Artifact

ISSN: 1749-3463 (Print) 1749-3471 (Online) Journal homepage: www.tandfonline.com/journals/saif20

Translations: Artifacts from an Actor-Network

Perspective

JohnShigaPh.D. Candidate

To cite this article: JohnShigaPh.D. Candidate (2007) Translations: Artifacts from an Actor-
Network Perspective, Artifact, 1:1, 40-55, DOI: 10.1080/17493460600658318

To link to this article: https://doi.org/10.1080/17493460600658318

Published online: 30 Aug 2007.

Submit your article to this journal

Article views: 8416

View related articles

Citing articles: 6 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=saif20
TRANSLATIONS: ARTIFACTS FROM AN ACTOR-NETWORK
PERSPECTIVE

John Shiga, Carleton University

iPods, MP3s and file-sharing networks perform a series of actions that are often reserved for Keywords: Actor-network theory,
human agents, such as the intellectual and taste-driven labor involved in selecting, sequencing, Nonhuman agency, Politics of design,
and rediscovering forgotten sound recordings. At the same time, the familiar understanding of Digital audio, File-sharing, Copyright
artifacts as stable, material, objective things ‘‘out there’’ is also being eroded by the infinite law
replicability, malleability, and ephemeral flickering of things online. These trends lead to questions
regarding the ontological status of artifacts and reopen the question of how to distinguish
technical and material artifacts from human and social relations. In this article, the author explores
actor-network theory’s (ANT) concept of translation, which advances an alternative framework for
understanding the role of artifacts in everyday life.

Introduction: Do artifacts ‘‘act’’? framework for engaging with these founda-

In the overlapping designs of compression tional questions regarding the role of artifacts
algorithms, online music distribution systems, in contemporary life. The article beings by
MP3 players, and features like random shuffle, reconstructing ANT around the concept of
there is a new space opening up for popular, translation, a step which, I argue, is necessary
journalistic and academic discussions of the in order to redirect analytic attention to the
activeness of artifacts in social and cultural life. processes through which claims about society,
These trends raise important questions con- nature, and technology are transformed into
cerning the ontological status of artifacts: How facts and artifacts. Despite the rhetoric of
can we distinguish technical and material dematerialization and virtuality, I argue that
artifacts from human and social relations? the ANTian concept of artifacts as embodied
How do we locate agency in a world where actions and knowledges remains useful for Correspondence: John Shiga, Ph.D. Candi-
capacities to act are distributed across a wide studying the politics of online artifacts. The date, School of Journalism & Communica-
array of materials? In order to recognize the article then turns to the controversy in the tion, Carleton University, 1125 Colonel By
role of artifacts in constituting the social sociology of science and technology concern- Drive, Ottawa, ON, K1S 2H4, Canada.
world, do we need a notion of nonhuman ing nonhuman agency. ANT’s notion of non- E-mail: [email protected]
agency? Using industry statements, news re- human agency is distinguished by its
ports, and technical papers on iPods, data ontological claim that the social world is Published online 2006-04-21
compression codecs, and copy protection tech- constituted by humans and artifacts and that ISSN 1749-3463 print/ ISSN 1749-3471
niques, this article demonstrates how actor- society should therefore be studied symmetri- DOI: 10.1080/17493460600658318

network theory (ANT) provides a useful cally with regard to human and nonhuman # 2007 Taylor & Francis

//ARTIFACT VOL 1 ISSUE 1 2007 40

entities. Finally, the article attends to critiques as a fixed property of certain entities.3 The do not) they talk of the technical. And,
of nonhuman agency that have important linguistic metaphor of translation emphasizes if it appears, the technical acts either as
implications for the application of ANT’s the manner in which entities’ interests, goals, a kind of explanatory deux ex machina
notion of artifacts as associations of hetero- or desires are represented, simplified, and (technological determinism). Or it is
geneous entities to the study of digital artifacts transformed in the production and mobiliza- treated as an expression of social
and technologies. tion of artifacts. Reconstructing ANT as a relations (social reductionism). Or
theory of translation puts into view the dis- (with difficulty) the two are treated
Reconstructing ANT around the concept tributional processes through which knowledge as two classes of objects which
of translation and action come to be embodied by a collec- interact and mutually shape each other.
What is today referred to as actor-network tive of humans and artifacts.4 (1991, p. 8)
theory or ANT was initially called the sociol- In early ANT literature, translation refers to
ogy of translation. As ANT became increas- The theory of translation was developed to
a process through which one or a few actors
ingly popular in sociology and other fields in overcome these problems in sociological ap-
become spokespersons for a multitude of
the late 1980s and 1990s, the notion of proaches to science and technology. It was
others by defining and linking their identities
translation receded to the background (Latour, introduced to English-speaking audiences with
in increasingly simplified and fixed forms. The
1999a; Law, 1999). Both concepts were devel- the translation of Callon’s (1980) essay on
concept of translation was quickly adopted
oped to stress the heterogeneity of the social scientific problematization, a process that he
and transformed into a broader theory of
world, the distribution of agential properties studied in the development of a public
private
social order and power by a group of sociol-
across the human/nonhuman divide, and the research program on electric vehicles in
ogists of science and technology, most notably
processes through which socio-technical col- France.6 Callon (1986a,b) later defined pro-
Bruno Latour, John Law, Madeleine Akrich,
lectives extend themselves.1 However, when blematization as the first stage in a series of
and Michel Callon.5 These theorists offered
divorced from the concept of translation, the actions by which an actor makes itself indis-
the sociology of translation as an alternative to
actor-network becomes at best a synonym for pensable to others. To make entities accept the
constructivist and realist explanations of scien-
other kinds of social, technical, or commu- inter-definition of their identities in the pro-
tific knowledge, which require the analyst to
nication networks and at worst a restatement blematization, the principal actor uses various
take either ‘‘nature’’ or ‘‘society’’ as a given
of the agency-structure debate.2 By relocating strategies and devices of interessement. Seduc-
(Callon & Latour, 1992). As John Law argues:
translation closer to the centre of ANT, this tion, force, and persuasion, for example, may
section attempts to recover the ANTian con- [Sociologists] talk of the social. And be used to stabilize the entities’ identities, set
ception of agency as a distribution rather than then (if they talk of it at all which most parameters for their interaction and cut them
1
‘‘In order to describe such heterogeneous worlds and their dynamics in general terms, we introduced the notion of the actor-network. This concept is important in
part because it overcomes the macro-micro distinction: actor-networks may either grow, or decline, in size. Indeed, the strategies of scientists, as well as those of other
actors, can be characterized as attempts to make relevant actor-networks grow’’ (Callon, Law & Rip, 1986, p 224).
2
One might say downplaying the role of the actor-network concept in ANT does more than merely reconstruct ANT and transforms it beyond recognition. But the
actor-network concept has been the weak link in ANT’s repertoire of concepts, as suggested by Collins and Yearley (1992a, 1992b). Compounding the difficulties of its
apparent resonance with ‘‘agency-structure’’, Callon, Latour, and Law frequently use ‘‘actor’’, ‘‘agent’’, and ‘‘actant’’ interchangeably, thus making it all too easy to
misunderstand the ‘‘network’’ in ‘‘actor-network’’ as a synonym for ‘‘society’’ or ‘‘social structure’’. Finally, Latour points out that the application of the actor-
network concept to digital communication networks has diminished the sense in which actor-networks are composed through the transformation of the ‘‘nodes’’
linked together, and not through the instantaneous transmission of information between nodes (Latour & Crawford, 1993).
3
In contrast to translation, the actor-network seems to impose a particular topology of agency or agency effects onto the actors (Law & Mol, 1994) and risks
overlooking the way actors use network metaphors themselves, often to exclude and define insides and outsides (Riles, 2000).
4
In addition to the problems mentioned here, the network metaphor imposes a particular shape or structure on social relations. This is completely contrary to
ANTian methodology. According to Latour (1999a): ‘‘ANT does not tell anyone the shape that is to be drawn
circles or cubes or lines
but only how to go about
systematically recording the world-building abilities of the sites to be documented and registered’’ (p. 21). Indeed, in ANT literature there are countless diagrams and
figures that are meant to demonstrate the variability of the shape and form that actor-networks may take (as well as the many possible ways analysts may depict them).
5
Since much of Akrich’s work has not been translated into English, I have concentrated primarily on the work of Callon, Latour, and Law. This unfortunately
replicates a blind-spot in Anglo-American work on ANT, which excludes Akrich from ANT’s ‘‘first wave’’. In fact, her descriptive techniques and semiotics of
machines had a significant influence on ANT (Latour as Johnson, 1988, p. 305
306). See especially Akrich and Latour’s (1992) repertoire of terms for describing the
association and substitution of action between humans and nonhumans through the use of scripts in material settings.
6
Callon (1991) traces the concept back to work published in 1976, although it has not yet been translated into English.

