Content uploaded by Kathryn Furlong
Author content
All content in this area was uploaded by Kathryn Furlong on Aug 22, 2017
Content may be subject to copyright.
Author's personal copy
STS beyond the “modern infrastructure ideal”: Extending
theory by engaging with infrastructure challenges in the
South
Kathryn Furlong
*
Département de géographie, Université de Montréal, Pavillon 520, ch. de la Côte-Sainte-Catherine, C. P. 6128, succursale Centre-ville,
Montréal, Québec H3C 3J7, Canada
article info
Article history:
Received 23 October 2013
Received in revised form 24 March 2014
Accepted 10 April 2014
Keywords:
Water supply
Global South
Disrepair
STS
Coexistence
abstract
With few exceptions, STS theories of infrastructure stability and change have not been
applied to circumstances in the South. Developed in post-War Europe, these theories are
often applied in ways that lack transferability to situations where infrastructure conditions
are precarious and hybrid. This article seeks to broaden these theories by relating them to
infrastructure challenges common to the South, drawing in particular on prevalent issues
in water supply. Such engagement helps to identify shortcomings in these theories, to push
their paradigms further, and to raise new questions related to infrastructure configuration,
stability, and transition. As such, the study of sociotechnical systems across a range of
contexts can be enriched. In particular, this article extends theory by placing coexistence
among sociotechnical systems, as opposed to the universality of a single dominant infra-
structure network, at the center of enquiry. Recognizing coexistence is important because
it enables one to decouple key concepts in STS from the presumption of universalized and
uniform networks, enabling them to become relevant for the South. Examples discussed in
this essay include stability or “momentum”and transitions.
Ó2014 Published by Elsevier Ltd.
1. Introduction
Improving access to potable water and sanitation is
recognized as one of the key challenges of our time. Still,
success continues to be limited [1]. Reasons are manifold
and include issues such as precarious land tenure, rapid
urbanization, poorly funded utilities, weak institutions, and
inequality [2,3]. However, important challenges also stem
from contradictions between uniform infrastructure net-
works, derived from experiences in the North, and diverse
realities in the South [4]. Instead of privileging a single
universal infrastructure network, greater attention is now
given to hybrid and disaggregated systems characterized by
coexistence. In the North, where the “modern infrastruc-
ture ideal”of universal and uniform coverage by a single
network was largely met [5], mounting pressures on the
environment and utility capacity have encouraged the
adoption of alternative infrastructure configurations [6].In
the South, water supply has long involved multiple systems
in varying degrees of coexistence (e.g. piped water aside
rainwater collection, tanker truck delivery, and communal
systems, etc.). What remains controversial is whether to
ignore what is external to the piped network, to regulate it,
or to subsume it under the management of a single water
provider [7].
At the same time, many local water systems are un-
dergoing change. For Bakker, efforts in the South have
focused on a transition from “artisanal”to “industrial”
water supply [8]. Yet, in many cases, such efforts have given
way to accepting diversity and coexistence among water
*Tel.: þ1 514 343 6111 28499.
E-mail address: kathryn.furlong@umontreal.ca.
Contents lists available at ScienceDirect
Technology in Society
journal homepage: www.elsevier.com/locate/techsoc
http://dx.doi.org/10.1016/j.techsoc.2014.04.001
0160-791X/Ó2014 Published by Elsevier Ltd.
Technology in Society 38 (2014) 139–147
Author's personal copy
supply alternatives [9]. Many prominent approaches to
cities and infrastructure networks, including those from
science and technology studies (STS), do not reflect this
diversity, given their tendency to be derived from circum-
stances in the North [10,11]. But, even these are shifting. As
such, certain analytical frameworks could be strengthened
though engagement with coexistence between –as
opposed to the uniformity of –sociotechnical systems. This
implies submitting theories to empirical circumstances in
the South, where such coexistence is more obvious [10].
In this way, there is mutual benefit to be derived from
engaging the STS literature with that concerned with water
supply access in the South. Geographers, for example, have
dealt extensively with a variety of impediments to service
extension in the South, often taking a political ecology
approach that underscores the historical and political ori-
gins of environmental and resource inequalities [12].
Increasingly, they have taken an interest in the material
components of water supply and how power is mobilized
and transformed through artifacts [e.g. Refs. [13,14]]. Here,
authors have drawn on the philosophical aspects of STS’
engagement with materiality, but not the theoretical
frameworks developed to analyze sociotechnical systems,
specifically Large Technical Systems (LTS) and Multi-Level
Transitions (MLT). The former focuses on explaining the
evolution, function and stability of large technical networks
like water and energy, while the latter is concerned with
the transition from one sociotechnical system to another,
e.g. from a transport system based on increasing highway
capacity to one focused on collective modes [e.g. Ref. [15]].
LTS and MLT have been important to our understanding
of sociotechnical systems. Yet, focused on the North, they
have tended to take post-War, western sociotechnical sys-
tems as the apex of infrastructural development. Moreover,
they are often apolitical in their application,
1
presenting a
near one size fits all model of sociotechnical configuration.
As such, LTS and MLT would seem to have little to say about
infrastructure conditions in the South, which are often
hybrid. Yet, at the same time, in addressing issues of soci-
otechnical configuration, stability and change, LTS and MLT
theories force new questions about sociotechnical systems
in the South with the potential for a broader understanding
of the challenges and possibilities for transitioning to more
equitable configurations.
Specifically, in engaging STS approaches with socio-
technical conditions in the South, I seek to question the
conflation of “system”with uniformity, “stability”with
universality, and “transition”with displacement. To do so, I
draw on the existing (although limited) literature on soci-
otechnical transitions in the South, recent research in ge-
ography and anthropology on the work of objects in
configuring water supply access, as well as aspects of my
own research in Colombia. As such, this paper does not
offer a comprehensive case study. Instead, it draws on both
theoretical and empirical work in geography and STS to
further current theorizations of large technical systems
(LTS) and multi-level transitions (MLT). Thus, while the
paper does not deal with a range of issues that are key to
understanding service provision in the South, it engages
with aspects that can shed new light on standard premises
in STS and our understanding of sociotechnical systems in
turn. In this way, the paper follows Geel’s call to “accept
some loss of empirical complexity”in order to introduce
new questions into STS theory and take it beyond the
conventional case study sites [17].
