The Australian feminist blogosphere: Methods of analysis
Frances Shaw
Paper to The Australian Sociological Association Annual Conference, Canberra, December 2009
Abstract: As part of a workshop dealing with different approaches to mapping social movement evolution, this paper addresses research into online feminist communities and the discursive legacy of the feminist movement in Australia. The Australian feminist blogging community is a discursive community that is reproducing existing feminist discourses, challenging mainstream discourses, and generating new discourses. This is an example of discursive politics in an online activist network.
The paper provides an overview of the methods of analysis to be used and a discussion of their appropriateness. These methods include participant observation, critical discourse analysis of particular discussions, network analysis using the IssueCrawler program with modifications, and finally semi-structured interviews of movement participants. Feminist standpoint theory, theories of discourse and communicative action, as well as the implications of some new developments in internet ethnographic research and cultural theory, are addressed.
It will provide a discussion of the particular problems raised in internet research, particularly in regard to approaches to ethics. It will also raise the theoretical problems and opportunities of using existing social movement theory to understand the presence of social movement activity in new media.
Key words:internet ethnography, discursive politics, online communities, contemporary Australian feminism
Introduction
My research into the discursive legacy of the women’s movement focuses on the uses of the internet for contemporary Australian feminists. The Australian feminist blogging community is a loose-knit and informal network of blogs that link to one other, list each other on ‘blogrolls’, and engage in debate and discussion from feminist and progressive perspectives (Bruns 2006:16; Boyd 2009; Lovink 2008; Rettberg 2008). The community also connects with the international feminist blogging network and various other progressive and minority blogging communities all over the world (particularly New Zealand and the United States). I have chosen this community for the study of discursive feminist activism in Australia because it is engaging in feminist activism in new ways, using new strategies for engagement with and criticism of mainstream public debate.
My research relates centrally to discursive activism. Research shows that the internet has become part of a process by which social, national, subcultural, generational, and even local discourses develop (Bahnisch 2006; Goggin 2004; Kolko 2001). Cohen (2006:162) argues that blogs have a history located in personal diaries and journals, and media forms such as zines and newsletters, as well as a social genealogy. Each of these different genealogies have resonances with the history of the women’s movement, with life writing traditions, feminist publishing initiatives, and the social networks that help to create and maintain social movement activity (Bell 2007:100; Duncombe 1997:70; Kitzmann 2004; Melucci 1989; Serfaty 2004; Young 1997).
Australian feminist blogs engage with politics of disability and ableism, race and racism, transgender rights and discrimination, queer politics and heteronormativity, and many other issues relating (in particular) to difference and exclusion, and also engage with mainstream political issues. The network frequently functions to critique the idealogy of mainstream discourses, raise awareness, and raise feminist and anti-discriminatory consciousness.
This network will provide a case study for research into discursive activism in online politics. While the medium is new, many of the discursive tools used within the feminist blogging community echo the strategies of feminist activists from earlier times. Consciousness-raising and linguistic interventions, for example, form part of the discursive work of the community (Katzenstein 1995). Feminist bloggers are also developing new and innovative methods of discursive politics specific to internet spaces.
The theoretical perspective that grounds this research draws upon feminist standpoint theory, new social movement theory, a politics of listening, affect theory, agonistic democracy and new discourse theory. Building on these approaches, I make use of a set of qualitative research methods that will provide a rich picture of the feminist blogging network. This paper briefly outlines these methods and the theoretical underpinnings that inform the approach, as well as providing some analysis of the ethical issues that are unique to online spaces, and the relative lack of established research methods for understanding social movements in online spaces.
Methodology and the theoretical underpinnings of my research
In brief, the methodological approach that informs my research combines an understanding of research as situated that is advocated by feminist standpoint theory (Bar On 1993; Brooks 2006; Devault 1990; Haraway 2004; Hekman 1995; Hennessy 1993; Ramazanoğlu & Holland 2002; Smith 1987; Sprague 2005; Taylor 2000), with the reflexive approach to internet research advocated by internet ethnographers (Baym 2000; Baym 2006; Hine 2000; Markham 1998; Senft 2008; Sundén 2003; Turkle 1995). I also make use of the concept of agonistic democracy (Laclau 1996; Laclau & Mouffe 1985; Mouffe 2005; Rancière 1999; Rossiter 2001), and the role of affect (Barker 2001; Brader & Valentino 2007; Flax 1993, 89; Goodwin et al. 2001; Jasper 1997; Neuman et al. 2007) and listening (Bickford 1996; Couldry 2006; Couldry 2009; Crawford 2009; Dreher 2009; Ratcliffe 2005) in the political. I emphasise the importance of developing a methodology for understanding political discourse in online settings that recognises the contingency and undecidability of the political (Laclau 1996; Mouffe 2005), as well as the inevitability of conflict, processes of exclusion and power relations in any understanding of discursive activism online (Cohen 2006:166; Coyle 1996:50; Dean 2002:45; Laclau 1996:52-53; Mouffe 2005:18; Smith 1987:20; Wajcman 2004:42).
