polity Press.
Lindberg, S. E. (2022) Simondon écologiste? In J. H. Barthélémy & L. Duhem
(eds.), Écologie et technologie: redéfinier le progrès après Simondon (pp. 49–62).
Paris: Éditions matériologiques.
Malm, A. (2018) The progress of this storm: Nature and society in a warming
world. Verso Books.
160 Jochem Zwier

Marshall, J. D., & Toffel, M. W. (2005) Framing the elusive concept of sustain-
ability: A sustainability hierarchy. Environmental Science & Technology, 39(3),
673–682.
Mariola, M.J. (2005) Losing Ground: Farmland preservation, economic utilitari-
anism, and the erosion of the agrarian ideal. Agriculture and Human Values 22,
209–223.
Mitcham, C. (1994) Thinking Through Technology: The Path Between Engineering
and Philosophy. Chicago: The University of Chicago Press.
Morton, T. (2013) Hyperobjects. Philosophy and Ecology After the End of the
World. Minneapolis and London: University of Minnesota Press.
Morton, T. (2018) Being Ecological. Cambridge, MA: MIT Press.
Nietzsche, F. (1986) The Will to Power (W. Kaufmann & R. J. Hollingdale, Trans.)
(W. Kaufmann, ed.). New York: Random House.
Plato (1997a) ‘Cratylus’, In J. M. Cooper (ed.). Plato: Complete Works.
Indianapolis: Hackett.
Plato (1997b) ‘The Republic’, In J. M. Cooper (ed.). Plato: Complete Works.
Indianapolis: Hackett.
Plato (1997c) ‘Timaeus’, In J. M. Cooper (ed.). Plato: Complete Works.
Indianapolis: Hackett.
Ploum, L., Blok, V. Lans, T., & Omta, O. (2018) Toward a validated competence
framework for sustainable entrepreneurship. Organization & Environment,
31(2), 113–132.
Salas-Zapata, W. A., & Ortiz-Muñoz, S. M. (2019) Analysis of meanings of the
concept of sustainability. Sustainable Development, 27(1), 153–161.
Sauvagnargues, A. (2013) ‘Crystals and Membranes: Individuation and Tem­
porality’, In A. de Boever et al. (Eds.) Gilbert Simondon: Being and Technology.
Edinburgh: Edinburgh University Press. pp. 57–70.
Simondon, G. (2010) The limits of human progress: A critical study. (S. Cubitt,
Trans.) Cultural Politics, 6(2), 229–236.
Simondon, G. (2017) On the Mode of Existence of Technical Objects.
(C. Malaspina & J. Rogove, Trans.). Minneapolis: Univocal Publishing.
Simondon, G. (2020) Individuation in Light of Notions of Form and Information.
(T. Adkins, Trans.). Minneapolis: University of Minnesota Press.
Stiegler, B. (1998) Technics and time: The fault of Epimetheus. (R. Beardsworth &
G. Collins, Trans.). Stanford University Press.
Stiegler, B. (2006) The Disaffected Individual in the Process of Psychic and
Collective Disindividuation. (P. Crogan & D. Ross, Trans.). Available at: https://
arsindustrialis.org/disaffected-individual-process-psychic-and-collective-
disindividuation (Accessed: March 2023).
Stiegler, B. & The Internation Collective (Eds.) (2021) Bifurcate: There is No
Alternative. (D. Ross, Trans.). Open Humanities Press. Available at: http://www.
openhumanitiespress.org/books/titles/bifurcate/ (Accessed: March 2023).
Visser, G. (2014) Heideggers vraag naar de techniek. Een commentaar. Nijmegen:
Van Tilt.
Voss, D. (2018) Simondon on the notion of the problem. A genetic schema of indi-
vudation. Angelaki. Journal of the Theoretical Humanities, 23(2), 94–112.
 Sustainability as Planetary Maintenance 161

Zwier, J., & Blok, V. (2019) Seeing through the fumes: Technology and asymmetry
in the Anthropocene. Human Studies, 42(4), 621–646.
Zwier, J., & de Boer, B. (2023) Earth becomes world? Scientific objects, nonmod-
ern worlds, and the metaphysics of the Anthropocene. Environmental Human­
ities, 15(1), 64–86.
7 Towards a Realist Metaphysics
of Software Maintenance
Keith Begley

7.1 Introduction
Software would seem to be an unusual object of maintenance because it is
not subject to degradation in the way that physical components are and so
the maintenance of software has a different character to the maintenance
of hardware. This has been recognised by software engineers themselves.
For example, consider the following passage on software maintenance
from a software engineering textbook.

[…] whiledo constructs do not wear out after 10,000 loops, and semi-
colons do not fall off the ends of statements. Unlike hardware, soft-
ware does not degrade or require periodic maintenance. Thus, software
systems are different from hardware, and we cannot use hardware
analogies successfully the way we can for other aspects of software
engineering.
(Pfleeger & Atlee 2010, pp. 535–6)

The ‘difference’ that is at issue here, put ontologically, is a difference in

ontological status between abstract and concrete objects. Similar observa-
tions have been made by others. For example, Gruner has asked the ques-
tion ‘Something that is not material cannot physically decay or wear out;
in what sense is it then possible to speak of “software maintenance?”’
(2011, p. 296). There is a presupposition here that software is abstract in
some way and also that maintenance is merely preservative or conserva-
tive. However, if software were abstract, in the sense of something non-
concrete and unchanging, then it seems that software maintenance would
not be possible at all, at least not in a way that would have some effect
upon an abstract object. Realism about software need not commit us to
such an otiose position. As we shall see, we can have our abstract software
and maintain it too, just not in any way that has an effect upon an abstract
object.

DOI: 10.4324/9781003316213-8
 Towards a Realist Metaphysics of Software Maintenance 163

Some things that computer scientists and users often call ‘software’, that
is, its copies, are in part physical and are involved in interactions with
hardware. Many philosophers and computer scientists have been con-
cerned with drawing attention to software’s dual nature, its being both
symbolic and physical, abstract and concrete (e.g., Moor, 1978; Colburn,
2000; Turner, 2018; Coeckelbergh, 2022), and some propose that it is an
‘abstract artifact’ that is temporal but non-spatial while its copies are con-
creta (Irmak, 2012). In Section 7.2, I will delineate and assess some of
these views.
There are strong connections to be found between this topic and recent
investigations in the philosophy of linguistics, particularly the ontology of
words, where software has often been mentioned in comparison. My ap-
proach draws upon a recent parallel project to reconstruct Katz’s meta-
physics of words (Begley, 2023). In Sections 7.3 and 7.4, I will argue,
contra Irmak (2012), that software can be best understood through a real-
ist type-token distinction that was put forward by Katz (1998). This ap-
proach employs both non-spatiotemporal abstract types and tokens of
those types. I show that employing and further developing Katz’s realist
approach can help us to make better sense of computational artefacts and
their maintenance than the alternatives outlined. In Section 7.3, I provide
an outline of Katz’s ontology of composite objects, before, in Section 7.4,
applying it to software and computational artefacts in general.
In Section 7.5, I discuss how we should make sense of the notion of
software maintenance in light of the realist view of software developed.
I also address a problem that has been raised by Young (2021) for theories
of artefacts that employ intentions, such as the one involved with the real-
ist view of software and software maintenance for which I shall argue.
I will show that software maintenance on this view is not solely conserva-
tive but also adaptive and enhancive, constituting a source of counterex-
amples and so is not susceptible to the problem.

7.2 The Ontological Dualities of Computational Entities

It has long been noted that computers and computational objects appear
to exhibit a dual or ‘mixed’ nature. This has been put in a variety of ways
by authors over the years. In this section, we will consider some of these
ways, coming to focus in particular on the use of the abstract-concrete
distinction. An early example that is often mentioned in the literature
comes from an article by James H. Moor.

Computers can be understood on two levels. Computers can be anal-

ysed on the physical level just as any physical system can. If the com-
puter is electronic, then the laws of physics concerning electronics can
164 Keith Begley

explain the computer activity. But, it is the understanding on the sym-

bolic level which makes computers calculating devices, for it is under
this kind of interpretation that various structures or processes of the
computer are understood as symbols. Similarly, a computer program
can be understood on two levels. Physically, computer programs may be
a series of punched cards, configurations on magnetic tape, or in any
number of other forms. Symbolically, computer programs are under-
stood as instructions to a computer.
(Moor, 1978, p. 213)

Here Moor makes a distinction between a physical and a symbolic level,

both of computers and their programs. The physical level and its laws
provide the explanation for how the physical components of the computer
end up in certain physical states,1 with, for example, quantities of charge
arranged thus and so. However, this is not sufficient for computation, be-
cause computation is a kind of abstract operation that has been character-
ized in various ways, for example, as involving abstract states and
transitions between those states on input or, equivalently, as general
recursive functions, etc. (Turing, 1937 [1936]; Church, 1936). That is,
nowhere in such abstract descriptions need quantities of charge, etc., be
mentioned.
Colburn (2000) takes the title of an undergraduate textbook in com-
puter science, Concrete Abstractions (Hailperin, Kaiser, & Knight, 1999),
as his starting point for reflection, and others have followed him (e.g.,
Irmak, 2012; Turner, 2018). However, the authors of the textbook do not
discuss ontology beyond the first few paragraphs of the preface in which
they attempt a brief initial justification for calling computer science ‘the
discipline of concrete abstractions’.2 Colburn examines the view that soft-
ware is a ‘concrete abstraction’ more closely.

While computer scientists often enthusiastically embrace this duality, a

metaphysician will view it as a puzzle to be explained. How can some-
thing, namely a computer program, be at once concrete and abstract?
(Colburn, 2000, p. 205)

Colburn approaches the puzzle by employing a framework that is ordinar-

ily used to classify responses to the mind-body problem, that is, as being
either monistic or dualistic. Nevertheless, he recognised that the approach
results in ‘not an entirely appropriate language, for describing programs as
concrete abstractions’ (2000, p. 205). The approach turns out to be rather
underproductive, as in the end, he entertains notions of pre-established
harmony. As Turner points out, this is merely a metaphor for a kind of
harmony that is better explained as being due to correct implementation
 Towards a Realist Metaphysics of Software Maintenance 165

(2018, p. 51). Furthermore, as we shall see presently, there are much more
promising lines of metaphysical theorising in neighbouring areas such as
the philosophy of linguistics, where we find very similar issues being dis-
cussed regarding the apparently mixed or dual ontological status of words
and other linguistic entities.
The ‘puzzle’ recognised by Colburn has been recognised by others. For
example, Turner raises a similar question.

[…] everywhere we turn in computer science we find the same pattern:

a fusion of an abstract thing with a concrete one. Does this herald the
introduction of a new kind of ontological entity? How are we to catego-
rize and conceptualize the entities of computer science?
(Turner, 2018, p. 15)

Despite this initial characterization in terms of the abstract-concrete dis-

tinction, Turner instead appropriates a conception of the dual nature of
what Kroes (2012) calls technical artefacts. Kroes’ approach takes the du-
ality involved in this case to be between structural and functional proper-
ties, with the structural side being satisfied by the physical objects involved
and the functional side being satisfied on the basis of intention (Turner,
2018, p. 25).3 Where I agree with Turner is that there must be intention
involved in the creation of a computational artefact. However, the func-
tional side is put forward in terms of a design specification only for what a
physical system should do, and so the required symbolic or abstract level
that was mentioned by Moor and Colburn appears to be deprecated. In
Sections 7.3 and 7.4, we will give further consideration to what is intended
in the intention, including what the ontological status of that thing in-
tended is.
This is an apt point to mention a way in which the conception of soft-
ware maintenance is related to the conception of software as an artefact
constituted through an intentional act. Young (2021) is rightly concerned
to show that the usual conception of maintenance as preservation or con-
servation fails to recognise the full role that maintenance plays in the pro-
duction of artefacts and their efficacy. He aims to show that the widespread
adoption of this conservative view is due to the acceptance of a conception
of the production of artefacts that involves intentions. He first points out
that ‘among philosophers […] artifacts are nearly always understood as
manifestations of the intentions of their creators’ (2021, p. 358). He then
argues that such an understanding ‘often encourages us to privilege the
moment construction finishes as the point in an artifact’s history where the
intentions of the designer are most fully realized’ (p. 359). The claim here
is that this privileging of the moment of full realisation of the intentions of
a design leads us to think of maintenance merely as a means of conserving
166 Keith Begley

this full realisation. Additional considerations would also appear to apply

to realist or essentialist views in particular.

Because the essence of an artifact, the form, is understood to be atem-

poral, time and change are cast as external influences against which a
technology must be protected, giving rise to a conception of mainte-
nance as conservation.
(Young, 2021, p. 400)

These arguments would appear prima facie to present a problem for any
view that maintains both that its artefacts are realisations of intentions and
that maintenance is not merely conservative, but also, for example, adap-
tive and enhancive. So, it would appear to be a potential problem for the
realist view of software and software maintenance that I will put forward in
the following sections, based on Katz’s ontology for composite objects. This
is because Katz too provides a role for intentions as aspects of creative rela-
tions (or patterns of relations) that token artefacts of a certain abstract type.
I will return to address Young’s problem again in Section 7.5, after develop-
ing the realist approach employing intentions in Sections 7.3 and 7.4.
Returning to our discussion of the dual nature of computational arte-
facts, we should note that Coeckelbergh (2022) has similarly recognised
the ‘hybrid’ nature of computational objects, with particular regard given
to the role of language.

Language is often literally connected to technology, or is even part of

what the technology is. Consider some of the technologies most of us
use daily, such as computers and mobile phones (or less common tech-
nologies such as robots). These are material artifacts, to be sure, but
they run on software, which is based on a software language, on code.
Both the material aspects and the linguistic aspects constitute the tech-
nology. For the technology to work, it is crucial that they are connected.
Code by itself cannot do anything in the world; the hardware (and other
software) is needed. Vice versa, the hardware needs the language of
software to do things. When it comes to contemporary information and
communication technologies (ICTs), it is clear that language and mate-
riality are entangled. The agency and ontology of such technologies can-
not be adequately described without taking language into account. Even
at the so-called technical level, these technological artifacts already have
a hybrid nature: they are material and linguistic at the same time.
(Coeckelbergh, 2022, p. 343)

Whether or not the hardware-software distinction is an ontologically sa-

lient one remains to be seen. That is, both hardware and software would
 Towards a Realist Metaphysics of Software Maintenance 167

appear to have dual natures. Further, both can be emulated by another

computer, a piece of software or a piece of hardware. I will return to this
point again briefly in Section 7.4.
Coeckelbergh focuses in particular on the mediating roles that language
plays in technological practices and argues that the insights of a particular
brand of use-based philosophy of language have been neglected in this re-
gard in the philosophy of technology. As is well-known,4 and is empha-
sised by Coeckelbergh, there was an ‘Empirical Turn’ in the philosophy of
technology roughly 40 years ago, that is, a limitation of considerations to
those regarding material artefacts and our embodied engagements with
them. However, it is clear that the questions regarding the dual nature of
software and other entities in computer science have not abated and that
the abstracta have stubbornly refused to be turned away.
In my view, which is especially concerned with the objects under consid-
eration, the insights of another neighbouring subfield have also been ne-
glected, that of the philosophy of linguistics, where metaphysical debates
regarding the status of linguistic entities have been ongoing since at least
the publication of Jerrold J. Katz’s Language and Other Abstract Objects
(1981).5 More recently, work has been done in the philosophy of linguis-
tics, and on the philosophy of words in particular, which treats of issues
similar to those that we come across when considering the dual nature of
objects in computer science.6 There should be no surprise in this, as the
domains overlap in many ways. I will discuss Katz’s approach further in
the next section.
One author who has treated of both areas in parallel ways is Irmak
(2012, 2019).7 Irmak claims that most philosophical accounts of software
have failed ‘to capture the nature of software as an artifact’ (2012, p. 56),8
mentioning Moor and Colburn in particular. Further, he says that ‘soft-
ware is a kind of abstract artifact which fails to fit the classical Platonic
picture of abstract objects [….] because it is not an eternal and mind-
independent entity but is created by human beings with certain intentions’
(2012, p. 56). A realist, such as Wetzel (2006) or Katz (1998), will at least
minimally agree that artefacts have temporal properties and so could not
be non-spatiotemporal abstracta. However, Irmak’s ontology recognises
both these created temporal abstracta that he calls ‘software’ and the con-
crete physical objects that he calls its ‘copies’. On Irmak’s account, it is the
temporal abstracta that he calls ‘abstract artefacts’, while presumably also
accepting that the concrete copies are also artefacts of some other variety.
In contrast, Irmak thinks that algorithms are indeed ‘Platonic’ abstracta,
that is, eternal and mind-independent, and says that he assumes this with-
out an argument for the position (2012, p. 59). He says that there is a
type-token relation between an algorithm and a particular instance of it,
but he is largely unconcerned with the natures of types and tokens and
168 Keith Begley

follows others in taking types to be abstract universals and tokens to be

concrete. Nevertheless, he does leave open the possibility that there may be
some abstract tokens (2012, p. 59). He also mentions that a similar distinc-
tion holds for the text or code of software that is written in a particular
programming language (p. 60).9
Irmak argues that software is created, and so is not an abstract object in
the ‘Platonic’ sense, and, further, that it can survive ‘changes in its algo-
rithm or text’ (p. 68). So, in the absence of a theory of changing types or
universals, he concludes that it is not a type or universal. Thus, he thinks
that ‘the relation between computer programs and their physical copies
cannot be understood in terms of the type/token distinction’ (p. 68). Why
he thinks that the distinction nevertheless applies to the code of these
changing programs is not addressed (cf. p. 60). He argues that software
can be destroyed if all of its copies, executions, and memories of it are de-
stroyed. Thus, he thinks that software is abstract because it lacks spatial
location, but nevertheless has temporal location (p. 70).

7.3 Katz’s Realist Ontology of Composite Objects

In this section, I will argue against Irmak’s view, which was presented in
Section 7.2, and in favour of Katz’s ontology of composite objects before
applying it to software in Section 7.4. Perhaps the greatest tool available to
philosophers is the principle of non-contradiction (PNC), first put forward
explicitly in terms of negation by Aristotle (Metaphysics, Gamma 3–4,
1005b19–1006b34). However, as employed in metaphysics, and not merely
as a logical principle, it is often not appreciated that PNC is not simply a
prohibition on the positing of contradictory objects or facts, which it cer-
tainly is, it is also a heuristic by which parsimonious ontological divisions
can be made. That is, contradictions must be contradictions with reference
to one and the same respect of a unitary thing.10
For example, take Irmak’s notion of software as not being identifiable
with a concrete object, but also being a created entity under certain inten-
tions, that is, having a temporal beginning. On the standard distinction
between abstract and concrete, he claims that this is a contradiction. Irmak
then effectively interprets PNC as being a reason to revise the distinction to
allow for abstract objects that are temporal (Irmak, 2012, p. 56). However,
this is not the only option available and should not be the first port of call
for theory modification in view of the fact that the distinction is so funda-
mental in metaphysics, and so revisions to it can have global theoretical
effects extending beyond the local problem under consideration.11 A
fortiori rejecting PNC is not considered to be a further option here either.
A potential third option is instead to interpret PNC as being, for example,
a reason to analyse merely the object in question with a finer grain and so
 Towards a Realist Metaphysics of Software Maintenance 169

to dissolve the contradiction by way of some further analysis. On this view,

it is correct to predicate being temporal of a piece of software relative to a
certain aspect or component of it or, rather, its copies, and also to predicate
being atemporal of a copy relative to a different aspect or component of it.
This is analogous to my standing straddling the threshold of a room and
being said to be both inside and not inside the room. It would be proper in
such a case to distinguish different parts of me and apply the predicates to
me but only relative to those parts (Politis, 2004, p. 145).
In his work Realistic Rationalism (1998, hereafter RR),12 Katz intro-
duces a category of objects that he calls composite objects, which are
heterogeneous with respect to the abstract-concrete distinction by having
both abstract and concrete constituents. Katz’s theory of composite objects
effectively provides us with a finer-grained analysis of tokens, including
artefacts, which is in accord with the latter kind of theorising described
above, and for which there is no need to revise the usual distinction be-
tween abstract and concrete. So, in addition to what he called objects that
are homogeneous with respect to the distinction, that is, abstract and con-
crete objects, Katz argues that there are objects that are heterogeneous
with respect to the distinction, those of a third category, namely, compos-
ite objects. Thereby, the distinction remains unrevised, but these three cat-
egories are nevertheless disjoint and jointly exhaustive (Katz, 1998, p. 145).
A composite object is formed by virtue of a creative relation, or a pattern
of such relations, holding among its component objects. That is, compo-
nents are the relata of creative relations (1998, p. 141). Katz says that ‘In
the case of artifacts, the intention of the producer or the intention of the
user can be an aspect of the creative relation’ (1998, p. 141). Nevertheless,
the intention must be of the right sort. Katz uses the example of a dog
swallowing biscuits because they taste pleasant, but which does not create
a new object having the dog’s stomach and the biscuits as components
(p. 141). In the case of tokens, the intention must be to form a token of a
specific type.
Further, some relations, such as ‘number of, identity, between, greater
than, and inside of’ (1998, p. 141), are uncreative because they do not in-
volve ‘the notion of some further object of which the objects satisfying the
relation are components’ (p. 154). Katz uses the example of 17 cherries
falling from a basket, which would merely instantiate the number 17 and
not create a further object with the 17 cherries and the abstract object, the
number 17, as its components (p. 154).
There are three kinds of composite objects: simple, complex, and com-
pound. Simple composite objects have only homogeneous components,
that is, only abstract and concrete objects (and, of course, at least one of
each). Complex composite objects have both homogeneous and heteroge-
neous objects as components. Compound composite objects have only
170 Keith Begley

heterogeneous objects as components, that is, at least two composite ob-

jects (Katz, 1998, p. 142). It is important to note here that the latter two
kinds of composite objects can be hierarchically structured objects having
many layers because these composite objects can have other composite
objects as components. This makes them very useful for explaining the
kind of linguistic structures that we find in natural and artificial languages,
and, indeed, as I shall show, in software.
The ontology of composite objects suggests the following general prin-
ciple for specifying identity conditions: ‘Composite objects are identical
when they have the same components and the same creative relation holds
of them in the same way’, that is, ‘corresponding objects saturate the same
places of the relation’ (1998, pp. 145–6; cf. p. 167). It also suggests a gen-
eral principle for specifying existence conditions: ‘The condition for the
existence of a composite object is the sum of the conditions for the exis-
tence of their components plus the condition that the relevant creative rela-
tion obtain among them’ (1998, p. 146; cf. p. 167).
Composite objects can be of different denominations depending on their
components and the relations involved. For example, impure sets are com-
posite objects formed by a containment relation holding between the null
set, which is an abstract object, and the other objects and sets that are
contained by the null set (1998, p. 142).13 On Katz’s account, tokens are
composite objects formed by a type-token relation and (in at least the sim-
ple case) have a type as an abstract component. This makes them a signifi-
cantly different kind of entity to those posited by other accounts of the
type-token distinction, which are often taken merely to be physical objects.
Tokens on this view are categorized in virtue of their type components or
constituents. That is, on the basis of whether the type or types involved are
mathematical, logical, linguistic, or musical, etc., or some combination of
kinds of type (1998, p. 146). As we will see in Section 7.4, we may also add
computational objects to this list. In Katz’s view, types are not universals
and tokening is distinct from instantiation (1998, pp. 154–5). In the case
of instantiation, an object instantiates a property by virtue of its relation to
a universal. In the case of tokening, a composite object co-possesses many
of its properties by virtue of having the components that it does. Tokens
have certain properties of their abstract components in a relative way, that
is, relative to those abstract components, whereas the abstract components
themselves have them in an absolute way. The reason that tokens appear
to have inconsistent sets of properties, and why they may seem to be both
abstract and concrete, such as in the case of the dual or ‘hybrid’ nature of
software discussed earlier, is that these properties are possessed relative to
the appropriate component. That is, it is as much of a genuine contradic-
tion as saying of me that I am standing both inside and not inside the room
when I am straddling the threshold, which is not a contradiction at all
under the correct analysis (Politis, 2004, p. 145).
 Towards a Realist Metaphysics of Software Maintenance 171

7.4 Types, Tokens and Software

In this section, I will apply Katz’s ontology to software and its copies. Most
of the remarks that Katz made about computation in his writing tended to
be included as parts of technical discussions of linguistic theory. However,
in a footnote of an article co-authored with Postal from 1991, in which the
ontological status of language is a central focus, they compare linguistics
to computer science. In their view, the core of theoretical linguistics is a
formal science of natural languages (NLs), but there are also other areas of
linguistics.

[…] historical linguistics, sociolinguistics, etc., are genuinely empirical

because they concern the relation between NLs and the people who
know them. […]
Note that computer science similarly has a dual nature: part of it
studies physical objects (e.g., computer hardware) and part abstract ob-
jects (e.g., algorithms).
(Katz & Postal, 1991, p. 538: n. 11)

At this stage, Katz had not yet developed the ontology of composite objects,
which had its first and only airing in RR (1998). There he also uses that
ontology to explain and expand further upon the distinctions between
grammar and other areas of linguistics (cf. Katz, 1998, p. 165). Later in
their 1991 article, while delineating two possible readings of what they take
to be an equivocation by Chomsky regarding the nature of I-languages,14
Katz and Postal make an analogy between I-languages as abstract, mathe-
matical objects and software.

One term of the equivocation is to define an I-grammar as ‘a character-

ization of a function in the mathematical sense’. The other term claims
that I-languages are physical objects existing in space-time and entering
into causal relations. On the former term of the equivocation, I-languages
are analogs of computer programs; they are software. On the latter, they
are analogs of physical states of computers which instantiate programs.
They are then hardware. But they can’t be both.
(Katz & Postal, 1991, p. 544)

Ultimately, their argument is that conceptualism, which is the position

that claims that NLs are I-languages, ‘must limit itself to the hardware
interpretation of both grammars and sentences. For the software interpre-
tation is tantamount to realism’ (p. 549), which is the position that NLs
are neither E- nor I-languages but something like P-languages.15 It is quite
striking here that software, computer programs, are used as paradigmatic
examples of an abstract object. This analogy was never pursued further in
172 Keith Begley

later work. However, I will now apply the later ontology of composite
objects to this underdeveloped position and attempt to extend it in line
with how Katz later develops the metaphysics of linguistic entities in RR.
The aim will be to have the main aspects of the theory in place, so that we
may then attend to the notion of software maintenance further on that
basis. Further details and developments extraneous to this aim will be left
for another time.
Notwithstanding his taking software to be an abstract object, Katz
would have recognised that software can have copies that are executed by
computers. In his ontology, these copies would be tokens of a certain type.
Irmak denied that this was possible because he claimed that types could
not accommodate change. This is for the most part accurate in that, for
Katz, types are unchanging abstract entities. However, Irmak’s general on-
tological position is not novel. Indeed, Katz responded to a very similar
position that was put forward by Hale (1987), which claimed that things
like games and languages, although abstract, have temporal location be-
cause they are created (cf. Katz, 1998, pp. 132–3). Katz also responds to a
very similar charge that was made by Hale against the notion that lan-
guages are abstract objects. Hale’s concern was that realism requires us
‘to regard what we are pleased to call the history of the English language
as really a record of a succession of replacements of one language by an-
other’ (Hale, 1987, p. 49; cf. Katz, 1998, p. 137). Katz responded that in
the final analysis of what is surely a matter for linguistics, the individuation
of languages will be explained in terms of individuating sets of sets of sen-
tence types. For example, in the case of English, it is such a set that includes
relations between idiolects, dialects, and languages. The history of English
is a history of its speakers’ changing competence in a succession of sets of
sentences (1998, pp. 137–8; cf. Katz, 1981, p. 9).
Similarly, we can think of, for example, the named items of software
that Irmak mentions, such as Windows 7 (no longer maintained), as a set
of sets of interrelated program types that collectively individuate that soft-
ware. In the same way that it would be up to the linguist to determine the
exact set in the case of a natural language, it is up to the software engineer
or other specialists to determine the exact set in the case of Windows 7.
The essential message here is that whatever abstract means of explanation
or whatever configuration of sets of types is needed is available to them
and this avoids the problems raised by Hale and Irmak, respectively.16 In
the case of software, the proper arrangement will of course be compli-
cated. It may resemble something like a branching Git tree,17 which has
many more actual versions and development lines than those that eventu-
ally become available to the public, some of which are simultaneous lines
of development. Indeed, there are many more possible trees, potential lines
of development and maintenance, which are never realised. Irmak, perhaps
 Towards a Realist Metaphysics of Software Maintenance 173

unwittingly, proposed to explain all this complexity by way of only a sin-

gle changeable abstract object. Although it is clear that such an ontology
would allow for a single line of change, it is unclear how, if at all, it could
account for copies of previous versions of the software as still being copies
of the software. The software, on Irmak’s account, a temporal abstract
object, has changed and so is no longer instantiated by the ‘copy’ in ques-
tion. In light of the above, we should adopt the distinction between the
software and the individual computer programs (or systems of multiple
programs and their documentation, etc.) that make up the tree, each dis-
tinct from the other but each a node of the tree. Software then is properly
identified as being a set of sets of program types (and other ancillary types)
to an arbitrary depth.
Attending to the program types themselves, we can see that they are
made up of algorithms, that is, generally, finite progressions of operation
types.18 As we saw earlier, although Irmak accepts that algorithms are non-
spatiotemporal abstract objects, he neglects to provide an argument for the
claim. Nevertheless, it is clear that they must be, because algorithms, and
progressions of operation types more generally, are what are under discus-
sion in, for example, formal program verification and analytic determina-
tions of computational complexity, just as other kinds of abstract objects
are under discussion in other kinds of mathematical proof (cf. Katz, 1998,
p. 167). One cannot with certainty formally verify a sequence of physical
events in this sense due to their dependence ‘upon the causal properties of
physical systems, whose presence or absence is only ascertainable by means
of inductive procedures’ (Fetzer, 1988, p. 1059). It follows that algorithms
are not sequences of physical events, rather, they are abstract progressions.
As we will see later, in Section 7.5, this also has the consequence that soft-
ware maintenance must be in part a formal discipline that enhances the
efficacy of software.
Program types are analogous to discourse types in Katz’s theory of lin-
guistic entities, which are progressions of sentence types.19 Discourses are
things such as novels, poems, speeches, etc., and compositions constructed
by arranging sentences and expressions of a language. Such types are to-
kened by a compound composite object, the third basic kind of composite
object mentioned earlier. A succession relation holds as the creative rela-
tion between the composite objects that are the components of the token.
In the case of discourse tokens, these are often sentence tokens or other
expressions arranged in a sequence that is in accord with the progression
of the discourse type (Katz, 1998, p. 167). Analogously, in the case of pro-
grams, operation tokens are arranged in a sequence that is in accord with
the progression of the operation types in a program type.
Program types may also be represented in the high-level code of a pro-
gramming language, which has the property of being able to be compiled,
174 Keith Begley

interpreted or otherwise translated into low-level assembly code and as-

sembled into machine code. The high-level code itself does not have (or
need not have) this low-level code as a component. Rather, the components
of the high-level code are the character set in use, including, for example,
alpha-numeric characters and various special characters, which are them-
selves tokens of those character types. At this level, such objects are close
to those of natural language representations, and may even contain some
natural language, which is what makes them useful to programmers. The
code at any level may also be displayed to a user via a printout on paper or
a monitor, etc., which is effectively a translation by some pointillistic algo-
rithm designed for this purpose. Each character is translated into a certain
sequence of flashes of light or dots on the page, firing positions, etc., which
together re-token the character in a new physical medium.
The operation tokens that are the components of token programs are
complex composite objects that are basic or atomic manipulations of data.
Their components are data objects, which are composite, and abstract op-
eration types. These can be more or less complex. Data objects may take
various forms depending on the use case. Usually, data objects are binary
sequences that are compound composite objects. At the most basic level,
each token binary value in such a sequence is a simple composite object. Its
components are certain physical quantities, be they charges, pressures,
voltage ranges, positions, punched holes, etc., and an abstract binary value
type. The concrete physical states may happen to instantiate the items of
the binary sequence, that is, be counted zero or one, but this is not what is
operative, because they could just as well be used to token the binary value
of the opposite type than they do. Rather, they token the values that they
do in virtue of the way in which the computer system was set up, that is,
as a product of the intentional acts of some company of designers and en-
gineers. Similarly, a certain sequence of tokens of binary values may token
a compound composite object, a data object that is a binary numeral of a
certain type or a coded instruction.
Notice here that the operation token itself may be a complex composite
object, while its coded instruction may be a long compound composite
object, for example, written in machine code. So, the ontology accords
with the well-founded intuition that these are distinct objects/events. The
arrangement of the hardware of the computer too may be coded as further
binary sequences, compound composite objects. After all, the states and
transition functions of a computer are abstract objects that can themselves
be encoded as a binary sequence. It should also be considered that comput-
ers can be emulated as part of a program running on another computer.
Famous examples include Turing machines built in Conway’s Game of Life
(Rendell, 2016) and computers in game worlds, such as redstone comput-
ers in Minecraft. Such cases can also be explained using the ontology of
 Towards a Realist Metaphysics of Software Maintenance 175

composite objects. In these cases, the role otherwise played by the concrete
physical layer is played by a layer of composite objects. For example, in-
stead of simple composite objects tokening binary values, complex com-
posite objects will play the same role. What matters is that tokens formed
in each of these ways will be of the same type. We will leave further discus-
sion of these more complex areas of investigation aside for present pur-
poses,20 except to say that ultimately the distinction between hardware and
software is really one based on what we choose to encode and token where
and in what medium. At a certain level of abstraction, the distinction
breaks down. This is shown in the ontology by there being a close similar-
ity between the composite objects involved in hardware and software, at
the particular level of their involvement.

7.5 Software Maintenance
With this ontology for software in hand, we can proceed to ask what it
can tell us about the metaphysics of software maintenance. On the realist
account, that I have developed above by employing Katz’s ontology, we
can see that software is an abstract object and its copies are tokened by
composite objects that have a hierarchical structure. Software mainte-
nance then is a discipline that on the one hand has a formal side that at-
tempts to discover and thereby token a range of valid program types
appropriate and peculiar to the software, that is, in accord with a specifi-
cation or specifications of its functions. The software that is being main-
tained in this sense is a set of program types that encompass the range of
valid realisations of a specification or, more broadly, an intention for a
computer to perform a certain computation or computations. The set and
the program types remain unaffected by this, as they are abstracta. All of
these are relations that are extrinsic to the types in question. On the other
hand, software maintenance is a discipline that constrains any copies pro-
duced to the tokening of those program types. For example, corrupted
copies must be repaired and, in the most usual case, lower versions must
be updated to higher versions.
Let us now address Young’s problem for this view, especially with re-
gard to how the nature of software maintenance relates to the nature of
software, or its copies, as artefacts that are constituted in part by way of
an intention. Young (2021) has been rightly concerned to show that the
usual conception of maintenance as preservation or conservation fails to
recognise the full role that maintenance plays in the production of arte-
facts and their efficacy. It is perhaps understandable why, say, a civil engi-
neer might have such a one-sided conception of the role of maintenance in
the production of, for example, a bridge. Such an engineer produces the
bridge to a certain specification and intention. At that point, the bridge is
176 Keith Begley

complete and, in their eyes, the primary maintenance role to be played is

conservative and restorative against the usual vicissitudes of physical mat-
ter (cf. Young, 2021, p. 359). However, the case of software engineering is
importantly different because the role of maintenance is difficult to ignore,
and software engineers are under no illusions about its importance and
character.

This distinction between development and maintenance is increasingly

irrelevant. Very few software systems are completely new systems, and
it makes much more sense to see development and maintenance as a
continuum. Rather than two separate processes, it is more realistic to
think of software engineering as an evolutionary process […] where
software is continually changed over its lifetime in response to changing
requirements and customer needs.
(Sommerville, 2016, p. 60)

From the very outset of their basic training – I happen to have been through
such training myself – software engineers are told that the initial software
development for any project is merely the tip of the iceberg, representing
only 20%, and that 80% of the effort will be spent doing maintenance
(Pfleeger & Atlee, 2010, p. 540), usually involving code that they did not
write themselves. Of this maintenance, approximately 24% will be the cor-
rection of errors, another 19% will be necessary for adaptation to chang-
ing environments, but 58% will be for the addition of new features and
other enhancements (Sommerville, 2016, p. 272; cf. Pfleeger & Atlee,
2010, p. 546). Hence, they will often expend over twice as much effort do-
ing maintenance that adds enhancements as they did on the initial develop-
ment. Software engineers are only too aware of the reasons for this and are
well ahead of the philosophers in this regard. One reason is that program-
mers generally prefer to work on new projects rather than on maintaining
old ones. This can lead to the break up of teams and thereby the loss of
experience and knowledge of the system to be maintained, which must
then be acquired by any subsequent maintainers at some not insignificant
cost of days or weeks of work. The distinction between development and
maintenance also results in there being little incentive to write easily main-
tainable software, because it will most likely be another team that ends up
maintaining it (Sommerville, 2016, pp. 90, 272–3). The general attitude to
software maintenance is evinced by the following passage.

[…] these problems stem from the fact that many organizations still
consider software development and maintenance to be separate activi-
ties. Maintenance is seen as a second-class activity, and there is no in-
centive to spend money during development to reduce the costs of
 Towards a Realist Metaphysics of Software Maintenance 177

system change. The only long term solution to this problem is to think
of systems as evolving throughout their lifetime through a continual
development process. Maintenance should have as high a status as new
software development.
(Sommerville, 2016, p. 273)

As we saw earlier, Young instead places the blame for the misunderstand-
ing of the role of maintenance generally on the ontological conception of
artefacts as realisations of design intentions, supposedly due to a perni-
cious kind of Aristotelianism that sees change as antithetical to a com-
pleted artefact that fits its form and essence (2021, p. 359).21 Many acts of
maintenance of artefacts, conservative or otherwise, do nevertheless fulfil
something of the initial intention. That is, an intention and what is in-
tended need not extend merely to the initial development of an artefact. It
is clear that an intention can also encompass an artefact’s maintenance.
Indeed, Young agrees that artefacts are often designed with the expectation
of future and unanticipated changes (Young, 2021, pp. 361 ff.). Although
I agree with Young that a rebalancing of our conception of maintenance
may be warranted in order to take account of it in all its aspects, the cause
of the imbalance has, I believe, been misidentified.
Young’s underlying claim is that there is an entailment between the con-
ception of the nature of artefacts as realisations of their producers’ inten-
tions and the conception of the maintenance of those artefacts as being
solely conservative.22 A corollary of this claim is that if Young is correct
that maintenance is not merely conservative, which he surely is, then it fol-
lows that many, if not all, artefacts are not realisations of their producers’
intentions. So, seen in this light, the claimed entailment constitutes a
broader claim about the nature of artefacts and their production. This
consequence could only be drawn if it were shown that such an involve-
ment of intentions in a theory of artefacts is inconsistent with maintenance
having a non-conservative aspect, which, I take it, has not been shown.
This is, I think, important for the possibility of the realist view of software
and software maintenance for which I have argued, because they involve
the employment of intention in the formation of tokens of an abstract type.
Furthermore, I take software engineering and software maintenance to
be sources of straightforward counterexamples to the claimed entailment.
In almost all cases, the producers of software intend for it to be periodi-
cally updated and/or enhanced, and in order to implement changes to
software, knowledge of its intended purpose and design is required
(Sommerville, 2016, p. 90). The exact nature of future changes could out-
strip the producer’s knowledge of the software (on the Katzian view out-
lined, a complicated abstract object) of which they intend to produce a
token. Sometimes, it is choreographed to some degree in advance what
178 Keith Begley

features, etc., will be added at certain stages, but often it is not. Nonetheless,
there is ab initio an intention to maintain the software in an adaptive or
enhancive way, or continually to develop it throughout the duration of its
use, as Sommerville advocates. This intention can also be observed more
widely in the practice of releasing alpha and beta versions, and sub-unitary
versions (for example, version 0.1), which are in one sense complete and
functioning pieces of software, but are also intended to be further devel-
oped and enhanced. That is, such realisations of an intention or intentions
need not be an all or nothing linear affair. The thing intended is an abstract
structure that can be more or less fully comprehended, adumbrated, and
realised due to its complexity, that is, both single-program complexity and
complexity of the tree of programs, etc., that make up the software accord-
ing to the view. The force of such counterexamples is that the entailment
between the intentional conception of an artefact’s production and its
maintenance being conceived of as conservative only holds in cases where
the intentions are restricted to the realisation of a ‘fully completed’ arte-
fact, so to speak, and where it is not in the nature of the artefact to be pe-
riodically enhanced, etc. Perhaps, the entailment holds for an ordinary
table or a bridge, so conceived, but it does not hold for most software,23
which, as we are all too aware, and expect, is continually patched, up-
dated, and upgraded. It is also questionable whether the entailment could
hold even for such artefacts as tables and bridges. The same question
should be asked of these cases, regarding whether the producers of these
artefacts intend for them to be maintained (whether or not they under-
stand that this can often be adaptive or enhancive and not merely conser-
vative). Perhaps, they even take intended maintenance into account when,
for example, choosing their materials and other design features and con-
straints. So, it seems likely that they often do so intend. In any event, a
single counterexample is all that is needed to show that a generalised ver-
sion of the claimed entailment is false.
This is not to say that there are no entailments of worth between the
nature of software, or the production of its copies, and the nature of its
maintenance. I have already pointed to examples of such entailments. The
reason why an aspect of software maintenance, an aspect that is involved
with increasing software’s efficacy, is a formal discipline aimed at verifica-
tion and determination of computational complexity is because software is
an abstract object made up of sets of sets of program types. The non-
abstract constituents of its copies are not amenable to such a formal disci-
pline because they depend upon causal properties that are ascertainable
only by inductive means. In that case, wider considerations regarding the
environment and use case are taken into account. For example, whether
the software’s ability to correct errors due to interstellar radiation impact-
ing a space probe is a pertinent factor or whether it will matter for the use
 Towards a Realist Metaphysics of Software Maintenance 179

case. All of this kind of adaptive maintenance is distinct from whether the
algorithms involved in the software are formally verifiable, because it is
about a distinct constituent of the composite object.
If Young’s claim is not the reason for the imbalanced view of mainte-
nance, then we should ask what is. It is more likely that something like the
situation that Sommerville presents, as outlined earlier, is the correct diag-
nosis of the tendency to view maintenance as solely or primarily conserva-
tive. So, on this view it is some feature or features of our society, market
economics and perhaps even our psychology, etc., which is the cause of
the imbalanced view. This becomes all the more stark when the imbal-
anced view is set against the extent of adaptive and enhancive mainte-
nance practices involved with something like software. The cause of taking
maintenance to be solely conservative is not any particular philosophical
understanding (or misunderstanding) of the nature of artefacts generally,
and especially not that they are beholden to intentions and what is in-
tended. The example of software maintenance, I think, shows that this
could not be the case because the software engineers have precisely a
design-oriented perspective on their artefacts, though they are pragmatic
about what can be achieved given the complexity of their task (cf. Young,
2021, pp. 361 ff.). They discuss such designs and specifications at length
and form elaborate practical theories about the best methods of formulat-
ing and achieving them (for example, Waterfall, Agile, etc.). It is also clear
that they are not under any illusions regarding the important role of main-
tenance, including its non-conservative, adaptive and enhancive aspects,
because this is what they spend the majority of their time doing.

7.6 Conclusion
In this chapter, we have seen that while many authors have taken compu-
tational objects like software to have dual or ‘mixed’ natures, this should
be taken as an ontological puzzle to be solved. Some proposals to solve this
puzzle have not been very fruitful or have skirted the main issue regarding
the abstract-concrete distinction. Other proposals, such as Irmak’s, have
taken more drastic measures in reformulating the traditional distinction
between abstract and concrete objects to include temporal abstracta. I ar-
gued that this was based on what is an unwarranted appeal to PNC, in
view of the fact that we could instead opt for a finer-grained analysis that
left the distinction unchanged. Such a finer-grained analysis is provided by
Katz’s ontology of composite objects. I argued that this ontology could be
used in a way roughly analogous to how Katz uses it to explain linguistic
entities in order to explain computational objects such as software.
This allowed us to see that the notion of a realist metaphysics of soft-
ware maintenance is not as otiose as the initial worries might have had us
180 Keith Begley

believe. We saw instead that software can be understood as being a set of

sets of related program types in accord with a specification or, more
broadly, an intention. Software maintenance, on this view, involves both
discovery of a valid range of program types in accord with this intention
and also the use of programming and engineering to token the appropriate
types that are in accord with this set. We also saw that software engineers
have an understanding of maintenance as being continuous with develop-
ment, and in which maintenance is not merely conservative and about
fixing errors but is in fact mostly adaptive and enhancive. Software main-
tenance thereby constitutes a source of counterexamples to Young’s
claimed entailment between the conception of the nature of artefacts as
realisations of their producers’ intentions and the conception of the main-
tenance of those artefacts as being solely conservative. The reason for that
imbalanced view of maintenance as being solely conservative is much more
mundane, and primarily due to factors stemming from our society, eco-
nomics, and psychology, rather than our understanding of the nature of
artefacts and their production.

Acknowledgements
The author wishes to thank three anonymous reviewers and the editors for
their comments on drafts of this chapter.

Notes
1 For the purposes of this discussion, I leave to one side the issue of laws and the
nature of physical objects, etc.
2 The philosophical view that they express is ambiguous, perhaps between a con-
ceptualism and a nominalism about abstracta. For example, they say that ‘the
term abstraction refers to an idea or general description, divorced from physi-
cal objects. [….] although these abstractions such as “word processors” and
“documents” are merely convenient ways of describing patterns of electrical
activity, they are also things that we can buy, sell, copy, and use’ (1999, p. ix).
This is at least an odd use of the word ‘merely’. Colburn seems not to recognise
this ambiguity in their position, for he is more clear that there is an apparent
incompatibility at play in the notion of a concrete abstraction.
3 It is perhaps worth observing in passing that there is a potential problem with
Turner’s appropriation of Kroes’ work because Kroes explicitly excluded soft-
ware from his notion of a ‘technical artefact’: ‘software programs fall outside
the scope of this book. I consider software programs to be “incomplete” tech-
nical artefacts; only in combination with the appropriate hardware that exe-
cutes software programs are they able to fulfil their technical function’ (Kroes,
2012, p. 2: n. 4).
4 See especially: Achterhuis (2001); and Verbeek (2022), for an overview.
5 Katz himself raises the question earlier, regarding meaning, in his Semantic
Theory, but leaves it without a ‘final answer’ (1972, pp. 16 & 38).
 Towards a Realist Metaphysics of Software Maintenance 181

6 For an overview of some of the central questions in the philosophy of linguis-

tics, see Stainton (2014). For a recent survey of the ontology of words, see
Miller (2020). Also, see the recent Synthese topical collection on ‘The
Philosophy of Words’: https://link.springer.com/collections/cgicihfcie
7 I respond to Irmak’s ontology of words (2019), and in particular his criticism
of Katz in that regard, in Begley (2023). So, I will not get into those arguments
here, except where there is a direct connection.
8 Irmak says that ‘an object is an artifact if and only if it is an intentional product
of human activity’ (Irmak, 2012, pp. 55–6). He claims that this is a widely
accepted definition and references Hilpinen (2011) for further discussion.
However, Hilpinen discusses a number of definitions and debates, and it is not
clear that Irmak’s definition is widely accepted. Further, Irmak’s definition
would seem to be an overproductive one. That an artefact is an intentional
product of someone’s activity is perhaps necessary, but is it sufficient? Counter­
examples and difficult cases abound. For example, are smiles or incredulous
stares artefacts?
9 Irmak quotes what he says is the text of a program for adding two values that
is written in the language C++, but what follows is the text of a program writ-
ten in assembly language (cf. Irmak, 2012 p. 60). Irmak here appears to mis-
read Colburn (2000, p. 202), without citation.
10 For a thorough explanation of this principle in the context of Aristotle’s
Metaphysics, see Politis, 2004, pp. 122 ff.
11 I believe that Katz provides such an example: ‘An ontology that says that all
concrete objects are corporeal would preclude concrete objects that are mental
in the Cartesian sense, and hence make the classical mind/body problem unstat-
able’ (1998, p. 125). By making spatial properties the only mark of the con-
crete, Irmak’s ontology is of this kind.
12 Much of my discussion here has benefitted from a parallel project to reconstruct
Katz’s metaphysics of words. For further in-depth discussion, see Begley (2023).
13 Katz is careful to distinguish the containment relation from membership and
inclusion, as it is more general than those relations and should not be confused
with them (1998, p. 142). See Begley (2023) for further discussion.
14 Chomsky introduced the distinction between E-language, I-language, and
P-language in part to provide clear and concise labels for categories introduced
in his earlier work. E-language or ‘externalized language’ is a construct ‘under-
stood independently of the properties of the mind/brain’, including ‘the notion
of language as a collection (or system) of actions or behaviors of some sort’
(1986, p. 20). I-language or ‘internalized language’ is ‘some element of the
mind of the person who knows the language, acquired by the learner, and used
by the speaker-hearer’ (p. 22). P-language or ‘Platonic language’ is the name he
gives to ‘something else additional […] existing in a Platonic heaven alongside
of arithmetic and (perhaps) set-theory’ (p. 33), which he says is claimed by Katz
and others (p. 49).
15 See footnote 14 above.
16 For a further discussion of Katz’s response to Hale on the issue of natural lan-
guage, see Begley (2023).
17 Git is an open-source distributed version control system that was invented by
Linus Torvalds in 2005 for the development of the Linux Kernel. It is now
widely used for software development, and on repository websites such as
GitHub. However, as Irmak rightly points out, there are many other methods
(Irmak, 2012, p. 63).
182 Keith Begley

18 Katz tends to reserve ‘progression’ for reference to the abstract types, and
‘sequence’ for the tokens.
19 Northover et al. (2008, p. 93) make a broadly similar analogy between token
performances or readings of a novel and token executions of computer
programs.
20 For a full discussion, and with regard to virtual reality, see Begley
(forthcoming).
21 Whether this is an accurate interpretation of Aristotle or not is a vexed issue in
the literature. Aristotle might not have taken artefacts to be oūsiai (primary
beings or ‘substances’) (cf. Metaphysics, Book Eta 3, 1043b22). In any event, it
does not appear to be essential to the point at issue about intentions.
22 As we saw in Section 7.2, Young’s argument as presented refers to our under-
standing of artefacts and how it ‘encourages’ us to think about maintenance
(2021, p. 359), rather than referring to an entailment between the two concep-
tions. However, it will be more charitable to assume that such an underlying
entailment is claimed, especially as Young has confirmed this to me (personal
communication).
23 There are of course edge cases. For example, if I were to install a copy of
Windows 7, or some other final version of now unsupported software, I would
be creating a token the maintenance for which will be conservative, if any
maintenance occurs at all.

