LIKE IT? Ritual Symbolic Exchange Using Facebook's 'Like'

MEDIA@LSE Electronic MSc Dissertation Series Compiled by Dr. Bart Cammaerts and Dr. Nick Anstead Like it? Ritual Symbolic Exchange Using Facebook’s ‘Like’ Tool Kenneth J. Gamage, MSc in Media and Communications Other dissertations of the series are available online here: http://www.lse.ac.uk/collections/media@lse/mediaWorkingPapers/ BIO Kenneth Gamage is a 2012 graduate of the MSc Media and Communications program at The London School of Economics. He holds a BDes (2010) from Thompson Rivers University/University of Toronto and a Diploma in Interpretive Illustration (2000) from Sheridan College both from his home country of Canada. Prior to arriving at the LSE, Kenneth has worked for a nearly a decade prior as an entrepreneur specializing in communication art and digital illustration for TV, print and online catering to 50+ international clients. Some of these clients include Disney, Hooked on Phonics, The Washington Post, Oxford University Press, Reckitt Benckiser, Scholastic and Hitachi. He has illustrated several children’s books and television animation series. E: [email protected] Published by Media@LSE, London School of Economics and Political Science ("LSE"), Houghton Street, London WC2A 2AE. The LSE is a School of the University of London. It is a Charity and is incorporated in England as a company limited by guarantee under the Companies Act (Reg number 70527). Copyright in editorial matter, LSE © 2013 Copyright, Kenneth J. Gamage © 2013. The authors have asserted their moral rights. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means without the prior permission in writing of the publisher nor be issued to the public or circulated in any form of binding or cover other than that in which it is published. In the interests of providing a free flow of debate, views expressed in this dissertation are not necessarily those of the compilers or the LSE. MSc Dissertation Kenneth J. Gamage Like it? Ritual Symbolic Exchange Using Facebook’s ‘Like’ Tool Kenneth J. Gamage ABSTRACT Web 2.0 social networking sites (SNSs) are characterized as interactive sites of collaborative content production and content viewing. Facebook’s release of its content marking ‘Like’ button affords its members with even finer-grained reading and writing capabilities. Despite its apparent simplicity, the meanings re/produced and the social needs satisfied by the ‘Like’ tool may be far more complex than we currently understand. Many qualitative studies have aimed their efforts at overt displays of active audience, or ‘prosumer,’ behaviour i.e. tweets, wikis, blogs etc., yet little on micro-tool use. Conversely, pinpoint data traces inscribed by micro-tools have ushered in large-scale quantitative social network analyses. This macro view comes at the expense of obscuring the atomized person and their social needs. These two strands of research require bridging. Using a symbolic exchange framework, this study empirically explores salient patterns of communicative rituals within a fourteen participant sample. It responds to issues raised by audience research through the lenses of microsociology and anthropology. The research methodology employs semi-structured interview, thinking-aloud and non-probability variable analysis. The purpose is to explore two interrelated research questions: (1) what inter-subjective meanings does this sample of Facebook users convey and interpret using the ‘Like’ button tool? and (2) how, if at all, are the identified meanings socially constructed and ascribed with perceived value? The findings suggest that the meanings that the sample ascribes to ‘Likes’ are polysemic. Using different strategies ‘Likes’ are used to signal language, paralanguage, emotion, shared memories and gifts largely for social support. The primary strategies of influencing the perceived value of ‘Likes’ are (1) keeping them scarce, and (2) varying combinations of the following elements: social tie, ritual type, reciprocity, human attention, and interaction mode (semi-/public performance awareness). The central limitation of this study is its presumption that ‘Like’ exchanges are intended to always communicate meaning. As evidenced, ‘Likes’ are also clicked for unknown personal reasons. -2- MSc Dissertation Kenneth J. Gamage INTRODUCTION On April 2010 the social network site (SNS) Facebook Inc. launched its ‘Like’ button technology to its 900 million members (Sengupta, 2012). Defined as a collaborative Web 2.0 platform, Facebook’s ‘Like’ button is embedded below each piece of members’ shared content such as comments, posts, and status-updates. By navigating their cursors over and clicking the ‘Like’ button, members can mark each other’s content. It is a tool for mediated collaborative reading and writing. Whatever is ‘Liked’ is semi-/publically communicated, or narrowcasted, on Facebook’s blog-like “newsfeed.” Aggregated and updated in real-time, the “newsfeed” displays the minutia of each social network’s interactions ad infinitum. Here, each member’s shared content intersects with one another’s forming a collective narrative. The ‘Like’ button combines with other content producing tools to mediate this flow of content as it synthesizes, collides, and sprawls. An astonishing 362, 861 ‘Likes’ are exchanged per second, reports Time magazine (2012). ‘Likes’ dot the virtual surface of 82, 557 status updates and 79, 364 posts per minute, notwithstanding that SNS use is predominant in wealthier regions. The American Civil Liberties Union is momentarily working to constitutionally protect ‘Likes’ as free speech (Dockterman, 2012), while across the globe an Israeli couple recently named their new born daughter ‘Like’ (Haaretz, 2011). In-between the fleeting moments of Facebook members’ everyday lives, these tiny signals of attention are exchanged amongst family, friends, and acquaintances—each a gesture of virtual arms outstretched, reaching for something so apparently mundane: to be liked. Facebook defines ‘Liking’ as “an easy way to let someone know that you enjoy [content]” and a way of “giving positive feedback” (Facebook, 2012). Nevertheless, is the motivation to simply say that we “like” something all that prompts users to click the ‘Like’ button? Can this really be all that ‘Likes’ mean? The recent domestication of Web 2.0 SNSs ushers in debates, both old and new, across the social sciences. Its ripples most pronounced in the discipline of audience research. The Facebook SNS is interactive and multi-modal: it combines multiple communication features such as email, photo sharing, blogging, instant messaging, status updating etc., within users’ social networks (Cohen, 2004: 5-6). Despite their slightness, ‘Likes’ could play a key role in creating and negotiating the flows of meaning in a semi-/public fashion. The semi/publicness of sharing content is what Castells calls “mass self-communication” (2009). Content produced is ‘durable’ or permanently inscribed (Mayer-Schönberger, 2011). In describing the active audience of the contemporary media climate in contrast to prior mass media conceptions of passive audiences, Press has offered the apt characterization of “the -3- MSc Dissertation Kenneth J. Gamage post-audience age” (2006). Several others, applying Toffler’s (1980) portmanteau of producer-consumer, refer to new media users optimistically as ‘prosumers’ (1980) that comprise the “network society” (Castells, 1996). These terms capture the simultaneous readwrite affordance of SNSs and importantly the ‘Like’ button as a content marking tool. This paper responds to issues identified by audience research focusing on processes of symbolic exchange. This approach moves outside of the audience research canon (Livingstone, 1998b), integrating the dual lenses of anthropology and micro-sociology. The theoretical pivot underpinning this conception of the audience is not necessarily new, but stems from two views of communication put forth by Carey (1975): the transmission and ritual models. The former characterizes the mass media conception of sender-receiver (see Lasswell, 1948). The latter characterizes a more deep-seated social function of symbolic exchange. As an innately social process, symbolic exchange is inextricably located within the etymological trinity of what Carey identifies as “community,” “communion,” and “communication” (Ibid). Carey maintains that ritual social interaction “is directed not toward the extension of messages in space but toward the maintenance of society in time … ” (1989: 18). This trinity comprises the areas of focus that this paper commits itself to as a rearticulation of audience research’s ‘text-reader-producer-context’ – the interface of potential meaning making using the ‘Like’ tool. Using a symbolic exchange framework, this study aims to explore the processes and meanings re/produced using the ‘Like’ tool within a fourteen participant sample. It argues that the potentially diverse meanings and uses of the ‘Like’ tool are perpetually re/produced by people through everyday ritual practices, and these engagements are neither the determinations of the ‘Like’ technology nor the signification of simply liking something. To do this, the paper advances its argument along three main stages. First, a theoretical framework of symbolic exchange will be assembled. Second, the methodological application of semi-structured in-depth interviews and ‘thinking-aloud’ activity will be discussed. Third, findings and interpretation through variable analysis of coded data will be presented. The discussion will conclude by suggesting avenues for future research, the limitations of the present study, and how these findings have responded to the proposed research questions. -4- MSc Dissertation Kenneth J. Gamage THEORETICAL CHAPTER In this chapter a theoretical framework will be assembled around the concept of symbolic exchange. It will advance along the three etymological areas of Carey’s ritual communication model (1989: 18): (1) Community: a comparative discussion of various approaches to theorizing perceived social networks in sociology, social network analysis, and anthropology. This will be followed with its relation to gift exchange theory. (2) Communion: a presentation of largely Goffman’s (1971; 1959) theory of ritual social interaction and its potential compliment to online attention economy discourse. (3) Communication: a discussion of Goffman’s notion of ‘signs’ within participatory units of interaction. This study will intermittently connect this framework to audience research and the overall justifications of my proposed research. Community: Social Structure The concept of ‘social structures’ or ‘networks’ is a means of conceptualizing, modelling and analysing large social group formations. Different academic disciplines envision these social formations in various ways. This section will comparatively discuss the sociology of social networks and its recent migration onto new media platforms as “networked society” and “virtual communities.” Finally, anthropological ‘kinship structures’ will be discussed as a means of tempering the limitations of the former schools of thought. Sociologist Simmel defined social networks as a “web of group affiliations” (1922). Later Granovetter (1973), using the twin metrics of social similarity and frequency of time invested (Homan, 1950), refined Simmel’s amorphous “web” into weak and strong relations of exchange. Strong ties comprise the inner layer of an individual’s social network (e.g. close friends) and weak ties the outer one (e.g. acquaintances). The more ‘ties’ someone has the higher their ‘density of ties’ is. Through qualitative interview, Granovetter discovered that weak ties more often than strong ties provided important information that led to obtaining new jobs. Later, Dinidia and Canary (1993) discovered that tie relationships needed to be actively maintained through exchange in order for the relationship to survive. -5- MSc Dissertation Kenneth J. Gamage Granovetter’s study laid the groundwork for several computer-mediated social network analysis studies (e.g. Haythornthwaite, 2005; Wellman et al., 1996; Adamic et al., 2003; Golder et al., 2007 etc.) With these works two methodological perspectives have emerged: one that focuses on the social individual, and the other on the individual’s data traces. Smith (2008) and Adamic (2008) focus on exchanges of social hyperlinks and ‘hyperties’ between SNS users. Given that ‘Likes’ are hyperlinks that link two members’ pages, they are classified as ‘hyperties.’ Due to new media’s inherent data recording capability, a deluge of miniscule quantitative data exchanges are easily available to researchers. As Smith comments, “ties can be systematically researched, analysed, searched” (2008: 167). As new technologies emerge, large-scale quantitative studies such as “Life logging,” “Jaberwocky” and “Slam XR” have proliferated, giving rise to terms like “network society” (Castells, 1996). Large-scale social network analysis studies pose two issues. First, an over-emphasis on quantitative data conflates the virtual network for society itself as opposed to a tool used by people. We can surmise that this issue stems from the macro level scale of data that obscures the atomized person. Concomitantly this negates key emotive factors. Second, social network analysis data’s ability to crystalize every node and subtle movement of an individual’s SNS network obscures the true perceptual ‘fuzziness’ of how people may actually imagine their social networks. Social network analysis data is immensely useful. However, its pinpoint clarity might benefit if complimented by analysis of how networks are perceived by people (see Kumar et al., 2006; Hsu et al., 2007) at the micro-level vis-à-vis the imperfections of human memory and perception (see Radstone and Hodgkin, 2005; Halbwachs, 1992). This emphasis on the individual would alleviate issues posed by large-scale quantities studies. In contrast to the mathematical precision of data-driven social network analysis, anthropological studies of kinship provide necessary nuance with a focus on how individual’s perceive, or cognitively imagine, their community. Anthropologists such as Sahlins (1972) and Lévi-Strauss (1949) refer to these social groups as kinship structures. These include ‘fictive’, non-blood related kin. Kinship structures are described as concentric rings of “kinship distance” spreading from close family and friends in the centre and strangers in the outer periphery. Sahlins focused on the tension of dependence and independence between relations. Explicitly, this appears to be similar to social network analyses. Implicitly, however, kinship rings are described from the perspective of one individual and thus subject to the imperfections of cognitive perception. More precisely, kinship is a socially re/produced construct of an individual’s mind. Lévi-Strauss believed that, “[t]he social world is the realization of the categories of the brain” (Deliège, 2004). This means that the human -6- MSc Dissertation Kenneth J. Gamage categorization of social groups is perceived in and enacted by the mind as a system of symbols. This links structural anthropology to its origins in Saussure’s (1916/1983) structural linguistics (Deliège, 2004: 20-1). Lévi-Strauss’s notion of symbolic social worlds provides a sound basis for analysing how the meanings of ‘Likes’ are hinged to the people that give, receive and perceive them. Considering that Anderson’s (1983) “imagined community” was intended to address how people imagine the abstractness of whole nations that they are part of, it nonetheless applies to Facebook users insofar as their imaginaries of social networks on- and offline are shaped by this social media platform. SNS algorithms that dictate what content appear on the “newsfeed” shuffles members’ social networks. Outer layer relations that would normally recede in an individual’s mind might be placed in the mind’s forefront through updating “newsfeed” content. Concomitantly, members that share little content comprise what Boyd (2010) has defined as the “invisible audience.” Interestingly, Dunbar’s (1993) anthropological study strikingly discovered that primate neocortex size can “predict the cognitive group size for humans” (Dunbar and Hill, 2003). The same pattern exists in both tribal cultures and post-industrial societies (54). Whatever the context, human group size is about 150 people (2003: 54). This figure is a mere 20 people off from Facebook’s reported average ‘friend’ count of 130 people (Miller, 2007). These 130 ‘friends’ comprise one’s “imagined community” as a symbolic system constituted with posts, comments, and ‘Likes.’ Curiously, part of this “imagined community” is the likely forgotten “invisible audience” (Boyd, 2010). Dunbar and Hill’s (2003) survey based study on annual Christmas card exchanges discovered that individuals feel that some effort must be made in maintaining sufficient communicative exchange with “outer layers” of the social network. The annual ritual of card exchanges is the one time their sample made contact with their entire network (mean size 153.5). The reason for this, as identified by Kana’Iaupuni et al. (2005) lies in the probability of eventually receiving social support. Given all of this, it makes sense to assume that the meaning of exchanges of ‘Likes’ may be in supporting strong tie relationships and maintaining sufficient contact with weak ones. Community: Gift Exchange Theory Mauss’ classic (1954/1923) anthropological ethnographies of Polynesian and Melanesian tribal societies identified a common social thread of kinship groups: the norm of reciprocity in non-capitalist gift exchanges. Lévi-Strauss also saw “exchange” as the first fact of social life: “a society is first of all an exchange network” (Deliège, 2004). We have established that -7- MSc Dissertation Kenneth J. Gamage individuals are imagined symbolically. In this section, I will discuss how, through exchange, ‘Likes’’ meaning might be inextricably hinged to these symbols through the Maussian principle of reciprocal exchange. Baudrillard’s ‘logic of exchange’ draws a distinction between the logic of symbol value and sign value (1981: 65).1 The former will be discussed in this section and the latter in the section entitled “Communication: Social ‘Like’ Signs.” This provides the present study with a nuanced view of the meaning(s) of ‘Likes.’ Recalling the etymology of Carey’s notion of symbolic exchange (1989), Baudrillard’s assertion allows us to closely articulate – and thus scientifically mobilize – how both sign and symbolic values are ascribed to objects and intrinsically intertwined within social relations and their guiding social codes. One of these codes is attributed to Mauss’s norm of reciprocity in gift exchange. For Mauss, the three obligations of gifts are: (1) to give, (2) to receive, and (3) to give again (1923: 50). Gifts for Mauss are never free and are inalienable. Weiner (1992) defines this inalienability as the bond created by the receiver and donor through the process of cyclical gift reciprocation. The act of reciprocation is governed largely by temporality, where gifts are exchanged at different semi-calculated points in time. As Gregory explains, “[g]ift exchange is an exchange of inalienable things between persons who are in a state of reciprocal dependence” (1982: 82). It is “the production of relationship” (1982: 43). Mauss’s ‘maná’ and ‘hau’ therefore meant a type of haunting of the object by the donor that beckoned for its reciprocation. The central criticism of Mauss’s work was that the enigma of reciprocity was a ‘pseudo-religious’ magic contained within the object itself—a point that Lévi-Strauss’s structuralism, and Carey’s and Durkheim’s ritual models, locate within the social contract. Moreover, gift giving can be either antagonistic or non-antagonistic. Antagonistic gifts are based on competitiveness where rivals give gifts that are difficult to match as a show of wealth. This typically culminates into Mauss’s potlatch: conspicuous spending to humiliate others by placing them “in the shadow of his name” (Mauss, 1923: 50). Conversely, nonantagonistic gifts are exchanged between non-competitive strong tie relations e.g. best friends. Using Facebook practices as an example, pressing ‘Like’ on someone’s status update indicates a social relation between donor and recipient within a social network that is semi-/publically performed. This process can take on many arrangements varying from the simple one-to-one exchanges (i.e. ‘Like’-for-‘Like’) to more complex exchanges between social clusters. My -8- MSc Dissertation Kenneth J. Gamage assumption is that reciprocity is instrumentalized by the social need for social order, rather than the determinations of pseudo-religion, linguistic liking, or ‘Like’ technology. As Carey notes religious determination is the antecedent of technological determination. Reciprocation takes on many forms. For example, between close friends and family gift exchanges enter a state of generalized reciprocity (Sahlins, 1972; Baudrillard, 1981; Bataille, 1967). As such, we can conceptualize one-to-one reciprocity and generalized reciprocity as two points along a continuum. It is important to note that Baudrillard was adamantly against the application of reciprocity onto language. Language exchange was rather an innate interaction. Conceiving of it terms of reciprocal language is a trope. In contrast to Mauss’s reciprocity there are some other perspectives that should be noted. Opposing Mauss’s reciprocity, Derrida’s “Counterfeit Money” (1994) believed that a gift must not have the appearance of a gift to be returned. Adding complimentary nuance to Mauss’s theory is Weiner’s notion of ‘keeping-while-giving” (1992). To paraphrase Weiner, a gift’s value is enhanced by simultaneously withholding it from circulation. Although not this paper’s focus, it is important to note that gift exchange theory is often argued as the antecedent to capitalizing commodity exchanges, despite that its anthropological roots are seeded in Marxism (see Gregory, 1982). Communion: Ritual and Ceremony In media and communication literature the term ‘ritual’ is theorised primary in two coexistent ways around Carey’s two communication modes (2009). First, in the transmission mode, research on media rituals focuses on the uneven power relations that are maintained and produced through the heightened centrality of media (see Dayan and Katz, 1992; Couldry, 2002 etc.). Second, ‘rituals’ also refer to everyday interactions between humans that cultivate social integration. This trajectory stems from the sociology of Durkheim (1955) and informs Goffman’s (1959; 1971) model of communication interaction. Goffman argues that despite its religious connotation, ‘rituals’ are a vital part of social organization. Couldry (2002), Goffman (1971), and Carey (1975) contend that ritual conventions have proceeded from both secular and religious lineage. Rituals can be performed through the social conventions of non-verbal and verbal communication (LéviStrauss, 1966/1962; Weber, 1958). In this framing, the ‘Like’ button could be conceived as a ritual artefact (Douglas and Isherwood, 1979). Rook (1984), paraphrasing Meddin (1980), 1 Although not the focus of this study, Baudrillard’s logic of signification (1981: 65) also includes: “A logic of -9- MSc Dissertation Kenneth J. Gamage says “ritual organizes, then ‘fixates’ or ‘freezes’ symbols in cultural variable ways, and standardizes symbolic activity.” Goffman’s cyclical, human communication model therefore rivals Laswell’s linear (sender-receiver) mass media model. While Goffman’s micro-sociology might seem far-flung from audience research, it lends description to the complex exchanges of audience/performers (Silverstone, 1990; Neuman, 1991) using the ‘Like’ tool. Reason being that its constructivist model of communication far precedes transmission models and prior conceptions of passive audiences. It articulates Press’ “post-audience” as performer/viewer in mutual symbolic exchange. Building on Durkheim’s articulation of negative rituals (keeping distance) and positive rituals (paying homage), Goffman illustrates two key forms of ritual that frame the present study: (1) Rituals of Ratification: when someone changes their status in some way. Responses to these changes are reassurance displays and provide positive ‘supportive interchanges.’ (2) Remedial Interchange: Are intended to keep social distance. The interaction is scarce. If one is given, the receiver must show a sign that the symbol or gift has been received or is sufficient. It is a negative ritual. Accordingly, each ritual likely has different perceived values for people. Goffman distinguishes between everyday social-recognition rituals and ceremonies i.e. birthdays, weddings. Ceremonies typically require a more formal set of interaction signals and microrituals. Accordingly, these interchanges will serve as the interaction units for ‘Like’ exchanges. Communion: Homage of Attention To quote Simon, “A wealth of information creates a poverty of attention” (1971: 41). While this quote pre-dates CMC and SNS by several decades it still rings true. Kollock adds that the “signal to noise ratio, it is said, is bad and getting worse” (1999: 220). “Signal to noise” characterizes Hindman’s (2009) notion of the Internet as site of high input/low output that could relate to SNS-mediated interpersonal interactions. We can surmise that this “poverty of attention” also dictates how content is written and read on the Facebook SNS. Paying attention to someone is a key component in Goffman’s interaction unit. As he puts it, “messages primarily serv[e] to establish, to prolong, or to discontinue communication …” exchange value” and “A logic of use value” that link to Marx’s critique of capitalist consumption (Marx, 1867). - 10 - MSc Dissertation Kenneth J. Gamage (1966: 99). For instance, running into an old friend from out of town on the street might garner a mutually inflected form of attention. Strong tied friends may have a long-standing commitment to pay attention to each other every weekend or after work. Combining this conception of interpersonal attention with the Internet’s “poverty of attention” provides a foundation to advance a key assumption: ‘Likes’ that mark the abundant content that circulates on “newsfeed” may signal an active choice made in paying attention to certain members’ content over others. Communication: Social ‘Like’ Signs Recalling Baudrillard’s distinction between the logics of symbol and sign, this section will focus on analysing the latter in terms of Goffman’s interaction order (1959). The reason for selecting this particular semiological brand stems from Baudrillard’s (1981) criticism over Saussurian structural linguistics (1916) as disconnected from social worlds. Goffman’s constructivist micro-sociology re-locates sign interactions within the micro-ritual fabric of human communication (1959). It is a genre of social semiotics and is a means of receiving and giving attention. My assumption is that this is where meaning is re/produced in social practices that circulate between on- and offline worlds. In Goffman’s model, social actors perform and respond to each other's ‘signals,’ or ‘signvehicles’ (1959: 13) at the interpersonal level of interaction units. Chandler (2002) adds that performed interaction exchanges contain signs that include language, gestures, bodily, and facial expressions (135). Building on a notion put forth by Icheiser (1949), Goffman states that the two different ‘sign activities’ are: “the expression that he gives, and the expression that he gives off” (1959: 14). To illustrate, a person sees someone that they know and like passing by on the street. They smile and say, “hi, how are you?” The smile and verbal utterance are the expressions that they “give.” Despite that the expressions were intended to be genuine, for whatever reason, the expressions “given off” might be read in two entirely different ways. First, they could be interpreted as sarcastic or mocking. The pivot between these two types of communication is interpretive differences. Audience research (i.e. Hall, 1974; Morley, 1980) more accurately presents this as a continuum between ‘encoded’ preferred reading and ‘decoded’ interpreted meaning that fills the gap between text and reader. Despite its absurd cumbersomeness, ‘read-write’ on Facebook is more accurately read-interpret-write-interpret-read. This forms a continuum of understood and misunderstood signals. Second, the expressions “given” and “given off” may match each other. Here, the genuine greeting is interpreted as one. The recipient smiles back and says, “fine, thanks.” The ‘glue’ that holds this interaction together is social contract. Goffman (1971: - 11 - MSc Dissertation Kenneth J. Gamage 81) points out when someone says, “how are you?” you reciprocate with “fine, thanks” because of an implicit social contract. It is unlikely that an actor would respond back repeating the initial verbal signal of ‘how are you?” or more awkwardly with silence. One might not actually care how the other is doing, but the social ritual is more-or-less performed. For Goffman the ritual interaction unit “pertain[s] entirely to the management of co-presence” (1971: 19). Goffman also identifies ‘tie signs’ such as holding hands as signalling to third parties a type of relationship. Boyd in her studies of teenage friendship on SNSs has coined the textually mediated version of Goffman’s ‘tie signs’ as “public displays of connection” (Boyd, 2004). Taking together, it seems appropriate to approach textual interpretation of audience research (i.e. Hall, Morley) with Goffman’s interactionist model as a disciplinary bridge. Given that Facebook ‘Likes’ are an asynchronous ‘textual interaction’ that cultivate polysemic meaning in combination within other Facebook signs, we could surmise that they may mediate in terms of Goffman’s social interaction signals. These sign combinations may act as tropes of physical gestures, non-verbal cues, and language. Due to ‘Likes’ blunt constraints as a trope of offline spatial communication there are likely events of communicative misunderstanding. Yet, ‘Like’ may have some interpretive flexibility. This is the tension – constraints/affordances – that I focus on. When two Facebook social actors exchange comments and ‘Likes’ on the semi-/public “newsfeed,” interactions units view and perform to each other and simultaneously to the partly “invisible audience” of the wider social network. CONCEPTUAL FRAMEWORK Research Objectives It should be made clear that this study does not examine the capitalizing aspects of the ‘Like’ tool. It focuses solely on person-to-person social exchanges that occur within the site alone. Despite the broad theoretical framework, the objective is not to excavate every crevice of what has just been discussed, but rather, to explore how patterns of symbolic meaning is re/produced in the social practices of ‘Like’ tool use in everyday life (de Certeau, 1984). It is explorative, not conclusive. By combining audience research with the multi-disciplinary lenses of micro-sociology and anthropology, the present study locates itself in the on-going debate of technological and signifier determinisms versus human agency of using/shaping technologies and signifiers for social needs. It is my assumption that this wider debate is modulated by public perceptions of - 12 - MSc Dissertation Kenneth J. Gamage the function of Facebook’s ‘Like.’ This wider assumption is set against the micro-level notion that the technology of the ‘Like’ button and the simple signification of ‘liking’ do not determine ‘Likes’ true communicative uses – a notion steeped in negative connotations of audiences as passive media users. ‘Likes’ do not inherently hold or emit meaning. Despite its facilitation of new forms of communication, i.e. mass self-communication, durability, timespace compression etc., ‘Likes’ are rather located and constructed within social worlds through economies of symbolic exchange, motivated by the social psychological need of “maintaining society in time.” Accordingly, I propose the following two inter-related research questions and their related assumptions: (Q1) Research Question 1: What inter-subjective meanings does our sample of Facebook users convey and interpret using the ‘Like’ button tool? (A1) Assumption 1: ‘Likes’ may convey multiple meanings both linguistic and paralinguistic. In varying degrees they signal that attention has been paid as a form of social support. (Q2) Research Question 2: How, if at all, are the identified meanings socially constructed and ascribed with perceived value? (A2) Assumption 2: ‘Like’s’ perceived value and meanings are influenced by social codes, context and relationship types. More precisely these variables informed by the literature review might be: (A2.1) Social Tie (A2.2) Ritual Type (A2.3) Semi-/public Awareness (A2.4) Reciprocity (A2.5) Human Attention Research question 1 (Q1) falls under ‘communication,’ while research question 2 (Q2) falls under ‘communion’ and ‘community.’ It is therefore my over-arching assumption that ‘communion’ and ‘community’ are the explanatory variables, while communication is the response variable. Ultimately, a variable analysis based on salient patterns will be applied. The table below has been provided as a point of reference. It links theories, concepts, and the research questions. - 13 - MSc Dissertation Kenneth J. Gamage Table 1 – Research Questions as they relate to the theoretical framework Research Q2 (Explanatory) Research Q1 (Response) Communion: Social Interaction Rituals (Goffman, 1971; 1959) Communication: Community: Signals/Sign-Vehicles (Goffman, 1959; 1971) Gift Exchange & Symbol (Mauss, 1923; Baudrillard, 1981) Attention Economy (Simon, 1971; Kollock, 1999) Tie Relations (Granovetter, 1973) Imagined Social Networks (Lévi-Strauss; Boyd, 2010; Dunbar and Hill, 2003) Rationale My rationale in taking on this project is both personal and academic. To the former, this study stems from my personal curiosity in observing diverse patterns of ‘Like’ use behaviour within my social network on Facebook. This has led to ‘naïve’ or tacit theorisation of the mechanisms at work. This has now been advanced by my examination of select literature in the theoretical chapter. I would now like to empirically test these advancing assumptions. From the perspective of its potential academic contribution, this study could act as a step in supplementing two strands of research. First, within active audience research, a number of recent studies have focused on overt displays of ‘prosumer’ behaviour on Web 2.0 in the context of vlogs, tweets, wikis, podcasts etc., (see Jenkins, 2006; Bruns, 2008; Gillespie, 2007). Aside from select studies on extra-textual favoriting on YouTube (Burgess and Green, 2009), and emoticon research (Walther, 1992; Utz, 2000; Herring, 2010 etc.), the distinct uses and properties of ‘Like’ exchange have not been thoroughly examined. This study methodologically compliments important large scale technologically and textual determined hyper-tie studies (i.e. ‘Life Logging,’ ‘nTag systems’ and ‘SensCam’ etc., see Smith, 2008) that rely largely on quantitative data ‘traces’ about people, but offer little on social produced meanings about the data. While my study does not offer statistically generalizable evidence - 14 - MSc Dissertation Kenneth J. Gamage on Facebook’s ‘Like’ data, it does identify variable patterns that can be later extended by further research. RESEARCH DESIGN AND METHODOLOGY This section will discuss the three main strategic stages that were taken to implement the research design. Each stage will advance chronologically from (1) research strategy prior to fieldwork, (2) methodological choices during fieldwork, and (3) post-fieldwork coding and analysis. Research Strategy Using qualitative ethnography, this study aims to explore the range of ‘Like’ button practices enacted by fourteen London, England inhabitants. Here, the subjective experiences of the aggregate sample will be scrutinized for salient patterns. These patterns will both test the assumptions and respond to the inter-related research questions. “The objective,” as Gaskell contends, “is a fine-textured understanding of beliefs, attitudes, values and motivations in relation to the behaviours of people in particular social contexts” (2000: 39). It’s often stated that no research is value free and this project is no exemption. My ontological view of reality that guides this study’s strategy stems from symbolic interactionism and constructivism (Gilbert, 2000: 138). Accordingly, the aim is not in locating objective truths, but rather socially produced meanings about and in relation to ‘Likes.’ My assumption is that these meanings are held inter-subjectively thus lending to pattern analysis across the sample. Given this, a qualitative approach seems appropriate. This study employs a mixed method approach by ‘complimenting’ (Green et al., 1989) two different qualitative methods: semi-structured interview that is directly followed by a thinking-aloud activity. Prior to fieldwork a baseline survey was employed to inform sampling decisions. A post-interview follow-up diary was provided as an optional activity, but yielded no responses and was excluded from the study. Semi-structured in-depth interviews have advantages and limitations. Due to its conversational, open-ended format, guided by a pre-determined topic guide, the participant is given space for their own thoughts to be heard with limited restriction. This approach facilitates personalized questioning for each participant and the exploration of new avenues of thought as they emerge. Additionally, this method captures non-verbal expressions (i.e. - 15 - MSc Dissertation Kenneth J. Gamage tonal infliction, glances etc.) that enrich verbal data. There are also limitations to semistructured interviews. Flexibility can lead to interviews meandering into tangential territory (which did occur), prompting me to bring the participant back to the task at hand. Additionally, as identified in a pilot study that I conducted on April 2012, participants and myself, as the interviewer, intermittently shifted from private to public selves. Bailey (1987: 177) refers to this as skewing the authenticity of the data. Nevertheless, his point begs the question, what is actually ‘authentic’ data? What I learned from my pilot study is that public/private is a continuum of social construction. Therefore shifting into public selves still produces vital data for my aims. The advantages and limitations of my complimentary thinking-aloud method are more-orless similar to what has just been discussed. Due to its unconventionality, a definition is in order. “[T]hink-aloud data requires the research participant to continually speak aloud the thoughts in their head as they work,” says Young (2005). Ericcson and Simon (1993) call this a “cognitive reflection.” In our case, participants will ‘read,’ and thus interpret, the ‘Likes’ that are recorded on their Facebook profiles. This is a retroactive process of positioning the participant as an audience to their Facebook ‘performance.’ Given that the decisions behind ‘Liking’ are assumed to be emotive and ephemerally driven, my reason for choosing this method is that it may enable a verbalized production of the meanings around ‘Liking’ practices. In combination with the interview, data is collected from two analytic levels of the same phenomenon: the apparent level of formed opinion and the latent level of emotion. Thinking-aloud however should be used sparingly due to its lack of evidence in its link between cognitive patterns and verbal expression (Wilson, 1994). Accordingly, I’ve limited it to 10 minutes. Because my goal is to capture the nuanced processes of how the participants uniquely construct their responses within their specific social contexts, and if any of these responses form patterns across the sample unit, several methods of data collection were excluded. For instance, surveys were only used for sampling. Surveys could detrimentally mirror my personal assumption by potentially leading participants towards pre-determined answers. They may also constrain emotion and surprise that require fine-grained probing. Content and semiotic analysis were also not chosen as they over-determine the Facebook-text, its technology, and my sole interpretation, contradicting this study’s aims. - 16 - MSc Dissertation Kenneth J. Gamage Pre-Fieldwork: Sampling The ethnographic sample frame consists of fourteen participants aged 18-64 that live in London, England. The sample was comprised of expats and native Londoners. Participants were selected through ‘purposive sampling.’ (Mason, 2002a). This means that participant selection was based on a criteria that developed and tested my argument (Glaser and Strauss, 1967). Because my chief aim was to maximize the topological range of socially produced meanings in and around the ‘Like’ tool, I sought to capture a spectrum of experience (see Appendix A). This included those that seldom use the tool to those that use it frequently. Maintaining this balance critically facilitated a fair testing of my assumptions through potential contradiction (Denzin, 1989). Moreover, this also provides space for the unanticipated, which ultimately created new inductive codes. The selection process involved several stages. First, posting an invitation to participate through Facebook’s email function to my personal Facebook network. This ensured that candidates were both Facebook users and possessed sufficient SNS literacy. The seventeen London residents that responded were emailed a hyperlink to a baseline survey. The survey was created using the web application ‘SurveyMonkey’ (see Appendix B). Questions were formulated from my own imagination combined with the questioning structure of Pew Internet & American Life Project’s study on SNS use (Hampton et al., 2011). The purpose of the survey was to identify suitable candidates that in aggregate possessed a spectrum of characteristics based on my pre-established criteria. Seven participants were selected. At issue is the bias of my personal network that could constrain variety. Locating the remaining seven participants through snowballing offset this. Here, I sought to maximize social distance away from me personally. Sampling ceased once a saturation point was reached and few new inductive codes were revealed. My sampling falls short in its aims for variety as very high frequency ‘Like’ tool users were not locatable (see Appendix A). This participant trait is unfortunately excluded from the research. Moreover, the study is limited to participants of a relatively privileged socioeconomic status. Pre-Fieldwork: Sampling The interview followed a pre-established topic guide (see Appendix C). The questions were engineered to prompt participants to reflexively analyse the meanings and strategies of their ‘Liking’ practices. In verbalizing these fleeting, tiny exchanges I contour the fabric of social production that the ‘Like’ rests within. Because this study takes a theory-driven approach, - 17 - MSc Dissertation Kenneth J. Gamage questions were built around five specific deductive codes (see Table 3). Each of these codes represents a broad concept from our theoretical framework. An example of the question-tocode link is provided below: Table 2 – Sample questions in relation to concept code Concept Code Questions (1) What is your reaction when you post something on a Obligations & Expectations friend’s wall that you thought they might enjoy and they ‘Like’ a month later? (1a.) What if they ‘Like’ it right away? Table 2 also exemplifies my strategy of employing the questioning tactic of what Huberman and Miles call “making if-then tests” (1994: 151). I used the above question to prompt the participant to consider the ‘obligations and expectations’ of the same scenario temporally contrasted: one month versus instantly. This tests the inter-relatedness of ‘temporality’ and ‘obligations and expectations.’ The ultimate variable analysis built into research question 2 of this study allowed me, question-by-question, in a conversational fashion, to test one variable against another. This tactic was used sparingly, but its use increased as it proved useful. Initial questions were asked in a fun and creative manner using free association in order to ‘break the ice’ (i.e. ‘Using your imagination, choose a real life object physical object that the ‘Like’ button might resemble?’) This prompted the participant to look at the mundane tool with ‘fresh eyes.’ As I later realized, a by-product of this was the cultivation of trust: by accepting any response that was given – no matter how strange – without passing judgement. It is worth noting that the topic guide and questions changed after the first four interviews as some questions proved to be ineffective and others more useful. Incidentally, I think my questions around obligations and performativity on Facebook were far too blunt (Do you feel any obligation … ?) This constrained a genuine response in favour of a publicly ‘acceptable’ response. - 18 - MSc Dissertation Kenneth J. Gamage Fieldwork: Interview and Think-Aloud Interviews were conducted from June 30 – July 16, 2012 in various locations in London. Locations ranged from quiet areas in participant’s offices, to homes and coffee shops appropriate for audio recording. The main requirement was a Wi-Fi connection to access the participant’s Facebook page using my personal laptop. Interviews began with light conversation to establish rapport. Often, I would divulge some mundane personal information about my day to hopefully reciprocate personal sharing in the interview. I then explained how the session would proceed, how their data would be used, followed by a signing of the virtual consent form (see Appendix D). Screen and audio recording was executed using the software ‘IShowU’ while simultaneously recorded on a smartphone application as a back up. Interviews were then conducted using the topic guide (see Appendix C) followed by the thinking-aloud activity. During the latter method, I kept my input to a minimum and allowed the participant to talk freely about their activities. My typical intervention was usually “what do you think the meaning of this is?” or “keep talking” (Young, 1997). A small portion of my interviews unhinged deep-seated emotional reactions (e.g. ‘Likes’ from old enemies, insecurities etc.) Conversely, some participants had a difficult time verbalizing their habits. On both accounts, this provided vital data, but equally challenged my skills as a novice interviewer. All highly emotive responses were documented on memos that were filed with each transcript. Eventually an emotive scale was applied during coding. Post-Fieldwork: Coding After interviews were completed, they were transcribed and coded using thematic and variable analysis. This was executed using the software ‘TAMSAnalyzer.’ As illustrated below, concept codes were generated from the theoretical framework. These were simplified into broad yet discrete and easy to apply ‘indexing categories,’ or themes (Mason, 2002b). These codes are listed below: - 19 - MSc Dissertation Kenneth J. Gamage Table 3: Concept code in relation to indexing categories Concept Code Indexing Categories Social Network & Symbol Social Tie Audience/Performer Interaction Mode Ritual Social Situation Obligations & Expectations Social Rule Human Attention Attention Sign Social Meaning Exchanged Using ‘TAMSAnalyzer,’ indexing codes were located and marked within each transcript; for example: [Social Situation] ‘She ‘Liked’ my graduation photo’ [/Social Situation]. Here, the inductively derived code is ‘graduation.’ A summary of these inductive codes is displayed on Appendices G - I. The decision to index code a segment of the transcript was based on it matching a set of pre-established indicators (Boyatzis, 1998: 104). Indicator lists were kept nearby on cue cards. Once all fourteen transcripts were coded, and the codes were reviewed and rewritten to more accurately fit the data (e.g. ‘event’ become ‘social situation’). This was followed by a subsequent analysis where indexing categories were located holistically across the raw information. For instance, codes like ‘social rule’ do not always appear in segments but can also come across the entire narrative of the interview as the participant forms their opinion out loud. Reflexively, recounting these steps comes across far more neatly than what occurred. Admittedly, some bias may have occurred in the difficulty of maintaining consistency in coding and focal lengths of text. Moreover, certain complex phenomenon (e.g. reciprocity [A2.4]) occurs intermittently across a lifetime and is not captured within a single transcript. Transcripts are ‘snapshots’ of a re/produced moment, but not representations of entire lives. A longitudinal methodology might have offset this. Finally strong emotive responses (i.e. surprise, anger) were also memoed. These were matched with my own recorded observations of physical behaviour (largely in role playing). During analysis codes were cast onto an emotive scale from 1-3 (Huberman and Miles, 1994). Because of the nature of research question 2, some further preparation was necessary to complete a “variable analysis.” Here, I used what Huberman and Miles term as “pattern coding” (1994: 69). This required two-stages. The first stage identified the salient themes of - 20 - MSc Dissertation Kenneth J. Gamage the inductive codes. Using ‘TAMSAnalyzer’ these inductive codes were counted and placed in a hierarchy of emotive response. Indexing codes that did not appear to occur in more than five out of the fourteen participants were excluded. In the second stage, variable analysis was used to recognize the patterning of inductive codes that occurred across the sample. Explanatory variables were separated from response variables. Variable analysis involved interpreting any salient inter-relationships of each inductive variable. These patterns were placed into two matrices of findings (next section). My application of codes falls short only insofar as they were not verified by another researcher for inter-coder reliability due to my lack of resources. RESULTS AND INTERPRETATION In this section we advance from coding to an analysis of salient variable patterns that influence the perceived value and meanings of ‘Likes.’ These variable patterns act as the explanatory variables that respond to research question 2 and the thematic meanings conveyed that respond to research question 1 (see Appendix H &I for full variable matrix). Each finding sub-section will be linked to a section of this table as a reference point identified by row and column number. This chapter will conclude with a reflexive discussion reflecting on the study as a whole. Generally speaking the assumption variables of ‘social tie’ (A2.1), ‘ritual type’ (A2.2), ‘semi/public awareness’ (A2.3), ‘reciprocity’‘(A2.4), and ‘human attention’ (A2.5) within the sampling frame, all inter-correlated to influence unique meanings of ‘Likes’. This incidentally supported A1 of ‘Likes’ being used for multiple meanings. What was unanticipated was the identification of new inductive codes voiced by the participants. The most prominent are as follows: • Keeping ‘Likes’ scarce is a secondary obligation to Maussian reciprocity • ‘Likes’ signal retroactive attention paid to past posts • ‘Likes’ signal the memory of others • Intentional antagonistic gifts from weak ties From a methodological perspective an unanticipated finding was also discovered: • Bodily and facial expressions that ‘Likes’ metaphorically signal are identifiable as participants physically act them out through retroactive interpretation. - 21 - MSc Dissertation Kenneth J. Gamage The sub-sections that follow will contain an interpretation of key quotations through the lens of the theoretical framework. Quotations were chosen based on their patterned similarity of content and emotive response to all other qualified quotes within their respective categories. Quotes that both contradict and support my assumptions were given balanced evaluation. Each categorical theme is based on frequency count, emotive scale across the sample and my subjective evaluation. One cautionary note is in order. As this an analysis of micro-rituals, all emotions and meanings presented here are ephemeral, tiny, and located within wider contextual narratives, symbolic super-systems, and the dynamism of human lives. Social Codes of ‘Liking’ In response to Q2, meanings and the variables that influence the perceived value of ‘Likes’ in the sample appear to be guided by the following two social strategies: (1) Maintaining Scarcity (2) The aforementioned strategy combines with ‘social tie’, ‘ritual type,’ ‘reciprocity,’ ‘interaction mode,’ and ‘attention’ to cultivate a perceived value of ‘Likes.’ What was not expected was that scarcity/abundance of ‘Likes’ seems to predict if a ‘Like’ is a gift or para-/language. Maintaining Scarce ‘Likes’ (Appendix H & I, Column B) Although the initial assumption of the primary social code of ‘’s reciprocation (A2.4) of ‘Like’ exchanges was evidenced within the sample (twelve of fourteen), the unexpected social code of maintaining a scarcity of ‘Likes’ (nine of fourteen) was also prominent. Although limited attention was paid to Weiner’s theory of ‘keeping-while-giving’ (1992) in the theoretical chapter, it best explains this social phenomenon. Participant conformity to the social norm links to Goffman’s (1971) and Carey’s assertion that ritualized norms contribute to social organization, even those virtually networked. Conforming to the code seems to imbue the signal “given off” (Goffman, 1959: 14) by the ‘Like’ with a higher perceived value (A1). For instance, as D contends “I just don’t give out my ‘Likes’ to anyone.” We can surmise that D’s social network has an imagined sense of this from her retroactive habits. Therefore piercing through the “signal to noise ratio” (Kollock, 1999) to pay attention to another member’s post may reciprocate attention to her unique ‘Like’. - 22 - MSc Dissertation Kenneth J. Gamage This tells us that a socially acceptable mean level of ‘Liking’ must be maintained. T mockingly calls excessive ‘Like’ clickers as “Like Happy” while others labelled the behaviour as being “a bit off” (Z),“insecure” (M), or “stalkery” (G). A states the simplicity of the rule: “you can only do it once in a while.” As Weiner contends, “in an economy based on gift-exchanges it is necessary to withhold from the giving process …” (1993: 33). Drawing from de Lauwe’s study on maintaining a mean level meat consum[ed] (1956), Baudrillard (1981: 70) points out that “one’s needs are to not over- or under- consume.” Comparing meat to Facebook ‘Likes’ might seem absurd. Yet the similarity of mean semi-/public consumption suggests that this pattern occurs even when the object exchanged has no material value and is infinitely replicable. Considered in relation to the initial assumption (A1), the ‘Like’s’ value is within this sample frame is likely perceived and socially re/produced through an inter-subjective sense of one’s social network members’ abundance/scarcity ratio narrowcasted on the “newsfeed.” Its value is measured by holding a tacit calculation of ‘Likes’ given compared to what is kept. ‘Likes’ that are kept scarce are more likely to attain the stature of ‘gift,’ as will be discussed in the next section. Conversely, some participants counter-strategized by managing the signals “given off” by the ‘Like’: I go through friends’ stuff just to see what they’re up to and stuff, but I think I'm conscious to not sort of you know um, I might see a photo of a friend doing whatever that is like 2 or 3 years old and I won’t say anything or ‘Like’ it because they'll think that I'm like a creepy stalker (G). Here, G represents Boyd’s “invisible audience” (2010) that others in the social network may forget are listening due to their strategic restriction of ‘Like’ signals. Maintaining invisibility is predicated on keeping entry into visible “participation units” at a minimum in favour of abundantly spending time in viewing the “newsfeed” text. One must strategically act as if one is not listening or risk implicating oneself as a “creepy stalker.” Signalling that one is viewing others content in excess by ‘Liking’ leads to a breach of the social code. ‘Likes’ as Gifts and Para-/Language (Appendix H & I, Column F) In response to Q1, ‘Like’s’ meanings fall into two categories of exchange: gifts and language as patterned across 8 of 12 participant interviews. Here, the aforementioned ‘keeping-whilegiving’ social code appears to combine with all of the non-probability variables to cultivate a perceived value of ‘Likes’ that exists on the continuum of para-/language and gift giving. - 23 - MSc Dissertation Kenneth J. Gamage ‘Likes’ that are kept scarce are more likely to be perceived as gifts, while ‘Likes’ that are abundantly exchanged in mutual and generalized reciprocation (Sahlins, 1972) are perceived to have a conversational banter that is both non-verbal (e.g. smiles, glances) or linguistic. The latter case resembled emoticons (e.g. LOL, smileys etc., see Utz, 2000; Walther, 1992). This interpreted continuum is the key finding of research question 1. For the most part this polysemic quality of meaning matches the initial assumptions (A1) For instance, as J says, “[m]aybe in the beginning it was literal but you use it to show something in different ways.” This notion, however, was extended by unanticipated inductive codes. The summary table below illustrates the response variable’s continuum of value and meaning that was ascribed to ‘Likes.’ The full list of inductive variables appears on Appendix G. Table 4 – The language and gift continuum ‘Like’-as-Language  ‘Like’-as-Gift Non-verbal communication (nods, Well wishes during monumental life events glances etc.) banter (i.e. births, weddings) * ‘I agree with you’ Affirming past shared memory * Minor emotional affirmation Being noticed by old friends * Conversation starter Antagonism or mockery from enemies * ‘I’ve seen it’ ‘I enjoyed this’ Conversational intrusion from acquaintances and enemies * ‘Let’s talk about this post later in person’ On-going social support from close friends & family Marking content as method of browsing During monumental life events there appeared to be a firm need to reciprocate. There is also moderate need to reciprocate in mundane dialogue with strong ties. This, however, is bearing - 24 - MSc Dissertation Kenneth J. Gamage on the fact that ‘Likes’ make up a lone and slight tool of people’s repertoire at certain times language (i.e. ‘thank you,’ and ‘I read that’), and others paralanguage (i.e. glances, nods etc.) This data was captured in response to the question: 'How would you describe ‘Liking’ to someone who had never come across it before?’ and ‘Can you think of anything that people do in real life that resembles “Liking”?’ Interaction Rituals This section will discuss the salient patterns of different scenarios where ‘Likes’ take on unique meaning, cultivated by the strategies just discussed. These core findings with variables in various combinations (see Appendix G & H) inform the following key patterns of symbolic exchange that create meaning and values along the extreme poles of the gift – para/language continuum: (1) ‘Likes’ as ceremonial gifts in major life events. (2) Antagonistic ‘Likes’ gifts from weak ties. (3) Para-/language ‘Likes’ between weak and intermediate tie relations are conversational intrusions or conversation starters. (4) Para-/language ‘Likes’ between strong tie relations as everyday conversational banter. ‘Likes’ as Ceremonial Gifts (Appendix H & I, Row 1) In the event of a ceremonial life event (A2.2) that is performed typically through a status update, ‘Likes’ seemed to have a higher value. This was evidence across the sample. The dominant variable here falls under Goffman’s “ceremony of ratification” (1971: 17). Here, one’s social status has advanced. This is evidenced in B’s account of receiving an abundance of ‘Likes’ upon posting her new job as a high school teacher on the “newsfeed” that she describes as “like an online virtual party, like a celebration party.” Using Goffman’s theory, these tiny reassurance displays means that although B has advanced in her social position those who have given ‘Likes’ will maintain their relationship with her (A1). From the sample it appears that strong ties (A2.1) do have some obligation to press ‘Like’ (i.e. J: ‘you saw what I posted??' check it!), but a phone call, a text or in-person interaction are all options. For weak ties (A2.1) there doesn’t appear to be an obligation (A2.4). Posting a social advancement on the “newsfeed” is essentially a solicitation of Goffman’s reassurance displays. Empirically evidenced, reassuring ‘Likes’ given meant a showering of “congratulations … yeah way to go!” (M), “clapping” (H) or a ‘high-five’ (V). In this instance, - 25 - MSc Dissertation Kenneth J. Gamage ‘Likes’ carried similar meanings to “flowers” (J), “cards’ (Z),“gifts” (B), and giving “jewellery” (S) or the celebratory collision of pints of beer during a toast (P). Eight of the fourteen participants confessed that receiving abundant ‘Likes’ ‘given off’ emotively swept them into a feeling of popularity. In their words, “getting a buzz,” (G) “SUCCESS!” (F), “an ego boost” (B) or a ‘high-five’ (V). These emotive meanings revealed themselves not just in their words but also in their gleeful smiles they unintentionally acted out in their descriptions. Five participants went so far as to metaphorically link it to a game show buzzer: ‘Buzzzzz’ ‘Ding!’ This implies the semi-/publicness (A2.3) of the ceremony as they refer to their content through the metaphor of mass media television. Or further, the “newsfeed” narrowcasting of the abundant ‘Likes’ and recipients’ increased popularity itself – a phenomenon where abundant ceremonial ‘Likes’ cultivates an inter-related virtual social advancement. The obligation of reciprocity (A2.4) seems to heighten in the subsequent interaction order of the thank you ritual (A2.2) that follows the ceremonial ritual. Here there appears to be a moderate social obligation for the receivers of ‘Like’ gifts to reciprocate a thank you signal (A1). Thank-you ‘Likes’ are often used here along with comments, phone calls, texts, and faceto-face interactions. In some cases, non-reciprocation garnered a negatively emotioned reaction: It was somebody's birthday and I just posted something funny up on his wall. It was just kind of a 'happy birthday, hope this makes you laugh' kind of thing … I was just kind of surprised that he didn’t ‘Like’ it. And I did feel like deleting it. I felt a bit embarrassed cause it is like telling a joke and nobody laughs and you kind of want to erase that moment … I think yeah in that instance I did feel a bit sheepish. And a bit [pauses] hurt is too strong a word [pauses] a bit disappointed, a bit deflated by it. I wanted a bigger reaction (H). H’s disappointment of the recipient’s disregard for their ‘gift-debt’ could be explained by Gregory’s comment that “what a gift transactor desires is the personal relationships that the exchanges of gifts creates, and not the things themselves” (1982: 19). This is exacerbated by both the gift’s durability (“you kind of want to erase it”) and the “newsfeed” narrowcasting of the ‘gift debt.’ The un-‘Liked’ post becomes a source of mediated embarrassment (A2.3). Under a microscope, this “disappointment” disrupts the alliance between H and her gift’s recipient. Stepping away from the microscope’s lens, we can surmise that these small alliances form wider patterns of human solidarity. Reassurance displays - 26 - MSc Dissertation Kenneth J. Gamage fortify solidarity regardless if they are signalled by ‘Likes’, a thank you card or even a hug from a friend. These “things” are used to fulfil primordial social needs. Antagonistic Like-Gifts (Appendix H & I, Row 5) Another pattern observed in the sample (eight of the fourteen participants) was that ‘Likes’ took on slightly sinister meaning (A1) as an antagonistic gift. Applying the appropriate variables, this mainly occurs when they are from weak ties (A2.1) and not given in a timely fashion (A2.4) after a virtual ‘ceremony’ (A2.2) has occurred. As observed, antagonistic gifts are sometimes unintentional due to possibly two things: first, the ‘Like’-donor misinterpreting the symbolic relationship distance; and second, the ‘Like’receiver misinterpreting the intended meaning of the signal/sign. As J reports: “it’s ambiguous … there’s some space to engage.” Simply put, a gap exists between intended meaning and interpreted meaning of the ‘Like’. This was the reported prime constraint of ‘Like’ communication. Consider P’s account: If I have some random person kind of coming around and deciding well I’m going to ‘Like’ that but I don’t really have on going connection with someone on a deep level it’s like 'what are they doing poking around in here??' That's great you ‘Like’ my frickin' wedding photos, did you even fucking know that I got married? This quote infers an intended supportive remedial interchange of P’s social advancement in marriage likely meant to signal ‘congratulations.’ It’s akin to Hill & Dunbar’s (2003) annual Christmas card exchange and relates to Goffman’s ‘remedial interchange’ (1971) in maintaining sufficient distance and contact in offering bare minimum social support. Unfortunately it has been interpreted as “spying” or paying retroactive attention to P’s life narrative (A2.5). We can surmise that this would not be the case if the donor gave the ‘Like’ during the initial posting as opposed to months later. Given this, either the donor has misinterpreted the relationship distance (their imagined social network holds P within an inner layer [Sahlins, 1972]) or they aren’t aware of the social code of timely gift giving (A2.4). Moreover, timely gift giving occurs on the semi-/public “newsfeed” and not on private “profiles.” P’s biting reaction perhaps stems from a collision of her “imagined social network” that she frequently exchanges with and the invisible audience (Boyd, 2010) that are the ‘fuzzy’ gaps in her symbolic mental conception. The unintended ‘Like’ gift is imbued with a - 27 - MSc Dissertation Kenneth J. Gamage foreign symbolic memory of a person that she seldom thinks about. D takes this one step further, in her anticipation of weak tie gifts altogether: I kind of want to tell just my friends that for example we’re expecting [a baby]. I really don’t want some guy that I worked with one day five years ago – who gives a shit what he thinks? And would I say anything to him in the first place. I wouldn’t if I saw him on the street. D resists posting ceremonial events altogether in her self-consciousness of perpetually being monitored by the “invisible audience” and intolerance for weak tie ‘Like’-gifts – a reoccurring theme throughout D’s interview. The key difference between D and P is that D mentally includes the “invisible audience” of weak ties in her imagined social network, while P may not, despite that they share in their intended audience. It is worth noting that in one particular instance an intended antagonistic ‘Like’ was given to an unflattering photo from a rivalled acquaintance that kept their ‘Likes’ scarce. As J explained, “a friend from school, not a close friend that I haven’t talked to in ten years. She put ‘Like’ in a photo where … I was looking bad and ridiculous [laughs] … and this person put this ‘Like’ and I was like ‘Bitch!” This was not represented across the sample (as most did not include rivals or enemies in their social networks); however, it is worth noting as it offers a converse view of the aforementioned increased value of ‘Likes.’ From the lens of Baudrillard’s sign, once this ‘Like’ enters Facebook’s textual space it presents an issue of interpretation at the textual level that mediates actual human interaction. Consider that when J tried to interpret what the donor of the ‘Like’ may have meant she acted out an interpretation of its meaning with facial and bodily expressions. As mentioned ‘Likes’ mediate these expressions metaphorically. Obviously, isolating just a ‘Like’ cannot be interpreted, but within its contextual sphere (i.e. against the photo, the nature of the relationship, the timeliness of the exchange etc.) meaning is interpreted. As Baudrillard notes, signs are interpreted through their differences with these others sign (1981: 75). In P’s case, the donor of the ‘Like’ may have had good intentions that attempted to pay homage in social support, but was misinterpreted as antagonistic lurking. These ‘Like’ gifts from weak ties are aimed at conversational starters. ‘Likes’ as Para-/language: Weak and Intermediate Ties (Appendix H & I, Row 4) Gifts from weak ties can also lead to brief conversations. These are either well received or interpreted as conversational intrusions. To the latter, conversational intrusions occur when - 28 - MSc Dissertation Kenneth J. Gamage participation units (Goffman, 1971) of two typically strongly tied Facebook members are engaged in conversational banter that is narrowcasted on the “newsfeed” and members of the outer ‘invisible audience’ mark themselves into the interaction space. Consider the following instance: I do think that it is just a bit like talking in a group and suddenly realizing that other people are listening to you [A2.3]. It’s not that it’s a particularly private conversation. You just didn't know that they were listening and that thing on Facebook everyone is always listening. Even today when I got here an hour early to enjoy a drink in the sun and my boyfriend responded, which I would expect, and then a GUY that I used to work with that I haven't seen for a couple of years pressed ‘Like’. And then my OLD BOSS pressed it and 'it’s not the people that I would expect to be interested … And it's just like 'oh okay' I wasn't REALLY talking to you but 'oh okay' I will listen to your input (N). Unlike our early discussion of not appearing to pay an over abundance of attention to weak tie relations through strategically restricting signalling ‘Likes’, the above scenario illustrates the converse breaching of this social code (A2.4). Again misinterpretation could likely be the culprit. The intended signal given by N’s “old boss” and the “guy that [she] used to work with” are different than N’s interpretation of the ‘Likes’ given-off. Both weak ties may have desired to start a conversation or state solidarity in being able to relate to the posting on some level that cannot be pinpointed. This all of course depends on the nuance of each weak tie relation. If a weak tie relation is an old friend that one has lost touch with (Haythornthwaite’s ‘intermediate tie’ [2005]), these intrusions are received more positively as conversational starters. For instance as Z describes: So I guess these are people that you don't see on a day-to-day basis and they are sort of acquaintances. It’s like if you went back home from London and you saw somebody on the street that you knew. Not like your best friend, but you know someone that you haven't seen in a couple of years you don't stop everything you're doing go out for lunch, go for dinner, go to a movie and then wake up three weeks later. No, you say ‘hi, how are thing’s’ and then you go off on your way. There’s that little interaction that’s normal. - 29 - MSc Dissertation Kenneth J. Gamage This evidences a remedial interchange—an exchange that is sufficient enough to maintain distance, but cordial enough to maintain each person’s reputation. As observed in the sample, those with high weak tie relations and an abundance of ‘friends’, in the outer regions of the imagined social network, are far beyond Dunbar’s (1993) prediction of neo-cortex size to social group size and thus increasingly ‘fuzzy.’ A minimal conversation was exchanged in the form of short comments, some ‘Liking’ etc. In these instances reciprocity of ‘Likes’ may momentarily heighten, but are short lived. This was a moderately observed pattern that influences this gift to conversation shift. The initial ‘Like-gift’ produced through the established scarcity rule is rare. The ‘Like-gift' inalienably symbolizes the old friend and the need to update one’s memory of them. This ensures taking stock of any basic social advancement (marriage, job etc.) and enacts the requirement to sufficiently amend the decaying social bond within minimal effort. This “minimum” is implied in Z’s use of the phrase “little interaction.” An unanticipated intervening variable here was the apparent mild obligation (A2.4) to ‘catch-up’ due to their retroactive paying attention to the recipient’s narrowcasted content. As B describes: Its just like I guess you think they aren't like noticing what’s going on with you, but they actually are. It’s a way of connecting too. It’s kind of nice if people you haven't talked to in a while comments or ‘Likes’ something. You get a little connection. Observing across B’s interview and many others, maintaining these remedial interchanges was made easy and convenient with the ‘Like’ button. As B continues: It’s the CONVENIENCE of it. You don't have to actually REACH out and CONNECT on a super personal level but that little connection you know … it’s like spreading the love, why not? It takes one second and it makes people feel good. Obviously “spreading the love” becomes thinner as one moves towards the outer peripheries of the social network, but the patterned benefit of ‘Likes’ in remedial interchanges is their obvious ease compared to the improbability of running into someone on the street to ask “hi, how are things?” Here, Hill and Dunbar’s (2003) annual Christmas card exchanges with one’s entire social network feels so much easier with ‘Likes’ where “you don’t have to actually reach out and connect” (B). - 30 - MSc Dissertation Kenneth J. Gamage ‘Likes’ as Para-/language: Strong Ties (Appendix H & I, Row 3) Between strong ties in everyday dialogue, ‘Like’ exchanges supplement offline exchanges, circulating in a wider framework of mediated and non-mediated exchange (full inductive code list on Appendix G). Unlike ‘Like’-gifts, ‘Like’-language, are given in uncalculated abundance and not kept to produce perceptual scarcity. They also do not appear to express many emotive surprises. ‘Like’-language interactions appeared to be expected, mundane yet refined in their diverse articulations. Practices between strong tie relations re/produce unique inter-subjective meanings (A1). In this abundant form, ‘Likes’ more closely resemble text message emoticons (e.g. smileys, LOL, OMG) In a recent meta-analysis of mediated emoticon research, Herring and Dresner (2010) discovered that emoticons are not used by people to signify emotion, but rather serve a linguistic function. As observed within this study, ‘Likes’ are used diversely to signify emotion, memory, language, and paralanguage (A1). In its later state participants reported that ‘Likes’ metaphorically mediate tiny bodily expressions/signals such as a ‘wink’ (J), ‘a glance’ (S) or an approving nod. As H comments ‘Likes’ are: … An electronic indication of agreeing on something … it feels like a nod [participant nods head], like agreeing … when you're agreeing with someone you almost don't need to say ‘yes I agree’ because you've just taken it forward by saying 'oh this and this and this' so you're kind of agreeing with them inherently by nodding. ‘Like’-language obligations (A2.4) here are identical to other asynchronous modes of conversation. For instance as V and S comment: You don't say anything for two hours and then you might say something along the same lines and it’s a completely unconnected conversation that just kind of happens throughout the day with random people. And the ‘Like’ is just an extension of that (V). An expectation from my close circle of friends that I see on a regular basis that I kind expect to get a ‘Like’ or make some sort of comment (N). Some typical meanings of para/language ‘Likes’ are listed on Appendix G. Without overlabouring each meaning, we can see select meaning genres: humour, affection, confirmation of receipt or being read, paying attention, banter, teasing etc. In many cases the - 31 - MSc Dissertation Kenneth J. Gamage instant appearance of a ‘Like’ signals that a close friend is online. This sometimes instigates a communication medium change to synchronous chatting (e.g. IM). These findings were expected and in line with the core assumption of multiple meanings (A1) beyond merely liking something. Discussion: Reflecting on the Study As the interpretation and findings might suggest, future research is necessary to alleviate some of the shortcomings of this study. Some of these issues will be discussed here in greater detail than presented in the methodology chapter. Due to the emphasis on variable analysis key areas such as social capital and identity politics were responded to only tenuously. For instance, further research that extends Goffman’s rituals of ratification vis-à-vis social advancement in hindsight would benefit by integrating a social capital framing. While variable analysis is necessary for an initial exploration, it tends to spread each variable too thinly due to its over-ambition with limited resources. Further study would be best suited if it isolated one element at a time and then united them into a variable analysis. The selection of participants should also be put into question. Although the use of my personal Facebook network was partially offset by snowballing for the latter half of participants there is still strong bias of limiting the sample frame to individuals of a relatively narrow socio-cultural sphere. Incidentally, there is a gender bias in that the research includes far more women than men (5/9). Also despite the wide age-range (18-64) more participants clustered around age 33.9. Sociologically, we can speculate that individuals around this age are shifting from wider informal social groups