White (1991:96-103) argued that feminist philosophy had potential to resolve the contradictions of postmodern philosophy that troubled understandings of subjectivity and political agency. Standpoint theory provides an understanding of subjectivity and agency that supports an understanding of discursive politics as activism, and that the capacity to act and notions of individual agency are informed by power relations rather than assumed to be equal and universal (Hennessy 1993:67; Hirschmann 1989). A compatible understanding of political discourse, agency, and power is present in the theory of agonistic democracy advocated by Laclau and Mouffe (1985) as well as Mouffe (2000; 2005), Laclau (1996), Žižek (2008), Butler (2000) and Rancière (1999). In agonistic democracy, power relations are politically constitutive, and counterhegemonic discursive struggle is fundamental to democracy (Laclau 1996:53; Mouffe 2005:18). This theory informs my understanding of discourse and the political, and will affect the approach to discourse analysis taken in my research (Howarth & Stavrakakis 2000).
Towards the late 1990s a tradition of internet ethnography began to develop among internet researchers. Markham’s (1998) groundbreaking work on internet ethnography, Life Online: Researching Real Experience in Virtual Space, enunciated some of the contradictions of ethnographical research in online communities, while emphasising the reality of online experience as lived experience (Markham 1998:115). Particularly at the time, and in much sociological research on the internet even today, this was a groundbreaking framing within a discipline that privileges face-to-face interaction, community, and research. Hine’s (2000) Virtual Ethnography advocates a similar approach, and challenges the face-to-face emphasis of ethnographic research historically. Internet ethnography necessarily takes place largely in online and networked spaces. Markham also emphasises the reflexive approach to online ethnography.
Feminist standpoint theory is compatible with this approach. The understanding of politics decribed above necessitates an awareness of the researcher’s own situatedness and a recognition of research itself as part of discourse (Ramazonoğlu and Holland 2002:63; Sprague 2005:52-54). Knowers are “specifically located in physical spaces, in systems of social relations, within circulating discourses” (Sprague 2005:47).
Recent work on “political listening” and the role of enjoyment and other affects also contributes to my research approach (Burgess 2006; Couldry 2009; Crawford 2009; Dreher 2009). This work enables me to take into consideration affective dimensions of communication such as fear, shyness, anxiety, kindness, anger, and hatred, as well as the ways that power relations affect social expectations of being heard and thus to engage in discursive activism. An assumption such as this is compatible both with the politicising aims of new discourse theory (Laclau & Mouffe 1985) and the ethical expectations of feminist standpoint theory (Davion 1998; Devault 1990).
In sum, my approach brings together theories of agonistic democracy, internet ethnography and feminist standpoint theory. These theoretical perspectives are compatible with a view of subjectivity that allows for discursive political agency, and also encourage an ethics of listening and respect for difference that doesn’t require the denial of affect, inequality and power relations in the study of online communities (Dean 2002; Marx Ferree & Merrill 2004; Mouffe 2000; Wajcman 2004, 42).
Methods
Online conversations and interviews will form the bulk of the qualitative data for analysis in this project. However the choice and analysis of these texts will be complemented and informed by network analysis. Network analysis in social movement studies involves the use of network maps to question and discover how the network location of individuals shapes their actions and behaviour (Diani 2002:174). Internet researchers have developed several ways to map and visualise online networks, with the development of software programs that assist in the visualisation and conceptualisation of these networks (Bruns 2007). In this way internet and social movement research agree on the usefulness of network theory for the understanding of social networks. Social networks are highly significant for the study of social movements because within these networks are developed the stories, discourses, practices and shared meanings that are required for collective action (Taylor 1996:68). Although social movement theorists have often been sceptical of the potential for the internet to facilitate community, online social networks should be taken seriously in the study of activist cultures (Wilson & Peterson 2002:456-457).
Network analysis helps researchers to understand blogging networks by making visible the structure of the community, as well as some of the ways that the community links together with other blog networks. It can help show what sections of the community or more closeknit and central (for example through dense and frequent linking patterns), and where factions and divisions may lie, which will leave room for analysis of whether these divisions are based on conflicts, affective relationships, demographics, or differing political priorities.
I propose to use IssueCrawler ( as the main program with which to analyse the network. I have chosen this program for its user-friendliness and because data can be exported for analysis using other network visualisation programs, which makes it a flexible option. Bruns (2007) has shown that this program can be successfully used for the purpose of mapping blog networks in Australia. The program provides basic network maps and information about the number of linkages, describing the main hubs in the community, and the sites and participants that have the closest linkages to one another. It can also show the ways that the community changes and develops over time. This network analysis method uses automated snowball sampling and analysis of link frequency to delineate the limits of the network (Diani 2002:177).
While network analysis may provide useful insights into the community itself, its main role will be a peripheral one; to assist in the identification and delimitation of the network itself, as well as helping to identify the main actors in the community. In this way network analysis will supplement my own observation of the community through reading and exploring the network itself. In internet research, Baym (2006:82-85) advocates a methodological approach that combines participant observation with other research methods, interviews in particular. These ethnographic research practices will inform my methods, particularly in terms of online participant observation and analysis of messages, but also the combination of message analysis and accounts of online experience gained through interviews. This is a common triangulation of methods in online spaces (Baym 2000; Baym 2006; Hine 2000; Jones 1999; Markham 1998; Nakamura 2008; Senft 2008; Sundén 2003; Turkle 1995). Face-to-face interviews will help contextualise the research and provide further opportunities to gain the perspectives of community members about their role in and experience of this community (Baym 2006:85; Orgad 2009:39).
