Enactment or Performance? A Non-dualist Reading of Goffman

. This paper contributes to the sociomateriality research orientation with a critical examination of two concepts – enactment and performance – that have been associated with the notion of performativity. While a preference for the term enactment has been expressed in influential IS literature, we argue that sociomateriality will benefit from an engagement with the body of research that focuses on Goffman’s notion of performance. We provide a critique of Mol’s reading of Goffman’s notions of “persona” and “mask”. We then show how a careful non-dualist reading of his work reveals his opus as relevant and useful for sociomateriality, because his notion of performance affords locating technology in differing roles within a performance. In doing so, we argue that Goffman’s work, largely overlooked within this stream of research so far, contributes important concepts and terminology for making sociomateriality actionable for IS.


Introduction
Sociomateriality has predominantly been grounded in a performative, non-dualist ontology [1,2,3,4,5].Such an ontology holds that reality does not exist independently of action but rather is brought into being and sustained through materialdiscursive practices [6,7].This understanding challenges our often taken-for-granted dualist understanding, which holds that reality is "objective and out there", existing independently of the models by which we know and represent it (in the mind) [8].
While performativity is often mentioned in sociomateriality research [1], [5], [9], the vocabularies and histories involved in this approach are sometimes slippery and present us with particular choices and challenges.
In this paper we examine two terms that are associated with performativity: enactment and performance.We locate in the performativity literature a suspicion against the term performance and a preference for the term enactment [1], [10].Intrigued by this preference we conduct a critical hermeneutic reading of this discussion.Our analysis reveals that the preference for the term enactment may inadvertently conceal a body of literature on performance that is of relevance to sociomaterial theorizing: the work of sociologist Erving Goffman [11].
Consequently, we come to reconsider the relationship between the two terms.We argue that while boundaries, materialities and agencies are enacted, these can productively be understood as effects of sociomaterial performance.We thus demonstrate that a sociomaterial reading of Goffman's notion of performance holds potential for enriching sociomateriality research.

