Presenting "pissis girls": Categorisation in a social media video

Abstract In Finland, a social category called “pissis girls” has evolved into a cultural concept, used widely in various discourses concerning young women. In the paper, the empirical focus is on how, in the context of a viral video disseminated in social media, a group of young girls are presented as representatives of this category. The video, called AM IRK, is a short edited clip of an emergent interaction between young men and girls meeting by chance in a park. Drawing on the notions of membership categorisation, cluster of indexes and indexical field, “enoughness” and authentication, the paper investigates how the edited video, through the means of de- and recontextualisation of the emergent interaction evokes the category of the pissis girls in the ways in which it represents young adolescent girls. More specifically, the paper shows how the edited video uses selected emblems and activities that have been interpreted as key indices of the pissis category, such as the use of alcohol and the appearance and vanity of the girls, as semiotic material evocative “enough” to make the category of pissisness salient. Furthermore, in order to categorise someone as an authentic pissis girl, one has to show how the girls deviate from the ones who are not, that is, the “ordinary” or “good” or “appropriate” girls. And, in turn, when some emblems and activities are considered bound to the pissis girl category, the norms concerning girls in general become visible. In the video, the dynamics of categorisation, authentication, and normativity merge.

Presenting "pissis girls": Categorisation in a social media video

Mia Halonen 1 Introduction
In August 2005, an edited video called ÄM IRK, presenting a random encounter of a group of young men and group of adolescent girls in a park in Helsinki, was launched in a chat forum, mIRC, an internet relay chat space.Soon after, the video quickly developed into a veritable media spectacle.The video depicts a group of girls, presenting them as members of a category of so called "pissis girls".In Finland, pissis girls constitute a controversial social category that is associated with superficiality, vanity and morally questionable behaviour.
In the chain of events following its publication, the video and the ways in which it represented the girls became a nexus for sets of norms concerning the limits of appropriate behaviour for modern girls as well as for norms concerning what is taken as "enough" for indexing membership in the pissis category.The reception of the video ranged from hilarious surprise to moral panic.Mostly, in Finnish social media, the video was treated as a basically harmless comic depiction of the ignorance, superficiality and vanity of the girls who, for example, naively and openly expressed their desire to become famous celebrities.At the same time, the Finnish Broadcasting Company, YLE, treated the video as proof of the morally questionable behaviour of modern girls.It also accused the authors of irresponsibility.By dramatically re-editing the video, YLE recontextualised the original video content so that it also served them as an example of how girls can be sexually exploited in the internet and, more generally, of the dangers of the internet for children (https://www.youtube.com/watch?v=bRO_7lVXtzQ).YLE's uptake of the video, only a month after it was originally aired in mIRC, actually contributed to its fame and triggered its development into a viral video.With it, the drama started to develop and expand.
In this article, my focus will be on how the ÄM IRK video was used to present and further nuance the features that are seen to characterize the category of pissis girls.More specifically, the aim of this article is to show how the categorisation of pissis girls is done in a social media video through processes of de-and recontextualisation of emergent interaction by editing.What motivates such an analysis is that, as many of the other articles of this SI also demonstrate, social media now function as flexible and innovative spaces for social classifications.In the following, I will first discuss the theoretical concepts and methods applied in the study.Secondly, I will introduce the cluster of indexes that characterize the social category of pissis girls.Thirdly, I will analyse how categorisation is produced and enforced, but also resisted in the video.

