Desire and confusion: A sociolinguistic ethnography on affect in the ethnic economy of Thai massage

Abstract In my contribution, I will look at the interconnections between language, work, ethnicity and gender in the exemplary site of the Thai massage studio as part of a larger sociolinguistic ethnography in Vienna, Austria. I argue that Thai massage therapists are trying to establish an independent and professional self, while being continuously repositioned along gendered and racial stereotypes based on post-colonial ideas of the “exotic woman”. In other words, their work empowers them on the local labour market, but simultaneously threatens to reinstall clear social and ethnical hierarchies. In order to unpack this complex, I propose to discuss two theoretical concepts from a critical sociolinguistic perspective: the ethnic economy and the affect of desire, as they both inform an understanding of Thai massage as a particular localised global practice. I will first discuss ambivalent opportunities related to language competences in the ethnic economy, and then turn to examine how male clients come to ascribe “confused affect” to their experience with desire in the Thai massage. Finally, I will discuss the issue of researcher positionality in dealing with the potential reproduction of exoticisation through research.


Introduction
Who has not encountered a Thai massage studio either in the centre or in a backstreet of any major city in Europe or, somewhat more unexpectedly, in a village in the middle of a countryside? Indeed, it almost seems as if the Thai massage studio has become an integral part of the European service sector, alongside other businesses of the so-called "ethnic economy", i. e. a specific form of business run by an "ethnic" group, e.g. the emblematic Turkish kebab stand or Chinese restaurant (cf. Light and Gold 2007). While the Thai massage studios may range from high-level luxury spas to run-down-looking establishments, they will offer a similar product. Thai massage is a traditional and medical practice (Salguero and Roylance 2011), which is widely known to have not only relaxing but also therapeutic effects (Iida 2010). However, sexual connotations have come to dominate the public imagination of Thai massage, unrelated to its therapeutic intentions, due to geopolitical developments within the last century. Thailand, for one, became a destination for sex tourism, thereby undergoing a process of "libidinalization" (Hamilton 1997), with Thai massage used as a cover for prostitution and other sexual services. This process has been explained, e.g. by Pruekpongsawalee (2004), as a result of the US troops deployed in the Vietnam War, who were sent for R&R (i.e. rest and recuperation) to Thailand, where they searched for sexual services. As a result, the affective ascription of Thai massage has changed from a purely therapeutic/relaxing massage to a sensual/sexualised physical experience (cf. Kogiso 2012).
To this day, (female) massage therapists in and outside of Thailand are confronted with erotic expectations, be they based on fantasy, mediatised narratives or actual experiences of clients (Sunanta 2020). Note that European massage therapists also experience sexualised demands from customers, which goes back to the "widespread popular elisions between massage and sex work" (Oerton 2004: 550). On the one hand, the sexualisation of massage work is due to the nature of work on "supine and frequently semi-clad or fully undressed bodies" (Oerton 2004: 550); on the other, it is related to the illegitimacy of advertising for sexual services in many European countries, which results in codified listings that are known to offer sexual services under the purported guise of massage (cf. Oerton 2004). Given this context, the question arises whether, under which conditions, and with which consequences Thai massage therapists (TMT) are in charge of their professional selves. How do they become entangled in ambiguous and shifting discourses and practices, which, in turn, can lead to clients' ambivalent affective interpretations of what TMT views as professional (i.e. non-sexual) interactions?
In this tension, TMT appear to find themselves in a bind. While attempting to avoid sexualised ascriptions and trying to create professional spaces, they also try to recreate an "authentic" environment for their customers, basically commodifying the "exotic" for business purposes. In this vein, they inevitably reproduce imageries of the "exotic" that are closely linked to post-colonially tinted images of the "erotic" (Manderson 1997). In most cases, the semiotic landscape of the studio will index Thai-ness through its business name, references to Thai topography, colloquial expressions (e.g. the greeting Sawadee, which literally means "good day"), the decorative staging of "exotic" imagery (e.g. the use of bamboo, Thai artefacts, images of the King of Thailand, Buddha statues, even an altar somewhere), and emblematic presentations of Thai writing, resulting in multilingually designed brochures, flyers, and signs. In terms of spoken language, the TMT usually speak Thai among themselves and a locally dominant language (e.g. German in Austria) or English with clients, albeit often on a rather functional level (Serwe 2015). They thus are multilingual and competent to employ a variety of communicative repertoires, depending on situation, domain, and interlocutor.
The relevance of language in this specific setting is closely connected to the conditions created by the political economy, i.e. "the technologies and processes governing the valuation of resources as well as their production, circulation, and consumption within a given place and at a specific moment in time" (Del Percio et al. 2017: 55). With respect to TMT, local migration policies and their focus on language requirements, as well as global marriage patterns between Western men and Asian women, become most relevant in this respect, leading to a gendered transnational mobility (Sunanta 2020). As a result, the migrated women's career options are restricted on the highly regulated labour market due to their limited language competences and their lack of (recognised) professional qualifications. The question thus arises as to how societally dominant ideas of language competences bringing about social and professional integration unfold in this environment, and how they interrelate with racialised and gendered stereotypes.
