‘We went from the anonymity of the internet into my private WhatsApp’: Rituals of transition among dating app users in Berlin

This article focusses on the courtship rituals and practices of intimacy among young dating app users, aged between 20 and 33, in Berlin. Dating app users participate in ‘rituals of transition’ as they signal mutual interest and heightened intimacy by moving conversations from dating apps to social media messaging platforms such as WhatsApp. These rituals of transition play a far more prominent role in signalling romantic interest than the matching-mechanisms inherent in the design of dating apps. Drawing on ethnographic data incorporating 36 semistructured interviews and 45 chat interviews across three popular dating apps, Tinder, Bumble and OkCupid, the study finds that users code the apps installed on their smartphones as hosting spheres of varying intimacy. These spheres are substantiated through the infrastructure of notifications on users’ devices. Rather than drastically altering how users communicate across different apps, rituals of transition are a key moment of communication in themselves.


Introduction
Dating apps, such as Tinder (see https://tinder.com) and Bumble (see https://bumble. com), have been the subject of recent scholarship investigating the mediation of courtship rituals and communication practices (among others, see Degim et al., 2015;Duguay, 2017;Duguay et al., 2017), highlighting that apps are utilised successfully for a variety of social objectives, whether seeking relationships, casual sex, companionship or simply entertainment (Timmermans and Caluwe, 2017). Indeed, while often associated in media narratives with casual sex, this is just one facet of the possible interactions between users, with more substantial, or long-lasting intimate encounters a common occurrence (Sumter et al., 2017). Dating app users are adept at utilising dating technology and manipulating design features to meet their own needs and desires, for example, by changing their search settings in order to force the app's cache to refresh, allowing them to see new sets of profiles (David and Cambre, 2016). In their geolocational functionality, dating apps provide an authenticity to connections lacking previously on dating websites, since users are aware of the proximity of potential partners, and their bodies are experienced as being present in shared city space (Bonner-Thompson, 2019;Veel and Thylstrup, 2018).
While it is clear that in the course of arranging a date through a dating app, some interaction must take place via the dating app itself, it is important to note that communication between potential partners is not limited to this platform after the initial match. This article builds on the growing field of research dealing with the social implications of dating apps (among others, see Chan, 2018;Hobbs et al., 2017;Licoppe et al., 2016;Ward, 2016), to address a gap in the literature: the fundamental role non-dating app messaging platforms play in communication practices between dating app users. While scholarship has addressed the fact that dating app users make use of other messaging services to communicate with one another outside of dating apps (MacKee, 2016; Ranzini and Lutz, 2017), there has been no detailed engagement with the ritualistic significance of such practices for courtship. 'Rituals of transition', which refer to the switching of communication from a dating app to a social media messaging platform, were examined through ethnographic research methods for 13 months in 2019 and 2020. These rituals of transition occur primarily prior to, but also occasionally during, or immediately after, a date.
Media and digital devices are woven into the fabric of everyday life, enmeshed in experiences, encounters, practices and spaces (Bareither, 2019a(Bareither, , 2020Bausinger, 1984;De Souza e Silva, 2006;Ito et al., 2005). As such, courtship rituals and intimate relations concerning love, sex and everything in between, are often mediated by digital technologies, whether such a relationship is initiated through a dating app or not (Gershon, 2010). While research investigating dating apps has focussed on the specificities of particular dating app platforms as mediators (among others, see Duguay, 2020;Miles, 2017;Newett et al., 2018;Timmermans and Courtois, 2018), this article focusses on dating apps as part of a polymedia environment and the significance of transitioning to different platforms within this. Madianou and Miller (2012) use the term polymedia to describe the environment of affordances which new media offers, where 'the primary concern shifts from the constraints imposed by each individual medium to an emphasis upon the social, emotional and moral consequences of choosing between those different media ' (p. 169). Indeed, smartphones, due to the variety of applications they may hold, and the variety of affordances they encompass, can be theorised as polymedia environments in themselves (Madianou, 2014). Focussing on users of three popular Berlin dating apps, Tinder, Bumble and OkCupid (see https://www.okcupid.com), the article highlights how for research participants, the act of transitioning away from these apps to a social media messaging service, such as WhatsApp (see https://www.whatsapp.com) or Signal (see https://signal.org), marks a key courtship ritual and highlights mobile phones as offering a multitude of spaces, or spheres, within which dating app users engage with one another in the course of their interactions. The term ritual is utilised, since dating is best perceived as a ritualistic activity, 'entailing multiple actions with underlying meanings' (Jackson et al., 2011: 630). Furthermore, research participants in Berlin all showed an awareness of the courtship rituals inherent in using a dating app, and the practices of signalling romantic or sexual interest (Greer and Buss, 1994), whatever the final outcome of a date.
While Tinder, Bumble and OkCupid all have slightly different features and designs, all three function via a 'card stack' of user profiles. The term card stack is fitting, since the interface is laid out so that the user sees a seemingly infinite stack of profile cards, one on top of the other. Users can swipe the topmost profile either left, to discard it, or right, to like it. If two users like each other, they are connected -they match -and that user becomes available to chat with. For Tinder, a mobile phone number is required to create an account and it is possible to link the resulting profile to various social media, and other platforms, such as Facebook (see https://www.facebook.com), Instagram (see https://www.instagram.com) and Spotify (see https://www.spotify.com/uk/). The interface of Tinder has changed throughout the years since its release; indeed, the version of the app being handled at the outset of the research project was different to the version that was available at the completion of the fieldwork. However, the core tenets of Tinder have always remained the same, reflected in the basic functionality of its design. The app is set out in four key interface points: one section where users can customise their profile and choose how other users see them; one section where users can change the settings of the app and choose who they want to see; one section where users can chat to other users they have matched with; and the central and most important tab, the card stack, known as 'Discovery', where users can browse the profiles of other users. Tinder's chat interface is not too different from other messaging platforms. One can send gifs, links to other social media profiles, and music, but not photos, to prevent unsolicited explicit material.