2. Beyond uniformity and universality: sociotechnical
coexistence in the South
The evolution of large technical systems (LTS) has been
an important topic in STS. Studies have focused primarily
on western contexts where universal infrastructure net-
works have become an unconscious backdrop to everyday
life [e.g. Refs. [18,19]]. Infrastructure, in this context, is
considered “stabilized”,“obdurate”and “black-boxed”, that
is, as essentially immune to change or external influence
[e.g. Ref. [20]]. In these formulations, transition from one
LTS configuration to another is possible, but rare. Many
studies, for example, have focused on a “battle of the sys-
tems”whereby transition involves the total displacement
of one technology by another, which thereby obtains a
virtual “natural monopoly”[21]. Stability is not absolute,
but major external shocks to a system are considered
necessary for reconfiguration to occur, resulting in new
stable configurations as opposed to dynamism [20,22].
To understand such reconfiguration, analyses in STS are
often framed in terms of “multi-level transitions”(MLT).
MLT theory posits three “levels”of interdependent activity
through which transitions occur. Briefly, these include the
level of technological development (niche), the in-
stitutions, norms and practices that surround a technology
(the regime), as well as broader societal trends (the land-
scape) [e.g. Refs. [23,24]]. A key limitation of MLT in
analyzing sociotechnical transitions in the South is that it
assumes a universal system as the end point, i.e. one system
is replaced by another [25]. Thus, while the MLT approach
recognizes a range of pathways to change [26], ultimately a
single, universal outcome is settled on. Consequently, the
approach has rarely been applied to cases where the
outcome it assumes is not already in place.
2
Authors in disciplines like geography and planning have
also been concerned with trends in infrastructure. Here, the
influence of political-economic change, notably neoliber-
alism, has been a prominent focus [e.g. Ref. [28]]. From an
MLT perspective, such research privileges the landscape to
the virtual neglect of the niche or regime. Most prominent
is the work of Graham and Marvin, which argues that
policies of economic liberalization led to widespread
network “unbundling”and the “splintering”of the urban
experience [5]. Unlike MLT, this work does not assume a
universal outcome to splintering, but it does assume uni-
versal networks as the starting point. For this reason, the
“splintering urbanism”thesis is argued to be of limited
applicability to experiences in the South [29]. There,
1
This is a critique of STS more broadly [16].
2
A recent exception is the application of Strategic Niche Management
(SNM) –a variant of MLT –to biogas implementation in India [27]. This
work is discussed in the final section.
K. Furlong / Technology in Society 38 (2014) 139–147140
Author's personal copy
infrastructure, existing in “archipelagoes”and not “net-
works”[8], could not splinter, as it was not universalized
[29].
When thinking through sociotechnical transitions in the
South, not only is it problematic to assume universal net-
works as either the origin or outcome of change, univer-
sality can be problematic as a goal. In some cases, the drive
to universalization has compounded exclusion by creating
service islands and encouraging sprawl [30], or by limiting
alternatives in contexts of diversity [9]. For water, impor-
tant public health issues can arise from efforts to displace
existing options with a single piped supply. Such networks
are often under-resourced and experience problems with
maintenance, disrepair and service interruption [31].
Although interruption may be a “normal”aspect of net-
worked infrastructure even in the North [10], with water
supply in the South, it can have important consequences in
creating what Gandy calls “landscapes of disaster”[32].
Public health research, for example, finds that even a few
days of service interruption can “destroy the health benefit
from the provision of clean drinking water”and increase
the annual risk of illness by 100% [33]. As such, some
advocate the improvement of existing water supply sys-
tems as opposed to their replacement [33].
There are good reasons for fostering coexistence among
water supply systems as opposed to the universalization of
a uniform and “black-boxed”network. First, in areas with a
high degree of poverty, it is important to retain free sources
of water for non-potable use [34]. To this end, rainwater
collection in combination with piped water is promoted in
many parts of the world [34]. By way of example, in Quibdó,
Colombia, residents have traditionally relied on rainwater,
where piped water has long been spatially limited, unreli-
able, and non-potable [35] and where rainfall levels are
among the highest in the world [36]. To improve potable
water access, a project to develop a functioning piped water
system in coexistence with extensive rainwater collection
was begun in 2010. Given a legal requirement for metered
billing in Colombia, attempting to displace rainwater and
restrict users to piped water could significantly raise
household costs, presenting new barriers to access.
Second, utilities may often need the support of alter-
native water sources in order to extend piped water to
unconnected households. This is considered to be among
the key benefits of integrating rainwater collection with
piped supply [37]: it reduces demand on limited treatment
capacity. High consumption by those already connected to
piped water systems has been a key impediment to projects
to extend water infrastructure in Medellín [38] and
throughout many parts of the world [39,40]. Continuing the
example of Quibdó, there, the utility lacks sufficient treat-
ment capacity to meet demand for all water uses for the
city’s entire population (approximately 100,000). As such,
utility staff view rainwater as an alternative source that can
help to reduce pressure on limited water treatment ca-
pacity, enabling network extension. This situation reflects
the finding that, in response to scarcity, cities in the South
must innovate in idiosyncratic ways that mix technical
solutions and social understanding [41].
Yet, Quibdó is not alone. In many cities in both the North
and the South, efforts are being made to develop rainwater
collection systems in order to reduce pressure on treatment
plants under conditions of population growth and/or water
scarcity. Although integration is not simple, energy and
environmental efficiencies have been found to increase
over time [42]. Studies conducted across 62 cities in Brazil
demonstrate that savings on piped water can range from
34% to 92% with the introduction of rainwater harvesting.
Savings are especially significant in areas with high rainfall
and low demand for piped water [37]. The decreased sys-
tem demand can lower system peeks [43], mitigate pres-
sure on water sources, as well as reduce pollution, flooding,
and pressure on sewer systems [34]. All of these can help to
facilitate service extension by reducing treatment and
provision costs and well as pressure on water resources.