Bibliography
Achterhuis, H. (ed.) 2001. American Philosophy of Technology: The Empirical
Turn. Bloomington: Indiana University Press.
Aristotle 1952. Metaphysics. Trans. Richard Hope. New York: Columbia University
Press.
Begley, K. 2023. ‘Katz Got Your Tongue? The Metaphysics of Words’, Synthese,
202(4): 107. https://doi.org/10.1007/s11229-023-04324-x
Begley, K. forthcoming. ‘The Metaphysics of Communication in Virtual Reality’.
In: T. Ahlers & K. Senkbeil (eds.), Virtual Reality in den Geisteswissenschaften –
Konzepte, Methoden und interkulturelle Anwendungen [Virtual Reality in the
Humanities: Concepts, Methods and Intercultural Applications]. Peter Lang.
Chomsky, N. 1986. Knowledge of language: Its nature, origin and use. New York:
Praeger.
Church, A. 1936. ‘An Unsolvable Problem of Elementary Number Theory’,
American Journal of Mathematics, 58: 345–363.
Coeckelbergh, M. 2022. ‘Using Philosophy of Language in Philosophy of Tech­
nology’, In: S. Vallor (ed.), The Oxford Handbook of Philosophy of Technology.
Oxford: OUP, 341–357.
Colburn, T. R. 2000. Philosophy and Computer Science. London: M. E. Sharpe.
Fetzer, J.H. 1988. ‘Program Verification: The Very Idea’, Communications of the
ACM 37(9): 1048–1063.
Gruner, S. 2011. ‘Problems for a Philosophy of Software Engineering’, Minds &
Machines 21: 275–299.
Hailperin, M., Kaiser, B., Knight, K. 1999. Concrtete Abstractions. Pacific Grove:
Brooks/Cole. https://www.gustavus.edu/+max/concrete-abstractions.html
 Towards a Realist Metaphysics of Software Maintenance 183

Hale, B. 1987. Abstract objects. Oxford: Basil Blackwell.

Hilpinen, R. 2011. ‘Artifact’, The Stanford Encyclopedia of Philosophy. https://
plato.stanford.edu/archives/win2011/entries/artifact/
Irmak, N. 2012. ‘Software is an Abstract Artifact’, Grazer Philosophische Studien
86: 55–72.
Irmak, N. 2019. ‘An Ontology of Words’, Erkenntnis 84: 1139–1158.
Katz, J.J. 1972. Semantic Theory. New York: Harper & Row.
Katz, J.J. 1981. Language and Other Abstract Objects. Totowa, NJ: Rowman &
Littlefield.
Katz, J.J. 1998. Realistic Rationalism. Cambridge, MA: MIT Press.
Katz, J.J. and Postal, P.M. 1991. Realism vs. Conceptualism in Linguistics. Lin­
guistics and Philosophy, 14(5): 515–554. https://www.jstor.org/stable/25001444
Kroes, P. 2012. Technical Artefacts: Creations of Mind and Matter. A Philosophy
of Engineering Design. Springer.
Miller, J.T.M. 2020. ‘The Ontology of Words: Realism, Nominalism, and
Eliminativism’, Philosophy Compass, 15(7): 1–13. https://doi.org/10.1111/phc3.
12691
Moor, J.H. 1978. ‘Three Myths of Computer Science’, The British Journal for the
Philosophy of Science 29(3): 213–222.
Northover, M., Kourie, D. G., Boake, A., Gruner, S., & Northover, A. 2008.
‘Towards a Philosophy of Software Development: 40 Years after the Birth of
Software Engineering’, Journal for General Philosophy of Science 39: 85–113.
Pfleeger, S.L. & Atlee, J.M. 2010. Software Engineering: Theory and Practice.
4th Edition. Upper Saddle River, NJ: Prentice Hall.
Politis, V. 2004. Aristotle and the Metaphysics. London: Routledge.
Rendell, P. 2016. Turing Machine Universality of the Game of Life. Cham: Springer.
Sommerville, I. 2016. Software Engineering. 10th Edition. London: Pearson.
Stainton, R. J. 2014. ‘Philosophy of Linguistics’, Oxford Handbooks Online.
https://doi.org/10.1093/oxfordhb/9780199935314.013.002
Turing, A.M. 1937. ‘On Computable Numbers, with an Application to the
Entscheidungsproblem’, Proceedings of the London Mathematical Society, s2-42:
230–265.
Turner, R. 2018. Computational Artifacts: Towards a Philosophy of Computer
Science. Berlin: Springer.
Verbeek, P.-P. 2022. ‘The Empirical Turn, In: S. Vallor (ed.), The Oxford Handbook
of Philosophy of Technology. Oxford: OUP, 35–54.
Wetzel, L. 2006. ‘Types and Tokens’. The Stanford Encyclopedia of Philosophy.
https://plato.stanford.edu/archives/fall2018/entries/types-tokens/
Young, M.T. 2021. ‘Maintenance’. In: D.P. Michelfelder and N. Doorn (eds.), The
Routledge Handbook of the Philosophy of Engineering. London: Routledge,
356–368.
8 Maintaining Perpetual
Actuality in the Digital Age?
Simondon’s Conception of
Maintenance and the Networked Era

Johannes F.M. Schick

‘Maintenance’ encompasses the control and repair of technical objects, a

prerequisite for this being that technical objects are constructed in such a
way that they can and must be maintained.1 This implies a distinction be-
tween living beings and technical objects; technical objects require mainte-
nance since they lack the self-sustaining and self-repairing structure of
organisms.2 While organisms are ‘concrete to begin with’, technical objects
are – as constructed beings – always perfectible (Simondon, 2017, p. 51).
These ontological and temporal aspects of maintenance are key to Gilbert
Simondon’s conception of technical objects (2017, p. 255; 2009, p. 19).
For Simondon, the ideal technical object is open and has a ‘reticular struc-
ture’ (i.e., networked), which allows it to be ‘completed, improved, main-
tained in the state of perpetual actuality’ (Schick, 2021a; Simondon, 2009,
p. 24). The construction of technical objects should already imply that
improved parts (e.g., a faster computer chip, a more efficient battery, etc.)
can and should be replaced. Technical objects should thus be open for im-
provements provided by the infrastructural networks of production.
For Simondon, ‘maintenance’ implies a temporal aspect that relates
technical objects with the present. Maintaining a technical object in this
emphatic sense means to continue the engagement with the technical ob-
ject so that it is kept – perpetually – up to date. The openness of technical
objects therefore enables a relationship of human beings towards tech-
niques in which technical objects are not something to be bought and dis-
carded, but instigate participation (Simondon, 2014a, p. 364).
Yet, while it is still plausible to think about open objects on a mechanical
level, evoking the image of bricoleurs that maintain machines and tinker
with objects of all kinds in their workshops, such objects are much harder
to come by in the digital age. ‘Maintenance’ in an etymological sense
means to hold things in one’s hands, which allows ‘manipulation’ and sus-
tainment of a certain ‘tension’ in relation to the present moment. Can these
two aspects of maintenance be more than an anachronism in the digital
age?3 On the one hand, it seems we are rarely without a smart device in our

DOI: 10.4324/9781003316213-9
 Maintaining Perpetual Actuality in the Digital Age? 185

hands, maintaining, establishing and creating relations with technical

ensembles – and with other human beings. On the other hand, we are not
manipulating the technical infrastructure directly, but rather using it to
participate in different forms of publics that can be either ‘calculated pub-
lics’ (Gillespie, 2014, p. 188) (i.e., generated through algorithms) or ‘data
publics’, where data is used for a different purpose than it was initially
generated for (Gray et al., 2018).
In Chapter 8, I will attempt to reconstruct the concept of maintenance
in Simondon’s philosophy of technology and furthermore explore how
this conception is relevant to the current technical phenomena of the digi-
tal age.
In the first part of this chapter, I will develop Simondon’s conception of
maintenance in terms of the process of ‘concretization’ and ‘technicity’ to
address the temporal dimensions of technical objects. Simondon shows
that technical objects evolve with the process of concretisation. Technical
objects have a temporal condition and tend to become ‘concrete’ (Simondon,
2017, p. 25ff.), that is, a state where the elements of the technical object are
in an intrinsic relation with each other, serve multiple functions and cannot
be replaced by any other part. The so-called ‘concrete technical objects’
increasingly run independently (they self-maintain) and create their own
milieu that allows their proper functioning. The actualisation of mainte-
nance for the digital age exemplifies the difference between ‘maintenance’
and self-maintenance (auto-entretien): smart devices are ‘concretized ob-
jects’. Their production relies on an infrastructure and their material ele-
ments are intricately linked to each other. Spare parts cannot be fabricated
by the user, programming languages need to be learned in order to access
multiple layers of the object and engagement with the technical object is
initiated through the user interface by mediated bodily movements.
‘Maintaining’ smart tools thus cannot mean manipulation of each of the
elements, but rather providing the conditions for their internal function-
ing. While specialised skills have always been necessary for maintenance,
smart tools are far harder for the user to maintain since the interaction
with the technical object occurs on the interface. The internal and external
relations with users, other technical objects and networks are what tech-
nicity refers to. These relations have to be captured to maintain technical
objects. I will thus introduce the conception of maintenance in the philoso-
phy of Simondon as a condition to entertain a relationship with technical
objects and to fathom their ‘technicity’, that is, the concrete technical as-
pects of an object’s functioning and its embeddedness in socio-technical
ensembles. Acts of maintenance do not merely refer to acts of construction,
but are intrinsically linked to their use, as Mark Thomas Young points out
(2022, p. 363f.).
186 Johannes F.M. Schick

Maintenance relies, however, not only on the skill of the user but also on
the structure of the technical object. Both the user and technical object
have to be in recursive relations with each other so that technical networks
participate in cultural networks and vice versa. To explore this reciprocity,
I will, in the second part of this chapter, introduce the analogy between
(hi)story-telling and maintenance. This analogy draws on Simondon’s ped-
agogical work and his call for ‘poets of techniques’ (Simondon, 2014b,
p. 416). I will combine these two aspects of Simondon’s philosophy to
show that maintenance ‘tells’ the (hi)story of an object and adapts it to the
actual situation. Furthermore, the notion of ‘story-telling’ provides the
means to understand Simondon’s call for a ‘technical mentality’ as an at-
titude towards technical objects and its relevance for the digital age.
In the concluding part, I will elaborate on whether Simondon’s notion of
maintenance can be actualised in an era where engagement is mediated in
the aforementioned data-driven way. This may be a point of departure
from which to understand and maintain a relation with the technicity of
the networks we are living in and is in line with current efforts to ‘reflect
over the temporal nature of technologies’ (Young, 2022, p. 363).

8.1 The Temporality and Evolution of Technical Objects

In this first part, I will turn to Simondon’s description of the evolution of
technical objects as a process of concretisation, that is, an evolution from an
abstract technical object, where each element can be replaced by another
regardless of material constitution providing this element fulfils its function,
to a more concrete technical object, where each element is in an intimate,
recursive relation with other elements. Each element of the ‘concrete techni-
cal object’ has a specific material structure and fulfils a specific function in
relation to the object’s other elements (Simondon, 2017, p. 27). This evolu-
tion thus concerns the interrelatedness of the technical object’s components.
This process has implications for the phenomenon of maintenance. On
the one hand, one could consider it an evolution of objects that increas-
ingly maintain themselves (auto-entretien) (Simondon 2017, p. 50). Less
human intervention is required to guarantee the functioning of these ob-
jects, while these enter multiple technical networks and rely on their infra-
structure (Simondon 2017, p. 50). On the other hand, maintenance of
‘concrete’ technical objects requires an understanding of how their ele-
ments relate to each other in order to guarantee proper functioning.
The evolution of a technical object establishes new relations in two direc-
tions: towards its inner constitution and milieus, as well as with the natural
world, the external milieus. These relations are modified over the course of
the evolution of technical objects. They are the expressions of a process of
 Maintaining Perpetual Actuality in the Digital Age? 187

organisation; as technical objects tend towards the self-organisation of an

organism, a network with other natural and technical objects is established
via adaption to the external and the internal milieu:

This object needed a regulative external milieu in the beginning, the

laboratory, workshop, or sometimes the factory; it gradually increases
its concretization, it becomes capable of doing without the artificial mi-
lieu, because its internal coherence increases, its functional systematicity
closes as it organizes itself. The concretized object is comparable to the
spontaneously produced object; the object frees itself from the origi-
nally associated laboratory and dynamically incorporates the labora-
tory into itself through the play of its functions; what enables the
self-maintenance (auto-entretien) of the objects conditions of function-
ing is its relation to other technical and natural objects, and it is this
relation that becomes regulative; this object is no longer isolated; it as-
sociates itself with other objects, or suffices unto itself, whereas at first
it was isolated and heteronomous.
(Simondon, 2017, p. 50)

Simondon describes a process where human beings intervene less and less
in the actual functioning of the technical object. This process can certainly
be seen – from both the perspective of use and of construction – as
progress.
The process of concretisation of the technical object has as its ideal the
organisational form of the living being. The understanding of the technical
object as a ‘unit of coming-into-being’ that is ‘present at each stage of its
becoming’ (Simondon, 2017, p. 26) is oriented towards the actual develop-
mental processes of technical objects: technical problem-solving is most
effective when new elements do not have to be added from the outside, but
new relations are invented within the technical objects, thus reducing the
complexity and the elements through ‘condensation and concretization’ to
a ‘minimum that is at the same time an optimum’ (Simondon, 2009, p. 19).
The example of the engine illustrates the concretisation process very
clearly:

One could say that the contemporary engine is a concrete engine,

whereas the old engine is an abstract engine. In the old engine each ele-
ment intervenes at a certain moment in the cycle, and then is expected
no longer to act upon the other elements; the pieces of the engine are
like people who work together, each in their own turn, but who do not
know one another.
(Simondon, 2017, p. 27)4
188 Johannes F.M. Schick

The relation of the elements to each other is transformed in the ‘concrete

engine’ and through this new relation, problems are solved. The function,
for instance, of ‘protecting engines from overheating to prevent deforma-
tion of the pistons inside the engine’ led to the problem that as long as ele-
ments for cooling were added to the engine from the outside, it necessarily
became larger, heavier and less efficient (ibid.). The solution to the problem
was to use new materials to integrate cooling fins within the piston design,
thus preventing piston deformation and providing cooling (Simondon,
2017, p. 27f.). Another example is the turbine of the GUIMBAL group:
there, the interaction of different elements (a very small alternator is placed
in a housing wrapped in oil, which in turn is placed precisely in the piping
and in the turbine) allows for more effective cooling than would be possi-
ble with generators in the air (Simondon, 2009, p. 19).5
The concretisation process describes the genesis of the technical object
as an individual. That is, as a being that possesses a genuine mode of exis-
tence, which consists of striving from an abstract to a concrete state. This
process relates the elements of the technical object to each other. Within
the technical object, natural elements (water, gas and fire) and technical
elements (cylinder pistons, etc.) interact. This recursive causality creates a
specific technical milieu that contains the potential for possible solutions to
problems, becoming increasingly stabilised and interacting with natural
elements. A technical individual therefore has an ‘associated milieu’
(Simondon, 2017, p. 59) that ensures its functioning.
While this milieu was – and still is – ‘regulative’ in craftsmanship and
relies on the tool-wielding human being, in the post-industrial age the as-
sociated milieu consists of technical ensembles and the relationship of the
human being with the technical object has been transformed. It is no lon-
ger necessary to maintain in the literal sense, that is to manually hold the
technical object in its existence, but it is possible to concentrate on differ-
ent aspects of their mode of existence: the focus can shift to the organisa-
tion of relations to other natural and technical objects. The ideal relation
of human beings with technical objects would be as a coordinator in the
midst of technical objects (Simondon, 2017, p. 17). This effort to be in the
midst of technical objects goes against the mode of production in the fac-
tory, where one either remains on the level of construction or on the level
of the manager, who deals with the integration of the technical object in
the networks beyond the factory, that is, on the level of the network
(Simondon, 2017, p. 78ff.).
Whilst the artisan may also be a ‘coordinator’ in providing the ‘regula-
tive milieu’ for his/her tools, a contemporary user of a concretised object (a
smartphone, for instance) maintains it by concentrating on a different op-
erative level, since the object ‘frees itself’ from its original space of produc-
tion and enters into a network where it can self-maintain (auto-entretien).
 Maintaining Perpetual Actuality in the Digital Age? 189

This means it creates its own ‘milieu that conditions it, just as it is condi-
tioned by it’ (Simondon, 2017, p. 59). It ‘mediates the relation between
technical, fabricated elements and natural elements, at the heart of which
the technical being functions’ (ibid.) It is thus this transformation of
milieus – both the regulative milieu, that necessitates the interference of the
human being, and the associated milieu created with the operations of the
technical object – that allows for self-maintenance (auto-entretien). The
concept of auto-entretien indicates the different relationships human be-
ings enter into with concrete technical objects. For concrete technical ob-
jects, it is no longer necessary to hold them literally in hand (main-tenir) to
keep them functioning. Instead, maintenance is directed at the relations
that allow for self-maintenance (auto-entretien). While the object was ‘in
the beginning isolated and heteronomous’ (Simondon 2017, p. 50) and
needed the support of the infrastructure provided by the laboratory, the
integration of these conditions in its functioning and the association with
other natural or technical objects generates its autonomy (ibid.).
Maintenance thus refers to an activity that provides and cares for
the requisite conditions for self-regulation and self-maintenance (auto-
entretien) of the technical object. It refers both to the internal relations of
the technical object as well as the relations with the external milieu.
This has a couple of consequences: maintenance becomes increasingly
inherent to technical objects and therefore specialised; technical objects
generally seem to tend towards an organisation that allows for specific
forms of participation. In order to understand the relationships within the
technical object, that is, the conditions of auto-entretien and the relations
with the world, specialised knowledge and skills are required. Simondon
even calls for a ‘relation of equality’ with technical objects ‘that is, a reci-
procity of exchanges; a social relations of sorts’ (Simondon 2017, 105).
Human beings should therefore aspire to understand technical objects and
their genesis.
I argue that maintenance plays a crucial role in this task. Maintenance
relates the technical object spatially and temporally within the network of
production and use. It implies a specific understanding of the ‘technicity’
of technical objects. Here, technicity does not simply refer to the inner
workings or construction of a technical object. Rather, technicity refers to
Simondon’s theory of individuation. ‘Technicity is one of the two funda-
mental phases of the mode of existence of the whole constituted by man
and the world’ (Simondon 2017, 173).6 It refers not only to the technical
constitution of the object but also shows that techniques and technical
objects are essential to the being-in-the-world of human beings and the
way in which they structure the world. Technicity refers therefore to the
relations within the technical object as well as to the relations of the techni-
cal object in the wider world. In technical objects, it is the ‘degree of the
190 Johannes F.M. Schick

object’s concretization’. It expresses the concrete relation of form and mat-

ter (Simondon, 2017, p. 72). Furthermore, technicity encompasses both
theoretical knowledge and the practices performed with technical objects
(Simondon, 2017, p. 171). As such, technicity is inscribed in technical ob-
jects. Tools and instruments can be maintained through time and even
be reinvented, since they contain crystallised human gestures (Simondon,
2005a, p. 340).
The technicity of technical objects allows for a continuous relation-
ship of technical objects with their socio-technical networks and is
rooted in Simondon’s philosophy of individuation. Technical objects – as
stated above – evolve and are conceived in analogy to organisms, that is,
living beings. Living beings are, however, open systems and entertain a
metastable relation with their milieus. This metastable relation allows
for adaptation and transformation. It is – in terms of Simondon’s
philosophy – a condition of life to remain metastable. As long as an or-
ganism is metastable, it can transform itself and has not yet become
stable; as such, it can create new ‘stable forms’ (Simondon, 2005b,
p. 161) without being prone to being reduced to one stable form. Living
beings are able to restructure themselves, they dispose of the potential of
plasticity to adapt to new situations and even to restructure their bodily
organisation (Simondon, 2005b, p. 161).
Even though a technical object cannot achieve the same level of self-
organisation as a living being, its evolution strives towards this goal
(Simondon, 2017, p. 50f.).7 Simondon warns, however, that technical ob-
jects should not be identified with living beings. The aforementioned self-
organisation of the living being cannot – according to Simondon – be
achieved by technical objects due to their status as constructed beings:
‘living beings [,] are concrete to begin with’.8 However, as an expression of
life (Simondon, 2014c, p. 321), technical objects are endowed with vitality
through their relation with human beings. According to Simondon, it is
maintenance that furnishes the technical object with a life in temporal and
spatial networks:

Similarly, the degradation of the technical occurs when the object is

isolated in time (by the break that is the end of manufacture and the fall
into the condition of venality) and in space (by the detachment that
isolates the manufactured object from the conditions in which it could
receive a perpetual regeneration maintaining it at the level of its full
functional meaning/signification). Technicity degrades as it becomes ob-
jectified because the object, as a closed object, becomes archaic and de-
grades when it is no longer maintained in the network of technicity
through which it was constituted. Technicity is a mode of being that can
only exist fully and permanently in a network, both temporally and
 Maintaining Perpetual Actuality in the Digital Age? 191

spatially. Temporal reticulation is made up of the object’s revivals, in

which it is updated, renewed, in the very conditions of its original man-
ufacture. Spatial reticulation consists in the fact that technicity cannot
be contained in a single object; an object is technical only if it operates
in relation to other objects, in a network in which it takes on the signifi-
cance of a key point; in itself and as an object, it possesses only virtual
characteristics of technicity that are actualized in the active relationship
to the system as a whole. Technicity is a characteristic of the functional
set that covers the world and in which the object takes on a significa-
tion, plays a role with other objects.9

Technicity is, according to Simondon, the central aspect of the technical

object that requires maintenance. Maintaining an object allows its technic-
ity to enter into a network. Simondon points out that the production of
objects merely to sell them creates closed objects. They become things (une
chose) as soon as they leave the factory since they are only ever new and
actual at this precise moment; after which they deteriorate and age, having
lost contact with the actual reality (Simondon 2014b, p. 401). Maintenance
thus calls for the ‘opening’ of objects in a literal and metaphorical sense:
literally it can mean attempting to fix and maintain a technical object by
opening it up and replacing components; metaphorically, it refers to a pos-
sible relationship with the technical object that goes beyond mere use and
attempts to understand its operations and mode of being. Maintenance can
thus restore the technicity of technical objects by reintegrating them into a
network. This reintegration is aimed at the temporal condition of their
genesis and the spatial condition that refers one object to others. Human
acts of maintenance thus actualise the virtual technicity within a technical
object. I, with Simondon, argue that the human technical interactions with
technical objects while maintaining them, (re-)create signification by relat-
ing technical objects to each other in a technical ensemble:

Man understands machines; for there to be a true technical ensemble

man has to play a functional role between machines rather than above
them. It is man who discovers significations: signification is the meaning
[sens] that an event takes on with respect to already existing forms;
signification is what makes an event have value as information.
(Simondon, 2017, p. 150)

Maintenance can provide signification to the role of technical objects

within an ensemble. Understanding technicity, that is, the spatio-temporal
conditions of human beings, technical objects and technical ensembles in
the world, creates information and meaning (sens).10 Maintaining any tool,
for instance, refers back to its past and relates it to future use. It is an
192 Johannes F.M. Schick

attempt – as rudimentary as this might be – to understand the technicity

and history of a particular object and to continue it.
The production of modern technical objects requires division of labour,
standardisation and, to a certain degree, automation.11 The changed condi-
tions of production also bring a new way of relation to the produced
objects – and thus also the changed form of maintenance. While the artisan
holds the entire production process – so to speak – in his hands and both
works on and understands the individual elements of the objects produced
as well as ensuring that the object fits into the technical ensemble, in the
factory these two aspects diverge. The manager and the worker of a fac-
tory each address only one level of the technical – either the ‘ensemble’,
that is, the level of the technical object is integrated into the larger fabric
of society and enters a network outside of the factory as a new whole, or
on the level of the element, where workers assemble the technical object
without having the perspective of the owner (Simondon, 2017, p. 80).
Both would tell very different stories about their relation to the technical
object, since each of them has only a limited perspective which does not
grant an understanding of the entire technicity of the technical object, but
rather mere aspects of it. For Simondon, the ideal would be the engineer,
who has simultaneously an understanding of the element and the ensem-
ble.12 Simondon was however well aware that already in his time, engi-
neers had been reduced to a mere service function keeping the status quo
of the functioning alive (Simondon, 2005a).
According to Simondon, concretisation also leads to different forms of
maintenance: the maintenance of a tool differs from a concretised technical
object. While honing a blade, for instance, acts directly on the material of
the tool and requires body techniques that are in direct relation to the tool,
maintaining a concretised technical object, a smartphone, for example,
might only involve an update in order to keep the self-maintenance (auto-
entretien) of the phone intact. This even holds when the phone is opened
and the battery is replaced. However, all three activities, the honing of the
blade, the software update and the replacement of a battery can all be
points of departure for a deeper engagement with technical objects. They
all share the temporal aspect of maintenance, that is, they provide a link
between the past of the object, the present and the future. Information is
inscribed in technical objects that can always be actualised and signifies a
potentiality that remains present through time:

The technical being realizes the summation of an availability that re-

mains always present; the effort spread out in time, instead of dissipat-
ing, discursively constructs a coherent being that expresses the action or
the series of actions that have constituted it, and keeps them always pres-
ent: The technical being mediates human effort […] The technical being
 Maintaining Perpetual Actuality in the Digital Age? 193

is participatory; as its nature lies not only in its actuality, but also in the
information it fixes and constitutes, it can be reproduced without losing
this information; […] it is open to any human gesture in order to use or
recreate it, and is inserted in a universal elan of communication.13

It is precisely this ‘universal elan of communication’ that allows us to call

the relation with technical objects ‘hermeneutic’.14 I understand hermeneu-
tic here not in a methodological or historical sense, but merely as the pos-
sibility to discover and create and express signification with technical
objects and ensembles. It can, but need not, use language in order to do so.
This provides the basis on which to develop a conception of maintenance
as ‘(hi)story-telling’. Human beings are in multiple relations with technical
objects and must address the technicity of technical objects when main-
taining them. That is, acts of maintenance require an understanding of the
history of an object – what elements require maintenance, what are the
conditions of auto-entretien and when are they met – as a prerequisite to
keeping the objects up to date. The temporal aspect of maintenance allows
for the revival and the constant actualisation of technical objects. In this
process, technical objects do not remain the same but are transformed due
to their temporality and also due to the effort of spatial reticulation, where
the temporal aspect is put to the test, that is, the virtual characteristics of
the technical object are actualised in a network.
This implies that the history of the technical object receives new signifi-
cation by means of the active relation with the network.15 Maintenance is
thus the active continuation of the life of a technical object. Maintenance
holds, as the German translator of Simondon’s ‘On Technical Mentality’,
Michael Cuntz puts it, technical objects in the present (Simondon, 2011).
One could thus say that maintaining a technical object always begs the
question: ‘What kind of future does this object have?’ Consequently, how
an object is maintained cannot merely refer to the object itself, but must
include a conceived virtual future. Actualising technicity in the present net-
work may thus require adaptation or invention, making it necessary to:

[…] know the language through which the human gesture that created
it [the technical object] re-actualizes itself. […] The technical being must
be considered as an open, polarized being that seeks its complement,
which is the working human, in the coincidence of the newly assembled
whole. The user must take the place of the constructor. For this it is
necessary that s/he falls at one with the essential schematism inscribed
in the technical being, that s/he is able to think it, to understand it, to
love it as if s/he had made it him/herself. The duality of human-nature is
resolved in the functional unity of the working human. 16
(Simondon 2014h, 252–253)
194 Johannes F.M. Schick

Even if objects are built to sell in the modern capitalist mode of produc-
tion, technical objects still contain technicity that can be accessed.17 Yet, an
ideal technical object would be in constant synchrony with the presence,
which would facilitate the user’s participation in the genesis of the techni-
cal object:

On the contrary, if the object is open, i.e. if the user’s gesture, on the one
hand, can be an intelligent, well-adapted gesture, knowing the internal
structures, and if, on the other hand, the repairer - who, moreover, can
be the user - if the repairer can perpetually maintain the parts that wear
out new, then there is no date, there is no aging. On a base which is a
base of durability or at least of great solidity, one can install parts which
will have to be replaced, but which, in any case, leave the fundamental
scheme intact and which even allow it to be improved; because one can
well think that at some point, if a better cutting tool is found for a ma-
chine intended for work involving cutting, this tool can be fitted, pro-
vided that it has the necessary standards, on the base, and thus the
machine will progress with the development of techniques. This is what
I call the open object.18

Even if the open object remains an ideal that might never or only seldom
be produced due to economic interests, it reveals the central aspect of
maintenance necessary for a harmonious relationship with technical ob-
jects and ensembles. On the level of use, the user’s gestures already ‘get to
know’ the internal structure of the object. Openness also refers to the ex-
ternal parts of the object, which can be improved and replaced if more ef-
ficient parts become available. The actualisation of an open object thus
moves from the internal ‘fundamental scheme’ of functioning to the parts
that can be replaced and connect the object with the contemporary en-
sembles of technical production (e.g., faster microchips, batteries, etc.).
The fundamental scheme and the contemporary networks of production
are thus in a recursive relationship with each other mediated by acts of
repair and maintenance that keep the object up to date. This process cre-
ates signification since the object is positioned in the network and can enter
relationships with other technical objects and ensembles.
In the foregoing, we observed that maintenance actualises the virtual
technicity of a technical object. Technicity refers to the spatio-temporal
integration in the technical ensemble and the practices of use. Maintaining
technicity refers simultaneously to both the internal structure of an object
as well as to its relation to other objects when functioning. Acts of mainte-
nance create signification by relating the past of a technical object to its
future operations, thus actualising its technicity. In the following, I will
explore the notions of ‘universal communicative elan’, ‘signification’ and
 Maintaining Perpetual Actuality in the Digital Age? 195

‘understanding of technical objects’ by introducing the analogy of ‘story-

telling’. The communicative plan implied in the production, use and main-
tenance of technical objects allows users to participate in the process of
becoming of the technical object. I will in Section 8.2 develop (hi)story-
telling as a mode of participation with and maintenance of technical
objects.

8.2 Simondon’s Philosophy of Maintenance: Maintenance,

Education and (Hi)story-Telling
I interpret Simondon’s claim of needing to ‘know the language’ of the hu-
man gestures used in the creation and use of the object as well as the aim
of the ‘user becoming the constructor’ in terms of story-telling. Even
though Simondon himself only explicitly mentions the ‘lack of technical
poets’ in an interview with the Canadian mechanologist Jean Le Moyne
(Simondon, 2014b, p. 422), this does imply a reference to maintenance,
since the poetic effort restores an original unity – the technicity – of out-
dated objects such as steam locomotives.19 This ‘poetry of techniques’ rec-
reates the history of an object and links it to the present and a possible
future. It is a specific form of poiesis that gives meaning to an object and
can be understood as a form of story-telling.
Like a great story, a technical object can remain perpetually actual and
in contact with the presence; a potentially timeless connection between the
story’s ‘audience’, in this case, the technical object’s users, the storyteller,
the repairer and the story itself – the technical object, may be established
and maintained. Even though the central message remains the same, the
expectations of the audience, their social status, their listening skills and
their knowledge of stories are all different and even the language they use
is transformed over time. Some parts of the story may no longer fit and
have been replaced and other parts may be completely omitted. Yet, in
order to do so, the storyteller has not only to know the story itself but also
its audience and the network in which the story is embedded. Both the
story and its network are furthermore integrated into a broader network.
This feature of the network does not only apply to external integration but
must also be an internal aspect of the ideal open object. As stated above, it
applies to both the story and the object.
In order to be actualised and maintained in a state of actuality, an object
needs to be both: stable and transformable. The ‘open object’ consists thus
of ‘parts designed to be as close as possible to indestructibility’ and of ‘oth-
ers … in which the finesse of adaptation to each use, or wear and tear, or
possible breakage in the event of a shock or malfunction, is concentrated’.20
This constitution of the technical object signifies that it can ‘allow the de-
velopment of the technical mentality and […] be chosen by it’ (ibid).
196 Johannes F.M. Schick

To actualise and maintain a technical object it is necessary to grasp

its technicity. To perform this task in its entirety, a form of abstract
knowledge – that refers to the internal operations and understands the
schemata of functioning – as well as the requisite skillsets to perform the
tasks of maintenance are necessary. These two forms of knowledge go
back to the distinction Simondon makes in On the Mode of Existence of
Technical Objects between craftsmanship and engineering. Each refers to
a specific mode of transferring knowledge. Craftsmanship is linked to the
technical training of a ‘master’, while the training of an engineer is based
on a transfer of knowledge via technical schemata that are laid out in an
Encyclopedia (Simondon’s example is Diderot and D’Alembert’s
Encyclopedia). On the one hand, craftsmen possess an almost mystical
knowledge and intuition of the matter they are dealing with and act al-
most like magicians, whose knowledge remains a secret ability that can-
not really be shared with others (Simondon, 2017, p. 106). On the other
hand, a technical object in its proper sense requires a sort of abstract,
generalised and rational knowledge, which can be represented in sche-
mata and relies upon the formula and scientific knowledge, which belongs
to the ‘encyclopaedic order of knowledge’ (Simondon, 2017, p. 123).
Even though Simondon describes these forms of knowledge in their ex-
treme and claims that the teaching of the form of craftsmanship is aimed
at the child, while the encyclopedism is aimed at the adult, the distinction
between the skill taught by means of a body technique (Mauss, 2006) and
a knowledge transmitted in an abstract form is still recognised today.
However, even though craftsmanship and the encyclopaedic spirit may
seem inverse to each other, both rely upon a quasi ‘magical form of initia-
tion’ (Simondon, 2017, p. 112).
This ritualistic initiation allows us to develop the analogy of (hi)story-
telling, as introduced earlier. According to each form of knowledge, a dif-
ferent story may be told: The craftsman’s story is one of a magical initiation,
an intimate relationship with matter and techniques that is passed down by
a master to the apprentice; Encyclopaedic knowledge, however, may be
seen as a story with a magical book at its core, which provides the key to
a general knowledge bringing a variety of topics, matters and technical
objects in relation to each other (Simondon, 2017, p. 112).21
The encyclopaedia generalises the initiation into technical knowledge –
and as such it becomes a danger to the institutionalised craftsmanship in
guilds, since if the encyclopaedia is clear and well understood, the initia-
tion performed by the master craftsman becomes redundant (Simondon,
2017, p. 111). The difference between encyclopaedism and craftsmanship
shows itself in the ways in which information and energy are distributed.
In craftsmanship, energy and information are bound to the body of the
craftsman. S/he is thus in direct contact with matter by means of the
 Maintaining Perpetual Actuality in the Digital Age? 197

technical operation (Simondon, 2014f, p. 303). A machine, on the con-

trary, is not dependent on the human being in the same way. Energy can
be provided by natural resources (oil, coal and wind) or by a technical
ensemble, that is, an infrastructure. Once a machine is switched on, it per-
forms its operations by itself, dependent on the human being only for in-
formation (Simondon, 2009, p. 20). The various scientific principles – the
encyclopaedic schemata – are crystallised in the machine and thus inform
the technical object. This process of informing cannot be reduced to the
utilisation of the object. It is concerned with understanding the operation,
providing it with direction and signification, which consists of maintaining
its ‘associated milieu’, that is, its conditions of auto-entretien and its inner
workings, and may also mean to transfer it into different milieus of use.
Comprehensive technological knowledge, however, requires both skills
on the concrete performative level and an understanding of the schemata
that allow for technical objects.
Technical objects crystallise a complex practice that emerges in its op-
eration. The operation relates different structures to each other and leads
to ‘circular knowledge’, that is, to a practical theory in its literal sense,
which informs the theory of the practical synergy of heterogenous
elements.22 This line of argument shows not only the complexity of
Simondon’s philosophy but also a general point on the relation to techni-
cal objects in contemporary times. We are faced with objects that are ex-
tremely complex, which cannot be fully understood and maintained in
their technicity by ourselves as users (at least not without some very spe-
cialised knowledge, which is simultaneously practical and theoretical, and
not without access to very specific infrastructure, software and hardware
components).
To elucidate a possible actualisation of maintenance, I would like to re-
turn to the idea of (hi)story-telling. Maintenance is conceived of as an ideal
relation to technical objects, the technical ensembles and the infrastructure
they rely on. Though it is difficult – near to impossible – to teach oneself
the skills needed to maintain a modern digital object, it is certainly possible
to enter into hermeneutic relations with technical objects. The encyclopae-
dic spirit and the spirit of craftsmanship serve as ideals for a technical edu-
cation. For Simondon, the seeds of a technical culture, its corresponding
education and the images required to create stories are planted in child-
hood. In childhood, an effort to understand technical operations takes the
form of mimicry. This mimicry provides the images needed for story-
telling: when children ‘become’ the airplane or the automobile and imitate
the technical object at work, they acquire a glimpse of understanding of the
technical object at an operational level. Obviously, this is not the encyclo-
paedic understanding of an engineer, but it provides a foundation for the
creation of mediating images and bodily schema that can aid technological
198 Johannes F.M. Schick

understanding in later life. Children grasp a technical scheme through play,

which can later be conceptualised and objectified (Simondon, 2014g,
p. 44). Initially, technical inventions are ‘schemata of behaviour, of opera-
tion’ (ibid.). Operations and images that served childhood play can be re-
discovered and reconstructed to understand technical objects. These images
and bodily schemata can be used to create stories that relate the human
being and the technical object with each other at an operational level.
This relation – according to Simondon – exists not only in children but
also on a higher level of understanding in other social groups that have an
intimate relation with technical objects (Simondon, 2014g, p. 48ff.).
Simondon uses the examples of the farmer, the housewife and the mariner
to show that social images and prejudices are connected to social groups,
the technical objects they utilise and the operations performed by the
groups with these technical objects (ibid.). Such social images are used to
implement stories about individuals, their social status in these groups and
their use of technical objects. These stories may perpetuate stereotypes,
which have to be countered by different accounts of how to relate and act
in technical ensembles. Simondon shows that the woman in the kitchen,
for instance, operates as a coordinator in the midst of technical objects
(Simondon, 2014g, p. 47). She has skills and technical knowledge that go
beyond the social image perpetuated by advertisements in the 60s. A con-
temporary example might be the hacker, who – even if s/he does not ‘live’
in the data – is by means of her/his programming skills a coordinator, em-
bedded in the technical objects and ensembles. S/he can hack into systems
and manipulate technical ensembles and infrastructures.
Going from these social images to the relation of the respective social
group with the technical objects, a different picture can be drawn: the toys
of children, the kitchen appliances of housewives and househusbands, the
farmer’s tractor and the mariner’s vessel all reveal a participatory, non-
dualistic relation with technical objects. The child becomes the airplane in
play, the cook operates in the kitchen amidst the household appliances and
orchestrates their interaction and the tractor is plurifunctionally used by
the farmer to perform a wide variety of tasks (Simondon, 2014g, p. 49f.).
The dualisms between body and mind, body and imagination, form and
matter and human and machine are suspended in the act of playing, cook-
ing, working with the tractor or navigating the vessel. The socio-cultural
meaning of the technical object in these relations is still present because the
child, farmer and cook correspond to this image precisely through their
activity, but the relation itself consists of operations that produce norms
that can in turn become part of a unified, technical culture.
This technical culture, where technical objects are treated as cultural
objects, on par with works of art, etc., becomes possible, if one establishes
a ‘participatory observation’ of technical objects beginning in childhood;
 Maintaining Perpetual Actuality in the Digital Age? 199

operations and images that one has played out in childhood can be redis-
covered and reconstructed in order to create meanings which free the tech-
nical object from the norms imposed on it from the outside. The child
observes the object and attempts to participate in its functioning (Simondon,
2014g, p. 43f.). The observed and played out schema gives access to the
history of the technical object, to maintain it in the present and to actualise
it in the later stages of life (Simondon, 2014g, p. 44). However, this does
not mean that one can and should return to the supposedly innocent or
natural state of the child. After all, understanding technical objects does
not only consist of physically imitating operational schemata and em-
pathising with them. Especially, when one has digital technologies in mind,
this notion is downright absurd. While bodily techniques remain the basis
of all techniques – including digital ones – since there is no strict, dualistic
separation of the mental and the physical, and all bodily techniques are
also imbued with mental operations, the understanding of technology
points beyond itself and implies an understanding of matter that evokes
theory and science.
The schemata of childhood are far from being sufficient to develop a
‘technical culture’. They merely reveal a very basic notion of a possible
relationship with technical objects that must then be fostered by proper
technical education. Simondon developed such an idea of technical educa-
tion in his work as a teacher. In two articles based on his own experiences,
he explains how this education would be both adequate for the respective
age of children and to enable new relationships with technical objects
(Simondon, 2014d, 2014h). It further exemplifies how maintenance is un-
derstood not only as maintaining technical objects but also as maintaining
(and regaining) a relationship to and with the natural world. Technical
objects serve as media since the technical ensemble is the ‘most stable and
universal mix of the natural world and the human world’.23 The goal of
technical education is, according to Simondon, to create an intuitive un-
derstanding of the technical being:

Our aim was the intuitive grasp of the technical being by the young in-
telligence. A child does not understand, in the deepest sense of the word,
what a tree or an animal is. S/he can, however, understand, in the tech-
nical sense of the word, why a newly planted tree must be watered, why
a tree needs light; this is because the child achieves an intuitive grasp of
the organisation of the tree; s/he does not understand scientifically what
assimilation and photosynthesis are, but s/he can understand what
grafting or layering is. It is this kind of understanding, intuitive but not
affective or animistic, which we call technical understanding. Between
the primitive mentality and the scientific mentality, there is technical
thinking.24
200 Johannes F.M. Schick

Here we can see precisely defined both the technical ensemble as a stable
entity and technical thinking, which allows the development of mainte-
nance skills through a specific kind of story-telling. Technical ensembles
unite the natural and the human world, they are key to understanding hu-
man operations in the arena between the ‘primitive’ and the ‘scientific
mentality’. Maintenance in Simondonian terms refers to the temporal rela-
tionship with the concrete technical object, the natural world and the so-
cial world, co-produced with technical objects. Technical knowledge is
therefore embedded within culture (Simondon, 2014d, p. 217). This is ap-
parent from the position Simondon assigns to technical thinking in con-
trast to the scientific and primitive mentality.25 Technical thinking may
begin in childhood since it is rooted in socio-technical operations.
It is thus unsurprising that, as a teacher, Simondon considered it his task
to contribute to the creation of a technical culture. In the classroom, he
addressed the threefold dimension of nature, techniques as medium and
culture and introduced his students to technical objects. The objects that
were studied did not merely serve the function of being opened, repaired
and tinkered with, but also to create a new, symmetrical social order in the
classroom: the students worked in different groups and they had to work
together in order to appreciate and understand the technical objects.
Working in small groups provided the students with a sense of autonomy,
initiative, invention and a sense of effort and solidarity (Simondon, 2014d,
p. 214). Furthermore, Simondon did not understand himself as ‘contremaî-
tre’, but as a teacher, who monitors the progress of his students (ibid.).
The analogy of story-telling exemplifies how Simondon introduced the
technical ensemble to his students. As stated above, Simondon played an
analogous role to a storyteller: the goal was not explaining the scientific
nature of the technical objects nor merely imitating the master’s move-
ments but rather to gain an understanding of the gist of the story; for the
students should find a first intuition beyond an imitation of a technical
object, but already pointing to its technicity and operations. Drawing the
analogy of the storyteller further we could say that Simondon’s story to his
students addressed how to become a ‘technician’, which means to ‘prefer
the knowledge of a human operation to the knowledge of an objective
structure’, that is, not to see the technical object as something that requires
explanation as a structure, but as something that operates and is the fruit
of human hands and invention.26
Technical activity is always related to human activity that mediates be-
tween nature, society and the individual. It thus goes beyond ‘mere work’
and must be distinguished ‘from alienating work’ (Simondon, 2017,
p. 255). It cannot be reduced to the use of the machine (ibid.), but rather
consists of ‘maintenance, adjustment, and improvement of the machine,
which continues the activity of invention and construction’ (ibid.). This in
turn shows that maintenance is a human operation that does not merely
 Maintaining Perpetual Actuality in the Digital Age? 201

operate on the technical object as an abstract and objective structure but

relates to it as an entity at an intimate, spatio-temporal level.
This intimacy, this love for technical objects, is what a technician should
aspire to. It is this message that Simondon conveyed to his students
(Simondon, 2014h, p. 252). This relationship with the technical ensemble
is exemplified in the telling image of complementary beings that fulfil a
task together:

The machine demands services from us and renders them to us, like a
friend; the exchange of services, preferable to slavery, is not yet even the
highest and most adequate relationship to the machine. It is necessary
to ‘pull the yoke’ with her, to know her well, to work taking her neither
as an end nor as a means, but as a fellow worker and as a complemen-
tary being […]. This horizontal relationship must replace any vertical
relationship.27

The story required for developing technical knowledge – which is in turn

necessary to maintain technical objects and their relationship with the
natural world – displays, at its core, a horizontal relation between natural,
technical and human entities. The aim of a technical culture and conse-
quently also of maintenance is to know and appreciate technical ensembles
as such. They should not even enter into the logic of means and ends, but
rather be ends in themselves that require care, since their well-being is
necessary to fulfil a task together. Human beings and machines should
work as an operative couple, each complementing the function of the
other, instead of reducing their relationship to economic or energetic val-
ues (Simondon, 2017, p. 135).
The openness and opening of objects corresponds to the openness of the
learning practice: the apprentice of a craftsman imitates operations and the
object is presented merely as an occasion to fulfil standardised tasks
(Simondon, 2014d, p. 214). Instead of a closed object, Simondon chose
technical objects that were open to understanding (ibid.). The first experi-
ence with the object reveals a strangeness and complexity to the students
(ibid.). This experience can be conceived of as an experience of a technical
other that has an irreducible quality and is the point of departure for the
horizontal, harmonious relationship with technical objects (Schick, 2021c).
Maintaining this relationship means, as already mentioned above, becom-
ing the constructor of technical objects instead of merely using them
(Simondon, 2014h, p. 252). The image of this relationship goes even fur-
ther. It does not suffice just to understand the technical being, but rather it
is necessary to:

coincide with the essential schematism inscribed in the technical being,

s/he must be able to think it, understand it, love it as if s/he had made it.28
202 Johannes F.M. Schick

This coincidence with and love for technical objects does not mean that
machines should be idolised, but it should prevent contempt and fear of
technology and furnish the students with a ‘healthy knowledge based on a
careful attentiveness’ for technical objects (Simondon, 2014h, p. 253).
This becomes ever more relevant since the attitude towards an ever more
rapidly increasing number of new technical developments oscillates be-
tween fear and idolatry. Especially, the robot or any automat – exemplified
in today’s digital realm by attitudes to artificial intelligence – is furnished
by the imagination of its users with powers and abilities that have little to
do with the actual technical object (Simondon, 2017, p. 16). To work
against this common fear, Simondon introduced an automat to his stu-
dents. It becomes quite clear that the basic notion of the automat still holds
for contemporary computers and smart tools. Simondon defines any au-
tomat as follows:

Every automaton, whether electronic or not, consists of four parts: a

system of receptors, similar to sense organs; one or more amplifier or
selector relays (the ‘brain’); one or more effectors, similar to muscles;
and finally, a source of energy (power supply).29

The main four ‘organs’ (sensory organs, brain, alimentary system and mus-
culoskeletal system) correspond to the sensors recording movement and
input of information (sensory organs), the selection of information on the
motherboard (brain), the transformation of signals into movement and
information as visual or audible output (muscles, movement) and the bat-
tery or power cable which powers the device.
To conceive the automaton in analogy to a living being with different
‘organs’ is a first step towards understanding the (hi)story of the technical
object. Technical (hi)story-telling is, however, not merely about telling sto-
ries about technical objects, but rather about engaging with technical ob-
jects and seeing how signification can be created with technical objects.
Maintenance can thus be conceived of as generating a (hi)story with the
technical object, that is, to take the social relationship with the technical
object seriously; while technical objects have a specific and genuine mode
of existence, they are the result of human invention and a store of informa-
tion and as such can always be actualised by them.
The (hi)story-telling relates human operations with the operations within
technical objects. The operations that take place in the machine can thus be
understood as operations that human beings also perform in a similar, but
different way. Even if the machine itself is not opened, a relation to the
machine becomes possible that understands it as something beyond its
 Maintaining Perpetual Actuality in the Digital Age? 203

utility. However, this should not lead to the anthropomorphisation of the

machine; its otherness must at all costs be preserved in order to realise the
desired horizontal, harmonious relationship: since the love and friendship
for technology takes the ontogenesis, that is, the genesis and evolution of
the technical object as an individual into account. However, most technical
objects are not constructed to be maintained by the user, which leads to
alienation from the object’s technicity (Simondon, 2017, p. 255f.).
We are faced with the question of how a concept of maintenance can be
actualised in a world where almost any technical object is built merely to
be sold and to be maintained, or more likely scrapped and replaced, by the
constructor. The concept of maintenance, as I have introduced it in this
chapter, can serve as a critical concept to tell different or new stories with
technical objects. Maintenance as story-telling entails entering into a dia-
logue with technical objects and searching for their responsiveness. It im-
plies an understanding of the history of the technical objects and the
development of a perspective for their future use. A starting point is the
re-evaluation of the relations with objects; understanding them as some-
thing containing human gestures and simultaneously irreducible to a spe-
cific human mode of operation, as they have their own mode of existence.
Technical objects that correspond to the ideal of maintenance may yet be-
come possible:

they are conceived so that the different organs constituting them can be
continually replaced and repaired over the course of use: maintenance is
not separate from construction, it continues it, and in certain cases,
completes it, for instance by means of breaking it in [rodage], which is
the prolongation and completion of construction through rectification
of the surface conditions during their functioning.
(Simondon, 2017, p. 255f.)

