Open Research Online The emotional organization and the problem of authenticity: The romantic, the pedagogic, the therapeutic and the ludic as liminal media of transition

Max Weber excluded the phenomenon of emotions from the idea of rational bureaucracy. Modern European organizational theories are on the other hand almost obsessed by emotions and especially affect. Emotion re-entered organizational theory around the limited topic of ‘ emotional labour ’ , but today, passion is generally praised as a driver in successful organizations. An important element here is the demand upon passionate employees to install the organization as their significant other. To the extent that they rely on this new concept of themselves and their employees, organizations become dependent upon the authenticity of the ‘ self-enrolment ’ expected of each employee. In the discursive field of organization, we therefore see a number of new communicative media, which centre upon emotion and upon helping the organization to attribute authenticity and inauthenticity to employees. This paper also makes the case that these media are liminal in nature and extend beyond the use of discursive symbolism in a Sisyphean effort to reach the authentic emotional ‘ heart ’ of each employee.


| INTRODUCTION
In Economy and Society, Max Weber wrote the famous words: 'Bureaucracy develops the more perfectly, the more it is "dehumanized," the more completely it succeeds in eliminating from official business love, hatred, and all purely personal, irrational, and emotional elements which escape calculation. This is appraised as its special virtue by capitalism' (Weber, 1978, p. 975). One corollary of this situation for modern organizations is a feature that has been called 'formal membership'. This can be illustrated by something Mr Lorry says to Mr Carton in Charles Dickens' A Tale of Two Cities: 'we men of business, who serve a House, are not our own masters. We have to think of the House more than ourselves' (Dickens, 2008, p. 78). The condition of membership is that the role of the employee (here, Mr Lorry) is decided by the organization (the House of Telson's Bank). For this to work, the employee must take the perspective of the organization, and that attitude must override any 'personal' perspective of their own. In this context, the dramaturgical notion of role designates a general motive, generalized both from the person who should fulfil the role and from the specific situation. The 'House' (or organization) thus functions as what Mead (2015, p. 154) famously called a generalized other. The role is decided by 'the organization' and in the name of the organization, and this is done through each employee adopting what they take to be the perspective of the organization (one must 'think of the House more than ourselves'). The generalization of motives afforded by this generalized other provides insight into the impersonal character of membership (Luhmann, 1982, pp. 75-78). Generalized motives are not directed at a specific person, whose job is to realize them, but is beyond the individual, like a mask that can be worn by any qualified person. In this way, membership also entails a zone of indifference between organization and individual (Luhmann, 1996, p. 341). This indifference is again neatly illustrated when Mr Lorry explains to Mr Stryver that as 'a man of business, I am not justified in saying anything about this matter, for, as a man of business, I know nothing of it. But, as an old fellow, who has carried Miss Manette in his arms, who is the trusted friend of Miss Manette and of her father too … I have spoken' (Dickens, 2008, p. 139).
This zone of indifference-in which the speaker of the role knows 'nothing of' that which is relevant to the personal speaker-is double-sided: it implies (a) the organization's indifference to personal motives for being in the organization and (b) the employee's indifference to the organization's instructions and distribution of responsibilities as long as they fall within the framework of the general motive (Barnard, 1968, pp. 167-171). To be granted membership implies that one has been appointed and recognized as a person. But membership also implies that one is assigned generalized motives, which establish one's relevance in relation to the organizational communication. One is employed for a specific purpose. It is only a particular part of the person who has been granted membership, and only this part is included as relevant for the organizational communication. The rest of the person is granted existence, but is considered irrelevant to the communication, and is typically given the indefinite label 'private' (Kieser, 1989, p. 547). Thus, the decision of membership perpetuates a principle of inclusion by exclusion: one belongs on condition that-in one's capacity as an employee-one publically disowns one's private desires and feelings and speaks instead through the mask. One is only relevant to the communication, and therefore included, through reference to the generalized motive. Everything else is excluded until further decisions are made. Formal membership, which became constitutive for modern organization, thus defines membership by clearly distinguishing role and person and by enrolling only those aspects of the person deemed relevant to the enactment of the role (as a unity of role and person).