Users also seek coexistence. Recent studies show that
users often continue to draw on a range of water sources in
addition to piped water, even after getting a household
connection. In low-income parts of Tijuana, Meehan finds
that users remake the state planned piped water network
by adding additional “objects”like rain barrels to their
newly acquired centralized supply. They do this in order to
maintain independence, control costs, and limit piped
water consumption [13]. Meehan’sfindings are reflected in
research in Medellín. There, a survey of neighborhoods
recently connected to the piped water network shows a
tendency to continue to draw on a variety of sources to
meet water needs [38].
After connection, water supply systems can remain
hybrid integrating the piped supply into existing systems
or merging it with new techniques and artifacts. This does
not necessarily reflect the inadequacies of a sociotechnical
system that has failed to achieve uniformity and univer-
sality. It can also be a sign of the empowerment of users vis-
à-vis the state system. For Meehan, it “make[s] new water
worlds possible”by “enable[ing] institutional flexibility,
local expertise, and technological malleability”[13].
Further, while such technologies like rain barrels may begin
as “stop-gap”solutions to punctual issues in service de-
livery, their “longevity .signals a proliferation of infra-
structure, and with that a variegated geography of power”
[13]. These artifacts and the collective social practices
they imply can “coexist with or even limit the spheres of
state power”[44].
Following from the above, fostering coexistence –as
opposed to uniformity –in a sociotechnical system can
facilitate service extension and enable people to gain
greater control over how they meet their water needs.
Opening LTS theories to coexistence is also of increasing
importance for understanding infrastructural change in the
North. Current approaches tend to be fixated on an ideal of
networked infrastructure as uniform and universal. Such
configurations have been challenged by neoliberal policy
reform as well as by environmental concerns. In fact, while
studies have tended to focus on how cities in the North
came to achieve an idealized infrastructure configuration, it
must be acknowledged that both the experience and the
stability of such configurations are relative. If I live in a
poorly maintained building with inadequate plumbing in
New York, I will not experience the water network as uni-
versal and invisible [see Ref. [45]]. Moreover, if I examine
other aspects of urban water systems such as drainage, it is
K. Furlong / Technology in Society 38 (2014) 139–147 141
Author's personal copy
not uniformity but coexistence –between technical infra-
structure and what Carse calls “nature as infrastructure”
[46] –that will dominate [47].
3
3. Decoupling stability from the “ideal”: momentum
as a feature of disrepair
The presupposition of a “universal infrastructure ideal”
in LTS and MLT theories can help to explain their lack of
traction for studies focused on the South, where infra-
structure is often characterized by “archipelagoes”and
disrepair or even dilapidation. Still, concepts that have been
reserved for the “ideal”have something to offer to other
contexts. Recently, malfunction and disrepair have occu-
pied new interest in the literature on urban infrastructure.
Graham and Thrift’s piece “out of order”stands as a critique
of LTS theory as focused on “utterly fixed hard technologies
.characterized by perfect order, completeness, imma-
nence and internal homogeneity rather than leaky, partial
and heterogeneous”[10]. This focus, as they note, has led
LTS studies to neglect systems in the South [10]. Still, the
authors are concerned with infrastructure in the North and
understand disruption as about “learning, adaptation, and
improvisation”rather than “catastrophe”[10]. Drawing on
Summerton’s idea of technical transitions through incre-
mental change [49], they argue that disrepair can engender
important system shifts through manifold incremental
experiments focused on adaptation [10].
This cumulative adaption and learning, I argue, is also
about custom and coping. When disrepair is normal and
everyday, as opposed to aberrant, it can generate its own
“momentum”. Momentum is a concept developed by
Hughes to describe how the links between society
(including users and a variety of institutions) and tech-
nology evolve to create a stabilized sociotechnical system
that is resistant to change [50]. In his reading of Hughes,
Geels argues that momentum is derived from the re-
lationships between people and things, “people adapt their
lives to artifacts”[15]. Once an LTS achieves momentum,
however, it is considered to uniformly dominate a
geographical space and to be black-boxed, whereby its
behavior is independent of context or user interaction
[51,52]. In these formulations, the sociotechnical system
has become embedded because of its invisibility and its
normalization [51].
Indeed, while the South is more or less ignored in the
application of STS theories of stability and change, the long-
term and widespread difficulties of “transitioning”to the
“universal infrastructure ideal”suggests a degree of mo-
mentum. That is, momentum can be a feature of socio-
technical systems characterized by coexistence,
malfunction, and co-production (as opposed to universal,
idealized, and black-boxed). Barnes and Alatout go some
way to acknowledging this possibility in their discussion of
the uniqueness of all sociotechnical systems, stating that
their evolution and “the ways in which they are variously
cemented, contested, and discarded is closely tied to the
production of social worlds”[53]. That is, the ways in which
they become stabilized –or “cemented”–is produced with
the social context; it is not dependent upon the existence of
a particular social context a priori.
How can this be understood? For Trentmann, like Gra-
ham and Thrift, disrepair and malfunction are normal ele-
ments of user relationships to infrastructure; they “are not
freak accidents or aberrations but natural, constitutive
features of lived normality”[54]. In the North, the
normalization of breakdown and service interruption is
derived from the adaptability (as opposed to the obduracy)
of people’s habits in the expectation that interruption is
temporary. Once service returns, so too do the regular
customs of use; alternate habits are only sustained through
long-term shifts in the sociotechnical system [54]. This
speaks to Star’s assertion that infrastructure constitutes
and is constitutive of communities of practice [55]: habits
and customs are resistant to change, as they are bound-up
in the wider social order. Such communities of practice
need not be restricted to the “ideal”case. They can
reasonably be expected to develop in conditions of
persistent disrepair and dilapidation, to which people also
“adapt their lives”[see the above citation of Ref. [15]].
Disruption and its assimilation by users are revealing in
terms of the mutual construction of customs and perfor-
mance. First, users “are not just recipients [of services] but
co-producers, whose behavior is vital for managing short-
ages”; they play active roles in absorbing, coordinating, and
even orchestrating disruption [54]. Further, user reactions
to system function and breakdown are defined by percep-
tions of what is normal or aberrant for the sociotechnical
system in question [54]. For example, the sometimes-
disproportionate expectations of high-income groups can
place undue strain on a network, engendering disruption
[54]. Similarly, in the South, middle class users may de-
mand a level of service that is unaffordable to other groups,
limiting network extension [7]. Below, I further make the
point that extremely low expectations of system perfor-
mance can also result in practices on the part those con-
nected to the network that place strain on it, complicating
efforts to increase coverage.