to more formal relations. This may explain some of the cynical accounts of receiving ‘Likes’ from weak ties. Finally, a comparative ethnography that moves beyond this study’s focus on a relatively affluent sub-sample from London, England towards a comparative focus of diverse socio-cultural spheres would be beneficial. There is some evidence that ‘Likes’ are not used for communicative purposes at all. Further research is required in this area. These findings contradict the initial assumptions (A1), but suggest the many genres of ‘Like’ usage. For instance, four participants reported that their ‘Like’ clicking was not intended to communicate anything to anyone. For them, these markings were reserved for personal reasons. For instance, G and N use ‘Likes’ as a conversational marker: a reminder for themselves to ask someone about a post in their - 32 - MSc Dissertation Kenneth J. Gamage preferred offline setting. H, when pressed for the intended interpretation in receiving a ‘Like’ for her humorous post, reported that it was only for the simple fact of marking something as funny, and not to convey to the recipient that they laughed. In these accounts ‘Likes’ enable an innate human behaviour of naming, categorizing online worlds (read-write). This also relates to ‘Liking’ as non-communicative symbolic activity, but an internal process of negotiating internal and external symbolic worlds. Finally, further research would benefit building from the anthropological social memory studies that the present study only superficially touches on. CONCLUSION It is worth taking a moment to recount this study’s steps. First, using Carey’s (1975) symbolic exchange as its conceptual framework the theoretical chapter discussed social networks and gift exchange (community). Following that analysis, I explored rituals of ratification and remedial interchange, and the mediation of attention (communion). ‘Likes’ were also examined as interaction signals (communication). Then the mixed methodology of semistructured interview and thinking-aloud was explained vis-à-vis the fourteen participant sample. Finally, the findings were presented and analysed using a thematic and nonprobability variable analysis. By integrating the theoretical framework and the findings, latent patterns were further interpreted. Narrowing our focus towards the two inter-related research questions, two statements can be made. Q1: The meanings that ‘Likes’ inter-subjectively convey and interpret are polysemic. Sometimes they are understood and other times they are misinterpreted. They are used to signal language (‘thank-you,’ ‘I read that,’ conversational intrusions etc.), paralanguage (‘nods,’ ‘glances,’ ‘high-fives’), emotion (‘I was thinking of you,’ ‘I remember that,’ paying attention to someone etc.,) and gifts (flowers, cards, presents, memories, retroactive attention being paid, conversation starters etc.) Q2: These identified meanings and perceived values are socially constructed by two strategies: (1) conforming to the social code of keeping one’s ‘Likes’ scarce as related to Weiner’s (1992) “keeping-while-giving theory” and (2) through un/intentionally varied combinations of the following inter-related variable elements: ‘social tie,’ ‘ritual type’, ‘reciprocity,’ ‘interaction mode’ (semi-/public performance awareness), and attention. The salient patterned themes that occur due to these variable combinations are: ‘Likes’ serve as on-going social support from strong ties and sufficient contact from weak ties, and ‘Likes’ act as gifts during the posting of monumental life events and language during everyday posts. In particular, these gifts can be antagonistic and nonantagonistic. Although the initial assumptions were observed there were several - 33 - MSc Dissertation Kenneth J. Gamage unanticipated processes (e.g. ‘keeping-while-giving’) and meanings (i.e. retroactive attention) that was derived from the data. Methodologically, it was discovered that some non-verbal signals are conveyed by ‘Likes’. This was revealed through retroactive interpretation, and was self-initiated. In sum, as observed within this study’s sample, Facebook members are active performers and viewers who use Facebook’s ‘Like’ tool to serve their innate social needs in symbolic exchange. This provides some evidence that the communicative use of ‘Likes’ is not determined by the technology and is likely rather shaped by people. Concomitantly, the meaning of a ‘Like’ is not determined by a simple linguistic statement of liking something. The ‘Like’ veers towards providing varying degrees of social support but in some instances it also is used to un/intentionally create social entropy. The present study’s isolated focus onto the ‘Like’ tool for Facebook comes with some issues. Considering the extremely wide textual field that ‘Likes’ interact with (i.e. profiles, comments, status-updates), largely ignoring the intervening text that the ‘Like’ intermingles with may have distorted the meanings that this study reports. Having said that, methodological techniques and studies have already approached this problem (see large scale studies of SNS comments e.g. Adamic et al., 2003 and Golder et al., 2007). Future research into meaning re/production of ‘Liking’ would benefit by integrating the following: large scale comment databases, quantitative hyperlink/tie data, and emoticon research (Walther, 1992; Utz, 2000) with small-scale qualitative studies like this one. Further, it would be wise to sample from broader range of gendered and socio-economically stratified participants. Crucially the inclusion of a political economy framework is required as ‘Likes’ market value is based on data harvesting of social interactions (see Fuchs, 2009). It would also be useful to conduct this project longitudinally to monitor how meanings emerge and submerge. Despite being beyond the purview of this study, it would be generative if future qualitative studies matched ‘Like’ receivers with their donors, perhaps integrating focus groups. One area of this study that should be improved is in tightening the lens of ‘Likes’ given and received. In this study, receivers and givers were all independent from each other. A more improved study would analyse actual units of interaction where data from a single exchange was collected and analysed from all members within the interaction unit. This could be analysed longitudinally involving focus groups from within smaller units and the wider “invisible audience” of the shared social network. Although a variable analysis has proved useful it may have been too pre-mature. Future research would benefit from a deeper excavation of key areas (i.e. misinterpreted ‘Likes’ between weak ties). - 34 - MSc Dissertation Kenneth J. Gamage ACKNOWLEDGEMENTS Dedicated to Sarah. You can finally have your husband back now. I can’t thank you enough for your resilience, support, love and inspiration throughout this arduous process. For my Mom, Dad, sister and extended family across Canada, I am forever grateful for your unending love and support. I would like to thank all fourteen participants who generously contributed their time and the stories of their private lives to this research study. This project would not have been possible without the richness of your words. I would like to express my gratitude to my supervisor Dr. Alicia Blum-Ross in helping to guide my all too often scattered and indecisive thoughts into a realizable project. Thank you to Dr. Sonia Livingstone for turning my attention to the key debates and scholarship in active audience research. I would also like to thank my Thursday dissertation study group (Claire, Manjula and Paula) and the entire London School of Economics, Media & Communications Dept. staff and students for their contagious passion. Last but not least, I would to thank my vices: beer, coffee and rap. The devil’s trinity. - 35 - MSc Dissertation Kenneth J. Gamage REFERENCES Adamic, L. A. (2008) The Social Hyperlink, in J. Turow, & L. Tsui (eds.) The Hyperlinked Society: Questioning Connections in the Digital Age (pp. 227-249). Ann Arbor, MI, USA: University of Michigam Press. Adamic, L. A., Buyukkokten, O., & Adar, E. (2003) A Social Network Caught in the Web. First Mondays , 8 (6), np. J. C. Green, & V. J. Caracelli (eds.) (1997) Advances in mixed-method evaluation: The challenges and benefits of integrating diverse paradigms, in New directions for evaluation. San Fransisco, CA, USA: Jossey-Bass. Alexander, D. (2008) Chapter 7: Mixed Methods, in N. Gilbert (ed.) Researching Social Life (pp. 126-144). London, UK: Sage. Anderson, B. (1983) Imagined Communities: Reflections on the Origin and Spread of Nationalism. London, UK: Verso. Bailey, K. D. (1994) Interview Studies, in Third (ed.) Methods of Social Research (pp. 173-208). London, UK: The Free Press: A Division of Macmillan Publishers. Batailles, G. (1988/1967) The Accursed Share: An Essay on General Economy. Volume 1, Consumption. New York, NY, USA: Zone. Baudrillard, J. (1981) For a critique of the political economy of the sign. St. Louis, MO, USA: Telos Press. Baudrillard, J. (1975) The Mirror of Production. St.Louis, MO, USA: Telos Press. Baudrillard, J. (1996) The System of Objects. New York, NY, USA: Verso. Baum, N. (2011) Social Networks 2.0, in M. Consalvo and C. Ess (eds.) The Handbook of Internet Studies (pp. 384-405). Malden, MA, USA: Wiley-Blackwell. Boyatzis, R. (1998) Tranforming Qualitative Information: Thematic Analysis and Code Development. Thousand Oaks, CA, USA: Sage. Boyd, D. (2004) Friendster and Publicly Articulated Social Networks, paper presented at the Conference on Human Factors and Computing Systems (ACM), Vienna, Austria 24-29/05. Boyd, D. (2010) Social Network Sites as Networked Publics: Affordances, Dynamics, and Implications, in Z. Papacharissi (ed.) Networked Self: Identity, Community, and Culture on Social Network Sites: 39-58, URL: http://www.danah.org/papers/2010/SNSasNetworkedPublics.pdf [Last Consulted 12 July 2012] Boyd, D., & Ellison, N. (2007) Social Network Sites: Definition, History, and Scholarship. Journal of ComputerMediated Communication , 13 (1): 210-230. Bruns, A. (2008) Blogs, Wikipedia, Second life, and beyond: from production to produsage. New York, NY, USA: Peter Lang. Burgess, J., & Green, J. (2009) Chapter Three: YouTube's Popular Culture, in J. Burgess and J. Green, YouTube, Online Video and Participatory Culture (pp. 38-57). Cambridge, UK: Polity Press. Carey, J. (1975) Communication and Culture. Communication Research (2): 173-191. Carey, J. (1989) Communication as Culture: Essays on Media and Society. London, UK: Unwin Hyman. Castells, M. (2007) Communication, Power and Counter-power in the Network Society, International Journal of Communication , 1: 238-266. Castells, M. (1996) The Rise of the Network Society. Cambridge, MA, USA: Blackwell Publishers. Cohen, N. (2008) The Valorization of Surveillance: Towards a Political Economy of Facebook, Democratic Communiqué , 22 (1): 5-22. Couldry, N. (2002) Media Rituals: a critical approach. London, Uk: Routledge. Dayan, D. and Katz, E. (1992) Media Events: The Live Broadcasting of History. Cambridge, UK: Harvard University Press. - 36 - MSc Dissertation Kenneth J. Gamage de Certeau, M. (1984) The Practices of Everyday Life, in Questioning the media: a critical introduction. London, UK: Sage Publications. de Lauwe, C. (1956) La vie quitidienne des familles ouvrières: Recherches sur les comportements sociaux de consommation (The daily lives of working families: Research on the social behavior of consumption). Paris, France: Centre national de la Recherche scientifique. Deliège, R. (2004) Lévi-Strauss Today: An Introduction to Structural Anthropology. Oxford, UK: Berg. Denzin, N. (1989) The Research Act: A Theoretical Introduction to Sociological Methods (Third ed.). New Jersey, USA: Princeton-Hall. Derrida, J. (1994) Given Time: I. Counterfeit Money (Vol. 1). Chicago, Ill, USA: University of Chicago Press. Dinidia, K., & Canary, D. J. (1993) Definitions and theoretical perspectives on maintaining relationships. Journal of Social and Personal Relationships, 10: 163-173. Dockterman, E. (2012) Are Facebook 'Likes' Free Speech? 