Published	15/06/2016
Type	Article
Author(s)	Christine Preiser
Corresponding Authors	Christine Preiser, PhD candidate, Max Planck Institute for Foreign and International Criminal Law
DOA
DOI

Conducting research on nightclub bouncers involves fieldwork with actors who have limited interest in having the details of their work become visible to third parties. Conversely, it is the specific interest of ethnography to make the invisible visible. Thus, the research process is a constant negotiation of two potentially conflicting logics of (in)visibility. Furthermore, it is shaped by potential risks and the requirements of ethical codes. Focussing on the study of an inexperienced researcher ‘entering the field’, this article provides insights into the themes of ‘risks’ and ‘(in)visibility’ and shows how the two were interconnected throughout the whole process of the project, from fieldwork through to writing and publishing. It also shows how the researcher became an overt, but discreet participant observer whose fieldnotes contained intended and unintended blanks. The article suggests that ethnographic data always stays incomplete as the researcher partakes in a balancing act regarding what is revealed and what remains hidden. Intentional blanks help to confirm the researcher’s trustworthiness in the eyes of the research subjects, and so are vital in making such research possible in the first place.

Introduction
What should you do if your code of ethics requires you to keep your participants from unforeseen consequences due to your research,18* while public and academic discourse tell you that your field – nightclub bouncers – is a field full of risks? Similarly, what is the correct approach when important publications within that and related fields suggest that you must put yourself at risk in order to gain valuable insights (Ferrell & Hamm, 1998; Calvey, 2000; Westmarland, 2000; Winlow et al., 2001; Marks, 2004), while others remind you to take your own risks into account (Scarce, 1994) and exercise caution? In such circumstances the task at hand becomes one of negotiating various interests and the logics of the work – on site, at the desk, and in publications.

My doctoral research involved 60 nights of participant observation with bouncers working at three nightclubs in two German cities. I experienced the action-packed side of the night-time economy – and the boredom of endless hours outside these places of amusement. I wrote 500 pages of fieldnotes. Yet arguably the tensions inherent in the work are at their starkest now, as I reflect on the experience of my research subjects and seek to reap the fruits of my labour through the process of publishing. Two antagonistic logics have to be balanced during this process. On the one hand are the interests of the bouncers who allowed me to access to their night shifts but sometimes found themselves conflicting with the law and, so, understandably wished to stay invisible. On the other hand, it is the logic of ethnography to make the invisible visible, and to reveal what is otherwise hidden. It is the nature of our system of research ethics to reduce potential harm for research participants as much as possible – a tool, a filter or a logic gate which helps us negotiate these antagonistic logics. The result of this negotiation is therefore an inevitably partial representation that focuses on some aspects of experience whilst leaving others out; in other words, we are left with a balance of (in)visibility.

This article aims to provide a series of insights into the process involved in creating this balance. Although there is a vivid body of literature on doing ethnography (e.g. Hammersley & Atkinson, 1983; Ferrell & Hamm, 1998; Pawluch et al., 2005; Fetterman, 1989) and on writing ethnographic works (e.g. Clifford & Marcus, 1986; van Maanen, 1988; Emerson et al., 2011), the literature tends to be particularly silent on the interplay between the field under study and the collected data that provide the basis for further analysis and the final representation of the subjects of the ethnography. Thus, the ethnographic process of making the invisible visible tends to stay largely invisible itself. In what follows, the focus will be on the actual process of doing research, and the piece will seek to communicate concrete aspects of my own ethnographic practice. Three sections will follow. In the first, the focus will be on the themes of ‘risks’ and ‘(in)visibility’, and how they are interconnected throughout the entire research process, from the initial forays into fieldwork right through to the publication of results. It will be argued that the interplay between risk and (in)visibility hinges on two basic prerequisites: the need for discretion and the need for trust. The second section will explore how the dimensions of openness, discretion and participation are interconnected at the actual physical site of the ethnography. The final section will show how some of the tensions apparent in the field are recreated and played out within the fieldnotes that are taken. Fieldnotes are often used as illustrative depictions of research, but it is important to ask what is absent from them, and why. Ultimately, the article will show how an individual ‘entering the field’ of research dealt with the challenge of balancing a number of competing logics in the process of ethnography. Furthermore, it will suggest that ethnographic data by its very nature is incomplete, but that in certain cases it is actually desirable for this to be the case. This is because participants will only agree to take part in such studies and develop the trust in the researcher necessary for a successful project if they know that not everything will be revealed. Ultimately, the central balancing act in research of this nature concerns what is revealed and what remains hidden.