41 //ARTIFACT VOL 1 ISSUE 1 2007

off from alternative definitions of their iden- Translating interests: iTunes, iPods and fying and displacing interests and goals into a
tities (Callon, 1986b). These tactics, if success- MP3s set of relations between cognitive and social
ful, lead to the enrolment of entities within a According to the theory of translation, it entities (p. 215). A problematization that states
program of action. The definitions of interests, should be possible to trace the iPod, like any that there are logical links between a set of
goals, and identities are then mobilized through other artifact, back to set of problematizations. problems (that their sequential solution will
representational techniques and the physical As Callon (1980) noted in his study of resolve the larger problem defined at the
displacement of entities. In general, the process scientific problematization, these zones of outset) ‘‘is to state that a community of
of translation tends towards the association, uncertainty are not defined through logical interests exists’’ between a set of groups and,
combination, and simplification of entities and deduction from the existing body of knowl- moreover, that ‘‘social interaction is conceiva-
the reduction of representatives to one or a few edge, nor are they the result of straightforward ble between them’’ (p. 211). Problematizations
actors. It transforms weak, provisional, and political or economic influences.8 Rather, a for Callon are attempts to define and mobilize
generally defined identities into durable and heterogeneous array of elements
technical social groups while simultaneously constrain-
seemingly irreversible ties. If translation is artifacts, notions of what sort of society would ing the number of possible responses (and
successful, the principle actor ‘‘speaks for need or support the production of portable sequences of responses) from these groups.
others but in its own language’’ (Callon, MP3 players, the attributes of the imagined Actors deploy strategies that work upon
1986a, p. 26). consumers that would be interested in these (transform, link, merge, displace) interests in
The boundaries between the various stages artifacts, and so forth
were deployed along- such a way that the other entities consent to
of translation are perhaps more fluid than side principles of psychoacoustics and the the imposition of the problematization because
Callon’s template implies, but it is useful behavior of different kinds of digital memory they already appear to be implicated in it. The
for describing the transformation of claims in constructing the grids of certainty that translation or inter-definition of interests in-
and projects into technological facts and frame research problems. duces consent by proposing relations between
artifacts in a way that does not take social, In principle, each engineer in Apple’s secre- problems/groups as well as by accommodating
natural, or technological reality as givens. tive ‘‘digital hub’’ project could have defined and constraining how the next actor will orient
Using the framework of translation, attention the problematic area differently due to her or him/herself towards the uncertainty imposed
can be directed towards the transformative his particular position within a set of relation- upon it. For example, Apple’s expansion into
processes through which entities are com- ships with laboratories, companies, areas of music, video, and photography in the early
bined and linked with others. As Latour expertise, and so on. This raises the question of 2000s revolved around a sequence of proble-
(writing as ‘‘Jim Johnson’’, 1988) put it: ‘‘I why actors come to accept others’ definitions matizations that established Apple as an ob-
use translation to mean displacement, drift, of the problem and the prescribed actions and ligatory passage point for those who wanted to
invention, mediation, the creation of a link roles within it. Callon’s (1980) answer, which collect, organize, and share digital content.
that did not exist before and that to some he derives from a combination of textual The popular iTunes music software can be
degree modifies two elements or agents’’ (p. analysis of laboratory documents and partici- traced back to a problematization of techno-
32).7 While one could use this framework to pant observation, is that problematizations economic reality that initially generated resis-
examine the various functions that an arti- follow one another not by logical deduction tance (rather than consent) in the upper ranks
fact acquires over time, it also directs analy- but by association, or the socio-logic of trans- of Apple. Sales of portable MP3 players lagged
tic attention on the stages prior to the lation. Each problematization is an attempt to behind the enormous popularity of the MP3
emergence of a discrete artifact or self- ‘‘induce consent and provoke resistance in format on the web because, as Apple’s vice-
evident technological fact. various groups’’ by defining, equating, simpli- president Greg Joswiak put it, ‘‘The product

7
Latour (1988) wrote under the pseudonym Jim Johnson in this particular article. The actor-network theorists used the term ‘‘translation’’ in a variety of ways. In
some cases, it seems to refer to a stage in a broader process of coordinating and mediating action. In other cases, it is used as a kind of shorthand for the entire process
of coordinating action. Latour (1994a) later used translation to refer to a stage in the extension of social fabric to nonhumans. Translation is ‘‘the means by which we
inscribe in a different matter features of our social order’’ (p. 45), setting the stage where nonhumans receive human properties, and are enrolled into and mobilized
within the collective. At other times, translation is used to refer to this whole process of collective formation, modification, and movement/action, as in the ‘‘socio-logic
of translation’’ (Callon, 1980).
8
Latour & Woolgar (1979/1986) similarly found that money itself was not the motivation for scientific activities, even though actors consistently understood their
work, associations, and productions as investments.

//ARTIFACT VOL 1 ISSUE 1 2007 42

[made by other companies] stank’’ (Barker, identity to certain legal, economic, and techno- sciences as residing within the human universe,
2005). Tony Fadell, an Internet consultant, logical problems, such as ‘‘shrinking revenues, not that of ‘‘things’’. Second, one might point
modified this problematization: how can MP3 downsizing work forces, and the threat of out that the tactics of interessement are
players foster the sale of music online (Barker, extinction from an apparently unstoppable deployed by human actors (e.g. Apple’s
2005)? Fadell’s problematization does not force called downloading’’ (Barker, 2005). The CEOs, marketing directors, etc.) and that
simply redefine the problem to be solved but most important point here is that the goals of counting the number of iPods that work for
changes the identities of various products and the principal actor drift as increasing numbers 10 hours as opposed to those that work only
services in a highly strategic way. iTunes of entities are drawn into the program of action. for three does not amount to a kind of
becomes an online music service and tool for In this case, Apple’s goals drift from the nonhuman ‘‘vote’’ for the program of action.
selling and promoting Apple’s other products production of computers and software towards While artifacts may in certain ways ‘‘act’’,
rather than an application for playing music the sale of digital music and the development of sociologists are hesitant to refer to them as
acquired elsewhere. MP3 players become ex- a line of portable MP3 players. social actors since they lack the (human)
tensions of an online music store rather than Before a collective can be mobilized, the capacity to choose between various courses
devices for playing music downloaded from links between entities and the summarization of action according to their intentions or
file-sharing networks or ripped from CDs. of their interests through representational interests. Finally, it could be argued that
Legal, economic, cultural, and technological chains (e.g. statistical analyses, prototypes, notions of nonhuman agency diminish the
realities were used in the construction of this etc.) have to be tested. Will file-sharers pay capacity of analysts to trace the origins or
problematization and durable links were estab- for music? Do subscribers to music services causes of collective action and therefore under-
lished between seemingly disparate entities (e.g. really want to ‘‘own’’ rather than ‘‘rent’’ music? cut attempts to make people accountable for
record companies, soft drink manufacturers, Will record companies accept their role as their actions and those of their artifacts.
musicians, football fans, and users of unsanc- ‘‘content providers’’ and relinquish their oli- ANT does not deny that there are differ-
tioned file-sharing networks). The enrollment of gopolistic control over the system of distribu- ences between human and nonhuman entities,
these entities required the modification of their tion? In this case, all the entities accepted their but it does challenge the asymmetrical view of
identities and the use of a variety of means of roles with one exception: the iPod. The por- the social world as constituted by human
interessement: seduction, persuasion, and the table MP3 player drew power from the bat- actors who impose their will upon passive
force of law. Perhaps the most visible device of teries even when it was switched off. As one artifacts. As Callon explains:
interessement in this case was Pepsi’s ‘‘I Fought engineer put it, ‘‘The production lines had
Considered from a very general point of
the Law’’ Super Bowl advertisement in 2004 in already been set up. That was a tense part of
view, [translation] postulates the exis-
which 16 teenagers sued by the Recording the project. For eight weeks they thought they
tence of a single field of significations,
Industry Association of American explained had a three-hour MP3 player’’ (Barker, 2005).
concerns and interests, the expression of a
how they could continue to get free music on the To persuade the iPod to accept its role as a
shared desire to arrive at the same result.
Internet by obtaining the winning codes printed portable player with vast memory and a long
Though translation recognizes the exis-
on soft drink caps and redeeming them at battery life, one of its characteristic design
tence of divergences and differences that
Apple’s iTunes music store (Walker, 2004, features
a small hard drive with a relatively
cannot be smoothed out, it nevertheless
p. F07). Apple thus sets itself up not so much large storage capacity
had to be compro-
affirms the underlying unity between ele-
as a computer hardware/software company but mised. The result was that the iPod did not
ments distinct from one another. Transla-
as an obligatory passage point for those seeking differ radically from other MP3 players since it
tion involves creating convergences and
‘‘free music’’. Subscribers to competing music too required a dose of flash memory to over-
homologies by relating things that were
services were attributed with interests, wants, come its hunger for power.
previously different. (1980, p. 211)
and desires that were only partially fulfilled by The last part of this analysis would be highly
these other music services. As iTunes marketing problematic for some social scientists on a The performance of a particular definition
director Peter Lowe put it, ‘‘They want to have number of grounds. First, it suggests that of reality depends on the strength of the
the flexibility to do what they want with it. They power is exercised upon nonhuman entities links between heterogeneous entities and their
don’t want to rent music’’ (Ryan, 2003). Apple and that they are therefore sites of potential enrolment and mobilization into a singular
also set itself up as an obligatory passage point resistance. This does not conform to widely course of action. When the iPod’s hard drive
for the music industry by reducing the industry’s held conceptions of agency in the social behaved differently from the way it had been
43 //ARTIFACT VOL 1 ISSUE 1 2007
represented, it threatened to shut down the online artifacts. In this section, I argue that the collective action to a relatively few ‘‘point
assembly lines as quickly as they had been set theory of translation remains relevant to the locations’’, that is, ‘‘places or points that last,
up. Our understanding of artifacts can be study of the politics of digital artifacts. An that keep on going for a time’’ (Callon & Law,
enriched by attending to the possibilities of analytic framework based on translation ex- 1995, p. 497). An actor, regardless of size,
resistance that emerge from the representation amines the design of digital artifacts as the complexity, or species, begins to acquire
and imposition of identities on entities that exchange of actions, skills, goals, and capa- agency (capacities to perform a list of actions)
seem merely technical. In this framework, cities between human and nonhuman entities, through the primary translation mechanism in
Latour (1999b) writes, ‘‘the prime mover of what John Law (1991) calls ‘‘heterogeneous which a principal actor forms associations with
an action becomes a new, distributed and engineering’’ (p. 9). Such a notion of design, it other entities. Agency may be ascribed to a
nested series of practices whose sum can add seems to me, is much more relevant to the point within that collective through a second-
up but only if we respect the mediating role of kinds of sociality, power techniques, and ary mechanism of translation. This secondary
all the actants mobilized in the series’’ (p. 181). empowerment networks emerging from con- mechanism has no necessary relation to the
By replacing the sociological notion of actor temporary socio-technical collectives. primary mechanism (Latour, 1987). In other
with the semiotic notion of actant, we have a The theory of translation leads to a quite words, attributions of responsibility, account-
framework in ANT for understanding agency specific notion of artifacts as constituents of ability, and intentionality do not always reflect
as a set of capacities that are structured like a (rather than supplements to) the social world. the actual processes through which the collec-
language and which can therefore be translated However, the theory of translation postulates a tive action was composed in the first place. In
into various material settings. The next section form of nonhuman agency that is distinct from this sense, an actor is a network; its identity is
explores the implications of this concept of that of posthumanism and cyborgism wherein constituted by its epistemic relations with
agency for ‘‘virtual’’ artifacts. nonhuman agency is understood to be a ‘‘others’’ that persist over time (Hetherington,
relatively recent cultural phenomenon. As 1997, p. 195).
Digital artifacts, materiality, and politics Latour puts it: According to Jon Rubinstein, the senior
Along with other fields and disciplines, media vice-president of hardware development at
The name of the game is not to extend
studies has recently become concerned with the Apple who assembled the design team for the
subjectivity to things, to treat humans
cultural, political, and economic consequences first version of the iPod, the development of
like objects, to take machines for social
of the dematerialization of artifacts in online the iPod ‘‘started with a clean sheet of paper’’
actors, but to avoid using the subject