This examination of the interconnections is part of an ongoing sociolinguistic ethnographic project on language, work, ethnicity, and gender in the exemplary site of a Thai massage studio in Vienna, Austria. My aim is to contribute to existing critical discussions on language competences and labour market integration within the discipline, while adding another analytical layer: drawing on empirical data, I will argue that their work empowers these women on the local labour market, while it simultaneously reproduces hierarchies that are constructed according to social and ethnical imaginaries. This means that even if these women establish an independent and professional self, they are continuously repositioned along gendered and racial stereotypes that portray Asian women working in the service sector as docile and sexually available. Language competences, as demanded by migration regulations and labour market developments, come to play a somewhat arbitrary and unexpected role in this context, as their value shifts depending on the situation and on factors such as qualifications, networks, and work experiences. We shall see how the women's interactions with clients are framed by the ethnic economy and the affect of desire, two theoretical concepts with explanatory potential. The former explains practices of exoticisation as well as the conditional access to the Austrian labour market that result in blocked mobility for TMT, the latter how the job as TMT is defined and devalued (at least in part) through practices of affect. In order to clarify this relationship, I will first discuss the ambivalent opportunities related to language competences in the ethnic economy, and then turn to examine how male clients come to describe their confusion with regard to affect (i.e. "confused" affect of desire) in their experience of Thai massage.
2 From ethnic economies to the affect of desire: A theoretical framework While the assumed linkage between language competences and labour market integration has been critically scrutinised elsewhere (e.g. Flubacher et al. 2018), this study brings into focus the particular ethnic economy of Thai massage. It is heavily marked by sexuality and affect, which comes with specific invocations of "body work" (Wolkowitz 2002) and is the result of a "gendered division of labour" (McDowell 2009: 45;cf. Sunanta 2020). Not only is Thai massage often confused with sex work (like any type of massage, actually, cf. Oerton 2004), it is anchored in post-colonial imageries of gender and ethnicity. I argue that it is productive to explore these intersections between gender, ethnicity and work with a critical sociolinguistic perspective, as multilingual competences and communicative repertoires have shifting value, meaning and consequences for Thai women working in this ethnic economy. Moreover, as I will argue, the women's multilingualism is a condition for the formation of a particular register within this ethnic economy, including its semiotic landscape, functionality, taboos and euphemisms. Therefore, the professional practice of Thai massage will be analysed as negotiated in the framework of the ethnic economy and as simultaneously overdetermined by its stereotypical reading, which is infused with the affect of desire.

Ethnic economies as an analytical lens
When groups of immigrants from the same country or region develop entrepreneurial profiles in specific industries or niches and draw on internal networks for economic solidarity, they are considered "ethnic economies" (Light and Gold 2007). While many of these niches make use of a specific ethnic cuisine or cultural product, they often have formed somewhat serendipitously. Local institutional, political and economic conditions offer opportunities to newcomers hailing from the same "ethnic", "national" or linguistic background, while the economic activity can be unrelated to the "culture" of the group in question (for example Vietnamese nail shops that originated in the United States, cf. Eckstein and Nguyen 2011). Research has pointed out that specific ethnic economies are informed by group-internal features (e.g. socio-economic background) as well as by the "opportunity structure" of the host country (Waldinger et al. 1990), i.e. migration policies, societal "openness" (which includes access to the labour market or to economic resources) and the availability of intra-ethnic social networks for group-internal recruiting of labour force or for other economic activities.
In a reciprocal manner, the opportunity structure heavily influences group-internal features. For example, the study on the emergence and spreading of the Vietnamese manicure industry in and beyond the US clearly emphasises the importance of local initiatives and professional training allowed in languages other than English. It was particularly this linguistic openness that made it possible for Vietnamese women to carve out a specific niche for themselves, i.e. the affordable walk-in version of the already existing manicure business (Eckstein and Nguyen 2011). The figure of the ethnic entrepreneur has been a special focus point for research. They have been studied extensively in their efforts of looking for employment, and as potentially providing employment for other members of the same ethnic group. Their primary motivation has often been understood as an attempt to overcome "blocked mobility" (Raijman and Tienda 2000) in the new political economy, i.e. to productively apply themselves on their own when their original professional qualifications are not valorised. However, whether the founding of an ethnic business really is a "stepladder experience" (Raijman and Tienda 2000) leading to upward social mobility not only depends heavily on individual aspirations and/or the resources available (e.g. financial, material, personal) but also on the specific business sector. In any case, networks based on shared "ethnicity" (or nationality) seem to repeatedly emerge as a functional instrument to overcome difficulties encountered on the local labour market, and can be ideally drawn on for solidarity and reciprocity. Furthermore, such networks provide support for individual or collective experiences of discrimination and exclusion, and facilitate access to information, capital and institutions. While the ethnic economy thus offers advantages to entrepreneurs and their employees, there are also potential drawbacks; namely, loss in salary because of social obligations to remain in a specific business sector (Sanders and Nee 1987) or the danger of being employed under informal and illegal work conditions, due to the avoidance of complicated and costly legal employment (Portes 1995). Finally, remaining in intra-ethnic economies can run counter to the legally widely implemented pressure to learn and improve the locally dominant language.