Theoretical and methodological background of categorisation as a social action
The the term pissis girls was coined in the 1990s.At the time of the making of the video in focus in this article, the category was thus already familiar to Finns.The authors of the video could thus rely on shared cultural knowledge when they decided to broadcast the video, as a contribution to the further establishment and representation of the category.
What makes the video interesting is how it highlights social categorization in action.Hence, social categorization, i.e. the social action of classification of a group of people, is the central theoretical concept and analytical tool adopted in this article.In my use of the notion I draw from the ways it has been used in ethnomethodological conversation analysis (see, e.g.Atkinson & Heritage 1984;Goffman 1983) and especially in the work of Harvey Sacks (1992Sacks ( [1964Sacks ( -1965Sacks ( ] & [1970Sacks ( ], 1972)).For Sacks, categories and categorising are central in all (inter)action.According to him, categorisation is the key device for the members of a culture to make meaningful interpretations, "story lines", of the often extremely minute and fragmental pieces of speech and other semiotic signs.This system of inferring he called the (membership) categorisation device (MIR;1992[1964-65], 1972).In the system, all referential entities are connected to each other in a frame of different sets of categories which is activated when some part of the category is presented.In his famous example of a minimal two clause story line "The baby cried, the mommy picked her up", the reference term "baby" activates both the category set of "family" and "mommy" as a member of the category of family, as well as the category of the "stage of life" of "early childhood".In addition to the entities included in the categories, the sets of categories also include category-bound activities, for example, in the case of the baby , the activity of crying.
In order to apply the categorisation device in the analysis of a relatively new category, such as pissis girls, it is essential to explore its membership categories, or its indexical field (Eckert 2008).The indexical field refers to a frame of semiotic emblems associated with a category that are thus capable of indexing it.In Eckert's words (2008: 464), the indexical field is "an embodiment of ideology in linguistic form".Eckert 's notion also draws on Silverstein's (2003) more general theory of indexicalisation process in which, in a chain of interaction(s), a linguistic form evolves from a marker (of some place, a locality) into an index bearing all the ideologically bound associations linked to the marker and its users.These indexical associations create the indexical field.Elsewhere, when discussing the markers associated with pissis girls (e.g.Halonen 2015;Vaattovaara & Halonen in press; Halonen & Leppänen forthcoming), I have described the indexical field as "a cluster of indexes" that, besides linguistic variants and variation, also includes other semiotic features.The cluster is a dynamic and bidirectional system of perceptions: only one or a couple of features are enough to activate the category as relevant when the category is sketched.At the same time, categorising someone as a representative of a particular category directs the forthcoming perceptions towards the features associated with the category.Further, a central device in categorisation is that, to index a particular category, it is not necessary to use a range of indexes, but only enough of them."Enoughness" is thus a key concept in (re)presenting a category.As a notion, "enoughness" has been discussed by Blommaert and Varis (2013).According to them, many categories (which they discuss in terms of "identities") can be claimed by using just "enough" of indexes of the category.Furthermore, as they argue, "enough" really is enough.Hence, in order to create as much authenticity as possible when (re)presenting an identity, it is actually important to present just enough, rather than too many indexes of the identity category.
Enoughness is a helpful analytical unit for describing (re)presentations of a category in a short video in social media.This is because in a video the resources available for sketching a category are limited, and therefore the categorisation is done mainly by editing, by picking up and presenting such semiotics and activities which are considered as authentic enough for the group.At this point it is important to note that in this context, authenticity does not refer to the traditional sociolinguistic view of authenticity or authentication.Within sociolinguistics the concept of authentication has traditionally been been discussed especially in relation to the evaluations concerning the authenticity of informant speakers as representatives of some specific social group (more about the tradition and its critique, see, e.g.Coupland 2003;Bucholtz & Hall 2005).Here, aligning with Eckert (2008), Blommaert and Varis (2013) and Westinen (2014), I refer to authentication as a discursive process whereby a person is constructed as a "genuine" representative of some category or group on the basis of some set of features generally perceived as emblematic ("enough") for the group.In the analysis of the video my focus will thus be on with what resources the girls in the video are presented as authentic representatives of the category pissis girls.