The concept of maintenance itself is part of a story worth telling. A story

that goes from concrete operations with technical objects in craftsmanship
to the maintenance of the conditions of the auto-entretien of complex tech-
nical objects in the age of information. In this age, human gestures still
organise the relationship with technical objects, but the ‘crystallized hu-
man gestures’ of the invention within the technical objects must be redis-
covered. It points to an ethical relation with and in nature that is mediated
by the technical ensemble: it also tells the story of a reciprocal relationship
where maintenance of a technical object also entails the human being do-
ing the maintenance keeps a relation to themselves as well as to the natural
and social world.
204 Johannes F.M. Schick

8.3 Maintenance in the Networked Era?

How can maintenance allow for an intimate, participatory relationship
with technical objects in the digital age? First, we must acknowledge that
the task is too much to handle for an individual to tackle alone. Technical
objects are constructed and sold as closed objects, users in the west are
deskilled and the maintenance infrastructure has been outsourced to the
global south (Graham and Thrift, 2007, p. 18f.). The genesis of a technical
culture and an ideal of maintenance that keeps technical objects in the
present and maintains a harmonious relationship with technology and na-
ture presupposes a transformation of the attitude towards technical ob-
jects. It was the goal of this chapter to develop some aspects of how
maintenance is integral to this transformation and how this different atti-
tude can be developed. The different aspects of maintenance as (1) holding
a technical object in the present, that is, to keep it up to date, (2) under-
standing its history, (3) being able to actualise its history in new and differ-
ent situations as well as (4) maintaining a relationship of human beings
with technical objects which is open and sees them as social beings, is
crystallised in the notion of story-telling.
As Elizabeth Spelman points out, repair and maintenance is what keeps
our social order together (2002, p. 1ff.) and is essential to human beings at
all levels of their existence: the homo reparans repairs relationships, infra-
structures and technical objects and restores natural habitats (ibid.). To
understand these relations implies an understanding of not only the signifi-
cance of the relations themselves but also of the mode of existence of the
objects and entities we have relations with. Maintenance as (hi)story-telling
provides a point of departure to relate to technical objects and technical
ensembles as social beings. It probes into the object’s past and relates it to
the future. The temporal aspect of maintenance is intrinsically linked to the
use of the object since it also implies the future perspective of what kind of
relations with the object will be possible. A smartphone relates the user via
the interface to its material structure, its internal, associated milieu that
allows its functioning. The user maintains this internal functioning via up-
dates which relate her to the technical ensemble of production. Using the
smartphone, s/he interacts with other human and non-human entities (e.g.,
social platforms, satellites, etc.) spanning from other users close by to sat-
ellites in the orbit of Earth. The maintenance of a smartphone thus refers
to the history of the infrastructure of production and use, the social con-
struction of meaning with the smart tool as well as to the operations with
the interface and the operations of functioning in the phone. The mainte-
nance of a smartphone must thus take into account these aspects to de-
velop a future relationship with the phone. Story-telling raises the questions
of how to participate and cooperate with nature and technical networks
The ‘homo reparans’ (Spelman, 2002) has thus in Simondon’s case to be
 Maintaining Perpetual Actuality in the Digital Age? 205

termed homo coordinans (Schick, 2021a), being in the midst of technical

objects, able to maintain and provide the relations with nature, society and
the technical ensemble with signification.

Notes
1 I would like to thank the anonymous reviewers, whose valuable comments
pushed me to clarify my argument and who pointed me to the important notion
of ‘auto-entretien’. The publication of the chapter was supported by the
German Research Foundation (DFG) – Project Number 262513311 – CRC
1187 Media of Cooperation, University Siegen.
2 Living beings can transform their internal structure – if necessary – while tech-
nical objects have to be maintained. Living beings are from the very beginning
of their existence endowed with the aptitude to reorganise themselves.
Artificiality, however, does not merely refer to the ‘fabricated origin of objects’
but to the engagement of human beings with artificial and natural objects that
have been ‘domesticated’:
As it evolves, this object loses its artificial character: the essential artificiality
of an object resides in the fact that man must intervene to maintain the exis-
tence of this object by protecting it against the natural world, giving it a
status of existence that stands apart. Artificiality is not a characteristic denot-
ing the fabricated origin of the object in opposition to spontaneous produc-
tion in nature: artificiality is that which is internal to man’s artificializing
action, whether this action intervenes on a natural object or on an entirely
fabricated one; a flower, grown in a greenhouse, which yields only petals
(a double flower) without being able to engender fruit, is the flower of an
artificialized plant: man diverted the functions of this plant from their coher-
ent fulfillment, to such an extent that it can no longer reproduce except
through procedures such as grafting, requiring human intervention.
(Simondon 2017, p. 49)
3 For an application of Simondon’s philosophy to the digital, see, for example,
Hui (2016); Rantala (2019).
4 This is also one of the few passages, where Simondon refers implicitly to his
work as a teacher:
Moreover, this is precisely how the functioning of thermal engines is ex-
plained to students in the classroom, each piece being isolated from the oth-
ers like the lines that represent it on the blackboard in geometric space,
partes extra partes. The old engine is a logical assemblage of elements de-
fined by their complete and unique function. Each element can accomplish its
own function best if it is, like a perfectly completed instrument, oriented
entirely to accomplishing this function.
(Simondon, 2017, p. 27)
5 The process of concretisation can be also applied to the digital phenomena.
The development of AlphaGo to AlphaGo Zero evolves from a design where
two different neural networks are each dedicated to a specific task (a ‘policy
network move that outputs move probabilities’ and ‘a value network that out-
puts a position evaluation’ (Silver et al., 2017, p. 354)) to a design where both
206 Johannes F.M. Schick

tasks are integrated in a single neural network (Silver et al., 2017). An internal
solution to the problem was found and improved the capabilities of the algo-
rithm. Another example in biotechnology is the development of CAR-T cells
(Schick, 2021b).
6 The other phase being ‘religion’. Simondon’s conception of ‘technicity as a
phase’ is itself embedded in a complex ‘phenomenological capture of the pre-
historical process of hominization’ (Bardin, 2015, p. 167). ‘Phase’ and ‘phase-
shift’ are notions that stem from physics and signify the transformations within
a system, where phases refer to each other (ibid., 173). It is not necessary for
my argument to develop these notions further. For a more detailed account, see
the article ‘Technicity’ in Jean-Yves Chateau’s ‘Vocabulaire de Simondon’
(2008, pp. 103–7).
7 For a comprehensive discussion of the distinction between living and non-living
beings, see DiFrisco (2014).
8 ‘What risks making the work of cybernetics partially inefficient as an inter-
scientific study (which nevertheless is the objective Norbert Wiener attributes
to his research) is the initial postulate concerning the identity between living
beings and self-regulating technical objects. Yet the only thing we can say is
that technical objects tend towards concretization, whereas natural objects,
such as living beings, are concrete to begin with. One mustn’t confuse the ten-
dency towards concretization with the status of entirely concrete existence’
(Simondon, 2017, p. 51).
9 All quotes are translated by the author if an original French text is referenced.
De même, la dégradation du technique se produit quand l’objet est isolé
dans le temps (par cette cassure qu’est la fin de fabrication et la chute dans
la condition de vénalité) et dans l’espace (par ce détachement qui isole
l’objet fabriqué des conditions en lesquelles il pourrait recevoir une perpétu-
elle régénération le maintenant au niveau de sa pleine signification fonction-
nelle). La technicité se dégrade en s’objectivant parce que l’objet, en tant que
fermé, s’archaïse et se dégrade, lorsqu’il n’est plus maintenu dans le réseau
de technicité par lequel il a été constitué. La technicité est un mode d’être ne
pouvant exister pleinement et de façon permanente qu’en réseau, aussi bien
de façon temporelle que de façon spatiale. La réticulation temporelle est
faite de reprises de l’objet en lesquelles il est réactualisé, rénové, renouvelé
dans les conditions mêmes de sa fabrication première. La réticulation spa-
tiale consiste en ce fait que la technicité ne peut être contenue en un seul
objet; un objet n’est technique que s’il opère en relation avec d’autres objets,
dans un réseau où il prend la signification d’un point-clef; en lui-même et
comme objet, il ne possède que des caractères virtuels de technicité qui
s’actualisent dans le rapport actif à l’ensemble du système. La technicité est
une caractéristique de l’ensemble fonctionnel qui couvre le monde et dans
lequel l’objet prend une signification, joue un rôle avec d’autres objets.
(Simondon, 2014b, p. 82)
10 The temporal dimension of understanding technical objects even goes further
in invention. According to Simondon, invention requires the imagination of a
problem being solved. This implies that the future condition – where the prob-
lem is already solved – has effects on the present situation – where the problem
is still to be solved:
 Maintaining Perpetual Actuality in the Digital Age? 207

Only a thought that is capable of foresight and creative imagination can ac-
complish such a reverse conditioning in time: the elements that will materi-
ally constitute the technical object and which are separate from each other,
without an associated milieu prior to the constitution of the technical object,
must be organized in relation to each other according to the circular causality
that will exist once the object will have been constituted; thus what is at stake
here is a conditioning of the present by the future, by that which is not yet.
(Simondon, 2017, p. 60)
11 Simondon claims that the process of concretisation made the ‘production line’
(Simondon, 2017, p. 29) necessary, since the elements of the technical objects
have to fit precisely and to correspond with each other. There exists a recursive
relation between the mode of production and the intrinsic values of technical
objects, which results in standardisation:
If technical objects do evolve toward a small number of specific types then
this is by virtue of an internal necessity and not as a consequence of eco-
nomic influences or practical requirements; it is not the production-line that
produces standardization, but rather intrinsic standardization that allows for
the production-line to exist.
(Simondon, 2017, p. 29)
12 ‘It is rather difficult for a worker to know technicity through the aspects and
modalities of his daily work on a machine. It is also difficult for a man who is
the owner of machines and who considers them productive capital to know
their essential technicity. It is the mediator of the relation between machines
alone who can discover this particular form of wisdom. Such a function, how-
ever, does not yet have a social place; it would be that of the production plan-
ning engineer if he wasn’t preoccupied by immediate output, and governed by
a finality external to the operating system [régime] of machines which is that of
productivity’. (Simondon, 2017, p. 160)
13 ‘L’être technique réalise la summation d’une disponibilité qui reste toujours
présente; l’effort étale dans le temps, au lieu de se dissiper, construit discursive-
ment un être coherent qui exprime l’action ou la suite d’actions qui l’ont con-
stitué, et les conserve toujours présentes: l’être technique médiatise l’effort
humaine […] L’être technique est participable; comme sa nature ne réside pas
seulement dans son actualité, mais aussi dans l’information qu’il fixe et qui le
constitue, il peut être reproduit sans perdre cette information; […] il est ouvert
à tout geste humain pour l’utiliser ou le recréer, et s’insère dans un élan de com-
munication universelle’. (Simondon, 2005a, p. 340)
14 Simondon does not use the term ‘hermeneutical’ himself. Language is not at the
centre of his philosophical research and his ontology is quite different than the
hermeneutics since nature is, for instance, ‘not part of the discourse’ (Bardin,
2015, p. 44f. & 105).
15 ‘Par contre, plusieurs élèves ont profondément compris et pratiqué cette prise
de conscience de l’être technique, saisissant dans la machine l’histoire humaine
déposée, et y ressentant la présence du monde’ (Simondon, 2014d, p. 204f).
16 ‘Il faut connaître le langage par lequel se réactualise le geste humain qui l’a
produit. […] L’être technique doit être envisage comme un être ouvert, pio-
larisé, qui appelle son complement qu’est l’homme au travail, dans la coinci-
dence du tout recompose. L’utilisateur doit prendre la place du constructeur.
208 Johannes F.M. Schick

Il faut pour cela qu’il coincide avec le schématisme essential inscrit dans l’être
technique, qu’il soit capable de le penser, de le comprendre, de l’aimer comme
s’il avait fait. La dualité homme-nature se résorbe dans l’unité fonctionnelle
de l’homme au travail.’ (Simondon, 2014h, p. 252–253)
17 Stephen Graham and Nigel Thrift draw a similar conclusion: most contempo-
rary technical objects (e.g., smartphones) are not built to be maintained but
rather imply obsolescence, which goes hand in hand with a certain deskillment
of the user (2007, p. 18ff.) They see possible solutions in ‘fundamentally alter-
ing the nature of tool-being’, which could change the problems of e-waste and
the commodification of the relation towards technology by reintroducing the
importance of maintenance and repair (2007, p. 19).
18 ‘Tout au contraire, si l’objet est ouvert, c’est-à-dire si le geste l’utilisateur, d’une
part, peut être un geste intelligent, bien adapté, connaissant les structures in-
ternes, si d’autre part le réparateur – qui, d’ailleurs, peut être l’utilisateur –, si
le réparateur peut perpétuellement maintenir neuves les pièces qui s’usent, alors
il n’y a pas de date, il n’y a pas de vieillissement. Sur une base qui est une base
de pérennité ou tout au moins de grande solidité, on peut installer des pièces
qui devront être remplacées, mais qui, en tout cas, laissent le schème fondamen-
tal intact et qui même permettent de l’améliorer; car on peut bien penser qu’à
un moment ou à un autre, si on trouve un outil de coupe meilleur pour une
machine destinée à un travail impliquant la coupe, cet outil pourra être monté,
à condition qu’il ait les normes nécessaires, sur la base, et qu’ainsi la machine
progressera avec le développement des techniques. Voilà ce que j’appelle l’objet
ouvert’. (Simondon, 2014c, p. 401f)
19 ‘Par ailleurs, on pourrait trouver aussi, par une plongée dans le temps, le pou-
voir poétique de ce qui était extrêmement parfait et qui, un jour ou l’autre, sera
détruit, est peut-être déjà détruit, par le cours d’une évolution, qui est extrême-
ment et très dramatiquement négatrice, de ce qui a été pourtant, un jour, une
nouveauté: voyez les locomotives à vapeur, voyez les grands navires, que l’on
met de côté parce qu’ils sont désuets. Ce que l’on appelle l’obsolescence, c’est
une réalité économique mais, à côté de l’obsolescence économique, il y a une
espèce de montée poétique qui n’a pas été, je crois, tout à fait suffisamment
mise en valeur. Nous manquons de poètes techniques’. (Simondon, 2014b,
p. 416) An example of such a poet is Jules Verne (ibid. p. 410), whose work
allows for an imagination of the future and inspires the imagination of the ado-
lescent (Simondon, 2014e, p. 191).
20 ‘Mais l’essentiel réside en ceci: pour qu’un objet permette le développement
de la mentalité technique et puisse être choisi par elle, il faut qu’il soit lui-
même de structure réticulaire: si l’on suppose un objet qui, au lieu d’être
fermé, présente des parties conçues comme aussi près que possible de
l’indestructibilité, et d’autres, au contraire, en lesquelles se concentre la fi-
nesse d’adaptation à chaque usage, ou l’usure, ou la rupture possible en cas
de choc, de mauvais fonctionnement, on obtient un objet ouvert, pouvant être
complété, amélioré, maintenu à l’état de perpétuelle actualité’. (Simondon,
2014d, p. 311)
21 An interesting example of such a story is The Diamond Age by Neal Stephenson
(2011), where in a fictitious nanotechnological world an engineer invents an
interactive, nanotechnological book to educate and foster the creativity of a
child. In this world, standardisation and technical development have resulted in
a new class system, where engineers correspond to the nobility of Victorian
 Maintaining Perpetual Actuality in the Digital Age? 209

times. However, creativity and the necessity to evolve and transform lack the
saturated class of engineers. Stephenson’s book shows multiple things: an imag-
ined new socio-technical world, a new nanotechnological relation with the
world and the effects technical objects have on social change.
22 Simondon provides an example from his time as a teacher:

Cet exercice fait acquérir des notions ayant une valeur culturelle parce qu’il
prépare la compréhension des connaissances scientifiques de thermody-
namique (c’est l’allumage qui a permis la combustion interne dans les moteurs
thermiques) sur le principe de Carnot, et surtout parce qu’il fait comprendre
comment la synthèse de recherches au début complètement séparées, de ther-
modynamique d’une part, d’électricité d’autre part, ont donné naissance à un
être technique viable. Le moteur à explosion est fils de la machine à vapeur et
de l’eudiomètre. C’est une chaudière dans un cylindre.
(Simondon, 2014e, p. 213)
23 ‘Il faut donc créer une nouvelle culture, un nouveau schématisme adéquat au
monde humain et au monde naturel. Le mixte le plus stable et le plus universel
du monde naturel et du monde humain, c’est l’ensemble des êtres techniques’
(Simondon, 2014f, p. 241).
24 ‘Notre but était la saisie intuitive de l’être technique par la jeune intelligence.
Un enfant ne comprend pas, au sens profond du mot, ce qu’est un arbre ou un
animal. Il peut pourtant comprendre, au sens technique du mot, pourquoi on
doit arroser un arbre qui vient d’être planté, pourquoi un arbre a besoin de
lumière; c’est que l’enfant réalise une saisie intuitive de l’organisation de
l’arbre; il ne comprend pas scientifiquement l’assimilation et la photosynthèse,
mais il peut comprendre ce qu’est une greffe ou un marcottage. C’est ce genre
de compréhension, intuitive mais non affective ou animiste, que nous nommons
compréhension technique. Entre la mentalité primitive et la mentalité scienti-
fique, il y a la pensée technique’. (Simondon, 2014e, p. 217)
25 To define technical thinking independently, especially of primitive mentality, is a
critique of Levy-Bruhl’s account of primitive mentality and its relation to tech-
niques and scientific thinking. This goes back to the homo faber debate at the
beginning of the twentieth century, where Durkheimians and followers of
Bergson controversially debated the genesis of human intelligence (Sigaut, 2013).
Durkheim, Mauss and other members of the Durkheim School were quite criti-
cal of the notion of primitive mentality since they argued that different sets of
categories can be developed in different cultures without claiming that the men-
tality of the colonial or cultural other is essentially different (see the discussion
between Mauss and Levy-Bruhl, 1923). In contrast to Levy-Bruhl, they argued
that the genesis of the categories is a socio-technical process (Schick, 2022).
26 ‘Est technicien celui qui préfère la connaissance d’une opération humaine à la
connaissance d’une structure objective’ (Simondon, 2014e, p. 229).
27 ‘La machine exige de nous des services et nous en rend, comme un ami;
l’échange des services, préférable à l’esclavage, n’est même pas encore la rela-
tion la plus haute et la plus adéquate à la machine. Il faut « tirer le joug» avec
elle, la bien connaître, travailler en ne la prenant ni comme fin ni comme
moyen, mais comme camarade de travail et comme être complémentaire […].
Cette relation horizontale doit remplacer toute relation verticale’. (Simondon,
2014h, p. 253)
210 Johannes F.M. Schick

28 ‘L’utilisateur doit prendre la place du constructeur. Il faut pour cela qu’il coïn-
cide avec le schématisme essentiel inscrit dans l’être technique, qu’il soit capa-
ble de le penser, de le comprendre, de l’aimer comme s’il l’avait fait’ (Simondon,
2014f, p. 252).
29 ‘Tout automate, électronique ou non, se compose de quatre parties: un sys-
tème de récepteurs, semblables à des organes des sens; un ou plusieurs relais
amplificateurs ou sélecteurs (le «cerveau»); un ou plusieurs effecteurs, sem-
blables à des muscles; enfin, une source d’énergie (alimentation)’. (Simondon,
2014e, p. 221)

Bibliography
Bardin, A., 2015. Epistemology and political philosophy in Gilbert Simondon: in-
dividuation, technics, social systems, Philosophy of engineering and technology.
Springer, Dordrecht Heidelberg New York London.
Chateau, J.-Y., 2008. Le vocabulaire de Simondon. Ellipses, Paris.
DiFrisco, J., 2014. Hylomorphism and the Metabolic Closure Conception of Life.
Acta Biotheor 62, 499–525. https://doi.org/10.1007/s10441-014-9233-9
Gillespie, T., 2014. The Relevance of Algorithms, in: Boczkowski, P., Foot, K.,
Gillespie, T. (Eds.), Media Technologies. MIT Press, Cambridge, MA, pp.
167–194.
Graham, S., Thrift, N., 2007. Out of Order: Understanding Repair and Main­
tenance. Theory, Culture & Society 24, 1–25. https://doi.org/10.1177/02632
76407075954
Gray, J., Gerlitz, C., Bounegru, L., 2018. Data Infrastructure Literacy. Big Data &
Society 5, 1–13.
Hui, Y., 2016. On the Existence of Digital Objects, University of Minnesota Press,
Minneapolis, Minn. London.
Levy-Bruhl, L., 1923. La mentalité primitive. (Scéance du 15 février 1923 à la
Société française de Philosophie, avec la participation de Mauss, Belot, Parodi,
Ashbourne. Weber, Renoir, Piéron, Fauconnet, Gilson). Bulletin de la Société
française de Philosophie 17–48.
Mauss, M., 2006. Techniques of the Body, in: Schlanger, N. (Ed.), Techniques,
Technology and Civilisation. Durkheim Press/Berghahn Books, New York/
Oxford, pp. 77–95.
Rantala, J., 2019. Blockchain as a Medium for Transindividual Collective. Culture,
Theory, and Critique 60(3–4), pp. 1–14.
Schick, J.F.M., 2021a. The Potency of Open Objects: (Re-)Inventing New Modes
of Being Human in the Digital Age with Bergson, Franco “Bifo” Berardi, and
Simondon. Techné: Research in Philosophy and Technology 25 (3), pp. 279–
406. https://doi.org/10.5840/techne2021723144
Schick, J.F.M., 2021b. “Mein fremder Killer in mir”: Menschen, Organismen und
Designer-T Zellen, in: Dzwidza, E., Speer, A. (Eds.), Philosophische Anthropologie
Als Interdisziplinäre Praxis. Max Scheler, Helmuth Plessner Und Nicolai
Hartmann in Köln – Historische Und Systematische Perspektiven. Brill/mentis,
Paderborn, pp. 364–384.
Schick, J.F.M., 2021c. On Technical Alterity. Foundations of Science 27 (2022),
pp. 515–520. https://doi.org/10.1007/s10699-020-09771-0
 Maintaining Perpetual Actuality in the Digital Age? 211

Schick, J.F.M., 2022. “Let us dare a little bit of metaphysics”: Marcel Mauss, Henri
Hubert and Louis Weber on the Categories of Causality and Technology, in:
Schick, J.F.M., Schmidt, M., Zillinger, M. (Eds.), The Social Origins of Thought:
Durkheim, Mauss and the Category Project. Berghahn, New York/Oxford.
Sigaut, F., 2013. Homo faber: documents d’un débat oublié. http://francois-sigaut.
com/index.php/inedits/53-ouvrages-inedits/527-homo-faber-documents
(Accessed 3 January 2024).
Silver, D., Schrittwieser, J., Simonyan, K., Antonoglou, I., Huang, A., Guez, A.,
Hubert, T., Baker, L., Lai, M., Bolton, A., Chen, Y., Lillicrap, T., Hui, F., Sifre,
L., van den Driessche, G., Graepel, T., Hassabis, D., 2017. Mastering the game
of Go without human knowledge. Nature 550, 354–359. https://doi.org/10.
1038/nature24270
Simondon, G., 2005a. Notes complémentaires sur les conséquences de la notion de
l’individuation, in: L’individuation à la lumière des notions de forme et
d’information. Millon, Grenoble, pp. 331–355.
Simondon, G., 2005b. L’individuation à la lumière des notions de forme et
d’information, Krisis. Millon, Grenoble.
Simondon, G., 2009. Technical Mentality. Parrhesia: A Journal of Critical
Philosophy 7, pp. 17–27.
Simondon, G., 2011. Die technische Einstellung, in: Hörl, E. (Ed.), M. Cuntz
(Tran.), Die Technologische Bedingung. Suhrkamp, Berlin, pp. 73–92.
Simondon, G., 2014a. Objet technique et conscience moderne (inédit), in: Sur La
Technique (1953–1983). Presses universitaires de France, Paris, pp. 363–365.
Simondon, G., 2014b. Entretien sur la mécanologie: Gilbertz Simondon et Jean Le
Moyne, in: Sur La Technique (1953–1983). Presses universitaires de France,
Paris, pp. 405–445.
Simondon, G., 2014c. Culture et technique (1965), in: Sur La Technique (1953–
1983). Presses universitaires de France, Paris, pp. 315–329.
Simondon, G., 2014d. Place d’une initiation technique dans une formation hu-
maine complète, in: Sur La Technique (1953–1983). Presses universitaires de
France, Paris, pp. 203–232.
Simondon, G., 2014e. Art et nature. La maîtrise technique de la nature (1980), in:
Sur La Technique (1953–1983). Presses universitaires de France, Paris, pp.
179–200.
Simondon, G., 2014f. La mentalité technique, in: Sur La Technique (1953–1983).
Presses universitaires de France, Paris, pp. 295–313.
Simondon, G., 2014g. Psychosociologie de la technicité, in: Sur La Technique
(1953–1983). Presses universitaires de France, Paris, pp. 27–129.
Simondon, G., 2014h. Prolègomenes à une refonte de l’enseignement, in: Sur La
Technique (1953–1983). Presses universitaires de France, Paris, pp. 233–254.
Simondon, G., 2017. On the mode of existence of technical objects. Univocal Pub,
Minneapolis, MN.
Spelman, E.V., 2002. Repair: the impulse to restore in a fragile world. Beacon
Press, Boston.
Stephenson, N., 2011. The diamond age, Reissued ed. Penguin, London.
Young, M.T., 2022. Maintenance, in: Michelfelder, D.P., Doorn, N. (Eds.), The
Routledge Handbook of the Philosophy of Engineering. Routledge, London,
pp. 356–368.
Part II

Ethics, Politics and

Aesthetics of Maintenance
9 Maintenance of Value and the
Value of Maintenance
Steffen Steinert

9.1 Introduction
While it’s common to view maintenance practices as a means of ensuring
proper functionality and usability of technology, this overlooks that tech-
nology is inherently value-laden. Technology can either facilitate or hinder
the realisation of values. For example, roads and dams embody the value
of safety, while artificial intelligence systems can promote transparency
and fairness. Moreover, technology plays a significant role in shaping soci-
ety and contributing to social development. Social media, for instance, can
affect people’s well-being and influence how individuals perceive and inter-
act with each other. Maintaining our technology maintains not only tech-
nology’s functional aspects but also its values.
In this chapter, we will expand upon existing discussions of value and
technology by highlighting the importance of maintenance. While current
discussions of value and technology typically centre on design and use, the
role of maintenance is often overlooked. We can gain new insights and
enrich our theoretical toolkit by bringing maintenance into the discussions
of the relationship between technology and value. Looking at technology
and value through the maintenance lens illuminates two crucial points of-
ten disregarded in these discussions. First, maintenance helps stabilise val-
ues by ensuring that technology continues functioning according to its
intended purposes. Second, the maintenance lens reveals that technology
has a life after the design stage and emphasises the importance of consider-
ing technology through time, including how embedded values may be
transformed through maintenance. By focusing on maintenance, we can
better understand the dynamic interplay between change and stability in
technology and values.
This chapter provides an initial exploration of the intersection between
maintenance and value and argues that the maintenance and value perspec-
tives can greatly benefit from each other. While the value perspective sheds
light on the role that values should play in maintenance considerations,

DOI: 10.4324/9781003316213-11
This chapter has been made available under a CC-BY-NC-ND license.
216 Steffen Steinert

it is often underdeveloped in discussions of maintenance. Maintenance can

help preserve social values by stabilising the values embedded in technol-
ogy. Conversely, the maintenance perspective encourages us to consider
technology over time and how the embedded values are stabilised and
evolve through maintenance. This long-term perspective is essential to dis-
cussions of value and technology, as it challenges the common assumption
that values are fixed in the design stage based on designers’ intentions and
that values remain stable. We can gain a comprehensive approach to tech-
nology maintenance and value by combining the maintenance and value
perspectives.
The chapter is structured as follows: we will begin by characterising the
maintenance perspective, and this characterisation will provide the foun-
dation for our subsequent discussions of technology and value. Specifically,
we will highlight three crucial aspects of the maintenance perspective rel-
evant to these discussions. First, the maintenance perspective enables us to
move beyond the design paradigm that characterises many accounts of
value and technology. Second, the maintenance perspective shifts the focus
from value disruption to considerations of value stabilisation, recognising
that the maintenance of values is an ongoing process. Finally, adopting
a maintenance perspective helps to reveal how values are transformed
through maintenance, emphasising the dynamic and evolving nature of the
relationship between technology and value.
With this groundwork established, we will turn to the examination of
socio-technical systems through the maintenance lens. The maintenance
lens reveals how the components of socio-technical systems and their em-
bedded values are maintained and possibly transformed over time.
Finally, in the last part of this chapter, we will explore the relationship
between the maintenance of technology and the maintenance of value.
Here, we will consider how a maintenance account of technology and value
could look that takes the entire life cycle of technology into account.

9.2 Maintenance: Beyond Design, Hardware and Conservation

Viewing technology from the maintenance perspective requires acknowl-
edging that technology is not static but evolves over time. Therefore, a
comprehensive approach that takes the dynamic nature of technology seri-
ously involves examining the entire life cycle of technology, from its cre-
ation to eventual obsolescence (later, we will delve into pre-design and
end-of-use considerations). In this section, we will briefly highlight three
critical aspects frequently ignored in discussions of technology and value
that become apparent when considering technology from a maintenance
standpoint. In Section 9.3, these three aspects and the maintenance
 Maintenance of Value and the Value of Maintenance 217

perspective will also guide the exploration of the relationship between

value and technology in socio-technical systems.

9.2.1 Maintenance: Beyond Design

The maintenance perspective illuminates the crucial insight that there is

more to technology than just design. Discussions about technology are
often guided by what Mark Young calls the design paradigm (2020), which
primarily focuses on a technology’s design and the role of the designer’s
intention in creating it. This paradigm ignores the entire temporal dimen-
sion of technology, as it places design as the primary focus and views what
happens to the technology afterward as secondary.
The maintenance perspective contrasts the design paradigm by inviting
us to take the entire temporal spectrum of technology seriously. Technology
is not finished after the design stage, and the designer’s intention does not
determine it. Instead, as many maintenance scholars have noted (Young,
2020; Weber and Krebs, 2021), technology has multiple temporal dimen-
sions that require ongoing consideration. Therefore, we must recognise
that technology is an ongoing process that extends beyond its initial cre-
ation and design phase. Many technologies can be considered liminal be-
cause they are continually reshaped, and their form must be negotiated
(Suboticki and Sørensen, 2021).
Looking at technology through a maintenance lens can help us reevalu-
ate and shift away from deeply sedimented ways of thinking about tech-
nology and value. One of these sediments is the design paradigm, which
often dominates the discussions of value and technology. The focus on
design and the designer’s intention has resulted in approaches to value and
technology that are often framed in terms of design and the intentionally
designed properties of technical artefacts. For instance, some philosophers
of technology, like Ibo van de Poel and Peter Kroes (van de Poel and Kroes,
2014; van de Poel, 2020), argue that technical artefacts embody values
when their intentionally designed properties facilitate the realisation of
these values. For instance, Michael Klenk (2020) has dubbed accounts that
focus on design intentions ‘intentional history accounts of value
embedding’.
Furthermore, the design paradigm also influences accounts that do not
focus on design intentions and their role in value. For example, although
Boaz Miller (2020) recognises the long-term implications of values embed-
ded in technology and technology’s material longevity, Miller’s treatment
of value and technology focuses primarily on design. Although Miller
briefly brings up redesign, practices of how longevity can be achieved, like
maintenance, are neglected.
218 Steffen Steinert

What can the maintenance perspective contribute to discussions of value

and technology? By emphasising the longevity and stability of values, the
maintenance perspective encourages a forward-looking approach that
complements traditional design-focused accounts of value. Instead of
thinking that design alone determines value, we must also consider the
ongoing processes that maintain or transform values over the entire life
cycle of technology. In this sense, the maintenance perspective challenges
the narrow focus on the designer’s intention that characterises many ac-
counts of value and technology. As we will explore in Section 9.3, applying
the maintenance lens to technology also has important implications for
understanding socio-technical systems. But first, let us examine a second
crucial aspect that adopting a maintenance perspective brings to the fore,
and which will contribute to a more comprehensive understanding of the
relationship between value and technology.

9.2.2 Maintenance: Beyond Hardware

Although hardware repair is a crucial part of maintenance, it is not the

only aspect that deserves attention. Many maintenance scholars argue that
maintenance encompasses the social dimension as well. For instance,
Steven Jackson (2014) has emphasised that repair can help to maintain
meaning. Maintenance and repair extend beyond material artefacts to our
relationships, trust, health, honour, and status (Henke and Sims, 2020).
Our desire to restore justice and repair our political system also falls under
maintenance practices (Spelman, 2002). Therefore, the maintenance per-
spective highlights the interplay between the material and social aspects of
maintenance and the broader implications of maintenance beyond the
technical realm.
Maintenance studies have long recognised the symbiotic relationship be-
tween the maintenance of technology and the maintenance of social order.
As Stephen Graham and Nigel Thrift (2007) suggest, it can be challenging
to determine whether the focus of maintenance is the technical artefact it-
self or the social systems surrounding it. Andrew Russell and Lee Vinsel
(2018) argue that the maintenance of technology often goes hand in hand
with the maintenance and reproduction of social structures. Maintenance
scholars also emphasise that the upkeep of social order requires continu-
ous effort involving material maintenance and repair practices (Denis,
Mongili, and Pontille, 2015, p. 7). Furthermore, Christopher Henke (1999)
argues that while repair work fixes and maintains technology, it also re-
pairs and restores social order in the workplace. Repair practices frequently
involve reordering and fixing social relationships and beliefs, which Henke
calls ‘people repair’ (p. 56). In their recent book, Repairing Infrastructure,
Christopher Henke and Benjamin Sims (2020) argue that maintaining
 Maintenance of Value and the Value of Maintenance 219

infrastructure can sustain social and political order. They coin the term
‘socio-technical repair’ to describe processes of repair that involve both
material and discursive interventions to restore meaning.
Expanding our thinking about values and considering how values are
maintained can inspire a new perspective on value and technology. The
inclusive conception of maintenance that looks beyond hardware and ac-
knowledges the maintenance of society can help us see that the values em-
bodied in technology are not static or fixed but are continuously negotiated
and maintained.
The maintenance perspective provides a valuable complement to many
approaches in the philosophy of technology that prioritise the destabilising
effects of technology on values. While these discussions frequently centre
around the destabilising social impacts of technology use, such as social
media’s impact on mental health or sustainability concerns regarding en-
ergy systems, scholars have also developed analytical tools and frame-
works for exploring how technology can transform morality. One such
concept is techno-moral change (Swierstra, 2013; Kamphof, 2017), which
highlights the transformative impact of technology on morally salient
practices. Other scholars have explored the potential for technology to
disrupt morality altogether (Nickel, Kudina, and van de Poel, 2022). In
contrast to these approaches, the maintenance perspective offers a crucial
counterpoint by emphasising how values are maintained over time, not
just how technology destabilises or transforms values.
Although it is essential to recognise the potential of technology to desta-
bilise morality and social order, it is equally important to acknowledge
technology’s role in stabilising and maintaining these values. Adopting a
maintenance perspective reminds us that technology has a socially and
morally disruptive potential and stabilises social and moral practices, insti-
tutions and orders. Stabilising technology by maintenance, then, also sta-
bilises and maintains a particular social and moral order. As Bruno Latour
(1990) famously argued, technology is society made durable. We can ex-
tend this idea and suggest that maintaining technology makes society and
values more durable.
The stabilising effect of technology is especially evident with entrenched
technologies, which are deeply embedded in the social fabric and can rein-
force certain values and social practices. For example, the combustion en-
gine and the infrastructure supporting it have stabilised a value regime
centred on individual independence, freedom and car ownership as a status
symbol. Car use and ownership are deeply socially ingrained in many cul-
tures and discussions about them are often highly politicised. This is evi-
dent in some countries, such as Germany, where attempts to establish
nationwide speed limits on highways face significant resistance from driv-
ers and the automobile lobby.
220 Steffen Steinert

Adopting a maintenance perspective helps us to gain a better under-

standing of the relationship between technology and value. While the de-
sign paradigm focuses on the creation and innovation of new technologies
and their corresponding values, the maintenance lens highlights the ongo-
ing work of preserving and sustaining technology and value over time.
This lens sheds light on often overlooked aspects of technology and value,
including the maintenance of social institutions and actors. At first glance,
maintenance may appear to be a conservative force that merely sustains
the status quo. However, as we will explore in Section 9.2.3, maintenance
can be transformative.
The maintenance perspective provides a broader understanding of tech-
nology and value by highlighting the need to maintain hardware and social
relations. This raises an important question: does the maintenance of tech-
nology also involve the maintenance of values? The answer, according to
the maintenance perspective, is yes. Later, we will explore how this insight
helps us better understand the maintenance of socio-technical systems.

9.2.3 Maintenance: Beyond Conservation

Many maintenance activities are forward-looking and focus on the stabil-

ity or continuity of some technical artefact or system over time. However,
maintenance is not always about restoring the status quo or conserving
something. For instance, maintenance can also involve upgrading technol-
ogy to meet changing demands or technological advancements. In addi-
tion, as discussed in Section 9.2.2, the maintenance perspective opens
space for thinking about how maintaining technology relates to the main-
tenance of value.
In the literature on technology maintenance, there remains a tendency to
emphasise conservation and preservation. For instance, Andrew Russell
and Lee Vinsel argue that maintenance is all the work to preserve particu-
lar technical and physical orders (2018, p. 7). Similarly, Heike Weber and
Stefan Krebs (2021) propose that repair, together with other forms of up-
keep, is an intervention that prolongs the time that technology can stay in
use. Furthermore, Gabriele Schabacher proposes that maintenance is a
‘prospective routine procedure to prevent all forms of disorder’ (2021).
Also, Steven Jackson (2014) stresses order and proposes that repair in-
volves acts of care that maintain order and meaning in socio-technical sys-
tems. This does not mean that these authors don’t allow for transformation
and change or consider maintenance exclusively in terms of conservation
and preservation. Still, these aspects seem to be the primary focus.
Although there is a tendency to view maintenance as a force that con-
serves and sustains social order and values, scholars have pointed out that
 Maintenance of Value and the Value of Maintenance 221

maintenance can be a creative, innovative and transformative process

(Graham and Thrift, 2007; Russell and Vinsel, 2018; Vinck, 2019; Young,
2020). Rather than exact restoration, maintenance often allows for impro-
visation and innovation (Graham and Thrift, 2007). Furthermore, Henke
and Sims (2020, p. 21f.) distinguish between repair as maintenance and
repair as transformation, highlighting the potential for maintenance to en-
able change in existing social orders, including power structures and social
practices. While repair as maintenance conserves and restores the status
quo of practices and structures of power, repair as transformation can en-
able changes to these power structures and social practices. Therefore,
maintenance should not be seen as solely a conservative force but also as a
force that can bring change and transformation.
The maintenance perspective offers a valuable framework for discussing
technology and values and goes beyond the traditional focus on value dis-
ruption. Not only does maintenance help maintain and stabilise existing
values but it can also be a transformative process that leads to value trans-
formation. This perspective encourages us to pay attention to value stabil-
ity and transformation through maintenance practices. We should consider
supplementing existing discussions about the relationship between tech-
nology and values with what could be called maintenance as value trans-
formation. This concept acknowledges that values other than those
intended by the designer can be embedded in technology through mainte-
nance practices, potentially contributing to significant changes in societal
values and norms.
In this section, we have highlighted how a maintenance perspective can
provide insights into important but often neglected aspects of value and
technology. We have identified three key features of this perspective. First,
the maintenance perspective encourages us to go beyond the design para-
digm to consider how values relate to technology after the technology has
been created. Second, maintenance is not just about hardware upkeep but
includes the maintenance and repair of social dimensions. The mainte-
nance perspective emphasises the importance of value stability and conser-
vation. Third, maintenance scholars have pointed out that maintenance is
not merely a conservative force but can also be transformative. This opens
possibilities for thinking about how values can be transformed through
technology maintenance.
In summary, adopting a maintenance lens gives us a valuable perspective
for analysing the relationship between technology and values. In the up-
coming section, we will put the maintenance perspective to work. We will
explore how the maintenance perspective can improve our analysis of
socio-technical systems and value, building upon the three aspects intro-
duced previously.
222 Steffen Steinert

9.3 Maintaining Technology, Maintaining Value: The Case of

Socio-technical Systems
This section will explore how the maintenance perspective can enhance
our understanding of the relationship between values and technology by
applying this perspective to socio-technical systems.
Taking a systems perspective on technology and focusing on socio-tech-
nical systems is warranted because even singular technical artefacts are
embedded in broader systems. For example, a single bicycle is not an
isolated technology but is embedded in the broader socio-technical trans-
portation infrastructure, including roads, bridges, cars and public trans-
portation. Furthermore, most highly impactful technologies, like transport
infrastructure and artificial intelligence, are socio-technical systems that
integrate social and technical elements. Therefore, by applying the mainte-
nance perspective to socio-technical systems, we can better understand
how the maintenance of technology and the maintenance of value and so-
ciety are related.
Maintenance scholars have long recognised the importance of socio-
technical systems, particularly infrastructure. For example, in their treatment
of infrastructure repair, Henke and Sims (2020) describe infrastructures as
socio-technical systems. Examining socio-technical systems brings attention
to the interwoven relationship between technology and social factors, high-
lighting the links between technology and social structures, including power
and privilege. This approach offers unique insights because it connects
maintenance to value considerations and emphasises the crucial role of the
maintenance perspective in the analysis of technology and value. By adopt-
ing a maintenance perspective that considers socio-technical systems, we
can better understand how technology and values are maintained and trans-
formed over time.
Before we go into the details of the relationship between maintenance,
socio-technical systems and values, it is crucial to understand what socio-
technical systems are. According to several scholars (Kroes et al., 2006;
Ottens et al., 2006; van de Poel, 2020), socio-technical systems have three
essential components. The first component is the material elements, which
include technical artefacts and other hardware necessary for the system’s
functioning. The second component is the agents involved in the system’s
operation, which are primarily human but can also include artificial
agents (van de Poel, 2020). Finally, socio-technical systems consist of so-
cial, legal and economic institutions, which are required for the system’s
functionality.
Institutions are structures of established rules, conventions or behav-
ioural principles that structure social interactions (Fleetwood, 2008; Wilfred
Dolfsma and Rudi Verburg, 2008). Social norms are institutions because
 Maintenance of Value and the Value of Maintenance 223

they prescribe what to do in particular circumstances (Bicchieri, 2005;

Brennan et al., 2013). Institutions are enforced by social sanctions that dis-
courage non-compliant behaviour. Laws and regulations are also institu-
tions but in contrast to most social norms they are specified in writing and
enforced by formal sanctions, such as fines. Understanding these compo-
nents is essential to understand how maintenance can affect socio-technical
systems and values.
To illustrate how the different components of a socio-technical system
work together, we can turn to the Tokyo subway system, as described in
Michael Fisch’s ethnographic study (2018). This massive socio-technical
infrastructure integrates hardware (such as trains and tracks), agents
(including passengers and conductors) and institutions (such as social
norms and municipal regulations) to ensure the functioning of the system.
For instance, passengers have developed efficient boarding norms during
rush hour that limit train delays and keep the system running smoothly.
The example of the Tokyo subway shows how the different components of
socio-technical systems are tightly interconnected and must work together
seamlessly to enable the system to function properly.
All three components of the socio-technical system embody values, and
the hardware of a socio-technical system comprises technical artefacts that
embody values. How value embodiment occurs is subject to philosophical
debate, and various accounts of value embodiments have been proposed,
such as the historical-intentional account (van de Poel and Kroes, 2014) or
the affordances account (Klenk, 2020). Regardless of one’s preferred ac-
count of value embodiment, it is plausible that technical artefacts in a
socio-technical system embody values.
Similarly, values are embedded in the institutional component of a socio-
technical system. We can follow Ibo van de Poel (2020) here, who has
suggested that values are embedded in the same way they are embedded in
artefacts. He proposed that an institution embodies a value when it is con-
ducive to this value because it has been designed for that value. Traffic
rules are one example that van de Poel gives to illustrate his argument. For
instance, the rule to drive on one specific side of the road embodies traffic
safety because it was intentionally designed to facilitate this value. If all
traffic participants follow it, traffic will indeed be safer.
Finally, human agents in a socio-technical system also embody values.
While technical artefacts may be designed with specific values and functions
in mind, human agents do not have such design features. Nonetheless, it is
reasonable to say that human agents embody values because they endorse
values, are motivated by them, and sustain them through their behaviour.
For example, engineers who endorse the value of safety will think that safe
technology is a desirable state of affairs and will consequently be motivated
to design a safe product. They will feel a sense of satisfaction when they
224 Steffen Steinert

succeed in designing a technology they consider to be safe. Different ap-

proaches have been proposed to characterise individual values, including
values as conceptions of the desirable (Kluckhohn, 1951), values as abstract
trans-situational goals or ideals (Schwartz, 2015; Maio, 2016) or values as
patterns of relatively robust attitudes, such as emotions and desires, that
provide reasons for actions (Tiberius, 2018). Despite the various accounts
of values, it is widely accepted that human agents hold or endorse values.
Now let us discuss how maintenance relates to socio-technical systems
and values. To ensure the proper functioning of the entire system, it is
necessary to maintain all three components of a socio-technical system –
hardware, agents and institutions. Furthermore, a maintenance perspective
can enhance our understanding of values and technology by recognising
that maintaining the components of a socio-technical system also entails
preserving values embedded in them. The maintenance of these compo-
nents and their embedded values are illustrated in Figure 9.1.
Let us now consider the maintenance of the components one by one,
beginning with the maintenance of hardware.
When it comes to maintaining a socio-technical system, the hardware
component is crucial. Technical components inevitably break down
or wear out over time and thus must be repaired or replaced to ensure
the system’s proper functioning. Take, for instance, the complex socio-
technical system of an airport. It comprises various hardware compo-
nents, including airplanes, airport buildings, air traffic control equipment
and runways, all requiring ongoing maintenance to keep the airport oper-
ating smoothly.
The maintenance perspective reveals that maintaining the hardware is
not solely about replacing parts or repairing broken components. It also

Figure 9.1 Maintenance and socio-technical system.

 Maintenance of Value and the Value of Maintenance 225

involves maintaining the values that are embodied in the hardware.