Recent research has pointed out that it is problematic to use the term "ethnic economy" for several reasons (e.g. Pécoud 2010). First of all, "ethnicity" can be an imprecise description, a shifting index of belonging that only sometimes coincides with nation-state boundaries, while in fact at times nationality becomes the most important common denominatoror, under other circumstances, sub-ethnic categorisations emerge as prevalent. In this line, I am aware of the debate surrounding the term "ethnic" economies, further contested by the proposition of "migrant" economies instead (Glick Schiller et al. 2006), which intends to broaden the discussion and to refrain from culturalising tendencies in research. Yet in the case of Thai massage, the ethnicity of the women is a central element of their economic activity and its ascriptions, which is why I will keep with this term for the purpose of this paper. Finally, research on ethnic economies has mostly focused on male entrepreneurs and/or on family structures in support such endeavours (Dannecker and Çakır 2016;Erdem 2005). The Thai community in Vienna, however, is largely female. There are concrete reasons for the female composition of this community, contributing to its specific ethnic economy, i.e. Thai massage, which has specific gendered and affective challenges and issues. In view of the existing bias toward male entrepreneurs, my research can contribute to making this strand of research less gender-biased. In sum, I understand the "ethnic economy" as a conceptual lens in order to gain a sociolinguistic ethnographic understanding of its challenges and issues that are further exacerbated by societal pressure to learn the local language.

The affect of desire as patterned practice
Affect has increasingly become the focus in research across the social sciences as a critical component that relates to, impacts on and intersects gender (Butler 1997), race (Ahmed 2004), labour (Hochschild 2012(Hochschild [1983), and language (Besnier 1990), resulting in its own bona fide "turn" at the beginning of the millennium (cf. Clough and Halley 2007). Affect can be described as "every aspect of emotion and sometimes it refers to just physical disturbance and bodily activity" (Wetherell 2012: 2), but its conceptual understanding actually implies a complex variety of theoretical approaches, according to the ontological and epistemological stances of the research(ers) in question (cf. Wetherell (2012) for a general overview). In the critical study of language, affect becomes relevant in that it emerges in relation to language practices and language ideologies, as described by McElhinny (2010) in her review article on the interplay of gender, race and affect. Sociolinguists have pointed to the relevance of researching the role of affect for language learning (Busch 2017), the ramifications of societal concepts of the legitimate speaker (Boudreau 2016) or, in relation to this, the linguistic insecurity of speakers (Park 2017).
Echoing the main tenet of critical sociolinguistic scholarship, i.e. to address the social dimension of language use (Del Percio et al. 2017;Kraft 2016), I will draw on Wetherell's (2012) proposal to understand affect not as a singular, isolated and individual "emotion", but as socially patterned and distributed. This implies that the patterns of affect are intrinsically linked with and embedded in political economic relations. Secondly, affect should be understood as practicesimilar to how critical sociolinguists address language as practice rather than abstract system. Affect should thus be approached as emerging in interaction and as enacted, situated and negotiated. Finally, affective practice is "bodily activity" (Wetherell 2012: 2), most clearly visible in affects displayed and enacted bodily, such as joy, fear, panic or desire. In short, affect becomes what Wetherell (2012: 4) calls "embodied meaning-making", i.e. "[a]n affective practice is a figuration where body possibilities and routines become recruited or entangled together with meaning-making and with other social and material figurations" (Wetherell 2012: 19), which, again, are framed by the political economy.
There is one particular form of affect that emerges in the context of the ethnic economy of Thai massage: the affect of desire. Subjected to critical investigation in sociolinguistics, desire has been understood as residing closely to a complex of politics and power, in that it is connected to subordination (cf. Cameron and Kulick 2003;Foucault 1978). The sheer drive to desire anything is thus determined by power structures and by one's position in the social structure: not everyone is deemed legitimate to desire something, someone, a certain action or productor to actually express this desire. In this sense, studying desire becomes highly informative of power structures, privilege and entitlement, and mechanisms of exclusion and inclusion that are informed by hierarchised ideologies of gender and race. Invoking post-colonial concerns raised by Fanon (2008), Milani (2016: 412) reminds us "that the power imbalances produced and justified on racial grounds cannot be fully understood unless one also teases out the sexual desires and fears generated by race". In sum, as any other affect or affective practice, desire is not located in an individual but is embedded in larger societal patterns and practices, in language practice(s), discourse(s) and sign(s). It is infused with power and with a claim of legitimacy to be able to desire and to take what one desires.
