Felt quality of sociomaterial relations: Introducing emotions into sociomaterial theorizing

Information and Organization 24 (2014) 156–175

Contents lists available at ScienceDirect

Information and Organization journal homepage: www.elsevier.com/locate/infoandorg

Felt quality of sociomaterial relations: Introducing emotions into sociomaterial theorizing Mari-Klara Stein a,⁎, Sue Newell b, Erica L. Wagner c, Robert D. Galliers d a

Copenhagen Business School, IT Management Department, Howitzvej 60, Frederiksberg 2000, Denmark Sussex University, Department of Business and Management, Jubilee Building, Falmer, Brighton, BN1 9SL, UK c Portland State University, School of Business Administration, 631 SW Harrison St., Portland, OR, United States d Bentley University, Information and Process Management and Sociology Departments, 175 Forest Street, Waltham, MA 02452, United States b

a r t i c l e

i n f o

Article history: Received 25 September 2012 Received in revised form 14 April 2014 Accepted 21 May 2014 Available online xxxx Keywords: Computer-mediated work practice Sociomateriality Critical realism Agential realism Emotion Felt quality Interpretive field research

a b s t r a c t Sociomateriality, in helping to overcome the longstanding dualism between the social and the technical, has become an increasingly popular theoretical perspective in Information Systems (IS) research. However, while recognizing the usefulness of sociomaterial theorizing, we contend that it also inadvertently perpetuates other kinds of dualisms—particularly that of objectivism–subjectivism and cognition–emotion. We argue that sociomateriality's current inability to express what it feels like to be a human agent, and the inadvertent perpetuation of the cognitive–emotional dualism, is problematic in terms of the limited practical insights these perspectives generate. To address this limitation, we propose and illustrate two different approaches for including emotions in sociomaterial theorizing. By proposing two approaches for the inclusion of emotions into applications of sociomateriality in IS research (one founded on critical realism, and the other on agential realism), we provide researchers with the conceptual tools to generate richer practical and theoretical insights. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction Sociomaterial theorizing (Orlikowski & Scott, 2008) has entered the arena of IT implementation studies relatively recently and is an increasingly popular choice for generating a nuanced understanding of this ⁎ Corresponding author. Tel.: +45 41 85 21 50. E-mail addresses: [email protected] (M.-K. Stein), [email protected] (S. Newell), [email protected] (E.L. Wagner), [email protected] (R.D. Galliers)

http://dx.doi.org/10.1016/j.infoandorg.2014.05.003 1471-7727/© 2014 Elsevier Ltd. All rights reserved.

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

157

phenomenon (cf. Leonardi, 2009; Wagner, Newell, & Piccoli, 2010). It does not attribute social change to either the material features of the technology or see it as a product of social agendas; rather it considers social change to be the emergent outcome of the mutual constitution of the social and the material. For example, packaged software is inscribed with standard templates from which to configure the product. The product becomes entangled with decision makers' choices and activities to buy rather than build; to enter long-term partnerships with vendors; to adhere to so-called software-based ‘best practices’, and to change long-standing work practices. Such activities are only relevant in the circumstance of configurable packaged software, and of shifting work practices and power relations that become bound up with the standard templates. A broadly sociomaterial perspective, therefore, focuses researcher attention and helps draw insights related to the joint agency of the social and the material, acting together. While sociomaterial theorizing can help overcome the longstanding dualism of the social and the technical in IS research (Latour, 2005; Leonardi & Barley, 2008; Orlikowski & Scott, 2008; Pickering, 1995), it inadvertently perpetuates other dualisms—particularly that of the objective/cognitive vis à vis the subjective/ emotional (cf. Thompson, 2012). This critique has been leveled against IS research in general, as well as against practice-oriented explanations of technology implementation more specifically. Thus, IS research is generally characterized by a relative lack of consideration of the emotional side of individuals (Bagozzi, 2007; McGrath, 2006; Zhang, 2013), partly due to its roots in computer science and rational scientific management (cf. Hirschheim & Klein, 2006). In practice-oriented research, it has been pointed out that, while processes of “meaning-making” are illuminated, the “subjective experience of agency”—how human agents feel about themselves and their circumstances—tends to be devalued (Thompson, 2012: 189). Most studies adopting a sociomaterial perspective have also tended to focus on social norms and standards, work goals and collective practices, as the social or human agency (e.g., Leonardi, 2011). The point that human agency often works through the reflexive intentionality and motivations of individuals has been recognized (e.g., Pickering, 1995); however, what tends to get neglected in such studies is that what often motivates humans are their emotions (Thompson, 2009; 2012). Notwithstanding, it is precisely this “biographical awareness (elaborate sense of self) that allows a person to position herself relationally against unfolding social reality, [and] this juxtaposition occurs within consciousness, which is felt affectively” (Thompson, 2012: 195). This suggests that in many IT projects key concerns for all stakeholders are related to their biographical awareness, which includes issues of self and identity on an individual and a collective level and which is reflected in how they feel their sociomaterial circumstances. In sum, many sociomaterial accounts of IT implementation offer an under-personalized view, where human emotions and biographical awareness play little or no role in the dynamic interplay between the social and the technical. Thompson (2012: 204) calls for the examination of the mutually constitutive relationships between, not only the social and material dimensions of reality, but also the subjective/biographical dimension. He only explores the duality (as against the dualism) of the social and biographical structures in detail, however. In order to investigate how sociomaterial theorizing could include emotions, we first need a better grasp of what sociomateriality actually means. In recent years it has been increasingly recognized that there are multiple appropriate and valid ways of conducting a sociomaterial inquiry (Feldman & Orlikowski, 2011; Leonardi, 2013). While researchers subscribe broadly to a relational ontology, they differ in the specifics of “how they theorize the status of nonhuman agency relative to human agency” (Feldman & Orlikowski, 2011: 1244). Furthermore, over the years many, slightly different, relational perspectives have been proposed, all of which can be broadly characterized as sociomaterial (Leonardi, 2013; Orlikowski & Scott, 2008). Examples include the “mangle” of practice (Pickering, 1995); actor– network theory (Latour, 2005); human–machine (re)configurations (Suchman, 2007); imbrication (Leonardi, 2011), and constitutive entanglement (Orlikowski, 2007). Recently, two broad philosophical foundations for sociomaterial theorizing have been discussed: critical realism and agential realism (Leonardi, 2013). The choice of foundation influences not only the kinds of theoretical arguments and contributions one can make, but is also expected to have pragmatic consequences in terms of research methods and empirical inquiry (Leonardi, 2013: 73). More importantly from the perspective of this paper, as we illustrate below, theories associated with both foundations have tended to ignore emotional aspects as they examine relations between the social and the technical. In this paper, then, we consider both of these ‘types’ of sociomaterial theorizing and examine how each perspective could include the emotive domain. In sum, our study focuses on the research question: How can sociomaterial theorizing include emotions?