Bumble is in many ways almost identical to Tinder. Bumble users can create a profile -using either their phone number, Apple ID or Facebook account -filter who they want to see by attributes such as distance, age and gender, swipe through a card stack of profiles and chat with matches. The options of profile customisation are a little more detailed than on Tinder. There are specific options to disclose details such as height, fitness level, religion and so forth, as well as prompts to answer -'My personal hell is . . .' -alongside the short biographical text one can write. The most obvious difference in design between Tinder and Bumble is that whereas on Tinder after a match either user can send a message, on Bumble, when the match is between a man and a woman, the woman has to send the first message to the man.
Unlike Bumble and Tinder, OkCupid was originally a dating website, before rebranding as an app due to the popularity of these new platforms. OkCupid's original premise as a dating website aligned itself with Chan's (2017) discussion of the affordances of dating websites versus apps, where websites offer a more detailed, slower process of partner selection, particularly in their emphasis on extensive textual profiles. OkCupid, in its original dating website form, prompted users to answer a series of multiple-choice questions -'How do you feel about documentaries?', for example -and based on these answers each profile a user encountered was marked with a 'match percentage', which suggested how likely two users were to get along with one another. 'Match percentage' is still a feature on the OkCupid app and sets it apart from other platforms in transparently highlighting predicted compatibility, as opposed to Tinder and Bumble's opaque matchmaking algorithm.
However, while Bumble, OkCupid and Tinder all have slight differences in their interfaces, all three function similarly as swipe-based, geolocational apps designed to facilitate matches based on mutual interest and proximity, with the option of subsequent chat functionality within each app. While it is possible to access many dating app platforms -including Tinder, Bumble and OkCupid -via Internet browsers on desktops and other non-mobile devices, the vast majority of research participants solely used dating apps in the form of applications installed on their smartphones. As such, the article is written primarily in the context of dating apps being utilised on a mobile device. The ethnographic evidence suggests users supplant the matchmaking framework of various dating apps, which is based on an 'emotional architecture' (Wahl-Jorgensen, 2018, where users like one another to signal interest, with their own matchmaking practices centred around the integration of potential partners into the most private communication spheres of their smartphone. The article highlights how such communication spheres are brought into existence through the perceived meanings of particular applications, and through the infrastructure of notifications on mobile devices, which create a hierarchy of access.

Methodology
The data discussed in this article stem from a 13-month ethnographic study into the practices of dating app users in Berlin, their relation to the affordances of various dating apps and the spaces of the digital city. For the purposes of clarity, when I use the term affordances I do so to refer to how 'technological artifacts or platforms privilege, open up, or constrain particular actions and social practices' (McVeigh-Schultz and Baym, 2015: 2). Berlin has a particular reputation for hosting fleeting intimacies and hedonism, often characterised as a place where 'anything goes' (Quest and Neild, 2018), and one which over the past century has been enshrouded in narratives of urban revolution, broken down, rebuilt and always striving for progress, as well as offering a home for 'sexual excess and experimentalism' (Evans, 2011: 3). It is a city with the ideal environment for dating culture to flourish, with over half the population classified as single (Amt für Statistik Berlin-Brandenburg, 2018). Consequently, it is also one of the most popular dating app cities in the world, according to Tinder (Gilligan, 2021). Berlin was selected as the field-site for the ethnography due to these factors, as a fertile ground for research engaging with dating practices and dating apps, alongside my fluency in German, which allowed dynamic interactions with research participants.
The ethnographic research entailed online and offline participant-observation, as well as 36 audio-recorded semi-structured interviews with dating app users, and 45 chat interviews on Tinder, Bumble and OkCupid. While the 36 audio-recorded interviews all lasted between 40 and 90 minutes, the timing of the chat interviews was more dynamic since research participants were not scheduled to a fixed interview time and replied as and when they were available. Chat interviews carried a time commitment of between approximately 10 and 90 minutes, yet it is difficult to gage the exact time spent communicating due to the fact that apart from the shortest of interactions, conversations were usually drawn out over hours, days, weeks and occasionally months. The average length of contact with research participants was a week, with a check-in a month later if the research participants felt they had some new insights to report. The chat interviews followed Kaufmann and Peil's (2019) notion that 'instant messaging fosters an informal style of communication' (p. 10), which can facilitate a more personal dialogue between interviewer and interviewee, outside of the formal setting of a recorded semi-structured interview, advantageous when discussing practices of an intimate nature. The other reason that chat interviews were employed was to allow dating app users to utilise the platforms they were discussing during the interview itself, easily making reference to any specific affordances or experiences associated with these, as a prompt to 'elicit the practices originating from the respective device' (Kaufmann, 2018: 238).
I engaged with dating app users through dating apps, observing profile characteristics and app design, as well as chatting to users about their experiences. Modelling my approach on Costa (2016), who characterises her ethnography as an online and offline 'hanging out', I immersed myself in the everyday life of my field-site, living in the city, making fieldnotes on the role of technology in my daily routines and those of the people I encountered, incorporating online platforms and physical city spaces. As such, due to the personal nature of ethnographic research, and participation in the same spaces and everyday practices as my research participants, this article often takes the form of a firstperson account of the field-site, as the most appropriate method of communicating with the reader.
