article

FCJ-155 EVEN WITH CRUISE CONTROL YOU STILL HAVE TO STEER: defining trolling to get things done

By ⋅ December 30, 2013 ⋅ Post a comment
Filed Under  , , , , , , , , , , , , ,

Andrew Whelan
University of Wollongong

[Abstract]

This is an article investigating trolling as an observable and reportable phenomenon, and how it comes to be sensible as such to those who describe interactional or discursive forms as trolling. The interest is not so much in what trolling ‘really is’ or what trolling ‘really means’ or what trolling ‘really says about where we are now’. Rather, it is an exploration of what might be the best means by which we can understand how trolling is identified, and what the intertwined moral, cognitive, and intersubjective processes at work in this identification are. What are we even talking about when we’re talking about trolling, and how do we come to understand this?

The argument is structured as follows. The first part of the article considers a particular representation of trolling in detail, a famous TV news segment, in terms of relevant literature on deliberative democracy, moral panic, and risk. It is a brief gloss of what I imagine an account of trolling and how it can be understood might look like in conventional academic terms. The second part of the article seeks to problematise this account, by situating it and the Seven News segment it is articulated through with respect to ethnomethodology and membership categorisation analysis. The emphasis throughout this close reading and discussion is on attending carefully and cautiously to what it is that people get done when they invoke trolling.

The analysis conducted is of the clip below from Seven News, ‘Charlotte Dawson fights back against trolls’. The segment aired on October 23rd, 2012, during The Daily Telegraph’s ‘Stop the Trolls’ campaign, for which the late Dawson was a spokesperson. It had a well-publicised backstory, involving Dawson’s previous ‘doxxing’ or ‘outing’ of a troll, which in turn led to a further bout of organised retaliatory trolling directed at Dawson, which led to her attempted suicide in August of that year. I will not go further into this story or the various ways in which media commentators and others (including trolls) contested Dawson’s position in it at that time, other than acknowledging that it likely provided some context for those who viewed the segment and were familiar with it. Rather, in what follows I attend specifically to the narrative of the segment itself and the moral logic it articulates: what is the definition of trolling mobilised here and what can be said about it? In what ways can a close and considered reading of the segment shed light on how trolling is represented and defined in mainstream mass media? What are the interpretive frameworks best suited to understanding this process?


Figure 1.

Aside from its poignancy, the segment is instructive and cautionary as regards the heated discussion of trolling in Australian mass media and how that has been conducted in recent years. The segment can be analysed for the definitions of trolling it mobilises, why trolling is (framed as) a problem and why it comes to be such, what its effects are, and what should or could be done about it. Specifically, we could begin to understand trolling as presented in the Seven News segment in the following ways:

as a threat to the public sphere, specifically, the public sphere as a space of deliberative democratic dialogue;

as the grounds for a moral panic: more precisely and interestingly, a moral panic the media has about itself; and

as a risk to (be managed by) those who engage in online media (and indeed, any media).

These distinct themes are woven together in the segment in subtle ways, and as such, it also sets up an interesting counterpoint between reason or rationality and emotionality or affect for the parties involved (the trolls and Dawson as ‘trollee’, respectively), articulated through and alongside a tacit model of the moral underpinnings of this counterpoint.

Implicit in the segment is the popular idea that trolling is radically disruptive to the ideal of the public sphere as a deliberative democratic space (where this may be read as subversive and emancipatory, or, as by Seven News, as negative and destructive). That the segment was aired at all is indicative of a kind of interest, felt presumably by staff at Seven News, and/or felt by them to be sufficiently present among the audience (perhaps on the basis of the broader media interest at that time) to warrant coverage. Somehow there was a mediated public sphere ‘before’, where we were safe from abuse, and now, along with the democratisation of voice social media seems to imply, there is danger and chaos: as previous reports attest, a ‘HATE CAMPAIGN’ (01:00), conducted by ‘TWEET ATTACKERS’, has put a ‘Star in HOSPITAL’ (01:03). Trolling is a vituperative discursive and interactional action without account or responsibility, a new pathology of democratic dialogue. It is a pathology because, as Seyla Benhabib puts it:

According to the deliberative model of democracy, it is a necessary condition for attaining legitimacy and rationality with regard to collective decision making processes in a polity, that the institutions of this polity are so arranged that what is considered in the common interest of all results from processes of collective deliberation conducted rationally and fairly among free and equal individuals (1996: 69).

‘Democracy’, ‘legitimacy’, ‘rationality’, ‘common interest’, and ‘collective deliberation conducted rationally and fairly’ – all laudable, and all ideals of the sort evidently under threat from the trolling contributions Seven News cites as problematic. These contributions – unlike Dawson’s expletives, which receive the more conventional bleeps – are overlaid with whistling sounds and asterisks (‘you fucking cunt’ [00:46:]; ‘ugly ass albino Ellen DeGeneres impersonator’ [00:52]; ‘no wonder people think your [sic] a slut’ [01:34]). In defining trolling in this sort of way and with reference to this sort of evidence, the segment produces and spectacularises a bracketed class of ‘abnormal’ or ‘deviant’ statements, the tenor of which, incidentally, is innocuous in comparison to the ferocity of some of the messages Dawson had previously received. From these, we can work back to the deviated ideal: an imagining of a space for public dialogue characterised by civility and propriety of the sort Benhabib also imagines, unsullied by these hateful eruptions of grotesquerie.

What to make of these statements? Gabriella Coleman puts it thusly:

lulz-oriented actions puncture the consensus around our politics and ethics, our social lives, our aesthetic sensibilities, the inviolability of the world as it is; trolls invalidate that world by gesturing toward the possibility for Internet geeks to destroy it – to pull the carpet from under us – whenever they feel the urge and without warning (2012).

For Coleman, trolling is indeed a radical counterpublic, a communicative, gestural and performative mode which indexes the contingency of rationality and of assumptions of rational and reasonable interaction and dialogue: a mode which tends to render such assumptions absurd, and as such is legible as radical political action.