Cluster of indexes of the category pissis girls
In Finland, pissis girls is a social category that has evolved into a cultural concept, used widely in various discourses concerning young women.The term is occasionally also used to refer pejoratively to the behaviour of older women with some modifying attribute, such as mummopissis ('granny pissis').Pissis is a shortened form of the word pissaliisa, freely paraphrased in English as 'Pee-Liz'.The name refers to one of the assumed and perceived features of the girls: the habit of consuming large amounts of alcohol, mostly cider, and consequently peeing in public spaces.Even though, and most probably exactly because of a widespread stereotypical assumption concerning Finnish culture, namely "getting heavily drunk", drinking by young girls is considered in every way very pathetic, sad and worrying.Consequently, for the girls, doing and performing drinking can also be a transgressive action.(For more about the category, see Leppänen 2012;Paumo 2012;Halonen 2015;Vaattovaara & Halonen in press;Halonen & Leppänen forthcoming.)Apart from heavy drinking, in Finnish public discourse pissis girls are associated with the use of heavy make-up and revealing or, alternatively, tomboyish clothing, preferably in pink and black.They are also perceived as "easy", loose and ambivalent in their sexual behaviour and orientation, as well as superficial, vain and pretentious.Furthermore, the category includes features related to verbal behaviour and activities, such as screamingfor fear or joy -, swearing and giggling.All of these are, in fact, perceived as categorybound activities of teenage girls, and when present in a cluster, of pissis girls.Due to these activities, seen as forms of willingly non-serious performances, pissis girls are also perceived as "stupid".In addition to the verbal activities, the indexes of pissisness include the use of specific linguistic features, such as some exclusive vocabulary (see, Halonen 2015) and phonetic features, such as nasality and the sharp and fronted /s/ sound.The /s/ sound, in particular, is often associated with pissis girls, but also more generally with "girliness" or femininity (as in stereotypical perceptions of homosexuals) (Vaattovaara & Halonen in press;cf. in Danish context Pharao, Maegaard, Møller & Kristiansen 2014 and in American English context Levon 2007).In the cluster of the indexes for pissis girls, the system can also work the other way round: the sound /s/ can be heard as sharp and fronted -even when it acoustically is not -when other features associated with the pissis category are present.Figure 1 summarises the indexes in the feature cluster, or indexical field, concerning pissis girls.As we can see in Figure 1, many of the indexes for the pissis category highlight sexuality.In this, they are similar to the kinds of indexes that are associated with various girl cultures around the world, such as, to name a few, Californian Valley girls (Eckert 2008), Mexican (American) Fresas (Mendoza-Denton 2008) and American Betches (Williams 2012).As such, this is not surprising, since, as Eckert has argued (1996Eckert has argued ( , 2002)), normative heterosexuality is one of the most prevailing ideologies of teenage life that girls encounter, as well as the notion that playing with and interest in heterosexuality is a signifier of maturity.In the same way, Skapoulli's description (2009) of ideological gendered categories of teenage girls in Cyprus resonates with how pissis girls are seen in Finnish society.According to her, cultural anxiety about teenage girls is reflected in the ways the girls are perceived as primarily sexual creatures who are normatively regulated, monitored, and evaluated as "good" or "bad" in relation to (often only imagined) sexual behaviour (pp. 87-88).This is also the case in the video in focus: the categorisation of the girls in it, as well as the reactions in social media to the video are all about the (either imagined or experienced) sexuality of the girls.What is particularly interesting here is that social media apppear as one particular arena where this kind of categorizing takes place.
Next, I will discuss what kind of a device an edited video published in social media is for social categorisation.I will then analyse how the authors use the video as a categorization device, both in the emergent interaction that took place when the video was filmed, and in the editing process of the material later.Finally, I will turn to the targets of the categorisation, the girls, and investigate with what kinds of options of agency they are left to act with.