These issues reflect Shove’s concerns with habits and
custom. For Shove, the focus on efficiency in terms of
resource use is misplaced and “overshadows prior ques-
tions about the framing and formulation of ‘normal’prac-
tice”. It misses “the point that much consumption is
customary, governed by collective norms and undertaken
in a world of things and sociotechnical systems that have
stabilizing effects on routines and habits”[56]. To under-
stand and alter service use, we have to understand the
“collective dynamic of normalization”[56]. The neglect of
custom, moreover, engenders a focus on “educating”users
[56]. Assumptions that the right information will induce
behavioral change neglect the complexity of consumption
[57] as well as the role of artifacts in constructing social
practice [58].
Notably, convenience is an important factor in consti-
tuting the ways in which sociotechnical systems develop so
that users are able to meet their needs. Convenience is not
only about accessible services, it is about interactions with
the sociotechnical system that allow users to manage their
3
On the point of infrastructure coexistence in the North, see also Refs.
[48] and [26].
K. Furlong / Technology in Society 38 (2014) 139–147142
Author's personal copy
time [56]. Thus, when a service is highly unreliable, users
may seek strategies to improve reliability, either by opting
out of the system, through clandestine connections, or
through storage [59]. Understanding users as seeking sys-
tem knowledge and reliability, not only in terms of timing
and quality but also in terms of costs, helps to comprehend
the potential “momentum”of sociotechnical systems of
malfunction. Indeed, according to Shove and her col-
leagues, convenience becomes especially important where
“there are problems of coordination”between users and an
LTS leading to the proliferation of individualized
solutions
4
[61]. These tendencies can place pressure on
collective modes of consumption as user practices diverge
[61].
In the context of persistent malfunction, however, such
tendencies can make it difficult to transition to universal-
ized systems; momentum can become a feature of disre-
pair. Recently, some STS scholars have sought to address
this issue using strategic niche management (SNM) [e.g.
Refs. [27,62,63]]. SNM, like much MLT work, privileges the
niche as the locus of sociotechnical change [see Ref. [64]]. It
understands “transitions”to emanate from the careful
development and introduction of alternative –often more
environmentally sound –technologies [23]. For these au-
thors, malfunction and disrepair provide a window for
change, but only up to a point. Malfunction opens oppor-
tunities for transition, but can also create barriers to
change. That is, what they call an “unstable”system can
prove rather stable; it can exhibit “momentum”. Barriers
arise because malfunction, and the governance issues from
which it emanates, create uncertainty for investment in
new technologies, limiting their diffusion [27,62]. Accord-
ing to Raven, persistent malfunction favors change until it
creates too much uncertainty, at which point the potential
for change is reduced [65].
5
Thus, for these authors, the
effect of disrepair on sociotechnical change is all about
degrees.
Recent studies in geography can help to contextualize
the momentum witnessed. In an analysis of electricity
services in Nicaragua, Cupples argues that “people’s
embodied experiences of electricity failure”explain the
difficulty of introducing meters [58]. Citing Latour, she
notes that an artifact “will work when all the relevant
people are convinced”[66]. In Nicaragua, neither utility
workers nor users are “convinced”and they assist each
other is subverting the prescribed change [58]. Further, for
Shaw and Meehan, “new political realities must be pro-
duced with objects, not in spite of them”[44]. This means
that, where a sociotechnical system is characterized by
malfunction, transitioning to “a configuration that works”
[67] may be contingent on fostering reconfiguration and
coexistence as opposed to displacement.
Returning to the Quibdó example, the long-term unre-
liability of the piped water system impedes the proposed
transition to coexistence, whereby a reliable piped water
system is supported by rainwater collection. Given the
legacy of intermittent supply, in Quibdó as elsewhere, the
logical response is to fill household storage tanks whenever
tap water is available. Not knowing when this might be,
taps need to be left open, which can lead to overflow.
Moreover, in cases where a household has a hose, but no
tap, the possibility to turn off the water does not exist.
According to utility staff interviewed in 2011, for these
reasons, a connected household can draw four to five times
its actual monthly consumption. Given the intermittency of
service, moreover, everyone who wants or needs water
must draw simultaneously –i.e. when it happens to be
flowing. Such demand on the piped system complicates
network extension given limited treatment capacity.
Yet, water storage is logical. It enables people to have
water when it is needed, and not simply when it happens to
be provided. In fact, it is the same choice that users in the
North make when faced with intermittent malfunction [see
Ref. [54]]. But when it is function that is intermittent, storage
becomes a dominant as opposed to an occasional practice.
This can make it difficult to provide either continuous or
potable water. It can limit service quality because the utility
is unable to treat enough water to maintain continuous flow.
Continuous flow is needed to ensure sufficient pressure
without which negative pressures, infiltration, and
contamination ensue. In piped water systems, microbio-
logical growth becomes dislodged from pipe surfaces in
conditions of intermittent flow [68]. Intermittency can also
lead to contaminant infiltration, further encouraging
bacteriological growth, consuming chlorine residuals, and
potentially introducing fecal contaminants from the soil or
nearby sewerage systems [69,70]. In response to intermit-
tent supply, a vicious circle develops whereby intermittency
makes water storage essential to bridge service gaps, which
in turn makes continuous flow difficult to achieve, leading to
contamination.
4. Decoupling transition from displacement: from
malfunction to coexistence?
A 24-hour system, necessary to ensure potability, is
dependent upon both the utility and upon users. Yet, the
transition to such a sociotechnical system demands that
users shift their practices before the promised service
quality can materialize. That is, users need to develop
habits consistent with a functioning system before the
system actually begins to function. This poses evident
problems given one’s dependence on water.
In attempting to understand these issues, the ways that
theories of LTS and MLT are generally applied pose certain
constraints. Beyond the presumed displacement of one
universal system by another, the theories have been
critiqued for their lack of attention to users and con-
sumption, a disregard for incremental change, and a focus
on the “niche”as the locus of transition. I address each of
these below. I do not pretend to offer an alternative theory
of sociotechnical transition for cases of persistent disrepair.