Interview participants were aged between 20 and 33; 44 of these identifying as women, and 37 as men. As such, the ethnography takes into account dating apps solely through the lens of emerging and young adults, and the conclusions drawn are applicable primarily within this demographic. Since this segment of emerging and young adults makes up the vast majority of dating app users (Iqbal, 2021), it lends itself as the focal point of research and provided the easiest point of entry in terms of accessing research participants. This age group of users all came of age in a world of digital technological proliferation and Internet access. While digital literacy is certainly no longer the domain of the young, it was expected that research participants' everyday lives would be firmly enmeshed in practices spanning digital communication platforms, a point essential for the approach of this article (Bareither, 2020). Furthermore, the demographic aligned itself with the sixth stage of Erikson's (1977) theory of psychosocial development, in that for young adults, major conflicts of self revolve around forming intimate relationships with others, a field navigated by my research participants particularly in their utilisation of dating apps and other social media. As someone who also belonged to the same age group as participants and who had some pre-existing connections within Berlin, conducting research on this demographic allowed me easy access and the ability to embed myself within the community that I was studying.
The research focussed primarily on men searching for women, and women searching for men, on Tinder, Bumble and OkCupid, the three most popular apps encountered in the fieldwork. 1 Out of 81 research participants, 62 searched only for members of the opposite gender, 12 searched for both genders, five men searched solely for men and two women searched solely for women. While there were initially 81 research participants, the focus of the study was on mixed-sex dating relationships, and therefore, the information gained from the seven research participants who were exclusively pursuing samesex dating opportunities was not included in this ethnography. While the observations in this article are focussed on relations between men searching for women, and women searching for men, there were no notable differences regarding the rituals of transition in those research participants who searched for profiles of the same gender as themselves. All research participants for interviews were sourced via snowball sampling through dating apps, Facebook groups, a newsletter and personal contacts in Berlin.
The research was conducted without seeking to create arbitrary boundaries between offline and online experiences. It is not fruitful to seek to establish a distinction between online and offline practices (Miller, 2013), particularly when dealing with research participants who occupy a multiplicity of online and offline spaces simultaneously and switch between different platforms with ease. As Miller et al. (2016) write, 'when the study of the internet began people commonly talked about two worlds: the virtual and the real. By now it is very evident that there is no such distinction -the online is just as real as the offline' (p. 7). This is particularly true when dealing with dating apps, which operate as media integrated into the most intimate elements of everyday life and facilitate tangible and intangible human connections, encompassing love, sex and friendship. To discard such connections as being virtual would be to misapprehend the role apps and digital devices play in users' lives.
Dating apps do not exist in isolation and most users were active across multiple dating apps and other social media. As such, to form a more holistic understanding of the environment of affordances presented by dating apps (Madianou and Miller, 2012), I set up supplementary research profiles on any apps mentioned by research participants that fell outside of my primary research parameters, such as Grindr and Hinge. These apps do not act as a basis for this research; however, being present on these platforms was necessary to understand any references made by research participants, and any drastic deviations in their features. In terms of their affordances, while Hinge and Grindr, for example, do not operate via a swiping mechanism in the same way as Tinder, Bumble and OkCupid, they do function as geolocational dating apps installed on smartphones and as such share distinct similarities, belonging to the same ecosystem of media. I practised 'iterative data collection', which, as Handwerker (2001) writes, entails that one 'collect some data, analyze it, and use your new understanding to help you choose what data next needs collecting' (p. 71). As such, while my focus was on the most popular trio of apps among my research participants, through ongoing evaluation of fieldnotes and interview data I was also able to identify and utilise less frequently mentioned apps. It was over the course of the 13 months of research that Tinder, Bumble and OkCupid emerged as the focal point of study, rather than being designated as such at the outset. To further promote a 'polymediatic' understanding of dating app users in my methodological approach (MacKee, 2016), I was active on the same social media platforms utilised by my research participants, for example, WhatsApp and Instagram, since, as this article will highlight, such platforms played a key role in courtship rituals. Ethical approval was granted by King's College London and pseudonyms have been used to maintain the confidentiality of research participants' experiences. Interviews were conducted in German or English, and, where relevant, have been translated into English for the purposes of this article. Research participants were not remunerated.
Since this article is primarily concerned with the transitions from dating apps to messaging services and social media such as WhatsApp and Instagram, it is pertinent to briefly give an overview of these platforms. The primary and most common point of transition from dating apps was WhatsApp, an app almost ubiquitous in its use by my research participants. WhatsApp is a messaging service functioning via WIFI or mobile data as an app on a user's smartphone, linked to their phone number and the contacts stored on their phone. It allows communication with individuals, but also via group chats, and can be used to send photos, videos, voice messages or make audio and voice calls. Similar messaging services such as Telegram (see https://telegram.org) and Signal offer almost identical features to WhatsApp but were less frequently used among my research participants. Both Signal and Telegram are also linked to a user's phone number and allow the sending of individual and group messages with a variety of different media options, such as photos and videos.
Alongside these apps which catered solely to private channels of communication between individuals and groups, other less common points of transition included apps such as Instagram, Snapchat (see https://www.snapchat.com) and Facebook. These three social media have far more expansive features than WhatsApp. Instagram is a platform that allows the sharing of photos via a newsfeed, or temporary 'stories' -videos or photos that disappear after 24 hours -and is not limited to contacts in one's phone; here one can also follow news outlets, celebrities and other content. Facebook, similarly, operates as a newsfeed through which users can scroll to view content, not limited to one's friends, but also various online publications, public figures and viral videos. Snapchat is a photo sharing service, where content is temporal, one can publicly or privately share images that are erased after a certain period of time, and as with Instagram and Facebook, the app allows users to follow a variety of public figures and organisations. However, while Instagram, Facebook and Snapchat offer more expansive features than WhatsApp, they do also each include a messaging platform within their design, which has affordances very much in line with those of WhatsApp, allowing private communication between groups and individuals in the form of text, videos or images. As such, when users transitioned from a dating app to Instagram, for example, they did so not to utilise this app's expansive feature set in their interaction, but rather for the messaging functionality, which mirrored that of WhatsApp. Indeed, Instagram in particular, was adopted by some research participants primarily as a messaging platform. However, while there were various ways users could communicate outside of a dating app, as this article will illustrate, WhatsApp was the most frequent point of transition for dating app users and conceived of as the most intimate sphere on the smartphone.