Becoming familiar with the logics of the field
At the outset of the research, I was inspired by the literature on methodology and the night-time economy, and a particular set of ethnographic studies that had greatly enthused me (e.g. Simon, 1991; Barley, 2000; Wacquant, 2004; Fassin, 2013). I also had the advice of my supervisors and the code of ethics of the German Association of Sociology to heed. I took the early decision to conduct overt participant observations. This was rooted in pragmatism, as there are only a small number of nightclubs of a very particular type that employ female bouncers, which would have limited my chances of success had I decided to research covertly.19* Employment of a different nature (such as in Rivera, 2010) in nightclubs would have been at the cost of social and spatial proximity to the bouncers. The decision to research overtly also had an ethical dimension, since the code of ethics under which I worked requires explicit informed consent if possible.

Some researchers have stressed that carrying out participant observations when embedded within bouncer communities is a risky endeavour (Calvey, 2000; Winlow et al., 2001; Monaghan, 2002; Sanders, 2005; Rigakos, 2008). Meanwhile others seem barely concerned at all with this issue (Rivera, 2010; Søgaard, 2013; van Liempt & van Aalst, 2016). Despite this difference, the research draws attention to an archetypal bouncer: a 20-45 year old male, hailing from various socio-economic and ethnic backgrounds, with a certain penchant towards using violence which they are given the opportunity to do through their occupation. The prominence of violence in the existing literature is enough to make one quite intimidated.

The manner in which researchers have legitimated their decision to engage in covert participant observations (Calvey, 2000; Winlow et al., 2001; Monaghan, 2002) is perfectly coherent, and yet I have always wondered how these researchers came to the basic assumption that bouncers were a group that would be inaccessible with overt participant observations. With the sense that my way into the field would yield initial insights into its basic logics (Breidenstein et al., 2013) I decided to challenge this assumption. In short, I simply knocked at the doors of three nightclubs and was not rejected at any. Initially, I picked a random nightclub on a random night to have a try. As it turned out, the first bouncer I talked to was not only the head bouncer but also a sociologist.20* Three months later I began my participant observations at that nightclub. I gained access to the second and third nightclubs through bouncers who connected me to the respective teams with a positive recommendation. What was particularly surprising was that I was permitted access to the second nightclub, as I knew that some of the bouncers working at that venue were also local gang members. Even in this case, and despite the perceived risks of the work, it still proved readily possible to gain informed consent in all three nightclubs.

Each of the nightclubs were quite different from one other, but during preliminary discussions and negotiations regarding the research the potential risks were of paramount concern in all three venues. Of particular concern was my risk of receiving an injury and their risk of being held accountable for it. In addition, the venues and the bouncers were concerned that my data might be used against them by third parties, mainly the police. Together, we clarified that they would not be held accountable for my risks and we agreed that I would be prohibited from taking pictures, making videos or tape-recording anything. This went for all three teams, who were all happy for me to take pseudonymised fieldnotes.

Despite this basic agreement, each of the teams had quite different approaches in dealing with their risks with regards to the research. The bouncers in nightclub #1 requested an official letter of informed consent. I offered the same to the bouncers of nightclub #2 and #3 – they laughed fairly hard at me. It was more relevant for them that somebody they trusted vouched for me. Each club employed its own methods of acknowledging risk and ensuring that I was not tempted to cross certain boundaries. For example, the head bouncer of nightclub #2 told me several ‘best of’ stories that sent a very clear – although never explicitly addressed – message during our first personal encounter: ‘We also beat up women, if necessary’. Meanwhile, the bouncers in nightclub #3 had an internal discussion after several night shifts on setting up an approval system of sorts in order to check whether every team member was okay with my presence.

Thus, my route(s) into the field immediately revealed two of its basic logics: Bouncers are familiar with working in plain view of hundreds of revellers (and potential witnesses). But they are simultaneously aware of the fact that some parts of their routine work are potentially of interest to the police, which can have negative consequences. As a result of this, bouncers are extremely keen to remain invisible as individuals whilst executing those parts of their job that might lead to collisions with legal regulations and authorities. Bouncers have a set of tactics designed to minimise the chances of such collisions, such as changing clothes, performing certain activities in dark corners, or quickly disappearing from scenes, which have been described in other studies on bouncers (Lister et al., 2000; Monaghan, 2004) and were also encountered in the course of my own research. I was allowed to ‘look over their shoulders’ and see those aspects of the job that they hide on the condition that I also used cover-up tactics to hide their identities. Simultaneously, my way into the field revealed the importance of trust(worthiness), which was assured hrough both the official consent letters, internal techniques of testing and the demands of gate-keepers such as the head bouncers.