An edited social media video as reseach data
The data for the present study originate from my research on the everyday linguistic practices of sixth graders (12 years in age) in the Helsinki area from 2009 to 2012 (see, Halonen 2012).This work made it clear to me that the category and practices of pissis girls is relevant for the young informants.This research then led me to to study the sociophonetics of /s/ variation (see, e.g.Vaattovaara & Halonen in press) in which, the sharp and fronted /s/, is, as discussed above, could be perceived as a feature typical of pissis girls.After extending the scope of my studies to include young people's social media practices, too, I started collecting different types of social media data concerning pissis girls, such as videos of pissis girls (many of which are pastiches of the ÄM IRK video) and short stories about pissis girls (Halonen & Leppänen forthc.).
Interestingly, despite the fact that, as a social media product, the ÄM IRK video is already fairly old and that the number of data about pissis girls has recently exploded, it is still considered as THE viral pissis girl video.This, in fact, proves that the video was able to capture something essential about the way in which pissis girls are perceived in the Finnish society.After its launch in 2005, the video has had a colourful presence for ten years in various media.After its publication, it started circulating, disappearing and reappearing in other social media, including YouTube, Vimeo and other video sharing sites.For example, in June 2014 it was removed from YouTube, but in the beginning of 2015, it was again available on the site which the original authors' launched for the video in 2012 (http://ihqdaa.kirjiz.net/index_en.php)-after seven years of the small-scale media drama around the video.The video was finally, at least for the time being, removed from internet by its authors in the beginning of February 2015, just before its decennial anniversary.
ÄM IRK is a 2.40 minutes long clip, edited and (in its first published version) also subtitled in Finnish and English.It depicts an interaction between two groups of people, initially strangers to each other, young men and adolescent girls.They meet by chance in a park where the young men are having a bachelor party, and where the girls are hanging around.Before the girls show up, the young men have been in the park all evening, recording their party with two video cameras.The cameras, in fact, have a crucial role in the interaction between the two groups.This is because the video cameras brought along the possibility that the events in the park could later be de-and recontextualised for, at that moment, unknown future audiences.All the participants orient to this possibility by addressing these imaginary audiences by for example commenting on the interaction to the cameras.The fact that the young men had the cameras with them in the first place might even have been the reason why the girls bothered to show up, as they seem to have been eager to become celebrities (see below).
In the following, I will show how the interaction was recontextualised through editing in order to present the girls as representatives of the pissis girl category.An edited video of an unplanned interaction, illustrated here by the ÄM IRK, is a complex object consisting of layers of (inter)action.In this article I will show how editing is the key means with which an emergent physical interaction between the young men and the girls was recontextualised as a presentation of a social category.More specifically, in my discussion of the ways in which fragments of emergent interaction were used in categorisation, I will draw on the analytical practices of en-, de-and recontextualisation suggested by Bauman and Briggs (1990: 72-76).As shown by Leppänen et al. (2014), these practices, together with resemiotisation, are recurrent means in social media for creating new meanings on the basis of appropriating and re-signifying existing semiotic material.
In the entextualisation process an emergent interaction has first to be recognised to be an example of some activity or category, a unit and a movable entity.In the case of the ÄM IRK video, the various fragments of interaction, for example, the question-answer-adjacency pairs in it, have first to be interpreted as pissis girl performances.In decontextualisation, the recorded material is physically extracted from the context of its original emergent situation.Through editing, the performances are then recontextualised as a presentation of pissis girls: the material is cut and reorganised to construct not only an interpretation, but also a new account of "what really happened".and then the recontextualized account is offered and disseminated to new and unknown audiences in social media.(See, Bauman & Briggs 1990;Leppänen et al. 2014.)In the de-and recontextusalisation procees, the local management of interaction, its turn-taking system (Sacks, Schegloff & Jefferson 1974: 710, 725-726), through which participants of a emergent conversation interpet all the action in relation to the action prefacing it and to the action following it, is thus broken.In the case of the ÄM IRK video, the young men who were the agents of the de-and recontextualisation through their shooting and editing of the video, had the power to use the material provided by the original interaction for their own purposes (cf.Bauman & Briggs 1990: 76).One clear purpose like this was to present the girls as pissis girls.For this, they had to choose from the filmed material instances of "enoughness" (Blommaert & Varis 2013, see the discussion above), in other words, recognizable semiotic features that could function as indices of pissis girls, such as signs of vanity, particular features of their appearance and specific category-bound activies, such as screaming and giggling.

Categorisation through editing: Framing the video
In the video, the situation is first described in an introductory text.After the introduction, there is a scene where the girls present themselves by saying their names and their social media aliases.This is followed by a commentary by the authors and seven separate action scenes of the girls' activities which have been edited so that they appear to be telling a chronological story.
As suggested by Wooffit et al. (2013: 103), when studying videos, it is of utmost importance to study them as mediated objects which are, through editing, modified to argue for a set of opinions and ideas, in this case for the categorisation of the girls as pissis.In accordance with this requirement, I have reconstructed the original order of the emergent interaction to find out what is edited in the chronology -in order to be able to analyse how the editing was involved in the categorising.That the video is edited can be seen in the cuts in the video: the scenes have been de-and reordered.To study empirically what it is that they aim to do, I will start by discussing what the authors have claimed the video "is about", explicated in the commenting text frames created in the video, either explicated or implicated in the emergent interaction shown in the video.
My analysis of the original chronology will focus on both linguistic and semiotic cues of the timeline.For example, the fact that a topic can be presented as "known" (e.g. with a definitive pronoun se 'it' preceding the topic; see, Laury 1997) while the same topic is presented as "new" later in the video can function as a linguistic cue for reconstructing the original order of events.Also the placement and clothing (e.g.whether the girls have a hoodie on or off, for example) of the participants can reveal the order of the actions in the park.In comparison to the original interaction, the most important change in the edited video, and one that is relevant for the analysis of video as a categorisation device is that after the introduction, all of the scenes included in the edited video actually took place after what is presented as the conclusion of the video.
The video starts with an introductory comment and a description of the situation (extract 1).We first see a black screen with a white text of (originally is only in Finnish, the paraphrase here is translated by the author).
Extract 1.The framing of the context.
We are having a bachelor party and are preparing food in The Park, when a herd of these girls appear from behind the tree and attack the space.Everybody's like wtf and sure, it was an arranged bachelor party humour show.Because nobody confessed having ordered such a group, gradually we came to believe that they were indeed serious...The video is framed also in the last concluding scene (extract 2).By placing this piece of interaction last in the video, the girls' aspiration to become celebrity gets highlighted: Extract 2. The end -the highlight.