158

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

Our contribution is to highlight the neglected area of emotion in sociomaterial theorizing. In doing this, we provide researchers with means to enhance their sociomaterial accounts based on either a critical or agential realism foundation. We first introduce the theoretical foundation for our paper, before we turn to our method. Finally, we discuss our findings, drawing out our contributions and conclusions as they relate to our research question. 2. Theoretical background The context for our paper is packaged software implementation. This topic has been studied from many different perspectives. Traditionally, IT implementations have been studied from a deterministic point of view (Perrow, 1967; Woodward, 1958), yielding insights, for example, as to circumstances in which technology (as a discrete material entity) produces particular forms of organizing (as a separate and also discrete but social entity). Criticisms of such deterministic approaches are well rehearsed (e.g., Robey & Boudreau, 1999), and have increasingly encouraged scholars to accept that the effects of technology and humans on each other are socially constructed (Leonardi & Barley, 2010). However, strong social constructivist positions have also been criticized in the literature: for example, because they tend to fade the technology into the background, or because they hinder us from being able to account for the complex interplays between IT and its human implementers and users (Leonardi, 2011; Orlikowski & Scott, 2008; Pickering, 1995; Sassen, 2002). This is where sociomaterial theorizing (Orlikowski & Scott, 2008) enters the arena as an increasingly popular choice for a nuanced account of IT implementation (cf. Leonardi, 2009; Wagner et al., 2010). The next sections introduce the different sociomaterial perspectives we explore in our study as well as our initial theorizing around emotions in sociomateriality. 2.1. Sociomateriality As discussed, the distinguishing feature of the concept of sociomateriality is the ontological integration of the social and the material. From this viewpoint, it is not that the ‘social’ influences the ‘material’ (as in social construction studies, which focus on how people enact a particular technology); nor is it that the ‘material’ influences the ‘social’ (as in technological determinism, which focuses on how a technology forces some kind of structure or practice outcome). It is not even that there is a recursive relationship between the ‘social’ and the ‘material’ (as with the socio-technical view, which sees technology both enabling and constraining people's actions). Rather, as Orlikowski (2007: 1437) indicates, “the social and the material are inextricably related”. Thus, the material and social are enmeshed in any work practice. As an example, the ultramarathon runner Anton Krupicka used to take an Exacto knife and slice thin layers of sole from his trail shoes, essentially whittling a customized pair of low profile trainers. What is interesting for us in this story is the entanglement of the runner and his technology as he runs. The investigation of practice will neither be found in the out-of-the-box shoe, nor in Anton's customized pair; rather it is in what they do together on the trails.1 While a relational ontology is a fundamental characteristic of a sociomaterial perspective, it has also been increasingly recognized that there are multiple appropriate and valid ways of conducting sociomaterial inquiry (Feldman & Orlikowski, 2011; Leonardi, 2011). For example, researchers may differ in the specifics of “how they theorize the status of nonhuman agency relative to human agency” (Feldman & Orlikowski, 2011: 1244). These may be viewed as symmetrical (Latour, 1987), entangled or fused (Orlikowski & Scott, 2008) or asymmetrically weaved together, with human agency in the lead role (Leonardi, 2011). Furthermore, three ways of studying sociomaterial practices have been proposed (Feldman & Orlikowski, 20112): adopting an empirical, theoretical and/or a philosophical focus. The empirical approach focuses on exploring and describing the everyday practices and performances of both human and non-human agents in producing their organizational realities. The theoretical approach extrapolates from the empirical descriptions of sociomaterial practices to theories that explain the 1

http://www.youtube.com/watch?v=8ofkrxZvyrI. see also Academy of Management 2012 Symposium “Theoretical Elements of a Sociomaterial Perspective in Organization Studies” (OCIS, OMT), Boston, MA. 2

Table 1 Overview of two different sociomaterial perspectives. Pickering (1995): Mangle of Practice (MOP)

Callon (1986), Latour (1987; 2005): Actor–Network Theory (ANT)

Ontology

The social and the material exist separately. Human and material agencies are continuously mangled through series of resistances and accommodations propelled by intentional and goal-oriented human actors.

Epistemology

Analysts make determinations about how and why the separate “social” and “material” become the “sociomaterial” and persist that way over time (Leonardi, 2013) “The arrangement of an artifact's physical and/or digital materials into particular forms that endure across differences in place and time” “Abstract concepts such as norms, policies, communication patterns, etc.”

There is no social and no material, only fused sociomaterial networks. Human and non-human actants are seen as working together in an ontological network to create environments. Nothing lies outside the network of relations, and each actant is merely the summation of smaller actants. Analysts follow the action that culminates around heterogeneous networks rather than particular individuals or artifacts.

What is materiality? (Leonardi, 2013) What is social? (Leonardi, 2013) What is 3sociomateriality? (Leonardi, 2013) Human and non-human agency

Dynamics of agents

Nature of the emerging outcome Consideration of existing “structures”

Practices involving material objects inter-relating with all phenomena we typically define as ‘social’, including roles, norms, discourses Agency: ability to continually do things. Both human & material agencies are temporally emergent: we can only see the contours of material/human agency in the resistances and knowledge/ goals/plans produced at the intersection of human/non-human agency. Human agency is distinctly characterized by intentionality and future-orientation. Human and material agencies are effectual in producing outcomes only when joined together. Human and material agencies are constitutively enmeshed in practice by means of a dialectic of resistance and accommodation. Resistances arise at the intersection of human/non-human agency, while accommodations are human agent's responses to these resistances. A snapshot of a particular configuration of human and material agencies, where the mangling process is not visible (depends on the timescale adopted) Interactions between human and material agencies produce durable patterns. Cultural field of practice accounts for prior sequences of resistances and accommodations (knowledge, models), which all become part of current mangling.

There is no separable materiality. There is only an actor network. There is no separable social. There is only an actor network The inherent inseparability between the material and the social forming an actor network Agency is not an attribute or an essence of either humans or non-humans. Agency is a capacity realized in their relations (so it is always emergent and shifting). Generalized symmetry—humans and non-humans are treated the same for analytical purposes because what is interesting is the emergent network of interests. Relations and boundaries between humans and technologies are not pre-given or fixed, but enacted in practice (performativity). Associations between human and non-human actors are translated over time to re-produce another actor–network.

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

Citation/descriptor

Irreversible actor–network (fused association of human and nonhuman actants) that has a degree of stability for a period of time. Structures are not spoken of, rather, more consistent practices (routines) are seen as the outcome of a stable actor–network.

159

(continued on next page)

Table 1 (continued) Callon (1986), Latour (1987; 2005): Actor–Network Theory (ANT)

Example empirical case

Donald Glaser's creation of the bubble chamber described by Pickering (1995) to demonstrate a dialectic of resistance and accommodation. As Glaser attempts to construct a new kind of device that can detect strange particles, his efforts to do this face a number of material resistances from the detector he was developing (the detector did things like explode that were separate from Glaser and his practice). Glaser then responded with accommodations so that there was a gradual emergence of a prototype chamber that worked. Time period: covers the invention time period (approx. 2 years)

What the perspective emphasizes

Well-suited to describe cases where humans revise goals (accommodations) based on the constraints of material agency (resistances) in relation to the previous goal. Well suited to describe the emerging contours of human and material agency (e.g., emergence of a bubble chamber by Donald Glaser) Human and material agencies can be separated.

What the perspective obscures

Cases where material agency accommodates (rather than constrains) are not explicitly considered.

Potential conceptual contributions

Showcase how organizations, people and technologies come to be as they are. Move technology into a constitutive role in organizing/society, while showing how organizing/society shapes technology (Leonardi, 2013). For example, the bubble chamber that emerges from Glaser's practice then allows other discoveries to emerge because of its ability to detect strange particles.

Consideration (or lack there) of emotions

Human intentionality is discussed purely in relation to goals and plans. Emotions are not mentioned.

Actor–network of marine biologists, fishermen, Japanese techniques for fishing, and towlines with attached collector bags attempt (A) to enroll the interests of St. Brieuc Bay scallops by encouraging the scallop larvae to anchor onto the bags (B) and secure themselves from predators. Alternative networks of tide currents, parasites and visitors are opposing forces (C) that inhibit the enrollment of the larvae (B) into A. Negotiations occur between the networks but also within network A as it modifies the nature and design of its actor–network (depth of line placement, material used in bag collectors, etc.). Time period: duration of the negotiations across actor–networks. Well-suited to describe the performativity of practices. Human and material agencies cannot and should not be separated. Materiality is integral to human activities (e.g., in the Japanese fishing techniques). General symmetry between human and non-human actors that join together through shared interest to form an actor–network. Actor–networks are not confirmed to place but are determined by alignment of interest. Representatives act and speak on behalf of other actants. The intentionality of human and non-human actors. It may be unclear what the goals and political positions of actors are. The currents of St. Brieuc Bay inhibit the attachment of larvae onto the collector bags. The intentionality of this action is not addressed. Showcase how all organizational processes are sociomaterial (networks of heterogeneous materials). Demonstrate that organizing occurs in practice. Ability to zoom-in and out and see “things” as networks at a macro or micro level. For example the scientists and their research presentation form a new network within which the scallops are actants represented by graphs and charts of findings. The macro is seen in the micro network. Not explicit. Networks are described through technical features, human knowledge and choices.