Becoming an overt but discreet participant observer
It quickly became clear that in the process of negotiating (in)visibility, the dimensions of openness, discretion and participation were extremely important and interconnected at the actual physical site of the ethnography. This section will deal with each in turn.

I saw openness – and transparency – as being vital to the success of the project from the outset and I aimed to ensure that my research process engendered these traits as much as possible. To each bouncer I worked with on site, I explained who I was, how I would collect data, how I would pseudonymise the data, and assured them that they would be allowed to read my fieldnotes to check the pseudonymisation. Some were interested in this information whilst others were not, although I always told them anyway. What did seem to spark more interest was when I told the bouncers that it was not my purpose to evaluate the quality of their work. This can be understood in the context of a public discourse on bouncers in Germany that is focused on violence, (organised) crime and discrimination against particular revellers. Most bouncers found my general research interest in social order and space rather abstract and rather “boooring”, but they seemed to value the fact that I was attempting to look beyond the popular subjects such as violence and the selection practices at the entrance to the venue. I kept the bouncers informed of the dissemination of my work such as my participation in workshops, colloquia and conferences, as well as of my progress with articles and my blog in order to give them a sense of my work on their work.

On the other hand, the fact that I was collecting data through fieldnotes alone reduced the possibility of the bouncers being able to actually see when I was collecting data and what kind of data was being gathered. Nevertheless, there seemed to be a general agreement amongst the bouncers that I too was at work – somewhat – during our mutual nightshifts, although there were some blithe jokes such as “We know that you aren’t writing a book. You just enjoy hanging out with us.” Sometimes bouncers were irritated by the amount of time it would take until “the book” would be published, or they confused my fieldnotes with “the book”. The comments of those who read my fieldnotes ranged from “You’re…creepy. How can you remember all this stuff?” to “200 pages?! So you actually did work the whole time.” In moments like these, it became apparent that to most of the bouncers the very notion of ‘research’ remained vague and somewhat abstract, despite my efforts at remaining as open as possible about the process. Yet my openness regarding my role and my fieldnotes – crucially, what they would not contain – was essential in the construction of trust.

Discretion also proved to be of great importance and the subject of a balancing act of sorts. I realised very quickly that there are subtle nuances in the manner that stories are related, depending on the space and the company in which they are told. The following section from my fieldnotes illustrates this:

Mille [bouncer] says, I’ll have an appointment with the cops soon, because of this incident when that guy threw a bottle. Edu [bouncer] looks at him without understanding, I don’t know what you mean. People threw so many bottles at us last year, which one are you talking about? Mille tells him some details, Edu answers, oh yeah… the incident when you smashed up that guy? Mille darts a disgruntled glance at him, of course I did not do that. When Mille tells me about the incident in a dialogue later that night, he confirms that he knocked a guy over. I get the impression that these things are not supposed to be told in public. [X-RAY, night #12]21*

What is important about this in the present context is the difference between a public version of the story in which the bouncer in question “did not do that”, and a privately told version in which the detail that had been denied before is reintegrated into the story. So, the story depends on who is asking whom in the presence or absence of whom. Silences and absences are vital here. This happens mostly in the context of stories that contain potentially incriminating information from a legal perspective. Furthermore, there are topics that bouncers avoided discussing even when asked directly. Both of these traits can be interpreted as a matter of discretion and of keeping their control over stories. This is also mirrored in the experiences of other participant observers when they tried to conduct interviews with their research participants (Simon, 1991; Marks, 2004).

This tendency to discretion forced me to adapt, and I tried to ask questions without giving bouncers the feeling of being interviewed, to observe people without staring, to listen carefully while sometimes looking in another direction, to delay until a more private moment to ask questions. In many cases I simply waited for people to talk about situations by themselves and then asked them inconspicuous questions in order to keep them talking. Simply put, I tried to be as discreet as possible and adjusted my methods to a field in which discretion plays an important role. As we will see in the next section, this sense of discretion fed into the fieldnotes themselves. Nevertheless, while I communicated very openly about the general process of my research, I veiled the moments of doing research. This meant that during my overt participant observations there were undeniably moments in which my own visibility as a researcher was reduced. Once again, I was involved in a balancing act of sorts.