Over the 13 months of the ethnography, I engaged with hundreds of people online and offline, investigating the role of dating apps as technology integrated in everyday life (Bareither, 2019b). During the last 3 months of my time living in Berlin, I also utilised dating apps in a personal capacity, to date. Since I belonged to the same age group as research participants and utilised dating apps as a man searching for women, my personal experiences overlapped with the milieu of dating culture that I was studying. These firsthand experiences of dating in Berlin had the side-effect of adding an auto-ethnographic element to the research, albeit this was primarily utilised to understand practices, rituals and converse with users as someone aware of how apps functioned and dates evolved, while I relied on data from research participants to draw conclusions. I did not interview any dates, nor treat these individuals as research participants. Rather, during the time I dated in Berlin, I gained a detailed awareness of my own practices as a dating app user. This was particularly useful in forming a more experiential understanding of dating rituals, and the way smartphones can act as vehicles for intimacy (Vincent and Fortunati, 2009). As such, the article is anchored in an understanding of dating apps and the everyday life of dating app users in Berlin, founded on long-term ethnographic immersion.
Ethnography is a highly subjective and selective form of conducting research, yet it also allows a reflexivity for the researcher and a fluidity in approaching subjects which enables the capture of rich experiential data; a focussed qualitative portrait of research participants, which would be expressed very differently, or indeed completely lost, through a more limiting methodology. More specifically, my methodology aligns itself closely with Race's (2015) 'ethnography of affordances', which he employs to probe dating app use in queer male communities. This framework 'entails the analysis and description of design features and functions of smartphones and interfaces [. . .] to trace how digital devices interfere with, transform or otherwise impact given practices and relations' (p. 500). Such an approach is particularly well suited to dealing with ethnographic work focussing on mediated interactions and allows a relational perspective towards user and technology. As this article draws on fieldwork from an expansive research project, I took fieldnotes during everyday life in the city, but also specifically while using dating apps in different locations, primarily at home. I made note not only of app interfaces, profile presentations and the interactions I had with research participants, but also reflected on my own experiences as a user of dating apps, as well as someone communicating across a whole range of social media and messaging services during my everyday life in Berlin. Interview data, transcripts, fieldnotes, photos, screenshots and social media posts or articles on dating in Berlin that I came across during my fieldwork were all imported into NVivo qualitative data analysis software. Here I coded the data to specific nodes in terms of emerging thematic points of interest and carried out analysis, from which I draw the arguments at the foundation of this article.

Results and discussion
In this section, I will address the results of the ethnographic fieldwork, presenting my findings alongside existing research to aid the discussion. The results are neither presented in chronological order nor categorised by dating app, but rather in terms of prevalence among research participants, aligned with the research themes that emerged during coding in NVivo. As previously addressed, OkCupid, Tinder and Bumble were the most popular dating apps I encountered in Berlin, with at least one of these apps being used by every research participant. Most users had accounts on multiple apps, with varying combinations of the three primary apps at the heart of this article. Tinder was utilised by nearly all research participants and was often cast as the original and founding dating app. For example, of the 36 semi-structured audio-recorded interviews, 32 had used Tinder, 22 had used Bumble and 17 had used OkCupid.

'Suddenly they are inside my private circle'
No matter which dating apps research participants used, the transition from communicating via the messaging function of the dating app to communicating on a self-contained messaging app such as WhatsApp -especially as this often occurred prior to a physical meeting -was seen as a pivotal moment in the development of the relationship and dating ritual between users. Gael, 28, showed an awareness of this ritual to the extent that he had a clearly formulated strategy which he applied in his interactions with dating app matches.
With dating apps, I usually just tell her along the lines of, 'Hey, I hate this app. Let's just switch this to WhatsApp and get to know each other better there'. That's exactly word for word what I use. And usually, yeah, the girls are like, yeah, text me and they like basically just send me their number as soon as possible, because they know, like, they don't want to be on the app for long and they want to meet someone who, who actually asked them out, you know? (Gael) Gael's comments epitomise traditional notions around gendered heteronormative dating rituals where encounters pivot around men taking the initiative (Rose and Frieze, 1989;Rudder, 2014). This standpoint illuminates the importance of securing a potential partner's phone number as a key signal of mutual attraction and the foundation for further intimacy. Such pre-existing gender dynamics constantly interweave themselves into new media (Handyside and Ringrose, 2017). Indeed, while Bumble actively seeks to reverse this practice of men taking the initiative, by only allowing women to send the first message to men, my ethnographic fieldwork, and the work of other scholars (see MacLeod and McArthur, 2019), highlights that after the initial contact the norm prevails, with men primarily asking women for phone numbers and instigating in-person meetings.