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

Pickering (1995): Mangle of Practice (MOP)

160

Citation/descriptor

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

161

dynamics between the social and the material, the conditions for the emergence and change of sociomaterial practices, etc. The philosophical or ontological approach engages with the position that reality is not made up of independent objects (social or material), but rather emerges from the joint agency of humans and material objects. This focus explores the implications of the ontological position on our theories of technology and organizations. There are, then, somewhat different approaches to the study of sociomateriality. To illustrate such differences we present two well-known sociomaterial perspectives—the “mangle of practice” (Pickering, 1995) and actor–network theory (Callon, 1986; Latour, 1987; 2005) in Table 1. As can be seen from Table 1, while both theories adopt a relational perspective, they differ in respect to the ontological status of ‘what’ is related. For Pickering, the materiality of the bubble chamber is independent of Glaser—the bubble chamber presenting resistances that Glaser has to accommodate. For Callon, the actor–network of marine biologists; fishermen; Japanese techniques for fishing, and towlines with attached collector bags is what matters: it is not relevant to look at the collector bags, for example, independent of the human actors who are using them to try and make the larvae safe. These differences underscore that sociomaterial theorizing can have different philosophical roots. In this regard, Leonardi (2013) identified two distinct philosophical foundations for the study of sociomateriality: agential realism and critical realism. These differ subtly, but significantly, in their ontological and epistemological assumptions and are argued to produce different kinds of contributions (Leonardi, 2013). ANT, for example, is closer to Barad's (1998, 2003) agential realist philosophy, while MOP has something of a critical realist orientation (Archer, Bhaskar, Collier, Lawson, & Norrie, 1998; Bhaskar, 1978). From an agential realist perspective, nothing (neither people nor objects) have any inherent properties; rather any properties that are attributed to a ‘thing’ are seen to be a relational effect, produced and performed in and through the network of relations. On the other hand, from a critical realist perspective human and material agencies exist independently, but it is their synergistic interaction, which produces effects as they ‘act on’ each other (Slife, 2004). Since both philosophical perspectives are presented as sociomaterial, we consider each as we turn to identify how we can incorporate emotions into our sociomaterial theorizing.

2.2. Emotions and sociomateriality What we observe from Table 1 is that both MOP and ANT highlight the importance of understanding the typical consequences or outcomes we observe in any kind of practice as emergent results of sociomaterial mangling or networks. How such negotiations and networks are typically described, however, tends to be in terms of, for example, rational alterations in one's goals and plans in the face of material constraints or affordances, and rational choices of attempting to enrol and translate other actors' interests to one's own. We can separate one practice or network from another based on the different human and non-human elements involved, the strength of associations between the elements (i.e., the stability of the practice or network), or different emergent outcomes. We are unable to differentiate between them based on their felt or emotive characteristics, however. Both approaches, then, provide us with little understanding about what emotive ‘coloring’ the sociomaterial associations have. We cannot see the disappointment and elation as Glaser fails and succeeds to develop the bubble chamber; nor do we see the frustration or satisfaction as the marine biologists fail or succeed in making the scallop larvae safe. Yet, when reading the case descriptions, one has the idea that, without these elements, the stories are somehow incomplete. This is an interesting omission since in recent years emotions have become widely researched in organization studies (Ashkanasy, 2003; Barsade & Gibson, 2007; Elfenbein, 2007). Although in the IS field, the topic has received more limited attention (see Zhang, 2013 for a recent theory-building attempt), authors have been calling for more in-depth investigations of various emotion-related phenomena (Bagozzi, 2007; McGrath, 2006; Ortiz de Guinea & Markus, 2009; Thompson, 2012). Our aim in this paper is to answer this call with respect to sociomaterial theorizing. To begin this exploration, we briefly introduce the various existing conceptualizations of emotions, followed by a discussion of two perspectives on how emotions can be seen as relationally produced—emerging not just from interpersonal social relations, but, as we argue and demonstrate, from sociomaterial relations.

162

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

2.2.1. Emotions as personal inner states and social performances Emotions have been defined in various ways depending on whether the individual physiological and psychological aspects or the collective social aspects of emotions are stressed. Commonly, emotions are defined as object-oriented (focused on a specific target), relatively intense and short-lived (Barsade & Gibson, 2007; Brave & Nass, 2002). On the more individual side, a component-based definition is often used and is gaining acceptance in the IS literature (cf. Zhang, 2013). According to this, “emotions typically arise as reactions to situational events in an individual's environment that are appraised to be relevant to his/her needs, goals, or concerns. Once activated, emotions generate subjective feelings … motivational states with action tendencies, arouse the body with energy-mobilizing responses … and express the quality and intensity of emotionality outwardly and socially to others” (Zhang, 2013: 251). Emotion is, thus, formally defined as an “episode of interrelated, synchronized changes in the states of all or most of the five organismic subsystems (cognitive, neurophysiological, motivational, motor expression and subjective feeling) in response to the evaluation of an external or internal stimulus event as relevant to major concerns of the organism” (Scherer, 2005: 697). Other, more sociological, definitions have also been offered. For example, Fineman (2008: 1) sees emotions as “produced through interpersonal work that is conditioned by cultural imperatives: the social rules that sanction what is appropriate to feel and express.” Emotions, therefore, inherently express a valuation and collectively produce emotionologies or emotion cultures— society's and organizations' understanding of how certain emotions should be directed and expressed towards certain groups, such as royalty, asylum seekers, sexual minorities, etc. These definitions reflect two broad approaches to emotions: one that sees emotions largely as individual bodily and mental states—personal, but measurable experiences accompanied by changes in bodily sensations, expressive gestures and action tendencies (Scherer, 2005)—and one that sees emotions largely as social, collective performances oriented towards an audience, possibly with a strategic goal (Fineman, 2008). Most of the few prior studies dealing with emotions in relation to IS phenomena in particular have adopted the more personal approach, defining emotions as mental states (Beaudry & Pinsonneault, 2010; Stam & Stanton, 2010), as a person's response to various external and internal stimuli (Stein, Newell, Wagner, & Galliers, 2012), or as valenced reactions to the emotion-eliciting objects/states being processed by the individual (Kim, Chan, & Chan, 2007). In other words, technology or materiality is considered to be a stimulus or event that elicits emotional responses in individuals. A more social approach has also been taken, seeing emotion as inextricably linked to our historical sense of person-hood—“temporally-based, affectively felt consciousness is deeply relational, and thus structuring of, our stance towards unfolding social reality” (Thompson, 2012: 195). 2.2.2. Theorizing emotions in sociomaterial relations: two possible approaches In our sociomaterial theorizing around emotions we draw on the concept of affective spaces (Navaro‐ Yashin, 2009; Thrift, 2008). Affect in this instance is seen as referring broadly to an emotive domain, but moving the scope of this emotive domain outside of the human subject and, therefore, beyond the traditional definitions of emotions and feelings (Navaro‐Yashin, 2009). In short, affect refers to “sensual intensities that may move through human bodies, but that do not necessarily emerge from them. The reference-point for affect (which used to be, singularly, subjectivity) has been radically altered and multiplied in this approach, making it possible to read many other things, such as space and the environment, as affective” (Navaro‐Yashin, 2009: 12). For example, one's childhood home may exude and generate an affect of nostalgia; a war memorial—an affect of sadness and anger; a church—an affect of reverence, and a bedroom and the objects in it—an affect of safety and comfort. The particular affective atmosphere of shopping malls (and how this is achieved through escalators, air conditioning, piped music, etc.) and its influence on shoppers and their behavior have also been examined (Healy, 2014). However, like Navaro‐Yashin (2009): 14–15, we recognize that the human subject is still important—a particular environment or object generates its own affectivity; those who inhabit the space experience it in particular ways and have corresponding emotions. Furthermore, in particular social settings, individual emotions may be mixed with collective emotionologies that influence how emotions become expressed. In sum, there are three important concepts to consider when theorizing emotions in sociomateriality: affect, emotion and emotionology (see Table 2). From a sociomaterial perspective the emotive domain is produced relationally or, in other words, it occurs “in an encounter between manifold beings”, which we take to include the individual, the social and