10/08, URL: http://newsfeed.time.com/: http://newsfeed.time.com/2012/08/10/are-facebook-likes-free-speech/[Last Consulted 28 July 2012]. Douglas, M., and Isherwood, B. (1979) The World of Goods: Towards an Anthropology of Consumption. New York, USA: Basic Books. Dunbar, R. I. (1993) Coevolution of Neocortex Size, Group Size and Language in Humans. Behavioural and Brain Sciences: 681-735. Durkheim, E. (1915) The Elementary Forms of the Religious Life. London, UK: Allen & Unwin. Ericsson, K. A., & Simon, H. A. (1993) Protocol Analysis: Verbal Reports as Data (Revised ed.). London, UK: MIT Press. Facebook (2012) Facebook Help Center: Like. from Facebook.com, URL: http://www.facebook.com/help/like [Last Consulted 8 July 2012] Fach, M. (2012) Stats on Facebook 2012. from Search Engine Journal, 17/02, URL: http://www.searchenginejournal.com/stats-on-facebook-2012-infographic/40301/ [Last Consulted 17 August 2012] Fuchs, C. (2009) Information and Communication Technologies and Society: A Contribution to the Critique of the Political Economy of the Internet, European Journal of Communication (24): 69-87. Gaskell, G. (2000) Chapter 3: Individual and Group Interviewing, in M. Bauer, and G. Gaskell (eds.) Qualitative researching: with text, image and sound. A practical Handbook (pp. 38-56). London, UK: Sage Publications. Gilbert, N. (2001) Researching Social Life (Second ed.). London, UK: Sage Publications. Gillispie, T. (2007) Wired Shut: Copyright and the Shape of Digital Culture. Cambridge, MA, USA: MIT Press. Glaser, B. G. and Strauss, A. L. (1967) The Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago, Ill, USA: Aldine Publishing Company. Godelier, M. (1999) The Enigma of the Gift. Cambridge, UK: Polity. Goffman, E. (1966) Behavior in public places: notes on the social organization of gatherings. Glencoe, Ill, USA: Free Press. Goffman, E. (1971) Relations in public: microstudies of the public order. London, UK: Allen Lane. Goffman, E. (1959) The Presentation of Self in Everday Life. London, UK: Penguin Books. Golder, S. A., Wilkonson, D. and Huberman, B. A. (2007) Rhythms of social interaction: Messaging within a massive online network, in B. Steinfeld, B. Pentland, M. Ackerman, and N. Contractor (eds.), Proceedings of Third International Conference on Communities and Technologies (pp. 41-66). London, Uk: Springer. Granovetter, M. (1973) The Strength of Weak Ties. American Journal of Sociology , 78 (6): 1360-1380. - 37 - MSc Dissertation Kenneth J. Gamage Gregory, C. (1982) Gifts and Commodities. London, UK: Academic Press. Halbwachs, M. (1992) On Collective Memory (Heritage of Sociology). Chicago, Ill, USA: University of Chicago. Hall, S. (1980/1974) Encoding/decoding, in D. Hobson, A. Lowe, and P. Willis (eds.) Culture, Media, Language (pp. 128-138). London, UK: Hutchinson. Hampton, K. N., Goulet, K. N., Rainie, L. and Purcell, K. (2011) Social networking sites and our lives. Pew Internet & American Life Project. Washington, DC: pewinternet.org. Haythornthwaite, C. (2005) Social networks and Internet connectivity effects, information. Communication & Society , 8 (2): 125-147. Herring, S. and Dresner, E. (2010) Functions of the Nonverbal in CMC: Emoticons and Illocutionary Force, Communication Theory , 20 (3): 249-268. Hill, R. A. and Dunbar, R. I. (2003) Social Network Size in Humans. Human Nature, 14 (1): 53-72. Hindman, M. S. (2009) The Myth of Digital Democracy. New Jersey, NY, USA: Princeton University Press. Homans, G. (1950) The Human Group. New York, NY, USA: Harcourt, Brace & World. Huberman, M. and Miles, M. (1994) Qualitative Data Analysis (Second ed.). London, UK: Sage Publications. Huxley, J. (1966) A Discussion on Ritualization of Behavior in Animals and Men. Philosophical Transactions of the Royal Society of London , 251 (ser. B, 772), pp. 247-271. Ichheiser, G. (1949) Misunderstandings in Human Relations. Supplement to The American Journal of Sociology , 6-7. Haaretz. (2011) Inspired by Facebook, Israeli couple names their daughter Like, from HAARETZ.com, 16/05, URL: http://www.haaretz.com/news/national/inspired-by-facebook-israeli-couple-names-theirdaughter-like-1.362118[Last Consulted 5 August 2012] Jenkins, H. (2006) Fans, Bloggers, and Gamers: Exploring Participatory Culture. New York, NY, USA: New York University. Kana'Iaupuni, S. M., Donato, K. M., Thompson-Colón, T., and Stainback, M. (2005) Counting on kin: Social networks, social support, and child health status. Social Forces , 83: 1137-1164. Kollock, P. (1999) The economies of online cooperation: Gifts and public goods in cyberspace, in M. A. Smith, and P. Kollock (eds.) Communities in Cyberspace (pp. 220-242). London, UK. Lasswell, H. (1948) The Structure and Function of Communication in Society, in L. Bryson (ed.), The Communication of Ideas (pp. 32-51). New York, NY, USA: Harper. Lévi-Strauss, C. (1963/1958) Structural Anthropology. New York, USA: Basic Books. Lévi-Strauss, C. (1949) The Elementary Structures of Kinship (Les Structures élèmentaires de la Parenté). (R. Needham, Ed., J. Bell and J. R. von Sturmer, Trans.) Boston, USA: Beacon Press. Lévi-Strauss, C. (1966/1962) The Savage Mind. Chicago, USA: University of Chicago Press. Livingstone, S. (1998b) Relationships between media and audiences, in T. Liebes and J. Curran (eds.), Media, Ritual and Identity: Essays in Honor Elihu Katz (pp. 237-255). London, UK: Routledge. Marx, K. (1976/1867) Capital: A Critique of Political Economy (Vol. 1). Harmondsworth, UK: Penguin. Mason, J. (2002b) Organizing and Indexing Qualitative Data, in J. Mason, Second (ed.), Qualitative Research (pp. 147-172). London, UK: Sage Publications. Mason, J. (2002a) Sampling and Selection in Qualitative Research, in J. Mason, Qualitative Researching (Second ed., pp. 120-144). London, UK: Sage Publications. Mauss, M. (1954/1923) The Gift: Forms and Functions of Exchange in Archaic Societies. London, UK: Cohen & West. - 38 - MSc Dissertation Kenneth J. Gamage Mayer-Schönberger, V. (2011) delete: The Virtue of Forgetting the Digital Age. Woodstock, Oxfordshire, UK: Princeton University Press. Meddin, J. (1980) Symbols, Anxiety, and Ritual: A Functional Interpretation, Qualitative Sociology, 3 (4): 251269. Miller, D. (2011) Tales from Facebook. Cambridge, UK: Polity Press. Morley, D. (1980) The 'Nationwide' Audience: Structure and Decoding. London, UK: British Film Institute. Neuman, R. W. (1991) The Future of the Mass Audience. Cambridge, UK: Cambridge University Press. Press, A. (2006) Audience Research in the Post-Audience Age: An Introduction to Barker and Morley. The Communication Review, 9 (2): 93-100. Putnam, R. (2000) Bowling Alone: the collapse and revival of American community. New York, NY, USA: Simon & Schuster. Radcliffe-Browne, A. (1952) Structure and Function in Primitive Society. London, UK: Cohent & West. Radstone, S. and Hodgkin, K. (2005) Memory, History, Nation: Contested Pasts. New Jersey, NY, USA: Transaction Publishers. Rook, D. W. (1984) Ritual Behavior and Consumer Symbolismm in C. Kinnear. (ed.) Advances in Consumer Research, 11: 279-284. Sahlins, M. (1972) Stone Age Economics. Chicago, USA: Aldine-Atherton. Saussure, F. d. (1916/1983) Course in General Linguistics. (R. Harris, Trans.) London, UK: Duckworth. Sengupta, S. (2012) Business Daily: Technology, from The New York Times, 14/05, URL: http://www.nytimes.com/2012/05/15/technology/facebook-needs-to-turn-data-trove-into-investorgold.html?_r=1 [Last Consulted 17 July 2012] Silverstone, R. (1990) Television and everyday life: Towards an anthropology of the television audience, in M. Ferguson ed.) The New Imperatives (pp. 173-189). London, UK: Sage Publications. Silverstone, R. (2005) The Sociology of Mediation and Communication, in C. Calhoun, C. Rojek and B. Turner (eds.) The Sage Handbook of Sociology (pp. 188-217). London, UK: Sage Publications. Simmel, G. (1922/1964) Conflict and the Web of Group Affiliations. New York, USA: Free Press. Simon, H. A. (1971) Designing Organizations for an Information-Rich World, in M. Greenberger, Computers, Communications, and the Public Interest (pp. 40-1). Baltimore, Maryland, USA: The John Hopkins Press. Smith, M. A. (2008) From Hyperlinks to Hyperties, in J. Turow and L. Tsui (eds.) The Hyperlinked Society: Questioning Connections in the Digital Age (pp. 165-276). Ann Arbor, MI, USA: University of Michigan Press. Spencer, H. (1971) Designing Organizations for an Information Rich World, in M. Greenberger, Computers, Communication, and the Public Interest. Baltimore, MD, USA: The John Hopkins Press. Strauss, A. (1987) Quantitative Analysis for Social Scientists. Cambridge, UK: Cambridge University Press. TIME Magazine. (2012) TIME Video, from www.time.com, URL: http://www.time.com/time/video/player/0,32068,711054024001_2037229,00.html [Last Consulted 28 July 2012] Toffler, A. (1980) The Third Wave. London: Collins. Utz, S. (2000) Social information processing in MUDs: The development of friendship in virtual worlds, Journal of Online Behavior, 1 (1): np. Walther, J. (1992) Interpersonal Effects in Computer-Mediated Interaction: A Relational Perspective, Communication Research , 19 (1): 52-90. Weber, M. (1958) The Religion of India. Glencoe, IL, USA: Free Press. - 39 - MSc Dissertation Kenneth J. Gamage Webster, J. G. (2008) Structuring A Marketplace of Attention, in J. Turow and L. Tsui (eds.) The Hyperlinked Society: Questioning Connections in the Digital Age. Ann Arbor, Michigan, USA: University of Michigan Press. Weiner, A. (1992) Inalienable possessions: the paradox of keeping-while-giving. Berkeley, CA, USA: University of California Press. Wellman, B., Salaff, J., Dimitrova, D., Garton, L., Gulia, M. and Haythorthwaite, C. (1996) Computer networks as social networks: Collaborative work, telework, and virtual community, Annual Review of Sociologu , 22 (1): 213-238. Wilson, T. D. (1994) The Proper Protocol: Validity and Completeness of Verbal Reports, Psychological Science , 6 (6): 249-252. Young, K. A. (2005) Direct from the source: the value of 'think-aloud' data in understanding learning, Journal of Educational Enquiry , 6 (1). * Please note that only select appendices have been included below for this published version. - 40 - APPENDIX A | PARTICIPANT SAMPLE AND KEY CHARACTERISTICS Participant Age Gender Occupation Facebook Use Frequency Viewer/ Performer Ratio Strong/Weak/ Intermediate Tie Ratio Friend Count Frequency of ‘Likes’ Given Frequency of ‘Likes’ Received A. 24 M Univ. Student At least 3 times per day 7/3 34 / 30 / 338 402 moderate low J. 31 F Lawyer Once a day 9/1 9 / 46 / 38 93 low low B. 29 F Teacher Always logged-on 7/3 31 / 107 / 359 497 high moderate G. 32 M IT Once a day 9/1 10 /56 / 106 172 low low F. 32 F Researcher Once a day 8/2 6 / 50 / 286 342 low low H. 36 F Office Admin. Always logged-on 3/7 19 / 24 / 106 149 high high M. 64 F Retail At least 3 times per week 9/1 58 low low N. 38 F Project Manager Once a day 6/4 28 / 23 / 65 116 moderate high P. 33 F Business Owner At least 3 times per day 5/5 17 / 74 / 803 894 moderate high S. 18 F H.S Student Always logged-on 5/5 15 / 36 / 285 336 moderate moderate T. 38 M Visual Artist Always logged-on 4/6 35 / 69 / 1079 1,183 high moderate V. 24 M Univ. Student At least 3 times per week 10 / 0 434 low low Z. 40 M Office Admin. Always logged-on 8/2 5 / 23 / 195 223 moderate low D. 35 F Media Always logged-on 4/6 7/ 39 / 201 247 moderate moderate Sample Mean 33.9 M: 5 F: 9 –– Always logged-on 6.7 / 3.3 16.2 / 44.8 / 305.5 367.6 moderate low 7 / 18 / 25 4 / 41 / 389 APPENDIX H | MATRIX OF VARIABLE INTERACTION (GIVING ‘LIKES’) Column A Column B Column C Column D Column E QQ2 Explanatory Variables Social Situation Monumental Life Event Mundane Social Rule Social Tie High Expectation to Give Strong Attention Column F QQ1 Response Variables Interaction Mode Social Meanings Given Row 1 Semi-/Public ‘Congratulations’ [Non-Antagonistic Gift] Abundant Viewing ‘Like’Recipient’s Content Row 2 No Obligations Weak Generalized Reciprocity Strong Moderate Dyadic Laughter, social support, banter, marking, paying attention, dialogue [Para-/Language] Preserve ‘Likes’ Intermediate (old friends) ** Various (selecting viewing of posted content) Dyadic Conversation Starter Re-affirm bond Friendship re-negotiation [Para-/Language] Preserve ‘Likes’ Weak (acquaintances) Scarce Dyadic & Semi/Public Minimal Social Support Boredom, Social Intrusion ** [Antagonistic Gift] LEGEND: Codes in Black are Indexing Categories | Codes in White are deductive | * Asterisk indicates emotive scale from 1-3 Row 3 Row 4 Row. 5 APPENDIX I | MATRIX OF VARIABLE INTERACTION (RECEIVING ‘LIKES’) Column A Column B Column C Column D Column E QQ2 Explanatory Variables Social Situation Monumental Life Event Social Rule Social Tie High Expectation to Give Strong Attention Column F QQ1 Response Variables Interaction Mode Social Meanings Received Row 1 Semi-/Public Popularity, celebration, Social support [Non-Antagonistic Gift] Abundant Viewing ‘Like’Recipient’s Content Row 2 No Obligations Weak Generalized Reciprocity Strong Moderate Dyadic Laughter, social support, banter, ‘I’ve seen it,’ affirming [Para-/Language] Preserve ‘Likes’ Intermediate (old friends) ** Scarce Dyadic Conversation Starter Re-affirm weak tie Friendship re-negotiation [Para-/Language] Scarce Dyadic & Semi/Public Edit past memory Mocking ** [Antagonistic Gift] Mundane Preserve ‘Likes’ Weak (acquaintances) LEGEND: Codes in Black are Indexing Categories | Codes in White are deductive | * Asterisk indicates emotive scale from 1-3 Row 3 Row 4 Row 5 Electronic MSc Dissertation Series Media@lse Electronic MSc Dissertations will: Present high quality MSc Dissertations which received a mark of Distinction (72% and above). Selected dissertations will be published electronically as pdf files, subject to review and approval by the Editors. Authors retain copyright, and publication here does not preclude the subsequent development of the paper for publication elsewhere.