Performativity in Sociomateriality Research in IS
According to Orlikowski and Scott [1], performativity is a central concept for sociomateriality because this worldview aligns with an interest in how boundaries and relations are actively brought into being rather than existing a priori and universally: For scholars of sociomateriality, the notion of performativity draws attention to how relations and boundaries between humans and technologies are not pre-given or fixed, but enacted in practice.( [1], p. 462, our emphasis) This view challenges an understanding of the world as being already furnished with entities having inherent boundaries and properties that define what they are.Instead, in a performative understanding, entities are what they are through their relations; they come to be recognized as objects, with clear boundaries and properties, only in and through "practice" [1].
The term "enacted" is used to describe how such boundaries and relations are brought into being.Enactment is further positioned as the preferred way of understanding how reality is produced in sociomateriality research, in contrast to the notion of performance: A central idea entailed in sociomateriality is the notion of performativity (Barad, 2003).While related to the notion of performance, performativity is not synonymous with it.Where "performance" refers to the doing of some activity (as when a physician "performs" a medical examination, or a musician "performs" in front of an audience), performativity refers to enactment.[1] (p. 460) It is clear from this account that "performance" is being associated merely with the "doing" of an "activity" and thus demoted in favour of "enactment" [1].
Intrigued by this distinction, we conducted a hermeneutic reading of the wider body of sociomateriality literature and found more evidence of an explicit preference for the term enactment.Barad [12] and Suchman [2] for example both theorized that boundaries, materialities, and even agencies are effects that are "enacted".Suchman specifically uses the word enactment in response to the linguistic challenges of talking about material agency and the fluid boundary that arises between human and machine in action: The problem is less that we attribute agency to computational artifacts than that our language for talking about agency, whether for persons or artifacts, presupposes a field of discrete, self-standing entities.As an alternative, we can take the interface not as an a-priori or self-evident boundary between bodies and machines but as a relation enacted in particular settings and one, moreover, that shifts over time.[2] (p.263) What is "enacted" here is a particular kind of phenomenon -an interface or boundary -and the term enactment captures well the relational, situated and temporal way in which boundaries are brought into being.Enactment is thus a useful term for sidestepping some of the more intentional connotations of performance, and lets us talk about the significant issue of how boundaries come into being and therefore how objects are stabilized in practice.
However, in the spirit of advancing the sociomateriality research agenda we consider how a re-reading of the concept of performance, from a relational, non-dualist perspective, might assist researchers in their empirical and conceptual development of sociomaterial understandings of everyday life.We recognize that an emphasis on the emergence and stabilization of boundaries is important to a sociomateriality research agenda.What is less clear, however, is how we are to study these processes of "materialization" [6].
There has, for instance, been some agreement that from a sociomaterial perspective, things, people, and practices are entangled [13,14].The ontological inseparability implied in this word is indeed a necessary starting point for challenging the dominant dualist position in IS.There is, however, a risk that we are left without vocabularies to talk about how what we here term collectives -of things, people and practices -come to cooperate in the sociomaterial production of reality.We recognize that one challenge that researchers face in investigating sociomaterial phenomena is in accessing the kinds of activities [7] that work to produce effects, such as interfaces and boundaries between entities.Another challenge lies in how to talk about the role of humans and non-humans in bringing about this ongoing activity, which from a performative understanding works to stabilize certain realities and marginalize others [3].
We suggest as a possibility here that the word performance, if treated carefully, offers a less recognized and potentially rich path for talking about empirical inseparability and its active, processual character.In the following hermeneutic reading, we put forward an argument where the concept of performance opens up a body of literature to researchers who are interested in sociomateriality that may otherwise be considered off limits, because of the extant literature's dismissal of the concept in preference of enactment.
We will show that performance has a rich history in feminist, sociological and existentialist literature, and demonstrate that in our readings of key texts, the concept of performance is not necessarily tied to the notion of an intentional, individualistic exercise of choice or whim.It is thus possible to divorce the notion of performance from a dualist preoccupation, which has traditionally either overstated the role of human agency or reduced all action to subjective experience.
We demonstrate an alternative reading, where the term performance is understood as a complex, collective activity that gives rise to "effects" [2], [11] that can be understood in performative terms as working to sustain certain realities and marginalize others.We draw on and interpret Goffman's dramaturgical framework to bring forward literature on performance that may otherwise be viewed as sitting outside of our emerging research tradition.Finally, we offer examples illustrating how such a reinterpretation of Goffman could inform the study of emerging IS phenomena.

Methodology: A Hermeneutic Reading
To understand how the notion of performance relates to performativity, and furthermore how it may align with sociomaterial IS research, we read back over key texts in an iterative manner [15].We have followed a "reference trail" in looking backwards in time to read into how these conversations emerged between authors referring to each other's texts [15].This hermeneutic process informs our critical reading of texts that have been in conversation with one another.We interpret them afresh, for an IS readership interested in sociomateriality research, as we read one text through another [16].Hermeneutics holds that understanding is an ongoing process of interpretation and re-interpretation where a final understanding is neither sought nor possible [15].Nevertheless, using a hermeneutic approach, a reader can compare interpretations and come to a different understanding of how a text and its ideas relate to contemporary conversations.Consequently, by conducting a hermeneutic reading we do not seek to unpack the true meaning of a particular text or quote.Following our sociomaterial orientation we do not hold that meanings are fixed universally but that they come into being through engagement with the text in a particular way and at a particular moment in time.Specifically, we argue that any text is always read on a particular prior understanding, in a certain context, and is therefore never value-free but rather historically charged.
In the following we will demonstrate the grip that the dominant, dualist understanding [17] of the world exerts in readings of texts and that an awareness of this influence can be productive, and is indeed necessary for the advancement of the socio -materiality project.While we do not want to imply that we can fully step outside and avoid this influence we argue that it is important for scholars of sociomateriality to be aware of the ways in which historical texts might have been read previously by others and consequently how they can be read differently when assuming a different stance.We thus seek to show where a sociomaterial background allows interpreting texts in a new way that will provoke further discussion and debate.