Depending on which account of value embodiment one finds most con-
vincing, the maintenance of values will take different forms. For example,
the affordances account of value embodiment (Klenk, 2020) asserts that
artefacts embody values because they have affordances, which are rela-
tional properties that create action possibilities for the agent and make
some actions likelier than others.1 Technology affords or enables some
valuable actions more than others. Following the affordance view of value
embedding, maintaining technology would entail maintaining its affor-
dances and, hence, its values.
In contrast to the affordance view, on the function account of value em-
bodiment (van de Poel and Kroes, 2014), an artefact embodies values due
to the functional properties it was designed for. If the maintenance of the
hardware keeps the technology function stable, then the embodied value is
also preserved. Section 9.4 will explore the maintenance account for other
value domains besides design values.
The debates about value in the philosophy of technology often overlook
the fact that the values of technology are not static but rather require ongo-
ing maintenance to remain stable. The predominant focus on design has
led to a neglect of the role of maintenance practices in upholding values.
By contrast, a maintenance perspective highlights the importance of main-
taining values through active practices. This perspective provides a valu-
able counterbalance to the prevalent emphasis on value disruption and
destabilisation in the philosophy and ethics of technology. By highlighting
that values are stabilised through maintenance, it offers a fresh angle for
exploring and understanding the complex relationship between technology
and values.
However, the maintenance perspective recommends attending to conser-
vation and transformation, especially focusing on transformation through
maintenance practices. By applying this perspective to value and technol-
ogy, we can open up intellectual space to consider how values can be
transformed through maintenance. This approach challenges the implicit
assumption in many debates that values embodied in technology are fixed
in design. In reality, not only are embodied values maintained but they can
also be transformed. Through maintenance practices, we can transform
the values embodied in technology, thereby enabling it to meet new or
changing requirements. For instance, to better align artificial intelligence
with our value of justice, the maintenance of an algorithm may change the
system components responsible for producing biased results while main-
taining the rest of the algorithm.
To illustrate how maintenance as value transformation is viewed from
the standpoint of different accounts of value embodiment, let’s take the
example of the abovementioned affordance view. According to this view,
226 Steffen Steinert

technology embodies values by virtue of the affordances it provides, which

are the potential actions or interactions it allows or invites. Hence, chang-
ing the affordances of technology can result in a transformation of its em-
bodied values. For example, consider a social media platform designed to
maximise user engagement, but it has been found to lead to negative social
consequences such as addiction, polarisation and disinformation. By
changing the platform’s affordances, such as reducing notifications, it be-
comes more aligned with the value of promoting healthy social interac-
tions. In this way, maintenance practices that modify the affordances of
technology can lead to value transformation.
Let us now turn to the other critical elements of a socio-technical system.
As we have seen, taking the maintenance perspective seriously involves
recognising that maintaining the entire system requires more than hard-
ware maintenance. All components must be maintained to ensure that
the system functions correctly. Therefore, companies, governments or so-
ciety must consider maintaining all the components to maintain a socio-
technical system.
For example, to maintain a socio-technical transportation system like an
airport, it is necessary to maintain social institutions like air traffic control
regulations, which govern the system and enable it to operate. Similarly, to
maintain a road traffic system, it is crucial to ensure that hardware compo-
nents like roads and signs are in good condition, but it is equally important
to maintain institutions like traffic rules.
Above, we have highlighted that institutions, like artefacts, embody
values. However, the maintenance of institutions is often overlooked in
discussions of technology and value. Taking a maintenance perspective re-
quires us to consider the maintenance of institutions and how it relates to
maintaining embodied values. This perspective also highlights the impor-
tance of paying attention to how institutions are maintained and stabilised
over time. Maintenance scholars have stressed the relationship between
maintenance and social order, emphasising the significance of institutional
maintenance for the stability of values (Graham and Thrift, 2007; Russell
and Vinsel, 2018; Henke and Sims, 2020). Adopting a maintenance per-
spective can help us to better understand how institutions maintain em-
bodied values.
Let us briefly consider how institutions can be maintained. Institutions,
such as social norms and traffic rules, can be maintained through various
means, including positive and negative reinforcements. Negative reinforce-
ment includes sanctions, which can take different forms, from public dis-
approval to verbal or physical discipline. In some cases, the state usually
enforces physical sanctions through the police and courts. Positive rein-
forcement of institutions includes incentives to encourage people to adhere
to norms and other means of ensuring compliance. For instance, drivers’
 Maintenance of Value and the Value of Maintenance 227

education ensures that drivers internalise and follow traffic rules. At this
point, the maintenance of institutions overlaps with the maintenance of
actors, which we will discuss below.
The maintenance of enforcement mechanisms is indeed crucial for the
successful enforcement of institutions. However, it is important to note
that maintaining a socio-technical system does not just mean maintaining
these mechanisms but also the broader social and technical components
that enable these mechanisms to function effectively. For example, main-
taining a traffic enforcement system not only involves hiring and training
police officers but also ensuring that the road infrastructure is in good
condition, traffic signals are functioning properly and drivers are educated
about the rules. All these components work together to create an effective
traffic enforcement system.
Furthermore, it is important to recognise that we should not ignore the
connectedness of different socio-technical systems. The maintenance of
one system can impact the maintenance of another system, and the failure
to maintain one system can have ripple effects on other systems. Therefore,
a comprehensive understanding of socio-technical systems and their inter-
connectedness is necessary for a comprehensive understanding of mainte-
nance, but we will leave this complexity aside here.
To fully embrace the maintenance perspective, we must move beyond
the idea that maintenance preserves the status quo. Maintenance can be a
tool for intentional change, including the transformation of institutions
and values embedded in them. In the case of socio-technical systems, insti-
tutions can be changed purposefully to embed new values. For example, if
there is a demand to make shipping more sustainable, the maintenance of
the system may require revising shipping routes to achieve this goal. The
new shipping routes will then embody the value of sustainability, reflecting
a deliberate effort to transform the institution to align with new values
(Künneke et al., 2015). This kind of intentional change of institutions is an
important aspect of maintenance that is often overlooked in discussions of
technology and value.
Another example of maintenance as institutional change has occurred
due to the rise of sophisticated language models like ChatGPT. In response
to concerns about how these models may reshape education and scientific
integrity (Cotton, Cotton and Shipway, 2023), many educational organ-
isations have sought to adapt their institutions, like policies and guidelines,
to maintain specific educational values, such as scientific integrity. By re-
vising their rules and guidelines for scientific integrity, these organisations
have adapted to the new reality and ensured that their values remain em-
bedded in the educational system. This educational change highlights how
the maintenance of socio-technical systems often requires the purposeful
transformation of institutions to embed new values.
228 Steffen Steinert

Before we move on, it is important to note that the above examples of

institutional change, such as the revision of shipping routes and changes to
scientific integrity guidelines, involved changes in formal rules and regula-
tions. However, not all institutions in a socio-technical system are formal.
Some institutions, like social norms, are informal, and their change can
often be unintentional.2
The maintenance perspective considers the temporal dimension and ac-
knowledges that institutions are not fixed. In addition to focusing on the
maintenance required to stabilise institutions and their embedded values,
the perspective also highlights the potential for institutional transforma-
tion. By recognising this dynamic aspect of socio-technical systems, the
maintenance perspective provides a more nuanced understanding of these
systems.
Finally, let us turn to the maintenance of agents as components of socio-
technical systems. Besides external factors like sanctions, agents follow in-
stitutions because they have internalised them through processes like
socialisation and habituation (Fleetwood, 2008). Because agents likely in-
ternalise certain institutions during their education and training, maintain-
ing internalisation processes contributes to maintaining institutions. Thus,
it is crucial to recognise that the maintenance of agents involves ensuring a
steady supply of agents and the continuous development and training of
existing agents. This can involve providing professional development op-
portunities, updating knowledge and skills and promoting a culture of on-
going learning. The maintenance of agents also encompasses recruiting
and retaining a diverse range of agents to reflect the diverse needs of the
socio-technical system and its users.
Moreover, the maintenance of organisations that shape agents involves
not just universities and vocational schools but also other institutions that
support agents, such as professional associations and networks. These or-
ganisations provide resources, mentorship and networking opportunities
vital for agents’ ongoing development and support. The maintenance of
these organisations requires a sustained effort to ensure they remain rele-
vant, responsive and accessible to the agents they support.
The link between agents and institutions highlights the importance of
maintaining internalisation processes. Agents’ internalisation of social in-
stitutions, like norms, rules and conventions, underpin the socio-technical
system. To maintain these internalisation processes, organisations must
promote a culture that emphasises the importance of following social insti-
tutions and provides agents with opportunities to practice and reinforce
these institutions regularly.
However, the agents in socio-technical systems include more than just
operators. Users are another significant group of agents. For instance,
when focusing on travel infrastructure, train passengers and pedestrians
 Maintenance of Value and the Value of Maintenance 229

should be considered. A maintenance perspective should pay attention to

this group because, as we have seen, it looks beyond design and considers
what happens after technology leaves the factory. For example, in the case
of the Tokyo subway discussed previously, how commuters use a transpor-
tation system, including norms of smooth entry and exit, can significantly
impact the system’s operation.
Earlier, we discussed how agents embody values. However, the question
of how agents’ values are maintained is often overlooked in the literature
on technology and values. This is a critical oversight because the mainte-
nance of values is relevant to technology’s smooth functioning and stabil-
ity. Adopting a maintenance perspective can help correct this oversight by
emphasising the importance of maintaining values in addition to other
components of the system.
To illustrate the role of values in the maintenance of a system, let us
consider the example of traffic infrastructure. Drivers who endorse the
value of safety will behave in a way that does not harm the material parts
of the infrastructure. For instance, in certain countries with colder cli-
mates, valuing safety means changing tyres between winter and summer,
which avoids damaging the road3. Similarly, a plane passenger who values
efficiency will likely behave in a way that contributes to quick boarding,
which helps with the smooth operation of the flight.
Now that we have established that the values of agents, operators, engi-
neers and users are relevant for maintaining technology, we must ask how
the values embodied in agents are maintained. A part of the answer lies in
considering organisations that shape agents and their values. First, let us
briefly consider the organisations that shape operators and engineers and
then turn to the topic of maintenance of users’ values.
To fully understand the maintenance of values in socio-technical sys-
tems, it is important to consider the role of various organisations in shap-
ing the values of the agents involved. While it may not be possible to
provide an exhaustive list of all the factors that influence the values of
these agents, certain organisations, such as universities and vocational
schools, play a particularly significant role in shaping the values of opera-
tors and engineers. During their education and training, these agents inter-
nalise values crucial for maintaining the socio-technical system, such as
efficiency, safety and sustainability. These values, because they translate
into behaviour, like designing an efficient product, are conducive to main-
taining the socio-technical system. Organisations play a key role in shaping
and maintaining the values of users because they influence the internalisa-
tion of norms. For instance, a driver’s education program is mandatory in
many countries, and drivers must pass a test before receiving their license.
This education likely instils values such as safety and responsible driving in
the prospective driver. In addition, formal rules and regulations and official
230 Steffen Steinert

sanctions, like fines, help maintain drivers’ values and keep their behaviour
in line with established institutions crucial for maintaining the socio-tech-
nical system.
The interdependence of the components of a socio-technical system is an
essential feature of such systems. As we have seen, agents’ values are closely
related to institutions and maintaining one component can contribute to
maintaining the other. However, it is also important to note that the com-
ponents can be maintained independently, and the intention to maintain
one component does not necessarily imply the intention to maintain the
whole socio-technical system. For instance, a government may wish to
maintain certain rules and regulations regarding the use of technology
without the intention to maintain the entire socio-technical system. For
example, a government may enforce strict data privacy laws without the
intention to maintain the entire digital infrastructure. Similarly, the
European Union may wish to maintain its rules about the use and design
of cars without the intention to maintain the socio-technical traffic system
in its member states.
So far, we have examined socio-technical systems from a maintenance
perspective, which proved to be a powerful tool for identifying key compo-
nents of the connection between technology and value previously over-
looked. However, our treatment was more like a snapshot view. For this
reason, our considerations were not entirely faithful to the dynamism em-
braced by the maintenance perspective. To fully align with the mainte-
nance perspective, we must consider technology through time, particularly
how values are maintained throughout a technology’s lifespan. Section 9.4
will focus on technology and maintenance over time.

9.4 Technology through Time: Maintenance and Value Domains

The maintenance perspective shifts our attention from the initial design of
technology to its ongoing trajectory through time. It encourages us to focus
on how technology is stabilised and transformed. In contrast to accounts
that emphasise the importance of design, the maintenance perspective
highlights the ongoing importance of maintenance, repair and adaptation.
While design is crucial, it is only the beginning of technology’s life cycle. In
this section, we will see how adopting a maintenance perspective can give
us insights into how values are maintained in different domains throughout
the life cycle of technology.
At this point, it is crucial to introduce how a maintenance account of
value and technology would consider the whole life cycle of technology. To
provide such an account, let us revisit the concept of embodied values we
discussed earlier. We previously noted that some philosophers argue that
values are embodied in technology through design. However, it is essential
 Maintenance of Value and the Value of Maintenance 231

to recognise that design is only one of many domains that contribute to

shaping the values embodied in technology. Many philosophers of technol-
ogy also emphasise the importance of the use domain and its associated
values4 (van de Poel and Kroes, 2014; van de Poel, 2020).
The values related to design are intended values, which are the values
that designers or developers aim to embody in the technology they create.
For instance, the designers may intend for the technology to embody val-
ues such as sustainability or safety. In addition to intended values, we can
distinguish realised values. These are the values that are actually realised in
the use of technology. Notably, intended values and realised values may
not always align. For instance, it may turn out that the technology is not
sustainable or safe once people use it. Finally, there are embodied values,
which we have previously discussed. Embodied values are intentionally
embedded in the technology and realised in use because the technology has
been designed for these values (van de Poel, 2020). Let us now turn to a
maintenance account of value and technology, which can provide a broader
analytical perspective than existing accounts of value and technology.
Distinguishing between design and use is a good start, but to fully cap-
ture the relationship between technology and value, a maintenance account
would emphasise the full temporality of technology. Adopting a mainte-
nance perspective means looking beyond just design and use and consider-
ing the value domains that span the entire life cycle of a technology.
The life cycle of technology involves more than design and use; it also
includes a domain that could be called discard. While discard may appear
simply as a form of use, it is prudent to distinguish between use and dis-
card because what happens to technology after it becomes obsolete may
not necessarily affect how people use it or how creators intend it to be
used. For instance, how to discard technology may not even be considered
in the use plan. A use plan describes an intended use of technology, includ-
ing a set of manipulations of the technology so that it can adequately fulfil
its function (Houkes et al., 2002). This is not to say that discarding a tech-
nology is never part of a use plan or that designers cannot translate their
intentions regarding how technology is discarded into design requirements.
For instance, if designers and engineers intend technology to be recyclable,
they can make devices that are easy to disassemble. Nevertheless, an out-
look that focuses on design and use, like the predominant paradigm in the
philosophy of technology, invites us to do, risks ignoring the entire life
cycle of technology. Adopting a maintenance perspective, which means
focusing on technology over time, helps to prevent this limited view of
technology.
It is important to acknowledge that discarding technology is not a value-
free process because it is connected to value-sensitive issues such as sus-
tainability, justice and safety. For example, many discarded clothes and
232 Steffen Steinert

computer technologies end up in markets or landfills in the Global South,

where they are often disassembled or repurposed. Therefore, the discard do-
main can realise specific values like the use domain. The values realised in the
discard domain are related to the intended values designers aim to achieve
when the technology is discarded. Designers can significantly promote sus-
tainability, safety and other values during the discard phase by making the
technology easy to disassemble and break down into recyclable parts.
Besides the domains of design, use and discard, there is yet another do-
main that we should consider. In taking the social embeddedness of use
and design seriously, our thinking needs to reflect that the intended values
of design do not arise from a social vacuum. Therefore, adding pre-design
as a fourth domain to our analysis is crucial.
The pre-design domain encompasses the social and cultural factors that
contribute to the values intended by designers. For example, the values
that engineering students internalise during their education play a crucial
role in shaping the intended values of the technology they develop.
However, the values they learn in their education are influenced by larger
social contexts, such as cultural norms and expectations.
Additionally, the domain of pre-design includes the demands of employ-
ers and purchasers of technology. These demands are not isolated from
larger social values and expectations and are often shaped by them. For
instance, a company may demand that its technology be designed to mini-
mise environmental impact or prioritise user safety.
A maintenance account of value and technology requires us to adopt a
comprehensive perspective that considers the entire social life of technol-
ogy. To achieve this comprehensive view, we must focus on the four do-
mains introduced earlier, namely design, use, discard and pre-design and
examine how maintenance relates to values in each domain. The central
idea is that maintaining each domain helps maintain the associated values.
The relation of maintenance, domains and values can be illustrated in
Figure 9.2, which I adapted from van de Poel and Kroes (van de Poel and
Kroes, 2014; van de Poel, 2020).
In the remainder, we will elaborate on the maintenance of pre-design,
design, use and discard. It should be noted that there may be an overlap
between the types of maintenance, and my analysis is not exhaustive.
Furthermore, maintaining one domain can contribute to the maintenance
of multiple values.
To elaborate on the maintenance of pre-design, it is important to con-
sider the various social domains that shape the intended values that go into
the design of technology. As mentioned, engineering education plays a sig-
nificant role in shaping students’ values and beliefs. In their courses, students
are often confronted with tasks, materials and projects highlighting the im-
portance of instrumental values related to efficiency and cost reduction.
 Maintenance of Value and the Value of Maintenance 233

Maintenance
Pre-design intended values

Design embodied values

Use realized values

Discard realized values

Figure 9.2 Maintenance of value domains.

Students internalise these values, later shaping their design decisions about
intended values. Maintaining the status quo of engineering education that
features courses and curricula that mainly focus on some values and not
others, the university contributes (unintentionally) to maintaining some in-
tended values and not others. This need not be the case, however, and uni-
versities may opt for what Henke and Sims call ‘repair as transformation’
(2020, p. 21), which means finding ways to repair while changing some
practices, instead of simply repairing to conserve the status quo. For in-
stance, by including ethics education, students may also internalise moral
and social values that can inform the design of their future projects.
To further illustrate the maintenance of pre-design, consider the impact
of social and cultural norms on the intended values of technology design.
For example, gender roles and stereotypes may be deeply ingrained and
could influence the intended values of technology design. Maintaining a
status quo that perpetuates these norms would likely result in technologies
that reinforce and perpetuate gender biases and inequalities.
Maintenance of design relates to the maintenance of social factors that
influence the design of technology and its embodied values. Embodied val-
ues are the values for which technology is intentionally designed to facili-
tate. To illustrate, let us consider a previous example of rules and regulations
that prescribe which values should be intended and promoted in technol-
ogy design, such as rules concerning safety and sustainability standards. By
motivating designers to create technology in a certain way, these institu-
tions make it likely that the technology furthers specific values. Maintaining
institutions that ensure the technology is adequately designed then main-
tains embodied values.
Recall that intended and embodied values can misalign, for instance,
when the technology is not designed correctly, is misused or has unin-
tended consequences. Some practices, such as the tendency or even require-
ment to choose the cheapest design option to save costs, can contribute to
234 Steffen Steinert

this misalignment since designing affordably may come at the expense of

other values, such as the health and well-being of users. Thus, maintaining
these practices through economic incentives perpetuates the mismatch be-
tween intended and embodied values.
Maintenance of use concerns the conservation of factors that affect tech-
nology use, thereby maintaining realised values. Like pre-design and de-
sign, social institutions such as customs, laws and regulations are crucial in
maintaining specific technology use. Maintaining certain norms of technol-
ogy use through incentives or sanctions can ensure that technology is used
according to its intended purposes and values. For example, laws that gov-
ern the use of vehicles and organisations that enforce sanctions, like the
police and courts, aim to ensure that cars are used safely. The government
may also reward specific technology use by providing tax breaks.
When technology is not used as intended, and intended values and re-
alised values misalign, society and designers must find ways to maintain
specific technology use. Municipalities, for instance, may create new regu-
lations to increase the safe use of E-scooters by restricting where they can
be ridden. Besides social maintenance of use, there is also technical main-
tenance of use. Designers have tools to maintain specific technology use,
such as apps needed to operate the scooter, which can restrict where the
scooter can be parked or limit the maximum velocity in certain areas of the
town, realising the value of safety in traffic. Designers can also achieve
technical maintenance of use by designing technology with specific affor-
dances (Klenk, 2020; Steinert and Dennis, 2022), which makes particular
forms of technology use more likely, thereby maintaining realised values.
There is some overlap between the different forms of maintenance,
which means that maintaining one domain, and its accompanying values,
can contribute to maintaining other values. This is particularly evident in
the maintenance of use. When social institutions or affordances maintain a
specific use of technology, it maintains the realised value and embodied
value because the latter is intended by the designers and realised in use. In
other words, by maintaining a specific use of technology, designers and
society can ensure that the technology’s intended values are realised.
Finally, the maintenance of discard concerns the factors that influence
how technology is treated after it loses its usefulness and hence, which
values are realised in this treatment. Discard shares many similarities with
use and could be viewed as a form of use. Therefore, the aforementioned
points about the maintenance of use applies equally to the maintenance of
discard. Social institutions, such as norms and laws concerning recycling,
contribute to how technology is discarded, and maintaining these institu-
tions maintains the accompanying realised values, such as sustainability.
Providing the necessary infrastructure to discard technology can also be
a way of maintaining the values associated with discard. For example,
 Maintenance of Value and the Value of Maintenance 235

providing convenient opportunities for people to discard their old elec-

tronic devices so that they can be collected and appropriately recycled fa-
cilitates the realization of values like sustainability and safety.

9.5 Conclusion
The chapter’s main purpose has been to demonstrate how examining tech-
nology and values from a maintenance perspective can enhance discussions
of the relationship between value and technology in the philosophy of
technology.
We have explored three fundamental aspects of a maintenance perspec-
tive, which have helped us to understand the relationship between mainte-
nance and value more fully. First, taking a maintenance perspective enables
us to move beyond the design-centric approach and ask how value stability
is maintained after the initial design. Second, the maintenance perspective
on technology and value allows us to recognise that technology can sustain
values rather than merely destabilise them. Third, maintenance is not lim-
ited to preservation but can also be transformative. Adopting a mainte-
nance perspective allows us to consider how the maintenance of technology
can contribute to the transformation of values.
The maintenance perspective offers a novel lens to examine technology
and value. To demonstrate the usefulness of this perspective, we applied it
to socio-technical systems, highlighting how maintaining such systems pre-
serves the values embedded in their components and the relationships be-
tween social and technological maintenance. The maintenance perspective
also allows us to explore value considerations in the whole life cycle of
technology, thereby opening up new avenues for philosophical inquiry into
technology and value.
An upshot of a maintenance account of value and technology is that
extant perspectives on technology and value should be expanded. Adopting
a maintenance perspective can also inspire new investigations within phi-
losophy and ethics of technology. For example, if maintenance practices
contribute to the stability of values, then responsible maintenance should
supplement responsible design approaches, such as value-sensitive design
or design for values (van den Hoven, Vermaas, and van de Poel, 2015;
Friedman and Hendry, 2019).
Responsible design aims to facilitate responsible practices of technology
design by considering technology’s impact on people, the environment and
society. As a supplement to responsible design, responsible maintenance
would extend these value considerations to technology maintenance.
Responsible maintenance differs from frameworks like design for main-
tenance and maintainability approaches (Dhillon, 1999). Design for main-
tenance aims to anticipate and plan for future maintenance needs to ensure
236 Steffen Steinert

that the technology remains in good condition or continues to promote

safety and health (Martinetti and Singh, 2019). Design for maintenance
refers to processes during the design phase of technology, while responsi-
ble maintenance focuses on the impact of maintenance. Although design
for maintenance is relevant and can include important value consider-
ations, responsible maintenance supplements design for maintenance in
that it infuses maintenance with value considerations by considering the
impact of maintenance on people and society. For instance, by questioning
what is maintained and by whom and which values are sustained and
transformed through maintenance?
Furthermore, to further develop a maintenance account of value and
technology, a possible next step is to investigate the mechanisms of value
stability and transformation. For instance, it would be valuable to consider
when and under which conditions maintenance practices contribute to sta-
bilising or transforming values. This could involve analysing specific case
studies of technology maintenance and their impact on values or develop-
ing theoretical frameworks to guide such analyses. Additionally, exploring
the ethical implications of maintenance practices and their effects on values
could be a fruitful area for further investigation.
One motivation for the chapter was to convince readers that the main-
tenance perspective can make valuable contributions to the philosophy
of technology. A maintenance account of value and technology, which
highlights the crucial role of maintenance in sustaining and transforming
values, helps to shake loose common assumptions in the field. More work
is necessary to establish a complete maintenance account of value and
technology. It is key that we maintain our efforts to develop and deepen
our understanding of the relationship between maintenance and value in
technology.

Notes
1 The idea of affordances originated in ecological psychology and was developed
by psychologist James Gibson. According to Gibson, affordances are what the
environment ‘offers the animal, what it provides or furnishes, either for good
or ill’ (1979, p. 127). Affordances can be characterised as the relation between
the abilities of the organism and the features of the environment (Chemero,
2003). As relational phenomena, affordances depend on the properties of the
environment and the features of the agent interacting with it. Design scholar
Don Norman applied the idea of affordances to human–technology interac-
tions. Norman proposed that affordances are ‘[…] perceived and actual proper-
ties of the thing, primarily those fundamental properties that determine just
how the thing could possibly be used’ (1988, p. 9). For instance, a doorknob
affords grasping whereas a computer mouse and a trackpad afford particular
ways of handling. The idea of affordance helps to analyse what technology
does and how it does it. Recently, Jenny Davis (Davis, 2020) has presented
 Maintenance of Value and the Value of Maintenance 237

mechanisms of affordances that help to discern how technology shapes social

behaviour and its political implications.
2 Further discussion of these types of norm changes can be found in Brennan
et al. (2013, p. 107ff.).
3 I would like to thank the editor for suggesting this example.
4 This is an oversimplification and the phases need not be strictly compartmen-
talised. There are feedback loops, and the use can inform the design (see, for
instance, van de Poel (2020)).

References
Bicchieri, C. (2005) The Grammar of Society: The Nature and Dynamics of Social
Norms. Cambridge: Cambridge University Press.
Brennan, G. et al. (2013) Explaining Norms. 1st ed. Oxford: Oxford University
Press.
Chemero, A. (2003) ‘An Outline of a Theory of Affordances’, Ecological Psy­
chology, 15(2), pp. 181–195. https://doi.org/10.1207/S15326969ECO1502_5
Cotton, D.R.E., Cotton, P.A. and Shipway, J.R. (2023) ‘Chatting and Cheating:
Ensuring Academic Integrity in the Era of ChatGPT’, Innovations in Education
and Teaching International, pp. 1–12. https://doi.org/10.1080/14703297.2023.
219014820
Davis, J.L. (2020) How Artifacts Afford: The Power and Politics of Everyday
Things. Cambridge, Massachusetts: The MIT Press (Design thinking, design
theory).
Denis, J., Mongili, A. and Pontille, D. (2015) ‘Maintenance & Repair in Science
and Technology Studies’, Tecnoscienza Italian Journal of Science & Technology
Studies, 6(2), pp. 5–15.
Dhillon, B.S. (1999) Engineering maintainability: how to design for reliability and
easy maintenance. Houston, TX: Gulf Pub. Co.
Fisch, M. (2018) An anthropology of the machine: Tokyo’s commuter train net-
work. Chicago; London: The University of Chicago Press.
Fleetwood, S. (2008) ‘Institutions and Social Structures’, Journal for the Theory of
Social Behaviour, 38(3), pp. 241–265. https://doi.org/10.1111/j.1468-5914.
2008.00370.x
Friedman, B. and Hendry, D. (2019) Value sensitive design: shaping technology
with moral imagination. Cambridge, Massachusetts: The MIT Press.
Gibson, J.J. (1979) The ecological approach to visual perception. Boston: Houghton
Mifflin.
Graham, S. and Thrift, N. (2007) ‘Out of Order: Understanding Repair and
Maintenance’, Theory, Culture & Society, 24(3), pp. 1–25. https://doi.org/10.
1177/0263276407075954
Henke, C. (1999) ‘The Mechanics of Workplace Order: Toward a Sociology of
Repair’, Berkeley Journal of Sociology, 44, pp. 55–81.
Henke, C. and Sims, B. (2020) Repairing Infrastructures: The Maintenance of
Materiality and Power. Cambridge, Massachusetts: The MIT Press (Infra­
structures series). Available at: https://direct.mit.edu/books/book/4962/Repairing-
InfrastructuresThe-Maintenance-of
238 Steffen Steinert

Houkes, W. et al. (2002) ‘Design and Use as Plans: An Action-Theoretical Account’,

Design Studies, 23(3), pp. 303–320. https://doi.org/10.1016/S0142-694X(01)
00040-0
van den Hoven, J., Vermaas, P.E. and van de Poel, I. (2015) ‘Design for Values: An
Introduction’, in J. van den Hoven, P.E. Vermaas, and I. van de Poel (eds)
Handbook of Ethics, Values, and Technological Design. Dordrecht: Springer
Netherlands, pp. 1–7. Available at: http://link.springer.com/10.1007/978-94-
007-6970-0_40 (Accessed: 15 October 2018).
Jackson, S.J. (2014) ‘Rethinking Repair’, in T. Gillespie, P.J. Boczkowski, and K.A.
Foot (eds) Media Technologies. The MIT Press, pp. 221–240. Available at:
https://academic.oup.com/mit-press-scholarship-online/book/14976/chapter/
169335302 (Accessed: 10 August 2022).
Kamphof, I. (2017) ‘A Modest Art: Securing Privacy in Technologically Mediated
Homecare’, Foundations of Science, 22(2), pp. 411–419. https://doi.org/10.1007/
s10699-015-9448-5
Klenk, M. (2020) ‘How Do Technological Artefacts Embody Moral Values?’,
Philosophy & Technology [Preprint]. https://doi.org/10.1007/s13347-020-
00401-y
Kluckhohn, C. (1951) ‘2. Values and value-orientations in the theory of action: an
exploration in definition and classification’, in T. Parsons and E.A. Shils (eds)
Toward a General Theory of Action. Cambridge, MA and London, England:
Harvard University Press.
Kroes, P. et al. (2006) ‘Treating Socio-technical Systems as Engineering Systems:
Some Conceptual Problems’, Systems Research and Behavioral Science, 23(6),
pp. 803–814. https://doi.org/10.1002/sres.703
Künneke, R. et al. (2015) ‘Understanding Values Embedded in Offshore Wind
Energy Systems: Toward a Purposeful Institutional and Technological Design’,
Environmental Science & Policy, 53, pp. 118–129. https://doi.org/10.1016/j.
envsci.2015.06.013
Latour, B. (1990) ‘Technology is Society Made Durable’, The Sociological Review,
38(1_suppl), pp. 103–131. https://doi.org/10.1111/j.1467-954X.1990.tb03350.x
Maio, G.R. (2016) The Psychology of Human Values. Routledge. https://doi.
org/10.4324/9781315622545
Martinetti, A. and Singh, S. (2019) ‘Design for Maintenance of Infrastructures: The
Lesson of the Morandi Bridge’, in S. Singh et al. (eds) Transportation Systems.
Singapore: Springer Singapore, pp. 43–54. Available at: http://link.springer.
com/10.1007/978-981-32-9323-6_4 (Accessed: 16 June 2023).
Miller, B. (2020) ‘Is Technology Value-Neutral?’, Science, Technology, & Human
Values. https://doi.org/10.1177/0162243919900965
Nickel, P.J., Kudina, O. and van de Poel, I. (2022) ‘Moral Uncertainty in
Technomoral Change: Bridging the Explanatory Gap’, Perspectives on Science,
30(2), pp. 260–283. https://doi.org/10.1162/posc_a_00414
Norman, D.A. (1988) The Psychology of Everyday Things. New York: Basic
Books.
Ottens, M. et al. (2006) ‘Modelling Infrastructures as Socio-technical Systems’,
International Journal of Critical Infrastructures, 2(2/3), p. 133. https://doi.
org/10.1504/IJCIS.2006.009433
 Maintenance of Value and the Value of Maintenance 239

Russell, A.L. and Vinsel, L. (2018) ‘After Innovation, Turn to Maintenance’,

Technology and Culture, 59(1), pp. 1–25. https://doi.org/10.1353/tech.2018.0004
Schabacher, G. (2021) ‘Time and Technology: The Temporalities of Care’, in
A. Volmar and K. Stine (eds) Media Infrastructures and the Politics of Digital
Time. Amsterdam University Press, pp. 55–76. Available at: https://www.
degruyter.com/document/doi/10.1515/9789048550753-004/html?lang=de
(Accessed: 10 August 2022).
Schwartz, S.H. (2015) ‘Basic individual values: sources and consequences’, in
T. Brosch and D. Sander (eds) Handbook of Value. Oxford University Press,
pp. 63–84. Available at: http://www.oxfordscholarship.com/view/10.1093/
acprof:oso/9780198716600.001.0001/acprof-9780198716600-chapter-4
(Accessed: 19 June 2019).
Spelman, E.V. (2002) Repair: The Impulse to Restore in a Fragile World. Boston:
Beacon Press.
Steinert, S. and Dennis, M.J. (2022) ‘Emotions and Digital Well-Being: on Social
Media’s Emotional Affordances’, Philosophy & Technology, 35(2), p. 36. https://
doi.org/10.1007/s13347-022-00530-6
Suboticki, I. and Sørensen, K.H. (2021) ‘Liminal Technologies: Exploring the
Temporalities and Struggles in Efforts to Develop a Belgrade Metro’, The Socio­
logical Review, 69(1), pp. 156–173. https://doi.org/10.1177/0038026120918166
Swierstra, T. (2013) ‘Nanotechnology and Technomoral Change’, Etica & Politica,
15(1), pp. 200–2019.
Tiberius, V. (2018) Well-being as Value Fulfillment: How We Can Help Each Other
to Live Well. Oxford, New York: Oxford University Press.
van de Poel, I. (2020) ‘Embedding Values in Artificial Intelligence (AI) Systems’,
Minds and Machines, 30(3), pp. 385–409. https://doi.org/10.1007/s11023-020-
09537-4
van de Poel, I. and Kroes, P. (2014) ‘Can technology embody values?’, in The
Moral Status of Technical Artefacts. Dordrecht; New York: Springer, pp.
103–124.
Vinck, D. (2019) ‘Maintenance and Repair Work’, Engineering Studies, 11(2),
pp. 153–167. https://doi.org/10.1080/19378629.2019.1655566
Weber, H. and Krebs, S. (2021) ‘The Persistence of Technology: From Maintenance
and Repair to Reuse and Disposal’, in S. Krebs and H. Weber (eds) Science
Studies. 1st edn. Bielefeld, Germany: transcript Verlag, pp. 9–26. Available at:
https://www.transcript-open.de/doi/10.14361/9783839447413-002 (Accessed:
19 July 2022).
Wilfred Dolfsma and Rudi Verburg (2008) ‘Structure, Agency and the Role of
Values in Processes of Institutional Change’, Journal of Economic Issues, 42(4),
pp. 1031–1054.
Young, M.T. (2020) ‘Maintenance’, in The Routledge Handbook of the Philosophy
of Engineering. Routledge, pp. 356–368.
10 An Eco-Ethics for the End of
the Anthropocene
Finding Ethical and Sustainable Paths
through Consumerism, Disposability and
Planned Obsolescence

Simon Penny

10.1 A Storm Is Blowing in From Paradise

From where we stand, two decades into the twenty-first century – in the
closing act of the Anthropocene we might say – we look back over the
detritus of 250 years of industrialism, colonialism and commodity capital-
ism. We might be reminded of Walter Benjamin’s meditation on history,
prompted by a small painting by Paul Klee:

A Klee painting named ‘Angelus Novus’ shows an angel looking as

though he is about to move away from something he is fixedly contem-
plating. His eyes are staring, his mouth is open, his wings are spread.
This is how one pictures the angel of history. His face is turned toward
the past. Where we perceive a chain of events, he sees one single catas-
trophe which keeps piling wreckage and hurls it in front of his feet. The
angel would like to stay, awaken the dead, and make whole what has
been smashed. But a storm is blowing in from Paradise; it has got caught
in his wings with such a violence that the angel can no longer close
them. The storm irresistibly propels him into the future to which his
back is turned, while the pile of debris before him grows skyward. This
storm is what we call progress.
(Walter Benjamin 1968)

That ‘progress’ has created greater storms. If we look ahead, we survey the
prospect of ever-increasing climate catastrophes and calamities: a swelter-
ing planet, a great extinction, dying oceans, wars over water in some
places, battles against water in others and apocalyptic fires and storms. In
2022, Pakistan was disastrously inundated. New York experienced mon-
soons, Hurricane Ian laid waste to large parts of Florida. California expe-
rienced droughts and floods simultaneously. The Colorado River is drying
up and the great Ogallala aquifer that feeds the US grain belt is nearly

DOI: 10.4324/9781003316213-12
 An Eco-Ethics for the End of the Anthropocene 241

empty. The forests of western USA and Canada, Siberia, Australia and
Amazonia erupt into fiery conflagrations. Enormous chunks of Antarctica
and Greenland ice are breaking off. Less obviously, arable land dimin-
ishes, aquifers salinate, oceans acidify, coral reefs die and fish populations
plummet.
The beginnings of global government, the League of Nations and then
the United Nations, were catalysed by the horror of the ‘world wars’. In
comparison, the climate crisis seems diffuse and abstract and the immedi-
acy of local social, economic and political issues, not to mention simple
profit incentives, pushes the larger crisis into the background. It may take
more dramatic disasters to galvanise nations into concerted action, but the
awareness of impending existential crisis is, at long last (and perhaps not
too late), growing across diverse communities and especially the youth.
One aspect of this end of the Anthropocene is species extinction. We
understand that it is ecologies as much as individual species that must be
saved. It is easy to evoke sympathy for turtles snagged in plastic but not so
easy for leeches or hagfish or mites. Likewise for grand forests or coral reefs
but not so much for saltmarshes or thawing tundra. It is a testament to ris-
ing consciousness that awareness of previously obscure(d) aspects of the
economy, like the waste stream is growing. Food waste, surprisingly, has
become a political issue. The re-use of municipal greywater is a topic of
public interest. Fast food, fast fashion and single-use plastics (SUPs) have
come under scrutiny. Recycling itself has been exposed as a sham. Terms
like Anthropocene itself, along with Circular economy, lifecycle planning
and carbon footprint, are more commonly used and understood.1
The idea of sustainability has attained a new urgency, but ‘sustainabil-
ity’, like so many overused catchphrases, is ripe for critique. Greenwashed
home and body products and appliances exploit sustainability concerns.2
Purchasing a new, ‘green’ appliance or smugly sporting clothing contain-
ing recycled fibres is evidence of being duped into further consumption.
In most cases, not being responsible for further extraction and manufac-
ture is the best choice. Keep the old thing and fix it. The ‘right to repair’
movement and ‘repair cafés’ attest to new interest in maintenance and
repair – the sad irony being that many appliances and consumer goods
are inherently irreparable.
Any thinking person recognises the urgent need for change – political,
social, technological and environmental – but few are ready to willingly
embrace deprivations. It is sobering to reflect that any middle-class west-
erner has at their disposal more energy than a king or emperor of the pre-
industrial world. The period of fossil fuel use is but a blip on the timeline
of humanity, but we in western culture are as addicted to fossil fuels as a
junkie is to heroin. As nuclear electricity was once touted as ‘too cheap
to meter’, we are naturalised to the assumption of cheap, convenient
242 Simon Penny

energy-consuming infrastructure: we flick on lights, we expect water to

flow from the faucet, thermostats run central heating or air conditioning to
keep us cosy or cool. Without thought we jump in the car to get groceries,
taking a ton of steel with us at speeds once thought deadly for humans
(5000 lb electric cars are still cars).
A central question of this chapter concerns the ethical reconciliation of
the global with the personal. We are all already compromised to the extent
that we take part in contemporary western culture. We travel in planes,
with some shame, but we still do it. The industrialised nations we live in
have profited from the production of greenhouse gases, that less-industrial
nations now suffer from the effects of. As Derrick Jensen argues in his essay
‘Forget shorter showers: why personal change does not equal political
change’ (2015),3 ethical behaviour on the personal level and effective (po-
litical) action are different things. Change at the personal level will not re-
configure the practices of major culprits of climate crime – industries,
corporations, cities and states.
So how should one conduct oneself in one’s daily, domestic life in order
to at least minimise hypocrisy? What kinds of activities ought one engage
and which eschew? In what follows, I offer suggestions for ways of recon-
ciling the global and the personal – the result of a lifetime of striving for
eco-ethical awareness in lived practice. This essay is written in two different
registers: in Part II, I map out a big-picture view of the production, distri-
bution and disposal of contemporary commodities in historical, economic,
regulatory and technological dimensions (I live in the United States; many
of my examples reflect that context). Parts III and IV are autobiographical,
even confessional, outlining my idiosyncratic approach to living that at-
tempts to satisfy, on a personal level, the ethical implications of Part II.4

10.2 Swimming in the Global Waste Stream

We exist in a consumer commodity economy that is predicated on contin-
ual re-purchasing of intentionally short-lifespan products. Industry and
marketing are oriented towards continuous manufacture, continuous (re)
consumption of roughly equivalent things and the untroubled assumption
of an endless waste stream. This is most obvious in clothing, automobiles,
domestic commodities and appliances. Big-Ag has embraced this logic with
successive new generations of pesticides and GMO crops, while the food
distribution industry deploys ‘sell-by’ dates that encourage unnecessary
disposal. Designed-in use-by dates have different contributing causes: tech-
nological change, materials lifetimes and the bald-faced market manipula-
tion we call fashion. ‘New! Improved!’ was a cliché of advertising in the
mid-twentieth century – this year’s model is just somehow sexier. The ex-
ploitation of psychological insecurities to boost sales was pithily captured
 An Eco-Ethics for the End of the Anthropocene 243

long ago in the lyrics of the Rolling Stones song ‘I can’t get no satisfaction’.
Manipulation of tastes in cosmetics, clothing, cars and appliances has been
a fixture in media for a century or more – in evidence since the early days
of the automobile industry – managed by the advertising industry, that
became an economic force in the early nineteenth century.
Planned obsolescence, like the growth economy itself, is a twentieth-
century notion. The term originated in the USA in the 1920s. It is achieved
in three main ways – goods become obsolete, unfashionable or irreparable.
As Rivera and Lallmohamed explain:

Creating goods with a limited lifetime led to increased consumption. It

was a business strategy to create mass consumption, which the country
needed in a time of economic crisis. Producers who once were produc-
ing quality products started to find ways to make goods either more
fragile, or difficult to repair so that people would be forced to replace
the older version sooner. Processes and devices have been created with a
predetermined lifetime to maximize economic outputs.
(Rivera and Lallmahomed 2016)

We are naturalised to a mindset that ownership of a commodity is inher-

ently short term and transitory – from 10 minutes in the case of SUP
packaging to 10 years for cars (with clothing and appliances in between).
The awareness of the fundamental problem of sustainability is growing.
But the global economy is a giant ship – turning it around will necessarily
take time. The reconfiguration of cities and lifestyles to lower energy con-
sumption and carbon neutrality will necessarily be a slow and incremental
process. All manner of infrastructure and livelihoods will have to be re-
configured and this reorientation will have deep ramifications for busi-
nesses, for the economies of states and for the health and welfare of
citizens.

10.2.1 Technological ‘Progress’

Those of us in a position to read this text are naturalised to relentless tech-

nological change, and the rhetoric thereof. Solid-state electronics, plastics,
chemical and pharmaceutical industries and chemical agriculture all
emerged or grew substantially in the post-WW2 period – all tightly inte-
grated into the growth of the larger petrochemical infrastructure. In the
same period, digital computing developed from an obscure research idea to
a ubiquitous technology – and a major consumer of energy globally.5 In my
lifetime, the number of artificial satellites orbiting the Earth has gone from
zero to around 8000 (the number continues to expand rapidly).6 Pushing
tons of metal up beyond the atmosphere must be the most grandiose
244 Simon Penny

display of energy use ever conceived – but who mentions the carbon foot-
print of rockets?
Among the industrially mass-produced artefacts and rapidly changing
technologies of commodity capitalism, improvement implies obsolescence.7
Commercialisation of technological developments implies commodities of
short lifespan: in periods of technological ‘revolution’ entire systems rap-
idly become obsolete: even if you could find a dial phone, there’s nowhere
to plug it in anymore. No one would use a 20-year-old desktop computer.
We have become accustomed to the assumption that Moore’s law will ex-
tend indefinitely.8 Unavoidable in the first decades of the computer era,
many developments in digital technologies are now seen as self-perpetuating
and sometimes frivolous.9 As often as not, ‘this year’s model’ is dressed-up
with fashionable shapes and colours while little has changed ‘under
the hood’. Bloated code demands faster processors and more memory.
Constructed desires drive technological escalations that lead to the natu-
ralisation of expectations. Technological peccadillos become infrastructure
or rights: people stream movies onto tiny screens of their phones while on
the train to work – oblivious of the vast, energy-consuming technological
infrastructure brought into play. In a commodity capitalist culture, the
only limit to the existence of such travesties is profitability. The perpetra-
tors of this nonsense should be called out as the environmental criminals
they are.

10.2.2 (Fast) Fashion

Today, many consumer commodities are designed to have a short cultural

relevance or desirability. Internet niche marketing and sophisticated digi-
tally coordinated manufacturing have moved the process into a hyper-
rapid temporal space – China-based international fast-fashion giant Shein
is a prime example. This fashion churn drives unsustainable production in
a number of ways – aside from stimulating the creation of petrochemical-
based fibres and textiles (nylon and polyester), the expected short cultural
lifespan of such goods predicts the use of cheaper materials that wear out
or break quickly. In addition, the vagaries of markets create environmental
outrages like the mountains of unsold fast fashion in the Atacama Desert,
where the high UV turns toxin-exuding synthetics into microplastics
(Duong 2021). If you make a good pair of shoes, I will buy one pair in
maybe 10 years. If you make shoes that fall apart, and cannot be fixed,
I will be forced to buy a new pair every year. You will get my money
10 times. Who can argue with that? Certainly not the bosses of industry
nor Nike stockholders nor the planners of national economies. The fash-
ion-conscious will not wear last year’s sneakers anyway.
 An Eco-Ethics for the End of the Anthropocene 245

10.2.3 Things Fall Apart

In the contemporary consumer commodity system, things are designed to

break, thus requiring (regular) repurchase and resisting both repair and
recycling (Slade 2007). Chemical and materials engineering is sophisti-
cated; materials are designed at a molecular level. An entire branch of the
discipline of materials science is dedicated to making things break on
time – manufactured materials with a designed-in lifespan that are engi-
neered to decompose at a certain age, seldom into Earth-friendly detritus –
timed to fall apart in accordance with the calculated buying power of the
customers for that product. The plastic that comprises semi-structural
parts in your car will crumble, fracture and decompose into toxic dust in
10 years (give or take for ambient UV). Tyres, hoses, wiring insulation,
interior panels, upholstery, window seals and ‘disposable’ air and oil
filters – when those materials begin to lose their structural integrity, there
is no fixing them.
Composite and synthetic materials are cynically designed to fail, with
reliability and precision, at a particular age, not unlike the replicants in
Blade Runner (or indeed, like all living things). We have succeeded in creat-
ing materials that fall apart that no known organism can metabolise.
The global plastics crisis – nonbiodegradability and the impossibility or
nonviability of recycling – has resulted in the global ubiquity of microplas-
tics and a recognition of their dire health impacts on individual organisms,
species and entire ecosystems. Petrochemically derived synthetic materials
dubbed ‘forever chemicals’ are particularly egregious: they enter the waste
stream but are not reintegrated into cycles of biological metabolism.
Acronyms for deadly chemicals such as BPA, PFAS and EDCs (endocrine-
disrupting chemicals) are now used in daily language.
Humans have battled the destructive qualities of weather, rodents and
micro-organisms on organic materials for millennia – wood rots, leather
gets hard and cracks and wool is chewed by moths. But that destruction is
also salvation, micro-organisms turning detritus into bioavailable molecules
to be truly recycled, in the great ecological chain. Planned obsolescence need
not imply irreparability nor unrecyclability nor non-biodecomposability,
but it usually does.

10.2.4 ‘Disposable’ Packaging and Single-Use Plastic

The infamous 1960s ‘Crying Indian’ ad campaign was a classic case of
corporate cynicism in response to the ‘litter crisis’ created by disposable
packaging – much of it the new technology of plastics.10 This faux PSA
(public service announcement) ad campaign produced by an organisation
called Keep America Beautiful in association with the AdCouncil shamed
246 Simon Penny

the public and made cleaning up ‘litter’ a civic responsibility (Summers

2019). If the inherent cynicism was not enough, the ‘Indian’, Iron Eyes
Cody, was actually a Hollywood extra of Italian descent who claimed
some indigenous heritage.11 The ‘Crying Indian’ campaign put what should
have been a governmental or corporate responsibility onto the shoulders of
consumers. The Indian is still crying.
The litter crisis was a result of the use of ‘disposable’ packaging compo-
nents that supported the new drive-through junk-food industry that
emerged in parallel with the rise of car ownership. Automobile-based
dining – only in America! Seen as a convenience half a century ago, dispos-
able packaging and SUP food containers (polystyrene foam, compressed
styrene, polythene, plastic-impregnated paper, etc.) have become a global
environmental disaster. Some things are necessarily consumables – my
food, the energy I use to cook it and the water I drink. But why should
I buy a new cup with every cup of coffee, just to discard it when the coffee
is gone? This is a kind of consensual socio-economic insanity. People all
over the world use and have used ‘disposable’ food containers – in Oceania,
banana leaves, and in Mexico, lightly fired terracotta – but crucially, these
materials bio-decompose.
Ludicrously, most of the plastic bottles we consume exist to move water.
Domestic and personal cleaning products – laundry soaps, floor cleaners,
shampoos and so on – that line the shelves of supermarkets are mostly water,
diluted for ‘convenience’ and no doubt for profit. So vast quantities of water
are trucked around the country in specially made plastic bottles. The solids
in a large bottle of shampoo would not fill a shot glass and a bar of soap will
make a gallon of ‘liquid soap’. Taking the water out would substantially
reduce packaging, retail floorspace and diesel consumed in transport.

10.2.5 There Is No Unscrambling Some Eggs

Composite materials, combined with planned obsolescence, make a mock-

ery of pretenses of recycling, maintenance and repair. Appliances and
furnishings – refrigerators, office chairs, printers, phones and synthetic
carpets – are all diabolical cocktails that defy demanufacture. The amal-
gams of materials in our commodities make them virtually impossible to
recycle and reclaim. There is no way, or no economical way, to tease out
and sort those molecules back into their categories. Such diabolical com-
posites are everywhere: combinations of metals and plastics in microelec-
tronics, combinations of paper and plastic in the humble milk carton,
laminations of different plastics in ‘disposable’ plastic water bottles or the
mixtures of synthetic and organic fibres in a polycotton shirt. In the hum-
ble sneaker, the sophistication of manufacturing processes and the bonding
of materials makes them irreparable and inherently unrecyclable because
they are composites. Miniaturisation and encapsulation of subcomponents
 An Eco-Ethics for the End of the Anthropocene 247

prohibit repair. No one repairs a microprocessor – there is no fixing such

things. In such contexts, repair amounts to pointing a diagnostic tool at the
system, throwing away a faulty component and ordering the replacement
component by the serial number in an Internet-linked database.12 It is
ironic that the most sophisticated forms of technology that humans have
made cannot be repaired.13

10.2.6 Recycling – The Big Lie

Corporate support for recycling policies can be interpreted as a way to

reduce consumer resistance to consumption of ‘disposable’ materials by
assuaging the potential consumption-guilt – business, apparently, knows
no shame. But ‘recyclable’ almost never means ‘will be recycled’. The per-
centage of plastic recycled in 2022 is estimated at 5%, recently downrated
from the 9% estimate of recent years.14 While we dutifully clean and sort
our recycling, most of the waste we put in the recycling bins ends up in
landfill.15 The remanufacture implied in recycling is ecologically fallacious –
the energy and material consumption of recycling is unsustainable, eco-
logically and economically. What little recycling does occur consumes
energy and new toxic materials (plasticisers and the like), creating toxic
waste and liberating potent greenhouse gases. Recycling is often predicated
on the availability of cheap energy. The processes involved in recycling
other materials we purport to recycle – aluminium, paper and steel – are
barely economically viable, due in part to high labour and huge inputs of
materials and energy.
Much so-called recycling is down-cycling. The ‘recycling’ of glass by my
local trash service amounts to crushing it for roadmaking – turning food-
grade glass into gravel hardly constitutes recycling. I know how much fuel
it takes to turn broken glass into new bottles. I had a friend who was a
craft glassblower, he had to save for three months to afford gas to fire his
kiln and lehr for one work session.16 After 10–12 hours firing the kiln,
burners roaring, a few kilos of glass was molten, so hot it glowed like the
sun. Having brought the glass to working temperature, he had to work
non-stop for 36 hours. Working on such a dangerous material while sleep
deprived was always a fraught proposition. Then, he would sleep while the
red-hot glass objects underwent controlled cooling (annealing) for days (or
they would explode).
When I was a child, our milk was home-delivered. Not so many years
before, the milkman had a horse-drawn cart – the horse knew to walk
slowly along the street, keeping pace with the milkman as he went back
and forth from cart to doorstep (who needs self-driving cars?). When the
milkman left the milk, he would take our empty milk bottles. At the dairy,
they would be washed and refilled. Some of our milk bottles were bur-
nished and scratched, evidence of hundreds, perhaps thousands of cycles of
248 Simon Penny

cleaning and refilling. How many bottles can you sterilise with the energy
it takes to remake a bottle out of melted crushed glass? Thousands?

10.2.7 Growth and Degrowth

Consumer spending currently comprises about 70% of US GDP.17 The

economy we know is predicated on repeated consumption, of the same
things, or the new model, over and over again – often driven by ‘fashion’,
deployed as a driver of consumption. Buy it today, buy it again tomorrow.
Incessant production, incessant consumption of resources and power and
incessant creation of waste: this engine of the economy is constructed
around the assumption of continuous, repeated consumption and continu-
ous disposal. Goods that wear out with no potential for repair. The waste
stream is the skeleton in the closet of commodity capitalism.
If we bought only what we needed, bought things that did not wear out
or become obsolete, local, national and global economies would crash. To
abandon disposable products and inbuilt obsolescence, to abandon con-
sumerism, would destroy the economy as we know it. Yet to continue on
this path is to ensure environmental destruction. This puts us in an awk-
ward position: the global commodity economy itself resists ‘maintenance’.
The growth economy is unsustainable.
What should be sustained? If the continued functioning of a system de-
pends on the consumption and throughput of resources that continuously
enter the waste stream, unreusable and unrecyclable, is that system sustain-
able? Viewed this way, the commodity economic system is like a voracious
animal, consuming and excreting. But unlike the cowpat, the diesel particu-
lates or the used toner cartridges are not input resources for another (bio-
logical) system. Externalities be damned! In a closed system, there are no
externalities – the biosphere is a closed system. Choosing to manufacture
the disposable, the unmaintainable is, in terms of eco-ethics, unacceptable.
What is ethically sustainable? The fallacy lurking at the heart of much
sustainability discourse is the assumption that the profligate consumption
we are used to is in any way sustainable. Just because we are accustomed
to taking a ton of steel with us to get the groceries or leapfrogging vast
oceans in a few hours, or for that matter, leveraging global digital net-
works so we can watch an episode of an old sitcom when we might other-
wise be looking out the window of the bus, does not mean that these can
or should be sustained.