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

163

the material (Thrift, 2008: 179). As in the study of sociomateriality itself, and consistent with our discussion of MOP and ANT, it is useful to conceptualize the study of emotions in sociomateriality utilizing two theoretical foundations: critical and agential realism. Table 3, below, outlines the characteristics of the two different possible approaches. If one were to adopt a critical realist stance, one could describe instances of subjective emotions and feelings accumulating over time, relating to the possible affectivity of objects and spaces, and being expressed according to particular emotionologies. Together, these elements would produce an emergent emotive domain characteristic of a sociomaterial relationship. We adopt the term, felt quality (Stein et al., 2012; see also Ciborra, 2006) to describe this emerging emotive domain and define it as an emergent characteristic of a sociomaterial relationship. From an agential realist stance, however, a different understanding of this felt quality would arise. If there are no properties that are distinctly social or material, only emergent characteristics of the sociomaterial, then the felt quality cannot be the result of the meshing of human emotions, material affect and social emotionologies. Rather, the felt quality is an emergent characteristic of a particular sociomaterial configuration, which can be dissected into elements of emotions, affect and emotionologies only for analytical purposes. We will illustrate both stances below using examples from our empirical study; we first, though, describe the methodological approach used in our study. 3. Methodology We analyzed data gathered at two universities (Public and Private) as part of an in-depth multi-site case study focused on understanding the dynamics in packaged software implementation and use. We considered the implementation and post-roll-out phases of a packaged software product that has been designed to manage faculty data in university settings so that these data are available in a standard format across departments and institutions, thus allowing for analysis of ‘productivity, assessment and accreditation’. We refer to this package as Faculty Output or FO. Our fieldwork comprised interviews with key participants regarding their project involvement and use of the software; observation of meetings/small group discussions, and systematic review of official project documentation. Following Walsham (1993: 15), the cases are designed to seek “validity … not [from] the representativeness of the case in a statistical sense, but on the plausibility and cogency of the logical reasoning used in describing the results and in drawing conclusions from them”. We used interpretive research techniques to gather, interpret and analyze field data (Klein & Myers, 1999; Walsham, 1993). The research focus was specifically on the sociomaterial practices or the coming together of social, material and subjective agencies (depending on the perspective taken) that characterized each university's packaged software implementation. Collecting data from various stakeholders in different university contexts helped us to become aware of multiple interpretations in order to improve the “plausibility and cogency of our interpretive accounts” Table 2 Relevant concepts for theorizing emotions in sociomateriality. Concept Affect

Definition

Sensual intensities that may move through human bodies, but that do not necessarily emerge from them. The reference-point for affect lies outside of the human body, making it possible to read many other things, such as space, the environment and physical objects, as affective (Navaro‐Yashin, 2009: 12). Similar to the concept of affective quality (Zhang, 2013)—affective property of a stimulus that has the ability to cause a change in a person's emotions (cf. Zhang, 2013). Whereas emotions exist inside humans (people feel upset), affective quality/affect exists in the stimulus (it is the environment/object that is upsetting). Emotion Episode of interrelated, synchronized changes in the states of all or most of the five organismic subsystems (cognitive, neurophysiological, motivational, motor expression and subjective feeling) in response to the evaluation of an external or internal stimulus event as relevant to major concerns of the organism (Scherer, 2005: 697) Emotionology Society's and organizations' collective understanding of how certain emotions should be directed and expressed towards certain groups (Fineman, 2008). Emotionologies provide standards or norms of “appropriate” emotional expression for different occupational groups, genders, etc. (Wright & Nyberg, 2012) and are, therefore, politico-ideological constructions (Fineman, 2010: 27).

164

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

Table 3 Two possible approaches to theorizing emotions in sociomateriality.

Overall ontological position What is affect?

Critical realism

Agential realism

Materiality has affect; humans have emotions; social has emotionologies. These become mangled to form the ‘felt quality’ of the sociomaterial Sensual intensity/affective property that emerges from a material object/space and moves through human bodies Emotion is a set of interrelated changes (bodily, motivational, subjective feeling, etc.) in an individual

Sociomaterial relations/practices/ actor–networks have a ‘felt quality’

If there is no materiality, there is also no affect that is distinctly characteristic of particular material objects/spaces What is emotion? If there is no social, there is also no emotions that are distinctly characteristic to individuals that are part of this social What is emotionology? Emotionology is society's/collective's understanding If there is no social, there is also no how these emotions should be directed and emotionologies that are distinctly towards whom characteristic of the society/collective What is the ‘felt quality’ An emergent emotive characteristic of sociomaterial An emergent emotive characteristic of of sociomateriality? activities; melding of subjective emotions, material sociomaterial activities; boundaries between affect and social emotionology subjective emotions, material affect and social emotionology are not pre-given or fixed, but enacted in practice Unit of analysis Social, material and subjective agency Sociomaterial practice, including its felt quality

(Klein & Myers, 1999). Following Levina and Vaast (2005), we present a cross-case comparison in the form of a table that summarizes the research methods employed in the field (Table 4). Although the two cases were conducted separately, all researchers were informed by a sociomaterial perspective, employed the same interview techniques, and focused data collection on the same unit of analysis—specifically, the organizational implementation and use of the same packaged software. Our analysis involved articulating the dynamics of each case study, thereby developing a shared perspective based on the themes that were present in either one or both settings. In doing so, we identified how sociomaterial perspectives could include emotions. Interviews were recorded and transcribed. The transcripts were shared via Dropbox and the research team reviewed all data. Initial interpretations were communicated during Skype-based conference calls. In our analysis we use the empirical data for illustrative purposes, rather than for developing new theoretical concepts. Accordingly, we coded the data to identify instances of emotive content and then considered how material affect, emotions and emotionologies were related to the ongoing sociomaterial dynamic. These emergent themes were shared with one another in order to consider the principles of dialogical reasoning and suspicion with a view to further improve our interpretive accounts (Klein & Myers, 1999). An iterative process of discussion and fine-tuning of the case descriptions helped us to identify the over-arching theme of emotion as related to the nature of sociomaterial thinking, and its value to IS

Table 4 Cross-case comparison of research methods. Methods

Public

Private

Field work Timing Narrative interviews (inc. group interviews and meeting recordings) Observations

Interviews and observations over 18 months Implementation and post-implementation 30 with 23 stakeholders. Recorded and verbatim transcriptions

Interviews and observations over 12 months Implementation and post-implementation 18 with 30 stakeholders Recorded and verbatim transcriptions

Field journal Documentation Follow-up contact

4 sessions with faculty filling out their annual reports (video recorded) + limited observations of faculty use Pre and post interview notes and observations, Pre and post interview notes and transcribed observations, transcribed Yes Yes Yes Yes 2 faculty advisory group sessions + limited observations of faculty use

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

165

scholarship in explaining implementation and use. The analytical themes were then organized around this issue, allowing us to develop a coherent sense of the implementations across the two universities. 4. Case descriptions Each case is organized into three key episodes in the implementation process: selection and configuration of the FO package, university-wide roll-out, and initial faculty use of and responses to FO post roll-out. In this section, we first describe the FO package and then describe the introduction of the package in our two sites (see overview in Table 5). 4.1. The Faculty Output vendor and package The FO vendor first offered the package in 1999. At the time of writing, FO employed approximately 30 staff and there were about 3000 FO adopters in over 25 countries. The FO package is somewhat different from many software packages in that it is not configurable by the user organization. Rather, it comes as a standard package with the vendor organization undertaking configurations in response to requests from its clients. FO claims that the software package is tailored to individual clients (given its flexible back-end) and that it is configurable for many different specifications. Configuration requests are handled uniquely for each client organization, based on a system of ticketing, which clients use to submit their requests to FO. Configuration requests can range from requesting particular reports to modifying input screens or output formats. The faculty activity information that is input into FO is held on the vendor's cloud-based repository. Individual institutions using FO can, however, get periodic back-ups of their data that they may hold on their own server. As a standardized software package, FO requires individual faculty members to record their activities according to a fixed classification system. In order to accommodate activities across academic disciplines, FO is based on broad classifications of faculty activities—typically teaching, research and service. Each individual activity must be manually entered into FO and categorized: for example, research activities can be categorized into journal articles/books/book chapters; conference presentations; grants/sponsored research; artistic performances; exhibits; patents or ‘Other’. Faculty members also need to provide detailed information on each item (e.g., name; date; authors). There are no formatting or spell-checking options in FO. Based on the data entered, FO also provides reporting functionality. Typically, faculty members can pull out a (standardized) vita and annual activity report, while administrators can also run aggregate reports across departments, schools, etc. The next sections describe the series of phases around FO roll-out as they unfolded at Public and Private. The phases, key implementation episodes in each phase, actors involved and outcomes are summarized in Tables 6 and 7 below. 4.2. Public 4.2.1. Selecting FO and working with the vendor to customize FO With the approval of the (now former) Provost, Public purchased the FO product, replacing a very simple home grown, web-based, data entry form used for gathering faculty activity information. The decision to purchase FO was mainly driven by the need for a central faculty vitae/activity database. The FO

Table 5 Overview of the research settings.