The most important methods for my research are discourse analysis and semi-structured interviews with community participants. Discourse analysis will involve the selection and analysis of texts and conversations within the Australian feminist blogging community (Paltridge 2006). I will analyse two types of discussion: those that occur between blogs around a particular issue, and discussions that develop around a single post.
Analysis of the second type of discussion is important because it includes both the arguments of the original poster, who sets the agenda for the debate, and also includes the critical or approving reactions of other participants. Analysis of comments on posts also allows for the responses of non-blogger participants, casual participants in the network, and responses from those from outside the network, whether interested outsiders or “trolls”; aggressive and disruptive participants who either provoke irrelevant debate to derail discussion, or harass the original poster or other participants (Brail 1996; Herring et al. 2002). This approach to discourse analysis recognises that often the meaning given to an utterance can become the meaning of the utterance (Wood and Kroger 2000:109-110). Thus the comment responses can serve to actively change the eventual meaning of the original text.
Finally, but most importantly, I will be conducting interviews with members of the community. I have chosen a method of interviewing that is semi-structured and conversational in style, so as to not standardise expectations and responses. Semi-structured interviews also allow for reflexivity (Baym 2006:86) and the ability to scrutinise the meaning of statements made by social movement participants (Blee & Taylor 2002:94-95). This corresponds to what Hine (2000:154) calls “adaptive ethnography”, and is also compatible with feminist standpoint methods.
The semi-structured interview approach serves to minimise "the voice of the researcher, aiming to bring the everyday worlds of activists to the fore” (Blee & Taylor 2002:95). It is also a reflexive and iterative method, whereby researchers can discover new research questions and approaches in conversation with participants.The interview technique involves the construction of loose a interview guide which reflects the aims of the research project. However, the guide would be adaptable to the situation of the people/person being interviewed (Blee & Taylor 2002:99).
Interview participants, in this method, are chosen specifically for their role in the social movement, rather than at random (Blee and Taylor 2002:100). In my case I would identify potential interview participants according to their place in the network map and the role that I have seen them take in discussions, to provide comparative perspectives on network participation. These choices would arise in the course of my network and discourse analysis as patterns emerged in the mapping and discursive interpretation of the network. The conversations I have with the movement participants will then inform my analysis of the network and discourse.
Problems
The main problems that I foresee in my research involve an adjustment to the differing construction of public and private spaces online. My research approach is complicated by the way that internet media crosses public and private designations. Understandings of privacy online must always be context sensitive (Baym and Markham 2009:xviii; Serfaty 2004:12-13; Sveningsson 2009). For example, on one level we may understand or define a blog as a public space or at the very least a form of published media, but on another level it may be felt to be a private or “home place” for the blogger, and permission may be required to republish, cite or quote online texts in another medium (Sundén 2003:38; Serfaty 2004; Sveningsson 2009). Assessments of whether online texts are to be considered public must be carefully made (Sveningsson 2009:74-75). These boundaries should not be assumed, but negotiated in the course of the research (Hine 2000:64).
As Baym and Markham (2009:xviii) argue, the role of the researcher is particularly uncharted in internet research, partly in participant observation, because a researcher can observe unnoticed online in ways that they cannot in face-to-face interaction. Therefore the necessity for authentic self-representation, building trust, respecting privacy, and obtaining consent is all the more imperative in online research (Baym and Markham 2009:xviii; Hine 2000:48; Wilson and Peterson 2002:461).
Lichterman (2002:126) argues for the importance of being clear about ones’ position as researcher from the very beginning. For this reason I have a research blog and various PhD-specific internet profiles, so that I am seen as participating as a researcher rather than as an everyday internet user. Occasionally this approach can complicate both my everyday use and research on the internet, because I am at all times both a researcher and a user with political stakes in the issues being discussed, and who has relationships in online spaces. Naples (2003:49) argues that ethnographers are never fully outside or inside of the community being studied – this is something that must be constantly renegotiated. This is particularly true of participation in online communities, where a researcher’s presence is virtual and lacks the physical “arrival” traditionally associated with ethnographic research (Hine 2000:43-45).
A possible solution is to only perform participant observation (in my case network and discourse analysis) in sites and discussions that would be deemed public according to the criteria set out in studies of online research ethics, unless I have received informed consent and thus made my presence known. According to Sveningsson’s (2009:75) criteria, a public site is one that is “open and available to everyone, that anyone with an internet connection can access, and that does not require any form of membership or registration”. However, I would prefer to obtain permission to quote people’s words in my analysis, due to the fact that users may understand their online contributions as semi-private because intended for a smaller or different audience than my research (Sveningsson 2009:75-78). Stern (2009:96) also advocates asking the posters of online content whether or not they consider their writing to be part of the public sphere, or intended for a specific intimate audience.