Gael enjoyed going out with his group of male friends and meeting women in clubs; he felt the same rules applied, whether he was in a club or on a dating app, getting a phone number and talking on WhatsApp was seen by him as a prelude for a potential date. McVeigh-Schultz and Baym (2015) note how in navigating digital communication services users are prone to 'mapping particular kinds of communication practices to particular platforms within a larger media ecology' (p. 8); however, in the case of Gael, and other dating app users I encountered who showed a similar awareness of the significance of transitioning from dating apps to other communication services, the communication practices themselves did not significantly change upon switching. For Gael, WhatsApp did not change how he communicated with his dating app matches, it simply acted to signal a change in the dynamic between him and his match -one which he interpreted as showing the match had interest in him and appreciated his agency. While there were some variations in affordances, for example, on WhatsApp he could send images, but not on Tinder, these did not impact the primarily text-based communication between Gael and his matches. Rather, it felt to him as if the surroundings and the context had shifted. This was not so much an implicit or subtle socialisation of technology (MacKenzie and Wajcman, 1985), but rather an explicit coding of an emotional dynamic onto a media practice -the act of switching was a symbolic gesture in itself. Sylvia, 28, showed an almost identical mind-set, musing 'I feel like it's the people that I'm not, that I don't have a lot of expectation of seeing them again, that like we don't get off the app'. Sylvia does not feel transitioning away from a dating app is key to scheduling a date in the same way as Gael; however, she does believe the lack of transition implies an unsuccessful date, connoting a lack of interest, and, in a sense, a breaking of the courtship ritual. Indeed, the gender dynamic here must not be understated, for, particularly in my interactions with men searching for women, the process of receiving a phone number was seen as a fundamental rule, and primary challenge, of courtship.
Mehdi, 28, told me 'I just want to get the phone number, that is because it just makes me, like, it's just how I secure myself, you know?'. Jürgen, 30, when questioned how swiftly he asks for a phone number, answered 'pretty quickly, and for me it's the first big hurdle in the process' (my translation from German). Mehdi and Jürgen, similarly to Gael, portray this 'process' as one of the masculine rituals of courtship, abiding by traditional western notions of male agency in dating (Rose and Frieze, 1989). Interestingly, while such an attitude in itself is not particularly surprising, it does bring to light an intriguing perception on the meaning and purpose of dating apps. While within dating apps, a match between two users is coded as signifying mutual attraction, or at the very least, interest, this technological affordance does not truly translate into a successful courting as perceived by the dating app users I met. Dating apps may afford their own system of digital courtship -founded on an 'emotional architecture' of likes and matches (Wahl-Jorgensen, 2018) -but interviews showed users have transposed their pre-existing understandings and practices upon this system, as laid especially clear by Gael who highlights the parallels between asking for a phone number in a club and on a dating app. Indeed, in the experiences shared by users, and my time dating in Berlin, it became apparent that while one can easily accumulate hundreds of matches across various dating apps, only a small fraction of these progress to a conversation, and an even smaller fraction of these lead to an in-person meeting.
Consequently, a match is only the first link in a chain of interactions, mediated across not only dating apps, but a wider sphere of communication services. As touched upon previously, the exchange of phone numbers did not always precede a first date, yet even when it occurred during or after a first date it was viewed through the same lens, as an indication of mutual interest and signal for continued intimate interactions. It was clear across all the dating app users that I interviewed that far more weight was applied to interactions that occurred outside of the confines of the dating app and that the process of matching on an app was not regarded with great significance as a factor in arranging a date -even while from a technological standpoint a match must be seen as the single most important element in enabling a date, since all future interactions rely upon it. One can attribute this to the subtle motivation among dating app users to position themselves as having more agency in determining their romantic partners than the app's technological functionality and algorithm. In this sense one can note an urge to safely distance one's intimate interactions, or more colloquially, one's love life, from the technology that has facilitated it. As such, especially among couples I came into contact with, emphasis was never placed upon the initial match in determining the relationship's success, but rather the courtship practices that followed, with one primary element being the transition away from the dating app. It is this transition which brings potential partners together into a more 'mundane' space, one not explicitly geared to dating, within which to explore their compatibility (Cabañes and Collantes, 2020).
Indeed, one should not understate the significance of users' compartmentalisations of the spaces they perceive to exist on their mobile devices, which contribute significantly to the way the move from a dating app to a different mobile messaging service is experienced. I asked Liza, 29, whether she talks to potential partners for longer periods of time on dating apps.
No, switch to WhatsApp for sure. Like as soon as I feel like this is something, switch to WhatsApp. It's kind of interesting. It's like a, it's a ritual of some sort, like it really, it changes the nature of the interaction. Like it feels different, you know, it becomes this really, like, taking the next step. You know, like now you're part of, like, my, you know, like we went from like the anonymity of the internet into like my private WhatsApp. (Liza) Liza does not differentiate this practice between the four dating apps she uses -Bumble, Tinder, Hinge and OkCupid. Whichever dating app she is utilising, the 'next step' is always the transition to WhatsApp. She not only explicitly highlights the ritualistic nature of this transition, and thus its importance in signalling intimate intentions, but also refers to the spatial segregation manifested across her digital devices. This is a key point in understanding such transitions not only as a ritual, but also as going from 'the anonymity of the internet into like my private WhatsApp'. WhatsApp is a personal messaging service and is often seen as facilitating a private form of communication, which affords a more intimate interaction (Staudacher and Kaiser-Grolimund, 2016). For Liza, WhatsApp is cast as a different sphere of activity on her phone, which does not necessarily mean that how she talks to potential partners changes when communicating via WhatsApp as opposed to Tinder, for example, but rather that the context of the conversation, and, indeed, the relationship itself has shifted. Different apps can be conceptualised as providing different spatial experiences, akin to Liza having a conversation in a barwhich one could liken to Tinder -or in her living room -which one could liken to WhatsApp. In this sense, it is possible to identify Cabañes and Collantes' (2020) notion of the difference between the 'exceptional' spaces of dating apps and the 'mundane' spaces of other social media platforms such as WhatsApp. Every dating app user I interviewed demonstrated an awareness of experiencing different spaces or spheres on their mobile devices, associated with varying degrees of intimacy. I do not mean this in the sense of a 'scalable sociality', as Miller (2016) terms it, for this experience was not a practical categorisation between sharing varying degrees of public or private information across social media, but rather a more abstract emotional attachment to the settings various apps provided for an interaction. Brigitte, 31, who had been using dating apps for years, spoke to me at length on this dynamic in relation to WhatsApp and dating apps.