10.2.8 The End of the Anthropocene

Increasingly, I feel like I am living in the final act of a global Shakespearian

tragedy. We stand on stage, appalled and aghast as the bodies pile up.
The longer-term results of impetuous avaricious behaviour are revealed.
 An Eco-Ethics for the End of the Anthropocene 249

Except perhaps for the very rich, luxurious lifestyles predicated on an

abundance of low-cost energy, access to information, freedom of move-
ment by land, sea and air, sophisticated medicine, disposable wealth, even
access to water, seem likely to abate. The global poor will, inevitably feel it
first and worst; indeed, they already do. Populations of climate refugees
will inevitably grow. Storms, floods, fires and droughts will contribute to
the failure of infrastructure.
We live immersed in highly interconnected systems that have built upon
other systems over generations. My home Internet comes on repurposed
cable TV infrastructure that literally hangs off power infrastructure in-
stalled 100 years ago. At some point, the interdependence of the systems
we depend on to facilitate new projects will begin to degrade and the eco-
nomic load of maintenance will itself become unsustainable. Components
of subsystems will fail, bringing down entire systems. In some US cities,
water infrastructure is failing – due in part to inadequate or deferred main-
tenance. Restoring century-old piping now lying under roads and buildings
is a gargantuan engineering task. Some things are so broken that there is
no fixing them.
Surviving with trucked and bottled water and porta-potties in anything
but the short term is unsustainable. Industries and services and stores will
shut down. The quality of life in those places will plummet and they will
depopulate. Collapse of some kind seems inevitable. The best we might
hope for is to mitigate, and hopefully survive, climate crisis and engineer
more parsimonious modes of living that maintain health, safety and dig-
nity. There is good reason to think that even this humble aspiration may
not be possible – that climate pressures will outpace us.
Russell Hoban published Riddley Walker (Hoban 1980) at the height of
the global fear of nuclear apocalypse – miraculously (that time), we es-
caped the nuclear annihilation that had so many of my generation living in
nihilistic dread. The novel is set in southern England a century or two after
a nuclear apocalypse: humanity has regressed to a kind of iron-age culture.
Radioactive bogs are dredged for C20th iron detritus – bits of buildings,
railways and factories. The scenario that Hoban laid out may as well de-
pict human life in a post-Anthropocene world, adapted to a depopulated
and detechnologised normal.

10.3 Bricolage Aesthetics: Maintenance, Repair and Salvage

Maintenance, at its most basic, implies continuation, as in maintaining a
vigil. Mark Thomas Young offers ‘a preliminary conception of mainte-
nance as an activity through which we provide stability to something over
time’ (2020). This stability clearly includes less-tangible things: institu-
tions, traditions, relationships and ‘the peace’. In the context of this
anthology and similar concerns, ‘maintenance’ is often associated with
250 Simon Penny

large-scale technological systems, civic infrastructure, industries and the

built environment (Henke &Sims 2020). My concern here is with tangible
things, artefacts and skills, in a more humble, domestic sense. Sweeping
and dusting and vacuuming and doing laundry: such (often gendered)
tasks do constitute maintenance. Maintenance implies managing wear, the
relentless assaults of entropy and decay. It seems to imply stasis: to bring
the artefact back to the as-new condition, as closely as possible – as often
as not, these days, with OEM replacement parts purchased online via a
part number. In some contexts, that might mean yet another coating of
paint or the replacement of a broken part with a new one.
Sewing on a button, sharpening the kitchen knives, re-grouting the tile-
work and pruning the fruit trees and grapevines – these maintenance tasks
blur into ‘repair’. The transition from maintenance to repair usually in-
volves replacement or addition. Repair implies know-how, skill and expe-
rience: ‘I know how to fix that’. At the end of repair that is furthest from
maintenance, the work has more creativity to it – improvisation and brico-
lage. Re-creation might not be an overly pretentious descriptor (restora-
tion is something else again: a fetishised, obsessive maintenance tinged
with nostalgia) (Figure 10.1).

Figure 10.1 Kettle repaired (by the author) with a piece of rosemary wood that had
grown in his garden, bound with marline.
Photo: S. Penny.
 An Eco-Ethics for the End of the Anthropocene 251

What happens when something is irreparable? A less-aware person will

consign whatever it is to the waste stream and the more aware will
repurpose – if we must buy yoghurt, we can at least use the plastic contain-
ers for something. Some well-meaning but misguided souls ardently craft
plastic waste into nasty handcrafted sustainable folk art – plastic bottles as
plant pots and so on – as if gilding such trash with folksy aesthetic somehow
redeems it from being the poison it remains: no silk purses from sow’s ears.
A more extreme form of repurposing is salvage – the territory of scaven-
gers, gleaners, lowly rag and bone men, scrap metal merchants, trash-
sorters and dumpster divers. My neighbour replaced her fence, the bottoms
of the palings were rotted. They were left in a roadside pile, destined for
landfill. I took the planks and trimmed them. I clad a shed with the per-
fectly usable, if weathered, lumber. The offcuts became firewood or mulch.

10.3.1 Salvage Eco-ethics

Beyond, or beside, an ethics of maintenance, is an ethics of repair, and be-

side those, an eco-ethical responsibility to salvage. Permaculture, the radi-
cal environmental philosophy developed by Australians Bill Mollison and
David Holmgren in the 1970s is mostly known as an organic gardening
and regenerative agriculture philosophy. A central concept in permacul-
ture is that there is no ‘waste’. Any output of any subsystem is the input to
another subsystem – even sewage. Salvage eco-ethics aspires to a condition
in which there is no waste. It is intentionally parsimonious in the sense that
it is always conscious of economising energy and material resources.
Self-evidently, in a sustainable world, making things once, robustly,
with a long life, is the way to go. If they break, they can be repaired. If they
come to the end of their useful life, their parts are salvaged and repur-
posed. An ecological ethics of poverty, of scarcity, is an appropriate
antidote to our naturalisation of the (illusory) overabundance of the fossil-
fuel-driven orgy of commodity economics. Why would anyone straighten
a used nail when nails are so cheap? But nails are not cheap by any real
accounting. Iron ore is mined on one continent with giant trucks and high
explosives, shipped across vast oceans in diesel-powered mega-ships,
smelted and refined and manufactured in factories with giant furnaces and
then shipped again. No, an eco-ethics demands that I repair and reuse that
nail if I can.

10.3.2 An Autobiographical Intermezzo

When I was about 12 years old, I had an epiphany regarding the finiteness
of the planet: you can’t throw anything away! You just put things
somewhere else. This must have been about 1967, before the iconic
252 Simon Penny

‘blue marble’ images of our planet from space that jump-started the envi-
ronmental movement and adorned the cover of the Whole Earth Catalog.
Who knows how this idea popped into the psyche of a boy in suburban
Sydney in the mid-1960s? Did some person, some book, put it in my head?
I don’t think so – more likely a Daffy Duck cartoon. A few years later, as a
teenager, I would wander the rocks of the estuary near where I lived and
find tangles of fishing line, which I would take home and painstakingly
untangle. I never used the line, I wasn’t that interested in fishing. On a
backpacking trip as a young man, I spent several days in the yard of the
hostel I was staying in, tidying accumulated detritus in a large garden into
organised categories – rocks, dead leaves, branches, bits of metal and junk.
I was driven by a deep, anti-entropic desire to reorder. I’m still not sure
why I did it. I thought myself an artist at the time – I thought it was a kind
of artwork.
At art school, I gravitated towards sculpture – there was a workshop
and tools. I had the great good fortune to become a student of one Owen
Broughton: sculptor, foundrymaster, anthropologist and tool collector. The
profundity of what he taught me has only grown with time. He had a
seemingly infinite knowledge of tools, artisanal practices and manufacture.
He had been Henry Moore’s foundry foreman, responsible for successfully
pouring those elephantine bronze castings. He could tell you how in India
they quarry granite with driftwood, how the Romans cast the bronze soles
of centurion’s sandals, how Australian colonialists surveyed land on horse-
back. He had a collection of barbed wire, meticulously set in hardboard
pages of an album 3 feet high. If you bothered to ask, he could tell you
how, when and where each sample was made and on what kinds of ma-
chines. I learned toolmaking at the forge from him. We annealed old files
and reworked them into chisels and drawknives – a rusty file is still a useful
billet of carbon tool steel, after all.
Throughout my career as a sculptor, I consistently upcycled materials,
decades before the term was coined. I scavenged scrap metal yards and
came away on my bicycle with useful materials and treasures – memorably,
on one occasion, towing a 20’-long bundle of aluminium tubing along
heavily trafficked roads: a bicycle semitrailer. I raided county dumps in the
hills for tree trimmings that would be fashioned into finely articulated
structures. I made paper out of dryer lint and dust vacuumed from the
floor – the very definition of waste. As a child, in somewhat impoverished
conditions, I cobbled together rudimentary vehicles from bits of wood and
tin can and wheels scavenged from old prams found in dumps, then ca-
reened down dangerous slopes, narrowly avoiding culverts, cars and seri-
ous injury (little boys can get themselves into no end of trouble). Have
I embroidered an ecological and theoretical justification around the psy-
chic effects of that impoverishment? – quite possibly. Be that as it may,
 An Eco-Ethics for the End of the Anthropocene 253

as long as I can recall, the work of repair has been present in my psyche as
a responsibility. The discarding of almost anything is difficult because ev-
erything, presumably, has a use.
I live in an increasingly desertified country, but I have less guilt about
water use because I know the water I bathe in positively contributes to the
plants I grow and what doesn’t percolates down into the depleted aquifer.
My bath drains into the garden, where the trees enjoy the phosphates in
the organic soaps I use. The trees, while photosynthesising, provide shade,
keeping the soil cooler and reducing surface evaporation. They provide
fruits containing some of that water, transformed into nutritious juices,
which I then consume and contribute my urine back to the soil (urine, un-
surprisingly, is an excellent 1,1,1 fertiliser). In this little universe, every-
thing has value, is used and is transformed. The yard of my current house
was arid when I found it. Now it has inches of mulch and a thriving soil
microbiome – a bustling ecosystem of bugs and worms and spiders. Worms
are turning up in places they never were before, and birds are coming after
them. Organic matter has never left the property – even the coffee-grounds
and used paper towels are not ‘trash’ – but the makings of mulch. Lizards
and hummingbirds snap at the fruitflies rising off the compost heap. It is a
humble triumph. Every week my neighbours line their curbs with enor-
mous trash and recycling rolling bins, filled, often with lids open, over-
filled. Offerings in the cult of the waste stream. Our bins are always almost
empty (we are compelled to discard some things – mostly toxic or plastic
things that have no use to any organism).

10.3.3 Shoes and Socks

Now I worry about old socks, worn at the heel or toe. Socks made from
sophisticated polymers derived from the residues of fossil forests, dispers-
ing festoons of toxic microfibres. What am I going to do with them? I do
not want to throw them away to add to the 95% of all plastic ever manu-
factured, now discarded, dispersed in the waste stream that now encom-
passes the entire planet. With the best will in the world, it may take months
before I darn them. So they sit, not usable but not waste, taunting me with
their poisonous existence. They can’t be recycled, at least not without the
consumption of more energy (probably more fossil fuels) and not without
the addition of more toxic chemicals. I can’t burn them – God forbid! They
won’t rot in the compost heap – no known creature will feed on them.
They are microscopically immortal. Their parts will divide to invisibility,
but they will continue to invade bodies and wreak metabolic havoc. So
must I keep these socks, and plan to one day, darn them? But who darns
socks when you can buy a new pair for a dollar? These damned cheap co-
lourful plastic socks with the fashionable logo – dragged across giant
254 Simon Penny

oceans in giant iron ships burning bunker fuel at hundreds of gallons per
hour, belching toxic fumes. What a bargain! I will wear them around my
neck, like a dead albatross, forever, to remind us of our sin as a culture, of
making things that are foreign to the Earth: a betrayal of the planet and
our responsibility to it.
And these shoes, all these shoes, bonded in industrial autoclaves that
cannot be repaired. I’ve tried, but glue and stitching and rivets just don’t
jive with industrial vulcanisation or plastics that are designed to fall apart.
The sad fact is that, even with my library of tools and encyclopaedic skills,
I cannot fix them, as much as I feel the responsibility to try. I have a pair
of work boots, with good solid soles with plenty of tread left. They have
well-made leather uppers, with good eyelets and durable laces. But be-
tween the insole and the sole is a layer of what was some kind of foam,
that has decomposed into toxic goo. There’s no fixing that. And there is
no obvious pathway to making the boot usable again. Irreparable. An
eco-ethical disaster. What am I to do with them, these lingering spirits of
commodities, ghosts of footwear, casualties of a war of price-cutting and
market share?

10.4 Doing the Eco-ethics of Salvage

In Section 10.4.1, I describe idiosyncratic personal solutions to the ques-
tion of ethically sustainable lifestyle. I set out to be this person. For de-
cades, I have strived to understand the composition and manufacture of
things. The knowledge or ‘literacies’ I draw upon are at least as much
know-how as know-that. As an artist and artisan, a latecomer to academia,
I am increasingly aware that my references and resources are different from
my primarily book-learned colleagues. I draw upon my extensive hands-on
experience in metal fabrication, electronics and electromechanics, boat-
building, home renovation, precision machining, sailing and gardening.
I know what it takes to design and make a machine that works, a structure
that stays up and a hull that withstands often violent natural forces. I know
the qualities of iron and steel, of timbers hard and soft, of rock and con-
crete and of plastics and composite materials. I know how to put them
together, with fasteners and adhesives and molten metals, and how to take
them apart again, with blowtorch, wrench, saw or sledgehammer. I know
the kinds of concentration involved in such work, and the fatigue of the
body, the calluses on the hands, the regular superficial wounds: cuts and
scrapes and bruises, splinters of metal and glass and wood. There is know-
how, and there is also a pragmatic, artisanal know-that – significantly un-
like scholarly know-that.
 An Eco-Ethics for the End of the Anthropocene 255

10.4.1 Life in the Workshop – Repair, Bricolage and Improvisation

Pursuing salvage ethics implies skill and experience with (and access to)
tools and materials – a kind of intimacy. It also calls for a kind of patient
humility. Such a path is not for everyone – some are just disinclined, some
are intimidated and some resist the accumulation of the necessary stuff. It
has its psychic rewards, not simply assuaging guilt, but, on good days, a
successful experience confers a sense of creative competence.
For better or worse, I see value in material detritus that most people
would toss without a thought. I straighten used nails I pull from wood and
I’ll recut a thread on a burred bolt. The broken drill bit – a precision hard-
ened shaft of a specific diameter. The usefulness of a used paint can or milk
carton does not escape me nor do scraps of wood, down to tiny pieces –
wood scraps are always useful. Ditto string and cordage and rope: when
you need a small piece, why cut a big piece? Even if an appliance is trash,
there are salvageable cables and fasteners and hose clamps and a switch to
scavenge. In my line of work, there’s no telling what will come in handy –
that scrap of wood may be just the thing to raise up a workpiece to level or
support a piece under the drill or mix the paint or spread the glue. After
which use, it might be even more useful or find another use. And if they’re
no good for that, they’re firewood, sawdust or mulch, of course.
I fix things. I do have a fondness for tools and restoring tools, not in
some fetishistic collector way, but to bring them back to a good usable
state. A friend once picked up a hammer in my shop. Its handle, cracked
near the head, I’d bound with copper sheet and bronze wire. That’s love,
he said. I guess he was right. I do feel a responsibility to make things better.
Especially things that have originally been built with skill and care and
have perhaps, seen hard use or abuse. I do not feel a compulsion to lavish
care on a damaged Ikea flatpack bookshelf. That material cannot be re-
paired. But a chair that has pieces that were identifiably once part of a tree,
perhaps some joinery made with skill and care – that deserves respect.

10.4.2 Taxonomy, Organisation and Sorting in the Shop

Maintaining the material resources for general repair is a not-insubstantial

undertaking. Having accepted the responsibility to maintain and repair,
and to accumulate relevant parts and tools according to salvage ethics,
maintenance of order is implied. There’s little more frustrating than poking
about trying to find that thing, or tool, you know you’ve got, somewhere.
Worse – infuriating – is when you can’t find it and go out and buy one, only
to find it a week or so later. It’s important to know what you’ve got and
what you might need to get more of, so you don’t run out at a crucial
256 Simon Penny

moment in a project. You need to know where it is when you need it. It’s
not useful if you can’t find it or have forgotten you have it. And it’s amazing
how quickly you forget – not that you have it but where it is, so part of the
work is regular reinforcement of memory. This occurs organically in the
regular working and ferreting around: ‘so that’s where I put that collection
of scraps of piano-wire – in the small chain and lightweight cable box!’.
There is always order, and there is always what looks like clutter and
chaos – the things in the process of sorting, the salvaged whatevers: a pair
of hinges, an electric motor, a sheet of glass – a standing reserve of other
people’s trash – have to be stored, in the right place in this demented
Wunderkammer. How well sorted is sorted? – must every screw be identi-
fied, labelled and stored in a knowable place? No, that would be going too
far. But everything in a big pile: that would be useless. Each individual item
requires some consideration – which category is for it? Do the categories
have to be altered? And categories are always mutable. Some systems are
absurd – only a fool would organise their bookshelf according to the co-
lour of the spines.
The contents of the workshop are too contingent and fluid to be exhaus-
tively notated. Like any library or warehouse, the system of organisation
has to have a certain spatial logic to it and a certain granularity. To sort is
to define categories at a pragmatic level of detail. But total organisation is
a Borgesian impossibility because things are like and unlike other things in
infinite ways. The organisation of animals in the Celestial Emporium of
Benevolent Knowledge related by Jorge Luis Borges is a wry epistemologi-
cal joke. It divides all animals into these categories:

Those that belong to the emperor

Embalmed ones
Those that are trained
Sucking pigs
Mermaids (or Sirens)
Fabulous ones
Stray dogs
Those that are included in this classification
Those that tremble as if they were mad
Innumerable ones
Those drawn with a very fine camel hair brush
Et cetera
Those that have just broken the flower vase
Those that, at a distance, resemble flies

This list (so admired by Michel Foucault), appears in Borge’s short essay
on John Wilkins, the Baroque taxonomist. Borges observes ‘there is no
 An Eco-Ethics for the End of the Anthropocene 257

description of the universe that isn’t arbitrary and conjectural for a simple
reason: we don’t know what the universe is’ (1942).

10.4.3 Pathological Maintenance

About a year ago, I went to an estate sale to buy an old machine lathe.
I found the residue of the working life of a man who, for 30 years or more,
had built model boats. Not my thing really, grown men making small-scale
models of boats they might have known or simply fantasised about as
kids. But I could read his backyard shop. In its disarray was the history
of a life dedicated to a craft. Scattered all about, customised tools, scraps of
careful drawings, hand-made cardboard patterns, jars and little drawers of
parts – a chronicle of a life and a working process. Each little thing, down
to the scars on the edge of the workbench, spoke of purposive practice.
I saw the little scraps, the offcuts saved, and (mostly) I knew why they’d
been kept and what they might come in handy for. I felt like a knowledge-
able detective or archaeologist.
I made an offer and bought it all. All that junk. Ostensibly, because
I knew that amongst it all might be parts for the lathe that might be other-
wise unobtainable. I had a complex feeling, a combination of respectful
empathy and a little mania, or at least compulsion. I drove away feeling
simultaneously that I had a treasure trove in the back of the truck; doubt-
ing my own sanity; and dreading the task I had now committed myself to.
Then, for over a year, sporadically, I sorted and categorised broken tools,
outdated technological components and tiny fasteners, that may or may
not have value for me or anyone. The decision-making often outweighs the
value of the object – the work becomes demented and obsessive. I am wast-
ing my time, but I cannot throw it away. I am yoked to this labour, not
Sisyphisean, but obsessive and of dubious value. What is it that against my
better judgement, persuades me that this is good work that needs to be
done? What dues am I paying and to whom? I became, by a combination
of intention, accident and predilection, the custodian, the archivist, of an
eccentric, encyclopaedic collection of diverse fasteners. I’m sure I could
write a long and boring book just describing their variety just as an ento-
mologist would describe beetles – a taxonomy of threads and head shapes,
exotic, special purpose and common. Some people organise dead beetles
and butterflies and some organise postage stamps or coins. I organise nuts
and bolts – ostensibly for utility, not so much in the interests of science or
history.
My fetishism is not that of the ‘collector’, looking for that elusive maga-
zine issue, a ‘mint’ ‘NIB’ action figure or the missing piece to complete the
set of whatever.18 Neither is it the lure of the nostalgic novelty – the old tin
toy or the bakelite comb. To say that my collecting is utilitarian and
258 Simon Penny

motivated by ecological ethics is only half of the story. I have an abiding

interest in the way materials fit together and the way forms can be coordi-
nated in movement: the taxonomy and topologies of mechanism, espe-
cially ingeniously simple ones: did you know you can use a coil spring to
constrain a shaft to rotate in only one direction? This collection is a library
of mechanical ideas.
I once roomed with a friend in his deceased grandfather’s suburban
house. At the bottom of the garden was a shed, untouched more or less,
since the old man demised. The great depression left its mark on that gen-
eration – the shed was cluttered, full of stuff, much of it in labelled reused
wooden cheese boxes: bits of old leather for shoe repair and parts for gar-
den hoses. A dusting of fine dust covered it all. My friend maintained it as
a kind of museum for the old man. At the time, I found it quirky and
anachronistic. Now I have become that man.
John Haugeland once boasted that he had more nuts and bolts than any
other philosopher. The remark endeared him to me, but philosopher
though he was, in the world of nuts and bolts, he was a child paddling in
the shallows. Wordsworth found infinity in a grain of sand. There is an
infinite variety of nuts and bolts: the thread size and type, the length, the
material and the kind of head. Eleanor Rosch’s ‘principles of categorisa-
tion’ apply, but there are always ‘boundary objects’ (1988).19 This organ-
isation is a kind of Goedelian game where contingencies and exceptions
undermine reassuring logical structures. There are threaded shafts that are
not quite bolts but bear a family resemblance. This taxonomical task is
fascinating precisely because it seems to exceed nomenclature. The order
of these things remains resident in the relationships between the things
themselves, and not in some symbolic descriptive order by which they are
described.

10.4.4 The Obsessive Bricoleur

My mania is practical. It begins, usually, with a thought: ‘that could be

useful for…’or ‘with a little work I could fix that’. I look at the lace hooks
on the pair of old hiking books whose soles were somehow defective and
cannot be repaired: I could reuse those lace hooks, they are good lace
hooks. So begins a design and planning process – how to detach them
without breaking them, how to reattach them and what special tools would
be needed – salvage ethics at work. Once this thought process is underway,
the tools and materials are marshalled, first mentally, then physically – the
rasp or the file or the wire brush, the sandpaper or the steel wool or the
abrasive pad, the pry bar, the chisel, several mallets and hammers, and so
it goes. The process depends on a pre-existing, more or less encyclopaedic
 An Eco-Ethics for the End of the Anthropocene 259

array of tools for working different materials, at different scales, from the
size of a coin to the size of a house.
I found myself regretting that I’d thrown away a plastic toilet brush that
must have cost almost nothing new – another nasty, short-lifespan, mass-
produced plastic object made to wear out and be tossed and replaced with
an equally diabolical piece of crap a year or two later. Not SUP in the usual
sense, but in geological time it’s all the same. But my regret was geometri-
cal: the handle was conical, finely tapered from about 3 cm to about 1.5
cm. A very handy object when you need to connect, say, slightly different
diameters of hose. So who would want that? Only someone who does that
sort of thing from time to time, like me (these machine-made commodities
are paradoxical – they are worthless, but exhibit a level of precision that
would have been regarded as miraculous barely a century ago). It is an
open question whether this work is a blessing or a curse. It is handy to have
the thing you need when you need it.
There is something deeply satisfying about making some broken thing –
whatever it is, a sapling with a broken stem, a rusty tool, a worn piece of
clothing – whole again or at least serviceable. There is some self-care in
this, combined with a compulsion about it that at times seems unbalanced.
Oh, if only, sometimes I could just throw something away. How I dream
of being so oblivious that I could simply dump the old printer in the trash.
How I wish I believed that recycling actually worked. But I know too
much. I wear my inability to throw things away like some penitent. It’s not
that I want to save it, but it’s not useless – that t-shirt with a hole or two.
That old sneaker – if it goes into landfill it will be there…forever? How can
I be responsible for that? I only keep things that are useful. But, unfortu-
nately, everything has a use. It’s just a question of whether you have the
knowledge to ‘see’, and if you have the skills, tools, time and inclination.
It is my misfortune to have all five. I was bemoaning my condition recently
and some well-meaning type said – you’re an artist – you could make junk
sculpture. I responded, with wry self-reflection, that this was not possible
because nothing is ‘junk’.

10.4.5 Hoarding, Collecting, Exhibiting and Purging

I knew a guy who could not throw away newspapers. His house was
stacked with piles of broadsheet – I was amazed the floors supported its
weight. Reading Katie Kilroy-Marac on collectors and hoarders, I was
heartened that (contrary to some suspicions) I do not appear to possess
Hoarding Disorder, as clinically defined (2018, pp. 20–38). Hoarders, who
tend to be lower income (at least when depicted on reality TV), are consid-
ered misguided or even ill. Yet, ‘collectors’ who tend to be more wealthy
260 Simon Penny

are tolerated as eccentrics or admired (especially if they are potential phi-

lanthropists). Some kinds of collections – archives museums – are posi-
tioned at the very pinnacle of culture. Bennett (1988) describes an
‘exhibitionary complex’:

…history and natural science museums, dioramas and panoramas, na-

tional and, later, international exhibitions, arcades and department
stores – which served as linked sites for the development and circulation
of new disciplines (history, biology, art history, anthropology) and their
discursive formations (the past, evolution, aesthetics, man) as well as
for the development of new technologies of vision.
(Bennett, 1988, p. 73)

In this juxtaposition of hoarder and collector, a third, possibly equally

pathological category – the purger – goes un(re)marked. While, superfi-
cially, embodying the virtue of tidiness (Marie Kondo style), the purger
presumes the waste stream, takes for granted the huge hidden industry of
disposal, which literally sweeps our trash ‘under the rug’, bulldozing a skin
of soil over vast toxic landfills. Immersed in commodity capitalism as a fish
is in water, we inevitably position ourselves, intentionally or unintention-
ally, in relation to the obscene accumulation of things and stuff that char-
acterises our culture.20 The conditions of the hoarder, the collector and the
purger bear an uncanny but perhaps unsurprising similarity to the array of
eating disorders. The lesson in both cases is the same: too much and too
little are both undesirable. Tread lightly on the Earth and tread lightly on
oneself as well.

10.5 Conclusion
In the foregoing, I have sketched the outlines of an argument for eco-ethics,
not in terms of a philosophical argument, but in relation to the climate
crisis, commodity capitalism and environmental sustainability. I’ve traced
an argument from the global: laying out the realities of the waste stream
and inbuilt obsolescence to the idiosyncratically autobiographical: describ-
ing the way I have come to conduct and position myself with respect to the
social, biological and technological aspects of my world. On a personal
level, my solution is simple – a permaculture eco-ethics: consume as little
as possible (especially of non-biodegradable materials); repair and repur-
pose; be hyper-conscious of energy and resource use and minimise waste: a
mindful salvage-ethics.
It is important to recognise that this is a two-way flow – the personal also
informs the political. My day-to-day grappling with tools and materials
informs and grounds my activism. But we are all (already) compromised,
enmeshed in systems that often have us doing things we would prefer not
 An Eco-Ethics for the End of the Anthropocene 261

to. We ought not to let the perfect be the enemy of the good.21 As Sami
Grover has noted, we are all climate hypocrites, but we must not let that
stymie us (2021). Given that modern economies run on consumer spend-
ing, Anna Lappés remark ‘Every time you spend money you’re casting a
vote for the kind of world you want’ (2003) is a recipe for quietly reconfig-
uring the big picture. But voting with your wallet is not enough. We should
not imagine that ethical personal practices or lifestyles are a substitute for
socio-political action.

Notes
1 The term carbon footprint, ironically, was coined by British Petroleum (BP) in
the early 2000s. It was based on the concept of ‘ecological footprint’ by William
E. Rees and Mathis Wackernagel (University of British Columbia, 1992).
2 Markets, apparently, will cynically work any societal trend for profit. Perhaps,
the most absurd example I’ve seen is a benchtop blender for food waste – shiny
new white plastic with electric motors that seems to confuse simple maceration
with bioactive composting.
3 This essay has been made into a powerful short film by Jordan Brown (2015).
4 I am not qualified to position this text in the context of the academic ethics
discourses, so with apologies, I leave that for my ethicist colleagues.
5 Different estimates suggest that power consumption of the Internet, globally,
exceeds that of all commercial aviation or all automobiles.
6 Of which nearly 5000 are ‘active’ and, as of mid-2023, over half belong to Elon
Musk.
7 We assume there will be no (new) internal combustion cars in 20 years. But
some things are less susceptible to such obsolescence: paper is paper, bricks are
bricks and knives and forks remain knives and forks. In old-world cities, people
commonly live in houses and walk on roads (and send their waste down sew-
ers) built hundreds of years ago.
8 Gordon Moore was Intel’s longest-serving CEO and chairman. In 1965, he
proposed that roughly every two years, the number of transistors on micro-
chips would double and hence computer performance would constantly in-
crease. Now known as Moore’s Law, this held roughly true for 50 years. More
recently, there have been arguments that we are reaching the physical limits of
such miniaturisation.
9 We do not need yet another style of miniature power plug. In 2022, the EU ap-
proved a ‘common charger directive’. By the end of 2024, all mobile phones,
tablets and cameras sold in the EU will have a USB Type-C charging port.
10 On corporate cynicism and advertising, see Oreskes and Conway (2010).
11 A national nonprofit public education organization named Keep America
Beautiful, Inc. (KAB) was formed in 1953 with the mission of ‘engaging indi-
viduals to take greater responsibility for improving their local community envi-
ronments’. KAB’s first PSA focused on litter prevention. It partnered with the
Ad Council in 1960 to produce a campaign focused on the harmful environ-
mental effects of litter and other forms of pollution. Ten years into the KAB-Ad
Council partnership, in 1971, an Italian actor playing the part of a Native
American (who became known as ‘The Crying Indian’) or ‘Iron Eyes Cody’
appeared in an anti-litter commercial. As he looks over a polluted landscape
and sheds a tear, a voice-over says: ‘People start pollution. People can stop it’.
262 Simon Penny

This powerful commercial won many awards, including being named one of
the top 100 advertising campaigns of the twentieth century by Advertising Age.
Its success inspired other environmental messages from other groups as well.
https://www.psaresearch.com/a-brief-history-of-public-service-advertising/
Accessed 20 June 2023. In a curious recent wrinkle to this story, on 23 February
2023, the NCAI announced ‘Today, Keep America Beautiful, Inc. announced
the transfer of ownership of the well-known “Crying Indian” anti-pollution
public service announcement that first aired on Earth Day in 1971, to the
National Congress of American Indians (NCAI) Fund’ https://www.ncai.org/
news/articles/2023/02/23/national-congress-of-american-indians-acquires-
rights-retires-crying-indian. Accessed 29 June 2023. The NCAI refused this
author rights to reproduce any image of the crying Indian for this chapter.
12 In November 2021, Apple reversed its policy that opening an iPhone voided
warranty and is allowing owner repair. But this ‘repair’ is limited to replacing
a defective component.
13 The evaluation of the public good of computing and the Internet (as opposed
to its role as a vehicle of advertising and the manipulation of public opinion)
and the idea of Internet access as a ‘right’ far exceeds the scope of this chapter
and would demand at least a book-length treatment (see, for instance, Broeders
(2016)). Likewise, strategising a greener information sector – assessing energy
use, materials extraction, cost and cleanliness of manufacture, designing for
recycling and waste management – is a vast topic, but see Anatomy of an AI
system (Crawford and Joler 2018).
14 ‘Plastic recycling was estimated to have declined to about 5–6% in 2021, down
from a high of 9.5% in 2014 and 8.7% in 2018. At that time, the United States
exported millions of tons of plastic waste to China and counted it as recycled
even though much of it was burned or dumped’. https://www.greenpeace.org/
usa/news/new-greenpeace-report-plastic-recycling-is-a-dead-end-street-year-
after-year-plastic-recycling-declines-even-as-plastic-waste-increases/ accessed 23
January 2023. See also https://www.greenpeace.org/usa/reports/circular-claims-
fall-flat-again/. Accessed 23 January 2023.
15 See, for instance, https://www.kqed.org/news/11901288/you-cant-recycle-your-
way-out-californias-plastic-problem-and-what-to-do-about-it. Accessed 24
January 2023.
16 A Lehr is an annealing kiln where blown glass objects are very slowly cooled,
over days.
17 This figure had been repeatedly cited in economic commentaries over recent
years. See, for instance, https://www.whitehouse.gov/cea/written-materials/
2023/07/27/the-advance-estimate-of-second-quarter-real-gdp/#:~:text=
Consumer%20spending%2C%20which%20amounts%20to,the%20strong
%20U.S.%20labor%20market. Accessed 23 August 2023, and https://www.
ceicdata.com/en/indicator/united-states/private-consumption--of-nominal-gdp.
Accessed 23 August 2023.
18 NIB: ‘New, In Box’ (ebay jargon).
19 See also Bowker, GC & Star, SL 1999. Sorting things out: classification and its
consequences, MIT Press, Cambridge.
20 I read once that an Amazonian indigenous culture identified only 300 different
kinds of artefacts – baskets, cutting tools, bows and arrows and so on. We have
300 in every room, and thousands more in the garage, the car, the supermarket
and the workplace.
21 An aphorism which is commonly attributed to Voltaire, also Montesquieu.
 An Eco-Ethics for the End of the Anthropocene 263

References
Benjamin, Walter. 1968. On the Concept of History IX. Illuminations. Harcourt
Brace Jovanovich.
Bennett, T. 1988. ‘The exhibitionary complex’, New Formations, no. 4, pp.
73–102.
Borges, J.L. 1942. The Analytical Language of John Wilkins, La Nación, Argentina.
Bowker, G.C. & Star, S.L. 1999. Sorting things out: classification and its conse-
quences, MIT Press, Cambridge.
Broeders, D. 2016. The Public Core of the Internet: An international Agenda for
Internet Governance. Amsterdam University Press.
Brown, Jordan. 2015. Forget Shorter Showers. https://www.filmsforaction.org/
watch/forget-shorter-showers/ accessed 23 January 2023.
Crawford, Kate & Vladan Joler, 2018. Anatomy of an AI system. https://
anatomyof.ai
Duong, T. 2021. ‘Chile’s Atacama desert: where fast fashion goes to die’, Eco
Watch, 15 November, accessed 26 June 2022, https://www.ecowatch.com/chile-
desert-fast-fashion-2655551898.html
Grover, Sami. 2021. We’re all climate hypocrites now. New Society pub.
Henke, Christopher & Benjamin Sims. 2020. Repairing Infrastructures, The
Maintenance of Materiality and Power. MIT Press.
Hoban, R. 1980. Riddley Walker, Jonathan Cape, United Kingdom.
Jensen, Derrick. 2015. Forget Shorter Showers https://orionmagazine.org/article/
forget-shorter-showers/
Kilroy-Marac, K. 2018. ‘An order of distinction (or, how to tell a collection from a
hoard)’, Journal of Material Culture, 23(1), pp. 20–38. https://doi.org/10.
1177/1359183517729428
Lappé, Anna. 2003. O Magazine (The Oprah Magazine) June.
Oreskes, Naomi, Erik Conway. 2010. Merchants of doubt: how a handful of scien-
tists obscured the truth on issues from tobacco smoke to global warming.
New York: Bloomsbury Press.
Rivera, Julio L. & Amrine Lallmahomed 2016. Environmental implications of
planned obsolescence and product lifetime: a literature review, International
Journal of Sustainable Engineering, 9(2), pp. 119–129.
Rosch, E. 1988. ‘Principles of categorization’, in A. Collins & E.E. Smith (eds),
Readings in Cognitive Science, a Perspective from Psychology and Artificial
Intelligence, Morgan Kaufmann Publishers, pp. 312–322.
Slade, Giles. 2007. Made to Break: Technology and Obsolescence in America.
Harvard University Press.
Summers, M. 2019. The shocking true story behind the ‘crying Indian’ commercial,
accessed 26 June 2022, https://www.ranker.com/list/crying-indian-commercial-
true-story/megan-summers
Young, Mark Thomas. 2020. “Maintenance” in Dorn & Michelfelder (eds) The
Routledge Handbook of the Philosophy of Engineering. New York: Routledge
pp. 356–368.
11 Aesthetic Values in the Maintenance
of Urban Technologies
Sanna Lehtinen

11.1 Introduction
Urban technologies and their maintenance provide an interesting and even
somewhat unexpected angle to study the city and its aesthetic dimensions.
This perspective is strongly influenced not only by the way in which we
define technology but also by how the city in question is imagined: whether
it is an actual place or the idea of the city in general. As a new topic for the
philosophy of technology, maintenance opens up new perspectives within
the existing discourse on urban aesthetics by refocusing our attention on
technologies over time. In order to begin bridging the approaches of urban
technological aesthetics and their maintenance, it is important to empha-
sise that urban technologies are understood quite broadly in this context.
This means that the focus is not only on digital technologies or the latest
smart city developments which have been largely driving the latest
technology-inclined studies of the city but also on more rudimentary tech-
nologies. The definition of technology used here follows that utilised in
recent work delineating a specific field of study for the philosophy of urban
technologies (Nagenborg et al. 2021) and thus probes more broadly into
the existing and fast-developing plurality of human–technology relations
in cities. The benefit of this is a broad understanding of urban environ-
ments and especially their aesthetic implications. Further analyses of differ-
ent types of technologies can build upon these insights by providing a more
detailed understanding of the contemporary urban lifeworld.1
It is possible to approach cities themselves as technological artefacts.
This is because they are, to a large extent, produced by human intentional
activity and considered prime examples of complex systems. The focus of
this chapter, however, is on technologies of varying scales which one can
encounter perceptually and experientially in cities either directly or through
their use and impacts on urban surroundings. Most of the time, these types
of technologies have some sort of perceivable presence and their imple-
mentation, use and upkeep have an impact on how the city looks and feels

DOI: 10.4324/9781003316213-13
 Aesthetic Values in the Maintenance of Urban Technologies 265

to its inhabitants and visitors alike. This type of experiential emphasis typi-
cal of philosophical urban aesthetics has strong phenomenological under-
pinnings. It is further brought together with a postphenomenological2
understanding of different types of human–technology relations that char-
acterise urban everyday life. This broader take is necessary in order to
show that the aesthetic dimension of cities affected by urban technologies
is not only a matter of design or a step in the early phases of their realisa-
tion but an iterative, ongoing process and a topic for an open deliberative
debate for which studying the basic tenets of the aesthetic experience of the
city is of crucial importance. In the philosophy of technology, there has
been a disproportionate focus on design which has led to the neglect of
maintenance and its further implications as a lucrative topic for the field.
Aesthetics has for the most part of the twentieth century been strongly
associated with the arts and the philosophy of art, especially in the Anglo-
American context. Topics such as intentionality and interpretation have
been developed in the specific context of western art forms. Although there
is an obvious link to the theme of maintenance through designers’ inten-
tions which already take a stance towards how the designed artefacts or
places are to be maintained, the central insights of contemporary aesthetic
theory have not been applied as broadly outside the sphere of the arts as
they could be. Since the turn of the twenty-first century, the overlooked
questions about how things should and could look like and how their aes-
thetic value is linked to other values and practices have, however, been
proven increasingly relevant as a much broader take on the aesthetic di-
mension of human activities has been developed in the form of urban and
everyday aesthetics. Ecological, environmental and everyday-focused ap-
proaches in philosophical aesthetics have become prevalent and subjected
to systematic analysis since the 1960s. The increasingly complex problems
and grand challenges that humanity is facing have shown the need to un-
derstand human aesthetic values far beyond the limited, albeit fascinating
context of the arts. Motivated by growing environmental concerns, this
broadened academic interest has increasingly been directed to the influence
and aesthetic impact of human activity on various types of environments.
Urban environments have gained more relevance in academic discussions
in the few last decades, when cities started to be perceived not only as the
unavoidable by-products of societal development but also as places of cen-
tral aesthetic relevance for humans. This shift in the mindset is important
for the interest in the aesthetics of urban technologies, as it leads to a more
all-encompassing idea of what constitutes quality of life in urban environ-
ments of various types (Lehtinen 2021a).
In its current orientation, as a relatively new branch of philosophy, ur-
ban aesthetics sheds light on how humans not only perceive and pay atten-
tion to but also further assess and evaluate the sensory and experiential
266 Sanna Lehtinen

features of contemporary cities globally. In traditional aesthetic theory,

aesthetic judgements are linked to aesthetic values that have usually strong
collective and societal importance and that come up in public discussion
on how cities normatively should look and feel like. The focus of inquiry
can be on either a more distanced, conceptualised perspective of contem-
porary cities or the so-called inhabitant’s perspective, which emphasises
lived everyday experience. Whichever perspective is taken as the main
starting point, it is obvious that the realm of technology and the extent to
which it becomes a quintessential part of the specifically urban experience,
has not been traditionally at the centre of attention in philosophical aes-
thetics. (Lehtinen 2021a) In the context of contemporary cities globally, it
is increasingly important to address this omission as we are faced with
questions concerning 1) how urban technologies are impacting the overall
look and feel of the city, 2) how different types of technologies make dif-
ferent values in the city perceivable and 3) how to understand changes in
the technological realm, that is, what actually changes in the urban life-
world when new technologies enter the picture?
This chapter has been developed around two central arguments. The
first one is that the maintenance of urban technologies is an aesthetic issue
in the first place, not just a purely practical or functional issue. The second
argument goes further to claim that the maintenance of urban technologies
is of key importance in showing care for future generations. This goes to
show how flexible aesthetic values can and could be when societal priori-
ties change. The first argument is made in order to broaden the disciplinary
landscape of the emerging intersection of urban aesthetics, the philosophy
of the city and the philosophy of maintenance as part of the philosophy of
technology. The second argument is developed further to show how the
care and aesthetic attention with which humans tend to urban technolo-
gies, can provide insight into how we may care for and maintain the planet
in a less destructive way for various human-affected ecosystems. The end
goal is to show that care in itself does not compromise aesthetic values: on
the contrary, it is crucial to ask what type of values and ideals are the main
drivers behind current maintenance practices and in what ways they may
need to be rethought and adjusted in order to achieve a future we desire.

