As I’ve been working on a research project to examine how people manage their online identities across social networking sites (it’s not too late to participate in a research interview, if you’d like to help!), I have been thinking more and more that what we’re studying is how people speak about how they manage their online identities – not how they actually do so. That is because identity work (performing one’s identity) is not entirely self-conscious and strategic. Granted, online we have more time and opportunity to engineer the identity we present – and many of our research participants really do think about this in sophisticated ways. But what people say they do and what they actually do are different things. I’ve been thinking about the need to complement self-reported data with observational data.

And then, while consuming my Monday dose of lolcats, I came across this sociolinguistic study of LOLspeak – which lead to a conclusion that LOLspeak is used to construct online identity – not only as a cat, but also as an Internet-savvy person.

I followed the line of thought and some of the references cited in this study, and I was attracted to the line of work by Bucholtz & Hall (not that Hall, but the name coincidence helped). So I read a paper that I feel will become a cornerstone in one of my future studies of online identity. I summarize it here and think how it applies to my research of online identity management.

Bucholtz, M., & Hall, K. (2005). Identity and interaction: A sociocultural linguistic approach. Discourse Studies, 7(4-5), 585-614.

The paper integrates several strands of sociolinguistic, linguistic, anthropological, and cultural studies to define an approach to identity. This approach defines identity as emergent through social interaction, local and situated in specific interactions and contexts. This view of identity is contrasted with that of identity as a stable concept that the person brings to the social interaction. Identity is performed and created in millions of micro-interactions – small acts of speech and interactions. In the authors’ words:

[…] identity is emergent in discourse and does not precede it. […] identity [is] an intersubjectively achieved social and cultural phenomenon. […] the process of identity construction does not reside within the individual but in intersubjective relations of sameness and difference, realness and fakeness, power and disempowerement. (pp. 607-8).

This social constructionist view of identity is not groundbreakingly new. It echoes the social interactionist view of the self advanced by Mead, for example. But the paper goes beyond proposing a view of identity. It proposes five principles that can guide the study of identity.

Principle 1: Emergence

The first principle states that language does not play the limited role of reflecting an individual’s internal mental state (or sense of identity) – but that language use actively constructs identity. People “do,” or perform, identity through discourse.

Identity is best viewed as the emergent product rather than the pre-existing source of linguistic and other semiotic practices and therefore as fundamentally a social and cultural phenomenon (p. 588).

Research implication: Examine discourse, and especially interaction, to see how people actively construct and perform their identities. In social media, the advantage is that this discourse is already inscribed. The examination of interaction means looking at actual conversations, not just profiles.

Principle 2: Positionality

Social science often associates identity with broad demographic categories, such as age, gender, race, etc. However, identities are much more local than that. Even in a group that seems homogenous according to demographic characteristics, local, temporary roles are enacted by participants in moment to moment interaction. Participants may assume and enact temporary roles, and orientations (stances) that depend on the particular context and situation, and can only be grasped through a detailed, ethnographic approach.

Identities encompass (a) macro-level demographic categories; (b) local, ethnographically specific cultural positions; and (c) temporary and interactionally specific stances and participant roles (p. 592).

Research implication: This principle points to detailed ethnographic observation as a necessary research method to understand identity construction. Survey research is not only self-reported, but assigns identity based on demographic categories. Ethnographic observation has access to understanding local contexts and the specific cultural position an actor takes in a given context; and to the transient, temporary roles and stances participants assume. This means, for example, that studying an interaction on Twitter is necessary, but not sufficient. To make sense of that interaction, one needs to understand the cultural context of the participants. To a great extent, the cultural context is defined by Twitter culture and social norms (whatever those are). So, an ethnographic, participant understanding of this culture is a prerequisite to being able to draw any conclusions about Twitter interactions (and one of the reasons why social media immersion is required in my social media seminar). It also follows that identity on Twitter is constructed through several micro-interactions. Each micro interaction is temporary and local, and in any give micro-interaction the participant may assume any given role of stance. These roles of stances need not be consistent and coherent across time, although I can see that patterns would emerge in time.

Principle 3: Indexicality

Indexicality gets down to the very mechanisms by which identity is constructed. Simply put, the way we talk use language positions ourselves and others as certain kinds of people. At the more obvious level, this is done by assigning labels – stating that someone belongs to a certain social category. At more subtle levels, the way we use language may be associated with a certain group of people – for example, scholarly language with academics, or certain linguistic structures with women, members of ethnic groups, etc. The study of LOLspeak mentioned earlier concluded that use of LOLspeak serves to build one’s identity as an Internet savvy person, one who is informed about memes. This is done exactly through the process of indexicality. Other mechanisms through which identity is produced through language use are implicatures and presuppositions; orientations or stances; footings and participant roles; style; etc.