environments. As Don Slater (2002) observes: (Schlender, 2001). The blank page implies that
object distinction at all in order to talk
‘‘The over-arching claim of much of that the iPod was dreamt up and realized by
about the folding of humans and nonhu-
literature has been that this dematerialization Rubinstein and his design team without bor-
mans. What the new picture seeks to

in particular, the purely textual presence of rowing ideas, theories, technical components,
capture are the moves by which any given
interacting participants, cut loose from mate- and expertise from other places. The secondary
collective extends its social fabric to
rial bodies and places
has allowed the mechanism is evident in Rubinstein’s account,
other entities. (1996b, p. 194)
possibility for creating new forms of social since he is suggesting that the iPod was built
order and identity’’ (pp. 227
228).9 If, as Design is a political activity because it breaks from scratch by a few talented humans. But his
Andreas Reckwitz (2002) claims, ANT was down a program of action into little scripts (or account erases key processes in the primary
an attempt to redeploy materiality in cultural prescriptions) which are distributed across, and mechanism of translation. For instance, one
theory and to rethink materiality in terms of embodied by, the heterogeneous materials that of the crucial tactics of interessement that
artifacts, then the emancipatory rhetoric sur- compose the social world. The production of led to the enrolment of file-sharers was the
rounding dematerialization and virtuality may an autonomous agent, whether human or compatibility of the iPod with music files
pose a formidable challenge to the application nonhuman, machinic or virtual, complex or encoded in the MP3 file format. Focusing on
of the theory of translation to the study of simple, is produced through the attribution of the primary mechanism of translation takes us

9
Lawrence Lessig’s work (1999, 2002) provides a cautionary note against the utopian strand of literature on digitalization. Lessig has repeatedly argued that
creative uses of online artifacts such as MP3 files are constrained and monitored in an unprecedented way through software patents and copyright law and through the
technical codes that regulate social and cultural activities online. In Lessig’s view, the reduced malleability and constrained replicability of online artifacts is a
consequence of the imposition of economic interests by certain powerful human actors, in particular, ‘‘largish’’ entertainment conglomerates that seek to maximize
their control over the circulation and use of online artifacts (Lessig, 2002, chap. 11).

//ARTIFACT VOL 1 ISSUE 1 2007 44

past Rubinstein’s ‘‘clean sheet’’ to the messy to the things themselves. In contrast, the theory mann, MPEG developed a series of audiovi-
world of MPEG-1 Audio Layer 3, or MP3. of translation, as discussed in the previous sual encoding standards. While Charles
Given all the talk about dematerialization, it section, troubles this view of technologies as Creusere (2003) notes that digital audio com-
is interesting that digital artifacts, like any ‘‘brute material artifacts’’ (Callon & Latour, pression has been a persistent problem for the
technical innovation, leave long paper trails 1992) that are political only insofar as they set telecommunications industry since the 1970s,
behind them as they emerge. In the case of constraints on human relationships (Latour, MPEG interested a wide variety of firms and
MP3, the paper trail is enormous, spanning 1994a, 1999b). The thrust of ANT case studies engineers by developing a new problematiza-
across the news media, law journals, computer is this: the politics of artifacts should be tion. As Musmann (1990) argued, ‘‘A coding
science articles, patent claims, and legal deci- reconceptualized to account for the actions technique which allows to reduce the bit rate of
sions. Much of this literature construes the that are routinely distributed or shifted be- a stereo sound signal down to 2 /128 kbit/s or
politics of MP3 in terms of legal controversies tween humans and nonhumans in the practices even to 2/64 kbit/s preserving a sound quality
over ownership and public access. But one does of scientists, technologists, and users. Each of comparable to that of CD would be very
not need to concentrate on corporate or legal these heterogeneous engineers acquires allies attractive to save transmission and storage
influences on copyright legislation or legal and draws power from them by transforming capacity and would facilitate the introduction
decisions to discern the power relations of ‘‘natural’’ differences between humans and of new services’’ (p. 511). Musmann sets up a
MP3. ‘‘The social’’ does not intrude upon ‘‘the nonhumans into distributions (Law, 1991).10 problem (How can sound be digitally repre-
technical’’ periodically. To break the world up Without the proliferation of unsanctioned sented using less data than CD encoding but
in this way would be to accept the modernist MP3 sites, anti-piracy campaigns, and efforts with the same sound quality?) and a grid of
notion that science and technology can and to build security features and copy protection certainties (particular input and output sam-
should be purified of politics. Translation features into music files, would MP3 have pling rates, lower bit rates, and so forth are
theory suggests instead that attempts to order politics? Using the framework of translation, necessary) to outline the ‘‘hunting grounds’’ or
the world, including the most specialized we can answer ‘‘yes’’. MP3 is political not areas to be analyzed by groups of engineers.
technical and scientific papers, are not disin- because it has led to juridical and technical Musmann then associates these problematized
terested activities but rather are highly strate- constraints on human interaction but because areas of the cognitive field with particular
gic attempts to bind humans and nonhumans it requires a redistribution of capacities across coding concepts that were developed and
together and make them accept particular the human/nonhuman divide in order to tested by groups composed of engineers em-
definitions of their capacities, roles, and empower a hybrid collective. To trace out the ployed by different electronics firms.
identities. politics of MP3, we need to revisit the scientific MP3
the audio layer of the MPEG-1
In Langdon Winner’s (1980) oft-cited article problematizations that established alliances compression standard
emerged within this
on the question of whether or not artifacts and possibilities of interaction between enti- socio-cognitive network of problematization.
have politics, he argued that the social deter- ties. The effect of these problematizations, MPEG-1 was the first ISO standard for audio
mination of technology overcompensates for according to Callon (1980), ‘‘is not to create compression, published in 1992 (Creusere,
technological determinism and leads to its own stability and order. It is to create local 2003). This standard was accepted not so
impoverished notions of causality. Social de- instability. With the creation of such instability much because it was more efficient than other
terminism suggests that ‘‘technical things do the possibility of autonomy arises’’ (p. 217). proposals, but because its proponents per-
not matter at all . . . there is nothing distinctive In 1988, the International Standards Orga- suaded their colleagues to change the defini-
about the study of technology in the first nization established a committee (ISO/IEC tion of efficiency. Until the emergence of
place’’ (p. 122). Winner offered an alternative JTC1/SC29 WG11), which was called the MPEG, efficient compression was defined by
framework based on the notion that the Motion Picture Experts Group or MPEG the maximum reduction of ‘‘quantization
politics of artifacts can be analyzed by looking (Brandenburg, 1995). Chaired by Hans Mus- noise’’, noise that results from the conversion

10
The clearest example of this point is in Latour’s (1988) study of Louis Pasteur’s mobilization of France (and much of the rest of Europe) into a program of action
which, at the time, would have seemed like a rather extreme measure (injecting people with fluids; heating all fermenting drinks, etc.). The success of Pasteurization
cannot be reduced to the sheer force of his discovery but is a result of modification of identities and the establishment of links between humans and nonhumans.
Sickness, for example, is caused by material overlaps between species. Although bacteria occupy a different place in moral, political, and ethical orders, they are
usefully understood as distributions across the human/nonhuman divide. The social world was redefined in terms of groups formed by that distribution (the sick, the
infected, the contagious, the immune, etc.).