The process of participation itself also raised issues. The bouncers did not know what being a participant and hosting a participant observer would entail, and I did not know what being a participant observer would entail, so we had to find out through practice. Three teams, three times. I stood next to them while they carried out their work, tried to follow them as much as possible, chatted with them, asked questions, felt tired, cold, hungry, and bored with them. Participation also meant to become familiar with each other. It was the quiet, subtle shift from “Will you come back next weekend?” to “Where were you last weekend?” It meant that bouncers got used to having me around and I learnt to find a good position to observe from without blocking the way. It also meant to become a node in the web of relationships on site. Of course, building relationships is not a single-sided process (Hammersley & Atkinson, 1983). Research participants are observing, too, especially when observations are an important part of their business. Bouncers also got to know me, be it on a professional level – How are her reactions in certain situations? Is she trustworthy? – or on a personal level. It was fundamental to the success of the project to make myself visible, to give personal insights, to share thoughts, and to accept mistakes and jokes at my expense. The bouncers’ comments about me and aspects of my life were sometimes closer to the truth than I would have liked and they clearly showed that I was not the only observer on site. Becoming a trusted part of the network of relationships also required me to keep my distance in some moments, to stay behind a little in heated situations so that nobody would need to take care of my safety. I also distanced myself if bouncers wanted me to take a position in internal conflicts. And I firmly refused to give information in some cases: Two teams worked in the same city and they asked questions about each other, so I told them that I would not be able to answer their questions due to data protection. All of this was fundamental to my role on site.

However, the fact that I had clearly communicated the purpose of my research and positioned myself as somebody with a somewhat detached professional interest provided me with some advantages. Whereas the few female bouncers I met during my research had to struggle in the predominantly male teams (see also O’Brien et al., 2008), I enjoyed the privilege of fools, the privilege of not needing to fit in as a colleague (Marks, 2004). I kept Barley’s (2000) ironical pragmatism in mind; the researcher would most likely be seen as an eccentric coot, and I expected to and accepted that I would feel embarrassed, clumsy or ‘uncool’ on a regular basis (which I did). I did not aim to establish a particular role but rather to simply establish myself as trustworthy. Taking this approach helped me to relax and become the “trusted outsider” (Bucerius, 2013). And, unlike in Bucerius’ case, my gender was hardly made a subject of discussion, especially in the all-male teams.22* I tended to be addressed as an observer at work. Sometimes this was in the sense of an extra pair of eyes that might provide additional information, sometimes it was in the sense of an extra pair of ears that could be fed with all the stories everybody else on site was tired of hearing, and sometimes it was in the sense of providing an external perspective on one’s work.

In hindsight, one of the most difficult situations occurred when the question was raised by some bouncers in nightclub #3 as to whether I had the potential to be an indiscreet and thus harmful observer. I had already spent several nightshifts with some bouncers at that nightclub and had showed up for another night. Unfortunately, no bouncer on that particular shift knew me in person, even though some had already heard of me. This caused some confusion, so one of the bouncers asked me to name him all of his colleagues who I had already met in person. Then he left the entrance area, whipping his cell phone out. Shortly afterwards:

Demir [bouncer] comes back and says loudly, she’s alright. He grins at me, sorry, but I had to counter-check this, not that you’re a cop in civvies. He goes back on his position and seems to receive a message shortly after, reads it, nods again and turns around to me with a grin, JC [bouncer] says hello to you. No offence, but you could be a cop or someone from the press. His phone will ring several times in the course of the night and it will always be bouncers who know me already. When Jan [bouncer] is on the phone, he just says, the issue is over, Jan! Thanks! and hangs up. He explains to me, I rang everybody you named. [X-RAY, night #4]

It probably took only a few minutes until Demir came back, but they seemed endless to me. I felt deeply uncomfortable and embarrassed and tried to console myself with gallows humour: “Well… deal with it. At least this is interesting data that came into being right now…” Then, in a stroke of bad luck, it was reported that a female agent’s cover had been blown in another city and the case gained some media coverage. Despite this development and the events of [night #4], the team decided to accept me as a participant observer. Still, this resulted in weeks of comments and jokes about me as “cop in civvies” which were meant to tease me but also to remind me that although I was the observer, I was also being closely observed. I emphasized several times that I would leave immediately if somebody ever felt uncomfortable with me, ensuring that the ball was in their court, but nobody took me up on the offer. So, I laid down my arms and dealt with the jokes.