Author1:
Jos The scene shows that the participants are well aware of the possibility that the video will end in social media.By presenting the girls' wish to become celebrities in the end of the video the authors also gives it extra emphasis.Thus framed by the presentation of the girls' relation to the social media in the beginning and the end of the video, it highlights the ridiculousness of the girls.
As the video got a great deal of attention, and various uptakes of it appeared in the media (see section 3.3 below), in 2012 the authors retrospectively explicated their take on the video by launching two other commentary sites, one in Finnish and the other in English.In the English site they state: Video shows a part of a bachelor party of some IRC-users.The last event of the party was making a dinner where the celebrity had to make dinner for the whole group using a trek stove.The party took a place in Kaivopuisto, Helsinki.In the middle of cooking five teenager girls appeared from almost nowhere and took part on the party.Girls acted as stereotypic teenagers [in the Finnish page literally: 'performed being stereotypic Pissis teenagers', author's remark], though in a humorous way, when two DV-cams were recording at the same time.Girls also enlighted these old IRC-users what is IRC-galleria, why it is not the IRC and what is iRC.IRC is of cource ÄM-IRK (finnish pronounciation of mIRC).If that does not sound funny to you, you are not the target group of this video.

and,
Because video was so good and it contained many cliché of many funny things like chatting in IRC-Galleria imagining that it is IRC and teenage culture [on the Finnish page, literally: 'pissisness' (pissismi) instead of teenage culture, author's remark], it would have been sin to not publish this video.
These later framings and commentaries by the authors also show clearly that for them, the video was about the humour that stems from the open vanity of the girls in relation to the practices of social media.Presenting the girls as openly vain in this way is crucial for the purpose of also suggesting that they are pissis girls; this type of ignorance of the interactional functions of social media combined with a keen interest in such a frivolous activity as dowloading pictures of oneself is often seen as highly indexical of pissis girls.