What I find extremely private, and also a little exhausting, is that they always ask for your phone number so quickly and 'let's meet as soon as possible'. Brigitte showcased a sensibility that slightly contrasted with my other research participants, in that she was a little more wary of her privacy online. However, her understanding of WhatsApp epitomises that of many of the other dating app users I encountered. It is apt that she refers to WhatsApp as a 'sphere' and 'room', for this demonstrates precisely the segmentation of intimacy afforded by the conceptualisation of apps as distinct spaces on one's phone. As such, while there is no doubt that smartphones in themselves are intimate devices, ubiquitous in their user's lives (Goggin, 2011), the experience of these devices is further segmented. Brigitte is typing on the same screen and the same keyboard when she communicates with a potential partner on Tinder or WhatsApp; however, it feels like a different, more intimate, environment to her as a user. To refer to Burrell's (2017) conceptualisation of the Internet as facilitating imagined spaces, these spaces are vividly constructed by many of the dating app users I encountered in their transition between dating apps and messaging applications such as WhatsApp. Such transitions can be cast as negative or positive, one may contrast, for example, Brigitte and Liza's comments; however, they always focus on experiences of heightened intimacy and it is clear, thus, that ideas and practices of intimacy span across platforms (Tang, 2017).
In terms of WhatsApp in particular, Brigitte specifically references that it is this app 'where all the contacts are', tying one of the key affordances of WhatsApp, namely, that it is directly linked to a user's personal contacts stored on their phone, to a feeling of heightened intimacy. The people one communicates with via WhatsApp are not simply profiles one has come across by chance online -one can only send messages to those whose phone number one has saved on one's phone. To be saved as a contact on Brigitte's phone is to have acquired a certain worth to her, an intimacy based on a practice that predates WhatsApp and smartphones -the act of writing down or storing the phone numbers of those one wishes to communicate with frequently. WhatsApp, via the link to the contacts on one's phone, taps into this feeling for Brigitte.
While WhatsApp was the most popular messaging app among the dating app users I encountered in Berlin, there were of course instances where 'the next step' from a dating app was a different form of social media. Anita, 25, seldom gave her phone number to dating app matches before a first date, but was happy to communicate via Instagram, which she felt, as a public-facing social media platform, connoted less intimacy, and indeed, privacy than WhatsApp. She told me of an incident where a dating app match she was going to schedule a date with completely broke off contact after she told him in which area of the city she lived -Charlottenburg -and that she preferred to chat via Instagram rather than WhatsApp. The user told her that Charlottenburg and Instagram were not 'his world', coding her area as elitist -it is a historically wealthy district -and her use of Instagram as signifying vanity and vacuity. While it is also fruitful to muse on the connotations of place in this context, here, solely looking at the perceptions of media platforms, it is interesting to see the transition away from a dating app being imbued with such significance that it may end the interaction altogether. Ben, 33, also attributes importance to the choice of communication service one transitions to from a dating app, for him it is 'mandatory WhatsApp or Instagram. I would say Facebook is okay, but Snapchat, no'. Cycling through a host of social media communication platforms, Ben notes that WhatsApp and Instagram are seen by him as acceptable to transition to from a dating app. Ben is sceptical of Snapchat because it is not for his 'generation' and he feels some people only use Tinder to get more Snapchat followers; he projects an imagined community upon this platform which he does not wish to participate in (Anderson, 2006). While my fieldwork showed users can have strong opinions about the connotations of social media platforms, it must be noted that Anita's experience of a user breaking off contact due to her use of Instagram was an unusual incident, both for her, and also in comparison to the other dating app users I interviewed; however, it acts as an interesting, while slightly absurd, example of the delicate nature of the ritual.