This interpretation of the segment, in terms of discrepant understandings of the norms of dialogue within a deliberative public sphere, is borne out particularly by an insistence on the part of Seven News on a certain model of immediate and direct referentiality. There is only one way to speak here and only one way to understand the practice of participation in mediated communities of speech. Utterances, so to speak, must mean what they say, for they are (to be taken as) determinedly real in their emotional effects, and are consequentially tied directly and accountably to those who produce them. They are also taken as avowedly, directly, and intentionally aimed at their singular recipients, rather than, for example, being performed for the benefit of audiences other than or in addition to these recipients. This is of course somewhat different to how Seven News and other broadcast mass media outlets operate, insofar as the audiences they speak to are not singular and the statements they produce are not* *to be understood as intended for singular recipients. This, then, is not just a ‘vernacular’ theory of meaning and representation (no Foucauldian subject positions in discourse here please); it is a theory of morality, and a theory of (authorial voice in) media to boot. Critically, this is not a model of public dialogue allowing for or endorsing anonymous contributions. The interpersonal and moral implications of statements are borne out emotionally at the site of reception, implying responsibility and accountability at the site of production. The possibility of statements without identifiable sources is here a particular moral problem. After all, Seven News identifiably mean what they say. Where would we be if mass media did not mean what they say? Should not any or everyone with access to media therefore identifiably mean what they say? Why should a model of free speech imply any right to anonymous speech?

Yet the three trolls ‘exposed’ all contest the moral accountability inherent in the model imposed by Seven News: ‘They’re just things that I say. They’re things that I say on Twitter and Twitter isn’t real life’ (00:52). This first troll, Jordan McGuire, elaborates further later on in the segment: ‘And I don’t necessarily mean what I tweet half the time!’ (02:05). What intention then could lie behind such invective: ‘Where does that come from?’ (Some psychological wellspring is perhaps implied here). ‘It doesn’t come from anywhere in particular, it just comes’, says the second troll, Caspian Shields (01:17).

Something like a psychological account, however, is engaged with by the third troll, Ian Cameron, who succinctly iterates the distinction between the real and the virtual and then assigns a particular subjective benefit to the virtual: ‘There’s real life you and internet you, I think, I gain a little bit more confidence on the internet’ (01:36). This is not quite a concession to authoriality or responsibility of the sort Seven News appear to be aiming for, however. Rather, it seems to frame the internet as a kind of cathartic identity playground. This sort of reasoning has been described with reference to the ‘greater internet fuckwad theory’: the rather deterministic idea (more precisely, alibi) that pseudonymity as a feature of online environments (rather than the people involved, their cultures of use and participation, and the social contexts which normalise them) somehow generates offensive behaviour (Nakamura, 2013). Ian Cameron elaborates further in the segment, in such a way as to differentiate ‘the internet’ from ‘the media’ (01:42) those in the latter are ‘fair game’ for what might be said in the former.

It is worth noting that the sequence of events and of troll contributions, and consequently the full apportioning of culpability, is somewhat unclear here. The first interaction is introduced by voiceover with: ‘Charlotte Dawson meets one of the trolls who sent her abusive messages on Twitter while she recovered from a suicide attempt’ (0:20). Were all of the featured trolls then latecomers to the scene, and not among those trolls who, the segment states, landed Dawson in hospital? Or only the first one? To what extent does this have implications in terms of the moral opprobrium due to the trolls, or the ‘healing’ aspects of Dawson’s journey in the segment?

The objective Seven News work to achieve involves liquidating what otherwise is continually threatening to collapse the apparent grounds of their moral warrant to condemn the trolls: the distinctions between ‘the media’ (TV) and ‘the internet’ (Twitter), and either or both of these (but perhaps especially the latter) and ‘real life’. These are to be considered synonymous, and rendering them so is presented as Dawson’s job. This is to be done by re-anchoring everything in a ‘real’ way, in ‘real life’. We learn that Dawson sought to exact retributive justice: her ‘response has been to expose the trolls by forwarding their abuse to her tens of thousands of followers’ (01:48). Trolls too can and will be made to learn that words are, in fact, like sticks and stones, and that the right to use them publicly entails the responsibility of facing their consequences for others publicly.

A moral high ground is implied: the trolls will be hoist by their own petard; their suffering will be on their own account, by their own cruel and venal hand. Like Julian Assange, Dawson is merely making available the record of the damning behaviour of evildoers. One might argue that, within the segment, Dawson is not above stooping to their level: after all, the action begins engagingly with her shouting ‘Fuck you you cunt!’ at Jordan McGuire (00:32). But this is not quite direct speech; it is couched with an explicit conditional which renders a kind of pseudo-simulated performance of trolling in ‘real life’: ‘I’m face to face with you now, if I turned around and said ‘Fuck you you cunt!’, how do you feel?’ This indexical prefacing utilising the ‘f2f’ serves to render contexts synonymous: however McGuire feels in this context (something the camera, if not the microphone, is interested in showing), is as Dawson felt in that one.

In a public sphere characterised by ‘collective deliberation conducted rationally and fairly among free and equal individuals’, forwarding instances of abuse to tens of thousands of followers might seem a reasonable enough move (if perhaps somewhat ‘agonistic’). But we learn immediately from a somewhat startled Caspian Shields that this also involved posting where he works to twenty–two thousand people. Was this information contained in his abusive tweets? We’re not told, though it seems implausible, given the well-worn lines we have already heard about trolls hiding behind keyboards (00.35). ‘It’s not bullying you’, Dawson is shown explaining, talking over Caspian Shields while reaching out to touch his arm, ‘it’s exposing you for what you are’ (02:00). Is this fair? Is it lawful? Could Charlotte Dawson and Caspian Shields ever have been said to be free and equal with respect to each other? Does Caspian Shields, by the mere fact of appearing on television, also become ‘fair game’?

The theory of deliberative democracy has of course been subject to extensive critique, some of which is salient here. Such critique can be framed in relation to the cryptonormative notion of ‘rationality’ mobilised by the theory and its elision of issues of power:

deliberative democracy does not deal with the normalising (coercion) and exclusion involved in the designation of a particular form of communication as the rational and democratically legitimate norm. In order to be considered legitimate deliberators, subjects must come to internalise the rules of the particular form of communication deemed democratically valid or be excluded from the public sphere (Dahlberg, 2007: 52–53).

Trolls, by this reckoning, are those who do not, cannot, or will not successfully achieve this internalisation. Chantal Mouffe develops Dahlberg’s position eloquently, in terms of a framework derived from Wittgenstein:

to have agreement in opinions there must first be agreement on the language used and this, as he [Wittgenstein] points out, implies agreement in forms of life. According to him, procedure only exists as a complex ensemble of practices. Those practices constitute specific forms of individuality and identity that make possible the allegiance to the procedures. It is because they are inscribed in shared forms of life and agreements in judgments that procedures can be accepted and followed. They cannot be seen as rules that are created on the basis of principles and then applied to specific cases … therefore, distinctions between ‘procedural’ and ‘substantial’ or between ‘moral’ and ‘ethical’ … cannot be maintained and one must acknowledge that procedures always involve substantial ethical commitments (1999: 749).