The fact that several other members of the team had already spent quite an amount of time with me and had come to trust me certainly helped as I could rely on those relationships that had been built during the nights before. I also considered my openness and the strength of my relationships with the teams as a crucial part of my self-protection. I was sure that my network outside the field would not be helpful in the case of threat, but that there were always bouncers on site who would put in a good word for me. Furthermore, I assumed that those bouncers who might have the potential to become a threat would either not agree to take part in the first place, or that they were less likely to become a threat if they had been given the opportunity to consent to being part of the research project. As my analysis has shown (Preiser, 2016), the feeling of being tricked is one of the main factors in bouncers turning to threat and violence, so my openness was an appropriate choice to help prevent this feeling.

About blanks in fieldnotes
Writing is always a process that involves decisions on what to include and exclude (Clifford, 1990; Marks, 2004) and protecting research participants is – among other things (van Maanen, 1988; Emerson et al., 2011) – one of the main factors that shape such decisions in this type of work. The main concern of the bouncers was that my fieldnotes could fall into the hands of the police or others who could do them harm of some sort, so I used this concern as a ‘vanishing point’ of sorts while writing the fieldnotes. Consequently, I pseudonymised all places and persons immediately while writing and did not include information that was a part of criminal proceedings, whether settled or ongoing. The quoted episode in the above section is an example of this. Whilst including an outline of the incident in my fieldnotes, I excluded all details regarding the forthcoming criminal proceedings as well as any details that might have affected the process and outcome of those proceedings. I wanted to avoid the possibility that somebody may be able to decode my pseudonymisation. Thus, I deliberately left blanks in my fieldnotes to minimize the risk for the research participants, which itself is an acknowledgement that simply partaking in this research carried an element of risk for the bouncers.

The following two episodes also contain blanks:

Arthur [bouncer] and I stay for quite a while outside in front of the door and talk about ideas for the future, our biographies, concepts of lifestyles, gender equality and having children. [DYSTOPIA, night #9]

Mille, Edu [bouncers] and I are sitting around the mushroom heater, Nando [bouncer] is leaning against the counter, his arms folded on the counter, his chin on his arms. Mille and I had been talking about books for a while, then the conversation switched to relationships and became pretty private because of a question Edu had asked Mille. We talk in the circle – mainly Mille and me, spiked with comments of the other two bouncers – about relationships, experiences, failure, and expectations. The till is closed in the meantime and Edu sits behind the counter, but the conversation is not disturbed by this interruption. We talk about an hour like this, sometimes guests leave, or two, three new guests arrive, that’s it. Finally, the conversation returns to jocularity. [X-RAY, evening #20]

The fieldnotes in both examples provide details on the protagonists involved, the spatial positions of the protagonists, the context of what is happening simultaneously (in the second example), the fact that nothing else seems to attract the bouncers’ and the researcher’s attention at this point, and the topics the protagonists are talking about. It is also possible to gather something about the nature of the relationships between the protagonists as they are involved in conversation about intimate subjects. Yet at the same time, the fieldnotes stay silent on the details of the conversations and the personal opinions and experiences of the persons involved. These conversations are both part of building relationships and manifestations of those relationships. Publications on bouncers have tended to focus on the action-packed aspects of their nightly routines (e.g. Winlow et al., 2001; Monaghan, 2002; Hobbs et al., 2003; Calvey, 2008; O’Brien et al., 2008; Preiser, 2016), but a lot of hours “working the door” have the distinction of being uneventful and boring (Sanders, 2005), such as in the two episodes above. It is these hours that leave room for extended, sometimes very intimate conversations, either between the research participants or between research participants and the researcher (Marks, 2004).

Over time, I learned much about each team. This was not only regarding their routines and working practices, but also some potentially sensitive details about indictable incidents, personal information, job-related information, informal work hierarchies and conflicts within the team, and information from other spheres of life; love, family, sorrows, hobbies and problems at the daytime job (see also Marks, 2004). The fieldnotes aimed to show the relationships and the topics that were covered in conversation, while at the same time handling the private information with which I had been entrusted, in a sensitive manner. Thus, the fieldnotes are highly detailed on some aspects of the fieldwork, whilst remaining vague on others. While the blanks in the episode quoted in the previous section exist to protect persons as legal subjects, the above two episodes are redacted to protect people’s privacy more broadly. In addition, the blanks are also there to be recognized by the bouncers, to further establish my trustworthiness even after the fieldwork on site and to show that their credit of trust was justified. On occasion this led to confusion: One bouncer who had read my fieldnotes offered to fill in some of the intended blanks until I explained their purpose.