Since WhatsApp was the most common point of transition from dating apps, I will address its conceptualisation as a more intimate sphere than dating apps in detail. WhatsApp is a messaging application that links to a user's phone number, and functions via the contacts stored on a user's phone. It allows communication in the form of primarily text messages, calls, videos, voice messages and photos. It was used in Berlin by most people I encountered, both throughout my fieldwork and in my everyday life in the city -I used it to communicate with friends, dates and research participants. As Brigitte stated, her WhatsApp was filled with 'constant activity', as opposed to her various dating apps, which she would not use continuously throughout the day to the same extent that she used WhatsApp. Thus, in terms of categorising WhatsApp as a more intimate sphere, in a practical sense this was linked to the idea of having more access to a potential partner -being able to communicate with them easily and consistently at all times of the day. As Jürgen, 30, elaborates, You want to get people off the Tinder platform because I think, if you're sitting next to someone, I don't know, you're in a meeting -yes, answering a question on WhatsApp, or writing a message on WhatsApp isn't a problem. But would you really open Tinder? If you open Tinder in a meeting and answer a message, yes that is certainly unprofessional. (My translation from  German) Utilising WhatsApp, rather than Tinder, to interact with a potential partner, allows communication to continue across spaces and scenarios where the presence of a dating app is deemed as inappropriate. In fact, we can see a paradoxical understanding of intimacy here among users. Tinder is cast as signifying too much intimacy, through its coding as a platform for sex and romance, to comfortably utilise it at work -it is 'unprofessional' -while the very fact that one can use WhatsApp freely in any setting -it is not 'unprofessional' -characterises it as a more intimate sphere than Tinder in the context of communicating with other dating app users, since WhatsApp grants potential partners continuous access to each other throughout the day. This access does not mean that users will constantly exchange messages, but rather that they know if they send a message via WhatsApp as opposed to a dating app, the recipient is far more likely to see it. If one views intimacy as existing through 'connections and boundaries', then it is clear that WhatsApp in contrast to Tinder affords less boundaries in the connections mediated through it, and, therefore, is experienced as a more intimate sphere (Sehlikoglu and Zengin, 2015: 22). Carli and Yannick, both 25, a couple who had met on Tinder, told me that 'as soon as we moved over to WhatsApp it was normal texting'. Moving their conversation to WhatsApp normalised their interaction, as they transitioned from an app -or sphere -on their phone geared primarily towards finding a potential partner, to a sphere of communication integrated into a wider array of social activity in their everyday life. WhatsApp was used by them continuously throughout their day, unlike Tinder, and as such afforded them more consistent access to each other, which, in turn, manifested the experience of more intimacy. Laima, 28, made a similar link between WhatsApp and what she felt was a more personal aspect of her life. I probably personally would have asked him [a partner she met on Tinder, who asked for her phone number] sooner or later, because I was always like deleting Tinder once in a while, like very often. Usually when I didn't have enough space on my phone I was deleting firstly Tinder, because who needs it? And I feel like, yeah, and, also, I think WhatsApp is more personal. You use WhatsApp usually to talk to your friends, some people use it even to talk to your family. You sometimes hear when the message pops out. Meanwhile Tinder is usually like some people use it just when they're drunk and it is something very distant. It is something not in your family or friend's circle. I think lots of people want to leave these casual conversations in that box, separated from their life and family and friends. (Laima) Again, Laima uses the recurring metaphor of spheres, rooms and circles to describe how different apps relate to her life. In transitioning from Tinder to WhatsApp, she frees the interaction from what she feels is a segregated area on her phone and allows her potential partner to occupy a place among her friends and family. This is true not only abstractly in the way that she projects certain emotive experiences onto the apps that she uses -categorising communication channels (Madianou and Miller, 2012) -but also in a practical sense; she is more likely to delete Tinder than WhatsApp and as such there is less likelihood for a connection to break off. Furthermore, Laima refers to the fact that on WhatsApp, unlike Tinder, she can 'hear when the message pops out'. She draws attention to the notification settings on her phone, mirroring many of the conversations I had with other research participants, where notifications played into this understanding of access and intimacy.

Notifications: accessing intimacy
Notifications were a ubiquitous part of the technological infrastructure of the devices research participants used to engage with dating apps, primarily their mobile phones. Notifications are alerts that appear universally across a user's device and can inform them of an activity or event corresponding to a particular app, even when that app is not currently open (Chandler and Munday, 2016). For example, a user will receive a notification when they are sent a message on WhatsApp -they can see the message without opening the app. It is possible to enable or disable notifications across the entire array of apps installed on a device and also to fine tune how and where they appear. This factored into the motivations behind transitions away from dating apps, for example, Nancy, 28, did not have notifications enabled for her dating apps, but did have them enabled for WhatsApp, and as such sought to communicate with matches she particularly liked on WhatsApp rather than via the dating app itself, again tying into the link between access and intimacy. Kaufmann and Peil (2019), in their study on using WhatsApp as a platform for interviews, note that the app offers an unprecedented level of access for the researcher, since 'participants would be unlikely to switch off the app or its notifications as this would isolate them from their everyday communication partners, too' (p. 6). While a different context to that of dating app users, the underlying notion here is the same, namely, that gaining access to a person's WhatsApp grants you continuous access to their most intimate sphere of everyday communication. Indeed, among my research participants, it was common not to have dating app notifications enabled, as users sought to keep their dating apps in 'that box', to use the term employed by Laima, while only matches seen to be worthwhile were granted access to the notification-enabled communication sphere of WhatsApp. Karsten, 24, was particularly explicit about the relationship between his transitions away from dating apps -he solely used OkCupid and Bumble -and the notification settings on his device.
It usually goes to WhatsApp pretty quickly, even before a meeting, simply because I notice with many people and also with myself, that I just don't check the app [OkCupid] a lot. And it's silly if you don't reply to people. [. . .] And you're on WhatsApp quite a lot. Also, what you have with WhatsApp is that if someone writes you, it shows you the message straight away. In the worst case you can, even if your phone is still locked, you can already write your reply and send it. And you can't do that with dating apps. There, all you see is, for example, 'Luisa sent you a message', but you can't see what exactly she wrote and that's crap. [. . .] You have to unlock your phone to get into the app and then you can reply there. And if someone sends you a message and you're on WhatsApp, then I quickly forget that someone else wrote to me on one of the dating apps. That's why I try to switch relatively quickly to WhatsApp. (Karsten; My translation from German) Karsten highlights the technological implications for communication inherent in the way different apps interact with notifications on his phone. As previously discussed, Jürgen felt that one cannot easily open dating apps in all places to the same extent as one can do so with WhatsApp, since it is not seen as appropriate to use Tinder openly at work, for example. Karsten expands on this point of access noting that the notifications he receives from dating apps differ in design to those he receives from WhatsApp. A notification from WhatsApp shows him any messages he has received -even without unlocking his phone -while a notification from OkCupid simply alerts him that he has received a message, but not what that message is. Tinder, Bumble and OkCupid do not allow users to read messages in notifications outside of the app. The technological affordances of dating apps propagate the segregated image of the 'box', or separate sphere, as experienced by Laima and other dating app users.