4
Mol's Objections to Goffman's "Performance" In this section we outline our hermeneutic reading of a conversation that we have located in texts by Mol [10] and Goffman [11] and other authors that Goffman draws upon.In our reading of Mol [10] we locate an explicit dismissal of the term performance.We iterate between texts to explore the charges brought against the concept and to suggest an alternative reading.In Mol's 2002 text The Body Multiple, we find a potential source of the strong distinction being made between performance and enactment, and an explicit argument against the noun "performance", and in favour of the verb "enact": In the literature there has been a lot of discussion about the term performance -a term that does not only resonate the stage but also success after difficult work and the practical effects of words being spoken.I do not want these resonances, nor do I want this text to be burdened with discussions that it seeks no part in […] It may be helpful to avoid the buzzword.To look for another term.A word that is still relatively innocent, one that resonates with fewer agendas.I have found one.And, even if I have been using the term performance elsewhere in the past, I have carefully banned it from the present text.I use another verb instead, enact, for which I give no references, precisely because I would like you to read it in as fresh a way as possible.In practice, objects are enacted.[10] (p.38-41, our emphasis) Mol goes on to give two reasons for her choice.Firstly, she is critical of Goffman's dramaturgical analysis of performance in everyday life, attributing to him the notion that "adults have real selves deep down, backstage" in which case the "identity people perform is not deep, it is a mere performance" [10] (our emphasis).Mol's second criticism is of Judith Butler's theorization of gender as performed [6].In both of these critiques, the main source of concern is the idea of performance being a "mere" putting on of a "mask" or external identity: an isolated act by a willful agent that is not contingent on physical realities [10].
In the following, we investigate Mol's objection further by looking closely at what she refers to as Goffman's "outdated text" [10], to see what we can learn about performance and what of the concept can be salvaged for sociomaterial theorizing. 1 We begin by looking more closely at Mol's critique of Goffman, before we go to Goffman's text, The Presentation of Self in Everyday Life [11].We come to appreciate that while Goffman's text is indeed in some ways "outdated" (for example, in his discussions of race and gender), his concept of "performance" is, when read in a particular way, not only compatible with a performative worldview but also adds important new distinctions to the sociomateriality project.

5
A Non-dualist Re-reading of Mol's Critique of "Self" in Goffman Given that The Body Multiple [10] is an influential text for many sociomateriality researchers (ourselves included), Mol's banning of the word "performance" in 2002 in favour of "enact" carries significance.This led us to investigate her reasoning further.

Mol's Critique of the Performed Self
Mol [10] characterizes Goffman's 1959 text as concerned with a distinction between the real versus performed selves that are presented in everyday life.According to Mol's critique, Goffman puts forward an account of human identity where a real self exists a priori and universally, but this real self is concealed in the presence of others through "performance".Mol [10] describes Goffman's work as follows: In 1959, Goffman borrowed the language of the theatre in order to talk about human subjects.When people present themselves to each other, Goffman said, they present not so much themselves but a self, a persona, a mask.They act as if they were on a stage.They perform … adults have real selves deep down, back stage … The identity people perform is not deep, it is a mere performance.[10] (p.36) In this critique, performance is associated with the concepts of "persona" and "mask", which are supposedly "less real".In other words, whatever a person performs "as if they were on a stage" is only a surface impression, while a "real" self is hidden "deep down, back stage".Such a reading of Goffman would indeed make him incompatible with a performative world view, because it implies a dualist understanding of the world as split into a persistent "real" world that is covered over by a kind of subjective, intentional, arbitrary surface -a "mask" that is taken on and discarded at will by "human subjects".
In the following we will challenge this reading of Goffman's work.In particular, we will demonstrate that such a reading already assumes a dualist position a priori, and that when taken on a non-dualist background his work is revealed very differently.For doing so we go to his text, to see what Goffman [11] has to say firstly about personas and masks, and then more fundamentally what he has to say about the topic of "reality" itself.