45 //ARTIFACT VOL 1 ISSUE 1 2007

of a continuous signal (analogue) into discrete sets itself up as an obligatory passage point for object that can be controlled, chopped up,
bits (digital). The MPEG group adopted an Phillips, Matsushita, and other firms to solve reassembled, and injected with noise. Finally,
entirely different way of compressing data by their problems regarding data storage, data MP3, like all sound reproduction technologies,
allowing or injecting as much quantization transmission, and the development of new does not merely depend on the extension of
noise as possible, that is, by associating the services. Second, each coding concept is a human hearing but rather, as Jonathan Sterne
coding scheme with a perceptual model im- claim about the acceptable amount of data (2005) puts it, ‘‘on us delegating to machines
ported from psychoacoustics. By 1995, Rault that can be lost in the encoding stage and the that hear for us’’ (p. 41).
et al. confidently asserted that ‘‘it is world-wide amount of noise that can be ‘‘injected’’ into The question that remains is whether or not
accepted that the more efficient audio com- digital recording without affecting its ‘‘subjec- artifacts delegate capacities or tasks back to
pression algorithm is the one that introduces tive quality’’. To bolster these claims, the humans. In the next section, I examine how the
the maximum noise provided that it remains working groups embedded them in a collective application of political notions of delegation,
perceptually inaudible’’ (p. 3/1). The bit-reduc- (or network) of private and public institutions, spokespersons, representation and so on led to
tion techniques developed by MPEG were mathematical theories of data compression, a shift in ANT from quasi-anthropological
based on the psychoacoustic fact that a sound acoustic facts, and artifacts such the filters- studies of science and technology towards a
can be ‘‘masked’’ by a louder sound. The banks, integrated circuits, and chips that controversial theorization of society as an
representation of human hearing in the psy- encode and decode digital representations of association of human and nonhuman entities.
choacoustic model was then shifted or in- sound (Rault et al., 1995). The engineers In short, we move from a notion of nonhuman
scribed into the codec (decoder/encoder) in enrolled a psychoacoustic model of human agency as an effect of human attributions of
order to ‘‘exploit the fallibility of the human hearing, which they associated with their agency to artifacts towards a notion of agency
ear’’ (Carey & Wall, 2001, p. 36). various coding and bit-allocation techniques. that is inseparable from the distribution of
The politics of MP3 are not reducible to the The fact that sounds can be masked by other action across heterogeneous materials.
legal controversy surrounding some of its uses sounds was translated into an encoding tech-
on the Internet. Scientific controversies con- nique called psychoacoustic masking. Since no The resistance and representation of
cerning coding techniques, bit-rates, defini- one contested psychoacoustic masking, it artifacts
tions of efficiency and so forth destabilize acted as a black box. Inside this black box, Since ANT attempted to theorize artifacts as
local areas of knowledge to interest and enlist there were many previous translations includ- distributions of actions across heterogeneous
others in the performance of a particular ing a model of sound based on certain materials, it leads to certain questions about
definition of reality. This does not mean that components of the human ear, the summariza- resistance: Do human and nonhuman entities
the politics of science is reducible to the tion of ‘‘subjective quality’’ in perceptual always accept these provisional ties instan-
standardization or rationalization of hearing. audio codes, and the notion of efficient tiated in the design of an artifact? Are
Long before the legal controversy erupted, compression as the maximization of impercep- ‘‘actants’’ equal in their capacity to resist their
Musmann’s (1990) report on the ISO plan to tible noise. This model was crucial to the spokespersons? More fundamentally, does the
develop a compression standard clearly stated distribution of the capacity to hear within the notion of nonhuman agency detract from the
that the act of hearing could be divided into hybrid collective. study of empowerment and resistance in social
scripts (codes, models, algorithms) which Compression algorithms are not political organization? Steve Fuller (as cited in Barron,
would then be distributed to various human simply because they impact on social organi- 2003) finds the absence of human resistance in
and nonhuman entities. zation or because they are inscribed with the ANT literature very disturbing: ‘‘The whole
The redistribution of hearing in the devel- normative values of a corporate engineering point of social organization is specifically to
opment of MPEG codecs is an exercise of community. MP3 embodies representations of combine in ways that go against the natural
power. First, it establishes a grid of certainties the (vibrating) world and definitions of effi- course of things. In that sense, resistance and
about a phenomenon of vibration (i.e. sound) ciency which render the ‘‘lossy’’ compression conflict are what characterize the distinction
to develop an obligatory passage point (i.e. format acceptable, desirable, and unavoidable. between the human and the non-human’’
maintaining tonal relations despite compres- MPEG also enrolled a number of previous (p. 83). Fuller thus takes the ANT researchers
sion). Accepting the problematization means translations such as the psychoacoustic model to task for failing to address the capacity of
accepting a framework for interaction with the of hearing, which abstracts the senses, detaches humans to resist those who attempt to order,
other MPEG working groups. MPEG thereby the ear from the body, and treats sound as an engineer or design interactions between hu-
//ARTIFACT VOL 1 ISSUE 1 2007 46
mans or between humans and nonhuman human action. Agency is a relational concept other action verbs (‘‘decide’’, ‘‘understands’’,
entities. because it refers to the capacity to perform a ‘‘giving’’) point to a form of nonhuman
Are humans distinguished by their capacity particular action or set of actions. Agency also agency. The distinction between human and
to resist ‘‘going with the flow’’? Is this what seems to vary according to species because nonhuman action, and the privileging of the
artifacts cannot do, no matter how sophisti- entities have different capacities in the realm of former, enables social researchers to maintain
cated they may be in their other capacities? In action. In the case of humans, agency is usually critical distance from their research subjects.
Fuller’s view, yes. To complicate this view, associated with the capacity to make the world But since this distinction is routinely disre-
however, ANT argues that nonhumans may meaningful (Casper, 1994), to constitute repre- garded by technologists and users, researchers
also contest the apparent unity of the artifact sentations (Smart, 1982), or to shape the world are faced with the problem of how to integrate
or design. The resistance of nonhumans is according to one’s intentions (Bruum & Lan- these accounts of nonhuman agency into their
evident in Latour’s (1987) study of the Diesel glais, 2003). More fundamentally, human understanding of artifacts. Are users mistaking
engine, particularly during its mobilization agency is distinguished by the capacity to the actions performed by artifacts with the
through patent law, promotion, cultural and choose between different courses of action. human capacity to choose between actions? Is
industrial use: ‘‘The reality of the engine According to Henrik Bruun and Richard the analyst’s role to unveil the objective social
receded instead of progressed. . . . From a Langlais (2003), ‘‘Within the social sciences, forces that generate the illusion of iPods as
factual artefact it became, if I may use the agency is often associated with the power to artifacts that act?
two meanings at once, an artefactual artefact, choose, that is, to the power to act in one way In conventional sociology, and in Western
one of those dreams the history of technics is even though one could have acted differently’’ culture more generally, agency depends on
so full of’’ (pp. 105
106). Nonhumans are not (p. 33). both intentionality and language use, neither
easily persuaded into relations with each other In this view, it is extremely problematic to of which was considered to be a property of
or with human beings. As my discussion of the suggest that artifacts have the capacity to resist ‘‘things’’ (Callon & Law, 1995, p. 490). At the
development of the iPod suggested, nonhu- the tasks delegated to them, to exert force same time, many human actions are consti-
mans can ‘‘defect’’ from the coordination of upon the manner in which they are repre- tuted by materials outside the human body.
roles set out in business plans, blueprints, sented, or to decide how to make the world ANT researchers were interested in the possi-
prototypes, and promotion; hard drives may meaningful. Nevertheless, artifacts seem to bility that agential properties emerge from the
demand too much power and batteries can go display some of these agential properties. material overlaps between human and nonhu-
‘‘on strike’’, shutting down the assembly lines. According to one iPod user, ‘‘There is some- man action and that technological design is
But the notion that the defection of the hard thing thrilling about setting the player on about arranging, regulating, and exploiting
drive from the iPod amounts to a form of Shuffle and letting it decide what to play these overlaps through representational tech-
resistance is deeply troubling for some re- next. The little machine often goes crashing niques. The term ‘‘actor’’ obscures the distri-
searchers because it blurs the foundations on through barriers of style in ways that change bution of actions, deflecting attention to the
which studies of resistance and power in how I listen’’ (Kahney, 2004). Michael Bull, human origins and causes of action.11 Thus, a
science and technology have traditionally who has written extensively about the way term other than ‘‘actor’’ was required to talk
been carried out. By exploring the hostility in personal stereos like the Walkman enhance the about the heterogeneous embodiments of ac-
the social sciences towards the idea of nonhu- user’s sense of control over her or his environ- tion. According to Latour (1987), ‘‘both peo-
man resistance, this section elucidates the ment, mood, and subjective behavior, has ple able to talk and things unable to talk have
conceptual boundaries within which artifacts found that users attribute agency to the iPod: spokesmen [sic]. I propose to call whoever and
are permitted to act in conventional socio- ‘‘Some users feel that the machine intuitively whatever is represented actant’’ (p. 84). The
logical explanations of technology. understands them by giving them just the type term ‘‘actant’’ allows the researcher to set aside
The claim that artifacts ‘‘act’’ may not raise of music they want to listen to when they want the question of origins and causes of social
many eyebrows in the social sciences insofar as it’’ (Kahney, 2004). Bull’s description conforms action and examine the techniques of repre-
it is accompanied by a recognition that this to the notion of human agency in the social sentation, association, and combination
kind of action differs in important ways from sciences (‘‘users feel that . . . ’’) even though the through which actions are distributed and

11
In the sociological concept of ‘‘actor’’, agency is unquestionably located in the human universe, not that of things (Pels et al., 2002, p. 2; Hetherington & Law,
2000).