Type of institution Orientation Size Founded

Public

Private

State university Broad: three colleges and four schools; both professional as well as arts & sciences Approx. 30,000 students; 1564 faculty members (888 full-time) Post World War II

Private university Narrow: business; arts and sciences Approx. 5500 students; 280 full-time members of faculty. Post World War I

166

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

Table 6 Overview of FO implementation at public. Key episodes

Actors involved

Early 2009: selecting FO; working with the vendor – FO implementation assigned to CA – office (CAO) – – Change in scope: collection of all faculty activities (historical data), not just – annual activities – Configurations (pre roll-out) Late 2009/2010: university-wide roll-out – FO officially announced as a vitae database – – Data entry support personnel is offered to faculty –

Provost, CAO, FO Provost, CAO, deans, faculty, FO CAO, vendor, FO

Outcome – CAO reluctantly takes on the role (having no prior experience with IT implementations) – Faculty are overwhelmed with the amount of data they have to enter – FO fulfills the basic requirements of Public

Provost, faculty, – Still very limited faculty use of system + negative FO backlash Faculty, CAO, data – Some faculty feel resentful for not getting support entry personnel, FO 2011/2012: faculty backlash and reverting back to original scope of the project – Entry of whole vita is no longer required – CAO, faculty, FO – Limited impact on already-formed patterns of use and non-use

implementation project was assigned to a central administrative office (CAO). The project team comprised two people: one staff member handling the configuration, communication with vendor and faculty requests, and a senior administrator/faculty member responsible for the initiative as a whole but not involved in day-to-day matters. Public had no prior experience with any kind of formal faculty activity database. Once the package had been purchased, the Provost changed the scope of the project—rather than simply using it to produce standardized annual activity reporting, FO could be used for many different purposes, provided that it included all faculty output data from all years.3 [1] “From the institutional research standpoint we wanted a better way to capture faculty activity. We had been using a questionnaire … Initially, we hadn't intended it [FO] to be the entire vita for a faculty member. It was supposed to be the information that covered the same period as the questionnaire … It was never quite clear, where it [FO] was supposed to land. So, because we had been administering the questionnaire, we said OK … Frankly … I want the reports out of it, but I don't want to implement the software and I don't want to deal with all the faculty stuff.” [(Senior CAO administrator)] Prior to rolling-out the software to the university community, a few major configurations took place. Most importantly, the CV report was modified to make it conform to Public's Promotion and Tenure (P&T) format guidelines. Additional edits/amendments were made after FO was in production, based on feedback from faculty users, date entry assistants, and the FO administrator's on-going review of screens and reports. The single CAO staff member was responsible for all aspects of planning and implementation. While she was allocated 100% to the project for the first year, 30% of her prior work responsibilities were not covered, leading to her being regularly over capacity. [2] “We didn't get additional resources in [CAO] to support [FO] … And when the Provost kind of changed the focus and said that he wanted faculty to put their whole vitae in there, that's when we said, ‘Okay, this is how many data-support people we would need.’ And I think neither one of us [staff member and boss] knew at the time what that really meant, and since then we've surveyed a long list of institutions and it seems a lot of people have fallen into that trap. And part of that, I think … is that they may have oversold their product and we were very naïve, having never done it before …”. [(Senior CAO administrator)]

3

Quotes are numbered so that we can refer to them in our analysis.

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

167

Table 7 Overview of FO implementation at Private. Key episodes

Actors involved

2009: Selecting FO; working with the vendor – FO implementation assigned to CT – Committee, CTO, FO office (CTO) – CTO, faculty, FO – One major customization (faculty web – CTO, vendor, FO profiles) – Configurations (pre roll-out) 2010: university-wide roll-out: FO use is voluntary – FO officially announced as a tool for – Provost, faculty, FO web profiles, CVs and annual reports 2011/2012: FO use becomes required (strongly encouraged) – All faculty are requested to fill out their – Provost, CTO, annual reports in FO faculty, FO

Outcome – CTO takes on the role naturally, having done similar projects before – FO automatically links to faculty web profiles – FO fulfills the basic requirements at Private

– Faculty try out the system, different use patterns form; continuing configuration efforts – Some frustration, but mostly neutral compliance

[3] “FO was not well received by our faculty … the faculty who were using it weren't pleased that it didn't produce a CV that looked like anything like a CV… So, it was constant customization of what we had already created …”. [(CAO staff member)]

4.2.2. Rolling out FO across the university At the beginning of the roll-out, Public's Provost attended an early committee meeting and explained the history and motivation of the initiative. Subsequently, he gave an explicit directive that faculty should use FO, where FO was described as a vitae database. Despite this announcement, significant numbers of faculty did not input data into FO. As a result some academic units and individuals received support for data entry while others were required to input CV data themselves. This created a negative backlash from those faculty members who had not been so supported. 4.2.3. Faculty response to and use of FO For those who did attempt to input data, many found that the input screen options were inappropriate for describing their activities. The antipathy towards the system was evident in angry comments about the amount of time being wasted on FO, mixed with community-mindedness of wanting to do the ‘right’ thing. Once collected, data could be used for a variety of different purposes other than that which had been presented to faculty as CV management. Exactly what those purposes might be was unclear and sometimes threatening to faculty, however: [4] “There's a widespread fear that these meaningless statistics will be used to determine budget allocations. Such fears are extremely demoralizing for faculty …”. [(anonymous faculty member)] [5] “The Provost is saying that we're not productive enough in the library. And that's based on some of these reporting measures, right? … So all of this starts to feel like it's really a big deal. Like I need to make sure to check this box or else, what, we don't get a new tenure-line position? … I'm not actively worried, but it's definitely being aware of the ramifications of the little boxes you choose to fill in … [X] did something really interesting, he probably mentioned this, he found out from somebody [in the CAO], which categories will be reported up. And then he sent it out to all of us and said, ‘These are the categories that are going to be reported up, so these are the most essential pieces of information.’”. [(Librarian/Assistant Professor)] There were a few exceptions, with some faculty appreciating the usefulness of manipulating data and not seeing the more sinister implications of the kind highlighted above:

168

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

[6] “My instincts say: ‘jump on board and go for it’, because it's really a nice tool. I love the idea of being able to roll up [data] … I think that's useful.” [(Director of a PhD program)] In response to continued dissatisfaction on the part of the majority, it was decided that only entry of the most recent year's activities would be mandatory for the 2011/2012 academic year. This had a relatively limited impact on the already-established patterns of use and non-use: [7] “… We decided to go back to just asking the faculty to populate, I think, it's seven different fields. In some departments, 90 percent of the faculty responded to that, in others, only 38 percent. Essentially, it's not really being used by very many people.” [(Senior CAO administrator)]

4.3. Private 4.3.1. Selecting FO and working with the vendor to customize the FO At Private, the decision to purchase FO was made by a special committee, comprising administrators, faculty representatives and technology support personnel from a central technology office [CTO] that supports faculty members both in terms of their research and teaching. This committee attended product demonstrations for two different software packages and decided on FO due mainly to the perceived better availability of customization support from the FO vendor. The CTO, having prior experience with IT implementations at the university, was put in charge of the implementation. In the event, they felt, like Public, that vendor support for product tailoring had not met their expectations: [8] “I think [FO] was better at selling the product and still only had one developer. And once we installed, we figured out [that] what was good for them in terms of generating sales was terrible for us because their own developer was overwhelmed and I had some frustrating conversations with … the owner. And we really got up to around here [points above his head] with requests that had been made and sitting in the queue for literally months. We can't touch the system ourselves; everything is a request. It was a horror show.” [(Project leader, CTO)] The CTO team handling the implementation consisted of four people responsible for overall implementation and configuration, including communication with faculty, administrators and the vendor; faculty training, and the like. The main customization was the in-house creation of standardized web profiles for all faculty members—a faculty member's data input into FO is displayed on the Internet webpage as a profile of their various activities, including a listing of all publications. Aside from this, “many hundreds” of other configurations were requested to the input screens and output reports, based on feedback from faculty members: [9] “We wanted to have as little information to fill out for the faculty as possible. We went through every screen and said ‘Okay, what fields are required, what are optional, what's not there that we need to have.’” [(FO project team member, CTO)] 4.3.2. Rolling out FO across the university At roll out, some of the requested configurations had not been made (estimated to be 15 to 20% by the FO project leader) because of the time it was taking the vendor to respond to requests. However, given the need to launch the system in preparation for an accreditation review, the Provost made a decision to push ahead with this, and a communication about the initiative came in April 2010. In this, FO was labeled as a way to maintain an attractive public web profile, generate a standardized CV, and produce an annual activity report. 4.3.3. Faculty response to and use of FO Approximately one year after the university-wide roll out, all departments at Private had been using FO to some degree. A similar pattern has emerged in most departments—two or three active users update