Today, this very radical exclusion of feelings and private personal issues by formal memberships is somehow challenged. For some time, the need for passion has been articulated in management handbooks and HRM policy papers. One may even speak of a discursive regime of intimized management (Andersen, 2013;Andersen & Born, 2007Bauman, 2008). An example of this articulation is the book Passion at Work from 1998, written by marketing professional Kevin Thomson. His basic assumption is that 'one of the strongest emotions of all is passion. It is a motivator which drives us to incredible limits' (Thomson, 1998a, p. 3). For Thomson (1998a), passion is the single most important quality in an employee: 'When the passion is gone, people either quit from believing in their job and go, or even worse, they 'quit' and stay. We're left with people with no enthusiasm or excitement for what they do ' (p. 4). Richard Chang's (2001) book The Passion plan at Work is in line with this idea and argues that 'passion is the single most powerful competitive advantage an organization can claim in building its success' (p. 5). Shaul Fox and Yair Amichai-Hamburger define a more practical set of questions. As they say, the challenge is: 'How to tap the emotional components of human behavior?' (Fox & Amichai-Hamburger, 2001, p. 87).
The implication of this is that the organization as generalized other must somehow shed its generality in order to recapture something of the particular quality of a significant other. This, as will be touched upon later when discussing feminist contributions to the literature, can be viewed as a certain becoming mother or becoming lover of the old austerely patriarchal 'House' described by Dickens. The feature noted earlier by which the employee must take the perspective of the organization remains constant, but the quality of that perspective goes through a transformation. Both the organization and its members together undergo a discursive intimization that is also captured by the term 'emotionalization' (Greco & Stenner, 2008). This emotionalization is also reflected in a growth of academic interest in emotions and organizations, including the establishment of new journals (Domagalski, 1999;Fineman, 2005), and Greco and Stenner (2008, pp. 1-21) show this to be a trend that has occurred across all significant social systems and associated academic disciplines. The early phase of this trend was received rather critically within organization studies via the 'emotional labour' approach (Hochschild, 1983), before mainstreaming into the form illustrated by the examples quoted above. In more recent critical organization studies, a number of effects of this discursive intimization of work relations have been discussed. Betina Rennison has shown how the question of pay is framed as reciprocity of love. It pays off to show initiative and engagement (Rennison, 2007). Asmund Born and Niels Å. Andersen analyse the history of the public employee and show how formal membership is partly transformed into a membership of self-enrolment coded in the media of passion and installing the organization as the significant other. In the media of passion, the employee is expected to continually strive to make herself loved by the organization by anticipating the organization's needs. Love becomes a managerial regime (Andersen & Born, 2007Andersen, 2013). In a quite different type of study, Arlie Hochschild (2004) analyses the work-life balance technology 'Family 360' and shows how management performance today includes measures of relationships to the manager's parents, spouse, children and close friends (see also Lewis & Simpson, 2007). Susanne Ekman (2013) discusses how task-specific concerns in interactions between managers and employees have been replaced by relational intensity where managers and employees try to seduce each other.
Anders Bojesen and Sara Louise Muhr (2008) have investigated how passion at work might lead to a kind of overinclusion where the employees 'end up experiencing the love that the organization shows transformed into a unifying act of assimilation-wanting to own you; absorb you, direct you to its needs-all in the name of love ' (pp. 84-85). André Spicer and Carl Cederström (2010) formulate it in psychoanalytical terms: 'Figures who are completely subsumed and passionately attached to an organizational love object can prove to be a serious risk for the organization. This is because they see no boundaries to the love relationship and will frequently engage in excessive behavior that disturbs the smooth functioning of the organizational machine' (p. 161). And also the sociologist Zygmunt Bauman is aware of some of the effects of the intimization of work. He writes: 'The job is never finished, just as the stipulations of love and recognition are never met completely and unconditionally. There is no time, successes tend to be forgotten a moment after being scored. Life in a company is an infinite succession of emergencies …. This is an exciting and exhausting life; Exciting for the adventurous, exhausting for the weak-hearted' (Bauman, 2008, p. 130).