There are also blanks of another type within the fieldnotes as the next example shows:

I had been roaming with Erik [bouncer] in front of the door, we stayed outside for a short time and then I follow him back inside. Simon [bouncer] paces up and down in front of the counter. I exchange a smile with Pete [bouncer] and am distracted for a moment. As I turn back, Erik is gone, he’s in front of the 361° [neighbouring nightclub] where a jam has formed and two men have been brought down to the floor. One of them is lying on his belly and is held down by one of the 361°’s bouncers, the other one is lying on the side, while another 361° bouncer has one knee on his side. [DYSTOPIA, night #10]

This episode shows how quickly situations can seemingly change from the routine and normal to red alert. However, this is always from the point of view of the observer, who has a necessarily partial view of matters. The speed of the incident in this example could have been a real phenomenon, or it could simply have been that the observer had failed to notice that something was going on until it had boiled over. Thus, the fieldnotes depict a reading of the incident, but not the reading of the incident. In many regards they remain frustratingly vague. The researcher (even having being witness to the incident) and the readers of any subsequent publications are left with many open questions based on the lack of information: Who are the two men? How and when did Erik take notice of the incident? When and why did the incident start? Who brought the two men down to the floor and how? Does Erik even know much more than the observer? On the other hand, the fieldnotes provide fairly detailed information: the number and spatial positions of the main protagonists, the imbalance in control and power between the protagonists, the potential cooperation between the bouncers of two different nightclubs.

Of course, it is inevitable that a single observer cannot capture every detail, and that their perspective is necessarily a partial one. This inevitability was however exacerbated by the conditions on site and the fluid and often simultaneous nature of events and number of protagonists involved. In many cases I joined an encounter after it had already begun because I had followed another bouncer somewhere else. I could not clearly hear what was being said because of the noisy environment. Other people blocked my line of sight or attracted my attention. Persons disappeared from the scene and came back. After some time I realized that bouncers dealt with the same problems. This became apparent when bouncers needed to orient within the venue, discussed incidents afterwards, and exchanged ideas and interpretations of how and why a situation had started and who was implicated. Thus, my own difficulties mirrored those of the field I was studying, even though many of the bouncers were definitely better than me at making sense of the jumbled environment.

Thus, unlike the earlier episodes referred to, the blanks in the previous episode are unintended, and result in large measure from essential features of the fieldsite as well as the partial perspective of a researcher working alone. This means that certain things remained invisible in the fieldnotes because I simply could not do better. I realized in the course of writing that my memory was unconsciously trying to fill in blanks and it was challenging to leave them as they were. However, the passing of time was an important factor in making the data more complete (Schatzmann & Strauss, 1973), as similar situations would repeat, resulting in similar reactions, even though each incident was unique in its details. Additionally, some situations had an aftermath, be it in actions or in discourse. Time made it possible to depict situations and their course.

Conclusion
It is clear that participant observation is shaped in a constant interplay with the environment of the field under study, an interplay that starts with the first assumptions and first interactions with potential research participants and continues on through to the publication of results. Previous work from various countries (e.g. Calvey, 2000; Winlow et al., 2001; Monaghan, 2002; Hobbs et al., 2003; O’Brien et al., 2008; Rigakos, 2008; Søgaard, 2014; van Liempt & van Aalst, 2016) has shown that a range of quite different approaches to participant observation with bouncers are available. Since I chose to conduct overt participant observation for practical and ethical reasons, the first task was to find out whether this was feasible and under what conditions. To my surprise, many bouncers were happy to let me take part in their professional lives and also opened up on a personal level in the course of time. However, they disapproved of any kind of digital recording, be it photographs, videos or audio recording, and were only comfortable with allowing pseudonymised fieldnotes. In other words, bouncers did not want to be clearly identifiable as individual persons (which would be possible with digital recordings), because this could potentially have consequences in the event that they became legal persons. As a result, the sole data source for the project was fieldnotes written from the personal memories of the ethnographer.