11.2 Traces of the Past, Traces of the Future

Urban technologies can be defined in various overlapping or even conflict-
ing ways. For the purpose of looking into their aesthetic character and ca-
pacities, they are sometimes part of a larger architectural ensemble or at
least implemented in a way that is indiscernible from buildings themselves
as objects. They can be built in from the beginning, as in the case of smart
 Aesthetic Values in the Maintenance of Urban Technologies 267

architecture, or implemented later on. The latter strategy is referred to as

retrofit practices, in which compromises in design solutions need to be
made in order to fit new technologies into the already existing structures of
the built environment. Oftentimes technologies are also distinctly separate
and physically clearly detached from buildings, sometimes in order to ease
the implementation of future upgrades. This implies already a temporal
disconnect between the basic building stock of the city (which is ap-
proached in aesthetic terms commonly as ‘architecture’) and the more
anonymous technologies that are either hidden or just bluntly attached for
the time being to the facades or other surfaces of buildings. Often urban
technologies are implemented in ways which suggest they are regarded as
uninteresting or as having low aesthetic value. The assumed anonymity
and replaceability of technology, for example, culminates in the way more
recent digital technologies are designed into the processes of the city. The
implementation of 5G antennas and small cell towers into urban neigh-
bourhoods is an example of this, showing a whole breadth of different
strategies of aesthetically ‘dealing’ with them, ranging from camouflage to
full-on takeover of public space both visually as well as spatially (Lehtinen &
Fantini van Ditmar 2022).
What makes digging into the implicit aesthetic assumptions regarding
‘urban technologies’ difficult is the common, shared understandings of the
definition and range of technology needs to be always specified in relation
to the particular task in question. Not all specifically urban forms of tech-
nology are accessible to everyone and often their functioning is not under-
stood or consciously paid attention to when engaged in activities they
enable. There is sometimes significant overlap with different categories of
urban technologies which makes analysing them difficult. Recent philo-
sophical accounts of urban technologies take into consideration their
broad and varied nature and their impact on urban life. Epting (2021), for
example, provides a taxonomy of urban technologies that is based on their
role in facilitating sustainability transitions and securing a livable environ-
ment for citizens. According to Epting, urban technologies can be thus ei-
ther ‘wicked’ or ‘saving’, depending on what their main impacts on cities
are and whether they are ethically justifiable. Nagenborg et al. (2021), on
the other hand, offer a view of new and emerging urban technologies that
focuses less on judging their moral character and more on how they func-
tion and are used to mediate the relationship between citizens and their
environment. This mediation can be seamless and unperceivable or more
articulated and impactful in shaping our experience of the urban environ-
ment. It is advisable to keep both ways of defining technologies in mind,
although the human-centric mediation role tends to be more central for the
aesthetic repercussions discussed further in this chapter.
268 Sanna Lehtinen

To investigate the aesthetic dimension of urban technologies and their

maintenance in particular, one can start with the existing elements of the
built environment. On the level of individual buildings, the scale in a city
varies from the smallest micro houses to high-rise tower blocks and sky-
scrapers, which could not be built without the very existence of multiple
technologies and innovations, for example, those enabling mechanical ver-
tical transportation in the form of the elevators or escalators (Graham
2018). Contemporary buildings, even in their most simple forms are highly
technologised artefacts produced by human intentional design to comple-
ment the current form and ideas of what a city is or should be. Buildings
as advanced technological objects are thus not detached from urban imagi-
naries and the aesthetic forms they often take. It is tempting to speculate
that these imaginaries have a far longer lifespan than what concrete urban
plans and designs specify which is to a great extent reliant on technological
possibilities, whether presently at hand or imagined towards the future.
The construction phase of an entirely new building attaches certain
meanings to the urban environment which becomes always to some extent
rearranged by the new building and its subsequent uses. How built places
are then further maintained and cared for, gives some insight into the val-
ues of the community that uses and maintains them. Concrete maintenance
practices can be telling of how a community relates itself to their techno-
logical surroundings in cities. Showing care to buildings, either in the form
of everyday maintenance practices such as cleaning and painting or through
symbolic acts through more speculative or artistic means, is a very concrete
way of making attachment and ownership of a place tangible (Lehtinen
2020). To what extent manifesting this care extends to the details and var-
ied examples of the technological realm, has been less easy to trace and
depends again on what types of technologies are in the focus. Tending and
caring for the technological realm and specifically their aesthetic range,
however, represents a potentially significant shift in thinking. This can
have repercussions in moving towards more sustainable technological
practices, in which continuous upgrading is not a goal in itself.
Even in cities, all technologies do not work in direct contact with archi-
tectural constructs: technologies are to be found also in a more indepen-
dent form in urban environments: we find examples of this on the streets
and in the so-called in-between spaces in cities. Infrastructural technolo-
gies are often hidden, either completely or partially and this affects to
which extent they draw attention or stimulate imagination through their
aesthetic qualities. Technological constructs of the particularly detached
type, such as electrical grids or telecommunications-related constructs, of-
ten take up space from non-human species when they are installed in places
in which animals could potentially find shelter or in which plant ecologies
 Aesthetic Values in the Maintenance of Urban Technologies 269

are affected by human intrusion. Technologies within urban ecology can

also offer refuge to animals, although this is most often an unintentional
consequence. Bees or birds can build their nests under the coverings used
to shield technologies against the elements or spiders weave their webs in
places where electricity produces extra warmth emanating to the surface of
structures. Technology thus sometimes provides habitats to animals in ur-
ban environments, but most often technology and natural ecology in cities
are kept intentionally apart. The presence of animals and vegetation is of-
ten seen as an uncontrollable element in the maintenance of technologies,
and the long-term impacts of urban technologies on non-human ecologies
are not fully known either (Hernan & Ramirez-Figueroa 2022). There is a
deep ontological divide in thinking about cities in these terms: the natural
and the human originating are easier to manage when kept apart whenever
possible. This goes to show, that human decisions on maintenance affect
other-than-human ecologies even in cases where there is an especially
strong intention to keep them separate on purpose.
Despite common tendencies to hide urban technologies from view, many
moments of everyday life in cities nonetheless involve a recognition of ur-
ban technologies in use. Not all technologies are perceptually discernible
as the aforementioned example 4G and 5G networks show but the in-
stalled structures and parts still have a noticeable presence in the urban
environment. This perceptual dimension requires attention, recognition
and interpretation in order to be correctly linked to particular technology
‘in action’. In the fourth wave of industrialisation and through the expan-
sive digitalisation of cities, the act of linking the perceivable structure of
urban technologies and their ‘hidden’ and often converged uses is becom-
ing exceedingly difficult and often requires a high degree of technological
expertise. This is especially pertinent in cases of breakdown which reveal,
not only the vulnerability but also the opaqueness of increasingly complex
technological systems (Graham & Thrift 2007).
The selected examples from the technological realm have had a clear
point of implementation to use and most likely will have also an expiration
date in the urban sphere. These types of perceptually discernible technolo-
gies can be found on many scales of the city, ranging from traces of indus-
trial founding phases of urban expansion to communication technologies
that have had significant shifts throughout the twentieth and twenty-first
centuries alike. It is also important to consider the future impact of the
implementation of such technologies. This goes to underline that the aes-
thetic impacts of maintenance decisions do not concern only how the ap-
pearance of technologies is kept as they have been designed in the first
place but in thinking and rethinking how they could and should look in the
future.
270 Sanna Lehtinen

11.3 Aesthetics of Care in Maintenance Practices

Maintenance, repair and upkeep are areas of human activity in which hu-
man values become explicit and gain a perceivable, even tangible, dimen-
sion of some type. This dimension is temporally important as well, as
maintenance is not only an act but a continuous cycle with consequences
that reach far in time: it is tempting to state that, to a certain degree, main-
tenance gives perceivable presence and temporal longevity to human val-
ues. Maintenance practices mean taking an active stance towards keeping
something – a more or less networked artefact of some scale – in good
material and functional condition, looking and working like it is generally
expected to. Sometimes maintenance might even take a form of ‘curated
decay’, in which the ultimate end goal is to align maintenance into collabo-
rating with ecological processes rather than working against them as the
logic of maintenance usually dictates (DeSilvey, 2017). In societies under
chaotic conditions, this might be a prolonged reality, when resources are
directed only to the bare essentials. I would argue, that even then the aes-
thetic value of urban technologies does not vanish but takes on another
meaning for the citizens or from the perspective of maintenance decisions.
The ingenuity with which citizens maintain living quarters in conditions of
scarce resources can develop into original forms of creativity. This is visi-
ble, for example, in the aesthetically rich favela culture of Rio de Janeiro in
which the creativity of the inhabitants often stems from a severe lack of
resources. The solutions employed for extending buildings, for example, or
accommodating for a lack of infrastructure, produce aesthetically distinct
forms and a layered overall look for entire neighbourhoods.
To what extent different maintenance practices take the aesthetic dimen-
sion into account varies according to the maintained object, artefact or
process. Depending on the extent the technology in question is considered
already an integral part of the make-up of the building, block or neigh-
bourhood, its aesthetic value might not be contested. Very visible (and
audible) historical mobility technologies are a good example, such as the
street elevators in some Lisbon bairros or trams in many other historic
downtown locations. Despite their very noticeable presence (or precisely
even because of it), old, creaky and rattling tram cars have become em-
blematic of the city and soundscapes of particular corners in cities such as
Lisbon. This type of established and culturally recognised aesthetic value is
also tied to the heritage value of mobility technologies, which captures the
complex interplay between current and past values implicit in human–
technology interactions in cities.
Recognised cultural heritage value leads to certain expectations towards
the object’s aesthetic value that characterise some large-scale infrastruc-
tural technologies in cities. Bridges, for example, are often considered a
 Aesthetic Values in the Maintenance of Urban Technologies 271

part of the rich, vibrant heritage of cities and thus cultural as well as aes-
thetic values are more obviously a part of the discussion on their mainte-
nance. It is relevant to point out that even cultural heritage is not enough
to pinpoint the aesthetic values attached to human artefacts, as the aes-
thetic dimension implies the possibility of a more deeply personal take on
how a particular artefact is perceived and how it contributes to its sur-
roundings and the ever-changing lifeworld in flux (Giombini 2020).
Aesthetic values can be a cause of extreme disagreement between commu-
nity members, whereas heritage value tends to be something that is gener-
ally agreed based on shared historical and cultural perspectives and thus
can often function as a foundation for long-term maintenance plans. The
relative flexibility of aesthetic value depending on variations in personal
experiences, interpretations and nuances makes it more difficult to grasp.
It is fair to assume, that this is one of the reasons why the idea of aesthetic
value has been difficult to grasp in negotiations over the maintenance of
such urban elements as technologies.
Recent years have witnessed an increasing interest in ‘aesthetics of care’
in philosophical and applied aesthetics, especially in relation to environ-
mental and everyday aesthetics. One of the proponents of this new
approach, Yuriko Saito, has proposed bridging Western philosophical aes-
thetics and Japanese aesthetics for a more comprehensive take on aesthetic
sensibility. In her recent work, Saito links aesthetic interest and values with
contemporary perspectives in care ethics (2020, 2022). The strong feminist
underpinnings of the care ethical tradition direct attention to the underap-
preciated work and practices that enable the functioning of a society which
are often performed by underprivileged social groups. This perspective is
useful for discerning dimensions of aesthetics that have previously been left
in the dark due to the common tendency to valorise innovation or the de-
sign phase in many cultures (Young 2020).
Care can be understood in this context at least from two different yet
often complementary perspectives. It can refer to 1) care towards material
condition in the form of ‘attentive care’ that is shown through mainte-
nance and repair practices or 2) care towards future generations, whose
living conditions are altered and maintained through decisions made today
on the basis of current knowledge. These two modes of care are closely
connected in most cases. Some important nuances can, however, be distin-
guished in concrete maintenance practices with the help of this distinction.
One way of determining which mode of care is more central is by consider-
ing the relationship with environmental harm and overall sustainability in
particular maintenance decisions. If certain maintenance decisions and
practices are likely to harm the living conditions of future generations of
humans and non-humans alike, they cannot be considered to show care
towards future generations. An example of this is the use and replacement
272 Sanna Lehtinen

of carbon-intensive materials, such as concrete or glass, in building main-

tenance, which are now understood to contribute to a warming climate
which threatens the livelihood of future generations. The realisation of this
type of temporal influence has recently led to a significant cultural shift
which is still ongoing in many of the fields in which long-lasting decisions
over maintenance are made. The increased recognition of our obligations
and influence on the living conditions of future generations has encouraged
a re-evaluation of current maintenance practices. Followed as a strict prin-
ciple in various maintenance practices, many of the previously made deci-
sions would not be feasible anymore.
The perceivability of care itself as a practice, and how it becomes visible
and sensed is another important perspective to understanding maintenance
practices and decisions from the care ethical and aesthetic perspective. It
has thus been easier to focus on the first mode of care as understanding the
material conditions and ramifications of maintenance practices is perhaps
the clearest point of entry to upkeeping elements of the urban sphere. It
would be, however, ultimately important to link these two instances of
care: focusing first on what is possible with what exists already and then
understanding what the further consequences of these actions will be in
times and locations far beyond one’s immediate reach. Articulating more
clearly, this relation between the different dimensions of care will help in
re-assessing further some of the maintenance choices that have previously
been considered obvious and uncompromised. Through the lens of inter-
generational aesthetics, in particular, a significantly new perspective can be
gained on debated issues such as the renewal of parts of nationally signifi-
cant built heritage (Capdevila-Werning & Lehtinen 2021).
With the intergenerational focus in mind, the goal of maintenance, re-
pair and care can sometimes be to make material conditions to some extent
better than before in clearly acknowledgeable ways. Perhaps faults have
been identified in the original design or new ways have been found to im-
prove on existing structures or systems. Instead of merely aiming at main-
taining an existing state, this kind of incremental amelioration also often
changes the appearance of the maintained artefact. An example of this
could be changing the façade material of a building into something entirely
new, after discovering that the original material has not proved durable or
sustainable enough for the long-term use of the building (Capdevila-
Werning & Lehtinen 2021). In such cases, even a significant change can be
justified if it enables a building to avoid demolition and replacement.
However, if the change in appearance is significant, it can often be impor-
tant to communicate to stakeholders the reasons motivating the change in
appearance or the overall aesthetic status quo. Sometimes even the slight-
est changes in well-known or otherwise familiar elements of the city can
produce discontent among the urban dwellers, especially in cases where
 Aesthetic Values in the Maintenance of Urban Technologies 273

the decision-making processes have not been transparent enough. As

pointed out by Mark Thomas Young in his chapter in this volume, the
Golden Gate Bridge is a good example of this – the debate about whether
to install suicide barriers often centred on the aesthetics of the bridge.
Many worried that the addition of nets or raised guard railings would ruin
the iconic look of the bridge, considered to be emblematic and beautiful in
its unchanged appearance throughout the decades.
As concerns over the ecological and social sustainabilities of many hu-
man practices have grown, understanding the deeper consequences and
implications of aesthetic choices in design have become increasingly im-
portant. This in turn has led many to scrutinise the aesthetic dimension of
decision-making implicit in maintenance and repair decisions (see e.g.
Young 2021). Aesthetics opens a way to assess not only the chosen meth-
ods and goals but also the deeper layers of values that have directed the
plans and designs. The aesthetic layer of human world-making does not
refer merely to surface qualities that are to be taken into consideration as
the last part of the design process. Conceptualising the care aesthetical
perspective and implementing it in the decision-making processes would be
taking a step further in acknowledging the aesthetic importance of the
various elements that create the urban lifeworld.

11.4 Deliberation of Aesthetic Values

It is safe to say that aesthetic choices of some calibre are made always when
maintenance decisions are made. Often, they are by-products of some
other choices but nonetheless, repercussions to the aesthetic realm will take
place when decisions are put into effect. Further on, there are many de-
tailed moments in the process of maintenance, which will result in some
type of aesthetically relevant outcome. While these outcomes are often nu-
anced they can be more significant than we may often think. With repaint-
ing a house, for example, nuances in colour choice might seem insignificant
but the change in perception can nonetheless be very significant when the
newly repainted house is contrasted with the colours of the adjacent build-
ings and the details of the streetscape. When it comes to architectural pres-
ervation and restoration, there are rules and guidelines as the preservation
and maintenance of architectural heritage is already an established field
in itself. The consideration of urban technologies, however, has a much
broader scope as it concerns a wider spectrum of different types of tech-
nologies than just those directly linked to architectural solutions. However,
much like architecture, many of these technologies also have a significant
effect on how the urban environment looks and feels to its various users.
Yet, while any changes to the appearance of a city risk disrupting its
aesthetic character, preserving cities as museums of outdated technologies
274 Sanna Lehtinen

is not a feasible goal either. So how should we keep track of and value past
uses of the city in a way that does not hinder and even supports the current
needs of the citizens? One way of approaching this question appeals to
conceptions of ‘authenticity’ which have recently been utilised to analyse
the way cities are perceived and characterised (Wittingslow 2021). In the
context of philosophical aesthetics, however, this concept can be risky to
use, as it implies the existence of some original, true-to-character form and
identity of the city. Yet, many cities seem to genuinely have centuries-long
histories and traditions that constitute a uniqueness to their character.
Implementing new technologies into these kinds of environments can be
especially difficult, as the layers of history often quite literally do not have
space left for new layers to emerge. Consider, for example, the mobility
problems currently faced by many historical European cities, in which new
metro lines are nearly impossible to build because of the constant discov-
ery of archaeological findings in the excavation sites of planned routes.
It is important to remember in this context that the aesthetic layer of cit-
ies should not be considered only in the broad sense but from the so-called
macro perspective with a strong visual orientation (Lehtinen 2021a). When
decisions are made, the city or a part of it is often approached in imagina-
tion, based on impressions and ideas about the place and its aesthetic qual-
ities. Even with detailed visualisations and plans, the so-called micro
aesthetic layer is easily overlooked in these deliberations. Understanding
how the aesthetic character of a place opens up through multisensory,
street-level experience, in the midst of the hustle and bustle of everyday
urban life, is of crucial importance when making aesthetically informed
maintenance decisions. This is, however, also very difficult in practice
because such processes often unfold on different timescales. In planning
or architectural design processes, participatory and citizen-oriented ap-
proaches have become more prevalent in recent years, but the described
tension between the macro and the micro layers of aesthetics cannot be
resolved only by making the inhabitants participate more in the processes.
One solution to alleviate the friction in favour of more quotidian experi-
ences is to expand the temporal scale of the discussions and to take them
into the locations themselves, during many different kinds of conditions
(seasons, hours of the day and so on).
Approaching urban technologies as processes that require ongoing care
and value deliberation to maintain, changes the perspective slightly and
enables us to see these technologies beyond their most immediate use value.
Another aesthetic dimension important to the continuous cycles of upkeep
comes from the creativity that is required to tackle the challenges that time
itself presents in the process. This applies both to the material world and
the social sphere of the use of urban technologies. Materials decay and
age in unexpected ways, especially newer plastic-based and composite
 Aesthetic Values in the Maintenance of Urban Technologies 275

materials that do not endure changing conditions, which complicates the

task of maintenance. The aesthetic appreciations of individual people often
change during their lifetimes and even societal aesthetic norms are prone
to evolve. Because it is considered fickle and prone to change, aesthetics
has often been considered a difficult area to tackle as a value domain. Such
perceptions arise at least in part through the influence of the long-standing
idea of the autonomy of aesthetic value, based on the Kantian notion of
aesthetic appreciation as disinterested. However, recent work on environ-
mental and everyday aesthetics has often challenged the notion of disinter-
estedness and tends to put forward an idea of aesthetic interest and
appreciation as deeply intertwined with ethical and social values (Lehtinen
2021b). At the very least, it should be acknowledged that the level of dis-
interestedness differs when considering the aesthetic value of something
that we use and interact with on a daily basis. In comparison to artworks
which form the traditional focus in the study of aesthetic judgement, aes-
thetic values in the realm of the everyday are part of a value network in
which different values are interlinked in more complex ways than in the
context of art.
Something that once was fashionable might once again come into vogue.
Such fluctuations in aesthetic trends might be a tempting cue for determin-
ing long-term aesthetic value. It is at least useful to postpone aesthetic
judgement until a fuller picture of the significance of the artefact in ques-
tion has been gained. Large-scale infrastructural technologies provide in-
teresting examples of this. In lower Manhattan, New York City, dilapidated
railroad tracks were turned into a hugely popular urban park, the High
Line. In Paris, sections of the La Petite Ceinture, an inner city circular rail-
way, were also rehabilitated and transformed into recreational areas. The
development of both of these sites stems primarily from a commitment to
saving the main traces of a city’s past. Yet in addition to forming parts of
notable post-industrial landscapes, the unused railroads also exhibited un-
usual biodiversity. Demolishing these structures would have therefore
meant not only erasing parts of the historical layers of each city but also
erasing places that now support urban nature and function quite literally
as breathing spaces within the city.
These examples show in significant ways how infrastructural technolo-
gies, as examples of urban technologies more broadly, are not discernible
from the spaces and environments in which they have been built. They
merge into new kinds of places that evolve towards new meanings as their
past uses become largely forgotten and obscured. In the case of the High
Line, such processes are employed in the design of the new park as the old
rail tracks are left visible and untouched, sometimes rendered into newer,
stylised design elements of the trail. Deliberation of aesthetic values within
the community takes place on multiple levels simultaneously and traces of
276 Sanna Lehtinen

past technologies act as cues for these deliberations. However, whereas the
aesthetic dimension can be directly addressed, often it is merely implied by
discussions on other topics. Similarly, decisions about aesthetics are often
masked as something else. This is recognised at least in the sphere of archi-
tecture where the buildings that are considered ugly are more likely to be
demolished than repaired. The decisions have aesthetic repercussions, so
the supported and maintained state of urban technologies manages to
make other central values perceivable.

11.5 Conclusions
When discussing the range and influence of urban technologies in relation
to the use and overall look and feel of cities, aesthetic values are still an
understudied element. This chapter has aimed to demonstrate how aes-
thetic value intersects importantly with other value considerations, having,
for example, further ecological and social consequences through the ethi-
cal values that the appearance of urban places is interpreted to represent.
The aesthetics of cities or elements of the built environment are not to be
treated separately from other values but are deeply tied to how these ele-
ments are valued and what type of decisions are made over their fate.
Technological artefacts, networks or their perceivable parts should not be
considered an exception to this. On the contrary, the maintenance of vari-
ous types of urban technologies can be seen as indicative of aesthetic values
more broadly extending to how well the society in question adapts to
changing conditions for the human lifeform. It can serve even as a testing
ground for increasing the flexibility of societally acceptable values.
Taking into consideration the ongoing shifts and tensions in aesthetic
values would help to develop the burgeoning field of the philosophy of
maintenance further in order to better address the complex reality of the
contemporary world. Especially, in relation to urban environments and
their development, aesthetics is rarely only about aesthetic values but also
intersect with social and ethical values. What there is to be sensed, per-
ceived and experienced in the first place is indicative of societal values
more broadly. This extends also to what is missing from the cityscape, as
in the example of the many technologies which are hidden or simply erased
when they become obsolete for most city dwellers. This implies that look-
ing at cities over time might reveal new dimensions to established problems
and issues in aesthetics.
While the ruins of the future might not consist solely of technological
constructs even in the most highly technologised imagination, the appear-
ance and perceptual qualities of human-designed technological artefacts
form a great deal of the legacy that we leave for future generations. These
should not be approached merely as traces of the past from a heritage
 Aesthetic Values in the Maintenance of Urban Technologies 277

perspective or as updatable elements from a technical perspective but also

through their impact on the overall aesthetics of cities. The aesthetic per-
spective undermines the continuity of human–technology relations over
time and also facilitates experimentation with contested aesthetic values
and new possibilities. Understanding what is aesthetically pleasing, nega-
tive, interesting or even merely tolerable could have further consequences
on how to appreciate cities as complex systems, in which there is ‘some-
thing for every taste’ and which conform and adapt to the increasing re-
quirements for sustainability transformations.

Notes
1 A concept derived from Husserlian phenomenology, for the use in the specifi-
cally urban sense, see Madsen and Plunz (2002).
2 Post-phenomenological referring here to the views in philosophy of technology
based on Don Ihde’s work, see, for example, Rosenberger and Verbeek (2015).

References
Capdevila-Werning, Remei & Sanna Lehtinen (2021). “Intergenerational Aesthetics:
A Future-Oriented Approach to Aesthetic Theory and Practice”, Philosophical
Inquiries, 9(2), pp. 175–198.
DeSilvey, Caitlin (2017). Curated Decay: Heritage beyond Saving. Minnesota
University Press.
Epting, Shane (2021). Saving Cities: A Taxonomy of Urban Technologies. Cham:
Springer.
Giombini, Lisa (2020). “Reconstructing Heritage: Places, Values, Attachment”,
Contemporary Aesthetics, vol. 18 article 9. Available at: https://digitalcommons.
risd.edu//liberalarts_contempaesthetics/vol18/iss1/9
Graham, Stephen (2018). Vertical. The City from Satellites to Bunkers. London:
Verso.
Graham, Stephen & Nigel Thrift (2007). “Out of Order: Understanding Maintenance
and Repair”, Theory, Culture and Society, 24(3): 1–25.
Hernan, Luis & Carolina Ramirez-Figueroa (2022). “A Luminous Abstraction: 5G
Narratives and Futuring Otherwise”, Mediapolis: Journal of Cities and Culture,
Dossiers no. 3, vol. 7. https://www.mediapolisjournal.com/2022/09/luminous-
abstraction
Lehtinen, Sanna & Fantini van Ditmar, Delfina (2022). “Introduction: The Rise of
5G”, Mediapolis: Journal of Cities and Culture, Dossiers no. 3, vol. 7. https://
www.mediapolisjournal.com/2022/09/rise-of-5g
Lehtinen, Sanna (2021a). “Urban Aesthetics and Technology”, in Oxford
Handbook of Philosophy of Technology, ed. S. Vallor. Oxford: Oxford University
Press, pp. 449–469.
Lehtinen, Sanna (2021b). “Aesthetic Sustainability”, Situating Sustainability: A
Handbook of Contexts and Concepts, eds. R. Toivanen & P. Krieg, Helsinki:
Helsinki University Press, pp. 255–268. https://doi.org/10.33134/HUP-14-18
278 Sanna Lehtinen

Lehtinen, Sanna (2020). “Buildings as Objects of Care in the Urban Environment”,

in Aesthetics in Dialogue: Applying Philosophy of Art in a Global World, eds.
Z. Somhegyi & M. Ryynänen, Berlin: Peter Lang, pp. 223–236.
Madsen, Peter and Richard Plunz (2002). The Urban Lifeworld: Formation
Perception Representation. New York & London: Routledge.
Nagenborg, Michael, Taylor Stone, Margoth González Woge and Pieter Vermaas,
eds. (2021). Technology and the City: Towards a Philosophy of Urban Tech­
nologies. Dordrecht: Springer.
Rosenberger, Robert and Peter-Paul Verbeek, eds. (2015). Postphenomenological
Investigations: Essays on Human–Technology Relations. Washington, DC:
Rowman & Littlefield.
Saito, Yuriko (2022). Aesthetics of Care: Practice in Everyday Life. London:
Bloomsbury.
Saito, Yuriko (2020). “Aesthetics of Care”, in Aesthetics in Dialogue: Applying
Philosophy of Art in a Global World, eds. Z. Somhegyi & M. Ryynänen, Berlin:
Peter Lang, pp. 187–201.
Wittingslow, Ryan (2021). “Authenticity and the ‘Authentic City’”, in Technology
and the City: Towards a Philosophy of Urban Technologies, eds. M. Nagenborg,
M. González Woge, T. Stone, & P.E. Vermaas. Dordrecht: Springer, pp.
253–270.
Young, Mark Thomas (2021). “Now You See It (Now You Don’t): Users,
Maintainers and the Invisibility of Infrastructure”, in Technology and the City:
Towards a Philosophy of Urban Technologies, eds. M. Nagenborg, T. Stone,
M. González Woge, & P.E. Vermaas. Dordrecht: Springer, pp. 101–120.
Young, Mark Thomas (2020). “Maintenance”, in Routledge Handbook of the
Philosophy of Engineering, eds. Diane P. Michelfelder & Neelke Doorn. London:
Routledge, pp. 356–368.
12 Negotiating Visions of Waste
On the Ethics of Maintaining
Waste Infrastructures

Joost Alleblas and Benjamin Hofbauer

12.1 Introduction: Waste, Ethics and Maintenance

The ethics of waste has become an important academic and societal topic
(e.g., Thompson 1979; Strasser 2000; Hawkins and Muecke 2002, 2010;
Hird 2022). The growing literature on this subject shows the many ways
in which waste has evolved as a normative category to control and form
peoples, bodies and identities. Literature also shows how waste has be-
come a global problem in which responsibility and accountability are dif-
ficult to assign. However, the relationship between the ethics of waste and
the maintenance of waste infrastructures in urban contexts is still underde-
veloped, despite the ongoing urbanisation of the global population and
the ever-increasing amounts of improperly managed municipal waste.
Accordingly, this chapter seeks to highlight how cultural understandings
of waste have become embedded in waste infrastructures, becoming in-
structive for their design, and explore what role maintenance plays in the
continued embedding of these understandings. Furthermore, a collective
reimagination of waste is crucial, especially in High-Income Countries
[HIC] (UN-Habitat 2010). Support for the behavioural and lifestyle
changes deemed necessary to deal with the current environmental crises is
only possible through a collective confrontation with waste in these urban
contexts. An analysis of the socio-technical systems of waste and the ways
in which these systems are maintained allows us to better grasp how emer-
gent, new interpretations of what constitutes waste and how to best deal
with it are hampered by the design of these systems and by a static concep-
tion of their maintenance and repair.
Previous chapters in this volume have discussed how the imagination,
design and construction of systems and artefacts are intricately related to
their maintenance and repair and how maintenance can actually be con-
ceptualised as an extension of the design process (see also Young 2021a,
2021b; Edwards 2003). However, regarding the combination of mainte-
nance and waste, this idea seems counterintuitive. Maintenance and waste

DOI: 10.4324/9781003316213-14
280 Joost Alleblas and Benjamin Hofbauer

are temporally related, but do not necessarily overlap: on the one hand,
maintenance seeks to uphold the functioning of a system or artefact, po-
tentially reimagining it in the process.1 Waste, on the other hand, repre-
sents the end of a system or an artefact’s life cycle. What is wasted is no
longer of use or its usefulness has been neglected (Hawkins and Muecke
2002). Waste, therefore, often marks an end of what has been maintained,
the end of value. Furthermore, though dirt and waste are distinct concepts,
improperly managed waste can lead to the creation of dirt. Dirt has been
amply discussed in the maintenance and repair literature (e.g., Dant &
Bowles 2003; Harmer, Cooper, Fisher, Salvia & Barr 2019; Lejeune 2019).
Here, dirt stands in an antagonistic relation to the maintenance and repair
of artefacts, such as cars, vacuum cleaners and cameras, bridges and server
parks, but also urban infrastructures.
Thus, three perspectives on waste and maintenance emerge: 1) waste as
the discarded, disvalued object or material (solid, liquid or gaseous, haz-
ardous or not); 2) waste as that which, if improperly managed, endangers
the functioning of machines and instruments, processes and systems, lead-
ing to more maintenance and repair and more hazards; 3) maintenance
and repair of artefacts as a counterforce in consumer societies, keeping
things from turning into waste. The effective management of waste then is
a critical service in a society. Waste management infrastructures appear as
critical for the maintenance of urban areas and urban living conditions
(Nagle 2013; Steele & Legacy 2017). Indeed, urban areas are the sites of
the most dense expressions of infrastructure and urban dwellers almost
completely depend on these networks (Graham 2010).
However, urban infrastructures are not always perceived as critical. The
invisibility of infrastructure, together with urban habits and routines re-
garding waste, obfuscates the flow of waste through a city and conceals the
relevance for urban dwellers of unhindered waste streams. This waste flow
takes place in sewers and pipes, grinders and incinerators, opaque bags
thrown in underground collectors and garbage trucks, processes modern
urban dwellers tend not to notice – except when driving behind a truck in
a small street or when waste stops flowing (Nagle 2013). Here, mainte-
nance creates one of the moments through which municipal waste infra-
structure becomes (temporally) visible. Breakdown and crisis, as instances
of failed maintenance, also create moments in which infrastructure reap-
pears. However, this appearance doesn’t necessarily lead to insights con-
cerning previously hidden features and embedded meanings (Young 2021b).
In this chapter, we investigate the maintenance of socio-technical sys-
tems that handle waste, especially but not exclusively in advanced urban
settings. These systems include sewage systems, drains and canals all the
way to recyclable and biodegradable waste, landfills replete with miscel-
laneous and electronic waste, waste-to-energy incinerators, the shipping
 Negotiating Visions of Waste 281

and transportation of waste, nuclear waste and less obvious instances of

waste, such as greenhouse gases. The focus of our investigation is on the
ethical implications of maintaining socio-technical systems that handle
municipal solid waste, to discuss how these systems have become problem-
atic from a moral perspective and to explain why updating these systems
to comply with new societal goals and visions is difficult, especially in HIC.
We do this on the basis of the subsequent two-step process of analysis and
normative argument.
The first step of the analysis consists of laying out the methodological
framework for assessing existing infrastructure, based on Henke and Sim’s
reflexive repair of infrastructure (2020) as well as Mark Thomas Young’s
dynamic conception of maintenance (2021a, 2021b). From this basis, we
explore how waste itself is being viewed and how that view is representa-
tive of current socio-political structures, assumptions, values and norms.
This gives a descriptive outlook of what waste is and can be, and how it
manifests itself in today’s industrialised societies. Here we draw on anthro-
pological, sociological and philosophical scholarship that has sought to
understand the phenomenon of waste through different theoretical lenses
(e.g., Thompson 1979; Hawkins 2010; Douglas 2002; Nagle 2013; Scanlan
2005). We then introduce the normative claim of how waste maintenance
should be viewed. Here we seek to evaluate our interaction with waste,
arguing how the engagement with and view of waste should and could
change. Specifically, we claim that the maintenance of municipal waste
infrastructures should be reimagined as an active and dynamic process,
rather than caring for a stable, rigid artefact.
Since we believe the issues discussed here are particularly relevant for
urban areas in HIC countries, because of high levels of stratified socio-
material organisation, in which infrastructure appears as stable and per-
manent (Graham & Thrift 2007), we give examples mostly from these
countries. We argue that first and foremost, socio-technical systems of mu-
nicipal waste management in the Western Hemisphere are made to be in-
visible, and hidden from the public, and their operation is taken for
granted, while the continuous effort to keep this system in operation is
often neglected. This feature of invisibility has co-evolved with modern
understandings of civilian culture and urban progress. Our claim is that
the invisibility of waste infrastructure fixes the socio-ethical perception of
waste as something to be avoided, something contaminated, thereby fixing
its disvalue for society. This modern perception of waste, we claim, is
problematic in terms of achieving a more sustainable relationship with
waste in light of ecological concerns, health concerns and climate change.
The modern perception, facilitated by the current design of municipal
waste infrastructures in HIC, hampers the realisation of new visions of
waste. A circular economy, for example, should be able to reconceptualise
282 Joost Alleblas and Benjamin Hofbauer

waste for its potential as a resource and something that has not lost its
value. Despite steps taken in this direction, a shift in the collective under-
standing of waste in advanced urban areas has yet to occur. With 56% of
the world’s population already living in cities (World Bank Website)2, the
need to change this understanding is urgent. One step towards achieving
this reconceptualisation is through the dynamic maintenance and reflexive
repair of waste infrastructures (Young 2021a, 2021b; Henke & Sims
2020). In this study, we combine these insights with a focus on the values
and meanings embedded in waste infrastructures. Our conclusions amount
to a first step that we deem essential for new visions of waste to gain a
foothold in advanced urban contexts.
The structure of the chapter outlining this argument is as follows. First,
we introduce our methodological background considerations, outlining
key concepts of maintenance and infrastructure. Then, we discuss waste as
a dynamic and normative concept. In Sections 12.3 and 12.4, we explore
how current waste management infrastructures tend to reproduce a spe-
cific waste imaginary, linked to a modern, urban ideal of tidiness. In these
sections, we look at municipal waste mostly from a HIC perspective.
Section 12.5 provides a further theoretical reflection by looking at waste
infrastructure through the lens of socio-technical systems. This section
highlights how values are embedded, afforded and resisted through the use
of such systems. The subsequent Sections 12.6 to 12.9 introduce how value
change affects these systems, subsequently presenting budding alternative
visions of waste realised through dynamic maintenance and reflexive re-
pair. The concluding section gives recommendations for further research.

12.2 Maintenance and Infrastructure

In recent years, maintenance has emerged as an important topic in the eth-
ics and philosophy of technology, challenging the dominant focus on de-
sign and innovation. Maintenance, scholars stress, should be understood
as more than simply preserving the function of artefacts (Graham & Thrift
2007; Jackson 2014; Young 2021a, b). Instead, maintenance often involves
(re)design and innovation, creativity and adaptivity. Therefore, rather than
as static objects in which designers’ ideas are realised and carefully guarded,
artefacts should be seen as processes of change in which maintenance
interferes with, reinforces, guides or blocks these processes (Graham &
Thrift 2007; Young 2021a).
Conversely, maintenance can also be understood as a challenge to the
ideology of consumer societies, in which innovation, creative destruction,
planned obsolescence and excessive growth (and thus lots of waste) are
ultimate goals. Instead of a narrative of technological progress, mainte-
nance offers an ideology of care for artefacts and, thus, for the people who
 Negotiating Visions of Waste 283

depend on these artefacts (Denis & Pontille 2015). The maintenance of

urban infrastructure is no exception. Rather than building new infrastruc-
tures that deliver new public goods, most of the time cities are places where
existing infrastructure is updated, altered, repaired and maintained, often
within tight budgets. This maintenance is an aspect of daily, urban life.
However, its relevance is often neglected, even though maintenance carries
the burden of keeping intact material and social orders in urban contexts
(Graham & Thrift 2007).
Urban infrastructure, that is, the pipes, drains, cables, road networks,
etc. plus the agents and institutions running and maintaining them, is
often invisible to the everyday user. Despite this common association, the
(in)visibility of infrastructure remains a contentious topic (Larkin 2013).
Star’s (1999) influential study on the ethnographic background of infra-
structure has reinforced the idea that infrastructure goes unnoticed, as a
seamless web (Hughes 1986), in order to serve its function. Likewise,
Edwards (2003) sees infrastructure as the invisible background and the
substrate of modernity. At the same time, however, infrastructures do
often become noticeable and apparent when we interact with them. This
might come in moments of breakdown, which open new perspectives on
the technologies we use, and are often instructive of innovation trajecto-
ries (Jackson 2014). Furthermore, as Jackson (2014) implies, it is in
maintenance and repair that we might encounter relations of value and
social order that are often obscured under the ‘smooth functioning of
complex sociotechnical systems’ (p. 231), although this encounter is con-
tested (Young 2021b). But breakdown and maintenance are not the only
ways in which infrastructures reappear. As repair scholars Henke and
Sims write:

Infrastructures are never truly invisible, but their salience to us, includ-
ing their role in shaping power and privilege, shifts in and out of our
consciousness based on their operation and to what extent we depend
on or are oppressed by them.
(Henke & Sims 2020, 143)

A second aspect of infrastructure that is relevant to our discussion, besides

the issue of (in)visibility, is fragility. Nagle’s (2013) analysis of urban sani-
tation workers in New York shows that garbage collection is a fragile as-
semblage of people, punching clocks, bureaucratic measures, weather
forecasts, noses, luck, intuitions, truck maintenance schedules and grunting
citizens. In her account, the ‘smooth functioning’ of this system is far from
obvious. Rather, it seems to be in a constant mode of crisis, always about to
come to a halt and disintegrate. It can only appear smooth from the out-
side. Indeed, maintaining a technological artefact or a socio-technical
284 Joost Alleblas and Benjamin Hofbauer

system can be understood as the continuous abatement of letting that ob-

ject or system go to waste, despite seemingly solid and robust infrastruc-
tures. In the case of urban sanitation and many other urban structures,
what needs to be maintained, furthermore, are not only machines but also
work relations and routines, a company culture and a precarious relation-
ship with other urban dwellers. What is maintained, then, is a socio-material
world (Sormani, Bovet & Strebel 2019). Furthermore, this maintenance is
not always successful. Rather, the system breaks down repeatedly, making
waste suddenly a public concern (Nagle 2013). As a consequence, the
politicisation of waste in urban contexts is often made possible by unsuc-
cessful and neglected maintenance and repair of waste management sys-
tems. The fragility of the system is precisely what enables its politicisation.
Maintaining socio-technical systems means ensuring their continuity,
alongside the inherent values and agents that underpin that system.3
Evidently, infrastructure manifests human values and norms (Edwards
et al. 2009). Furthermore, infrastructures, given both their centrality and
invisibility, continuously reproduce certain values and norms, certain ways
of social being, cooperation and existence. Their repair in the form of
maintenance often leads to those values and norms being reified through
adaptation and slight, which in turn makes the infrastructure ever more
resilient to change. In a similar vein, Henke and Sims problematise the fact
that repair as maintenance reproduces and embeds current ‘… material
and discursive investments and assumptions embedded in those infrastruc-
tures …’ (2020, 121). Repair as maintenance could, in this sense, lead to
the further entrenchment of the status quo.
This resistance to change is a third aspect important for our analysis,
besides invisibility and despite fragility. Resistance is closely related to
Henke and Sims’ (2020) idea of resilience in infrastructure design: the pro-
tection against destabilising feedback loops in infrastructure use. According
to Henke and Sims, growing resilience is a consequence of the two ways in
which infrastructure is reflexive (2020, 124f). One, infrastructure is caus-
ally reflexive, that is, infrastructures are shaping the very environments
within which they operate. For instance, the required additional infrastruc-
ture to maintain sewer pipes further entrenches the initial infrastructure on
a physical and societal level. Two, the infrastructure’s causal reflexivity
requires the engineers, operators and policymakers involved in the design
and operation of the infrastructures to be ‘self-aware’ (Henke and Sims
2020, 125). Accordingly, anticipating disruption paradoxically enables
disruptive behaviour. For example, waste infrastructure that anticipates
poor recycling behaviour (e.g., through the introduction of further track-
ing systems or additional sorting after the collection) and seeks to remedy
those disruptions reflexively becomes more resistant to a necessary over-
haul or rethinking of its central assumptions on how people view waste.
 Negotiating Visions of Waste 285

The disruptions are reflexively fended off, and life can continue business-
as-usual without the need to reflect on the values that drive the behaviour
and attitudes towards waste, as well as the need for the infrastructure in
the first place. The degree to which disruptions succeed in discontinuing
urban flows seems to determine, to a large extent, the degree to which citi-
zens are able to engage with the politics, ideology and embedded values of
waste and waste infrastructures (Graham 2010). A steady and uninter-
rupted urban flow eliminates the need for reflection on a socio-political
level.
Concluding, urban infrastructures are essential to (modern) society and
resistant to change. However, the age of the Anthropocene and environ-
mental reckoning of societies built on ever-increasing consumption has
shown the limitations and dangers of these infrastructures. Despite their
continued function and delivery of energy, urban sanitation and stability,
these systems are also ‘broken’. Fossil-fuel-based energy systems, for ex-
ample, show all of the above characteristics of an incredibly resilient piece
of infrastructure that at the same time has become an existential threat to
humanity. Similarly, albeit to a much lesser existential degree, current
waste infrastructures are struggling to break free from now problematic
and unsustainable perceptions of resource depletion – what we refer to as
a ‘modern’ waste imaginary (see Section 12.6). The question then arises
whether we can change these current infrastructures, whether we can rei-
magine waste and waste management to account for meaningful sustain-
ability and social justice – all this without proposing a design from scratch
or letting waste management systems break down and disintegrate com-
pletely. Maintenance seems to fit this role well. It allows the questioning of
socio-technical regimes and socio-material routines and practices in urban
environments without necessarily falling for the intrusive dogmas of in-
novation and design. Maintenance, breakdown and repair of infrastruc-
ture, furthermore, are able to engage citizens. They might successfully
bring to the fore the inconspicuous culture and politics of waste.

12.3 Reflexive Repair and Dynamic Maintenance

Scholarship on maintenance and repair provides us with at least two pos-
sible answers to the question of how to change critical infrastructure, such
as municipal waste management systems. In both these answers, the focus
is on challenging the existing material and social order through the mainte-
nance of infrastructures that always already reinforce these orders. One,
through ‘reflexive repair’ (Henke & Sims 2020), and two, through a more
dynamic and procedural conception of maintenance (Young 2021a, 2021b).
The guiding principle behind ‘reflexive repair’ is mainly discursive, that
is, making the agents maintaining infrastructure projects aware of the
286 Joost Alleblas and Benjamin Hofbauer

underlying power dynamics as well as the values purported through these

projects. Reflexive repair means ‘… asking questions that bring the some-
times obscure but always present properties of infrastructures to the sur-
face, critically appraising the embedded dynamics of power, discourse, and
materiality that are built into these sociotechnical structures’ (Henke &
Sims 2020, 143). Reflexive repair opens up a space for debate, for interac-
tion among those stakeholders who otherwise would not communicate
with one another. Such interactions might lead to the realisation that main-
taining a given kind of infrastructure in a specific way through repair is not
a value-neutral undertaking. Rather, it is the re-assertion of specific world-
views and specific ideas of urban living and cooperation. A re-assertion of
ways of disclosing the city and disclosing urban life. Preceding that re-
assertion by a reflexive and discursive engagement could highlight prob-
lematic power dynamics and injustices.
Henke and Sims (2020) summarise their views by proposing that reflex-
ive repair will be able to ‘repair infrastructural repair itself’, providing a
critical, political form of repair. Thus, reflexive repair proactively considers
the limitations and (unintended) consequences of repair and maintenance,
effectively negotiating maintenance amongst stakeholders with different
aims and needs. Henke and Sims are aware of the difficulties of putting
their approach into practice. For instance, we can question the manner in
which a representative group of stakeholders is brought together in a glo-
balised economy with complex supply chains. As an example, vast amounts
of discarded ‘fast fashion’ from the Global North end up in Africa and
increase economic inequality (Brooks 2019). Second, the unprotected dis-
assembling of container ships in Bangladesh (Jackson 2014) and informal
recycling of e-waste in developing countries (Ádám et al. 2021) show that
discarded artefacts have become a global issue with a global supply chain,
in which questions of justice are still disregarded. Can reflexive repair con-
sider all these global consequences? And how would a hierarchy of repair
preferences be justified? We do not attempt to answer these questions con-
cerning ‘waste justice’ here, but realising its complexity is important.
The second approach to changing infrastructures is based on Young’s
reconceptualisation of maintenance as a dynamic procedure, rather than a
static continuation of a given socio-technical system. The main thrust of
this reconceptualisation of maintenance stems from a fundamental shift in
how technology is viewed. Many current approaches in the philosophy of
technology still frame technological artefacts as expressions of human
(usually engineers’ and designers’) intentions. For instance, the value-
sensitive design approach (Friedman, Kahn & Borning 2002) prioritises
intended values in the design process of HCI technologies. The proponents
of Constructive Technology Assessment have stressed the importance of
 Negotiating Visions of Waste 287

broadening the design, development and implementation processes (Schot

& Rip 1997). Verbeek’s (2011) mediation approach focuses on the design
of human–world relationship through technology. In all these approaches,
the focus is on the design stage of artefacts: how an object is brought into
the physical world through construction, following an ideal, preconceived
form. The artefact is complete and ready to use at its creation, and any
deviation from the designed ideal form is remedied through repairs or ad-
justments. Maintenance here plays an auxiliary role, meant to aid a fin-
ished design to cope with reality. This is what Young (2021a) dubs
‘technology as form’.
Thus, maintenance, under this framing, is conservative of an initial de-
sign and function. In this understanding, artefacts are discarded when this
initial design and function can no longer be restored. These artefacts are
beyond repair – the change that time and use have created have pushed
them over the brink of ‘preservation’ (Young 2021a, 360). With the intent
to move the focus from the designers to the maintainers, Young proposes
maintenance as the primary drive and functioning of technology. Rather
than highlighting the design and innovation phase of an artefact, more
emphasis should be placed on the continuous process of maintenance
through repair, readjustment and upkeep. Technology should be seen as
‘a process sustained by constant human activities of maintenance and re-
pair’, that is, ‘technology as a process’ (Young 2021a, 102).
Hence, we should think of the process of making, using and maintaining
a technology as a process of growth, guided throughout different, fluctuat-
ing phases of interaction with the artefact. The following quote exemplifies
this idea:

Technologies are understood as fluid entities which constantly respond

to the changing environments in which they exist, and which therefore
require guidance through time in order to appear for us as stable
entities.
(Young 2021b; 364)

This fluidity and room for growth, coupled with reflexive repair is in focus
when we argue that waste management systems can be reimagined to rep-
resent different visions of waste. In other words, we claim that a dynamic
and reflexive conception of technology as a process gives theoretical room
to re-conceptualise waste infrastructure/management systems and open up
these systems for the negotiation of new, sustainable visions of waste. In
the coming sections, we also show that this reconceptualisation encounters
considerable sociomaterial resistance. However, we first need to establish
what we mean by ‘waste’.
288 Joost Alleblas and Benjamin Hofbauer

12.4 A Dynamic Concept of Waste

Having established the concepts that we need to address and problematise
municipal waste management, we now turn to waste itself. For the focus
of this chapter, we chose municipal solid waste as a distinct category of
waste, aside from fluid and gaseous municipal waste. Municipal solid
waste has several characteristics that we will explore in relation to munici-
pal waste management and waste infrastructures. Additionally, we show
different perspectives on waste and then highlight how different valuations
of waste have emerged over time. Finally, we show how the modern imagi-
nary of waste hampers the materialisation of new visions of waste, as, for
instance, proposed in the European Union.
Based on philosophical, anthropological and sociological scholarships,
we identify four main characteristics of waste. Waste is 1) a socially deter-
mined product, 2) a normative category, consisting of ethical and aesthetic
components, 3) an environmental agent and 4) inevitable. These four as-
pects give respective explanations of waste’s ontological standing and ethi-
cal significance (1 & 2), its practical implications (3) and why waste merits
debate in the first place (4).
First, waste is a socially determined product. Importantly, waste is not a
predetermined, ontological category, something that can be objectively
distinguished. Rather, waste is the product of a relational process of cate-
gorisation (Hawkins 2010; Douglas 2002). Structurally, we can argue that
waste, like dirt, always is ‘matter out of place’ (Douglas 2002). However,
that doesn’t determine what counts as matter out of place in a certain (cul-
tural) context. Thus, what is considered waste is culturally dependent (e.g.,
Edgerton 2007). The category of waste is itself a social product, something
that has to be agreed upon and negotiated to be ‘waste’. This means that
waste itself is a dynamic concept that changes based on historical context,
societal norms and values and the possibilities present in societies to rein-
terpret what constitutes waste. However, this dynamism of waste is cur-
tailed by fixed processes, institutions and infrastructures for waste. Thus,
while waste is nominally dynamic, real processes and structures fix its
meaning through reproduction and static engagement.
This nominally dynamic character of waste becomes especially poignant
when considering the ethical and aesthetic relevance of waste. Waste is an
ethical and aesthetic category. From an aesthetic perspective, waste is
something that is unpleasant, even disgusting and unclean. We want to
avoid contact with waste. The idea of waste evokes sentiments of disgust,
uncleanliness, potential danger through contamination, etc. These aesthetic
judgments also have moral implications. Especially, if these judgments are
mobilised to produce emotional reactions and relationships with waste,
they fall clearly into the category of the ethical (Roeser & Todd 2014).
 Negotiating Visions of Waste 289

In those cases, these reactions have real-world effects that bear upon other
people, close-by or far-off, affecting their lives and livelihoods.
Waste surrounds us, while always being something undesirable.
Designating something as waste entails that it no longer has use, has lost
its initial purpose or is spent. On the ethical significance of waste, we fol-
low Hawkins when she writes that ‘[w]aste is now a field of activity struc-
tured by legislated and normative moralities, by disciplinary codes that
order conduct in the interest of wider objectives: from reduction of landfill
to global ecological survival’ (2010, 22). The way we engage with, catego-
rise and maintain waste has important moral implications, since waste is
a normative category: it tells us how to interact with a given artefact or
phenomenon. We focus on this criterion of waste in the following
sections.
Thus, the ethical and normative characters of waste highlight the third
and fourth practical and moral criteria of waste tied to its maintenance.
From this third perspective, waste is relevant as an environmental phe-
nomenon. The production and effective management of waste is seen as
crucial in dealing with environmental degradation. Spilled and improp-
erly processed waste can emerge as pollution and environmental hazard.
This aspect of waste is further compounded by environmentalists’ and
climatologists’ realisations that one of the biggest threats to humanity
yet – the continuous emissions of CO2 through combustion – is the ulti-
mate kind of waste through its double invisibility. Not only is it invisible
in the physical sense, as it cannot be perceived by the human eye and is
odourless, that is, our senses cannot make sense of it. It is also invisible in
its direct impact.
Finally, waste seems inevitable. There will always be matter out of place,
as waste and dirt establish material and social orders and borders. The
very human condition seems trapped in the necessity to produce waste and
to create excess. The inevitability of waste production is particularly glar-
ing in industrialised consumption-based societies, where waste rears its
head undeniably through packaging, sewage and broken gadgets – societies
in which, as Nagle (2013) puts it, ‘San[itation] workers are key players in
maintaining the most basic rhythms of capitalism’ (46). This connects to
Schumpeter’s (2008 [1950]) more general idea of ‘creative destruction’, the
destruction of old forms of wealth to create new ones, that is, the constant
upheaval of economic structures in capitalist societies. The production of
waste becomes necessary for economic progress.

12.5 Modern Waste Management: An Institutional Perspective

So far, we have argued that the dynamic nature of waste is most properly
complemented by a dynamic concept of maintenance that allows
290 Joost Alleblas and Benjamin Hofbauer

infrastructure to be aligned with new visions of waste. However, as we

have indicated, infrastructures provide resistance to change as well. The
coming sections pay attention to some of the forms of resistance that ur-
ban infrastructures exhibit.
Given the sheer production and amount of waste, modern municipal
waste management systems need to be invisible. In this aspect, they differ
from other systems operating in our urban backgrounds. Insensibility is
deliberately designed into municipal waste management systems (Hawkins
2007). Rather than retreat into the background of modern urban life, mu-
nicipal waste management infrastructure always already has retreated.
Recent garbage crises that affluent societies have faced mostly concerned
the re-emergence of the sensibility of waste and the re-emergence of the
question of how to best deal with this sensibility. More philosophical and
ethical questions about waste often escape the political agenda. That is,
infrastructural innovation is directed towards technological efficiency and
efficacy regarding a set of predetermined goals taken as exogenous ‘givens’:
health, tidiness and, therefore, insensibility. The existing rationale for
waste management was, and to a large extent still is, ‘expand and up-
grade’, based upon design principles committed to universal, reliable and
affordable service (Moss & Marvin 2001, p. 5/6). Figure 12.1 shows a
schematic overview of the four phases of solid waste management in urban
areas in HIC, phases that are (in)formed by infrastructures, technologies,
habits and institutions.
One way to understand the neglect of philosophical and ethical ques-
tions concerning waste is provided by the literature on socio-technical re-
gimes. In this institutional perspective, the focus on efficiency, expansion
and upgrading is seen as part of the socio-technical regime of urban solid
waste management in HIC, constituting its paradigmatic centre. This re-
gime is the result of the co-evolution of infrastructure and institutions.

Figure 12.1 Schematised Municipal Waste Infrastructure.