Personas and Masks in Goffman
We find that Goffman explores the concepts of persona and mask in his book mainly through reference to other authors.We will thus provide three quotes by way of example and demonstrate in each instance that dualist and non-dualist readings draw very different boundaries, highlight different parts and thus lead to very different interpretations.In a second step we show that Goffman utilizes the notion of perfor-mance to develop a non-dualist reading that, we argue, has not only been marginalized in sociomaterial theorizing so far, but offers useful distinctions and concepts to IS scholars.Firstly, Goffman [11] quotes a text by Park [18] to show that the etymology of the word "person" is strongly linked to the word "mask".In contrast to Mol's reading, a mask is considered here as preceding the organic being inhabiting it: It is probably no mere historical accident that the word person, in its first meaning, is a mask.It is rather a recognition of the fact that everyone is always and everywhere, more or less consciously, playing a role … It is in these roles that we know each other; it is in these roles that we know ourselves … In a sense, and in so far as this mask represents the conception we have formed of ourselves -the role we are striving to live up to -this mask is our truer self, the self we would like to be.In the end, our conception of our role becomes second nature and an integral part of our personality.We come into the world as individuals, achieve character, and become persons.[18] (pp.249-250, in [11] p. 30) When read on a dualist background the focus is likely to be on the distinction between person and role, evident in statements such as "everyone is […] playing a role", which might lead to the conclusion that "role" or "mask" conceal what is otherwise the real person.However, when we take the text on a non-dualist background, what stands out is the way in which the "mask" is what we know each other as and what we become.Contrary to Mol's [10] interpretation, this fragment then becomes an argument for an understanding of identity as being i) relational and ii) an ongoing process.By pointing out that in a sense, "this mask is our truer self", a performative understanding of identity can be assumed that is quite distinct from the dichotomized understanding of real self vs. fake (performed) self.
Secondly, Goffman [11] explores this issue further by quoting an older text by Durkheim [19], who explains that "personas" are masks that become a materialized constant, that distracts us from the precarity of organic existence: Everyone who is sure of his mind, or proud of his office, or anxious about his duty assumes a tragic mask.He deputes it to be himself and transfers to it almost all his vanity.While still alive and subject, like all existing things, to the undermining flux of his own substance, he has crystallized his soul into an idea … Our animal habits are transmuted by conscience into loyalties and duties and we become "persons" or masks.[19] (p.272, in [11] p. 65) In this fragment the contrast between a dualist and non-dualist reading becomes even more pronounced.On a dualist view what stands out is that everyone is "his own substance", but becomes "a tragic mask" in the process of engaging with the social world.Note that when one already presupposes the existence of a real self, the text merely reaffirms the split between real self and mask.Conversely, if we read the same text on a non-dualist pre-understanding, where we do not assume that one's self is fixed a priori, what stands out is how a "person" is always performed in ongoing so-ciomaterial practice of "holding office"; in other words, through participating in life.In this second reading, the fragment very much exhibits a performative understanding of reality, where reality is stabilized over time through iterative activity [7].The notion of an idea being "crystallized" against a background of temporal organic flux is reminiscent of how, in sociomateriality research, boundaries and materialities are considered to be stabilized in practice as effects of activities and routines [5].
Goffman makes further reference to masks, this time in commenting on a text by existentialist feminist philosopher de Beauvoir [20]: Through social discipline … a mask of manner can be held in place from within … [but] we are helped in keeping this pose by clamps that are tightened directly on the body, some hidden, some showing.[11] (p. 65) Again, on a dualist account, the focus will be on the mask that literally becomes something held before the body to conceal the self.Yet when taken on a non-dualist background what stands out is the reference to "social discipline" and the involuntary nature of masking, in that personas or masks are not only put on at will by a human subject as agent, rather they are "held in place from within" and from the pressures of being a part of a social practice.This account can thus be read as a sociomaterial treatment of how society and physicality (the mask, the body, clamps) are entangled in constructing identities and practices, where one entity does not have sole custody over reality.Rather, entities are entangled in practice and co-define one another, in a manner that is politically charged.
When read in this way, the above quote contains a further elaboration on the way in which things are involved in social performances, which are important for how we stabilize our identity in the world.Goffman supports this understanding by arguing that inhabiting a role requires us to engage in practices which are deemed appropriate by others -in other words, we do not merely and knowingly take on a role, as we would put on a mask.Instead, we work at being known as our roles through performance, in a social process wherein our very identity is at stake: To be a given kind of person, then, is not merely to possess the required attributes, but also to sustain the standards of conduct and appearance that one's social grouping attaches thereto.The unthinking ease with which performers consistently carry off such standard-maintaining routines does not deny that a performance has occurred, merely that the participants have been aware of it.[11] (p. 81) Central to this argument is firstly that a role is a collective involvement: it is active, even if the performer is not aware of their activity.Secondly, the framework from which the performance is derived and against which its success is judged sits outside of the individual performer.Consequently, we argue that, if we free ourselves from dualist ontological baggage, we are able to interpret this argumentation to mean that performances are always inherently at once active as well as both social and material, where one category informs and depends upon the other.
As a result, the routines and accessories that Goffman discusses as being part of performance are not mere frivolity and fancy, rather they are centrally involved in the work that is required for society to affirm, legitimize and bestow one's identity.We suggest that such a reading of performance moves well past the notion of a "mere" performance, involving the willful putting on and taking off of a surface-level mask as artifact.Rather, we come to interpret Goffman and those he cites as portraying performance as the collective work that goes into sustaining reality; the materialization of which we all in turn depend upon for our understanding for participating in practice.