In the second section of this paper below, we shall return to this kind of thinking in a somewhat more concrete and analytically focused form.

The criticisms of the Habermasian ideal elaborated by Dahlberg and Mouffe underscore the local, contextual and normative specificity the ideal entails – a specificity trolling (as described by Seven News) disregards, is unable to recognise, or actively and wilfully assaults.

The problematisation of the normative notions of procedure assumed by theorists of deliberative democracy can be further contextualised in terms of the tacit assumptions around rationality they instantiate, particularly with reference to how rationality is predicated and juxtaposed with emotion in the Seven News segment. According to the narrative presented in the segment, reasonable, normal people such as Dawson, attempting to get on with their everyday business, are ‘troubled’ by trolls. Indeed, as Dawson herself inquires at 02:14, speaking on behalf of mystified reasonable persons everywhere: ‘Most reasonable people find them highly, highly offensive and they, they can’t understand the mindset behind them or the logic behind them, how do you feel about that?’ Arguably, this is how ideology functions ‘in the wild’ in television news confrontations of this sort. Given the small minority of Australians who actually use Twitter, most viewers probably didn’t think too much of anything about mindsets or logics to offensive tweets until Dawson invoked and thereby came to stand for ‘most reasonable people’.

Of course, Dawson’s role was not just to be reasonable and normal: she was also a celebrity, and that would seem, according to the segment – notably the intervention by the Seven News reporter, Jodie Speers, at 01:19 – to entail a certain deference in her treatment. This is notable by its absence among the featured trolls’ tweets. Speers refers to ‘people like Charlotte’, expressing incredulity at what they are apparently supposed to accept: they should just cop ‘whatever you put out there’. The implication of ‘out there’ is of course that the trolling in question takes place in a public place. And as with the third troll, Ian Cameron, the category ‘people like Charlotte’ is explicitly oriented to by Shields in terms of ‘being in the public eye’. This is an extremely specific definition of the trollee position: as we shall see presently, it is almost immediately negated within the segment itself (and, it must be said, within other media reports of well-publicised troll-celebrity interactions).

Also of interest is the emotional register of this ‘troubling’ and how it should be accounted for or responded to by reasonable, normal people. This presentation by Seven News is interesting for what it demonstrates about the occluded role and standing of emotion in the public sphere:

emotion is thought of ultimately as the completely other of political reason; that is to say, as a sort of atavism or primitive remainder, as a symbol of everything that has been left behind by civilization and progress, and that has no proper place in the enlightened realm of liberty created by the moderns (Máiz, 2011: 34).

The reported tweets produced by the trolls highlight this: apparently casual expressions of loathing, ridicule and contempt (all strongly gendered) are framed as taboo in the social media public sphere on account of their emotional repercussions (which is to say, Twitter is framed as a dialogical public sphere with moral and affective entailments, rather than, for instance, a public repository of latrinalia or Billingsgate). Yet Dawson’s outburst at 00:32, and its position within the segment itself, both confirms and validates the normative and constitutive emotionality upon which the reasonable and rational is predicated.

This brings us to the second point: the aspect of trolling as a moral panic. As an instance of such, the Seven News segment is notable in that this is expressed, as is customary, through mass media, but relating to the perception of an assault on mass media, its norms, and its personnel (rather than on some other space or collective – behaviour in public places, standards of sexual conduct, the sanctity of childhood etc.). One could argue that Seven News (particularly in terms of the ‘name and shame’ strategy) is expressing a certain form of ressentiment about what trolls appear to be getting away with:

This complex sentiment has three interlocking elements. First, diffuse feelings of hate, envy and hostility; second, a sense of being powerless to express these feelings actively against the person or social stratum evoking them; and third, a continual re-experiencing of this impotent hostility. The essential point distinguishing ressentiment from rebellion is that the former does not involve a genuine change in values. Ressentiment involves a sour grapes pattern which asserts merely that desired but unattainable objectives do not actually embody the prized values – after all, the fox in the fable does not say that he abandons all taste for sweet grapes; he says only that these particular grapes are not sweet. Rebellion, on the other hand, involves a genuine transvaluation, where the direct or vicarious experience of frustration leads to full denunciation of previously prized values – the rebellious fox simply denounces the prevailing taste for sweet grapes. In ressentiment, one condemns what one secretly craves; in rebellion, one condemns the craving itself (Merton, 1957: 155–6).

Seven News, then, do not precisely ‘rebel’, although one could debate whether or not the trolls who they sought out do so. The moral panic Seven News articulate is compellingly shot through with the image of the troll as abject scapegoat. It is not just that the positions of trollee and troll are so strongly and robustly indexed to gender, class, and status. Trolls, we are invited to understand, are ‘defiled selves’, dysfunctional social miscreants, simultaneously hiding pathetically behind their keyboards and omnipotent in their capacity to wound their social betters.

the defiled self is imagined as deficient in those key human traits that make a moral life possible (conscience, compassion, altruism). Defiled selves are driven by an excess of otherwise ordinary human traits, for example aggression, self aggrandisement or grandiosity. Deficit and excess are two sides of the same coin. A self lacking in moral and behavioural control engages in excessive boundary-crossing, unruly conduct. This deficit/excess disequilibrium is imagined as the governing disposition of the Other. Lurching between a state of incoherence and uncontrollable self-aggrandisement, the defiled threatens to unleash a wave of chaos and ruin in civil life. Ultimately, the Other’s extreme sociopathic and sadistic profile risks the collapse of a human world into a de-humanised object world (Seidman, 2012: 5).

It is this dehumanisation which Dawson is presented in the Seven News segment as combatting and ultimately overcoming, by demonstrating to her trolls that their behaviour is morally consequential. Not only this: in the segment, her victory is in some elliptical way related to the reformulation of the very institutional fabric of the social media space: a subtitle at 01:58 informs us that ‘TWITTER HAS RECENTLY CHANGED RULES Users can now be removed for abuse’.