Yet, by underlining how the above issues inhibit engage-
ment with common infrastructural conditions in the South,
I hope to offer some considerations for extending STS
theories to the study of sociotechnical change in situations
characterized by shifting forms of coexistence and long-
term disrepair.
4
Here, the authors refer to Ref. [60].
5
Referred to in Ref. [27].
K. Furlong / Technology in Society 38 (2014) 139–147 143
Author's personal copy
First, the lack of attention to human agency has been
widely identified as a shortcoming in STS approaches to
technical networks [61,71].
6
Recalling Latour’s assertion
that an artifact will function “when all the relevant people
are convinced”[66], what is made obvious by an engage-
ment with conditions of disrepair is the need to expand the
notion of “relevant people”within both LTS and MLT. Ac-
cording to Geels, the literature posits two basic groups of
human actors: the “system builders”or “heros”who can
affect transitions (e.g. scientists and policy makers); and
the “faceless automata following iron rules, institutions or
given roles/functions”that cannot affect change [75].In
general, the influence of users is considered only in the
stabilization of a system through “habits”, or in achieving
change through mass protest [15]. This ignores how sys-
tems are continuously remade through everyday interac-
tion [61].
The above treatment of human agency raises questions
about how to think about networks. It focuses on infra-
structure as a physical network but ignores the Latourian
view of “networks”as relations between people and things.
The concern with “heros”obfuscates the fact that often
there are “no obvious ‘system builders’.no key in-
stitutions enlisted and enrolled and no well-defined states
of sociotechnical development”[61]. Here, Shove is talking
about the laundry, but this could also be said of systems
characterized by persistent malfunction: they are marked
by continual individual innovation [see Ref. [10]]. This fol-
lows Latour’s point that networks are generally absent of
central nodes of control; they are made of relations
[76].
7
This position also informs Cupples’critique of looking
to protest to explain sociotechnical change. For Cupples, the
approach does not “come to terms with how it is that
power is exerted across space”, because it ignores the role
of artifacts in explaining how relations of power and
resistance are produced [58].
A second impediment to analyzing conditions of disre-
pair that arises from common applications of LTS and MLT
is the lack of attention to incremental change. LTS transi-
tions are typically defined as involving major “trans-
formations in the way societal functions .are fulfilled”
and as being “systemic in nature”[62]. For Shove, change is
much more organic. It involves the shifting relationships
between people and combinations of artifacts that yield
changes in how services are conceptualized and thereby
“reconstructed”[61]. According to Geels, this oversight
results from the case studies chosen; a more diverse se-
lection might show that “the new grows out of the old .
through cumulative adjustments in new directions”
[15].
8
Again, these interventions invoke a Latourian vision
of networks, whereby the very premise of “durability”in
relation to networks must be contested, as it is their current
links to other “actants”(human and non-human) –and not
their size –that construct their “dominance”[76].
A third area where LTS and MLT ought to be broadened
is with respect to the focus on the niche as the locus of
sociotechnical change. By privileging niche innovation,
economism –a bias toward “the primacy of economic
causes and factors”
9
in the explanation of social phenom-
ena –is fostered. Specifically, innovation comes to be syn-
onymous with commercial development and economic
growth [41]. Consequently, the “technical niche”is often
reduced to the “market niche”. Verbong et al., for example,
assess the potential for a transition to biogas in India
through the targeting of “market niches currently not well
served by the power regime”[27]. Similarly, Berkhout et al.
see sustainability transitions in the South as occurring
through the “co-evolution of new innovative capabilities
within specific market and institutional setting”[62]. This
prioritization of economic factors in explaining socio-
technical change may not be unique to LTS. For Hess, STS is
generally marked by an inherent neoliberalism [79].
Beyond economism, the focus on niche innovations
raises other issues. It has been criticized for ignoring the
role of user interaction in “reconstructing”LTS [61]. For the
particular case of water supply, the focus ignores the
importance of simple, and well-known technologies in
enabling sociotechnical change, such as rain barrels or low
flow fixtures [74]. Whether for improved environmental
performance in the North or for improved system reliability
in the South, often it is the successful introduction of
readily accessible household technologies as opposed to
new technological innovations that are sought. Transition
in such cases is less about innovation and more about
integration into the physical as well as the Latourian
network.
In light of the above discussions, I suggest a possible way
forward for considering the circumstances through which
“people are convinced”. How can users whose “experiences
embody”disrepair be “convinced”by proposed socio-
technical reconfigurations? Three strategies witnessed
during my research offer potential pathways for investi-
gation. First, in Medellín and Quibdó both utility personnel
and representatives of communities newly connected (or
awaiting connection) to the water network emphasized the
importance of demonstrating project results, i.e. consistent
improvements in service provision, over things like public
education. Experiences elsewhere in Colombia echo this,
finding that reducing the time-lag before concrete im-
provements in living conditions become apparent is crucial
to garner project credibility [80].
Second, in Quibdó, the utility uses scheduled service
provision in order to reduce the need for water storage. In
2011, the utility began to provide water all day by sectors,
each sector getting one to two hours of service a day. The
strategy of maintaining scheduled if intermittent service
recalls Trentmann’s assertion that planned disruption,
through rolling blackouts or rota water disconnections, can
serve “to reduce stress on crisis-prone systems”[54]. It can
also help to addresses Shove’s problem of convenience.
6
Users, however, receive significant attention in other STS debates,
most notably in relation to particular objects [72]. For examples where
users have been important in analyses of networked infrastructure, see
Refs. [56,73,74].
7
The discussions of Latour’s philosophy in the section draw on Har-
mon’s (2009) see Ref. [76] analysis of Latour’sIrréductions [77].
8
See also Ref. [78].
9
Source: Oxford English Dictionary.
K. Furlong / Technology in Society 38 (2014) 139–147144
Author's personal copy
Although water cannot be obtained at all times, users know
that at particular times it is available. They can plan. Third,
mediating technologies, that help users to regulate their
consumption, can also be useful in improving convenience.
In Quibdó, as in Tijuana [see Refs. [13]], this means
providing floaters so that the water will automatically shut
off when storage tanks are filled.