A Non-dualist Re-reading of Goffman's Notions of Performance and Reality
Having challenged Mol's critique of Goffman's work through a hermeneutic rereading, we will now take a closer look at Goffman's own statements about reality and performance.When taken on a non-dualist account, we find evidence that Goffman himself challenges a dualistic understanding of self and reality in subtle yet forceful ways.We argue that there are instances in the text that might be missed if one has already concluded that his talk about "masks" and "personas" must presuppose the existence of a "real self" and hence a dualist view of the subject matter.We have selected two key quotes that we believe demonstrate that Goffman [11] applies a more nuanced approach than has been portrayed.

The "Self" in Performance
The first relevant component of Goffman's 1959 thesis here is that he made a distinction between "performer" and "character".Importantly, Goffman does not attribute either of these categories to the category of "real self".Instead, he shows how "the self" emerges in a relational sense between the two, in the process of performance: A correctly staged and performed scene leads the audience to impute a self to a performed character … The self, then, as a performed character, is not an organic thing that has a specific location, whose fundamental fate is to be born, to mature, and to die; it is a dramatic effect arising diffusely from a scene that is presented, and the characteristic issue, the crucial concern, is whether it will be credited or discredited.[11] (pp.

244-245)
This explanation of performance does not at all rely on a separation between real and surface self, rather, the self is the "dramatic effect" that arises diffusely from a scene in which a performer is involved in performance.The performer does their best to "correctly" stage and perform the scene, "in order to lead the audience to impute a self to a performed character".The audience can, however, as a collective, credit or discredit the scene that is being presented.The performer thus relies not only on their own work but also on their team mates and their audience in the process of becoming a self.