Categorisation in the emergent interaction
In the interaction shown in the video, the categorising of the girls is carried out in an interplay between the interaction and the camera shots and angles, the 'gazes', presented by them (Mulvey 1975).As we will see, the girls are categorised as pissis girls both implicitly by evoking other categories also associated generally with them, by presenting them cinematically with the help of particular camera gazes, and explicitly by using the category term to refer to them.After the textual introduction and the initial presentation of the girls, the authors move on to commenting on, describing and analysing the situation in order to give the audience an idea of what they think is happening in it (Extract 3).
They have a beautiful angst on and Go Sweden patches there, on their (.) bellies.
((The camera shows the girls without their heads, focusing on their bodies and the patches sewed onto the bottom of their short shirts or on the back of their jeans.The last scene captures a patch on the bottom of one the girls.)) In this scene, the girls are categorised as tämmösii yhdeksänkymmentä luvun tuotoksia 'these products from the nineties'.This introduction of the girls already shows that the authors orient to possible (future) audiences outside the immediate interaction.The girls are presented as the topic, labeled as "products", and as something known and familiar to the audiences with the help of the pronoun tämmösii 'like these' as a modifier (Etelämäki 2005).Thus, the authors invite the audiences to recognize the description as already a familiar one to them, too.Furthermore, also the formulation 'these products from the nineties', which leaves the description of the girls implicit and open for interpretation, ,suggests that the authors and the audiences are assumed to share a common ground.
What the authors might be referring to with this description can have at least two possible interpretations.First, they might refer to the girls' actual time of birth: as they are all adolescents, they were born in the 1990s, a time of a severe recession in Finland.The 1990s were also the time when the downshifting of the Finnish welfare state began, which had a strong effect on society, especially the children of the time (Blom 1999).Thus, the authors of the video may be suggesting that as 'products of the decade', the girls have in some way been affected by the harsh realities of the 1990s.Second, the authors might refer to the 1990s, because it was the time when the concept of pissis girls and term became widely known.It was first used in 1994 in a talk show which voiced the contemporary moral panic about the behaviour, and, consequently, the future of girls, a recurrent theme in Finnish media (cf.Aaltonen 2011;Laukkanen & Mulari 2011).
What is also interesting in the decsription of the girls is the use of the term angst, a German word for 'anxious' widely used in other languages (including English), especially when referring to teenage anxiousness.In the context of the dire economic conditions of the 1990s, referred to above, the choice of the word angst to describe the girls can be interpreted to further emphasize the gloomy meanings associated with the decade.After all, the 90s were also a period of the angst embodied, for example, in the grunge music genre and its icon, Nirvana.However, it may also be possible that the angst the girls are described to display can be a euphemism for their intoxication, in particular as the formulation has a modifier, kaunis 'beautiful' and an adverbial päällä 'on', niillä on kaunis angsti päällä 'they have a beautiful angst on', instead of simplier niillä on angsti 'they have angst'.Again, this possible reference to the girls intoxication can thus function as yet another means of presenting them as pissis girls, as drinking alcohol and public drunken behaviour are taken as key indices of the pissis category.
As suggested above, in moving images, the semiotics of the camera gaze is a prominent resource in the construction of the story.Consequently, it also participates in categorisation.In this particular video, the camera gazes include some that focus on features that are considered strereotypically as indexes of pissis girls.For example, these include the tight, shortish and partly pink shirts and patches sewn on the bottom of the girls' shirts and jeans.The camera zooms on these details at the same time as the verbal commentary explains what the girls look like.
The representation of the girls is done here in a way that the feminist film theorist Mulvey (1975: 11) has argued to be typical of films made in a world characterized by sexual imbalance.According to her view, Hollywood films, in particular, are commonly designed as if the events and actions are seen through the male gaze, whereas women are positioned as the passive focus of the male gaze.This is true of the present video as well.From the innumerous angles from which the girls could have been looked at, the most traditional, an erotisising one is chosen.This is illustrated by the following scene, in which the girls are shot without their heads, only as "talking bodies".However, despite this, the girls do not appear to be only naïve foci of the male gaze, but quickly detect the way they are being presented by the young men.
03 Smacky: HEII HEYY 05 Smacky: ÄLÄ KUVAA SEN PERSETTÄ!SIL ON POIKAYSTÄVÄ!DON'T SHOOT HER ASS! SHE HAS A BOYFRIEND! ((While shouting to one of the young men, Smacky jumps in front of Girly seemingly "protecting" her.The camera then starts to shoot the girls as a group.Smacky and Girly are smiling the whole time.)) Here the camera gaze is revealed and protested against.Interestingly, however, the girls' contestation of the male gaze does not draw on the girls' own agency or sovereignty, but on a reference to a boyfriend .As a result, it is also possible to see the whole set-up of this scene as a categorising device.It highlights such indexes of pissis girls as the importanece of good looks and erotic appeal and, consequently, the girls' readiness to place themselves as a focus of the male gaze.
In the next extract (5), the pissis category is explicitly referred to and explained: Extract 5. Bottle of cider.
((leans towards the camera with a bottle of cider in his hand.))Girl11: =MÄ EN OLE PISSIS.*Anna se mulle.*=I'M NOT A PISSIS. *Give it to me.* ((Girl1 grasps the bottle vigorously pretending to take it from one of the young men; then smiles, grins and looks at the man and at the camera.)) Example 5 shows a performance in which one of the authors explicitly stages himself as a pissis girl: using the first person singular personal pronoun, he talks and looks straight to the camera, showing the bottle of cider in his hand.From this position and using the girls' voice, he states that the girls are being pissis girls.Again, the use of alcohol as a pissis index is exploited in the presentation.What one of the girls does after this performance is typical of the girls: as a response, to the young man's performance, she uses an "ambiguous negation" in which she first explicitly denies the claim made of her, ('I'm not a pissis!'), but simultaneously, in contradiction to the negation, does something that proves the claim "right".In this case, she grasps the bottle eagerly whispering 'give it to me', thus, indexing that she actually likes cider.In this way, she ends up reinforcing one of the key indexes of pissis girls.In the next section, I will discuss this kind of action in more detail, alonside with other agentive actions which the girls are able to engage in situations in which their agency in constructing categories of their own is restricted.