Approaching the affect of desire as a socially patterned and distributed practice, we are confronted with several layers of questionsmost importantly: Who is in charge of the meaning-making and the interpretation thereof? How is the meaning-making distributed across participants, and in which social and material figurations? And, finally, in what way and for which reason does this bear importance for the critical study of language? With these questions in mind, I argue that the analysis of affect as practice will shed light on the social practices, power constellations and political economic conditions that engender or frame this very affective practice. In my contribution I will focus on the affective practice of desire, which emerges at the intersection of ethnicity and gender, most notably in the work context of Thai massage, a particular case of the ethnic economy. First, however, I will briefly describe the research this contribution draws on, before discussing the positionality and required reflexivity in ethnographic research.
3 Ethnography, its data and context

A critical sociolinguistic ethnography of Thai massage in Vienna
The research is conceptualised as an ethnography of long duration with a critical sociolinguistic perspective. In practical terms, this means that I have been hanging out at a particular Thai massage studio in one of the lesser affluent areas of Vienna for two years at the time of writing. The visits are of irregular manner and diverse with respect to interval, weekday, time of day, duration and interactants, but generally I spend 2 to 5 h there in the morning on a weekday every two to three weeks. Present are normally two Thai massage therapists, Naa and Dohng, 1 who are working, cleaning up the studio, cooking lunch, doing laundryor simply waiting for customers, napping, and watching Thai TV shows. The elderly boss joins later around noon when the clients tend to drop by, so she can be in charge, i.e. opening the door, picking up the phone, settling the bill, etc. Occasionally, I will help out with small tasks before the boss' arrival, as well as with documents, instructions, and letters in German if they have issues understanding them. My visits and interactions in this studio are documented in the form of field notes, complemented by a collection of documents, websites, online discussion forum on Thai massage, photographs, etc. The analysis of this data follows the ethnographic attempt of thick description (Geertz 1973) that ideally leads to a detailed, contextual, and situated understanding of meaning-making in situ. Indepth qualitative interviews serve as complementary windows into specific histories, narratives and positionings of interlocutors. For this purpose, I have talked to and interviewed a broad variety of informants (e.g. other TMTalso of Austrian origin -, clients, and Thai officials in Austria), resulting in about 20 audio-registered and transcribed interviews. 2 Two interviews with male European customers are most interesting in terms of understanding the affective practice of massage. I will therefore draw on these interviews for exemplary reasons (Sections 4.1 and 4.2), as well as on field notes in order to reflect on the women's ambiguous positioning (Section 4.3). The analysis will be guided by the interest to understand how the affect of desire is semiotically (Section 4.1), discursively (Section 4.2) and reflexively (Section 4.3) evoked and negotiated in the guise of "confused affect"with different stakes, aims and consequences for the individuals involved.
In my ethnographic visits to the studio, my Asian appearance has led to a constant scrutinisation by customers and visiting acquaintances alike: am I a new colleague, a fellow Thai compatriot? While this negotiation of my person and positionality helps break the ice, more importantly, it is telling about the automatic positioning that is going on in everyday life on the basis of ethnicity and gender, which will manifest in language choice also (e.g. Kubota 2014). Arguably, this does not come as a surprise, but it still constitutes an important part of ethnographic research into these exact sociolinguistic components of ethnicity and gender, and forces us to reflect on our positionality as researchers. We might be first and foremost just another person in an interaction, but we always come with a specific, politically inscribed and marked body. Repeated ambiguous experiences in my personal life regarding the racialisation of my body have sensitised me to recognise and question sexist Orientalist stereotypes. Thus, I critically reflect on the potentially offensive undertone of my research, most importantly about running the risk of reinforcing such harmful stereotypes. My primary aim is to understand the intersectionality of language, work, ethnicity, and gender and how it is infused with particular ideologies and processes that might marginalise these women. Yet, I argue that employing a solely critical perspective that seeks to identify how this particular economic activity is pervaded by orientalising stereotypes, also runs the danger of diminishing the agency of these women and painting their lives in overly negative terms, victimizing them as a result. This is why an ethnography is the most suitable research stance, since it allows me to understand how these women make sense of their lives, how they attain agency, and what meaning they attribute to their everyday social practices without overly prioritizing preconceived notions (cf. Heller 2008).