6.2
Performance and "Reality" The "scene" that is performed by a collective is what Goffman [11] refers to as the "impression of reality" that is "fostered" or "sponsored" by the performing group.It is this fostered version of reality that is at stake in matters of performance before an audience.We note that it is also important to realize that a performer depends on many others (e.g.team mates, equipment, setting) in fostering a particular version of reality -the smallest betrayal could interrupt and discredit the scene and thereby the credibility of the performer's character.Paying attention to the role of the audience in performance is thus crucial to understanding why performance is never entirely in the hands of the performer, or even in the collective efforts of the performing team.
In the following quote, Goffman addresses how such a conceptualization of a collectively fostered impression of reality relates to a conventional scholarly preoccupation with locating the real "reality"; in other words, Goffman here gives us an explicit account of his position on the matters discussed above: While we could retain the common-sense notion that fostered appearances can be discredited by a discrepant reality, there is often no reason for claiming that the facts discrepant with the fostered impression are any more the real reality than is the fostered reality they embarrass.A cynical view of everyday performances can be as one-sided as the one that is sponsored by the performer.For many sociological issues it may not even be necessary to decide which is the more real, the fostered impression or the one the performer attempts to prevent the audience from receiving.The crucial sociological consideration, for this report at least, is merely that impressions fostered in everyday performances are subject to disruption.[11] (pp. 43-44, our emphasis) We see that Goffman directly addresses his position in conceptualizing "reality" by challenging the "common-sense notion" that there is a real "discrepant reality" behind the performance [11].What he takes on here is nothing other than what we have so far referred to as the dualist position that always proceeds from the assumption of an a priori and universally existing essential reality.Conversely, he seems to argue that performance, far from being more or less "real", is all that we have access to: performance is all there is.In contrast to the dualist understanding, Goffman demonstrates that there is no ground that is "real" as an objective yardstick against which performances can ultimately be judged.Rather, some ways of doing things become solidified as the practice, and these then act as a background against which other performances are judged as being appropriate or inappropriate.
Goffman [11] further suggests that rather than focus on "what reality really is", studies of performance should investigate the interesting question of how a particular version of reality becomes sustained at all and, in turn, what competing impressions of reality might unsettle it.In other words, certain boundaries that make up reality become the object of study, not the given ground from which to proceed, which is precisely the project of sociomateriality as we understand it: We will want to know what kind of impression of reality can shatter the fostered impression of reality, and what reality really is can be left to other students.We will want to ask, "What are the ways in which a given impression can be discredited?"and this is not quite the same as asking, "What are the ways in which the given impression is false?"[11] (p. 44) In conclusion, we do not read Goffman's research agenda as claiming that "adults have real selves deep down, back stage" [10] (p.36).Rather, we find that pursuing the question of what is "really real" is unproductive at best and a remnant of a particular common-sense, dualist grounding at worst.Instead, we find that Goffman [11] suggests that the ways in which social activities bring about and sustain realities, through what he calls performance, is worthy of attention.In drawing our efforts to the issue of how particular versions of reality are potentially threatened by other discrepant impressions, we are further prompted to consider the fragility of local realities; what is marginalized from them and what threatens them, as well as the ongoing collective work that goes into staging the performances that sustain what we take to be "real" in everyday life.

Re-interpreting Goffman for Sociomateriality Research in IS
In this section we first summarize the line of argument we have woven through our hermeneutic reading: that Goffman's notion of performance is compatible with a contemporary, performative sociomaterial perspective in IS.We then demonstrate what we believe a contemporary interpretation of Goffman's work on performance can add to sociomateriality research.Specifically, we adapt and interpret Goffman from a post-humanist perspective, and argue that it is possible and productive to consider how technologies play various roles in sociomaterial performance, for example as not only "prop" but as setting, sign-equipment, team-mate, director, or audience.We present this perspective not as a finished research tool but as a starting point for further discussion and debate as to whether and how Goffman and others' notions of performance can enrich sociomateriality research.
Before we consider what Goffman [11] may have to offer IS scholars with his conceptualization of performance, we first freely recognize that his text upon which we draw is not concerned with technology.It was written before many of the kinds of technologies we are interested in today existed.We acknowledge this fact but do not deem that this precludes us from considering how we might adapt concepts from his work for a contemporary research agenda.We also note that previous adaptations of Goffman have already been made in IS studies that consider his better known theorizing of "back stage" and "front stage" regions [21,22,23,24].