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

169

their information almost monthly, while the rest of the faculty input their recent activities into FO once a year (at the time when annual reports are due). Faculty responses to the system have been more diverse at Private than at Public. The reaction towards maintaining a public web profile through FO was rather positive, and the most active FO users seem to be driven by the desire to keep their on-line profile up-to-date: [10] “I know if someone's going to be looking for me on the web, they're going to want to find what I've done most recently. And from looking over other faculty members’ [profiles], probably about 75% of folks keep theirs updated … I'm very gratified with the results …” [(Associate Professor)] In relation to using FO for annual activity reporting, faculty responses ranged widely. Some department chairs found the standardized report to be helpful: [11] “I can understand [FO's] desire to bring all that information together and to somehow mechanize the generation of the report. I'm not without guilt when it comes to trying to put faculty into “boxes” to some extent. I don't know how else you'd do it if you're going to make some kind of evaluation. … Evaluating people to me is a very scary thing, so I was looking for some mechanism whereby I could … justify my own answers with some degree of confidence …” [(Department chair)] Others grudgingly incorporated FO into their annual activity reporting, but kept using other, more familiar, formats and documents on the side: [12] “My faculty were extremely resistant to [FO] because, in their opinion, [FO] was clumsy, ineffective and didn't present them in a way that they wanted to be presented. Most of my faculty now maintain two CVs … And I want it all on paper. So it's [the faculty member's] job to present to me a complete picture of the work they've done over the last year and I do not take it to be my job to enforce a particular format on their presentation.”. [(Department chair)] Following this relatively successful introduction of FO at Private, in late 2011 the use of FO for annual activity reporting became required of all faculty members. This mandate created some frustration and annoyance amongst the faculty who were used to producing their reports in MS Word. However, because FO is associated with non-essential administrative tasks, many faculty members also felt neutral or even positive about FO: [13] “I put in my activities and then apparently [FP] makes an annual report. I'm not real happy about that magic behind the scenes. It's not an essential part of my work though, so it's not such a big deal” [(Associate Professor)]

5. Analysis Having described the FO implementations, we next analyze the results by applying the two sociomaterial perspectives previously introduced to explore how emotions can be incorporated into each. Before we turn to this, however, we briefly illustrate how a ‘traditional’ MOP and ANT study might analyze our data, focusing on the pre-roll out phase only (Table 8). What this analysis shows is that neither account really addresses the emotional aspects of our cases, even though these were clearly evident in our case descriptions. For example, the MOP perspective draws attention to the series of resistances and accommodations through which particular sociomaterial configurations develop, while the ANT perspective highlights the dynamic development of configurations of human and non-human actants through enrolment and translation. Little of the various emotive elements that were evident in the quotes were present in either analysis. The frustrations, fears, worries and guilt, so evident in our case descriptions have fallen away to produce descriptions that are accurate in basics, but that paint a monochrome and passionless picture, failing to capture what it means to act in this

170

Empirical observation

MOP perspective

ANT perspective

Pre roll-out practices: working with the vendor at Public (see [1][2][3])

Series of (material) resistances and accommodations at Public:

Enacted sociomaterial relations at Public: CAO actor–network translates a significant number of Public faculty into the story of the importance of capturing faculty productivity which is inscribed in the questionnaire. The rhetoric that counting ‘something’ is imperative helps further enroll actors into the network when FO is introduced. An alternative network emerges when conversations about effectively representing faculty CV-s/productivity ensue. These networks are vying for dominance through each configuration that slightly alters the CAO network through the modification of the FO software.

Pre roll-out practices: working with the vendor at Private (see [8][9])

1. CAO administers questionnaire to capture faculty productivity. a. Resistance: Questionnaire does not enable the capture of the nuances of activities b. Accommodation: Find a better way of collecting this info (i.e., FO) 2. FO is implemented and CAO starts working with the vendor a. Resistance: CAO has no experience, is understaffed; FO lacks certain capabilities (e.g., CV report does not match Public's needs) b. Accommodation: CAO asks vendor to make configurations Series of (material) resistances and accommodations at Private:

1. Committee decides to implement FO a. Resistance: FO screens, etc. do not fit the need of Private's faculty (+FO has no built-in web profile element) b. Accommodation: CTO builds web profiles in-house; vendor is asked to configure screens according to faculty input c. Resistance: Vendor response is slow; FO cannot be configured locally by CTO d. Accommodation: CTO keeps pushing vendor (80% of configurations are done for roll-out time)

Enacted sociomaterial relations at Private: The Private–FO actor–network attempts to enroll and translate the interests of faculty network. Mobilization fails because faculty resist the enrollment and demand technical changes be incorporated into the FO software. In response, the Private-FO network attempts to enroll the Vendor network which includes the software code base that enables customizations. The Vendor network prioritizes other interests and is slow to respond to the Private–FO network. All three networks remain unable to translate interests of the other, although some configuration changes are enacted in this period.

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

Table 8 Comparison of insights gained from different sociomaterial perspectives.

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

171

world. It is possible that emotional elements are so difficult to include in sociomaterial theorizing because we associate emotions purely with humans, and, therefore, consider them to be outside the scope of an account that emphasizes sociomateriality. However, as indicated above, and as illustrated next, there is no need to restrict our understanding of the emotive domain in such a manner. Table 9 provides an overview of how one could analyze the emotive elements in our empirical data from the critical realism and agential realism perspectives. In this table we have analyzed the same pre roll out practices as in Table 8 and in addition we also include post roll out practices. We show the results of the analysis according to a critical realism and agential realism perspective side-by-side in order to demonstrate the similarities and differences. From Table 9, we see that, while there are obvious differences, both approaches are adequate for examining the emotive domain unfolding in our two cases. Both the critical and agential realism

Table 9 Illustrative comparison of applying two approaches to develop emotions in sociomateriality.

Empirical observation

Pre roll-out practices: working with the vendor at Public (see [1][2][3])

Pre roll-out practices: working with the vendor at Private (see [8][9])

Critical realism

Agential realism

Felt quality as the emergent emotive characteristic of sociomaterial activities, resulting from the melding of existing subjective emotions, material affect and social emotionology. 1. Material affect: FO generates an affectivity of a burden 2. Subjective emotions: frustration, confusion 3. Emotionology: it is OK to feel that way when you have no prior experience with IT implementations 4. Emerging felt quality of the CAO–FO– vendor relations: harassed 1. Material affect: FO generates an affectivity of a ‘too-good-to-be-true’ 2. Subjective emotions: frustration 3. Emotionology: this is what happens when a vendor over-sells their product 4. Emerging felt quality of the CTO–FO– vendor relations: ‘horror show’

Felt quality as the emergent emotive characteristic of sociomaterial activities, where boundaries between subjective emotions, material affect and social emotionology are not pre-given or fixed, but enacted in practice. As the alternative networks vie for dominance, the emerging CAO–FO–vendor network is not stable and feels harassed. FO is enacted as a burden; it fails to enroll either CAO or faculty; CAO, failing to enroll FO and, to a lesser extent, the vendor, enacts a naïve implementer, entitled to feel frustrated and confused.

Post roll-out practices: faculty 1. use and non-use patterns of FO at Public (see [4][5]) 2. 3. 4.

Post roll-out practices: faculty 1. use and non-use patterns of FO at Private (see [11][12]) 2. 3. 4.