The (female) Thai community in Vienna
In Austria, about 4000 Thai nationals were officially registered in 2013 (the latest numbers available), in addition to about 5500 Austrian nationals of Thai origin (Butratana and Trupp 2011: 183). The largest percentage of Thais (both groups included) live in Vienna, about 1200 and 1550 respectively; 85% of this population are women, with 1500 in Vienna alone (the gender distribution is similar in other European countries, e.g. in Germany, Denmark or the Netherlands, cf. Sunanta (2020); Trupp and Butratana (2016)). 3 As with every other migrant group, the women in Vienna are of different socio-economic backgrounds and from different regions in Thailand. While there are highly educated urban women working for the United Nations in Vienna, there is also a large group from rural (often northeastern) Thailand with lesser professional qualifications, mostly employed as TMT or cleaning staff. While these different groups do not necessarily interact on a daily basis, there are certain spaces that offer a sense of community irrespective of social standing, e.g. Buddhist temples or the local open-air picnic taking place everyday in summer, where Thai food and drinks are sold, games are played and massages are offered. In any case, according to my observations, Thai women in and around Vienna seem to form a close network, engaging in various forms of informal economy (e.g. selling food door-to-door), and recommend each other, e.g. when a replacement for a TMT is needed.
Of the Thai women living in Austria, 60% are married to an Austrian resident, with marriage one of the main possibilities to enter the country. This rate has actually gone down due to stricter financial (net-income) and linguistic requirements (cf. Butratana and Trupp 2011: 186), which is related to the migration policies of the European Union, making it increasingly difficult for "citizens" of socalled "third countries" to get a visa. The policies have resulted in only two possibilities for entry: (1) family reunification (with an officially certified A1-level in the local language, hence German in Austria) or (2) highly valued professional skills, in which case requirements are waived. As the majority of Thai women have no (officially recognised) qualifications, they arrive in the framework of family reunification. Yet, the most important factor for this female "surplus" is the global marriage pattern between Western men and Asian women, which is deeply entrenched in the (post-)colonial "logics of desire" still present today. While marriage patterns are highly complex and individual, researchers (e.g. Lapanun 2012) have discerned veritable entrepreneurial motivation on the side of many Thai women (e.g. to be able to send remittances to the family/community at home) as well as idealised views of Asian wives on the side of European men. Lapanun (2012: 23-24) concludes that "the logics of desire driving these women and men to opt for transnational marriage is [sic] rather complex, transcending both economic motivations and intimate relationships". Still, even if each individual story has many nuances and layers, I argue that the logics of desire transcend marital arrangements to impact on interactions also in the Thai massage studio, potentially over-determining them.

The Thai massage studio as a gendered and ethnicised site of conflicted meaning-making
The Thai massage studio is a specific exemplar of the ethnic economy in that it is highly gendered, which is contingent on migration policies on the one hand, and on global gendered and ethnicised flows of interpersonal relationships and "logics of desire" on the other. While there are a few male massage therapists in Vienna (usually partners of Austrian men), most are women who find employment due to an inter-ethnic network and shared language background, either bringing with them the required (embodied) skills or learning them on the job. While the demand for good and reliable TMT is high and the number of studios is on the rise, the employment conditions remain rather poor irrespectively: the women are present most days of the week, sometimes for almost 12 h. Often, they are declared as parttime or self-employed, which lowers their social insurance and pension. Sitting in a studio all day, chatting, cooking and eating together, might turn the studio into a domestic space of socialisation, but it also denotes waiting for clients. This can have a draining effect on the women, as I have witnessed many times, as well as the physically strenuous massage itself. The TMT who have arrived under the new migration regulations in force since the early 2000s are supposed to learn German for their residence permit, but hardly get to practice it as a result of being stuck inside the studio day in, day out. Even if they manage to pass the required test, they might not fulfil the language competences required on the monolingual labour market even for low-level jobs. As it will be hard for them to apply for other jobs, they are, in a way, "blocked" (cf. Raijman and Tienda 2000) within this strenuous and potentially precarious line of work. We could thus argue that in effect, the Thai massage studio is a product of the linguistically regimented migration policy and labour market. In addition, Thai massage is marked by the general processes of globalisation and feminisation affecting all body work (McDowell 2009; Vidmar Horvat and Pušnik 2019). Traditionally Thai massage was practiced both by men and women, but as it spread as a commercial and feminised practice across the globe it became infused with sexist and sexual fantasies. Unsurprisingly then, the two male clients I interviewed described experiencing some confusion relating to desire in the context of massage. This has led me to propose to read Thai massage as a site with confusion regarding the intention of the affective practice of desirehence, as confused affect. In order to probe a bit further, I would like to go back to the questions raised in the beginning, i.e. who has the power of meaning-making when it comes to desire within the massage event, and how is this distributed across participants? In order to address these questions, I will first turn to the two different clients' accounts of Thai massage, followed by the interpretation of the TMT themselves.