Goffman and Inseparability
We have shown that Goffman [11] described "character" as an effect that arises from a scene of action.In his conceptualization, the performer is involved in the scene, but the character is a much more precarious category that depends on a successful staging of the performance.This staging involves the setting, props, team-mates, routines, and conventions, and, most importantly, a scrupulous audience that ascertains whether the performance "comes off".In this way, the performer owes much of their characterthat is, their social role, as for example lawyer or manager -to a host of "others".Performance is in this conceptualization inherently collective.This is in keeping with a performative worldview which recognizes the interdependency of the work that goes into sustaining reality.As a way of considering the world, sociomateriality emphasizes this inseparability of humans and technologies in practice [1].The point of the concept of inseparability is that all entities rely on one another for their identities.This notion is well expressed in this excerpt from a study that Goffman [11] cites, about pharmacist practice: The store is, in a sense, a part of the pharmacist.Just as Neptune is pictured as rising from the sea, while at the same time being the sea; so in the pharmaceutical ethos there is a vision of a dignified pharmacist towering above shelves and counters of bottles and equipment, while at the same time being part of their essence.[11] (p. 99) This evocative example echoes the earlier fragments that Goffman cites.We take this excerpt to demonstrate further the argument that humans could not inhabit "personas" without a host of "things", which in turn find a place in the world through their role in the stabilization of roles and identities in practice.Performance is here understood as inherently collective: any character is stabilized only in a collective, successfully staged scene that relies on a host of "others".This appreciation of the collective effort of performance, which is required for reality to be brought into being and stabilized, is we argue compatible with a performative sociomaterial perspective.

What Goffman's Notion of Performance Contributes: Locating Technology
Sociomateriality is often associated with the term "entanglement" [13], [16], referring to an ontological inseparability between things and people, and sociality and materiality, in practice.While we acknowledge the significance of this concept, we also recognize that the language of entanglement can be challenging for the research process because it leaves us with the question of exactly what is entangled and how entanglements play out in practice.In Goffman's conceptualization of performance, we are given a number of useful terms that we argue can help us in teasing out the different ways in which humans and non-humans may be implicated in and contribute to collective performances.We here offer a brief introduction of these terms and speculate on how they could apply to sociomaterial studies of IS phenomena.
Goffman [11] uses the term setting for the environment in which a performance takes place; sign-equipment for props; team and team-mates for the performing group; roles such as the director for members of a performance who hold special access to the impression of reality that is being fostered; and audience for those who are being performed to.Some of these terms have been previously explored in IS literature (e.g.[25]).What we believe is underdeveloped, however, is a sociomaterial interpretation of these terms.
While any analytic language will have to cut a phenomenon in a certain way (and the performance language is no different), the performance notion as we interpret and present it here does not make any assumptions a priori about the being or identity of entities involved in a particular performance; it is the project of the researcher to provide their interpretation of the situation aided by the performance lens.Importantly for the study of the sociomateriality of IS phenomena, the lens we present does not make a priori assumptions about whether characters or entities involved in the performance are filled by human or non-human actors.This allows locating technology in various places and roles within a performance.In the following we interpret and illustrate a sociomaterial application of selected terms.
Technologies as setting: A more conventional way to consider technologies in a performance might be to focus on their role as setting.This is how IS literature has most commonly made use of Goffman [11], to show how a particular software platform, such as Facebook for example, can act as a front stage or back stage space for performance of identity.In this analogy we might consider how a technology provides a space within and upon which "team-mates" can plan and stage their performance.For example, an Enterprise Social Network may become a stage upon which employees can demonstrate their allegiance to the organization, with the understanding that managers are monitoring the platform.An instant messaging system on the other hand might provide a back stage space within which more covert actions can be planned.In this way the concept of technologies as setting can be shown to sustain action in a way that is inherently sociomaterial.
Technologies as sign-equipment: A further straightforward application of Goffman's performance terminology in IS would be to say that technology can act as sign-equipment that helps actors to display their status, position, and identity.For example, a doctor may carry an iPad as a signal that they are engaged in contemporary healthcare, or a laptop could be a signal that a worker is not tied to a fixed desk.
While these two conceptual categories of technologies as setting and sign equipment are potentially useful, we suggest that a more radical re-reading of Goffman is possible, and that this reading is of interest to researchers examining emergent phenomena in IS from the perspective of sociomateriality.We use a call centre setting in the following examples to illustrate opportunities for future research.
Technologies as team-mates: In some research contexts work performance may be investigated by positioning technology as playing the role of a team-mate in the staging of a performance.Team-mates bestow other actors in a performance with credibility, as they support the impression of each other's characters.An algorithmic Decision Support System (DSS) in a call centre, for example, could therefore be considered a team-mate to the phone operator in the staging of a performance where the phone operator's credibility and identity is in no small part in the hands of the DSS.
Technologies as director: In stricter call centres, where the operator has very little autonomy in their role, such an algorithmic DSS companion may even be understood to be playing the role of director in the performance.That is, the concept of technolo-gy as director can be used to describe a scene where, for example, an algorithm controls the development of the scene's action to a greater extent than the human actor.
Technologies as audience: Equally in a call centre context, an emphasis on sociomaterial performance from a Goffmanian perspective would also require the researcher to consider the audience's contribution to the scene because Goffman [11] reminds us that all performances are vis-à-vis another party.Is the customer who is served by the call centre the audience of the performance?Certainly, but in most call centres, technology introduces further audiences, such as when the call itself is recorded for later review or aspects of call "performance" (such as length) are automatically recorded and measured.As a result, the phone operator's performance becomes staged in relation to an entire collective of human and non-human audiences.
We thus point out that multiple audiences often shape performance.Of particular interest to IS researchers might be when technology itself becomes the audience of a particular performance.Think further of sensors in a factory.These may well be usefully understood in compliance practices as the audience towards which certain work practices are performed (e.g.[26]).Or take Introna and Hayes' [27] study on plagiarism detection software; here the software becomes a major audience for and even changes and shapes the performance of essay writing, in that the success of the performance hinges in no small part on whether or not it is "appreciated" by the algorithm-as-audience, thereby performing identities such as the "successful student" or the "plagiarizing cheat".
Through these examples we demonstrate, firstly, the usefulness of the performance lens, as we have interpreted it, in locating technology in particular empirical phenomena without making a priori assumptions about what exactly technology is; and secondly, that technology can be located in any role or part of a performance.This sensitization to different elements of performance thereby opens the researchers' attention to the possibility that a performance may involve more than what is immediately visible or intelligible with taken-for-granted notions of technology as merely a "tool" or "platform".Thus, a research emphasis on performance here encourages a differentiation of what is entangled, in a way that tries to avoid deciding in advance how a sociomaterial collective is involved in the maintenance of a particular reality.
By interpreting Goffman's [11] notion of performance in light of a more posthuman understanding, that remains open to how technologies and people are configuring each other in practice, we can potentially make sense of these scenes of action in a more nuanced way, by carefully considering in-situ what roles various people and technologies play in the scenes that we bear witness to as we research IS phenomena.It further stands to reason that adopting a serious performative interest in sociomateriality research requires researchers to reflect on how their own presence in the research setting is involved in the scene of action.Our audio recorders and notebooks for example introduce unspecified audiences into the performance arena.We suggest that such a reflexive approach to research methodology [28] is a necessary component of conducting research from within a performative ontology, but this goes beyond what we can consider in detail in this paper.