5. Conclusions
This analysis sought to open a discussion about socio-
technical coexistence, stability, and transition beyond the
“modern infrastructure ideal”. It sought to identify oppor-
tunities to extend theories of infrastructure in STS to
include a wider variety of contexts and sociotechnical
configurations. Despite the limitations of current theories
for sociotechnical systems in the South, moving away from
the presumption of universality/dominance in favor of
coexistence reveals that LTS and MLT offer relevant insights
for sociotechnical systems beyond the “ideal”case. This is
important not only for applications in the South. If the
theories are to continue to be relevant for studies of the
North, they will likewise have to adjust to evolving condi-
tions of coexistence [see Ref. [6]].
Starting from a premise of coexistence, not only enables
an engagement with a wider variety of sociotechnical re-
alities, it complicates and extends the analytical utility of
certain key premises in STS. In particular, theories of sta-
bility and change, central to LTS and MLT, can have rele-
vance beyond situations characterized by the “modern
infrastructure ideal”. In many cities in the South, large
numbers of people have lived with persistent disrepair.
They may have no service, sporadic service, and/or poor
water quality [59]. Disrepair is the essential character of the
network. But that network remains part of a sociotechnical
system. As such, the continuous experience of unreliability
can result in its own “momentum”; it can become stabi-
lized and embedded. Recalling Hughes, momentum is not
inherent in a technology but develops through the variety
of social institutions and practices that have co-evolved
with it [50]; infrastructure “shapes and is shaped by the
conventions of a community of practice”[55]. Following
this reasoning, the actual functionality or dominance of an
LTS is not the sine qua none of momentum. Rather, it is the
fact that society has adapted itself to that system, whatever
its configuration, whatever its condition.
This is equally relevant for extending conceptualizations
of “transition”. By accepting that transition need not entail
the displacement of one universal system by another, one
can consider a broader range of configurations that may or
may not “work”. Moreover, and perhaps even more
importantly, in tempering the need for significant change,
one can consider pathways beyond niche innovation. Spe-
cifically, one can seek more modest changes that involve
integration as opposed to innovation. For example, inte-
grating simple and well-known technologies such as effi-
ciency devices into existing sociotechnical configurations
can act as “mediating technologies”that enable users to
shift their relationships to existing infrastructures without
making access to essential services more burdensome or
uncertain. Such technologies can enable transitions in
network performance without the wholesale displacement
of one system by another [see Ref. [74]]. Accounting for this
means an engagement with the social dimensions of
transitions as well as those that are technical and economic.
Concepts of momentum and transition can be translated
from their common usage in relation to studies of the North
to enrich analyses of networked infrastructure in the South.
Yet, for these studies, the role of the social in the socio-
technical must take both a geographical and a critical turn.
That is, studies must recognize that the manifold and
sometime conflicting relationships that people have to
networks reflect the inequities in a given society [4,39].In
LTS studies, these relationships are generally neglected even
where the role of users is given precedence. In socio-
technical systems of disrepair, divers actors develop not only
“customs”but also interests in relation to the status quo;
users are not alone in creating momentum around mal-
function. In India, for example, when local councilors also
act as private water vendors their conflicting economic and
political interests inhibit the improvement of water supply
[81]. As such, the power relations that perpetuate mo-
mentum and regulate transitions are not simply the purview
of “system builders”but extend to historical and contem-
porary political and economic actors at a variety of scales.
Recent studies of access to infrastructure services in ge-
ography have been particularly apt in addressing these is-
sues. Studies, for example, have underscored the
importance of historical and contemporary practices of
injustice in the perpetuation of sociotechnical inequality,
including colonial and contemporary forms of discrimina-
tory planning [39,82]. Engaging with such issues is neces-
sary to enrich studies of sociotechnical relationships, and to
mobilize insights from STS without losing sight of the power
and exclusion that is inherent in water’s production [39].
This paper emphasized the importance of coexistence
and the transferability of certain concepts like momentum
and transition, while underscoring the need to decouple
them from assumptions of uniformity, dominance, and
invisibility. This is of relevance in the North as well as the
South. For the particular case of water supply, widespread
concepts that are challenged by experiences common in
the South include “natural monopolies”[see Refs. [2]],
universal networks as opposed to sociotechnical co-
existence, black-boxing versus user co-production, and
the conflation of stability (or momentum) with dominance
and uniformity. By engaging LTS and MLT with infra-
structural conditions common in the South, steps can be
taken toward a more critical approach to infrastructure that
continues to account for the mutual construction of social
and technical worlds.
References
[1] WHO/UNICEF. Progress on sanitation and drinking-water 2010 up-
date. Geneva: WHO/UNICEF; 2010.
[2] Budds J, McGranahan G. Are the debates on water privatization
missing the point? Experiences from Africa, Asia and Latin America.
Environ Urbanization 2003;15:87–113.
[3] Bakker K, Kooy M, Shofiani NE, Martijn E-J. Governance failure:
rethinking the institutional dimensions of urban water Supply to
poor households. World Dev 2008;36:1891–915.
[4] Jaglin S, Zérah M-H. Eau et ville: Repenser des services en mutation.
Introduction. Revue Tiers Monde 2010;3:7–22.
K. Furlong / Technology in Society 38 (2014) 139–147 145
Author's personal copy
[5] Graham S, Marvin S. Splintering urbanism: networked in-
frastructures, technological mobilities, and the urban condition.
New York: Routledge; 2001.
[6] Coutard O. Services urbains: la fin des grands réseaux? In:
Coutard O, Lévy J-P, editors. Écologies urbaines. Paris: Économica:
Anthropos; 2010.
[7] Jaglin S. Vingt ans de réformes dans les services d’eau urbains
d’Afrique subsaharienne: une géographie de la diversité. Cybergeo
Eur J Geogr; 2004:2–5.
[8] Bakker K. Archipelagos and networks: urbanization and water pri-
vatization in the south. Geogr J 2003;169:328–41.
[9] Jaglin S. Differentiating networked services in Cape Town: echos of
splintering urbanism? Geoforum 2008;39:1897–906.
[10] Graham S, Thrift N. Out of order –understanding repair and
maintenance. Theory Cult Soc 2007;24:1–25.
[11] Robinson J. Ordinary cities: between modernity and development.
Abington, UK and New York: Routledge; 2005.
[12] Monstadt J. Conceptualizing the political ecology of urban in-
frastructures: insights from technology and urban studies. Environ
Plan A 2009;41:1924–42.