The narrative arc of the segment culminates with Dawson’s ‘closure’: ‘The thing that I got out of visiting these people and them agreeing to talk to us is the fact that their online bravado is completely polar opposite to what they are’ (02:42). Dawson is presented as having ‘gotten something out of this’: correcting her previous misunderstanding of trolls and their power relative to hers. She is shown as having come to understand that trolls ‘in real life’ are weak, abject, pathetic, and cowardly; seeing the deficit which is the dialectical flipside of their pathological online excess. The distinction between the real and the virtual is transcended, by morally tethering utterances in the latter domain to bodies in the former.

This consequentiality applies to both trolls and trollees: trolls are to be taken to account for their behaviour, but trollees are also invited to take responsibility. In this sense, trolling is produced as a risk. Another subtitle at 02:13 notifies viewers: ‘ADVICE FOR PEOPLE TARGETED BY TROLLS: Block user, report to Twitter or the police’. Through these means, at a ‘meta’ level within the segment, the audience is quite literally framed as reasonable people subjected, like Dawson, to the risks of being trolled through their social media use. ‘IF YOU NEED HELP COPING’, viewers are informed, presumably with the abuse they are subject to online, they can call Lifeline (02:43). Viewers are thus democratically ‘moralised’ to take active responsibility for the management of their sadness and pain, as is now customary. As alluded to above, there is a further twist on the idea of the deliberative public sphere here. Whereas the interactions between Dawson and her various trolls have bore the implication that people in the public eye are or are not obliged to ‘cop it’ – the contested trade-off for their visibility is that they are legitimate targets for invective and abuse from members of the public who take umbrage with them – here invective and abuse become suddenly and abruptly democratic, indiscriminate, and egalitarian in their directions and targets. Viewers are to understand that they too are involved in the policing of the crisis, and that moreover their emotional vulnerability in the face of trolling is, like Dawson’s, a risk to be managed (with the assistance perhaps of Lifeline, Twitter, or the police), and to be managed particularly by them:

one common feature of the process of moralisation in everyday life is that people are called upon to engage in ethical forms of individual risk management, and these forms of self-conduct exist in tension with collective subject positions of ‘harmful others’. What this implies on a conceptual level is that moralisation in everyday life contains a dialectic that counterposes individualising discourses (which call on people to take personal responsibility to manage risk, e.g. drinking responsibly) against collectivising discourses (which represent more broadly harms to be avoided, e.g. the drunk driver) (Hier, 2008: 174).

The distinction between Hier’s example of drink-driving and trolling as covered in the Seven News segment is that at no point in the latter is any advice offered on how to not troll – how to not occupy the subject position of the ‘harmful others’. The risk for the viewer is not that of being, for instance, ‘exposed’ as a troll, the risk is exclusively that of victimisation at the hands of these harmful others. In a sense trolls are presented as only partially capable of taking on such a process of becoming responsible. Even through the sort of exposure Seven News, as an instrument of justice, can engage in, the trolls are presented as morally defective: unable, like Ian Cameron, to traverse the distinction between real and virtual, to deploy the ‘confidence’ present in internet trolling in such a way as to achieve moral reasonableness in real life. As Jordan McGuire puts it, employing a generational logic immediately endorsed by Jodie Speers’s voiceover: ‘Me, I have a very, very dull sense of what is disrespectful and what isn’t because I’m just desensitised to it and that’s what the majority of Gen Y is’ (02:27).

Within the moral logic of the segment itself, then, Seven News is not concerned with the risks or consequences of being labelled a troll, through either deliberate or unwitting statements interpreted as trolling by recipients or witnesses. Nor are they particularly interested in exploring whatever the imputed or acknowledged deficiencies or vulnerabilities are which might render one a troll. To do so would both humanise the trolls and embed their behaviour in an alternate scheme of meaning, as well as running the risk of seriously undermining the moral certainty of the approach the segment, like so much other mass media reporting, takes on the issue. But of course, the entire segment itself can be taken as an ostensive cautionary message to potential or actual trolls: suffer the consequences of your actions (where, like drink-driving campaigns which concentrate on terrible repercussions, perhaps the most significant consequence is the subjective experience of shame as an element in a mechanism of collective risk management). You could get something a bit like (albeit not exactly like) a taste of your own medicine: not only might you find yourself on the national evening news being taken to task for your behaviour, twenty-two thousand people could be told where you work and the terrible things you have thought and said, and who knows what they might say or do? The segment does not instruct viewers in how to not be trolls; it demonstrates the moral consequences of trolling: ‘That’s what I do’ Dawson is shown saying, ‘I expose people like you’ (02:03). It becomes the moral and professional obligation of media personalities to ensure the interactional norms of the public sphere are abided by.

We learn this, not just from what is said and how it is edited, sequenced, and presented, but from how that saying is also a kind of doing, because, like trolling in the segment, interaction itself is also action: ‘exposing’ trolls constitutes them as such.

Thus far, I have presented an interpretive gloss of a news segment broadcast at a particular moment in the public debate about trolling in Australia. I argued that this segment can best be conceptualised in terms of what it tells us about the ideal of the deliberative public sphere and how this is framed as undermined by trolling. I suggested the segment could be understood as an interesting instance of moral panic, and that the segment presented trolling as a significant risk to viewers, a risk viewers, in turn, are positioned to take responsibility for. In presenting this account, I sought to unpack the moral logic according to which the segment operates and the values it articulates, and in making this case, I also sought to demonstrate how such a reading or interpretation might be conducted with respect to what transpires in the segment.

Before developing an alternate line of argument with which to take this account further, I want to point to some potential problems with this unpacking. Perhaps the most immediate response to such an account could be: ‘that’s not trolling’ (it is certainly not trolling as described in the literature, for example Donath, 1999; Herring et al, 2002; Shachaf and Hara, 2010). Seven News is making a reductive error in nomenclature. Abuse directed at celebrities (which has a dismayingly long history), public or otherwise, is not trolling, or at least, contemporary trolling is broader than this, extending perhaps to attacks on the Church of Scientology, or fans of Justin Bieber, or those who would curtail the operations of Wikileaks, or perhaps those tasked with providing medical care to the Duchess of Cambridge. Alternately, such an account actually lets trolls off the hook: trolling of this sort, at least, involves systematic and targeted abuse, often directed at young or vulnerable women, and in a growing number of cases (including Dawson’s) associated with suicide. Either way, therefore, it would be a mistake to ground an account of the politics of trolling in a mass media representation, particularly a sensationalistic, ‘tabloid’ representation. We still don’t know what trolling ‘really is’, and are still not in a position to make any judgment about it.