Conclusion
We have carried out a hermeneutic reading [15] of interconnected texts to critically consider how "performance" is understood in sociomateriality research.We find that while this term has been put aside in influential performativity texts, it is caught up in literature that is potentially useful to IS researchers interested in sociomateriality.As with any text, Goffman's work in particular is a product of its era and we acknowledge that adaptation and interpretation is necessary to make his notion of performance useful to a more post-human orientation [12].We argue, however, that Goffman's [11] original conceptualization of performance is already largely compatible with a performative understanding, which holds that reality is brought into being and sustained in action.Finally, the word "enact" no doubt has its place, particularly when referring to how boundaries, objects, and materialities come to be stabilized in practice.It is the work that goes into these practices, however, that we argue is usefully conceived of in terms of performance.We have exemplified a sociomaterial reading of Goffman using the concepts of setting, sign-equipment, team-mates, director, and audience.We further propose that when researchers conduct ethnographies in particular, they are involved in a scene, and that a greater sensitivity to how this scene is playing out may offer a starting point for meaningfully taking into account the complexity of the sociomaterial performances we witness, and help shape, through the research process.In sum, we put forward for further discussion the notion that boundaries are enacted, but that they are enacted in and through performance.In taking this as a premise for debate we looked to Goffman for an illustrative starting point and argued that his notion of performance can be usefully conceptualized for a sociomaterial IS research agenda.