The design choice of not disclosing messages in the notifications themselves suggests that these messages may be too intimate to openly reveal outside of the sphere of the app. The affordances of the app thus imply a similar notion of public-facing intimacy as that described by my research participants, namely, the feeling of not wanting to open the app in the presence of onlookers. The app interweaves intimacies, not only in its emotive association of use, but also in the way it is designed. One must note that the design and infrastructure of dating apps continues to evolve, but during the fieldwork in Berlin it was indeed the case that the three primary apps investigated in this thesis, Bumble, Tinder and OkCupid did not reveal the content of messages in their notifications. This, however, is of course not applicable to all dating apps. Another dating app, Hinge, for example, did show messages in its notifications. Yet, in both cases, whether messages are revealed or not, it is fascinating to see the way the affordances of dating apps mirror or brush against the accepted patterns and ideas of use projected upon them by users. As such, one can see the fact that Hinge allows messages to be read in notifications as an attempt to infiltrate the normalised sphere of communication occupied by WhatsApp on a dating app user's phone, or as a bid to distance the app from the public-facing intimacies and secrecies associated with dating apps, for, as Lamb and Kling (2003) write, 'the technical and the social are inseparable' (p. 202). It is clear that users categorise the experience of intimacies on their mobile devices, and the design choices of dating apps will naturally feed into, disrupt or mirror such user practices.

Conclusion
The role of dating apps in courtship rituals and dating practices occurs not in isolation but as part of a complex polymedia environment. This article has argued that while a match on a dating app is the first encounter between potential partners, it is not perceived as carrying much significance in signalling romantic interest or intimacy. In contrast, switching channels of communication away from a dating app serves as a symbolic shift in the relationship between two users and is cast as solidifying a connection. The article terms this switching, usually in the form of an invitation to connect via private messaging services, most prominently WhatsApp, as a ritual of transition. The article argues that the significance of this ritual is twofold. First, the invitation to leave the dating app, and the acceptance of this request, acts as a signal of mutual interest among users, and often a precursor to an in-person meeting -it is a form of communication, a symbolically coded gesture of intimacy, in itself. Second, transitioning away from a dating app, within the polymedia environment of the smartphone (Madianou, 2014), highlights the compartmentalisation of social life that occurs within a user's device. Research participants repeatedly used terms of segregation, such as spheres and boxes, to describe their smartphones not simply as intimate devices (Goggin, 2011), but rather as technology incorporating experiential spaces of varying intimacy. Existing scholarship has highlighted the way technology is socialised, particularly in recent ethnographic studies of social media (among others, see Costa, 2016Costa, , 2018Miller et al., 2016;Spyer, 2017), and the fieldwork similarly finds such socialisation in Berlin, where WhatsApp in particular was cast as the most intimate form of mediation between dating app users. Exchanges with a dating app match via WhatsApp were seen as integrating potential partners into a space characterised by an intimacy derived from WhatsApp's role in the everyday 'mundane' interactions with family and friends (Cabañes and Collantes, 2020).
WhatsApp, and other messaging services such as Signal, also led to the experience of heightened intimacy between potential partners as they represented a hierarchy of access. Due to the infrastructure of notifications ingrained in research participants' devices, Tinder, Bumble and OkCupid offered less immediate access to potential partners in a practical sense, since notifications from these apps did not reveal messages. Some research participants actively disabled notifications for their installed dating apps, further separating them from other social media installed on their smartphones. On the whole, research participants did not report feeling stigmatised for using dating apps, indeed stigmatisation of online dating has decreased as apps have replaced traditional dating websites (Degim et al., 2015). Rather, the practice among some users of disabling notifications for dating apps was rooted in the perception of dating app matches as strangers, who had not yet merited the greater access offered by notifications. Paradoxically, in the eyes of onlookers, dating apps were seen by research participants to be more intimately -specifically sexually and/or romantically -charged than social media such as WhatsApp, and thus not deemed appropriate for use in professional environments, further highlighting the increase in access to a potential partner post transition, since they could be reached no matter where they were currently situated.
While the data on which conclusions are based are rich, insightful and were collected over a substantial period of time, it relates specifically to young and emerging adults; thus, it cannot be said to be wholly representative of all users. Second, while the sample of research participants did include varying nationalities and cultural backgrounds to reflect the cosmopolitan nature of Berlin, the great majority were of white ethnicity, and as such data are particularly representative of this group. It would be interesting for future research into dating practices across smartphone platforms to incorporate the perspectives from different demographics to those discussed in this article, particularly older users, since these have not featured prominently in dating app scholarship. A comparative study between cities would also have the potential to provide fascinating insights into how dating rituals on dating platforms shift perceptibly in different cultural environments.
There is scope for further research into how such questions of access play out across city space (see Miles, 2019;Roth, 2014;Veel and Thylstrup, 2018), and the wider research project from which the data in this article are drawn engages with this facet at length. Informed by a polymedia framework, this article instead has focussed in detail on the practices of dating app users specifically in their first transition away from a dating app with a potential partner, identifying this as a key ritual of courtship enmeshed within a compartmentalising of intimacies across a mobile device. Technology remains deeply socialised, to the extent that the affordances of various applications at times jar with more abstract conceptualisations of their purpose. The ethnographic data have shown the importance of treating media and user via a relational framework, leaving room for the interplay between design and user interpretation in the study of the social. Future research on dating apps, but also smartphones and mobile media more broadly, will benefit from retaining this perspective, ensuring that devices and media are approached from the perspective of their users, and that affordances are understood not only as shaping or limiting actions but also as being shaped and limited by the location specific socialisations of technologies.