47 //ARTIFACT VOL 1 ISSUE 1 2007

embodied by various kinds of materials in perform certain practical knowledges, compe- lasting they all need (we all need) the
addition to the human body. tences, or skills. Latour’s (1994a) genealogy of actions of others. But what will these
There are two key senses in which artifacts techniques made this project explicit in his actions be? Many things, most of them
act in ANT. The first sense of nonhuman claim that a society characterized by divisions unpredictable, which will transform the
agency derives from ANT’s methodology and of labor, tremendous increases in scale, and so transported object or statement. So we
is congruent with more conventional socio- on, is only possible through translation tech- are now in a quandary: either the others
logical approaches to ‘‘things’’. However, over niques, here understood as the socialization of will not take up the statement or they
time, ANTian case studies gave rise to ontolo- nonhumans whereby scripts for social action will. If they don’t the statement will be
gical claims about nonhuman agency, which are embodied and performed by nonhuman limited to a point in time and space,
have generated considerable resistance within entities. The ontological and historical claim myself, my dreams, my fantasies. . . . But
the sociology of science and technology.12 concerning the agency of artifacts became if they do take it up, they might trans-
Following Foucauldian notions of power as increasingly apparent in ANT literature in form it beyond recognition. (Latour,
dispersed through bodily techniques, Law the 1990s.14 1987, p. 108)
(1986b) suggested that agency in the case of The notion of translation as the socializa- To prevent transformation beyond recognition,
the Portuguese expansion in the sixteenth tion of nonhumans was a response to an one might deploy what Latour calls ‘‘social
century needs to be understood as a network ambiguity in notion of the ‘‘social construction skills’’ by physically excluding or manipulating
of heterogeneous embodiments of knowledge of technology’’ and social deterministic views bodies (Latour, 1994a). For Latour, however,
that formed a ‘‘protective envelope’’ around of artifacts more generally. ANT bypassed the human societies are characterized by the
Portuguese ships in uncharted waters. Only in question of how social relations are mediated enrolment of extrasomatic resources into social
conjunction with mariners drilled in astron- by artifacts to examine the more fundamental ties, enabling a particular definition of society
omy and modifications in ship design did the question of why artifacts proliferate in human to become more durable. Artifacts result from
printed documents (maps, logs) acquire such societies. ‘‘If artifacts are social relations, then the delegation of social roles to nonhumans
tremendous importance in the way the Portu- why on earth has society to pass through them that perform provisional bonds between ac-
guese ‘‘interacted’’ with (and exercised power to inscribe itself onto something else?’’ asks tors: ‘‘the durability of the definition of the
upon) others without actually occupying their Latour (1994b). ‘‘Why not inscribe itself di- clan depends upon the duration of the re-
territories.13 Rather than viewing the roles of rectly?’’ (p. 793). Many of Latour’s works sources used to make it hold together’’ (1986,
humans and nonhumans in the constitution of
provide extended responses to this question. p. 275). In this framework, iPods and other
the social world asymmetrically, Law’s account
The following passage captures the ‘‘quand- ‘‘high technologies’’ are more durable and
of the Portuguese expansion tilts towards a
ary’’ most eloquently in his account of how intimate forms of the same elementary
symmetrical view of the social world as con-
Rudolf Diesel struggled to transform a muta- human
nonhuman relationship instantiated
stituted by humans and artifacts.
ble and isolated project into an immutable and by the extension of social skills to materials
ANT’s concept of nonhuman agency thus
mobile artifact: (Latour, 1994a, p. 62). In Latour’s (1993) view,
arises not only from empirical studies of
high technology can be understood in the
scientific knowledge practices, but also from If Diesel is the only person who believes
following terms:
its ontological claims that were designed to in his perfect engine, the engine sits in an
show the impossibility of a social order like office drawer in Augsburg. In order to A shifting network of actions redistribut-
our own without artifacts that embody and spread in space and to become long- ing competences and performances

12
‘‘As a principle of methodological symmetry’’, Alex Preda (2000) writes, ‘‘it just states that the sociologist has to analyze the human beings and artifacts
embedded in such a nexus as knots of socially sanctioned (and primarily tacit) knowledge, and that these kinds of knowledge are contingent upon each other’’ (p. 286).
Preda notes that a variety of approaches including ethnomethodology and the practice theory of Pierre Bourdieu accept this methodological principle and view things
as social actors insofar since they act as ‘‘knowledge-bearers’’.
13
‘‘The right documents, the right devices, the right people properly drilled
put together they would create a structured envelope for one another that ensured
their durability and fidelity’’ (Law, 1986b, p. 154).
14
This is particularly evident in Karin Knorr Cetina’s notion of postsocial relations, which has been developed on the basis of ANT’s arguments about the way
science and technology have been crucial sites for the socialization of nonhumans (Knorr Cetina, 1997; Knorr Cetina & Bruegger, 2002). She argues that the
importance of nonhumans in sociality has increased over time due to knowledge practices in science, economics, and elsewhere, and requires a redefinition of social
relations.

//ARTIFACT VOL 1 ISSUE 1 2007 48

 either to humans or non-humans in predictable. At first sight, this solution texts and other forms of inscription enable the
order to assemble into a more durable seems so contradictory as to look un- relations that make up an object to be fixed
whole an association of humans and feasible. If others are enrolled they will even while they are mobilized within a network
things, and to resist the multiple inter- transform the claims beyond recognition. by different actors according to their own
pretations of other actors that tend to Thus the very action of involving them is projects and goals. The concept refers to
dissolve this association. (p. 379) likely to make control more difficult. The techniques and materials of representation
solution to this contradiction is the such as typographical fixity and linear per-
Artifacts act not just because our research
central notion of translation. (p. 108) spective, which have enabled the creation of
subjects say they do but because the construc-
larger, more inclusive ‘‘meeting places’’ while
tion and organization of a society requires Understood as strategy of translation (or a set
at the same time conserving ‘‘a constant
more durable agents than values, ideologies, or of such strategies), design involves the enrol-
ment of others as well as the redistribution, through successive transformations of the
transient interactions between human beings.
delegation, and fixing of roles to social, medium’’ (1998, p. 425). Immutable mobiles
The theory of translation (especially the
natural, and technical entities.15 However, allow different entities to be combined, super-
stage of mobilization) postulates that actions
according to the theory of translation, as imposed, compared, simplified, and so forth
have to be physically displaced between bodies,
formulated by Callon (1986a,b), enrolling by accelerating the mobility and by enhancing
texts, and machines to empower the collective
these nonhumans will also tend to introduce the immutability of inscriptions. In short, they
and to increase the durability of particular
more uncertainty and unpredictability. Any simultaneously fix knowledge and render it
definitions of society. In this sense, more
attempt to represent or speak for others is an more portable. One particularly important
artifacts have become increasingly socially
exercise of power, which opens up the possibi- immutable mobile in the cultural industries is
active over time. However, according to the
lity of resistance. In short, translation can lead copyright law. Copyrights translate cultural
theory of the translation, the enrolment of
to ‘‘treason’’ (Callon, 1986b, p. 219). Latour’s practices and works into objects of property;
anything/anyone increases the possibility of
answer to this problem is that the continuous they enable all the materials, techniques, and
resistance or defection. Is nonhuman resis-
work of disciplining humans can be replaced (most) employees that contributed to an arti-
tance somehow more ‘‘manageable’’ than that
by the discontinuous work of installing more fact to be black-boxed or forgotten. The results
of human beings?
nonhumans. ‘‘By involving nonhumans’’, of this collective action
a discrete object like
Latour (1994a) asserts, ‘‘the contradiction a sound recording
can then be attributed to
Intellectual property and immutable
between durability and negotiability is re- a legal person, whether an individual or
mobiles
solved’’ (p. 61). Whether it is called translation, corporation.
The last section suggested that there is an
mediation, or inscription, the process within In principle, copyright ensures that
no
ambiguity in ANT’s approach to artifacts. On
which nonhumans acquire tremendous impor- matter what form the work takes
there will
the one hand, nonhumans are enrolled into
tance in social life is the gradual shift in the be durable relations between the legal subjects
social ties because they enable transient inter-
type of links between entities from ‘‘a provi- and objects of property. However, the prolif-
actions, provisional ties, and reversible trans-
sionally less reliable one to a longer-lasting, eration and popularization of file-sharing net-
formations to become more durable and
more faithful one’’ (Latour as Johnson, 1988, works have demonstrated how tenuous these
seemingly irreversible. On the other hand, the
p. 306). In this section, I will demonstrate how relations are in practice, and how dependent
enrolment and mobilization of these entities
this framework, and in particular the notion of these relations are on a supportive environ-
makes the artifact unpredictable and unstable.
immutable mobility, can be applied to explain ment composed of legal precedents, licensing
‘‘To get out of this quandary’’, Latour (1987)
the shift from MP3 as a relatively open systems, rights management, drilled clerks,
writes:
standard to the development of ‘‘copy protec- lawyers, and accountable employees. Traveling
. . . we need to do two things at once: to tion’’ technologies embedded in digital audio in MP3 form on the Internet, copyrighted
enrol others so that they participate in the files. artifacts perform new functions, working, for
construction of the fact; to control their Bruno Latour’s (1990, 1998) notion of im- example, as the currency of peer-to-peer music
behavior in order to make their actions mutable mobility describes the way in which exchange.

15
‘‘Engineers constantly shift out characters in other spaces and other times, devise positions for human and nonhuman users, break down competences that they
then redistribute to many different actants’’ (Latour, 1988, p. 309).