This sort of criticism is indicative of some of the broader problems of understanding what is meant by ‘trolling’, including the issue of contextualising trolling in relation to the range of available terms with which it is now being conflated (such as ‘flaming’, ‘griefing’ or ‘cyberbullying’).

Moreover, in order to begin showing a direction forward from here, another line of critique could be developed. The interpretation above draws on three well-established concepts for which there are vast bodies of literature: deliberation in the public sphere, moral panic, and risk. These are ideas with intellectual cachet. As such, not only is it reassuring for us to understand discussions of trolling in this way. Mobilising such a conceptual vocabulary, the analysis bolsters our sense of being able to grasp a deeper, broader, ‘bigger picture’ meaning to footage of Charlotte Dawson shouting ‘Fuck you you cunt!’ at a twenty-year-old man in a residential street. As interpretive frames for this kind of material, then, these are relatively conventional, and this sort of work could be done in relation to any number of contemporary mass media accounts of trolling. It is straightforward and reassuring (and satisfyingly mobilises particular forms of intellectual capital), to assert that the thing to understand about trolling is really a thing to understand about deliberation in the public sphere. Does this mean such interpretations are correct?

I want to argue that such interpretations only get us so far, and that this is because they bring with them extensive, albeit largely implicit, baggage with respect to how the meaningfulness of a term like ‘troll’ is produced and what an appropriate academic interpretation of this would look like. In much the same way that Seven News constitutes trolling as a morally sanctionable kind of interaction, to say that the best way of understanding how and why they do so is with reference to the public sphere, moral panic, and risk, is also to constitute and delimit trolling as explicable in a particular way. This produces another layer of interpretation and abstraction, and thereby moves us further away from the social logics of what is being done with the category ‘troll’, rather than closer to it.

Making this argument requires a brief excursion through an alternate set of resources, specifically, those to be found in the research program of membership categorisation analysis and its ethnomethodological underpinnings. At the risk of making a rather subtle position appear both simplistic and prescriptive, a few basic tenets of ethnomethodology germane for present purposes can be laid out. This is an approach drawing on themes found in Schutz (1962), Winch (1990), and Wittgenstein (2001), albeit with a particular methodological and empirical bent.

As its name suggests, ethnomethodology is concerned with ‘members’ methods’: the ‘common-sense’ methods people use in an indefinite range of routine activities; what ‘anyone would be expected to know’. Harold Garfinkel coined the term, in the course of analysing jury deliberations:

Here I am faced with jurors who are doing methodology, but they are doing their methodology in the ‘now you see it, now you don’t’ fashion. It is not a methodology that any of my colleagues would honor if they were attempting to staff the sociology department … ‘Ethno’ seemed to refer, somehow or other, to the availability to a member of common-sense knowledge of his society as common-sense knowledge of the ‘whatever’ (1974: 16).

As this origin story makes clear, ethnomethodology is first and foremost an analytical orientation, it entails the study of naturally occurring practical activities and the reasoning that is expressed through them and used to account for them. It is:

above all else, a policy towards enquiry, an analytic mentality, that insists on (1) doing studies, by (2) working on materials to see what can be discovered in and from them, rather than selecting problems and data on the basis of some theoretically-specified agenda. In this way it is homologous with its own subject matter, namely social order as the ongoing achievement of members of society conceived as practical actors who are themselves (1) practical analysts of, and inquirers into, the world, (2) using whatever materials there are to hand to get done the tasks and business they are engaged in (Hester and Eglin, 1997: 1).

Ethnomethodology thus seeks to frame as the proper area of inquiry what conventional academic accounts treat as a tacit resource: the competencies of mundane practical reasoning as these are displayed or made evident.

Another way of putting this is to gesture to the distinction between ‘studies about’ and ‘studies of’ particular practices, where the abundance of studies about some practice or setting does not tell us very much at all about how that practice or setting is accomplished by its members – they miss the ‘quiddity’ or ‘just thisness’ of the practice in question (Heritage, 1984: 298–299; ten Have, 2004: 22). For instance, Jane (2012) argues about trolling that it is objectionable, while Phillips (2011) argues about it that it is resistant, but in neither case do we learn very much about how trolling as such is identified and made sensible by trollees, trolls, academics, or anyone else. Rather, the idea of trolling is ‘fixed’ and used straightforwardly as a springboard, to critique appalling misogyny in the first instance, and the vacuity of Facebook memorial pages and their relation to the 24-hour news cycle in the second. In neither case is the logic used to move ‘up’ to these imputations presented or accounted for.

Ethnomethodology thus draws a distinction between topic and resource:

Beware of confounding the topic of one’s studies with the resources for studying them … sociologists have naïvely taken for granted the self-same skills, practices and suppositions as members of the society. The confounding has the consequence … of rendering sociology a folk discipline: sociology becomes naïvely ensnared in the very practices it ought to be describing (Pollner, 1987: xi-xii).

‘Indigenous’ understandings, which is to say, accounts presented by members (including academics) in the ‘natural attitude’, are not to be taken as resources providing the basis for more elaborate theoretical explanations of what is ‘really happening’ (for instance, the public sphere, or moral panic, or risk). Rather, they are the topics of inquiry in themselves (Gubrium and Holstein, 1997: 42). This has some implications insofar as it applies to conventional academic work:

Aside from opening up a field of substantive research, the idea of investigating methods has reflexive implications that problematise the division of labor between social scientist and native practitioner. In classic social science investigations, this division of labor often serves to distinguish how the social scientist amasses knowledge from how the natives organise their beliefs in a particular domain. This distinction is one that ethnomethodologists explore rather than adopt (Lynch, 2002: 486).

This cuts both ways: the ethnomethodological orientation implies a radical critique of conventional mainstream work in the social and human sciences, and it also implies that mundane practical reasoning of the everyday variety (such as that conducted and expressed by all parties to the Seven News segment) is itself sociological in character.

Notoriously, ethnomethodology is also ‘indifferent’:

Ethnomethodological studies are not directed to formulating or arguing correctives. They are useless when they are done as ironies … They do not formulate a remedy for practical actions, as if it were being found about practical actions that they are better or worse than they are usually cracked up to be (Garfinkel, 1967: vii).

An ethnomethodological account, therefore, would not presume to say that trolling was good or bad, or that trolling should be defended or condemned, or that ordinary members should be corrected as to their use of or understanding of trolling, or that some other position should be arrived at with respect to it. It would attempt rather to show how competent members might arrive at such positions, and how the reasoning behind such arrival is occasioned, made relevant, and displayed. This is not to imply that ethnomethodology is morally or politically apathetic, or that its analyses cannot or do not have moral or political implications. As Eglin and Hester put it: ‘any proposal for change presupposes a description of what is in need of change, and any description will have been produced by the parties to it with the use of members’ methods’ (2003: 127–128).