Confused affect and shifting indexicalities
Dimitri and Frank both frequent Thai massage studios to treat physical pain induced by intense sport activities. In fact, they exercise together, and it was on Frank's recommendation that Dimitri called on a specific TMT for his back pain. Dimitri, with whom I conversed in English, is a researcher at a local university in his thirties, originally from Southern Europe. Leading up to the following interaction (Excerpt 1), we were discussing how at times different services were offered under the cover of "massage". Having this in mind, Dimitri recounts ambivalent impressions when first arriving at the studio recommended by Frank. His own confusion related to the use of signs, as he explains: 4 As it appears, in Dimitri's account certain signs transmit sexual connotations, i.e. meaning "something else". This "something else" is induced as "the same sign of open and closed", in "colourful letters". At no point does he elaborate what "something else" refers to, keeping with the proposed euphemismbut it immediately becomes clear in our interaction what he means, especially when considering that he referred to "the two different types of […] let's say massage" mentioned by me just before this excerpt. In a joint interactional effort, we had classified studios into two types, "serious" (i.e. regular massage studio) and "nonserious" (i.e. with sexual services). To the latter he now attributes specific semiotic markers: an LED open-closed sign. In this recounted instance, Dimitri's interpretive frame is challenged due to a recommendation without which he would not have entered this studio. It becomes evident that interpretative frames are connected to specific semiotic markers, which are rooted both in individual and collective orders of knowledge but have shifting indexicalities (cf. Spitzmüller and Warnke 2011). For Dimitri, confusion begins even before entering the establishment, as he claims that signage may be a common means by which massage studios communicate to clients about their status ("serious" or sexual). The LED sign here serves as a prototypical example in that it apparently indexes different practices and persona to different people and is, thus, rendered ambiguous as a semiotic variable. The confused affect of desire resulting from this has potentially material consequences for the TMT of this studio, who are not necessarily aware of the interpretative potential of such signs. They might lose potential regular clients on the one hand, and attract clients who are looking for sexual services on the other. Yet, the confusion does not stop with this semiotic marker. In trying to unpack the reasons of why Austrian men might misinterpret massage as "something else", he states: As becomes evident in Excerpt 2, Dimitri sees the Thai female performance of "politeness" as a source of potential confusion or misinterpretation for Austrians without intercultural experience with Thai people, i.e. confusing politeness as a legitimate entry point for desire ("this may be perceived a bit differently"). He sets himself apart from such misinterpretation, telling me at an earlier point in the interview about working with a colleague from Thailand as a PhD student. With this strategic move, he discursively deflects any desire on his part, as he positions himself as familiar with the meaning of "Thai politeness". Emphasizing the Excerpt : Interview with client Dimitri, December  (Lines -): Dim: yeah yeah \ ehm (-) but the politeness i saw this here too \ at the at the massage place [m = h \] and this is probably i mean that's common in Thailand to be very polite \ ah but this may be perceived a bit differently in the different cultures \ so like in Austria for example [m = h \] i don't know \ i mean okay i could tell this is a cultural thing \ eh (--) but yeah may be in Austria (--) but this is my interpretation [yeah \] people might perceive it in a different way \ and they might expect something more that that that this is probably intimacy or something \ i don't know Desire and confusion friendly and polite demeanour of Thai people, he seems to imply that this Thaispecific politeness might lead inexperienced people to "expect something more", thus over-interpreting it as an invitation or openness to "intimacy", potentially engendering for clients a sense that the massage could become a sexual event. So, while Dimitri locates the confusion in the act of interpreting the "purpose" of Thai massage, he turns to explaining this pragmatic feature as a "cultural thing". Politeness thus becomes a culturally shifting index. This points to the argument proposed by Blommaert (2007) that "orders of indexicality" cannot always be transposed across (cultural, linguistic and political-economic) contexts, which will lead to varying if not contradicting interpretations of signs and pragmatics. What Dimitri does not consider is the post-colonial power structure, in which (Asian) women are not only expected to smile to and for White men, but are also expected to comply with enactments of desire.