As the FO–CTO, FO–Vendor and FO–Faculty networks each attempt to enroll others in their interests, the emerging CTO–FO–vendor network is not stable and feels like a ‘horror show’. FO is enacted as not delivering on its promise, it has difficulty enrolling CTO and the faculty; CTO has difficulty enrolling the vendor and enacts a frustrated implementer that has been fooled to buy a product without sufficient support. Material affect: FO generates an affectivity As the FO–library and the FO–Administration of threatening and unfair ramifications networks attempt to enroll others in their Subjective emotions: fear interest, the emerging library–FO–admin Emotionology: one should not show active network temporarily stabilizes in a fear or worry ‘demoralizing’ felt quality. FO is enacted as threatening and unfair, it fails to enroll the Emerging felt quality of the Library–FO– library; the administration has the power to Administration relations: demoralizing enforce library mobilization, so the library enrolls a procedural manual to deal with FO threat, while at the same time expressing their lack of moral support for this configuration. Material affect: FO generates an affectivity As the FO–faculty and the FO–chairs networks of mechanistic “boxing” of faculty attempt to enroll others in their interest, the members emerging faculty–FO–chairs network Subjective emotions: fear, guilt, defiance temporarily stabilizes in two contentious Emotionology: one should be fair in forms. FO is enacted as a ‘boxing’ mechanism, evaluating faculty it enrolls some chairs, but not others; some Emerging felt quality of the Faculty–FO– chairs enroll/do not enroll FO as arguably a Chairs relations: contention fair/unfair way to do evaluations and assuage their fears.

172

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

approaches can expose the emerging felt quality of particular sociomaterial configurations, even while they go about analyzing the emergence of this felt quality in distinct ways. A critical realism approach allows for the teasing out of clearly separable elements of material affect, human emotions and social emotionologies; these are then observed coming together in particular ways through the joint acting of human and machine agencies. An agential realism approach emphasizes the overall felt quality of the actor–network or sociomaterial configuration, but can also accommodate the examination of how boundaries between humans, machines and social aspects, as well as their emotions, affective qualities and emotionologies, are drawn in practice. Comparing these analyses to the ones given in Table 8 provides descriptions that are both accurate and animated, presenting the world in its full colors. Importantly, this can be done while staying true to the foundations of sociomaterial theorizing (i.e., without re-introducing the dualism between the social and the material, and importantly also, without creating another dualism between the subjective and objective). We discuss this more fully in the next section. 6. Discussion We began this paper by outlining the differences between two sociomaterial perspectives (“mangle of practice” and actor–network theory) to illustrate the diversity in this new line of theorizing. It has been argued that the choice of theoretical foundation will have a practical influence on the kind of empirical study undertaken, and consequently, on the kinds of contributions made to the study of sociomateriality (Leonardi, 2013: 74). Our comparison of the two perspectives/foundations in Table 1, in line with previous comparisons (Leonardi, 2013), showed that there are considerable differences in how various elements are conceptualized, the kind of aspects the perspectives emphasize and obscure and, correspondingly, the kinds of overall contributions the perspectives can make. Moving beyond the specific perspectives of MOP and ANT, we also discussed broader philosophical differences between critical realism and agential realism (Leonardi, 2013). While critical realism emphasizes the joint working of human and non-human agency and can showcase the temporal emergence of particular sociomaterial configurations, agential realism is more suited towards understanding sociomaterial practices and their performativity (i.e., the idea that everything we see as a social or a technical consequence is actually a function of sociomaterial practices). However, while both perspectives are useful for analyzing our data (as illustrated in Table 8), and can help overcome the longstanding dualism of the social and the technical in IS research (Latour, 2005; Leonardi & Barley, 2008; Orlikowski & Scott, 2008; Pickering, 1995), the exclusion of emotions limits our understanding of the cases as they are experienced in practice. Both perspectives, therefore, contribute to the continued prevalence of the dualism between subjective–emotional and objective–cognitive (Thompson, 2012). We argue that this is problematic both from a theoretical and practical point of view, as we discuss next. Table 9 demonstrates how emotional elements can be included in both an agential realist and critical realist sociomaterial account. This is further shown in more abstract terms in Fig. 1 below. This figure indicates that the mangling or imbrication of human and non-human agencies can be described, not just in terms of plans, intentions and constraints, but also in terms of human emotions, social emotionologies and material affects (Panel B). Similarly, the dynamics of an actor–network theory or a sociomaterial practice can be described, not just in terms of actants, enrolments, etc., but also in terms of the overall felt quality of the network/practice (Panel A). The two approaches, therefore, provide researchers with the capability to analyze, describe and explain sociomaterial configurations in a more nuanced way, including the full spectrum of what actually transpires “in an encounter between manifold beings” (Thrift, 2008). Interestingly, a sociomaterial perspective on emotions also allows us to theorize about those emotive characteristics that are typically black-boxed or excluded in studies of emotions (e.g., the overall felt quality of a sociomaterial association or material affect). By seeing emotions as a purely subjective or intersubjective phenomena, researchers from all paradigms and perspectives miss out on much of what characterizes our being-in-the-world (cf. Ciborra, 2006). While there are some subtle differences, therefore, in terms of how emotions are examined and interpreted depending on the particular theoretical foundation selected, the main point of our analysis is to demonstrate how emotions can be ‘added’ to sociomaterial theorizing without reverting to the traditional mind/body dualism. From all sociomaterial perspectives, emotions can be treated as practices (i.e., something that we do rather than something that we simply feel as an afterthought of some event

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

173

Fig. 1. Possible inclusion of emotions into key theoretical foundations for sociomateriality. (Adapted from Leonardi, 2013)

that is a product of our mind's interpretations of that event). Others have similarly argued for an ‘emotion-as-practice’ perspective, examining emotional practices as socially contingent and as performative in the sense that ‘putting a name’ on a practice (e.g., ‘this is horrible’) produces the experience (Scheer, 2012). However, the sociomaterial perspective alerts us to the idea that emotional practices are not only socially contingent but also materially contingent. Gherardi (2012) introduces the idea that practices take place in an ‘equipped context’, with material objects fundamentally part of the practice. This allows those involved in the practice to appear to practice effortlessly because objects are ‘ready-to-hand’, as with a doctor with a stethoscope always around his/her neck. However, we would argue that the context is not only ‘equipped’ in the sense of functionality—allowing people to do the work at hand—but also equipped in the sense that it can produce emotional experiences, so that the stethoscope, as part of the practice of treating patients, has affective and emotionological properties as well as goal-oriented properties. This emotional sociomaterial practice view, then, directs our attention as researchers away from trying to discover the ‘truth’ about particular emotional experiences associated with, for example, IT implementation, to focusing our attention on what people are doing when they say they experience certain emotions. It is in the sociomaterial practice, which includes the social scripts that allow us to put a name on certain feelings, and the social objects, which carry emotionological meanings, that subjectivities are produced and reproduced. As Bourdieu (1990: 53) indicated, stimuli (like a new IT package) only ‘elicit’ emotional responses when they “encounter agents conditioned to recognize these”. Paying more attention to the emotional aspect (including both affect and emotionologies) of practices can therefore help us to further unpack the embodied and socially and materially contingent aspect of our experience as feeling subjects with biographical awareness (Thompson, 2012).