[13] Meehan K. Tool-power: water infrastructure as wellsprings of state
power. Geoforum. See http://www.sciencedirect.com/science/
article/pii/S0016718513001747; 2013.
[14] Loftus A. Reification and the dictatorship of the water meter. An-
tipode 2006;38:1023–45.
[15] Geels FW. Transformations of large technical systems –a multilevel
analysis of the Dutch highway system (1950–2000). Sci Technol
Human Values 2007;32:123–49.
[16] Fuller S. Thomas Kuhn: a philosophical history for our times. Chi-
cago & London: The University of Chicago Press; 2000.
[17] Geels FW. Feelings of discontent and the promise of middle range
theory for STS –examples from technology dynamics. Sci Technol
Human Values 2007;32:627–51.
[18] Tarr JA, Dupuy G. Technology and the rise of the networked city in
Europe and America. Philadelphia, PA: Temple University Press; 1988.
[19] Coutard O. The governance of large technical systems. London:
Routledge; 1999.
[20] Hughes TP. The evolution of large technological systems. In:
Bijker WE, Hughes TP, Pinch TJ, editors. The social construction of
technological systems new directions in the sociology and history of
technology. Cambridge and London: Cambridge University Press;
1987. pp. 51–82.
[21] David PA. Heros, Herds and Hysteresis in Technological History:
Thomas Edison and ‘The Battle of the Systems’reconsidered. Ind
Corp Change 1992;1:129–80.
[22] Mayntz R, Hughes TP. The development of large technical systems.
Boulder, CO: Westview; 1988.
[23] Kemp R, Schot J, Hoogma R. Regime shifts to sustainability through
processes of niche formation. The approach of strategic niche
management. Technol Anal Strat Manag 1998;10:175–95.
[24] Geels FW. Co-evolution of technology and society: the transition in
water supply and personal hygiene in the Netherlands (1850–1930)
–a case study in a multi-level perspective. Technol Soc 2005;27:
363–97.
[25] Genus A, Coles AM. Rethinking the multi-level perspective of
technological transitions. Res Policy 2008;37:1436–45.
[26] Van Vliet B, Chappells H, Shove E. Infrastructures of consumption:
environmental innovation in the utility industries. London: Earth-
scan; 2005.
[27] Verbong G, Christiaens W, Raven R, Balkema A. Strategic niche
management in an unstable regime: biomass gasification in India.
Environ Sci Policy 2010;13:272–81.
[28] Swyngedouw E, Kaïka M. Urban water: a political ecology
perspective. Built Environ 2002;28.
[29] Coutard O. Placing splintering urbanism: introduction. Geoforum
2008;39:1815–20.
[30] Pflieger G, Matthieussent S. Water and power in Santiago de Chile:
socio-spatial segregation through network integration. Geoforum
2008;39:1907–21.
[31] Jaglin S. The right to water versus cost recovery: participation,
urban water supply and the poor in sub-Saharan Africa. Environ
Urbanization 2002;14:231–45.
[32] Gandy M. Landscapes of disaster: water, modernity, and urban
fragmentation in Mumbai. Environ Plan A 2008;40:108–30.
[33] Hunter PR, Zmirou-Navier D, Hartemann P. Estimating the impact
on health of poor reliability of drinking water interventions in
developing countries. Sci Total Environ 2009;407:2621–4.
[34] Angrill S, Farreny R, Gasol CM, Gabarrell X, Viñolas B, Josa A, et al.
Environmental analysis of rainwater harvesting infrastructures in
diffuse and compact urban models of Mediterranean climate. Int J
Life Cycle Assess 2012;17:25–42.
[35] COMPES. Documento COMPES 3470: Importancia estrategica del
plan de inversiones para garantizar la prestación de los servicios de
acueducto, alcantarillado y aseo en el municipio de Quibdó. Bogotá:
Consejo Nacional de Política Económica y Social, República de
Colombia y Departamento Nacional de Planeación; 2007.
[36] Poveda G, Mesa OJ. On the existence of Lloró (the Rainiest Locality
on Earth): enhanced ocean-land-atmosphere interaction by a low-
level jet. Geophys Res Lett 2000;27:1675–8.
[37] Ghisi E, Montibeller A, Schmidt RW. Potential for potable water
savings by using rainwater: an analysis over 62 cities in southern
Brazil. Build Environ 2006;41:204–10.
[38] Furlong K. The dialectics of equity: consumer citizenship and the
extension of water Supply in Medellín, Colombia. Ann Assoc Am
Geogr 2013;103:1176–92.
[39] Swyngedouw E. Social power and the urbanization of water. Ox-
ford: Oxford University Press; 2004.
[40] Gulyani S, Talukdar D, Kariuki RM. Universal (non)service? Water
markets, household demand and the poor in urban Kenya. Urban
Stud 2005;42:1247–474.
[41] Srinivas S, Sutz J. Developing countries and innovation: searching
for a new analytical approach. Technol Soc 2008;30:129–40.
[42] Mithraratne N, Vale R. Rain tanks or reticulated water supply?. In:
Proceedings of Rainwater and Urban Design. Sydney, Australia;
2007. pp. 1–8.
[43] Vaes G, Berlamont J. The effect of rainwater storage tanks on design
storms. Urban Water 2001;3:303–7.
[44] Shaw IGR, Meehan K. Force-full: power, politics and object-oriented
philosophy. Area 2013;45:216–22.
[45] Gandy M. Concrete and clay: reworking nature in New York city.
Cambridge, MA: MIT Press; 2002.
[46] Carse A. Nature as infrastructure: making and managing the Panama
Canal watershed. Soc Stud Sci 2012;42:539–63.
[47] Karvonen A. Politics of urban runoff: nature, technology, and the
sustainable city. Cambridge, MA: MIT Press; 2011.
[48] Moss T. Battle of the systems? Changing styles of water recycling in
Berlin. In: Guy S, Marvin S, Moss T, editors. Urban infrastructure in
transition: networks, buildings, plans. London: Earthscan; 2001.
pp. 43–56.
[49] Summerton J. Introductory essay: The systems approach to tech-
nological change. In: Summerton J, editor. Changing large technical
systems. Boulder, CO: Westview Press; 1994. pp. 1–21.