At the same time, by adapting to the conditions of the bouncers it was possible to become part of the scene and to collect rich data. As other researchers have shown time and again, relationships of trust play a crucial role in this field (Calvey, 2000; Winlow et al., 2001; Monaghan, 2002; Hobbs et al., 2003). Those who decided to conduct covert participant observations had to negotiate how they would be trustworthy colleagues and ethical researchers, while constantly fearing the risks that accompanied their deception of their research participants. In contrast, by using openness, discretion and participation I was able to largely protect myself from some of the risks that other researchers have faced. Of course this did not save me from the risk of receiving an injury on site, but this was relatively low compared to the risks faced by those who have researched covertly. Indeed, it was part of my trustworthiness as an observer to learn and to know when to better stay behind.

Finding a role and becoming a node in the web of relationships on site was an essential facet of the research. It was just as important to safeguard some of the knowledge that went along with these relationships. This meant that I included intended blanks within my fieldnotes in order to protect research participants as legal persons. Furthermore, intended blanks helped to protect the intimate aspects of these relationships and to reassure the participants that they were not risking exploitation (Marks, 2004). Yet these blanks also brought into sharp relief some other important aspects of the work, such as the necessarily partial view of the researcher. Ultimately, the aim was to ensure that none of the research, including outputs such as this one, could contribute to any negative impacts upon the research participants.

Finally, in negotiating the rather broad requirements of the ethical code of my relevant association, an aspect inherent in all ethnographic studies was revealed: ethnographic data always needs to stay incomplete. Partly this is because the researcher cannot do better, but also because the researcher needs to leave intended blanks in order to protect privacy and ensure that, as far as possible, no harm is inflicted upon participants. A willingness to adapt to the logics of the field and to apply cover-up tactics made it possible to do overt research within it. Only a few people will be able to fill these blanks – and hopefully they will feel satisfied in their trust in me. The entire process – right through to the publication of results – is a finely tuned balancing act that involves bringing many things to light, while leaving others in the dark. That some things must remain in the dark may be seen by some as a weakness in the approach; I would argue the contrary, it is the very thing that makes research such as this possible.

*I would like to thank Dr Jaime Waters, Prof. Didier Fassin, Sheraz Khan, and the two anonymous reviewers for their helpful and supportive comments. Den beteiligten Türstehern vielen Dank und ruhige Schicht.

18* http://www.soziologie.de/de/die-dgs/ethik/ethik-kodex.html

19* Only nightclub #1 employed female bouncers. Nightclub #2 never had a female bouncer in its more than 10 years of existence and nightclub #3 only employed female bouncers for particular events

20* Who told me months later: “I have to admit, I’m a bit jealous that I didn’t come up with the idea of writing a PhD thesis about bouncers by myself. I guess it was just too… obvious.”

21* All places and persons are pseudonymised; X-RAY and DYSTOPIA are the pseudonyms of two nightclubs.

22* A bouncer shrugged his shoulders when we were discussing this: “Come on, bouncers are not a bunch of guys that had been living on a lonely island for months. Bouncers work in nightlife, there’s plenty of women. And you were obviously not interested in us as individual men but in our work and us as a group.”

References

Adler, P. A. and Adler, P. (1987) Membership Roles in Field Research. Newbury Park, Beverly Hills, London, New Delhi: SAGE Publications.

Barley, N. (2000) The innocent anthropologist. Notes from a mud hut. Prospect Heights, Ill.: Waveland Press.

Bucerius, S. M. (2013) Becoming a “Trusted Outsider”: Gender, Ethnicity, and Inequality in Ethnographic Research, Journal of Contemporary Ethnography 42 (6): 690-721.

Breidenstein, G., Hirschauer, S., Kalthoff, H. and Nieswand, B. (2013) Ethnografie. Die Praxis der Feldforschung. Konstanz, München: UVK Verlagsgesellschaft.

Calvey, D. (2000) Getting on the Door and Staying There. A Covert Participant Observational Study on Bouncers. In: G. Lee-Treweek, S. Linkogle (ed.): Danger in the Field. Risk and Ethics in Social Research. London: Routledge: 43-60.

Calvey, D. (2008) The Art and Politics of Covert Research. Doing ‘Situated Ethics’ in the Field, Sociology 42 (5): 905-918.

Clifford, J. (1990) Notes on (Field)notes. In: R. Sanjek (ed.) Fieldnotes: The Makings of Anthropology. Ithaka New York: Cornell University Press: 49-70.

Clifford, J. and Marcus, G. E. (eds.) (1986) Writing Culture: The Poetics and Politics of Ethnography. Berkeley: University of California Press.