One such method ordinary members use to get things done is membership categorisation. Consider the range of categories of persons named in the Seven News segment discussed above. By order of appearance, they are:

  • TV host
  • Online bullies
  • Twitter trolls
  • Online tormentors
  • Trolls
  • Victims
  • People they’ve never met
  • You cunt
  • Twenty-year-old
  • People he doesn’t know
  • Celebrities like Guy Sebastian … and Jack Vidgen
  • Fucking cunt
  • Ugly ass albino Ellen DeGeneres impersonator
  • Trolls’ targets
  • People like Charlotte
  • Sydney truck driver
  • Serial troll
  • Slut
  • Real life you
  • Internet you
  • Fair game
  • Users
  • People like you
  • People targeted by trolls
  • Reasonable people
  • Gen Y
  • A generation
  • The world
  • Anyone who has taken offence

Some of these are demographic categories (‘a generation’, ‘Gen Y’, and ‘twenty-year-old’ as an instance thereof), while some are occupational (‘TV host’, ‘Sydney truck driver’). Some are universal (‘the world’); some (‘internet you’, ‘real life you’, ‘users’) are open to incumbency, while some are indexical and particularised (‘you cunt’). The most compelling categories are not precisely synonymous (as e.g. ‘online bullies’, ‘online tormentors’, and ‘Twitter trolls’ would initially appear to be), rather, they are ‘transforms’ of each other, and serve to co-elaborate each other in relation to their various predicated activities. For instance, ‘victims’, ‘people targeted by trolls’, and ‘people he doesn’t know’ do this – with respect to each other, with respect to a troll, and thereby, with respect to a moral definition of trolling as problematic. To conduct categorisation is to assemble morally consequential descriptions. It is out of this relational co-elaboration or transforming that the ‘socio-logical’ and moral fabric of the segment is woven.

These, then, are membership categorisations:

commonsense units of identification for referring to people in speech. These membership categorisations (along with the rules for their application) are conventionally grouped together into membership categorisation devices (M.C.D.s). So, for example, the M.C.D. ‘gender’ collects together the categorisations ‘male’ and ‘female’, the M.C.D. ‘family’ collecting together the categories ‘mother’, ‘father’, ‘daughter’, ‘son’ etc (Wowk, 1984: 76).

Through this set, we have a membership categorisation device, which we could call ‘parties to a trolling’. On the one hand is the larger set, the trollees (‘victims’, ‘trolls’ targets’ and so on), of which a particular subset (‘celebrities like Guy Sebastian and Jack Vidgen’, ‘people like Charlotte’) is newsworthy. This is a category generalised ‘up’ out of the particular. On the other are the trolls, who are also composed from a larger set, ‘a generation’ in fact, where this has certain implications for the future, given they possess ‘the ability to scream whatever they want to the world with complete anonymity and often no repercussions’ (02:35). Trollee and troll comprise a relational pair. The exchanges occurring between these two groups are to be assessed by an assumed audience (‘reasonable people’, ‘the world’, ‘anyone who has taken offence’), which likely extends to the viewers of the segment. These categories are ‘available to anyone to see’. They are recognisable and as it were ‘canonical’, and not unlike other membership categorisation devices we know, for instance, ‘parties to an offence’ in crime reporting, where there is an offender, a victim, a witness and so on (Watson, 1997: 83).

The idea is not that we are all somehow walking around with an indefinite number of such devices in our heads, waiting to have them activated. What will stand as a category and what will stand as a relation and thereby a device is a local members’ matter, artfully accomplished and parsed:

categories do not reflect pre-discursive entities that are ‘out there somewhere’ and which members use to make sense of what is happening. Rather, what constitutes a category, and the predicates (i.e., expectable features, characteristics, behaviours, states of mind etc.) that accompany categories, are locally produced and are designed to ‘do’ social actions … there is nothing a priori about the association of certain predicates with certain categories (Clifton, 2009: 3).

Categories are ‘inference-rich’, and it is in the unfolding of events that particular inferences and relations are topicalised and rendered relevant. This is done with extraordinary economy. Even from the title of the segment, ‘Charlotte Dawson fights back against trolls’, we can understand that ‘parties to a trolling’ is coming into effect, and that Dawson occupies the trollee role. We can work inferentially back to her categorial incumbency from the activity she is engaged in: her ‘fighting back’ allows us to understand she has been attacked at some previous point by trolls, and even allows us to understand that this previous attack was somehow both public and (until now) obscured from our view. This, after all, must be at least one of the reasons why the fighting back is of current interest: because events on Twitter are being imbued with a novel moral character by being presented (and rectified) in another medium.

These three principals to the drama, then, as members of their respective categories and as a collective in the device ‘parties to a trolling’, have category-bound predicates, agencies, and activities attributed to them, for which they are (and can be shown and held to be) responsible. The establishment of relations between categories, and the moral development and inflection of categories through their various predicates, is the means by which the segment does its work (as when ‘troll’ is modified by ‘serial’). Consider again the opening statement: ‘Charlotte Dawson meets one of the trolls who sent her abusive messages on Twitter while she recovered from a suicide attempt’. Charlotte Dawson is an individual. The troll is a representative of a larger group. ‘Trollness’ is articulated through the predicate ‘sending abusive messages’, where the recipient being predicated as ‘recovering from a suicide attempt’ compounds the abuse. That trolling should be held to be morally repugnant is evidenced not only in this framing, or by the offending tweets, but also in predication of Dawson’s response, which is to ‘expose’. Predicate and category are then conflated: ‘It’s just exposing the nasty. It’s not bullying you, it’s exposing you for what you are’ (01:58). ‘What you are’ here, what trolls are, is ‘nasty’ (and asserting as much, Dawson is shown pointing out, is not bullying).