Confused affect as discursive strategy
In the first excerpt, Dimitri recounted how he overcame his confusion, maybe even feeling the need to legitimise entering a massage studio with a semiotic surface indexical of "something else". Another potential confusion regarding affect is elaborated on further by Frank, an Austrian in his late forties and a self-declared seasoned client of different kinds of massage styles and studios (from sports or Swedish massage to Chinese and Thai massage). He professes to choose specific massage therapists for particular muscular pains related to his sporting activities and positions himself as an expert persona in this regard. I had several conversations with Frank about Thai massage before he agreed to be formally interviewed (in German; translations are all mine). Contrary to Dimitri, who referred to culturally differing interpretations of semiotic markers and pragmatics, Frank locates the source of potential "confusion" on the side of the (male) client simply in being undressed. His confusion thus refers to the act of massage itself: Excerpt : Interview with client Frank, November  (Lines -): Fra: und ja () natürlich ist es manchmal auch wieder merkwürdig na / weil eine gewisse erotische geschichte wenn man so entkleidet ist immer mitschwingt \ aber das ist kein typisches thai problem \ English translation Fra: and yes () of course sometimes it's strange once again huh / cause a certain erotic thing is always there when one is there undressed \ but that's not a typical thai issue \ After a pause of 2 s, Frank muses on the state of being "undressed" and concludes that this alone inevitably entails a "certain erotic thing"; which, however, in his view, is inherent to any massage and not particular to Thai massage. What he does not mention is whether he feels the same way with male massage therapistsa point neither raised by me. At the same time, his argumentation legitimises his reading of the situation as potentially erotic, due to his being undressedan "issue" probably not occurring when finding himself undressed at the doctor's, for example. As such, this assessment echoes the basic problem of service work on naked or semi-clad bodies that results in "elisions between massage and sex work" (Oerton 2004: 550), clearly positioning this issue as emerging at the power-infused intersection of gender and the ethnic economy. Even if Frank sets the massage scene as inherently potentially "erotic", he still voices confusion as to whether and when the interpretive frame shifts from therapy to include sexual pleasure: In Excerpt 4, Frank openly describes his continuous confusion with regard to desire in the process of a massage. Tellingly, he uses a deflection strategy to distance himself from this confusionor in fact, from engaging in desire: unlike Dimitri, he does not use the first pronoun, but continues to use the more generic "you" ("du" in German). And, indeed, Frank had told me before that he was often offered an "extra" at different places, but always declined (in spite of his professed attraction to Asian women), as in Austria any paid sexual activity not licenced as prostitution is illegal. In my view, this excerpt clearly shows how inscribed this confused affect related to desire is in accounts of Thai massage. Rather than probing into what happens behind closed doors, I aim to illustrate the implication of the confusion, i.e. the sexualisation of the massage practice and the women, and the discursive casualness with which Frank delivers this. More importantly, he frames it as if it was an intentionally created ambiguityagain, related to the fact Excerpt : Interview with client Frank, November  (Lines -): Fra: äh blöd komisch wirds dann nur wenn das ganze irgendwie so verwirrend ist na / wenn wenn sie halt dann jedes mal irgendwie so den eindruck bei der massage erwecken als wollten sie mehr und du weißt dann halt nicht ja weil wenn wenn du schon drei vier mal dort warst und die genau wissen dass es kein extra gibt und dann aber trotzdem so massieren English translation Fra: erm it turns awkward just when it all is somehow so confusing huh when when they well every time somehow create the impression during the massage as if they wanted more and you just don't really know cause when when you've had been there already three four times and they know exactly that there's no extra but still massage you this way Desire and confusion that offering paid sexual services is illegal, which prohibits explicit offers on the side of TMTwithout considering the possibility that this ambiguity actually could be his imagination. Finally, he construes his confusion about the sexual nature of the embodied practice without considering questions of power. In this vein, the verbalisation of confusion regarding the articulation of desire couldin its extremeactually serve as a discursive strategy that naturalises gendered sexual abuse or violence. Further, it projects any invitation to desire onto the TMT rather than the clients. The affect in question, as confusing as it is, is thus a patterned and clearly distributed practice involving gendered and ethnicised bodies that are inscribed in specific political economic conditions and logics of desire, which furthermore allows to imagine certain practices and utter certain words. It is with these issues in mind that I turn to the account of the TMT themselves.

Desire as an intersectional exercise of power
In the countless conversations with both TMT Naa and Dohng, they shared story after story about sexualised desire and inappropriate behaviour from clients and their reactions. Oftentimes they use raunchy humour to discuss these instances among themselves (which they translate from Thai for my benefit), arguably to diffuse their discomfort of being positioned as sex workers. Naa, for instance, has complained repeatedly to me about her situation over the last years. She arrived in Vienna a few years ago after marrying a German living in Austria, and has been working as a TMT since. After her arrival, she haphazardly called on a large Thai massage studio for employment, which hired her immediately due to her training and experience as TMT in Bangkok. As her husband does not approve of her profession (i.e. its reputation), they have gotten into fights repeatedly. Yet, other efforts to find employment have proven futile, as many jobs ask for higher German competences. However, without the energy to study on her own in the studio and without much opportunity to practice her German due to limited interactions (Serwe 2015), she has not managed to improve her linguistic level. She has thus pushed for the possibility to take time off work to attend German classes, even if this is not well received by her colleague Dohng, who remains alone when she attends morning classes. By now, she has retaken the B1-level German course twice and is thinking of taking it again as a refresher. From when we first met, she has confided in me that she will apply for other jobs (ideally in cleaning) once her language competences suffice the demands of the labour market. The following Excerpt 5 recounts a conversation with her in mid-December 2018, in which I tried to understand how she makes sense of the reasons for the taking hold of the logics of desire.