49 //ARTIFACT VOL 1 ISSUE 1 2007

 The theoretical immutability of the copy- piracy public relations campaign over the last 2005; Thorwirth et al., 2000). However,
righted work thus unravels. In practice, each three years targeting Canadian youth, a neces- DRM-plus-MP3 does not simply transmit the
translation of the musical work introduces the sary measure, CRIA claims, because ‘‘Consu- meaning of the order ‘‘do not copy’’ in a
possibility that it may defect from its role as an mers have to know that if they want a wide different form to the same entities. The se-
object of property. These displacements can choice and variety of music, that if they want quence of translations into new material ar-
drastically alter the ownership relations embo- their favourite artists to succeed, they must rangements leads to a gradual shift in the
died by the work. In the case of MP3, the sound support them by buying their music’’.16 Until distribution of roles, skills, and competences
object can be altered or simplified to make the recently, copyright ‘‘notices’’ were inscribed in assigned to the various entities. Rather than
file smaller and more easily copied and sent fine print on CDs in legal jargon. Now they are pointing towards an already-powerful actor (a
between computers. MP3 maintains the tonal transformed into multimillion-dollar anti-pi- class, a corporation, etc.) whose quantitative
relations of the sound through compression racy campaigns designed to capture the inter- difference from other actors (having more
(MPEG’s obligatory point of passage), but in est of the news media, educators, and youth. power, money, prestige, influence, etc.) ex-
order to maintain the links between the subjects File-sharing continues to be a popular plains the alignments in the uses of artifacts
and objects of copyright law, ownership rela- cultural practice in Canada despite the high- and thus the stability in the artifact’s identity,
tions need to be translated into more durable profile lawsuits, copyright reforms, and quasi- the emphasis in an ANTian framework is on
and mobile forms. For example, when the media educational anti-piracy programs. CRIA is the local practices through which power is
conglomerate Bertelsmann bought the Napster now faced with two choices since files continue produced by enrolling others into particular
file-sharing system in 2002, it also installed to be swapped long after the message ‘‘do not definitions of what is practical, unavoidable, or
rights management systems in the network copy unless authorized’’ has been amplified real.17
architecture to maintain the ownership relations and translated into various moral and eco- Two important points can be drawn from
of copyright. When copyrights travel, they nomic discourses: (1) either ensure that every- the attempt to make MP3s more resistant to
transform the places in which they move, turn- one reads the order/rule in the same way and copying. First, the translation of the statement
ing them into supportive environments or responds in the same way, or; (2) ‘‘load’’ the ‘‘do not copy’’ in anti-piracy campaigns,
protective envelopes to maintain the integrity statement by anticipating and incorporating compression algorithms, and other security
of the representational and proprietary func- ‘‘anti-programs’’ (Latour, 1991, pp. 104
105). devices transforms the interests enrolled by
tions of the copyright. CRIA can then make a second translation of the principle actor and distributes competences
Immutable mobility suggests that the med- the file-sharers’ interests (they do not like and actions to a new configuration of materi-
ium does matter. In MP3 form, sound record- transferring music from their computer to a als, bodies, and machines. The metadata in
ings are used differently and towards different CD and back again
a simple but annoying digital audio files, which controls the number
ends than they are in the form of CDs or in anti-circumvention technique), which leads to of copies that can be made from the file,
broadcasting. Similarly, copyrights change and a minor technological innovation (copies that internalizes the subject of copyright law who
are changed by their movement through dif- regulate their own replication) and a shift in accesses music via his or her computer con-
ferent computer networks. We are also begin- the materials used to enroll file-sharers into the nected to the Internet. Second, many of the
ning to see how the enrolment of MP3s, file- music industry’s program of action (digital prescribed actions on the web are not read by
sharing networks, and computer users into rights management or ‘‘DRM’’). humans. Like a speed bump, instructions are
copyright law alters the way in which owner- Recent technical papers on MP3 compres- translated into a form that makes the action of
ship relations are maintained. In Canada, the sion have set up new problematizations that the hybrid person-machine more predictable.
music industry lobby group, CRIA, has at- are designed to interest copyright owners The various scenes or settings that one en-
tempted to sue the end-users of file-sharing (rather than just electronic firms and telecom- counters on the web (identity checks, verifica-
networks, whereas, a few years earlier, com- munications companies) by translating those tion tests, blank search fields, etc.) are
mercial pirates were the main targets of such interests into the language of compression arranged as seriated actions: ‘‘The result of
litigation. CRIA has also developed an anti- codecs and metadata (e.g. Egidi & Furini, such an alignment of set-ups’’, Johnson (1988)

16
CRIA. (2005). http://www.cria.ca/freemusicmyth.php#mythsanswers. Accessed February 23, 2005.
17
Latour (1986): ‘‘Those who are powerful are not those who ‘hold’ power in principle, but those who practically define or redefine what ‘holds’ everyone together.
This shift from principle to practice allows us to treat the vague notion of power not as a cause of people’s behavior but as the consequence of an intense activity of
enrolling, convincing and enlisting’’ (p. 273).

//ARTIFACT VOL 1 ISSUE 1 2007 50

suggests, ‘‘is to decrease the number of occa- ontological claims regarding the agency of claim), other theorists were more concerned
sions in which words are used; most of the artifacts and their progressive immutability about the implications of ANT’s notion of
actions become silent, familiar, incorporated and capacity to form relations with one nonhuman agency for epistemology. Monica
(in human and nonhuman bodies)
making another: Casper (1994), for example, noted the merits of
the analyst’s job so much harder’’ (p. 308). taking nonhuman agency seriously by moving
What [historians] miss, however, is that
However, as the next section explains, this is beyond the study of human attributions of
each of these inventions, of more immu-
not the only difficulty facing ANTian studies agency towards the study of how action is
table more mobile elements is creating a
of artifacts. constituted by heterogeneous materials. How-
new specific type of space that allows
ever, she pointed out that such studies are
them to merge with the other in a specific
Problematizing ANT usually premised on ‘‘a decontextualized and
homogenizing way. The question of their
Not surprisingly, since ANT attempted to ahistorical definition of agency’’ (p. 840).
obvious differences is thus less pertinent
problematize key sociological methods and Moreover, assuming a priori that entities are
than that of their ability to tie in with
concepts, it has been the target of many human or nonhuman is problematic since
one another. (1988, p. 29)
criticisms. Three of these critiques need to be ‘‘human is a constructed (and often contested)
examined here because they have altered As discussed in the previous section, the theory identity or subject position’’ (p. 841). Her
ANT’s core concepts over the course of the of translation offered methodological princi- critique applies to both the ANTian frame-
past decade and have affected the way in which ples that required the analyst to abandon ‘‘all a work and the more conventional approach of
scholars study and theorize the role of artifacts priori distinctions between natural and social Collins and Yearley because neither camp
using ANTian models: (1) the ontological and events’’ and, moreover, to refrain from impos- examines the historical specificity of notions
epistemological implications of ANT’s replace- ing a grid of identities upon the entities in of agency. However, her position is closer to
ment of the social/technical dualism with the order to see how ‘‘actants’’ emerge through the that of Collins and Yearley since her concern is
human/nonhuman dualism; (2) the tension distribution of actions in translation (Callon, with accountability: ‘‘in many instances nonhu-
between immutability and mobility; and (3) 1986b, pp. 200
201). For Collins and Yearley, man agency deflects attention from human
the assumption that artifacts act within al- this means that ANT obscures the role of the accountability to other entities, whether hu-
ready existing places. (human) analyst in attributing agency to non- man, nonhuman, cyborg, or what/whomever’’
Despite its famous methodological slogan humans via analytic symmetry. Affirming the (p. 853).
(‘‘follow the actors’’), some critics have pointed determinative role of human action is not only In my view, Casper is critiquing a particular
out that ANTian case studies do not ade- a matter of recognizing the effects of one’s own formulation of ANT that became prominent in
quately attend to the use of artifacts, focusing analytic categories and methods upon the the 1990s. Initially, the concept of translation
instead on their construction. The paucity of subject matter; for Collins and Yearley as well enabled analysts to examine the temporality of
data related to use was a key reservation about as many other sociologists agency must be artifacts by approaching them not so much as
ANT that Collins and Yearley (1992a) put recognized as a human property in order to products impacting on society, but as processes
forth in their lively ‘‘epistemological chicken’’ differentiate true and false statements and to (e.g. distribution and association) that co-
debate with Latour and Callon. The manner in trace out the contours and consequences of produce social, natural, and technical realties.
which Collins and Yearley privileged the mo- human action. Collins and Yearley worry that By the 1990s, the concept of the actor-network
ment of use as some sort of confrontation nonhuman agency would collapse critical dis- became the centerpiece of a theory that
between interests or between nonhuman action tance in studies of science and technology by transformed relations of time into relations
and human agency can be contested on a relying on scientific expertise to decipher ‘‘how of space. In some sense, the resulting approach
number of grounds. However, they correctly much’’ agency nonhumans have in a given resembled a more structuralist approach to the
pointed to an ANTian tendency over the past situation. power of science and technology.18 Whereas
two decades to smooth over disjunctures and While Collins and Yearley were defending a the sociology of translation examined the
gaps in the history of technical artifacts. This particular view of agency in sociology as a transformation of identities in the stages of
tendency is most evident in Latour’s bold uniquely human capacity (an ontological research projects and technological design, the

18
I refer here to Michel Foucault’s (1986) definition of structuralism: ‘‘the effort to establish, between elements that could have been connected on a temporal axis,
an ensemble of relations that makes them appear as juxtaposed, set off against one another, implicated by each other
that makes them appear, in short, as a sort of
configuration’’ (p. 22).