Emerson, R. M., Fretz, R. I. and Shaw, L. L. (2011) Writing Ethnographic Fieldnotes. 2. ed. Chicago, London: The University of Chicago Press.

Fassin, D. (2013) Enforcing order. An ethnography of urban policing. Cambridge, UK: Polity Press.

Ferrel, J. and Hamm, M. S. (eds.) (1998) Ethnography at the Edge: Crime, Deviance, and Field Research. Boston: Northeastern University Press.

Fetterman, D. M. (1989) Ethnography: Step by Step. Newbury Park, London, New Delhi: SAGE Publications.

Hammersley, M. and Atkinson, P. (1983) Ethnography: Principles in Practice. London, New York: Tavistock Publications.

Hobbs, D., Hadfield, P., Lister, S. and Winlow, S. (2003) Bouncers: Violence and Governance in the Night-time Economy. Oxford: Oxford University Press.

Hobbs, D., O’Brien, K. and Westmarland, L. (2007) Connecting the gendered door: women, violence and doorwork. In: The Britsh Journal of Sociology 58 (1): 21-38.

Lister, S., Hobbs, D., Hall, S. and Winlow, S. (2000) Violence in the night-time economy. Bouncers: The reporting, recording and prosecution of assaults, Policing and Society: An International Journal of Research and Policy 10 (4): 383-402.

Marks, M. (2004) Researching Police Transformation. The Ethnographic Imperative, British Journal of Criminology 44 (6): 866-888.

Monaghan, L. F. (2002): Hard men, shop boys and others: embodying competence in a masculinist occupation, The Sociological Review 50 (3): 334-355.

Monaghan, L. F. (2002): Regulating ‘unruly’ bodies: work tasks, conflict and violence in Britain’s night-time economy, British Journal of Sociology 53 (3): 403-429.

Monaghan, L. F. (2004) Doorwork and Legal Risk: Observations from an Embodied Sociology, Social & Legal Studies 13 (4): 453-480.

O’Brien, K., Hobbs, D. and Westmarland, L. (2008) Negotiating Violence and Gender. Security and the Night Time Economy in the UK. In: S. Body-Gendrot, P. Spierenburg (ed.): Violence in Europe. Historical and Contemporary Perspectives: Springer: 161-173.

Pawluch, D., Shaffir, W. and Miall, C. (eds.) (2005) Doing Ethnography: Studying Everyday Life. Toronto: Canadian Scholars’ Press.

Preiser, C. (2016) Gewalt im Arbeitsalltag von Türstehern. In: Equit, C.; Groenemeyer, A.; Schmidt, H. (ed.) Situationen der Gewalt. Weinheim: Beltz: 329-347.

Rigakos, G. S. (2008) Nightclubs: bouncers, risk and the spectacle of consumption. McGill-Queen’s University Press, Montreal and Kingston.

Rivera, L. A. (2010) Status Distinctions in Interaction: Social Selection and Exclusion at the Elite Nightclub, Qualitative Sociology 33 (3): 229-255.

Sanders, B. (2005) In the Club. Ecstasy Use and Supply in a London Nightclub, Sociology 39 (2): 241-258.

Scarce, R. (1994) (No) Trial (But) Tribulations. When Courts and Ethnography Conflict, Journal of Contemporary Ethnography 23 (2): 123-149.

Schatzman, L. and Strauss, A. L. (1973) Field Research: Strategies for a Natural Sociology. Englewood Cliffs, New Jersey: Prentice Hall Inc.

Simon, D. (1991) Homicide. A year on the killing streets. Boston: Houghton Mifflin.

Søgaard, T. F. (2014) Bouncers, Policing and the (In)visibility of Ethnicity in Nightlife Security Governance, Social Inclusion 2 (3): 40-51.

Wacquant, L. J. D. (2004) Body and soul. Notebooks of an apprentice boxer. Oxford, New York: Oxford University Press.

Westmarland, L. (2000): Taking the flak. Operational policing, fear and violence. In: G. Lee-Treweek, S. Linkogle (ed.): Danger in the Field. Risk and Ethics in Social Research. London: Routledge: 26-42.

Winlow, S., Hobbs, D., Lister, S. and Hadfield, P. (2001) Get Ready to Duck: Bouncers and the Realities of Ethnographic Research on Violent Groups, British Journal of Criminology 41 (3): 536-548.