Ambiguity as a strategy for resisting categorisation
Even though categorising is action that is always taken in some specific interactional situation and context, once people get categorised as members of some social group, the classification may become a frame in which all their actions are evaluated.Above, we saw how acted in an ambiguous way by simultaneously denying being pissis girls while acting stereotypically like pissis girls are expected to do.On the basis of the analysis so far of the girls' actions and the ways in which they are represented in the video, the pissis could be argued to look like an "omnirelevant" category (Sacks 1992(Sacks [1967]]), in other words, a constantly valid category from which there is no escape.ENDNOTE 1.However, for pissis girls, the use of ambiguous negation may actually offer a means for problematizing their membership in the category.This pattern is also shown in extract (6).Extract 6. G-string.

Author1:
Käytätteks te stringei.Do you wear G-strings.Girl1: No vittu ei.Fuckit no fuck.((Girl1 turns around and shows the top of her G-string by pulling those high from her jeans.)) Here, simultaneously with the explicit denial of using G-strings, Girl1 shows hers -and thus displays her stance as ambivalent.Using such a contradictory strategy in answering can also function as a way to mark the question -often posed by an older, out-group, personas stupid or irrelevant.The particle no ('well') in the beginning of the verbal answer also indexes that either the question or the answer is dispreferred and unexpected (Raevaara 1989).Interestingly, as shown by Mendoza-Denton (1999, 2008), it may be that such a strategy is a more universal one: for Latina girls in California denying and opposing, "conflictive corroboration", can also be a general interactional strategy and stance, a way of being.
An ambiguous denial is not the only way to "do ambiquity" (cf.Sacks 1992Sacks [1970]]) in a strategic sense.What is also typical in this girl group are performances of an ambiguous sexual orientation.As shown in the video, the girls are also doing "being lesbians" -which, on the one hand, is in contrast with the (hetero)sexual desirability associated with pissis girls, but which, on the other hand, again contrastively, only strengthens the stereotype by playing to the heteronormative lesbian fantasy (cf.e.g.Mayne 1995) )) The question 'who do you like' (in extract 7) is constructed as a specific wh-question (Fox & Thompson 2010) which seeks for a one-item answer.It is formulated with the pronoun kenestä 'who' as seeking a specific person.Instead of naming a person, Girl2 chooses to pick a category term, 'girls'.Her choice to use the plural form, also makes relevant the whole category of girls, and not just some specific girl.By these category formulations, she is thus able to construct herself as a "lesbian".
As was seen in Figure 1, lesbian performances can belong to the indexical field pissis girl.
Because pissis girls are associated with sexual attractiveness in general, any kind of talk and performances highlighting sexuality can be included in the cluster of indexes for pissisness.In addition, the formulation tytöt on ihqudaa 'girls are ihqudaa' in the justification of the liking, Girl2 uses a pissis exclusive slang expression for 'lovely' (ihqudaa).In extract 8, the author has already adopted the lesbian orientation performed by the girls in asking about the sexual orientation of one of the girls.The answer is followed by a small kissing performance to further establish the orientation.Similarly to the authors drawing on the indexes "enough" to sketch a stereotype of pissis girls, the girls draw on the indexes they apparently judge to be enough for a lesbian performance, that is explicating the orientation and smacking each other.However, in addition to the mere performing, the scene offered the girls a stage not only to resist for example the oversexualisation of them but to present themselves and their sexuality on their terms.As Leppänen (2008) has pointed out, girls can also use social media to explore themselves as well as resist the various suggested categories.
We thus saw in extracts 7 and 8 there are several performances pointing to "lesbian" sexuality in the short video.When explored in detail, the performances of lesbianism have to be interpreted, however, as shared play involving all the participants, both the men and the girls.Before the extracts 7 and 8, for example, a boyfriend of one of the girls is explicitly mentioned (see extract 4) and discussed.Later the girl performs being a lesbian (see extract 6).Again, I wish to argue, ambiguity can function here as the discursive defence strategy (cf.Kramsch 2012: 498-499) that the girls are left with in a situation where it is possible and, because of the presence of the video camera, even likely that their actions will be filmed and thus potetnially edited and recontextualised and published.Thus, after the emergent situation, they will no longer have any power to influence the ways in which they are represented in it, and the only place and time for the girls to act in their own favor was their actual interaction with the young men in the park.