51 //ARTIFACT VOL 1 ISSUE 1 2007

concept of the actor-network implies that sies as contestations of representativity where and that bring about changes as they go
identity is determined by a spatial configura- heterogeneous engineers scramble to develop along’’ (p. 150). However, her ethnographic
tion oriented around centers of translation (i.e. more sophisticated techniques to speak for and study of patents in action contests the stability
laboratories). negotiate with an array of entities, many of implied by the immutable mobile concept:
The ANT researchers have altered their whom, argues Latour (1987), ‘‘do not look like
Following the tracks of patents into
approach in the wake of such critiques. Latour men or women’’ (p. 121).19
those worlds renders it strange that in
(1999a,b), for example, attempted to move Latour’s immutable mobile is one of the
different places patents would be consid-
away from the notion of translation as associa- most popular concepts in the ANTian reper-
ered the same things. For they emerge in
tion and instead used the term ‘‘mediation’’ to toire. Ironically, immutable mobility has been
different capacities. They have different
examine the exchanges of properties between modified through its movement into various
effects. They operate in different prac-
entities. This emphasis on the traffic between disciplines outside the sociology of science and
tices. They ride different vehicles. And
entities through which they acquire agential technology. As discussed in the previous sec-
they are deliberately used for different
properties offsets the tendency in ANT to tion, the immutable mobile concept suggests
purposes. This variety counters the no-
think of technoscience in terms of a fixed set that certain artifacts and techniques enable the
tion that patents are ‘single objects’,
of socio-technical relations and as associations world to be collected and summarized in
‘immutable mobiles’. (p. 166)
between pre-existing human and nonhuman durable and portable forms. In contemporary
entities. capitalist economies, many artifacts (including De Laet concludes that this multiplicity ‘‘sup-
While Casper and Collins and Yearley were software) acquire stability with the aid of ports a vision of patents as agents in their own
concerned about the disconnection of agency immutable mobiles called patents. Patent law right’’ (p. 166).20 However, she argues that
in ANT with accountability, commitment, translates the messy material activities of de- immutability obscures the actual manner in
responsibility, and intentionality, it is worth sign labs and the heated exchanges of scientists which artifacts spread out over time and space.
pointing out that the theory of translation engaged in a ‘‘proofs race’’ into the flat, two- Similarly, Susan Leigh Star suggests that to act
attempted to topicalize these attributes as dimension space of a patent claim. These like an immutable mobile artifacts need to be
processes within the mobilization of hybrid representations are designed to work like ‘‘both plastic enough to adapt to local needs
collectives (the secondary mechanism of trans- Latour’s immutable mobiles since they pre- and constraints of the several parties employ-
lation). Accountability is an important element scribe roles to human and nonhuman entities ing them, yet robust enough to maintain a
of current legal attempts to re-order the socio- in, for example, the role of the ‘‘user’’ of the common identity across sites’’ (as cited in
technical relations of, for example, file-sharing invention, or the ‘‘public’’ to whom the patent Nigel Thrift, 1999, p. 37).
systems on the Internet. While the theory of supposedly discloses the inner working of the The difficulty in deploying the immutable
translation may not offer the necessary tools to invention. mobile concept without introducing these im-
contest the manner in which legal systems In practice, patent claims are extremely portant modifications (or mutations) may
make individuals accountable for swapping mobile but are not quite as immutable as point to a more fundamental problem in the
copyrighted sound recordings, it is extremely Latour’s concept would suggest. In her study relationship of artifacts to space and place in
useful for tracing out the strategies through of how patent systems have been deployed in the ANT framework. Kevin Hetherington
which collective action is attributed to one or a Zimbabwe, Marianne de Laet (2000) uses the (1997) argues that, while ANT has destabilized
few individual humans beings. Moreover, it notion of the immutable mobile to show how the humanist division of place and space and
allows the analyst to examine these controver- patents are ‘‘events that perform connections
enables analysts to engage with agency as

19
In approaching such controversies, the analyst may attempt, as I have done in this article, to trace the actions attributed to individuals back to the alliances
established in the primary mechanisms of translation. However, pointing to the heterogeneous materials that perform a particular definition of socio-technical reality
does not resolve the problem of representativity. As Latour (1998) points out, ‘‘We cannot simply say that ‘all of them’ count in the making of an observation. If we
were stopping at that, something would be missing from the mere deployment of heterogeneous associations’’ (p. 434). Although ANT advocates a ‘‘distributed
monism’’ (Barron, 2003), Latour argues that new distinctions are necessary: ‘‘If sociology is the study of society it has to take full account of those crowds of non-
humans mingled with humans. To take full account of this retinue of delegates, sociologists have to look carefully at their conflicts over who is the most representative’’
(1988, p. 16).
20
Like de Laet, Jonathan Murdoch (1998) finds the link between immutability and mobility overly simplified when applied to patent law. He suggests that patents
perform a connection between the standardized, classificatory schemes of law or genetic science and the more fluid and heterogeneous relations of other modes of
ordering.

//ARTIFACT VOL 1 ISSUE 1 2007 52

distributions, the notion of immutable mobility technologies and artifacts such as the notion ble mobility and actor-networks now seems
‘‘tends to leave the question of the places of instantaneous information transfer, direct problematic because it takes for granted the
involved as already established prior to this political/economic influences as the prime construction of the categories of human and
network of heterogeneous agents’’ (p. 188). mover of technological change, and nonhuman nonhuman and does not emphasize the traffic
Building on Latour’s formulation, Hethering- agency as the imitation of human action. In between them (Reckwitz, 2002, p. 847). If this
ton suggests artifacts are more than just their place, the theory of translation points us is the case, it reinforces my argument that
immutable forms of knowledge that travel towards a cascade of increasingly simplified translation and its emphasis on the co-produc-
through space. They transform spaces into inscriptions, the glacial drift in the goals of tion of social, natural, and technological
places.21 prime movers, and subtle but strategic shifts of reality should be relocated to the center of
By combining Leigh Star’s and Hethering- actions, capacities, and other identity attri- ANTian understandings of artifacts.
ton’s modifications to the concept, one can butes between heterogeneous materials. The ANT has attempted to displace the narrative
move beyond the notion that artifacts prolif- theory of translation provides a language to of increasingly sophisticated technology with a
erate in contemporary society because knowl- describe and narrate the story of nonhuman new ‘‘mythology’’ about exchange of proper-
edge must be embodied by durable materials to agency in a way that offsets the explosive ties between human and nonhuman organiza-
organize socio-technical relations. Artifacts revolutions associated with intelligent ma- tion. The most significant implication for
perform place mnemonically and affectively chines like iPods and other evidently active studies of artifacts
whether they be scientific,
by sustaining a multiplicity of associations artifacts. The agency of things in the theory of religious, or aesthetic
emerges out of this
with other artifacts. In this modified ANTian translation has less to do with automation, ontology symmetry; it provides a theoretical
framework, agency is not just the capacity to artificial intelligence, or technical sophistica- basis for re-orienting historical and cultural
stay the same, to resist ‘‘the flow’’, as Fuller tion than it does with the continuous expan- studies of artifacts towards the many cross-
puts it, but to multiply associations and keep sion of ‘‘social-ness’’ to entities that have been, overs and exchanges with nonhumans that
them in place over time. This is a form of and continue to be, considered ‘‘merely’’ make durable forms of social interaction
agency that collectives gain from artifacts as technical or material. As Callon and Latour possible.
much as they do from human beings. (1992) write in regards to the rather low-tech
speed bump: Acknowledgements
Conclusion: Towards alternative I would like to thank Dr. Sheryl Hamilton,
To claim that only the humans have
narratives of technological artifacts Dr. Charlene Elliott, and Dr. Paul Théberge, as
meaning and intentionality and are able
One of the strengths of contemporary social well as the anonymous reviewers, for their
to renegotiate the rules indefinitely is an
research on artifacts is its focus on socio- insightful comments on earlier drafts of this
empty claim, since this is the very reason
cultural contexts. This attentiveness to context article.
why the engineers, tired of the indisci-
is a deliberate attempt on the part of cultural
pline and indefinite renegotiability of
and social theorists to offset technological
drivers, shifted their program of action
determinism and the causal roles attributed References
to decrease this pliability. (p. 361)
to technical artifacts in ‘‘impact narratives’’ Akrich, M., & Latour, B. (1992). A summary of a
(Sterne, 2005, p.7). Similarly, translation pro- Concepts like the immutable mobile and the convenient vocabulary for the semiotics of
human and nonhuman assemblies. In W. Bijker
vides an antidote to other forms of reduction- actor-network were useful insofar as they & J. Law (Eds.), Shaping technology/building
ism, simplistic notions of causality, and impact directed attention to the movements, circula- society (pp. 259
264). Cambridge, Mass.: MIT
narratives that now circulate around digital tions, and distributions across the divisions of Press
Barker, G. (2005, May 24). The god of iPod: Why
rights management, MP3s, iPods, and copy- human/nonhuman and social/technical (La-
music? The Age, 1 (cover story).
right law. The theory of translation problema- tour & Crawford, 1993, p. 263). The notion Barron, C. (Ed.). (2003). A strong distinction
tizes dominant themes in studies of new of nonhuman agency associated with immuta- between humans and non-humans is no longer

21
‘‘These representations assemble and transform the similitude of the materiality of space into the ordered arrangement of a place. They arrange, order, include
and exclude, they make knowable a space to everyone who might choose to look at these representations and also make it possible to compare it with another space.
The subjective world of memory, image, dream and fantasy, so often associated with place, operates by assembling materials into representations and using those
representations to establish the difference between one place and the next. Those representations contain truth claims (not necessarily scientific) about a space’’ (p.
189).