 Negotiating Visions of Waste 291

It has its own ‘logic’ and ‘grammar’ (Fuenfschilling & Truffer 2014).
Problems and their solutions come about as salient through this logic,
which forms ‘coherent arrangements of beliefs, norms, values and prac-
tices’ (ibid. p. 773).
We have seen, furthermore, that waste management systems (Figure
12.1) come into the public eye when they break down or when their con-
tinued use and operation leads to morally problematic situations, such as
disease, pollution, the involvement of organised crime and increased global
inequality. In a socio-technical regime, these problems are often interpreted
and dealt with in preconfigured ways, along already-established innova-
tion pathways. The robustness of this institutional logic, its endurance, is
determined by both internal and external events, such as the arrival of new
actors and niche technologies, changes in the distribution of power and the
social impact of crises and disasters (Geels 2010).
The institutional perspective adopted in the socio-technical regime’s lit-
erature explains how institutions and technologies co-evolve in socio-
technical systems, such as municipal waste management systems. From
this perspective, the extent to which a complete reimagination and system
transition is deemed necessary depends at least in part on the success of
other (and earlier) attempts of the regime to cope with problematic situa-
tions caused or endured by the system. These attempts might have left the
material design of the municipal waste management system largely intact.
We can think here of failed attempts at the improvement of the institu-
tional context (new rules, guidelines, laws, etc.) or failed efforts at educat-
ing city dwellers about good and bad waste habits (stickers on garbage
bins, codes on packaging materials, waste separation folders, school trips
to landfills, etc.). In these cases, new objectives and goals in waste manage-
ment are pursued at the supply side of waste, changing the waste behav-
iour of citizens through education and regulation. In the case these new
goals amount to substantial revisions of the ultimate objectives of waste
management, we refer to these new objectives as ‘values’. The pursuit of
new values is often associated with socio-technical transitions, in which
new practices, rules and technologies emerge (Geels 2004).
We depict the changes in these practices, rules and technologies in Figure
12.2. In this visualised trinity of (re-)education, institutional (re)design and
material (re)design, the latter option, the material (re)design of parts of the
socio-technical system of waste management, is often the most time-
consuming and costly. We are dealing in these infrastructural systems with
a materialisation of norms and values associated with waste, that have
‘locked in’ certain waste practices. The underground sewage pipes, the
closed-off garbage trucks or covered trash cans, all represent physical man-
ifestations of a certain understanding of waste that leaves its mark on
waste practices. At the same time, physical infrastructures embedding
292 Joost Alleblas and Benjamin Hofbauer

Figure 12.2 Visions of changing (waste) practices in socio-technical systems.

entrenched norms are very difficult to change, which is why acting on them
is often the last option to be considered.
This leads to a paradoxical situation: citizens are instructed to change
their behaviour and practices while existing infrastructure keeps reinforc-
ing that behaviour and those practices. Institutionally, therefore, citizens,
are motivated to change, while materially they are motivated to keep on
doing what they did. A purely institutional perspective on change is there-
fore not enough. If cities want to successfully implement new waste vi-
sions, all three aspects of Figure 12.2 need to be put into operation.
Institutional change, citizen education and motivation and material change
need to go hand-in-hand. This means the inclusion of diverse stakeholder
groups, such as citizens and maintainers, and the development of new
ideas concerning the interaction of these groups. In our view, maintenance
offers an underestimated opportunity for effectively bringing to the fore
the material aspects of socio-technical transitions that are inspired by new
visions of urban waste.
Finally, upholding a modern vision of municipal waste (‘Burning,
Burying, Be rid of it’) through the static maintenance of waste infrastruc-
tures is costly. Most of these costs might be borne by future generations,
the environment or subaltern social groups. Some of these costs might be
invisible (air pollution and CO2 emissions of waste transport) or might not
 Negotiating Visions of Waste 293

be experienced collectively (the shipping of waste into other regions/

countries). Hence, local invisibility is often associated with affluent parts of
the city or affluent parts of the world, while waste becomes more visible in
marginalised localities. Affluence, in this sense, determines the period dur-
ing which municipal waste is visible and still affirmed and a confrontation
and negotiation can take place.
This is a development long in the making. As a number of scholars have
pointed out, the modern imaginary of urban living and waste focused
around conceptions of tidiness that emerged throughout the nineteenth
century in cities in Northern America and Europe (Chakrabarty 1992;
Melosi 2004; Hawkins 2010; 2007). These conceptions have solidified in
infrastructural design choices for modern urban waste management sys-
tems and continue to inform contemporary choices.

12.6 Modern Imaginary of Waste

The modern imaginary of urban living and waste emerging in the nine-
teenth century is the ‘tidy city’ (Chakrabarty 1992; Hawkins 2010; 2007).
In this tidy city, the management of waste (and health) has become a public
concern. This means that urban sanitation is no longer an individual re-
sponsibility, as it was prior to the nineteenth century, but a communal re-
sponsibility and a service that the city provides (Melosi 2004). Furthermore,
responsible waste practices become an important aspect of civic life and
citizen culture. The administrative treatment of waste and public health
becomes a sign of urban progress. As the Dutch environmental ministry’s
homepage puts it: ‘It is the role of governments to prevent the waste of raw
materials and energy and to ensure that human health and the environ-
ment are not harmed by waste’(Ministerie van Infrastructuur en Waterstaat
2022). Modern waste management is a public concern.
A wide range of scholars of waste have shown how this nineteenth-cen-
tury imaginary of waste is both persistent and resistant in the western
hemisphere. It has set societies on socio-technical pathways of waste man-
agement systems that have led to both technological and institutional lock-
in. The force of this imaginary of tidiness is dependent upon several of its
features. First, its geopolitical power. As Newell (2015) and Chakrabarty
(1992) argue, the assumed connection between cleanliness and civilisation
has justified colonisation and subjugation of the ‘dirty native’ in the nine-
teenth and early twentieth centuries. Filth today is still associated with
lower social classes (Berg 2015; Forty 1986), and with ‘othering’.4
Second, overconsumption, and therefore waste, started to be taken as a
fact of civilian life. Nineteenth-century affluent citizens began showing
their wealth by buying things they wanted, but didn’t need and could dis-
card at will (Benjamin 1999).5 Third, the medical sciences of virology and
294 Joost Alleblas and Benjamin Hofbauer

bacteriology emerged at the end of the century, as modern public health

science established relations between (human) dirt, disease and death that
are still prevalent today (Melosi 2004). Finally, the emergence of munici-
pal sanitation had both ethical and aesthetic consequences. The ‘tidy city’
still is an attractive epithet. It attracts tourists, yuppies and supports in-
creasing levels of gentrification (Hawkins 2007). Clean, tidy and orderly
urban districts preserve the otherness of waste – as something that is both
aesthetically and morally disavowed.
On this basis, we introduce the modern waste imaginary. In this imagi-
nary, waste is something to be discarded; something that is both aestheti-
cally unpleasant, morally bad and unwanted. It is out of place and should
be put away, either geographically or excised and obliterated as a whole.
Even apparently more reflective approaches, such as the Zero Waste plans
of cities like San Francisco and Singapore uphold this image: ‘Zero Waste’
here does not refer to a completely circular (urban) economy or 100%
biodegradable consumer goods. Rather, these Zero Waste plans rely heav-
ily on waste-to-energy incinerators. These incinerators reduce waste mass
up to 90% (Setoodeh Jahromy et al. 2019). Since the leftover material can
be used for road construction and leftover metals can be separated and
recycled, these Zero Waste plans almost realise Zero Waste.
However, as we detail in the latter sections of this chapter, these plans
fall short of realising a new vision of waste on the supply side of waste, for
example, with urban dwellers. In the next section, we introduce how socio-
technical systems can entrench or hinder the expression and materialisa-
tion of new values. This provides theoretical tools to explore how current
waste infrastructures either reproduce the modern waste imaginary or do
not go far enough where they seek to go beyond it, given the inherently
static underlying assumptions of current waste management systems.

12.7 Value Change
A growing amount of recent scholarship recognises that socio-technical
systems, since they’re bound to exist for multiple generations, will see
some level of value change. In short, although these systems might have
been designed while keeping certain social, ethical and public values in
mind, the relative importance of these values, their conceptualisation or
the (design) norms associated with their realisation, might change (van de
Poel 2021; van de Poel & Taebi 2022; van de Poel & Kudina 2022). We
can even imagine completely new values emerging in society that start to
weigh down on the operation of certain systems, such as energy or waste
systems (Taebi et al. 2020).
The literature on value change argues that morally problematic situa-
tions, whether caused by these socio-technical systems or not, create an
 Negotiating Visions of Waste 295

urgent call to reassess the design of those systems, with this value change
in mind. These might be values new to the context (e.g., new to the context
of waste management systems, such as ecological sustainability) or com-
pletely new values as abstract goals (e.g., intergenerational justice). That is,
the redesign of parts of the system of waste management, deemed neces-
sary because of the problematic situations the current design has caused, is
often guided by new requirements that stem from new values. As is evident
from our discussion so far, the realisation of these new values cannot be
achieved through institutional change only. Despite a recognition that mu-
nicipal solid waste management needs to transition to more sustainable,
just and ecological practices, urban waste management infrastructures
continue to express an imaginary of tidiness that trumps the expression of
these new values. Both education and institutional (re)design (Figure 12.2)
have been unable to realise major shifts in collective beliefs and practices
concerning waste in European urban areas. To date, the focus has been
primarily on how to deal with waste, and visions are built around answers
to this question. Yet, another question often goes unmentioned: to what
extent should we change what we consider waste in the first place?
For instance, in Europe, the Smart City concept developed in the early
2000s, hardly pays attention to the reimagination of municipal solid waste,
despite a focus on issues such as urban pollution, sustainability and sanita-
tion. Smart waste collection does receive attention in urban contexts (e.g.,
Ali et al. 2020), but seems to focus more on Internet-of-Things solutions in
which garbage cans communicate with garbage trucks (e.g., Bhor et al.
2015) and sensors determine the level of air pollution. The focus here is on
efficiency and reliability, rather than new values emerging in relation to
waste management.
Furthermore, the EU directive of 2015 established a hierarchy of goals
in relation to waste management: prevention, reuse, recycling, recovery of
energy and disposal. However, many member states have difficulties in
achieving these goals and often focus on waste-to-energy processes instead
of higher order goals (EU 2018). The failure to achieve a more radical
transformation of municipal waste practices can be understood to stem in
part from limitations in the political directives themselves. For example,
one of the research foci of the Horizon 2020 programme was the circular
economy [CE] and urban metabolism, which developed valuable insights
into the challenges of CE in urban contexts, mostly from a governance
perspective (e.g., Obersteg et al. 2019; Remøy et al. 2019). At the same
time, however, the programme failed to address the affordance and resis-
tance of waste management systems or the potential for a collective rei-
magining of waste and related practices.
Hence, waste is often assumed as a given fact of urban life. Contemporary
debates concerning the ethics of waste highlight problematic attitudes to
296 Joost Alleblas and Benjamin Hofbauer

waste, delivering a moral appeal to change these attitudes (Hawkins 2010).

Yet, the extent to which the current treatment of and attitude towards
waste are seen as problematic is as much the result of the history of waste
(see Section 12.6), as of the persuasive power that circulating new perspec-
tives on waste are able to exercise. This persuasive power itself, we argue,
depends both on the possibilities to render visible the problematics of the
current operation of municipal waste management systems and on the his-
torical meanings of waste still designed into these systems, in the form of
embedded values. We call these two aspects infrastructural affordance and
infrastructural resistance.
A collective reimagination of waste, deemed necessary for collective sup-
port of changes in socio-technical systems of waste management and the
burdens these changes might impose on citizens, requires a collective con-
frontation with municipal waste in which new values can be brought for-
ward as essential for the accumulation, collection and disposal of solid
waste in urban contexts. For such a confrontation to take place, it will be
essential to explore new ways of experiencing waste. However, despite a
range of art projects involving waste, the conditions for such positive
‘waste experiences’ are still largely absent in affluent societies. One of the
reasons for this may be found in the physical aspects of waste infrastruc-
tures, where the resistance presented in the infrastructurally embedded his-
torical meanings of waste weakens the persuasive power of new visions.
Reconceptualising the maintenance of municipal solid waste manage-
ment systems as reflexive repair and a dynamic process can play two roles
here. First, it can undermine the resistance to value change designed into
these systems by rendering them visible during the act of maintenance and
allowing for new forms of political engagement with them, thus bringing
to the surface the problems associated with their operation. Second, main-
tenance can be seen as a process of redesign of urban infrastructure instead
of replacement/repair and short-term fixes, making maintenance an impor-
tant strategy for bringing political and ethical questions to the fore and
thus creating infrastructural affordance. However, the redesign of critical
urban infrastructures is often avoided because of costs and risks to urban
flows of utilities. While we do not claim to have a solution to this problem,
we argue that the creation of public acceptance for new visions of waste
can only be achieved through the ‘visibilization’ of waste infrastructures, a
process of creating infrastructural affordance in which maintenance can
play a crucial role.

12.8 Some Politics of Waste Infrastructures in the EU

As noted above, the affordance and resistance of contemporary municipal
solid waste management systems determine the extent to which new mean-
ings of waste can establish themselves in urban contexts. In a very basic
 Negotiating Visions of Waste 297

sense, this would encompass the transition from a waste and dirt-rejecting
system to a waste and dirt-affirming system of beliefs and practices
(Douglas 2002) in the city, supported by a waste-affirming vision.
Anthropologically, dirt affirmation refers to societies in which some forms
of pollution and corruption are ‘enshrined in sacred places and times’
(Douglas 2002), whereas in dirt-rejecting societies dirt loses all identity
and cannot acquire new functions before it is completely disintegrated and
undifferentiated, when it has stopped being dirt.6 Importantly, waste serves
as a proxy for dirt in the modern waste imaginary. Most advanced urban
areas are dirt rejecting in a non-religious sense, as places where ‘order, ef-
ficiency, and perfection’ (Scanlan 2013, p. 2) leave no room for the reintro-
duction of garbage as an object or even commodity before it has completely
disintegrated in landfills or incinerators. Before it has stopped being gar-
bage, that is. Nonetheless, we argue, that some level of dirt affirmation
must occur to effectively negotiate waste and open up a critical space for a
collective reimagination of what constitutes waste.
However, existing waste infrastructures are both robust and durable and
infrastructure tends to become more resilient over time (Henke & Sims
2020, 121). Late twentieth-century problems with landfills and levels of
toxicity and contamination of water tables have led many European coun-
tries to focus on waste-to-energy incinerators (McCauley 2009). The cur-
rent ubiquity of incinerators is a consequence of the mixed nature of most
solid municipal wastes, stalled efforts at sorting out different types of
wastes, protests against shipping waste abroad and failed attempts at
large-scale recyclable plastics. Nevertheless, many of these countries have
also seen (local) forms of resistance against these incinerators that, in some
cases, have led to further innovations (e.g., reduction of fly ash and better
filters). The EU directive 2018/851 recognises that many member states
have not yet developed the necessary waste management infrastructures to
achieve a circular economy. It does not, however, provide reasons for this
failure to adapt. Overall, the directive 2018/851 pays little attention to
raising public awareness (and visibility) of waste. Action point (30) states:

The promotion of sustainability in production and consumption can

contribute significantly to waste prevention. Member States should take
steps to make consumers aware of that contribution and encourage
them to participate more actively in order to improve resource effi-
ciency. As part of measures to reduce waste generation, Member States
should include continuous communication and education initiatives to
raise awareness on the issues surrounding waste prevention and littering
and may include the use of deposit-refund schemes and the setting of
quantitative targets, and provide, as appropriate, adequate economic
incentives to producers.
(EU 2018)
298 Joost Alleblas and Benjamin Hofbauer

The EU directive 2018/851 focuses on education and institutional design

to raise awareness (see Figure 12.2) about the negative effects of waste.
Awareness, we have argued, is something that is hampered by the design of
modern waste management systems and characterised by the processes of
‘invisibilization’ of municipal waste. Efforts at prevention, reusage and re-
cycling (EU directive 2015) are all hindered by these processes.
Problems of citizen education and the collective reimagination of
waste come together in the design of waste infrastructures in affluent
countries. To gain collective support for the proposed and much-needed
changes in the way these countries deal with waste, waste perceptions in
advanced urban contexts must first be laid bare as both historically con-
tingent and antiquated. This means that anthropological, sociological,
psychological and urban geographical insights concerning waste and its
management must be combined to address local, economic, environmen-
tal and social issues. Here we believe that both the dynamic reconceptu-
alisation of maintenance as well as the reflexive repair approach can be
of use.
What both of these approaches to maintenance highlight is the fact that
infrastructure is more than a simple network of technologies, and that its
upkeep is a value-laden process. This realisation, we believe, is paramount
to making room for the re-imagination of waste infrastructure, and the
way we experience waste itself. As Henke and Sims put it,

[I]nfrastructural repair is not just about fixing things but also relation-
ships and negotiation. An approach to reflexive repair that focuses only
on technical fixes and eschews conversation misses opportunities to
build common discourses and identities around a complex and urgent
problem; talk also allows us to listen and learn when we disagree and
misunderstand.
(2020, 134)

Thus the remaining two sections of this chapter explore the potential ave-
nues of reconceptualising our modern vision of waste through waste
affirmation.

12.9 Waste Affirmation and Municipal Solid Waste Management

It is difficult to imagine what a waste-affirming modern society would look
like. Indeed, such a society might be considered ‘postmodern’ in the sense
that it would make previously established boundaries and strict dichoto-
mies between concepts and categories permeable and porous. In this light,
Douglas (2002) asks how it is possible that something as destructive as dirt
can also have a creative force. Douglas (2002), citing William James (1902),
 Negotiating Visions of Waste 299

claims that the most complete systems of thought need to find some way of
affirming what has been rejected, whether it is matter out of place (Douglas
2002), objects that have lost their aura (Benjamin 2021) or the abject ex-
crements of the body (Kristeva 1982). A more complete philosophy of the
city would account for the presence of secular rituals in which the ‘mixing
up and composting of polluting things’ (Douglas 2002) takes place without
recourse to technological perspectives on waste. The practical question is
how such rituals or, more generally, practices of waste affirmation can be
firmly established in urban environments. Do we need to design crises of
waste, or are other venues possible?
As we have shown in the previous sections, cities are not designed for
zero waste visions, but rather for zero visibility waste visions. This zero
visibility paradigm problematises the efforts to challenge the modern imag-
inary of tidiness, with its hygienic and aesthetic connotations, and prob-
lematises the efforts to remind citizens that waste is an integral aspect of
urban living that, at the same time, has negative consequences. Waste in-
frastructures resist new interpretations and visions of waste. Before be-
coming waste affirming, we need to become waste admitting. Besides
education and institutional (re)design (see Figure 12.2), waste infrastruc-
tures should also play a part in this admission. Several authors have ex-
plored what dirt affirmation would look like from a material, architectural
perspective (e.g., Campkin 2013; Shonfield 2014). In these elaborations,
dirt affirmation is related to Foucault’s (1984) thoughts on heterotopias,
places of otherness that function as mirrors to the city, as counter-sites. In
these ‘othering’ spaces, we could find the celebration of dirt and waste, the
denial of purity and tidiness, an inverted city, turned inside out, the ‘gar-
bage cities’ presented in documentaries such as Waste Land (Walker 2010)
or Plastic China (Wang 2016). However, despite the existence of heteroto-
pias (Foucault names places of ‘deviation’ such as psychiatric hospitals,
rest homes, cemeteries and zoological gardens), it is unclear how these
sites, except as places of contestation and experimentation, could contrib-
ute to changing urban waste practices and beliefs.
The ‘broken world’ thinking of Jackson (2014) may offer a new para-
digm here for thinking about urban problems and solutions. We refer to
his oft-quoted idea that we should aim at:

an appreciation of the real limits and fragility of the worlds we inhabit –

natural, social, and technological – and a recognition that many of the
stories and orders of modernity (or whatever else we choose to call the
past two-hundred-odd years of Euro-centred human history) are in the
process of coming apart, perhaps to be replaced by new and better sto-
ries and orders, but perhaps not.
(221)
300 Joost Alleblas and Benjamin Hofbauer

One such story that needs to be dismantled is the myth of urban infra-
structures as stable, and permanent. This myth of fixed and stable infra-
structure (Graham & Thrift 2007), while cities expand, and urban flows
change, obscures the extraordinary everyday effort of keeping the city
‘running’. The ongoing process of infrastructure, exempt from a narrative
of progress and innovation, needs to be brought into focus. Post-modern
cities, in this sense, are ‘cities of repair’ (Graham & Thrift 2007, 10.)
Visibilisation of waste, we have argued, comes in two forms. On the one
hand, maintenance can serve to lay bare waste infrastructures by enhanc-
ing the visibility of maintenance activities. However, this first form of visi-
bilisation only temporally allows for critical engagement with waste. The
second form is more permanent. It focuses on dynamic maintenance as an
act of redesigning waste infrastructures to allow for engagement with
waste. The resensibilisation of waste does not mean that we return to
filthy streets filled with manure and all kinds of litter. It would mean that
through this resensibilisation, waste becomes something relatable, positive
and affirmative. What is repaired and maintained are positive meanings of
waste that have become obfuscated by modern imaginaries of tidiness and
unlimited progress. We can think here of maintenance and repair as a con-
tinuous laying bare of waste streams in urban contexts comparable to the
way the Centre Pompidou in Paris lays bare its inner tubing – its flows of
air, energy and waste.

12.10 Conclusion
Waste and maintenance are inadvertently connected. We hope to have
shown that this connection can serve as a point of exploration to prob-
lematise current western urban societies’ relationship and imagination of
waste. Perceiving waste as something aesthetically and ethically problem-
atic inhibits the transition to a more environmentally aware vision of
consumption patterns. Our current vision of waste rejection is further
entrenched into the very infrastructure we rely on to rid ourselves from
our waste, which makes the re-imagination of waste all the more difficult.
A dynamic conception of maintenance alongside a reflexive approach to
infrastructure repair could aid us in reconfiguring our understanding of
what waste is and how we ought to treat it. It would create infrastructural
affordance for new practices related to new values. However, these ap-
proaches alone will not be sufficient. Instead, it will take radical new vi-
sions of waste that go beyond the modern understanding on institutional,
societal and technological levels in order to re-imagine waste as a neces-
sary reminder and remainder of our consumption, as something that we
can gladly claim.
 Negotiating Visions of Waste 301

Notes
1 See Steinert (this volume).
2 See https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS [last accessed
January 2024]
3 See Steinert (this volume).
4 Indeed, modernity is often associated with these strict categorisations and dual-
isms, such as the distinctions between mind and body, nature and culture, pub-
lic and private and agency and structure (e.g., Latour 2012; Douglas 2002;
Bauman 2003; Jackson 2014; Giddens 2020).
5 For example, the word ‘gadget’ started being used late in the nineteenth century
as a term for an object one cannot remember the name of, although the exact
origins are debated (Merrin 2014).
6 We are aware that dirt and waste are not synonymous. In Douglas’ work, the
two concepts are not always easy to keep apart. On the one hand, dirt seems to
be an interpretation and evaluation of superfluous things, materials, words and
bodily aspects, as excessive ‘things’ that confuse and endanger accepted social
classifications. On the other hand, as we have noted in the introduction, waste
is always in immediate danger of becoming dirt. Douglas herself gives examples
of human matter in food such as hairs in soups. For the sake of our argument,
we do not think that a strict conceptual boundary is necessary.

Bibliography
Ádám, B., Göen, T., Scheepers, P.T., Adliene, D., Batinic, B., Budnik, L.T., Duca,
R.C., Ghosh, M., Giurgiu, D.I., Godderis, L. and Goksel, O., 2021. From ineq-
uitable to sustainable e-waste processing for reduction of impact on human
health and the environment. Environmental Research, 194, p. 110728.
Ali, T., Irfan, M., Alwadie, A.S. and Glowacz, A., 2020. IoT-based smart waste bin
monitoring and municipal solid waste management system for smart cities.
Arabian Journal for Science and Engineering, 45(12), pp. 10185–10198.
Bauman, Z., 2003. A sociological theory of postmodernity. In Intimations of
Postmodernity (pp. 215–232). Routledge.
Benjamin, W., 1999. The arcades project. Harvard University Press.
Benjamin, W., 2021. One-way Street: And other writings. Verso Books.
Berg, A., 2015. Waste streams and garbage publics in Los Angeles and Detroit. In
Global Garbage (pp. 92–110). Routledge.
Bhor, V., Morajkar, P., Gurav, M., Pandya, D. and Deshpande, A., 2015. Smart
garbage management system. International Journal of Engineering Research &
Technology (IJERT), 4(03).
Brooks, A., 2019. Clothing poverty: The hidden world of fast fashion and second-
hand clothes. Bloomsbury Publishing.
Campkin, B., 2013. Placing “matter out of place”: Purity and danger as evidence
for architecture and urbanism. Architectural Theory Review, 18(1), pp. 46–61.
Chakrabarty, D., 1992. Of garbage, modernity and the citizen’s gaze. Economic
and Political Weekly, pp. 541–547.
Dant, T. and Bowles, D., 2003. Dealing with dirt: servicing and repairing cars.
Sociological Research Online, 8(2), pp. 1–17.
302 Joost Alleblas and Benjamin Hofbauer

Denis, J. and Pontille, D., 2015. Material ordering and the care of things. Science,
Technology, & Human Values, 40(3), pp. 338–367.
DIRECTIVE (EU) 2018/851 OF THE EUROPEAN PARLIAMENT AND OF THE
COUNCIL of 30 May 2018 amending Directive 2008/98/EC on waste (Text
with EEA relevance) https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?
uri=CELEX:32018L0851&from=EN
Douglas, Mary. 2002. Purity and danger: An analysis of concepts of pollution and
taboo. Routledge. https://doi.org/10.4324/9780203361832
EC, 2015. COM (2015) 614 final. COMMUNICATION FROM THE
COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL,
THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE
COMMITTEE OF THE REGIONS Closing the loop - An EU action plan for the
Circular Economy. Brussels 2.12.2015. Available at https://eur-lex.europa.eu/
resource.html?uri=cellar:8a8ef5e8-99a0-11e5-b3b7-01aa75ed71a1.0012.02/
DOC_1&format=PDFFoucault, M. and Miskowiec, J., 1986. Of other spaces.
Diacritics, 16(1), pp. 22–27.
Edgerton, David. 2007. Creole technologies and global histories: rethinking how
things travel in space and time. Journal of History of Science and Technology,
1(1), 75–112.
Edwards, P.N., 2003. Infrastructure and modernity: Force, time, and social organi-
zation in the history of sociotechnical systems. Modernity and Technology, 1,
pp. 185–226.
Edwards, P.N., Bowker, G.C., Jackson, S.J. and Williams, R., 2009. Introduction:
an agenda for infrastructure studies. Journal of the Association for Information
Systems, 10(5), p. 6.
EU 2018. Directive (EU) 2018/851 of the European Parliament and of the Council
of 30 May 2018 amending Directive 2008/98/EC on waste (Text with EEA rel-
evance) https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:32018L0851
Forty, A., 1986. Objects of Desire: a history of commodity design. Pantheon
Books.
Friedman, B., Kahn, P. and Borning, A., 2002. Value sensitive design: Theory and
methods. University of Washington technical report, 2, p. 12.
Fuenfschilling, L. and Truffer, B., 2014. The structuration of socio-technical
regimes—Conceptual foundations from institutional theory. Research Policy,
43(4), pp. 772–791.
Geels, F.W., 2004. From sectoral systems of innovation to socio-technical systems:
Insights about dynamics and change from sociology and institutional theory.
Research Policy, 33(6–7), pp. 897–920.
Geels, F.W., 2010. Ontologies, socio-technical transitions (to sustainability), and
the multi-level perspective. Research Policy, 39(4), pp. 495–510.
Giddens, A., 2020. Modernity and self-identity: Self and society in the late modern
age. In The new social theory reader (pp. 354–361). Routledge.
Graham, S., 2010. When infrastructures fail. In Disrupted cities (pp. 13–38).
Routledge.
Graham, S. and Thrift, N., 2007. Out of order: Understanding repair and mainte-
nance. Theory, Culture & Society, 24(3), pp. 1–25.
 Negotiating Visions of Waste 303

Guy, S., Marvin, S. and Moss, T., 2001. Urban infrastructure in transition:
Networks, buildings, plans. Earthscan.
Harmer, L., Cooper, T., Fisher, T., Salvia, G. and Barr, C., 2019. Design, Dirt and
Disposal: Influences on the maintenance of vacuum cleaners. Journal of Cleaner
Production, 228, pp. 1176–1186.
Hawkins, G., 2007. Waste in Sydney: Unwelcome Returns. PMLA, 122(1),
pp. 348–351.
Hawkins, G. 2010. The ethics of waste: How we relate to rubbish. Lanham:
Rowman & Littlefield Publishers.
Hawkins, G. and Muecke, S. eds., 2002. Culture and waste: The creation and de-
struction of value. Rowman & Littlefield Publishers.
Henke, C.R. and Sims, B., 2020. Repairing infrastructures: the maintenance of
materiality and power. Mit Press.
Hird, M.J., 2022. A public sociology of waste. In A public sociology of waste
(pp. 83–98). Bristol University Press.
Hughes, T.P., 1986. The seamless web: technology, science, etcetera, etcetera.
Social Studies of Science, 16(2), pp. 281–292.
James, W., 1902. The varieties of religious experience, New York: The Modern
Library.
Jackson, Steven J. 2014. Rethinking repair. In Media technologies: Essays On
Communication, Materiality, and Society, 221–239. MIT Press.
Kan, S., 2015. Symbolic immortality: The Tlingit potlatch of the nineteenth cen-
tury. University of Washington Press.
Kristeva, J., 1982. Powers of horror: An essay on abjection, New York: Columbia
University Press.
Larkin, Brian. The politics and poetics of infrastructure. Annual Review of
Anthropology 42 (2013): 327–343.
Latour, B., 2012. We have never been modern. Harvard University Press.
Lejeune, C., 2019. Interruptions, lunch talks, and support circles: An ethnography
of collective repair in steam locomotive restoration. Repair Work Ethnographies:
Revisiting Breakdown, Relocating Materiality, pp. 221–251.
McCauley, D., 2009. Wasting energy? Campaigns against waste-to-energy sites in
France. Environmental Politics, 18(6), pp. 917–938.
Melosi, M. V., 2004. Garbage in the Cities: Refuse reform and the Environment.
University of Pittsburgh Press.
Merrin, W., 2014. The rise of the gadget and hyperludic me-dia. Public Culture,
10(1), pp. 1–20. https://doi.org/10.1215/17432197-2397209
Ministerie van Infrastructuur en Waterstaat. 2022. Rijksoverheid. 02 09.
Accessed 09 02, 2022. https://www.rijksoverheid.nl/ministeries/ministerie-van-
infrastructuur-en-waterstaat
Moss, Timothy, and Simon Marvin. 2001. Urban infrastructure in transition:
Networks, buildings and plans. Routledge.
Nagle, R., 2013. Picking up: on the streets and behind the trucks with the sanita-
tion workers of New York City. Macmillan.
Newell, S., 2015. Dirty familiars: colonial encounters in African cities. In Lindner,
C., Meissner, M. (eds.) Global Garbage (pp. 47–63). Routledge.
304 Joost Alleblas and Benjamin Hofbauer

Obersteg, A., Arlati, A., Acke, A., Berruti, G., Czapiewski, K., Dąbrowski, M.,
Heurkens, E., Mezei, C., Palestino, M.F., Varjú, V. and Wójcik, M., 2019. Urban
regions shifting to circular economy: Understanding challenges for new ways of
governance. Urban Planning, 4(3), pp. 19–31.
Raggi, Virginia, interview by Mark Lowen. 2021. Can Rome be rescued from the
rubbish? - BBC News (3 October).
Remøy, H., Wandl, A., Ceric, D. and Timmeren, A.V., 2019. Facilitating circular
economy in urban planning. Urban Planning, 4(3), pp. 1–4.
Roeser, Sabine, and Cain Todd. 2014. Emotion and value. OUP Oxford.
Scanlan, J., 2005. On Garbage. London: Reaktion.
Scanlan, John. 2013. Introduction to Aesthetic Fatigue: Modernity and the
Language of Waste, 2–24, edited by Clark, John F. M., and Scanlan, John.
Cambridge Scholars Publishing.
Schot, J. and Rip, A., 1997. The past and future of constructive technology assess-
ment. Technological Forecasting and Social Change, 54(2–3), pp. 251–268.
Schot, Johan, and Arie Rip. 1997. The past and future of constructive technology
assessment. Technological Forecasting and Social Change, 54(2–3), 251–268.
https://doi.org/10.1016/S0040-1625(96)00180-1
Schumpeter, J. A., 1976. II. Capitalism, Socialism, and Democracy, 1942.
Schumpeter, Joseph A. Capitalism, Socialism, and Democracy: Third Edition.
Harper Perennial Modern Thought, 2008 [1950].
Seale, K., 2015. The paradox of waste: Rio de Janeiro’s Praça XV flea market. In
Global Garbage (pp. 80–91). Routledge.
Schumpeter, Joseph A. 2008. Capitalism, socialism, and democracy: Third edition.
Harper Perennial Modern Thought.
Setoodeh Jahromy, S., Jordan, C., Azam, M., Werner, A., Harasek, M. and Winter,
F., 2019. Fly ash from municipal solid waste incineration as a potential thermo-
chemical energy storage material. Energy & Fuels, 33(7), pp. 5810–5819.
Shonfield, K., 2014. Two architectural projects about purity. In Architecture (pp.
29–43). Routledge.
Sormani, P., Bovet, A. and Strebel, I., 2019. Introduction: When things break
down. In Repair work ethnographies: Revisiting breakdown, relocating materi-
ality (pp. 1–27).
Star, S.L., 1999. The ethnography of infrastructure. American Behavioral Scientist,
43(3), pp. 377–391.
Steele, W. and Legacy, C., 2017. Critical urban infrastructure. Urban Policy and
Research, 35(1), pp. 1–6.
Strasser, S., 2000. Waste and want: A social history of trash. Macmillan.
Taebi, B., Kwakkel, H.J., and Kermisch, C., 2020. Governing Climate Risks in the
Face of Normative Uncertainties. WIREs Climate Change 11 (5): e666. https://
doi.org/10.1002/wcc.666
Taylor, A. (Director), 2008. Examined Life [Documentary].
Taylor, C. 2004. Modern social imaginaries. Duke University Press.
The World Bank. 2022. Urban population (% of total population). Accessed 01-
03-2024. https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS
Thompson, M. 1979. Rubbish theory: The creation and destruction of value.
Oxford University Press, 1979.
 Negotiating Visions of Waste 305

Un-Habitat. Solid waste management in the world’s cities. Un-Habitat, 2010.

Van de Poel, I. 2021. Design for value change. Ethics and Information Technology
23, no. 1: 27–31.
Van de Poel, I., and Kudina, O., 2022. Understanding Technology-Induced Value
Change: a Pragmatist Proposal. Philosophy & Technology 35, no. 2: 1–24.
Van de Poel, I. and Taebi, B., 2022. Value change in energy systems. Science,
Technology, & Human Values, 47(3), pp. 371–379.
Verbeek, P. P., 2011. Moralizing technology: Understanding and designing the mo-
rality of things. Chicago/London: University of Chicago Press.
Walker, Lucy, dir. 2010. Waste Land [Documentary]. Midas Filmes.
Wang, J. 2016. Plastic China [Documentary].
Weinberg, Alvin M. 1966. Can Technology Replace Social Engineering? Bulletin of
the Atomic Scientists 22 (10): 4–8. https://doi.org/10.1080/00963402.1966.
11454993
Young, M. T. 2021a. Maintenance. In The Routledge handbook of the philosophy
of engineering, edited by Michelfelder, Diane P., and Neelke Doorn. (pp. 356–
368). Routledge Handbooks in Philosophy. New York London: Routledge,
Taylor & Francis Group.
Young, M. T. 2021b. Now You See It (Now You Don’t): Users, Maintainers and the
Invisibility of Infrastructure. In Technology and the City, edited by Michael
Nagenborg, Taylor Stone, Margoth González Woge, and Pieter E. Vermaas
(vol. 36, pp. 101–119). Philosophy of Engineering and Technology. Cham: Springer
International Publishing. https://doi.org/10.1007/978-3-030-52313-8_6
13 Repairing AI
Taylor Stone and Aimee van Wynsberghe

13.1 Introduction
This chapter is a call to see our thinking about artificial intelligence (AI)1
as broken and in need of repair. Not literally broken in the sense of no
longer being usable or functioning. Quite to the contrary, we are seeing an
increasingly rapid uptake of innovative applications for AI, especially
with advances in machine learning (ML). Rather, we will argue that
when AI is positioned as an increasingly integral component of digital
infrastructures – and a critical infrastructure itself – there is something
that urgently requires fixing. This fix, however, is not one of efficiency,
usability, transparency or other concerns often debated in AI ethics. Nor
is it only about literal fixes to hardware or software (although how we
approach such acts of maintenance and repair are implicated here). Rather,
it is repairing how we think about AI: as a material infrastructure that is
shaping, and shaped by, its physical environment. AI carries an ecological
footprint and its increasing proliferation, even in the name of laudable
targets such as achieving the Sustainable Development Goals, will have
material impacts that will affect its own ‘operating system’, namely the
built and natural environments in which it is embedded. To understand
and address these material impacts, we need a re-orientation of AI ethics.
This chapter builds directly on calls for a new conceptualisation of the
ethical and political facets of AI (e.g., Crawford, 2021). This re-orientation
requires that we look beyond calculable or checklist-friendly issues such
as explainability, privacy and accountability, and instead, appreciate that
‘… AI applications must not be conceived in isolation but within a larger
network of social and ecological dependencies and relationships’
(Hagendorff, 2022, p. 852).
Repair as a conceptual lens can have far-reaching implications regarding
the re-orientation of AI ethics.2 Here we focus on how the philosophy
of maintenance and repair can help to frame the burgeoning idea of
Sustainable AI and the intertwined facets of utilising AI for sustainable

DOI: 10.4324/9781003316213-15
 Repairing AI 307

ends and appreciating AI as an infrastructure that itself needs to be sustain-

ably designed, maintained and decommissioned. Towards this goal, the
chapter will build on the recent distinction made by van Wynsberghe (2021)
between AI for sustainability and the sustainability of AI (Figure 13.1). The
former utilises or develops AI in the service of sustainable goals; the latter
concerns the sustainability of AI systems themselves. Building on the two-
part definition of Sustainable AI, this distinction will be expanded to ex-
plore Sustainable AI as acts of repair and repairing AI. The former concerns
the practical use of AI for the monitoring, upkeep and adaptation of critical
infrastructures, whereas the latter requires conceptualising AI as an in-
creasingly critical urban infrastructure itself requiring care and repair.
Through this lens, AI can be considered ‘broken’ in two ways. First, as be-
ing an increasingly unsustainable and unjust infrastructure that is itself a
cause of the environmental degradation which it is attempting to mitigate.
Second, as being conceptualised as an externality, or worse a ‘cloud’, that
is not itself embedded within and reliant upon the material world. We will
therefore argue that you cannot have AI for sustainability without first ad-
dressing the sustainability of AI.
As AI is increasingly used for achieving the Sustainable Development
Goals (i.e., AI for sustainability), the upkeep of those systems becomes in-
creasingly critical (i.e., the ongoing repair and maintenance of AI). This
becomes especially clear as AI-enabled systems and artefacts are deployed
as solutions to the pivotal challenges of the twenty-first century, including
combating climate change, preserving biodiversity and managing unprece-
dented urbanisation (Rolnick et al., 2022). While this can include efforts
focused on efficiency and optimisation (e.g., smart energy grids), we are also
seeing innovations utilising AI for the management and monitoring of natu-
ral areas and/or ‘natural infrastructures’ (McDonald and Beatley, 2021). In
these instances, AI becomes integral to the management and monitoring of
wilderness, wildlife or (urban) ecosystems and a core driver of restoration

Sustainable AI

AI for sustainability Sustainability of AI

Figure 13.1 Sustainable AI as AI for sustainability and the sustainability of AI.

Adapted from van Wynsberghe (2021).
308 Taylor Stone and Aimee van Wynsberghe

and rewilding strategies (Cantrell et al., 2017). We return throughout to the

case study of utilising AI for ecological restoration efforts – as an example
of Sustainable AI as an act of repair, broadly construed.
The focus of this chapter is the interrelated notions of maintenance and
repair and their relevance to AI ethics. Maintenance is typically framed as
an activity through which we provide stability to something over time,
although more recent scholarship also emphasises a process-oriented view
that highlights the changes to artefacts and infrastructures such practices
facilitate (Young, 2021a). This leads towards questions of repair, which
can be defined as ‘the subtle acts of care by which order and meaning in
complex sociotechnical systems are maintained and transformed, human
value is preserved and extended, and the complicated work of fitting to
the varied circumstances of organizations, systems, and lives is accom-
plished’ (Jackson, 2014, p. 222). What will be emphasised throughout is
the layered meaning of repair, namely both its technical and social
functions – something critical when considering the convergence of AI and
sustainability. We thus follow Henke and Sims (2020, p. 3), who situate
the process of repair – and in particular repairing infrastructures – as one
of ‘restoring both social and material order’. In this sense, repairing an
artefact or system is not only an act of restoring functionality but also
often serves to repair social relations and values. For our purposes, we
build on this conceptualisation to delineate two levels of repair: as the
literal upkeep of the hardware and software of AI and as a conceptual lens
for thinking about the trajectory of AI as a material infrastructure. We
will show that this second, conceptual level is critical and must accom-
pany any literal interventions in AI (development, design, maintenance or
decommissioning), if we are to make significant progress towards achiev-
ing Sustainable AI.
A focus on repair expands existing scholarship within the ethics of AI,
where the design stage is largely prioritised. This comes at the cost of un-
derstanding how technologies change and evolve through time and the
processes and practices through which this change occurs. As we will dis-
cuss, the continual process of (almost) breakdown and repair can serve to
reinforce the status quo or provide opportunities to re-consider the envi-
ronmental impacts of AI. Situating AI as a material infrastructure offers a
different ontological perspective, emphasising the dynamic, complex and
evolving nature of this system. Combined with the literal upkeep and de-
velopments to hardware and software, this reminds us that AI is never
‘complete’ and can never reach a static end goal of ‘sustainable’. Rather, it
is intertwined with the systems, structures and environments it will
impact – necessitating a reflexive approach to the relationship between
sustainability and AI (see Henke and Sims 2020).
 Repairing AI 309

Maintenance
Conservation of current
structures, practices

Repair Repair as:

material + social function

Transformation
Changing of current
structures, practices

Figure 13.2 Two potential targets for acts of repair.

Adapted from Repairing Infrastructures (Henke & Sims, 2020).

Following Jackson’s notion of ‘broken world thinking’ (2014), we

should remember that such a perspective is not just a pessimistic outlook,
but is also hopeful in appreciating the potential for change carried within
these acts of maintenance and repair. It is these very processes and prac-
tices that can offer a reorientation of AI. For this, the contrasting notions
of repair as maintenance and repair as transformation as proposed by
Henke and Sims (2020) are applied and extended to the topic of AI. Rather
than striving to maintain current structures and relations, repair as trans-
formation instigates processes that go beyond the upkeep and conservation
of current practices and structures, instead spurring deliberate change
(Figure 13.2). Within transformative repair exists the possibility to ques-
tion and re-think the physical structures that infrastructures support.
Taking a reflexive and interactional approach to the relation between AI
for sustainability and the sustainability of AI, it must be appreciated that
the innovative uses of AI for goals such as ecological restoration must oc-
cur in tandem with a transformation of the digital infrastructures, power
dynamics and sociopolitical systems that shape this technology.

13.2 Sustainable AI
With the growing number of AI/ML models in use across the globe, the
ethics of AI has emerged as a transdisciplinary area of research at the inter-
section of philosophy, policy, innovation and practice to address the po-
tential for negative outcomes associated with the pervasiveness of AI/ML
applications. The field of AI ethics has observed trends in analysis from
dystopian scenarios in which AI is thought to be the end of mankind to AI
and robots taking all our jobs (see Coeckelbergh, 2020, 2021; van
Wynsberghe, 2021). More recent trends have focused on how to design AI
310 Taylor Stone and Aimee van Wynsberghe

in ‘value sensitive’ ways that mitigate specific sociopolitical concerns raised

from one application to another – for example, developing frameworks
and policies promoting positive goals such as fairness and accountability
or aiming to mitigate privacy infringements or the reinforcement of harm-
ful biases in data sets. While these sociopolitical issues continue to domi-
nate discourse in/on AI ethics, systemic and structural concerns have
more recently been given increasing attention (e.g., Bolte, et al., 2022;
Coeckelbergh, 2021; Crawford, 2021; Hagendorff, 2022; Kaack, et al.,
2022; Strubell et al., 2019). Practical issues have been raised regarding
energy usage for data processing, data storage and the energy required
(with resulting carbon emissions) for training AI, particularly for large ML
algorithms. Further, often-hidden issues such as mining for rare minerals
raise concerns over resource depletion and the life cycle of electronic waste.
This can lead to political contestations over environmental justice, as well
as individual autonomy and freedom. As a result, scholars have urged a
shift in focus within AI ethics towards environmental justice issues, which
is referred to broadly as Sustainable AI (van Wynsberghe, 2021).
To be sure, impacts such as racial discrimination resulting from the un-
questioned use of historical data sets in training models are very real and
very dangerous risks; however, they are the risks for those exposed to AI as
a product. They are the risks for a society that believes they have, or should
have, power over the way in which technologies are made and used. They
are not the risks raised by those who are further up in the supply chain of
AI. The concern above is not (yet) a concern for those working in the mines
to procure the minerals necessary for the physical infrastructure of AI nor
is it a concern for those who have electronic waste mountains in their
backyards. The field of Sustainable AI is meant to target the latter set of
concerns and ask: ‘What is it that we, as a global community, want to
sustain?’. This is a descriptive question in so far as we must know the ac-
tual environmental damages caused by the infrastructure of AI and norma-
tive in so far as we must be willing to suggest what are (in)appropriate
consequences.
Sustainable AI has been defined by van Wynsberghe (2021) as a move-
ment towards greater ecological integrity and social justice throughout the
design process of AI. In focusing on the interrelations and interconnections
between the use of AI for sustainable ends and the sustainability of devel-
oping and using AI in and of itself, Sustainable AI presents a kind of dual-
ity to the discussion. It is not possible, nor desirable, to look at only the
positive or negative aspects of AI; rather, they are two sides of the same
coin. For example, in a country where mining is a main source of income
for millions of families, the social structure of entire villages revolves
around the practice of mining. Moreover, the social conditions of life (clean
drinking water, contamination of land) are impacted by these practices.
 Repairing AI 311

Thus, reframing AI ethics to focus on issues such as ecological impacts,

energy usage and environmental justice forces one to dive deeper into the
hidden costs and hidden demographics.
As the literature on Sustainable AI grows, we see a refining of the con-
cept. Recently, it has been argued that the very notion of sustainability –
when applied to something as complex as AI – must be understood as a
dynamic property of systems (Bolte et al., 2022). Crucially, this requires
conceiving of AI as a key node of our digital infrastructures and novel in-
frastructure itself, that is, AI will use existing infrastructures to function
and will itself also become an infrastructure upon which businesses, public
services and individuals will rely (see Robbins and van Wynsberghe, 2022).
Recognising AI as an infrastructure calls upon us to investigate the main-
tenance of said infrastructures. What needs to be maintained, is it possible
to maintain all aspects, and critically what needs to change? In what fol-
lows, we expand on the ideas proposed in the abovementioned work,
namely that Sustainable AI, as a conceptual framework, can provide tools
that AI ethics alone cannot or has not. In particular, we will show how an
examination of Sustainable AI through the lens of repair allows for a fur-
ther elucidation of AI as a dynamic and embedded infrastructure and high-
lights a set of challenges and opportunities for the future of AI.

13.3 Maintenance, Repair and Sustainable AI

Ethical analyses of technology are often focused on the design phase, over-
looking how artefacts and systems change, adapt and evolve over time. For
example, value sensitive design has developed a series of methods for iden-
tifying and incorporating moral values into the development and design of
(novel) technologies (Friedman and Hendry, 2019). While a useful ad-
vancement for the ethics of technology, there is an implicit assumption
that, if successful, moral values will become embedded in the artefact in
question. A specific ontological view of technology is presumed, in which
artefacts have a set of stable characteristics (Young, 2021b). However,
there are many critiques to this assumption, including the ‘multistability’
of objects (Albrechtslund, 2007), concerning the disjunction between the
design and use phases of an artefact, as well as the notion of value change
and its effect on the evaluation and use of technologies (van de Poel, 2021).
As van de Poel (2020) points out, we should avoid overly determinist and
constructivist perspectives but rather appreciate that technologies such as
AI will be shaped by the co-evolution of technology and society. This re-
quires an abandonment of completeness as a substantive goal for value-
sensitive analyses of technology, instead appreciating that any interventions
will change the relative meaning, importance and priority of the values in
question (Stone, 2021).
312 Taylor Stone and Aimee van Wynsberghe

Yet, with a complex and dynamic technology such as AI, we should not
presume that this interaction ends with the design phase. Where van de
Poel discusses the possibility of unanticipated consequences and changing
evaluative judgements as drivers of change, he does not elaborate on the
practices and processes through which this coevolution occurs. Any inno-
vation must necessarily be sustained through repair, and likewise, repair
and maintenance become sites of creativity and innovation (Jackson, 2014).
Further, as Graham and Thrift (2007) point out, what begins as ‘repair’
may quickly evolve into ‘improvement’ or ‘innovation’. To look beyond
the design phase is to appreciate repair as a key mechanism for society–
technology interaction and co-evolution, and as a moment where the social
and environmental impacts of AI can either be challenged or reified.
Before moving on to discuss repair as a driver of the co-evolution of AI
and society, it is first useful to offer a brief clarification and general defini-
tion. By way of clarification, it is important to appreciate that repair as
discussed here is not only about moments of complete breakdown. There
is a tendency within the study of infrastructures to over-emphasise mo-
ments of breakdown, as this makes their use and function temporarily
visible (Jackson, 2014; Star, 1999). For example, our infrastructure of
artificial lighting, while formative to – and ubiquitous throughout – our
world after dark, readily fades into the background of lived experience. It
is largely unnoticed until moments of breakdown occur in the form of
blackouts, and we are temporarily plunged into what feels like an ‘artifi-
cial’ darkness (Nye, 2010). Through these experiences, the invisibility of
infrastructure is temporarily disrupted, and they are momentarily brought
into the spotlight. Yet, a focus on such (often sensational or dramatic) mo-
ments overlooks the constant upkeep, monitoring and incremental changes
that are critical to the longevity and functionality of infrastructures
(Young, 2021b). Hence why Jackson (2014) situates broken world think-
ing at the generative and continual interface between breaking down and
restoring.
How, then, to approach repair as a constant and critical – if taken for
granted – practice? As a first step, it is useful to understand the relation
between repair and maintenance, conceptualised as ‘an activity through
which we provide stability to something over time’ (Young, 2021a, p. 356).
However, what exactly this stability entails, as Young points out, is an
open question requiring philosophical analysis – and, one particularly rel-
evant to discourse surrounding sustainability. A common view of mainte-
nance is as an act of conservation (Young, 2021a), in the sense of acting
against the passage of time; working to combat the inevitable processes
of breakdown, erosion, corrosion, decay, malfunction, obsolescence, etc.
This aligns with the conservative notion of repair as maintenance
 Repairing AI 313

(Henke and Sims, 2020), aiming to maintain existing infrastructures and

relations and to upkeep systems and artefacts in largely the same way as
they currently operate.
While often focused on the restoration and preservation of existing sys-
tems and relations, maintenance and repair can also instantiate processes
of change and evolution through adjustment, tweaking, adaptation or
more radical changes to systems and artefacts. This highlights the temporal
and dynamic nature of technology, as processes extended and guided
through time. Put otherwise, it asks us to appreciate that no technology or
system is completely static, but necessarily evolves and changes. The re-
placement of brakes and tyres on your bicycle, the rust growing on tools
left outside and the need for software updates on computers all remind us
that artefacts are not static. Such a perspective blurs the boundary between
the ‘designing/making’ and ‘using’ phases of a technology. Instead, we see
the continuous processes of adaptation and change via external influences
and deliberate choices as fundamental to the interaction of artefacts with
their world (Young, 2021a, 2021b). This is a useful reminder that all arte-
facts are dynamic and necessarily interact with the world in some ways,
and the results of these interactions often necessitate maintenance and re-
pair. Deliberate acts of repair, however, need not only be concerned with
maintaining the status quo. Henke and Sims (2020) contrast repair as
maintenance with the more ambitious and idealistic notion of repair as
transformation. While the former aims to maintain current structures and
relations (both social and material), repair as transformation instigates a
process of repair that goes beyond the conservation of current structures,
instead striving for change. Within transformative repair exists the possi-
bility to question and re-think the physical and social structures that infra-
structures support, including those related to environmental justice and
power relations.
What then does a focus on repair offer to the nascent concept of
Sustainable AI? And in particular, what does a process-oriented, transfor-
mative perspective on the potential of repair surface regarding the ethical
framing of (Sustainable) AI? To analyse AI through this lens, we propose
to extend and expand the overarching framework proposed by van
Wynsberghe (2021). We will argue that repair is where AI for sustainabil-
ity and the sustainability of AI interact (Figure 13.3) – at times literally as
implied above, but critically also in how we position the relationship be-
tween AI and sustainability. Practically, we cannot reap the full benefits of
utilising AI for sustainable goals without transforming the infrastructure
of AI. Conceptually, this requires reflexivity as a central consideration for
the ethical impacts of AI, particularly in how we choose to maintain those
systems which are themselves utilised for sustainable applications.
314 Taylor Stone and Aimee van Wynsberghe

Sustainable AI

AI for Sustainability Repair Sustainability of AI

AI used for acts of repair material + social function AI as target of repair

Figure 13.3 Repair as the interface between AI for sustainability and the sustain-
ability of AI.