Discussion
The empirical aim of this article was to demonstrate how a group of young girls are presented as representatives of the category of pissis girls in an edited video.The key means for evoking this perception were argued to be de-and recontextualisation of an emergent interaction, achieved through the editing of the original footage with varied social media audiences as a target.In the editing process, the emblems and activities that have been and can be interpreted as indexing the category were chosen to be presented in the edition.In particuar, the authors relied on a couple of indexes, especially the use of alcohol and the girls' appearance and vanity, as evidence evocative "enough" to activate the category in both the production and reception of the video.In addition to these emblems, the lesbian performances by the girls also could be interpreted to act within the frame of heterosexual fantasies, adding to the perception of general attractiveness and sexual appeal of the girls, emblems also included in the category.In my analysis I also argued that also the male gaze of the camera, that is, the televisual choices made by the young men, could work as means of categorisation.This is because being an object of a gaze is an essential part of being sexually attractive.Furthermore, the fake, lesbian performances also added to the perception of the girls as superficial and ambivalent.These features, too, align with the general understanding of the indexes of the category of pissis girls.
The editing of the video thus highlighted and further enforced the generally known features of the pissis category.These features were thus treated as authentic emblems of the category of pissis girls.Nothing in the presence of or the actions by the girls themselves as such would suggest them being pissis girls.Categorisation was thus an achievement by the authors, a result of the authentication of the girls as representatives of the category, involving modification of the interaction through de-and recontextualisation.For the girls, in turn, ambivalence, the simultaneous practices of denying category membership and aligning with it, served as a strategy of resisting the way in which they are treated as objects of categorising.
The reason why it was possible for both the young men and the girls to evoke the category of the pissis was due to the fact that pissis girls were already a established category associated with girls in the Finnish culture.For the young men, using the category to frame the video to stimulate interest in their video was thus probably useful.Another reason why the video became popular is probablythe fact that it is a video about girls.As such it engages and plays with various intertwining norms concerning girlhood in Finland in the 21 st century.To be able to categorise someone as a pissis girl, one has to show how the girls deviate from the ones who are not pissis girls, the "ordinary" or "good" or "appropriate" girls.And, in turn, when some emblems and activities are considered bound to the pissis girl category, the norms concerning other girls become visible.For example, since pissis girls drink cider, good girls do not, since pissis girls wear revealing clothing, good girls do not.In the video, the dynamics of categorisation, authentication, and normativity merge.
The creation and construction of a category is always done in relation to other categories, it is an interplay between positioning "what is there" and negation "what is not there".
As discussed in the beginning of the article, the reception of the ÄM IRK video ranged from amusement to moral panic.The video has been highly influential in having further established and nourished the cultural notion of pissis girls as a category.In general, categorisation flourishes in social media.Categorising others is a typical activity for humans, and it is frequently used to entertain us, often by ridiculing others (Leppänen & Häkkinen 2012).As Sacks (1992Sacks ( [1964-65]-65]) has argued, when we try to make sense of the world and others, we do not want to know that much who the other people are as individuals, instead, we wish to know what category they represent.In this respect, social media offer us innovative and efficient devices and spaces for this purpose.

ENDNOTES
ENDNOTE 1. Comparable categories could be, for example, "madness".When one gets categorised as mad, the only option they are left with is accepting the label -a denial would only be a stronger proof of the state.(See, e.g.Jäntti 2012: 42;Ussher 2012: 3.)

Figure 1 .
Figure 1.Cluster of indexes of pissis girls.