Of course, everyone is a member of many categories at any particular point in time. In addition to the categories she occupied in the segment, Charlotte Dawson was also a resident of Sydney, a daughter, a person raised in New Zealand, a reality TV personality and so on. But part of the dynamic drive of the segment is around disjunctive incumbency. That is to say, incumbents of the categories ‘TV host’, ‘people like Charlotte’, ‘celebrities like Guy Sebastian and Jack Vidgen’ are simultaneously occupying the categories ‘victims’ and ‘people targeted by trolls’. They become so by being predicated as ‘sluts’, ‘cunts’, and ‘ugly ass albino Ellen DeGeneres impersonators’. It is even suggested that they might be ‘fair game’ for this. That such things could happen is clearly indicative of a problem in the world warranting attention for Seven News; it is what Baker (using as an example a headline description of ‘Killer Sheep’) refers to as ‘category-predicate anomaly’ (2000: 103). But this is compounded by the incumbencies held by the trolls: an ‘online tormentor’ can be shown to be a ‘twenty-year-old’ or a ‘Sydney truck driver’ (as opposed to say, ‘university student’, ‘loved son’, ‘forthright media critic’ or whatever other categories the trolls could be described as occupying).

These categories then are imbued differentially with status and other moral attributes and hierarchically organised, such that they are embedded in the very structure of the interaction we are shown in the Seven News segment. Jodie Speers, at the point of her intervention in defence of Dawson, acts as aligned and critical witness to the exchange between Dawson and Shields. The other witness to the interaction, who does not speak, would appear to be Shields’s employer (who might conceivably have had an interest in the twenty-two thousand now apparently aware of the location of his business and the context of this awareness). In the very articulation of what the entitlements of parties to a trolling are, it is their differential categorial status which gives warrant for Charlotte Dawson to speak over Caspian Shields, as it is differential status which renders salient a particular topicality to Ian Cameron’s membership of the occupation ‘Sydney truck driver’. It is also their differential status which justifies Dawson’s summation having, in moral terms, the ‘last word’.

Then, of course, there are the typifications which instantiate and thereby define trolling in the segment: ‘you cunt’ (in Dawson’s ventriloquised ‘real life’ trolling of Jordan McGuire), ‘fucking cunt’, ‘ugly ass albino Ellen DeGeneres impersonator’, and ‘slut’. These are also categories. They are, effectively, derogatory categories of the device: ‘gender’. Within the segment, their public application is the predicate allowing for the adequacy of the categorisation of trolls as such. In this instance, this is how the work of trolling is done and recognised as being done.

Where Dawson was presented as comfortably ‘giving as good as she got’ in this respect, the potential scope for problematising this use of gender in these tweets was passed over by Seven News. The question for them was essentially the question of motive: why troll? Why troll ‘people like Charlotte’? The search for motive is expressed and given normative shape across the two axes of particularisation and categorisation, where specific trollees are identified and individualised (Guy Sebastian, Jack Vidgen), and yet where particular individual trolls can be made to stand for their collective category: trolls, who are in turn of ‘a generation’. Trolls are an undifferentiated mass; celebrities are uncommon individuals. Where ‘victim’ is an incongruous, troubling, and unseemly category for ‘celebrities like Guy Sebastian’, ‘twenty-year-old’ or ‘Sydney truck driver’ are incidental incumbencies for trolls.

The interesting possibility for the viewer, and for us, is that these categories are precisely those which constitute ‘reasonable people’ and ‘the world’. As Rapley points out, ‘ambiguity is a central resource for both speakers and analysts’ (2012: 325). What will be the social and moral consequences, Seven News allow us to ponder, if, when ‘reasonable people’ have the means of publicly expressing their views, they choose to utilise these means, and thereby contribute to public discourse, with such actions as calling Jack Vidgen an ‘ugly ass albino Ellen DeGeneres impersonator’?

We are in a position now to take stock, and consider what it is that membership categorisation analysis has to offer as opposed to the more conventional account presented in the first part of the paper.

It should be evident from the above that talk about trolling is a way of describing an activity in the world which gives a certain moral shape to the world. Any analysis of the ascription of trolling does membership categorisation, as does any invocation of trolling. We can identify the means of conducting the conventional form of interpretation in the first section of this paper through specifying the categories through which it is conducted in the second. That is to say, the play of categories, predicates and relations constituting the device ‘parties to a trolling’ is anterior, tacit to, and mobilised in any account of what trolling ‘really means’. Just as trolling is produced meaningfully in the segment, the work of justifying an argument that the segment shows how the public debate about trolling in Australia is ‘really about’ the public sphere, or moral panic, or risk (or indeed something else), lies in the use of this device. To say with reference to the Seven News segment that it is about a perceived threat to norms of deliberation in the public sphere is to point, for example, to how trolls are predicated as having ‘the ability to scream whatever they want to the world with complete anonymity and often no repercussions’. To say with reference to the segment that it is about moral panic is to point, for example, at the transformation of ‘troll’ into ‘user’, and the implication that any user is a potential troll. To say that it is about discourses of risk is to point, for example, to the announcement directed at ‘PEOPLE TARGETED BY TROLLS’, in a context where such targeting has been associated with ‘at risk’ populations (where suicidality as a state is being consistently and repeatedly predicated to such people).

Talking about trolling is not a neutral ‘capture’ of the world, it is part and parcel of that world and a way of shaping the world as well. This is how we accountably talk the world into existence. Disregarding this, or arguing otherwise, is ‘very much like complaining that if the walls of a building were gotten out of the way one could see better what was keeping the roof up’ (Garfinkel, 1967: 22). Naming behaviour as trolling is not deploying an objective and stable descriptor to convey a meaning about a social practice which is somehow itself before we get to it, it is a means of producing social practice itself as meaningful. This goes for any ordinary members ‘in the wild’ whose accounts are available to us, including accounts which seek to explain what trolling ‘really means’ in conventional academic terms. In any instance, constitutive categories will be invoked, topicalised, and assigned in order to get that work done. As networked interactional phenomena, trolling and discussion of it are notable in that, occurring as they do and where they do, they are amenable to such scrutiny and analysis.

In turn, how we interpret the segment, other instances of trolling, and the issue of trolling at large, depends contingently on our incumbency of or affiliation with various categories: men and women, social conservatives, trolls, recipients of verbal abuse, proponents of freedom of speech, regular Reddit readers, mental health survivors, members of ‘Gen Y’ or whatever. Without even a rudimentary grasp of how these kinds of description are invoked, applied, and rendered salient and sensible, we have no means of determining what is happening when behaviour is categorised as trolling, whether such categorisation is appropriate, or perhaps most importantly, what that categorisation is being used to effect. If we want to understand what trolling is and what people are using the category to do (for example, what kinds of changes in the world the category might be used to advocate for), it seems a good idea to attend to the work that we and other members put in to producing it as a category.