In this narrative, documented in my field notes, Naa clearly holds the precarious economic conditions responsible for the sexualisation of her trade, simultaneously demarcating this experience particularly to Vienna. She claims to have encountered this illegitimate side of her trade only in Vienna, and in blaming the enacting of desire on the local context, she positions herself and her fellow TMT as legitimate professionals. The sexual interpretations and demands voiced by some clients thus become even more prevalent in her lived and recounted experience. According to her, the logics of desire are what drives the clients to probe the TMT's frame of conduct, sometimes by the use of euphemistic language or, in the extreme case, even by boldly and outright asking for sexual services. This moment can be read as an interactional attempt at meaning-making of a particular embodied affective experience, highly insidious precisely because of its recurrence. Naa's resigned assessment, "That's the competition", implies that her clients approach her inappropriately because of the illegitimate practices enacted by others who are trying to make some money on the side. The emergence of a vicious cycle of supposed supply and demand centring around the inciting and acting on desire each reinforce the other, exposing TMT to its logics, with which Naa, Dohng, and others are forced to deal on a daily basis. Due to the political economic situation, most importantly the linguistic regime of migration and the labour market, it could be even argued that desire has structuring and structured effects in this particular Excerpt : Conversation with Naa (Field notes,  December ; translated from German): It is a quiet moment in the studio after lunch, with Dohng working and the boss absent. Naa tells me that she's had it with her work in the studio. It's become boring for her to always have these discussions with clients who are trying to feel her up. She is stroking her own arms and legs in order to show me how they would attempt to fondle her. Apparently, it has reached the point that is has become uncomfortable for her, as so many clients were displaying such behaviour. Some clients would even outright ask her why she didn't offer any sex, and argue that everyone else was. We talk about this for a bit, I ask her where this demand was coming from. She thinks that it has to do with the increasing competition between studios and TMTmany TMT would offer additional services to retain clients. She explains to me that in many studios women often only received about  EUR for  h of massage (which cost between  and  EUR), with the rest going to the bosses. It was thus that the TMT were courting clients in order to earn extra, which would go into their own pocket. In fact, Naa recounts a conversation with a client who had told her about another studio, in which one TMT did not wear any underwear during massage. Asking him about his reaction, he apparently replied to Naa that he had fondled that TMT during the entire massage. Naa shrugs her shoulders: "That's the competition". I shake my head and reply that this was giving massage a bad name. She agrees and adds, exasperated, that she had not known about this in Thailand, where she only gave massagesit was only in Vienna that she had learnt about this "extra". Furthermore, she does not believe that this is happening outside of Vienna, where the competition is less steep in her opinion.
ethnic economy, as it is clearly distributed in interactions that are deeply intersecting gender, ethnicity, thus becoming scripted in terms of power exercise.
5 Concluding thoughts on affect, the ethnic economy and multilingualism from a sociolinguistic perspective Considering Naa's experience as representative for many other TMT, I argue that the Thai massage studio is an epitomised example of both the advantages and pitfalls of the ethnic economy, which becomes even more blatant due to is gendered and ethnicised infrastructure. On the one hand, it offers an easy passageway into economic activity for Thai women recently arrived in Austria, guaranteeing some sort of independence from their Austrian husbands. It does not demand high competences in German and draws on skills which are either already there or at least culturally familiar. In this vein, the shared Thai background and language serves as an entry point for them. On the other hand, working as TMT keeps them in this highly contested working environment, on top of which they have no real opportunity to improve their German competences. Inevitably, they are coming face to face with the affective logics of desire, i.e. racialised and hierarchical sexual logics, their structuring of interactions and its impacting on expectations of Thai massage in Vienna. What follows from this is that the practice of Thai massage, considered a specific ethnic economy, cannot be dissociated from the global post-colonial pattern of sexual power inherent in articulations of desire and, following from this, its gendered nature. As the interactions are likely overdetermined by the affective practice of desire that is a constant undercurrent of the stereotypical reading of this particular ethnic economy, we can observe the formation of a specific register inherent to the site of Thai massage, which includes semiotic markers, bodily practices, and taboos and euphemisms. For us as critical sociolinguists, the question of who is in charge of the meaning-making sheds light on the distribution of power within society, exactly through the patterned enactment of desire. After all, the affective practice of desire to consume the fantasies of the eroticised "Orient" (Lapanun 2012: 6;cf. Said 1978) cannot be reduced to sex tourism, but is reproduced in interactions between Western men (and, to a certain degree, women) and Asian women (or men) in Europe. Against this backdrop, the body of the TMT becomes the site of desire, and thus the site of the negotiation of desireand its enactment. It is thus crucial to understand who has the power to utter desires and who is then forced to react; in other words, how desire is distributed and how we can interpret diffusing strategies such as "confused affect". Issues of power begin with me as researcher, who feels entitled to study the daily lives, practices and struggles of these women, probing into their exoticisation and sexualisation. Nonetheless, it is my aim to contribute to an understanding of the lived experiences of these women that also takes into account the social and material figurations that emerge. It is in these instances that we get a glimpse of broader processes of gendered and ethnicised meaning-making in contemporary society, e.g. in the ethnic economy of the Thai massage, with its interconnections between language, ethnicity, work and gender. The linguistic dimension is interwoven in these interconnections, manifesting itself in everyday encounters and lived experience, and on a different scale becoming highly relevant through migration regulations, as it determines access to visa and work. As such, the linguistic dimension sharpens the relevance of these interconnections even more.