Identity relations emerge in interaction through several related indexical processes, including: (1) overt mention of the identity categories and labels; (b) implicatures and presuppositions regarding one’s own or others’ identity position; (c) displayed evaluative and epistemic orientations to ongoing talk, as well as interactional footing and participant roles; and (d) the use of linguistic structures and systems that are ideologically associated with specific personas and groups (p. 594).

Research implications: Research should include microscopic attention to how language use positions the participant and others in social categories, or exhibits styles and stances associated with social categories. For example, an accumulation of forceful stances may be equated to a display of a masculine gendered identity. These social categories can only be understood in the context of socio-cultural values (aka the ideology) of the group. For example, many people on Twitter post recommendations – mostly, recommendations for things to read. How does this practice position them and others? Is the implication here that one is the recommended and the others are the audience, that one is the expert, and the other ones the students?

Note how in this tweet, the user is positioning himself both as a student, and as a recommender to other twitter users.

Principle 4: Relationality

This principle emphasizes that identities “acquire social meaning in relation to other available identity positions and other social actors” – in other words, identity is relative to the others’ positions. One can define oneself in relation to others – for example, as the same or different. In addition to similarity/difference, the Bucholtz and Hall propose two more types of relations: genuiness/artifice and authority/delegitimacy.

Identities are intersubjectively constructed through several, often overlapping, complementary relations, including similarity/difference, genuiness/artifice, and authority/delegitimacy (p. 598).

Bucholtz and Hall expand on these moves participants in social interaction can make to position themselves in relation to others, and discuss several “tactics of intersubjectivity;”playing up similarities or differences, establishing authority/credibility or sabotaging it by delegitimizing it, etc.

Research implication: It takes a trained, careful eye to begin to notice these tactics in the fast flow of everyday discourse. The fact that these tactics can be complementary and simultaneous makes things even more difficult. All that comes to mind right now is a type of discourse I often see on Twitter, that against people who self-define as social media experts or gurus. With this type of tweet, the author may do all 3 tactics at the same time: differentiate self from these self-proclaimed experts; point out the artifice/fakeness of the other; delegitimize the other while building up own authority.

Principle 5: Partialness

This is best explained in the authors’ own words:

Because identity is inherently relational, it will always be partial, produced through contextually situated and ideologically informed configurations of self and other. Even seemingly coherent displays of identity, such as those that pose as deliberate and intentional, are reliant on both interactional and ideological constraints for their articulation:

[Principle 5]: Any given construction of identity may be in part deliberate and intentional, in part habitual and hence often less than fully conscious, in part an outcome of interactional negotiation and contestation, in part an outcome of others’ perceptions and representations, and in part an effect of larger ideological processes and material structures that may become relevant to interaction. It is therefore constantly shifting both as interaction unfolds and across discourse contexts.

The authors relate this partialness principle to the concept of agency – the degree of autonomy and intentionality behind a social act. However, a more productive view of agency is that of accomplishment of social action, whether that action was deliberate or not.

Research implications: To me, this seems that identity cannot be fully known – because it is a moving target, constantly shifting and evolving, and its meaning depends not on internally defined states, but on a lot of external meanings – those of other people, of the context, culture, ideology. So, while it is possible to capture tactics and processes by which individuals construct identity, even observe longitudinal patterns, in may be impossible to capture one’s identity fully. It would be simple and simplistic to consider each social network a “discourse context” – e.g. Twitter is a discourse context, and Facebook another one. And while this may be partially (partially!) true, it may also be true that one can experience – participate in, even create, several discourse contexts on Twitter alone. What are the defining boundaries of a discourse context?

It also means that self-reports are insufficient for capturing identity processes, because not all identity actions are fully conscious, and because there are many other relational and contextual factors that need to be taken into consideration.

And finally, it means that the factors that contribute to identity construction are multiple: the self, others the self interacts with, the larger context, culture, ideology, material structures. Any study of identity that looks at only one of these is partial. Any study that looks at all of them must be… an ethnography. It may not be feasible or desirable to look at all of them, but it is important to acknowledge and articulate the partialness of the very knowledge we produce.