Biographical note

Andrew Whelan is a Senior Lecturer in Sociology at the University of Wollongong in New South Wales, Australia. He has research interests in subculture, popular music, digital culture, social interaction and organisation and social theory. He is co-editor of Zombies in the Academy: Living Death in Higher Education (2013), author of Breakcore: Identity and Interaction on Peer-to-Peer (2008), and has contributions in the edited collections Being Cultural (2011), Dichotonies (2009), and Cybersounds (2006). His current work addresses talk and discourse around popular music.

References

  • Baker, Carolyn. ‘Locating culture in action: membership categorization in texts and talk’, in Alison Lee and Cate Poynton (eds.) Culture and Text: Discourse and Methodology in Social Research and Cultural Studies (St Leonards, NSW: Allen and Unwin, 2000), 99–113.
  • Benhabib, Seyla. ‘Toward a Deliberative Model of Democratic Legitimacy’, in Seyla Benhabib (ed.) Democracy and Difference: Contesting the Boundaries of the Political (Princeton: Princeton University Press, 1996), 67–94.
  • Clifton, Jonathan. ‘A Membership Categorization Analysis of the Waco Siege: Perpetrator-Victim Identity as a Moral Discrepancy Device for ‘Doing’ Subversion’, Sociological Research Online 14.5 (2009), https://www.socresonline.org.uk/14/5/8.html
  • Coleman, E. Gabriella. ‘Our Weirdness Is Free: The Logic of Anonymous—Online Army, Agent of Chaos, and Seeker of Justice’, triplecanopy 15 (2012), https://canopycanopycanopy.com/15/our_weirdness_is_free
  • Dahlberg, Lincoln. ‘The Internet, Deliberative Democracy, and Power: Radicalizing the Public Sphere’, International Journal of Media and Cultural Politics 3.1 (2007): 47–64.
  • Donath, Judith. ‘Identity and deception in the virtual community’, in in Marc A. Smith and Peter Kollock (eds.) Communities in Cyberspace (London: Routledge, 1999), 27–57.
  • Eglin, Peter and Hester, Stephen. The Montreal Massacre: A Story of Membership Categorization Analysis (Waterloo: Wilfrid Laurier University Press, 2003).
  • Garfinkel, Harold. ‘The Origins of the Term ‘Ethnomethodology’’, in Roy Turner (ed.) Ethnomethodology (Harmondsworth: Penguin, 1974), 15–18.
  • Garfinkel, Harold. Studies in Ethnomethodology (Englewood Cliffs: Prentice-Hall, 1967).
  • Gubrium, Jaber, and James Holstein. The New Language of Qualitative Method (Oxford: Oxford University Press, 1997).
  • Heritage, John. Garfinkel and Ethnomethodology (Cambridge: Polity Press, 1984).
  • Herring, Susan, Job-Sluder, Kirk, Scheckler, Rebecca and Barab, Sasha. ‘Searching for safety online: managing ‘trolling’ in a feminist forum’, The Information Society 18.5 (2002): 371–383.
  • Hester, Stephen, and Eglin, Peter. ‘Membership Categorization Analysis: An Introduction’, in Stephen Hester and Peter Eglin (eds.) Culture in Action: Studies in Membership Categorization Analysis (London: University Press of America, 1997), 1–23.
  • Hier, Sean. ‘Thinking Beyond Moral Panic: Risk, Responsibility, and the Politics of Moralization’, Theoretical Criminology 12.2 (2008): 173–190.
  • Jane, Emma. ‘Your a Ugly, Whorish, Slut’, Feminist Media Studies (2012), https://www.tandfonline.com/doi/abs/10.1080/14680777.2012.741073
  • Lynch, Michael. ‘Ethnomethodology’s Unofficial Journal’, Human Studies 25.4 (2002): 485–494.
  • Máiz, Ramón. ‘The Political Mind and Its Other: Rethinking the Non-Place of Passions in Modern Political Theory’, in Marcos Engelken-Jorge, Pedro Ibarra Güell, and Carmelo Moreno del Río (eds.) Politics and Emotions: The Obama Phenomenon (Wiesbaden: VS Verlag, 2011), 29–70.
  • Merton, Robert. Social Theory and Social Structure (New York: The Free Press, 1957).
  • Mouffe, Chantal. ‘Deliberative Democracy or Agonistic Pluralism?’, Social Research 66.3 (1999): 745–758.
  • Nakamura, Lisa. ‘Glitch Racism: Networks as Actors within Vernacular Internet Theory’, Culture Digitally // Examining Contemporary Cultural Production (2013) https://culturedigitally.org/2013/12/glitch-racism-networks-as-actors-within-vernacular-internet-theory/
  • Pollner, Melvin. Mundane Reason: Reality in Everyday Life and Sociological Discourse (Cambridge: Cambridge University Press, 1987).
  • Rapley, Tim. ‘Order, order: A ‘Modest’ Response to Stokoe’, Discourse Studies 14.3 (2012): 321–328.
  • Schutz, Alfred. ‘Common-Sense and Scientific Interpretation of Human Action’, in Alfred Schutz (ed.), Collected Papers I: The Problem of Social Reality (The Hague: Martinus Nijhoff, 1962), 3–47.
  • Seidman, Steven. ‘Defilement and Disgust: Theorizing the Other’, American Journal of Cultural Sociology 1.1 (2013): 3–25, https://www.palgrave-journals.com/ajcs/journal/vaop/ncurrent/full/ajcs20123a.html
  • Shachaf, Pnina, and Hara, Noriko. ‘Beyond vandalism: Wikipedia trolls’, Journal of Information Science 36.3 (2010): 357–370.
  • ten Have, Paul. Understanding Qualitative Research and Ethnomethodology (London: Sage, 2004).
  • Watson, Rod. ‘The Presentation of Victim and Motive in Discourse: The Case of Police Interrogations and Interviews’, in Max Travers and John Manzo (eds.) Law in Action: Ethnomethodological and Conversation Analytic Approaches to Law (Aldershot: Ashgate, 1997), 77–97.
  • Winch, Peter. The Idea of a Social Science and its Relation to Philosophy (London: Routledge, 1990).
  • Wittgenstein, Ludwig. Philosophical Investigations (Oxford: Blackwell Publishing, 2001).
  • Wowk, Maria. ‘Blame Allocation, Sex and Gender in a Murder Interrogation’, Women’s Studies International Forum 7.1 (1984): 75–82.