[50] Hughes TP. Technological momentum. In: Smith MR, Marx L, edi-
tors. Does technology drive history? The dilemma of technological
determinism. Cambridge, MA: MIT Press; 1994. pp. 103–13.
[51] Callon M. Techno-economic networks and irriversibility. In: Law J,
editor. A sociology of monsters: essays on power, technology and
domination. London and New York: Routledge; 1991. pp. 132–64.
[52] Hinchliffe S. Technology, power, and space –the means and ends of
geographiesof technology. Environ Plan D Soc Space 1996;14:659–82.
[53] Barnes J, Alatout S. Water worlds: Introduction to the special issue
of social studies of science. Soc Stud Sci 2012;42:483–8.
[54] Trentmann F. Disruption is normal: blackouts, breakdowns and the
elasticity of everyday life. In: Shove E, Trentmann F, Wik R, editors.
Time, consumption and everyday life: practice, materiality and
culture. Oxford & New York: Berg; 2009. pp. 67–84.
[55] Star SL. The ethnography of infrastructure. Am Behav Sci 1999;43:
377–91.
[56] Shove E. Comfort, cleanliness þconvenience: the social origins of
normality. Oxford & New York: Berg; 2003.
[57] Barnett C, Cloke P, Clarke N, Malpass A. Consuming ethics: articu-
lating the subjects and spaces of ethical consumption. Antipode
2005;37:23–45.
[58] Cupples J. Shifting networks of power in Nicaragua: relational ma-
terialisms in the consumption of privatized electricity. Ann Assoc
Am Geogr 2011;101:939–48.
[59] McFarlane C. Infrastructure, interruption and inequality: urban life
in the global south. In: Graham S, editor. Disrupted cities: when
infrastructure fails. New York: Routledge; 2010. pp. 131–44.
[60] Warde A, Shove E, Southerton D. Convenience, schedules and sus-
tainability. In: European Science Foundation Workshop on Con-
sumption, Everyday Life and Sustainability. Lancaster University;
1998. pp. 27–9. March.
[61] Shove E. Converging conventions of Comfort, cleanliness and con-
venience. J Consumer Policy 2003;26:395–418.
[62] Berkhout F, Verbong G, Wieczorek AJ, Raven R, Lebel L, Bai X. Sus-
tainability experiments in Asia: innovations shaping alternative
development pathways? Environ Sci Policy 2010;13:261–71.
K. Furlong / Technology in Society 38 (2014) 139–147146
Author's personal copy
[63] Berkhout F, Angel D, Wieczorek AJ. Asian development and sus-
tainable socio-technical regimes. Technol Forecast Soc Change
2009;76:218–28.
[64] Smith A. Transforming technological regimes for sustainable
development: a role for alternative technology niches? Sci Public
Policy 2003;30:127–35.
[65] Raven RPJM. Strategic niche management for biomass: a compara-
tive study on the experimental introduction of bioenergy technol-
ogies in the Netherlands and Denmark. Eindhoven: Technische
Universiteit Eindhoven; 2005.
[66] Latour B. Science in action: how to follow scientists and engineers
through society. Cambridge, MA: Harvard University Press; 1987.
[67] Rip A, Kemp R. Technological change. In: Rainer S, Malone E, editors.
Human choice and climate change. Columbus, OH: Battelle Pres;
1998. pp. 327–99.
[68] Ayoub GM, Malaeb L. Impact of intermittent water supply on water
quality in Lebanon. Int J Environ Pollut 2006;26:379–97.
[69] LeChevallier M, Gullick R, Karim M, Friedman M, Funk J. The potential
for health risks from intrusion of contaminants into the distribution
system from pressure transients. J Water Health 2003;1:3–14.
[70] Tokajian S, Hashwa F. Water quality problems associated with
intermittent water supply. Water Sci Technol 2003;47:229–34.
[71] Spaargaren G, Van Vliet B. Lifestyles, consumption and the envi-
ronment: the ecological modernization of domestic consumption.
Environ Polit 2000;9:50–76.
[72] Akrich M. De-scription –how technologies are used to attempt to
inscribe user behaviours. In: Bijker WE, Law J, editors. Shaping
technology/building society: studies in sociotechnical change.
Cambridge, MA: MIT Press; 1992. pp. 205–24.
[73] Ormerod KJ, Scott CA. Drinking Wastewater: public trust in potable
reuse. Sci Technol Human Values 2012;00(on-line first):1–23.
[74] Furlong K. Small technologies, big change: rethinking infrastructure
through STS and geography. Prog Hum Geogr 2011;34:460–82.
[75] Geels FW. Technological transitions and system innovation: a co-
evolutionary and socio-technical analysis. Cheltenham, UK:
Edward Elgar; 2005.
[76] Harman G. Prince of networks: Bruno Latour and metaphysics.
Melbourne: re.press; 2009.
[77] Latour B. Irréductions in Pasteur: guerre et paix des microbes suivi
de Irréductions. Paris: La Découverte & Syros; 2001 (First published
in 1984 as Microbes: guerre et paix by Éditions Anne-Marie
Métailié).
[78] van de Poel I. The transformation of technological regimes. Res
Policy 2003;32:49–68.
[79] Hess DJ. Neoliberalism and the history of STS theory: toward a re-
flexive sociology. Soc Epistemol 2013;27:177–93.
[80] Trejos Gómez CL, Garcia-Zamor J-C, Castaño Rojas JM. Domestic
wastewater management in a rural community in Colombia. Comp
Technol Transf Soc 2008;6:212–35.
[81] Angueletou-Marteau A. Les petits opérateurs privés dans la chaîne
d’approvisionnement d’eau potable dans les petites et moyennes
villes indiennes. Revue Tiers Monde 2010;3:141–58.
[82] Kooy M, Bakker K. Splintered networks: the colonial and contem-
porary waters of Jakarta. Geoforum 2008;39:1843–58.
Kathryn Furlong is an Assistant Professor in Geography at the Université
de Montréal. She holds the Canada Research Chair in Urban Water and
Utility Governance. Her work engages with utility policy, urban geogra-
phy, and infrastructure management.
K. Furlong / Technology in Society 38 (2014) 139–147 147