174

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

7. Conclusion Sociomateriality offers considerable promise for researchers who want to examine IT implementation and use as emergent. It allows for the tracing of the practices and outcomes of implementation and use to relationships between social and material things and the exploration of how, in a particular context, these relations produce a more or less expertly choreographed performance. Our analysis of a particular instance of such practice (introducing a new IT tool for completing annual activity reports in university settings) demonstrated how two different ontological views of sociomaterial relations were equally useful in terms of enabling us to see the emergence, with both encouraging us to refrain from either overly social or overly technical accounts. However, we also showed how both accounts remained rather flat, in that they missed out the essence of the situatedeness or ‘being-in-the-world’ of the practice or, as we have named it – the felt quality of practice. We demonstrated that, without reverting to a personalized view of emotions as something purely individual, it is possible to include an emotional register in our sociomaterial theorizing; this, we argue, brings the account to life. By introducing this felt quality of practice we contend that, not only is our theorizing improved, but the relevance of our research in enhanced. Thus, without explicit attention to this felt quality, we might more readily ‘throw’ new objects into a context, forgetting that every context is not a neutral container, but an ‘equipped context’ (Gherardi, 2012). Injecting or removing new objects into any context, therefore, not only changes the extent to which the environment is prepared for competent practice, but also relationally produces a new felt quality—one that might well include negative emotions that can stabilize around a particular sociomaterial assemblage and so make it extremely difficult for a productive practice to emerge. References Critical realism: Essential readings. Archer, M., Bhaskar, R., Collier, A., Lawson, T., & Norrie, A. (Eds.). (1998). London: Routledge. Ashkanasy, N. M. (2003). Emotions in organizations: A multi-level perspective. Research in Multi-Level Issues, 2, 9–54. Bagozzi, R. P. (2007). The legacy of the technology acceptance model and a proposal for a paradigm shift. Journal of the Association for Information Systems, 8(4), 244–254. Barad, K. (1998). Getting real: Technoscientific practices and the materialization of reality. Differences: A Journal of Feminist Cultural Studies, 10(2), 88–128. Barad, K. (2003). Posthumanist performativity: Toward an understanding of how matter comes to matter. Signs: Journal of Women in Culture and Society, 28(3), 801–831. Barsade, S. G., & Gibson, D. E. (2007). Why does affect matter in organizations? Academy of Management Perspectives, 21(1), 36–59. Beaudry, B. A., & Pinsonneault, A. (2010). The other side of acceptance: Studying the direct and indirect effects of emotions on information technology use. MIS Quarterly, 34(4), 689–710. Bhaskar, R. (1978). A realist theory of science. Hemel Hempstead: Harvester. Bourdieu, P. (1990). The logic of practice. Stanford: Stanford University Press. Brave, S., & Nass, C. (2002). Emotion in human–computer interaction. In J. A. Jacko, & A. Sears (Eds.), The Human–Computer Interaction Handbook: Fundamentals, Evolving Technologies and Emerging Applications (pp. 81–93). New Jersey: Lawrence Erlbaum Associates Inc. Callon, M. (1986). Some elements of a sociology of translation: Domestication of the scallops and the fishermen of Saint Brieuc Bay. In J. Law (Ed.), Power, Action and Belief: A New Sociology of Knowledge? (pp. 196–233). London: Routledge and Kegan Paul. Ciborra, C. (2006). The mind or the heart? It depends on the (definition of) situation. Journal of Information Technology, 21(3), 129–139. Elfenbein, H. A. (2007). Emotion in organizations: A review and theoretical integration. Academy of Management Annals, 1(1), 315–386. Feldman, M. S., & Orlikowski, W. J. (2011). Theorizing practice and practicing theory. Organization Science, 22(5), 1240–1253. Fineman, S. (2008). Introducing the emotional organization. In S. Fineman (Ed.), The emotional organization: Passions and power (pp. 1–14). Malden, MA: Blackwell Publishing. Fineman, S. (2010). Emotion in organizations—A critical turn. In B. Sieben, & Å. Wettergren (Eds.), Emotionalizing organizations and organizing emotions (pp. 23–41). Basingstoke: Palgrave Macmillan. Gherardi, S. (2012). Sociomaterial practices and technological environments. In S. Gherardi (Ed.), How to conduct a practice-based study: Problems and methods (pp. 77–102) (ElgarOnline). Healy, S. (2014). Atmospheres of consumption: Shopping as involuntary vulnerability. Emotion, Space and Society, 10, 35–43. Hirschheim, R. A., & Klein, H. K. (2006). Crisis in the IS field? A critical reflection on the state of the discipline. In J. King, & K. Lyytinen (Eds.), Information systems: The state of the field (pp. 71–146). Chichester, UK: John Wiley & Sons. Kim, H. W., Chan, H. C., & Chan, Y. P. (2007). A balanced thinking–feelings model of information systems continuance. International Journal of Human–Computer Studies, 65(6), 511–525. Klein, H. K., & Myers, M. D. (1999). A set of principles for conducting and evaluating interpretive field studies in Information Systems. MIS Quarterly, 23(1), 67–93. Latour, B. (1987). Science in action: How to follow scientists and engineers through society. Milton Keynes: Open University Press. Latour, B. (2005). Reassembling the social: An introduction to actor–network theory. Oxford, UK: Oxford University Press. Leonardi, P. (2013). Theoretical foundations for the study of sociomateriality. Information and Organization, 23(2), 59–76.

M.-K. Stein et al. / Information and Organization 24 (2014) 156–175

175

Leonardi, P. M. (2009). Why do people reject new technologies and stymie organizational changes of which they are in favor? Exploring misalignments between social interactions and materiality. Human Communication Research, 35(3), 407–441. Leonardi, P. M. (2011). When flexible routines meet flexible technologies: Affordance, constraint, and the imbrication of human and material agencies. MIS Quarterly, 35(1), 147–167. Leonardi, P. M., & Barley, S. R. (2008). Materiality and change: Challenges to building better theory about technology and organizing. Information and Organization, 18(3), 159–176. Leonardi, P. M., & Barley, S. R. (2010). What's under construction here? Social action, materiality, and power in constructivist studies of technology and organizing. Academy of Management Annals, 4(1), 1–51. Levina, N., & Vaast, E. (2005). The emergence of boundary spanning competence in practice: Implications for implementation and use of information systems. MIS Quarterly, 29(2), 335–363. McGrath, K. (2006). Affection not affliction: The role of emotions in information systems and organizational change. Information and Organization, 16(4), 277–303. Navaro‐Yashin, Y. (2009). Affective spaces, melancholic objects: Ruination and the production of anthropological knowledge. Journal of the Royal Anthropological Institute, 15(1), 1–18. Orlikowski, W. J. (2007). Sociomaterial practices: Exploring technology at work. Organization Studies, 28(9), 1435–1448. Orlikowski, W. J., & Scott, S. V. (2008). Challenging the separation of technology, work and organization. Academy of Management Annals, 2(1), 433–474. Ortiz de Guinea, A., & Markus, M. L. (2009). Why break the habit of a lifetime? Rethinking the roles of intention, habit, and emotion in continuing information technology use. MIS Quarterly, 33(3), 433–444. Perrow, C. (1967). A framework for the comparative analysis of organisations. American Sociological Review, 32(2), 194–208. Pickering, A. (1995). The mangle of practice: Time, agency and science. Chicago: University of Chicago Press. Robey, D., & Boudreau, M. -C. (1999). Accounting for contradictory organizational consequences of information technology: Theoretical directions and methodological implications. Information Systems Research, 10(2), 167–185. Sassen, S. (2002). Towards a sociology of information technology. Current Sociology, 50(3), 365–388. Scheer, M. (2012). Are emotions a kind of practice (and is that what makes them have a history)? A Bourdieuian approach to understanding emotion. History and Theory, 51, 193–220. Scherer, K. R. (2005). What are emotions? And how can they be measured? Social Science Information, 44(4), 695–729. Slife, B. D. (2004). Taking practice seriously: Toward a relational ontology. Journal of Theoretical and Philosophical Psychology, 24(2), 157–178. Stam, K. R., & Stanton, J. M. (2010). Events, emotions, and technology: Examining acceptance of workplace technology changes. Information Technology & People, 23(1), 23–53. Stein, M. K., Newell, S., Wagner, E., & Galliers, R. D. (2012). Continued use of IT: An emotional choice. Proceedings of the International Conference on Information Systems, Orlando, Florida, December 2012. Suchman, L. (2007). Human–machine reconfigurations: Plans and situated actions (2nd Ed.). Cambridge, UK: Cambridge University Press. Thompson, M. (2009). Where has everyone gone? Re-integrating people into accounts of organizational practice. Working Paper. Judge Business School. Thompson, M. (2012). People, practice, and technology: Restoring Giddens' broader philosophy to the study of information systems. Information and Organization, 22(3), 188–207. Thrift, N. (2008). Non-representational theory. Space, politics, affect. London, UK: Routledge. Wagner, E., Newell, S. M., & Piccoli, G. (2010). Understanding project survival in an ES environment: A practice perspective. Journal of the Association for Information Systems, 11(5), 276–298. Walsham, G. (1993). Interpreting information systems in organizations. New York: John Wiley and Sons, Inc. Woodward, J. (1958). Management and technology. London, UK: Her Majesty's Stationery Office. Wright, C., & Nyberg, D. (2012). Working with passion: Emotionology, corporate environmentalism and climate change. Human Relations, 65(12), 1561–1587. Zhang, P. (2013). The affective response model: A theoretical framework of affective concepts and their relationships in the ICT context. MIS Quarterly, 37(1), 247–274.