53 //ARTIFACT VOL 1 ISSUE 1 2007

 required for research purposes: A debate Communications Magazine, 43(5), 90
97. change (pp. 225
258). Cambridge, MA: MIT
between Bruno Latour and Steve Fuller. His- Creusere, C. D. (2003). Quantifying perceptual Press.
tory of the Human Sciences, 16(2), 77
99. distortion in scalably compressed MPEG Latour, B. (1993). Ethnography of a high-tech
Brandenburg, K. (1995). Audio coding for TV audio. Conference Record of the Thirty-Seventh case: About Armais. In P. Lemonnier (Ed.),
and multimedia. International Broadcasting enth Asilomar Conference on Signals ystems Technological choices: transformations in mate-
Convention, 95, 159
164. and Computers, 1, 265
269. rial cultures since the neolithic (pp. 372
398).
Bruum, H., & Langlais, R. (2003). On the De Laet, M. (2000). Patents, travel, space: London: Routledge.
embodied nature of action. Acta Sociologica, Ethnographic encounters with objects in Latour, B. (1994a). On technical mediation:
46(1), 31
49. transit. Environment and Planning D: Society Philosophy, sociology, genealogy. Common
Callon, M., & Latour, B. (1992). Don’t throw the and Space, 18(2), 149
168. Knowledge, 3(2), 29
64.
baby out with the Bath School! A reply to Egidi, L., & Furini, M. (2005). Bringing multi- Latour, B. (1994b). Pragmatogonies: A mythical
Collins and Yearley. In A. Pickering (Ed.), media contents into MP3 files. IEEE account of how humans and nonhumans swap
Science as practice and culture (pp. 343
360). Foucault, M. (1986). Of other spaces. Diacritics, properties. American Behavioral Scientist,
Chicago: University of Chicago Press. 16(1), 22
27. 37(6), 791
808.
Callon, M., & Law, J. (1995). Agency and the Hetherington, K., & Law, J. (2000). After net- Latour, B. (1996a). Do scientific objects have a
hybrid collectif. South Atlantic Quarterly, works. Environment and Planning D: Society history? Pasteur and Whitehead in a bath of
94(2), 481
507. and Space, 18, 127
130. lactic acid. Common Knowledge, 5(1), 76
91.
Callon, M. (1980). Struggles and negotiations to Hetherington, K. (1997). In place of geometry: Latour, B. (1996b). Response: Pursuing the dis-
define what is problematic and what is not: The the materiality of place. In K. Hetherington & cussion of interobjectivity with a few friends.
socio-logic of translation. In K.D. Knorr, R. R. Munro, Ideas of difference: Social spaces Mind, Culture and Activity, 3/4, 270
289.
Krohn & R. Whitley (Eds.), The social process and the labor of division (pp. 183
199). Mal- Latour, B. (1998). How to be iconophilic in art,
of scientific investigation: Sociology of sciences, den, MA: Blackwell. science, and religion. In C. Jones, & P. Gallison
Vol IV (pp. 197
221). Dordrecht: Reidel. Kahney, L. (2004, April 16). Music magic found (Eds.), Picturing science, producing art (pp.
Callon, M. (1986a). The sociology of an actor- in the shuffle. Wired News. Retrieved Novem- 418
440). London: Routledge.
network. In M. Callon, J. Law, & A. Rip ber 23, 2004, from http://www.wired.com/ Latour, B. (1999a). On recalling ANT. In J. Law
(Eds.), Mapping the dynamics of science and news/culture/0,1284,63068,00.html & J. Hassard (Eds.), Actor network theory and
technology (pp. 19
34). London: Macmillan. Knorr Cetina, K. (1997). Sociality with objects:
after (pp. 15
25). Oxford: Blackwell.
Callon, M. (1986b). Some elements of a sociology Social relations in postsocial knowledge
Latour, B. (1999b). Pandora’s hope: Essays on the
of translation: Domestification of the scallops societies. Theory, Culture and Society, 14(4),
reality of science studies. Cambridge, MA:
and fishermen of St. Brieuc Bay. In J. Law 1
30.
Harvard University Press.
(Ed.), Power, Action, Belief A New Sociology of Knorr Cetina, K., & Bruegger, U. (2002). Tra-
Latour, B. as Johnson, J. (1988). Mixing humans
Knowledge? (pp. 196
233). London: Routledge ders’ engagement with markets: A postsocial
with non
humans: Sociology of a door
& Kegan Paul. relationship. Theory, Culture & Society, 19(5/6),
Callon, M. (1991). Techno-economic network 161
185. opener. Social Problems, 35, 298
310.
Latour, B. (1986). The powers of association. In J. Latour, B., & Crawford, T. (1993). An interview
and irreversibility. In J. Law (Ed.), A sociology
of monsters: Essays on power, technology and Law (Ed.), Power, action, belief: A new sociol- with Latour. Configurations, 1(2), 247
269.
ogy of knowledge? (pp. 264
280). London: Latour, B., & Woolgar, S. ([1979], 1986). Labora-
domination (pp. 132
161). New York: Routle-
dge. Routledge & Kegan Paul. tory life: The construction of scientific facts.
Callon, M., Law, J. & Rip, A. (1986). Putting Latour, B. (1987). Science in action: How to follow Princeton, NJ: Princeton University Press.
texts in their place. In M. Callon, J. Law & A. scientists and engineers through society. Cam- Law, J. (1986a). Editor’s introduction: Power/
Rip (Eds.), Mapping the dynamics of science bridge, MA: Harvard University Press. knowledge and the dissolution of the sociology
and technology (pp. 221
30), London: Mac- Latour, B. (1988). The Pasteurization of France of knowledge. In J. Law (Ed.), Power,
millan followed by irreductions (A. Sheridan & I. Law, action, belief: A new sociology of knowledge?
Carey, M., & Wall, D. (2001). MP3: The beat Trans.). Cambridge, A Harvard University (pp. 1
19). London: Routledge & Kegan Paul.
bytes back. International Review of Law. Com- Press. Law, J. (1986b). On the methods of long-distance
puters and Technology, 15(1), 35
58. Latour, B. (1990). Drawing thing together. In M. control: Vessels, navigation and the Portuguese
Casper, M. (1994). Reframing and grounding Lynch & S. Woolgar (Eds.), Representation in route to India. In J. Law (Ed.), Power, action,
nonhuman agency: What makes a fetus an scientific practice (pp. 19
68). Cambridge, belief: A new sociology of knowledge? (pp. 234

agent? American Behavioral Scientist, 37(6), Mass.: MIT Press. 265). London: Routledge & Kegan Paul.
839
856. Latour, B. (1991). Technology is society made Law, J. (1991). Introduction: Monsters, machines
Collins, H. M., & Yearley, S. (1992a). Epistemo- durable. In J. Law (Ed.), A sociology of and sociotechnical relations. In J. Law (Ed.), A
logical chicken. In A. Pickering (Ed.), Science monsters: Essays on power, technology and sociology of monsters: Essays on power, tech-
as practice and culture (pp. 301
326). Chicago: domination (pp. 103
131). New York: Routle- nology and domination (pp. 1
25). New York:
University of Chicago Press dge. Routledge.
Collins, H. M., & Yearley, S. (1992b). Journey Latour, B. (1992). Where are the missing masses? Law, J. (1999). After ANT: Complexity, naming,
into space. In A. Pickering (Ed.), Science as A sociology of a few mundane artifacts. In W. and topology. In J. Law & J. Hassard (Eds.),
practice and culture (pp. 369
389). Chicago: E. Bijker, & J. Law (Eds.), Shaping technology/ Actor network theory and after (pp. 1
14).
University of Chicago Press building society: Studies in sociotechnical Oxford: Blackwell.

//ARTIFACT VOL 1 ISSUE 1 2007 54

Lessig, L. (1999). Code and other laws of cyber-
space. New York: Basic Books.
Lessig, L. (2002). The future of ideas: the fate of
the commons in a connected world. New York:
Vintage Books.
Murdoch, J. (1998). The spaces of actor-network
theory. Geoforum, 29(4), 357
374.
Musmann, H. G. (1990). The ISO audio coding
standard. Communications: Connecting the
Future, GLOBECOM ’90. IEEE, 1, 511
517.
Pels, D., Hetherington, K., & Vandenberghe, F.
(2002). The status of the object: Performances,
mediations, and techniques. Theory. Culture &
Society, 19(5/6), 1
21.
Preda, A. (2000). Order with things? Humans,
artifacts, and the sociological problem of rule-
following. Journal for the Theory of Social
Behaviour, 30(3), 269
298.
Rault, J. B., Dehery, Y. F., & Lever, M. (1995).
The ISO/MPEG Audio MISICAM family. IEE
Colloquium on MPEG-2
What it is and What
it isn’t, 3/1
3/14.
Reckwitz, & Andreas (2002). The status of the
‘material’ in theories of culture: From ‘social
structure’ to ‘artefacts’. Journal for the Theory
of Social Behaviour, 32(2), 195
217.
Riles, A. (2000). The network inside out. Ann
Arbor: University of Michigan Press.
Ryan, M. (2003, May 11). Online music: Are we
finally buying it? Chicago Tribune (Arts &
Entertainment), 1.
Schlender, B. (2001). Apple’s 21st-century Walk-
man. Fortune Magazine, 144(9), 213
218.
Slater, D. (2002). Making things real: Ethics and
order on the internet. Theory, Culture &
Society, 19(5/6), 224
245.
Smart, B. (1982). Foucault, sociology, and the
problem of human agency. Theory and Society,
11(2), 121
141.
Sterne, J. (2005). The audible past: Cultural origins
of sound reproduction. Durham, NC & Lon-
don: Duke University Press.
Thorwirth, N. J., Horvatic, P., Weis, R., & Jian
Zhao (2000). Security methods for MP3 music
delivery. Conference Record of the Thirty-
Fourth Asilomar Conference on Signals, Sys-
tems and Computers, 2, 1831
1835.
Thrift, N. (1999). The place of complexity.
Theory, Culture and Society, 16, 31
70.
Vedres, B., Bruszt, V., & Stark, D. (2005).
Organizing technologies: Genre forms of on-
line civic association in Eastern Europe. Annals
of the American Academy of Political and
Social Science, 597(1), 171
188.
Walker, L. (2004, Feb 1). The cola giants align
with music vendors. Washington Post, F07.
Winner, L. (1980). Do artifacts have politics?
Daedalus, 109(1), 121
136.

55 //ARTIFACT VOL 1 ISSUE 1 2007