13.3.1 AI for Sustainability as Acts of Repair

Considered through the lens of repair, AI for sustainability concerns the

practical use of AI-enabled innovations for the monitoring, upkeep and
optimisation of critical infrastructures. This can include the use of AI-
enabled services and autonomous robots to better organise mundane but
necessary maintenance, such as road repair (Mukherjee et al., 2021) and
garbage collection (Zhang et al., 2020). ML techniques are also being
developed for preventative actions, for example, the World Bank’s project
to identify vulnerable housing in the Global South (Wahba, n.d.). ML re-
search has also focused on predictive maintenance of digital infrastruc-
tures via autonomous systems (Khan et al., 2020). Here, AI is a tool
to improve the practices of maintenance – either via providing stability
through continual and efficient upkeep or as a driver of the process of
change via monitoring and intervention or by serving to adapt artefacts
over their lifespan.
AI can be utilised for a variety of tasks falling under the broad categori-
sation of maintenance, but here we are particularly interested in how the
application of AI for sustainability instantiates moments of repair. We note
that, arguably, many acts of maintenance and repair aimed at optimising a
system or gaining efficiencies (in the sense of reducing energy demands or
materials used, extending lifetimes of a product, etc.) can fall under an
instrumental notion of sustainability. However, here we will look more
specifically at how Sustainable AI can be seen as acts of repair as transfor­
mation. Henke and Sims (2020), in discussing the IPCC reports, describe
them as effectively being ‘massive planetary repair’ manuals (p. 140).
Building on this notion, we can see that many of the applications and ini-
tiatives that utilise AI towards environmental ends are essentially attempts
to enact planetary repair on some scale.3 This arguably also underlies in-
creasing calls for interdisciplinary approaches to climate change mitigation
 Repairing AI 315

and adaptation with the aid of ML (e.g., Rolnick et al., 2022), including
the initiatives of organisations and movements such as AI for Good (aifor-
good.itu.int), Climate Change AI (www.climatechange.ai), Microsoft’s AI
for Earth (www.microsoft.com/en-us/ai/ai-for-earth) and Global Forest
Watch (globalforestwatch.org).
Within and beyond the work of these organisations, an increasing num-
ber of projects and initiatives take aim at more specific problems, such as
biodiversity preservation and ecological restoration. While this may ex-
pand the notion of repair as typically used in infrastructure studies, it does
align with Henke and Sims’ (2020, p. 3) definition of repair as acts of ‘re-
storing both social and material order’. In this sense, repairing an artefact
or system is not only an act of restoring functionality but also often serves
to repair social relations and values. Utilising AI for ecological restoration
and preservation then further extends repair to also include environmental
relations and values. As Dauvergne discusses in AI in the Wild (2020),
there are already many such initiatives underway. This includes the use of
ML for the monitoring of threatened or endangered areas, such as tracking
illegal fishing boats via remote sensing to see where overfishing is taking
place, monitoring the depletion of rainforests or tracking the movement of
wildlife – a tool also increasingly employed by groups such as the WWF
(World Wildlife Fund, 2020). It also includes various interventions, such as
re-seeding areas using drones and culling invasive species along the Great
Barrier Reef using semi-autonomous submarines.
Monitoring and interventions have also been proposed for urban areas
and the assistance of urban rewilding efforts. This may open up new uses
for ML, with ramifications for urban planning and design – for example,
Owens and Wolch (2019) discuss how urban wildlife can be tracked and
monitored via camera traps with deep learning pattern recognition, to-
wards better understanding their movement and the creation of wildlife
corridors. More speculatively, they discuss how autonomous vehicles
could serve to reduce roadkill deaths. This could be coupled with other
design directions for autonomous vehicles, such as a reduction in road
lighting (Stone et al., 2020) towards the preservation of nocturnal habitats.
As another example, there are proposals to create an Internet of Nature
through which embedded sensors, urban robots, remote sensing and AI
algorithms work together to monitor and improve the health of urban
forests (Galle et al., 2019; Galle and Nitoslawski, 2020).
The main takeaway from these varied examples is that, at its core, the
use of AI for sustainability is in many ways acts of repair; of aiming AI
innovation towards the target of planetary repair. This necessitates a con-
scientious reflection on the ethics and politics of said initiatives. First, are
316 Taylor Stone and Aimee van Wynsberghe

these applications truly instantiating instances of repair as transforma­

tion or serving to reinforce existing power structures and exploitative
practices (Dauvergne, 2020)? For example, the proposed Internet of
Nature mentioned above, as a type of programmed green infrastructure,
has been critiqued for perpetuating existing power imbalances along with
socially and environmentally unsustainable practices (Gabrys, 2022). A
second question to ask is how exactly sustainability is framed and
bounded in these contexts. Improving the efficiency of current systems is
undoubtedly desirable and worthwhile in many contexts, and necessary
given that we have an urgent need to reduce greenhouse gas emissions.
However, to simply optimise the systems, infrastructures and power rela-
tions that helped lead us into our current crises can hardly be considered
sustainable in any meaningful sense. Granted, an Internet of Nature
could lead to profoundly positive changes for urban ecosystems and wild-
life, as well as the well-being of humans living in those cities. Yet, the full
extent of the social and ecological impacts of the infrastructure that
would power this system must also be considered. Put otherwise, there is
a need to ask how to repair the very systems we are utilising to enact
planetary repair.

13.3.2 Repairing AI

The above discussion highlights the opportunities and challenges created

by the utilisation of AI for sustainable goals. However, this is only one side
of the coin. The sustainability of AI requires a conceptualisation of AI as a
critical infrastructure itself (Crawford, 2021; Robbins and van Wynsberghe,
2022). The process-oriented perspective on technologies (Young, 2021a)
reminds us that AI artefacts are not stable but will necessarily adapt, evolve
and ultimately be decommissioned and demolished. This draws attention
to the adaptability, vulnerabilities and ongoing upkeep of both the digital
and physical infrastructures underlying AI systems. Equal consideration
must be given to critical material components such as IoT devices, data
centres, transcontinental cables and satellites alongside codes and algo-
rithms. Yet these considerations cannot be viewed in isolation from their
uses and applications, especially in the name of sustainability. We must
appreciate that AI does not actually operate within a disembodied ‘cloud’
but rather within material artefacts and infrastructures that themselves re-
quire monitoring, upkeep, replacement, etc. In our ethical analyses of AI,
we too readily forget that AI includes physical components and human
labour. This complicates the analysis of AI as a discrete category rather
than one closely intertwined with larger digital infrastructures and reliant
upon seemingly external critical infrastructures (e.g., energy systems). This
requires an ethics of (Sustainable) AI that is material – in the sense of
 Repairing AI 317

interrogating physical artefacts and effects. This fact has been emphasised
in infrastructure studies but seemingly overlooked in AI ethics:

… it is by concentrating on the material architectures of electricity

generation and supply that, in many ways, we can begin to ‘see’
‘cyberspace’ for what it is – not an ethereal domain of ‘virtual’ bits
and bytes, but a gigantic, materialized and electrically powered system
requiring massive amounts of continuous and concerted maintenance
and repair.
(Graham and Thrift, 2007, p. 13)

This integration of digital infrastructures brings with it vulnerabilities and

cascading effects that need to be considered, and which fall under the um-
brella of repair.
As an example of the material presence of digital infrastructure, consider
data centres. Due to issues of security and competition, companies actively
strive to keep data centres – the material home of ‘the cloud’ – largely in-
visible. This also makes them difficult to research. The companies that
have released images of their data centres often present staged pictures in
which massive buildings are seemingly in harmony with their surrounding
environment (Holt and Vonderau, 2015). Yet, this overlooks both the po-
litical and environmental impacts of these material homes of our digital
world. A recent report from the International Energy Agency (IEA, 2022)
estimates that data centres are responsible for 1–1.5% of global electricity
use and approximately 0.6% of global greenhouse gas emissions. Likely,
only a fraction of this is directly attributable to AI; however, the exact
amount is not known due to lack of data and challenges in identifying the
boundaries of AI uses (Kaack et al., 2022). Data centre energy demands
are projected to increase rapidly in the coming years, as part of more wor-
rying energy forecasts for ICT generally. Some models predict ICT-related
emissions could jump from about 2% to over 20% within a generation
(Jones, 2018). To put into perspective, in 2010 it was estimated that if ‘the
cloud’ were a country, it would have the fifth largest energy consumption
in the world (Greenpeace 2012; Holt and Vonderau, 2015). While energy
required for computing power is a major issue, again the material upkeep
of data centres requires broader consideration: air conditioning can ac-
count for up to 40% of data centre energy usage. This has geographical
and political implications for data centres as they are often situated at sites
with cheap electricity and where climatic conditions offer ‘free’ cooling
(Holt and Vonderau, 2015).
The above numbers are indicative of the sheer scale of the materiality
required for our current digital infrastructure. Again, we must begin to see
critical nodes such as data centres not as static artefacts, but as complex
318 Taylor Stone and Aimee van Wynsberghe

arrangements of servers and cables housed within specialised buildings –

which will necessarily undergo moments of breakdown and repair. Such
a perspective necessitates a shift in how we approach the ethics and poli-
tics of AI. It is imperative that the ongoing upkeep and required mainte-
nance of data centres, including the servers themselves and the surrounding
built environment (e.g., cooling and other features of the building), are
included in a holistic conception of Sustainable AI specifically, and AI
ethics generally. These are nodes within a complex and dynamic system.
And if we indeed see the system supporting AI as in a constant state of
(almost) breakdown and repair rather than stable, we can begin to ap-
preciate that the maintenance of AI’s material components is critical to
any conception of Sustainable AI. While the increasing material impact of
digital infrastructure can be concerning, this is not meant to be a condem-
nation of all uses and applications of AI. Indeed, many – including some
examples of ecological restoration mentioned above – may very well be
desirable despite their energy or material demands. However, the more
we rely upon AI for sustainability, the more critical the sustainability of
AI becomes.
Yet, there is also a more profound implication underlying the increasing
concerns regarding the energy demands of AI and associated digital infra-
structures. It is headline-grabbing to point towards these quantifiable and
growing numbers, which can indeed serve as useful in calls for greater
awareness, new policies focused on reductions in energy usage, etc. However,
this focuses solely on the literal notions of maintenance and repair as de-
fined above. As Kaack et al. (2022) emphasise, it is the system-level impacts
that are both harder to quantify and predict and that will likely have larger
impacts on critical challenges such as greenhouse gas emissions. Kaack et al.
mention important considerations such as rebound effects and the risks of
‘lock-in’, which point towards the need for proactive and anticipatory ac-
tions. Because of this, the processes and practices that sustain our AI infra-
structure cannot be seen in isolation from the downstream uses and
applications, for this reinforces dualistic thinking and overlooks the mate-
rial embeddedness of AI. This serves as a useful reminder that the environ-
mental problems we currently face are not ‘out there’ to be addressed and
solved, but a product of our own choices and actions.

It is not a question of our encountering the crisis and resolving it through

technology. The crisis is not simply something we can examine and re-
solve. We are the environmental crisis. The crisis is a visible manifesta-
tion of our very being… The environmental crisis is inherent in everything
we believe and do; it is inherent in the context of our lives.
(Evernden, 1993, p. 128)
 Repairing AI 319

Accepting this means we must then see AI and digital infrastructure as in-
separable from the problems they are purporting to solve.
This is not just a practical necessity, but one with profound potential
to re-shape our digital infrastructure. Here, we again invoke the idea
from Henke and Sims (2020) that repair has a material and social func-
tion. Consider again the example of the Internet of Nature briefly men-
tioned above (Galle et al., 2019). This can be seen as an innovative (and
in many ways promising) use case of urban AI for ecological restoration,
utilising data analytics, predictive and real-time monitoring and urban
robotics to improve urban ecosystems, biodiversity, the health and well-
being of citizens, etc. But it is also introducing a new layer of infrastruc-
ture to the already complex urban fabric – it will necessarily respond to,
and in turn impact, the urban environments in which it is deployed.
Further, its successful integration into a city will make it hard to define
exactly where the boundaries of the Internet of Nature lie. This makes
quantifying the exact impacts of such interventions difficult (e.g. in at-
tempting any sort of cost–benefit analysis that compares energy and ma-
terial demands with local benefits). It will also engender a social change,
affecting the relationship between ‘natural’ and ‘artificial’ components
of a city. These changes could indeed offer a desirable urban future;
the point is that the ethical questions that arise extend well beyond is-
sues such as explainability or transparency, or even quantifying energy
usage.
Through such an example, we are forced to confront the reality that we
must continually reflect upon the systems that are themselves increasingly
being used for planetary repair. This highlights a paradoxical feature of
repairing infrastructure during the Anthropocene, that ‘we cannot de-
velop data and theories about “the world” without the very systems that
are creating wide-scale change’ (Henke and Sims, 2020, p. 122). To con-
front our current situation, we need to think about what Henke and Sims
call ‘repairing repair’, or ‘… a reorientation of repair itself, toward an
ethic that considers the consequences of infrastructural globalism through
the complex interactions of power, scale, and time that are built into the
sociotechnical structures of modernity’ (p. 137). Here, repair can be con-
sidered as a reflexive act on two levels: in a causal sense (i.e., that the op-
eration of infrastructures cause impacts that (re-)shape their own operating
environment) and in the sense of self-aware thinking (i.e., of those design-
ing and managing the systems to appreciate potentially negative feedback
loops). Put otherwise, to adopt a repair as transformation approach to
Sustainable AI, we need a reflexive and iterative approach to how the sus-
tainability of AI affects the efficacy of AI for sustainability, and vice versa
(Figure 13.4).
320 Taylor Stone and Aimee van Wynsberghe

Sustainable AI

Sustainability of AI
AI as target of repair

Repair as transformation

AI for Sustainability
AI used for acts of repair

Repair as Repair as
maintenance transformation

Figure 13.4 Repair as reorienting the relationship between AI for sustainability

and the sustainability of AI. While the use of AI for sustainability can
target maintenance or transformative repair, it relies on a transforma-
tive approach to the repair of AI and associated digital infrastructures.
Once a transformative approach is initiated, AI-enabled innovations
can themselves be used to support and improve the sustainability of AI
itself.

13.4 Conclusion
This chapter outlines a research agenda that takes AI ethics out of the
cloud and into the world of material things – as a tool to repair our mate-
rial world, and likewise as a material infrastructure itself.
For this, it investigated the notion of repair and its usefulness for think-
ing through the conceptualisation of Sustainable AI. It asks us to under-
stand technologies not as stable entities, but rather as evolving and dynamic
processes. Because of this, the constant upkeep of technologies – and the
resulting change and adaptation this instigates – are central considerations
for the ethics of AI. Repair should be understood to have both a material
and social function, with the potential to either maintain and reinforce the
status quo or to become a means of transforming AI in more sustainable
and socially just directions. This requires understanding the notion of re-
pairing AI as having two interrelated meanings: the literal upkeep of AI
 Repairing AI 321

infrastructure and a conceptualisation of AI as embedded within its own

operating system. The latter opens a necessary shift in thinking about the
ethics of AI that focuses on how it shapes – and in turn is shaped by – the
material environments in which it operates. This layered notion of repair-
ing AI can advance the nascent notion of Sustainable AI specifically and
support more general calls for a re-orientation of AI ethics towards sys-
temic and structural issues. Building on the definition of Sustainable AI as
having two branches, namely the application of AI for sustainability and
the sustainability of AI itself (van Wynsberghe, 2021), we have shown that
repair is a moment where the of and for come together. You cannot have
the for without the of; the benefits of utilising AI for sustainability cannot
be fully realised without also repairing the digital and material infrastruc-
ture of AI itself. This serves as a stark reminder that the tools we increas-
ingly use for planetary repair are themselves embedded within their own
operating systems. Overlooking this important reality risks reinforcing a
dualistic mode of thinking that has arguably contributed to the environ-
mental crises we now face.

Notes
1 According to the European Commission’s High-Level Expert Group on
Artificial Intelligence (2019, p. 1): ‘Artificial intelligence (AI) refers to systems
that display intelligent behaviour by analysing their environment and taking
actions – with some degree of autonomy – to achieve specific goals’. Following
from this, there are a range of methods in which this so-called intelligence can
be achieved from symbolic reasoning to machine learning (ML) and neural
networks, to name a few. Here, we will not go further than this broad defini-
tion when referring to AI or ML, as this chapter is not directly addressing the
development or use of specific models but rather our framing of this rapidly
proliferating technology – particularly how we situate its material processes,
structures and impacts.
2 Other sociopolitical aspects of AI could also be assessed through the concep-
tual lens of repair, such as the exploitation of workers, threats to democracy
or the treatment of animals (e.g., Crawford, 2021; Coeckelbergh, 2021;
Hagendorff, 2022). We acknowledge these social dimensions and see usefulness
in the framing of AI developed here for these impacts and ethical problems, but
will focus explicitly on AI’s environmental and material impact – using the re-
current case study of AI for ecological restoration – to provide an initial sketch
and framing of how ‘repairing AI’ can advance and extend Sustainable AI.
3 The concept of planetary repair has previously been used to describe ‘… a con-
ceptual provocation for an era of global environmental destabilization and as a
conflictual turn in mainstream conservation practice – particularly as technofu-
turist visions of ecological design and geoengineering gain new adherents’
(Knuth, 2019, p. 488). In this context, however, we draw this term from the
discussion of reflexive repair in the Anthropocene by Henke and Sims (2020,
chapter 5). Practically, it reflects contemporary thinking around ecological res-
toration as a means of repairing the planet, required alongside sustainable tran-
sitions of socio-technical systems (Cooke et al., 2023) – an idea arguably
322 Taylor Stone and Aimee van Wynsberghe

reflected in the AI-focused initiatives and organisations referenced above. We

acknowledge that this term can be suggestive of viewing Earth as an artefact.
While an in-depth discussion of this issue is outside the scope of this chapter,
the terminology of planetary repair is meant to evoke critiques of nature–
artefact dualistic thinking generally (Vogel, 2015), and specifically the need to
better appreciate the material impacts of digitalisation on the Anthropocene
(Creutzig et al., 2022). Further, it can be seen as a response to recent calls to
question our ontological categories of ‘nature’ and ‘technology’ as they become
increasingly entangled in the Anthropocene, and the need for philosophers of
technology to adopt a ‘planetary horizon’ given that ‘… technology is not just
embedded in ecology anymore but is increasingly becoming its very founda-
tion, such that the future habitability of the Earth is becoming the technological
question par excellence’ (Lemmens et al., 2017, p. 124).

References
Albrechtslund, A. (2007) ‘Ethics and technology design’, Ethics and Information
Technology, 9(1), pp. 63–72. Available at: https://doi.org/10.1007/s10676-006-
9129-8
Bolte, L., Vandemeulebroucke, T. and van Wynsberghe, A. (2022) ‘From an ethics
of carefulness to an ethics of desirability: going beyond current ethics approaches
to sustainable AI’, Sustainability, 14(8), p. 4472. Available at: https://doi.org/
10.3390/su14084472
Cantrell, B., Martin, L.J. and Ellis, E.C. (2017) ‘Designing autonomy: opportuni-
ties for new wildness in the Anthropocene’, Trends in Ecology & Evolution,
32(3), pp. 156–166. Available at: https://doi.org/10.1016/j.tree.2016.12.004
Coeckelbergh, M. (2020) AI Ethics. MIT Press.
Coeckelbergh, M. (2021) Green Leviathan or the Poetics of Political Liberty:
Navigating Freedom in the Age of Climate Change and Artificial Intelligence.
Routledge. Available at: https://www.routledge.com/Green-Leviathan-or-the-
Poetics-of-Political-Liberty-Navigating-Freedom/Coeckelbergh/p/book/
9780367747794 (Accessed: 12 July 2022).
Cooke, S.J. et al. (2023) ‘Ecological Restoration in Support of Sustainability
Transitions: Repairing the Planet in the Anthropocene’, in M.U. Hensel et al.
(eds) Introduction to Designing Environments: Paradigms & Approaches. Cham:
Springer International Publishing (Designing Environments), pp. 93–112.
Available at: https://doi.org/10.1007/978-3-031-34378-0_6
Crawford, K. (2021) Atlas of AI. New Haven: Yale University Press.
Creutzig, F. et al. (2022) ‘Digitalization and the Anthropocene’, Annual Review of
Environment and Resources, 47(1), pp. 479–509. Available at: https://doi.
org/10.1146/annurev-environ-120920-100056
Dauvergne, P. (2020) AI in the Wild: Sustainability in the Age of Artificial
Intelligence. Cambridge: The MIT Press.
European Commission’s High-Level Expert Group on Artificial Intelligence (June
2019) ‘A definition of AI: Main capabilities and scientific disciplines’, European
Commission. Available at: https://digital-strategy.ec.europa.eu/en/library/definition-
artificial-intelligence-main-capabilities-and-scientific-disciplines (Accessed 5 March
2022).
 Repairing AI 323

Evernden, L.L.N. (1993) The Natural Alien: Humankind and Environment.

University of Toronto Press.
Friedman, B. and Hendry, D. (2019) Value Sensitive Design: Shaping Technology
with Moral Imagination. Cambridge: The MIT Press.
Gabrys, J. (2022) ‘Programming nature as infrastructure in the smart forest city’,
Journal of Urban Technology, 29(1), pp. 13–19. Available at: https://doi.org/
10.1080/10630732.2021.2004067
Galle, N.J. and Nitoslawski, S.A. (2020) ‘Biophilic Cities for the 22nd Century’,
Biophilic Cities Journal, 3(2), pp. 12–21.
Galle, N.J., Nitoslawski, S.A. and Pilla, F. (2019) ‘The Internet of Nature: How
taking nature online can shape urban ecosystems’, The Anthropocene Review,
6(3), pp. 279–287. Available at: https://doi.org/10.1177/2053019619877103
Graham, S. and Thrift, N. (2007) ‘Out of order: Understanding repair and mainte-
nance’, Theory, Culture & Society, 24(3), pp. 1–25. Available at: https://doi.
org/10.1177/0263276407075954
Greenpeace (April 2012) ‘How Clean is Your cloud?’, Greenpeace International.
Available at: https://www.greenpeace.org/static/planet4-international-stateless/
2012/04/e7c8ff21-howcleanisyourcloud.pdf (Access 5 March 2022).
Hagendorff, T. (2022) ‘Blind spots in AI ethics’, AI and Ethics, 2(4), pp. 851–867.
Available at: https://doi.org/10.1007/s43681-021-00122-8
Henke, C.R. and Sims, B. (2020) Repairing Infrastructures: The Maintenance of
Materiality and Power. Available at: https://doi.org/10.7551/mitpress/11771.
001.0001
Holt, J. and Vonderau, P. (2015) ‘“Where the Internet Lives”: Data centers as cloud
infrastructure’, in L. Parks and N. Starosielski (eds) Signal Traffic: Critical
Studies of Media Infrastructures. University of Illinois Press, pp. 71–93.
IEA (2022) ‘Data centres and data transmission networks’, International
Energy Agency. Available at: https://www.iea.org/reports/data-centres-and-data-
transmission-networks (Access 10 May 2022).
Jackson, S.J. (2014) ‘Rethinking Repair’, in T. Gillespie, P.J. Boczkowski, and K.A.
Foot (eds) Media Technologies: Essays on Communication, Materiality and
Society. The MIT Press, pp. 221–239.
Jones, N. (2018) ‘The Information Factories’, Nature, 561, pp. 163–166.
Kaack, L.H. et al. (2022) ‘Aligning artificial intelligence with climate change miti-
gation’, Nature Climate Change, 12(6), pp. 518–527. Available at: https://doi.
org/10.1038/s41558-022-01377-7
Khan, S. et al. (2020) ‘On the requirements of digital twin-driven autonomous
maintenance’, Annual Reviews in Control, 50, pp. 13–28. Available at: https://
doi.org/10.1016/j.arcontrol.2020.08.003
Knuth, S. (2019) ‘Cities and planetary repair: The problem with climate retrofit-
ting’, Environment and Planning A: Economy and Space, 51(2), pp. 487–504.
Available at: https://doi.org/10.1177/0308518X18793973
Lemmens, P. et al. (2017) ‘Toward a Terrestrial Turn in Philosophy of Technology’,
Techné: Research in Philosophy and Technology, 21(2), pp. 114–126. Available
at: https://doi.org/10.5840/techne2017212/363
McDonald, R. and Beatley, T. (2021) Biophilic Cities for an Urban Century. Cham,
Switzerland: Palgrave MacMillan.
324 Taylor Stone and Aimee van Wynsberghe

Mukherjee, R. et al. (2021) ‘AI driven road maintenance inspection’. arXiv.

Available at: https://doi.org/10.48550/arXiv.2106.02567
Nye, D.E. (2010) When the Lights Went Out: A History of Blackouts in America.
Cambridge, MA: MIT Press.
Owens, M. and Wolch, J. (2019) ‘Rewilding cities’, in N. Pettorelli, S.M. Durant,
and J.T. du Toit (eds) Rewilding. 1st edn. Cambridge University Press, pp. 280–
302. Available at: https://doi.org/10.1017/9781108560962.014
van de Poel, I. (2020) ‘Three philosophical perspectives on the relation between
technology and society, and how they affect the current debate about artificial
intelligence’, Human Affairs, 30(4), pp. 499–511. Available at: https://doi.org/
10.1515/humaff-2020-0042
van de Poel, I. (2021) ‘Design for value change’, Ethics and Information Technology,
23(1), pp. 27–31. Available at: https://doi.org/10.1007/s10676-018-9461-9
Robbins, S. and van Wynsberghe, A. (2022) ‘Our New Artificial Intelligence
Infrastructure: Becoming Locked into an Unsustainable Future’, Sustainability,
14(8), p. 4829. Available at: https://doi.org/10.3390/su14084829
Rolnick, D. et al. (2022) ‘Tackling Climate Change with Machine Learning’, ACM
Computing Surveys, 55(2), p. 42:1–42:96. Available at: https://doi.org/10.1145/
3485128
Star, S.L. (1999) ‘The Ethnography of Infrastructure’, American Behavioral
Scientist, 43(3), pp. 377–391. Available at: https://doi.org/10.1177/0002764
9921955326
Stone, T. (2021) ‘Design for values and the city’, Journal of Responsible Inno­
vation, 8(3), pp. 364–381. Available at: https://doi.org/10.1080/23299460.2021.
1909813
Stone, T., Santoni de Sio, F. and Vermaas, P.E. (2020) ‘Driving in the Dark:
Designing Autonomous Vehicles for Reducing Light Pollution’, Science and
Engineering Ethics, 26(1), pp. 387–403. Available at: https://doi.org/10.1007/
s11948-019-00101-7
Strubell, E., Ganesh, A. and McCallum, A. (2019) ‘Energy and Policy Considerations
for Deep Learning in NLP’. arXiv. Available at: https://doi.org/10.48550/
arXiv.1906.02243
Vogel, S. (2015) Thinking like a Mall: Philosophy after the End of Nature.
Cambridge: The MIT Press.
Wahba, S.N. (n.d.) ‘Housing Policies That Save (and Improve) Lives, Protect Assets,
and Shield Economies’. Available at: https://thedocs.worldbank.org/en/doc/
185371540492697571-0200022018/original/WorldBankGlobalProgramfor
ResilientHousing.pdf
World Wildlife Fund (2020) ‘Employing AI to evaluate wildlife populations on a
global scale’, World Wildlife Fund. Available at: https://www.worldwildlife.org/
magazine/issues/winter-2020/articles/employing-ai-to-evaluate-wildlife-populations-
on-a-global-scale (Accessed: 12 July 2022).
van Wynsberghe, A. (2021) ‘Sustainable AI: AI for sustainability and the sustain-
ability of AI’, AI and Ethics, 1(3), pp. 213–218. Available at: https://doi.
org/10.1007/s43681-021-00043-6
 Repairing AI 325

Young, M.T. (2021a) ‘Maintenance’, in D. Michelfelder and N. Doorn (eds) The

Routledge Handbook of the Philosophy of Engineering. New York: Routledge,
pp. 356–368.
Young, M.T. (2021b) ‘Now You See It (Now You Don’t): Users, Maintainers and
the Invisibility of Infrastructure’, in M. Nagenborg et al. (eds) Technology
and the City: Towards a Philosophy of Urban Technologies. Cham: Springer
International Publishing (Philosophy of Engineering and Technology), pp. 101–
119. Available at: https://doi.org/10.1007/978-3-030-52313-8_6
Zhang, S. et al. (2020) Regaining Amsterdam Canals for Waste Collection. MIT
Senseable City Lab.
Index

Pages in italics refer to figures and pages followed by “n” refer to notes.

adaptation 15, 24, 67–69, 71–72, 74, 217–218, 220, 222–223, 224;
76, 92, 96, 117, 128, 163, 166, technological 6, 11, 24–26, 76,
176, 178–180, 186–187, 190, 87–88, 90, 95–96, 100, 166, 235,
193–195, 208n18, 227, 230, 232, 257, 264, 268, 276, 283, 286
249, 276–277, 282, 284, 297, 307, artificial intelligence (AI) 2, 8, 27–28,
307, 311, 313–316, 320 202, 215, 222, 225, 306–322;
aesthetics 7, 26–27, 94, 106, 109n2, ethics 7, 23, 27–28, 225, 235, 282,
213, 249, 260, 264–277, 288, 294, 306–322; repairing 27, 306–322,
299–300; urban 27, 264–277 314; for sustainability 307, 307,
Anthropocene 8, 26, 141, 144–145, 309, 313–315, 314, 318–321,
147, 150, 154–155, 285, 319, 321n2; sustainable 27, 306–322
321n3; end of 26, 240–262
Aristotle 24–25, 58–59, 61, 63, 73, biology 59–60, 72, 74–75, 260
98–99, 140–147, 149, 151, 153, breakdown 10–11, 13–15, 19–20, 28,
155–156, 158n13, 168, 177, 29n10, 41–42, 50, 54, 56n14,
181n10, 182n21 86–88, 93, 97, 104, 113–114,
artefacts 5–7, 11, 13, 15–25, 27, 116–117, 120–123, 269, 280, 283,
30n35, 58–81, 86–110, 114–134, 285, 308, 312, 318
135n7, 136n17, 137n21, bricoleur 184, 249–250, 255, 258
145–146, 163, 165–167, 169, bridge 1, 3, 16, 20, 39, 44–51, 53–54,
175, 177–180, 180n3, 181n8, 55n5, 55n8, 55n12, 67–68, 71, 75,
217–218, 220, 222–223, 225–226, 77–80, 81n5, 88–89, 100–101, 123,
244, 250, 262n20, 264–265, 268, 175, 178, 222, 270, 273, 280;
270–272, 275–276, 279–283, George Washington Bridge 44–48,
286–287, 289, 307, 307–308, 45–46, 55; Golden Gate Bridge
311, 313–317, 321n3; disposition 55n5, 67, 71, 78–79, 273
of 4, 26–27, 71, 74, 77, 116–117, broken tool 41, 257
240–262, 295–296, 300; function broken world thinking 14, 56n14, 113,
of 19–21, 86–109, 109n1, 299, 309, 312
110n4–110n5, 200, 282;
maintenance of 16–17, 19–22, care 3–4, 15, 22–23, 25, 28n6, 29n8,
24–25, 58–81, 86–110, 120, 30n38, 108, 113, 115, 129,
162–182, 186, 192, 195, 131–134, 135n3, 135n5, 142, 155,
203–204, 215–236, 282; technical 189, 201–202, 220, 255, 259, 266,
7, 21–22, 88, 165, 180n3, 195, 268, 270–274, 282, 307–308
 Index 327

change 4–5, 15–17, 22–24, 26, data centres 2, 316–318

48–49, 58–81, 88, 97–98, 103, degrowth 248
105, 107, 117, 129–130, 135n5, design 2–28, 48, 50, 65–66, 71, 73,
135n7, 141, 148, 157, 166, 168, 76, 78, 86–109, 113, 115–132,
172–173, 176–177, 179, 189, 192, 145, 165, 174, 177–179, 188, 195,
208n17, 208n21, 215, 219–221, 205n5, 215–236, 242, 244–245,
225, 227–228, 237n2, 241–244, 254, 258, 262n13, 265, 267–269,
266, 272–275, 279, 281–282, 271–276, 279–287, 290–299,
284–285, 287–288, 290–292, 292, 307–315, 319, 321n3; of artefacts
294–298, 300, 307–309, 311–316, 7, 22, 78, 87, 98, 100, 104, 230,
319–320 232–233, 275, 279, 311
circumspection 40–44 digital 26, 56n16, 135n9, 184–210,
climate crisis 140, 157, 229, 240–241, 230, 243–244, 248, 264, 267, 269,
249, 260, 272, 281, 304, 307, 306, 309, 311, 314, 316–321,
314–315 321n3; age 26, 184–210;
Coeckelbergh, Mark 1–31, 81n17, infrastructure 230, 306, 309, 311,
103, 135n8, 163, 166–167, 314–320, 320
309–310, 321n2 disposability 4, 26, 71, 74, 77,
communication 1, 10, 64, 79, 103, 240–262
128, 152–157, 166, 182n22, disposition 27, 104, 116–117, 190
193–195, 207n13, 268–269, 272, DIY 86–110, 130
286, 295, 297
composite objects 25, 27, 60, 63, 65, Earth 25, 54n1, 108, 141, 144–157,
70, 73, 75–76, 104, 155, 163, 166, 158n15, 204, 243, 245, 252, 254,
168–175, 179, 207n11, 210n29, 260, 261n11, 315, 321n3
245–246, 254, 274, 299 ecological 133, 140, 156, 158n15,
composites 246 236n1, 245, 247, 251–252, 258,
concretisation 26, 185–188, 190, 192, 261n1, 265, 270, 273, 276, 281,
205n5, 206n8, 207n11 289, 295, 306–322
conservation 13, 88, 102, 154, 156, empirical turn 7, 167
162, 165–166, 175–180, 182n23, engineering 7, 19–22, 66, 77, 88, 93,
207n13, 216, 220–221, 225, 99–100, 109, 154, 162, 176–180,
233–234, 287, 309, 309, 312–313, 196, 232–233, 245, 249, 321n3,
321n3 322; philosophy of 22, 322
consumerism 5–6, 26–27, 29n19, 91, environment 2–28, 43–44, 49, 52, 54,
97, 102, 145, 158n15, 240–262, 66, 72, 79, 89, 91, 101, 103, 106,
280, 282, 285, 289, 293–294, 297, 113–120, 127, 130, 135n3–135n5,
300, 317 144, 147, 176, 178, 232, 235,
consumers 6, 26–27, 29n19, 91, 102, 236n1, 241, 244, 246, 248,
241–248, 261, 262n17, 280, 282, 250–252, 260, 261n11, 264–277,
297 279, 284–300, 306–322
consumption 26–27, 97, 241–243, epistemic privilege thesis 114, 119,
247–248, 253, 261n5, 262n17, 121–134, 135n7, 136n12, 136n18
285, 289, 293, 297, 300, 317; epistemology 22–25, 37–210; social
cycles 26–27, 247, 280, 310 22, 184–210; standpoint 24–25,
craft 10, 27, 29n10, 41, 43, 88, 99, 56n19, 113–137, 225
101, 103–104, 146, 158n3, 188, ethics 3, 7, 15, 22–28, 29n8, 86, 99,
196–197, 201, 203, 248, 251, 257 132, 143, 145, 213–322; care 3, 15,
creativity 15, 93, 208n21, 250, 270, 22, 29n8, 132, 271; eco-­ethics 26,
274, 282, 312 240–262; of maintaining waste
curated decay 270 infrastructures 27, 279–301
328 Index

ethnomethodology 10–12, infrastructure 13–16, 21, 23–24,

30n23–30n24 27–28, 30n35, 39–56, 88, 90, 104,
experience 3–4, 12, 14, 27, 43, 46–47, 113, 116–118, 122–123, 129,
72, 98, 102, 104–105, 114–116, 134n2, 135n4, 185–186, 189,
119–134, 135n9, 136n13–136n14, 197–198, 204, 218–219, 222–223,
136n18, 152, 176, 199, 201, 240, 227–230, 234, 242–244, 249–250,
250, 254–255, 265–267, 271, 274, 270, 279–301, 306–322; digital
276, 293, 296, 298, 312 230, 306, 309, 311, 314, 316–320,
320; material 28, 306, 308, 320,
failure 11, 14, 19–20, 30n39, 40–41, 320, 321; traffic 20, 47–48, 68,
47, 50–52, 54, 56n15, 88, 92, 223–224, 226–227, 229–230, 252;
104–105, 115–117, 123, 135n9, waste 27, 279–301
144, 156–157, 227, 249, 295, 297 inspection 4, 48
fashion 241–244, 248, 252–253, 275, institutions 1, 10, 23, 26, 88, 109,
286 196, 219–220, 222–234, 249, 283,
fixer 114, 117–123, 126–134, 135n6, 288–295, 298–300
136n18 intergenerational 102, 249, 266,
fixing 17, 86–113, 116, 130, 180, 218, 271–272, 276, 292, 294–295
245, 247, 249, 254, 281, 298, 306 internet of nature 315–316, 319
fragility 13–15, 30n33, 94, 113, 115, inversion thesis 121, 124–125,
133–134, 283–284, 299 127–129, 132–134, 136n12
invisibility 10, 12–13, 27, 30n28,
global 13, 26, 28, 68, 131–132, 144, 39–42, 44, 46, 50, 53, 55n2,
148–150, 168, 204, 232, 241–249, 113–114, 119–122, 127, 135n8,
260, 261n5, 266, 279, 286, 289, 253, 280–284, 289–293, 298, 312,
291, 310, 314–315, 317, 319, 317, 322
321n3
Graham, Stephen 2, 10–11, 14, 29n16, Jackson, Steven 6, 9–10, 13–14, 17,
55n3, 55n9, 56n14, 56n18, 69, 24, 28n6–28n7, 29n10, 30n20,
123, 204, 208n17, 218, 221, 226, 56n14, 56n19, 113–115, 120, 126,
268–269, 280–283, 285, 300, 312, 134n1, 136n11, 218, 220,
317 282–283, 286, 299, 301n4,
308–309, 312
Heidegger, Martin 6, 12, 23, 29n16, justice 1, 23, 148, 218, 225, 231,
30n25, 40–43, 49, 55n13, 90, 142, 285–286, 295, 310–311, 313
145, 154–155, 158n3, 158n9
Henke, Christopher R. 11, 14, 16, 23, know-­how 41, 103, 250, 254
29n18, 30n22–30n23, 30n25, knowledge 2–4, 22, 24, 29n11, 54, 61,
81n7, 81n14, 88, 116, 123, 218, 86–90, 94–109, 113–114, 117–137,
221–222, 226, 233, 250, 281–286, 142, 151, 176–177, 189–190,
297–298, 308–309, 313–315, 319, 195–202, 228, 252, 254, 256–257,
321n3 259, 271; operational 100–103,
hermeneutic 90–93, 100, 193, 197, 197–199; procedural 98, 103,
207n14 108–109, 285; repair 22, 87, 95,
hylomorphism 25, 141–142, 144–147, 97–105; situated 124, 127, 133,
150–157 136n12–136n13; subordination
114–115, 118–120, 124, 126–128,
Ihde, Don 6, 24, 86, 90–93, 108, 130–132
110n3, 277n2
individuation 25, 151–157, 158n3, labour 12–13, 23, 28, 30n28, 42–44,
158n15, 172, 189–190 46–47, 51–54, 55n4, 55n7, 87, 99,
 Index 329

101, 103–104, 109, 117–118, 120, 136n17; realist 25, 162–182;

130–131, 192, 247, 257, 316 Western 24, 58–64, 72–73, 136n17
language 30, 61, 88, 164, 166–168, metastability 151, 153–156, 158n14,
170–174, 181n9, 181n14, 181n16, 190
185, 193, 195, 207n14, 227, 245, Mitcham, Carl 18, 28n1, 98, 158n8
304 modification 4, 7, 13, 16, 20, 60,
life-­cycle 26, 73, 75, 95, 216, 218, 64–72, 76–77, 90, 94, 105, 107,
230–231, 235, 241, 280, 310 168, 186, 208n17
lifeworld 86, 90–94, 96, 102–106, morality 103, 148, 158n12, 219, 233,
109, 264, 266, 271, 273 267, 281, 288–289, 291, 294, 296,
311
maintenance: artefact 16–17, 19–22,
24–25, 58–81, 86–110, 120, nature 14, 44, 47, 52–53, 56n14, 74,
162–182, 186, 192, 195, 203–204, 87, 133, 142–145, 148–149, 153,
215–237, 282; as conservation 13, 155–156, 158n6, 193, 200,
25, 88, 102, 154, 156, 162–166, 203–205, 207n13, 208n17,
175–180, 182n23, 207n13, 209n23, 301n4, 315–316, 319
216–221, 225, 233–234, 287, 309,
312–313, 321n3; of design 2, 4, objects: abstract versus concrete 25,
7–8, 109, 121, 128, 177, 217, 230, 162, 167–179, 181n11; composite
232–233, 265; of discard 60, 78, 25, 163, 166, 168–175, 179;
115, 184, 231–234, 246, 253, 280, technical 26, 184–210
286–287, 293–294; dynamic 27, obsolescence 6, 26–27, 77, 208n17,
215–216, 228, 281–282, 285–287, 216, 240–262, 282, 312; planned
289, 296, 298, 300; epistemology obsolescence 26–27, 240, 243,
of 23–25, 37–210; metaphysics of 245–246, 282
24–25, 37–210; ontology of 21, 90, ontology 21, 24, 58–62, 65, 72, 74,
141; planetary 25, 140–157, 79–80, 90, 141–142, 144, 151,
314–316, 319, 321, 321n3; 154–155, 158n12, 163–164,
software 25, 162–182; structural 166–175, 179, 181n6, 181n7,
41, 89, 115, 130, 165, 245, 310, 181n11, 207n14; of composite
321; technical 131, 141, 145, 147, objects 163, 168, 170–172, 179
150, 152, 154–155, 157, 234; as organisms 72–75, 155, 184, 190, 245
transformation 15–16, 71, 221,
233, 309, 313–314, 316, 319, 320; persistence 24, 58–59, 61–66, 71,
of use 2, 4–5, 7, 24, 29n8, 68–69, 75–77, 79–80, 108, 110n6, 118,
80, 90–94, 97, 115–116, 118, 123, 120, 135n5, 293
167, 187, 194, 197, 203, 216, 229, philosophy: process 60, 72–75, 79–80;
231–232, 234, 280; of value substance 58–59, 72; of technology
(domains) 26, 152, 215–237, 266, 1–31, 55n3, 72, 80, 93, 137n21,
273, 275, 280, 283, 294, 311; value 167, 185, 219, 225, 231, 235–236,
of 23, 26, 215–237 264–266, 277n2, 282, 286
maintenance and repair studies (MRS) phronesis 24, 98–99, 101
9, 11–22, 28n6, 29n16, 30n23, Pirsig, Robert 1, 5–6, 29n8
30n30, 59, 68–69, 81n13–81n14 planetary: maintenance see
malfunction 11, 21, 90–91, 94, 104, maintenance
107, 122, 195, 312 plastics 96, 241–247, 251, 253–254,
Marx, Karl 23, 42–43, 55n6, 120–121, 259, 261n2, 262n14, 274, 297,
124 299; crisis 245–246; single-­use 241,
metaphysics 7, 23–25, 37–210; of 245–246
artefacts 24, 58–81, 109, 110n6, politics 14, 23, 26, 133, 213–322
330 Index

potentiality 150–157, 192 154, 158n14, 215, 218, 220–221,

process: philosophy see philosophy; 226, 229, 235–236, 249, 285, 308,
versus substance see philosophy 311–312, 314
pump 48, 53, 66, 67 standpoint epistemology see
epistemology
realism 162, 171–172 stories 101, 192, 195, 197–198,
recycling 27, 69, 234, 241, 245–247, 202–203, 299
253, 259, 262n13, 262n14, 284, story-­telling 186, 193, 195, 197, 200,
286, 295, 298 202–204
repair: of AI 306–322; as critique 115, STS 3, 7–8, 10, 15–16, 79, 81n13,
129–134, 312; epistemology of 135n8
24–25, 56n19, 109, 113–137; sustainability 8, 25, 119, 140–157,
formal 101, 104–105, 107, 109; 158n3, 219, 227, 229, 231–235,
informal 95–109; infrastructural 241, 243, 248, 260, 267, 271, 277,
275, 286, 300; as maintenance 285, 295, 297, 307–322;
versus transformation 221, 233, transformations 277, 306–322; see
284, 309, 312–314, 319–320; also maintenance
reflexive 27–28, 281–287, 296,
298, 300, 308–309, 319 techne 24, 98–99, 142, 145, 150, 154
technical objects; see also
Seibt, Johanna 58–59, 62–63, 79–80 objectsconcretisation of 26,
sewers 261n7, 280, 284 185–188, 190, 192, 205n5, 206n8,
ship 60, 64, 69–70, 142, 243, 251, 207n11; temporality of 19, 22, 26,
286; of Theseus 60–61, 63–64, 66, 47, 62–63, 74, 80, 90, 163, 166–169,
70, 72, 95 172–173, 179, 184–195, 200–201,
Simondon, Gilbert 25–26, 141, 204, 206n10, 217, 228, 231, 267,
151–158, 184–210 270, 272, 313
situated knowledge thesis 124, 127, technicity 26, 185–186, 189–197, 200,
133, 136n12, 136n13 203, 206n6, 207n12
skill 3, 5, 12, 19, 24, 43, 86–88, technics 141–148, 151, 154–157
90, 93, 95, 98–109, 110n5, technological individuation 25, 140–158
118–119, 127, 129, 132, 134, technologies 1–31, 68, 73, 75, 81n6,
135n9, 185–186, 189, 195–198, 81n13, 88, 104, 110n3, 122, 150,
200, 204, 208n17, 228, 250, 154, 166, 186, 199, 217, 219–220,
254–255, 259 222, 232–233, 244, 260, 264–277,
social construction of technology 283, 286–287, 290–291, 298, 308,
(SCOT), the 9 310–311, 316, 320; domestic 12,
social order 10–11, 104, 200, 204, 250; discarding 115, 184, 231–234,
218–221, 226, 283, 285, 289 246, 253, 280, 286–287, 293–294;
socio-­technical system 26–27, 73, 113, life cycle of 216, 218, 230–231,
120, 128–129, 216–218, 220–230, 235, 310; see also life-­cycle; and
235, 279–282, 284, 286, 291–292, time 24, 269; see also temporality
292, 294, 296, 321n3 of technical objects; urban 27,
software 2, 25, 28, 50, 56n16, 88, 264–277
135n9, 162–182, 192, 197, 306, temporal 4, 17, 19, 26, 47, 62–63, 74,
308, 313; maintenance see 80, 90, 163, 166–169, 172–173,
maintenance 179, 184–186, 190–194, 200–201,
stability 10, 14–15, 17, 26, 56n14, 67, 204, 206n10, 217, 228, 244, 267,
74–78, 81n13, 90–93, 100, 270, 272, 274, 313; see also
113–114, 135n5, 147–148, 151, temporality of technical objects
 Index 331

temporality of technical objects values: aesthetic 27, 94, 264–277;

186–195 ethical 27, 276; see also artificial
token 4, 78, 94–95, 98, 100, 104, 163, intelligence; ethics, eco-­ethics;
166–175, 177, 180, 182n18– maintenance of see maintenance;
182n19, 182n23 transformation of see
transformation 15, 26, 42, 67–69, 71, transformation
76, 91, 101–102, 106, 127, 129– verticality of form 145–158
130, 189–190, 202, 204, 206n6, vulnerability 13–15, 30n32, 40, 71,
220–221, 225–228, 233, 236, 277, 115, 133, 269, 316–317
295, 309, 313–314, 316, 319, 320;
of artefacts 101; see also artefacts; waste; see also ethics;
repair; maintenance; of values 26, infrastructureglobal 242–249;
226–227, 235; see also maintenance infrastructures 27, 279–301;
Tronto, Joan 3, 29n9 management 27, 262n13,
type 87–89, 94–95, 98–99, 103–104, 280–301
107, 109, 117, 163, 166–175, Winner, Langdon 97, 104
177–178, 180, 182n18, 207n11, Wittgenstein, Ludwig 12, 30n27
232, 237n2, 258–259, 264–266,
268–270, 272–273, 276, 297, 316
Young, Mark Thomas 1–31, 55n2,
urban; see also aesthetics; 56n18, 58–81, 108, 110n6, 122,
technologiesinfrastructures 39, 41, 135n9, 136n11, 137n21, 163,
47, 49, 54, 55n9, 280, 283, 285, 165–166, 175–177, 179–180,
290, 296, 300, 307; life 23, 39, 264, 182n22, 185–186, 217, 221, 249,
266–267, 273–274, 283, 286, 290, 271, 273, 279–283, 285–287, 308,
295; technologies see technologies 311–313, 316