The question is: What is this new organizational cultivation of feelings and emotions an answer to? To answer this question, it is necessary to structure our paper into two parts: First, as essential theoretical preparation, we discuss more fully the change in the form of membership alluded to above as a move from formal membership to membership by self-enrolment (this draws upon Andersen, 2013, pp. 38-106 & Andersen & Pors, 2016. In the first form of membership, expectations of the organization are secured by the distinction between role and person, which is drawn within a communicational medium characterized by means-ends rationality. The distinction between role and person secures organizational expectations by assuming a stable set of organizational objectives and by leaving the private individual 'outside' the role. The employee, we might say, derives their authority from the organizational objectives demonstrably embodied in their formal role. Indeed, in this form, the private person, complete with their emotions, is observed to be a source of instability or noise (stereotypically associated with all things feminine, working class, foreign, childish, ethnic, gay, etc.). If the private 'person' side of the distinction 'intrudes' into the role side, this undermines authority. In the second form, the person is expected to enrol herself and to do so recurrently within an organizational context framed as ever-changing. Here, emotions acquire a new value as a source of motivation, responsiveness and vitality and so can no longer be excluded via the conventional role/person distinction. In this context, the distinction between role and person cannot secure (ever-changing) expectations, and this is compensated for by what we call the authenticity of the selfenrolment, that is, the enlistment of certain aspects of 'the person' that are excluded within formal membership. In an important sense, it is a misnomer to refer to selfenrolment as a new 'form' of membership, since strictly, it is an informalization (perhaps a de-formation but certainly a transformation) of formal membership. The clear conceptual distinction between role and person is blurred, and with this, blurring comes a mobile indistinction between personal feelings and work duties that can be described as liminal. Arising in response to this 'puncturing' of the formal zone of indifference, the new emphasis on authenticity can be understood as compensating for the authority that is inevitably lost through this transformation. As will be explored later, this shift constitutes a qualitative change in the communicative medium of meaning, a change that gives new value to forms of symbolic communication associated not with the employee as a formal and hence rational subject, but as an emotionally authentic vital subject (Stenner, 2017b). This is expressed both in a transformed conception of the organization-which mutates from 'generalized other' into 'significant other'-and of the 'self' of the employee-who must recognize herself through the gaze of the organization as passionate, self-governing, change embracing and playful. Both must now display that they operate within a mobile zone of intensive engagement.
Second, we will show the emergence of new media of symbolic communication, which are suited to these displays of authentic and vital subjectivity. Based on a semantic analysis of the recent literature on emotions in organizations, we identify how this growing semantic on the organizational cultivation of feelings and emotions can be differentiated into four distinguishable but closely related media of transition. Appropriate to the premise of permanent change highlighted above, we call these 'media of transition' because each is characterized by a concern with and for transition. Each in its own way is also preoccupied with managing the problem of authenticity, which arises along with the shift to self-enrolment under premises of permanent change. The four semantics identified in our analysis are (1) romantic media of transition (which facilitate the relational transformations entailed by a concern with love and passion), (2) pedagogic media of transition (which facilitate personal development and the emotional issues that this raises), (3) therapeutic media of transition (which facilitate movement beyond 'stuck' existential circumstances, and the emotional fears and phobias these entail) and (4) ludic media of transition (which encourage change through playfulness and the emotions of joy and excitement associated with play).
Finally, in our conclusion, we offer some theoretical reflections on these findings and point to the value of considering these semantics as liminal media of transition. Management of liminality is demanding and leads towards a trembling organization haunted by the incommunicability of authenticity.

| FROM FORMAL MEMBERSHIP TO MEMBERSHIP OF SELF-ENROLMENT
Many have pointed to the fact that change towards more change has become a discursive imperative in modern organizations since the 1980s. The temporality has evolved from change on a background of stability to a temporality where 'constant change' is seen as the only stable background (Andersen, 2013, pp. 27-38). It is this new regime of temporality that challenges organizational membership to shift from formal membership to membership by self-enrolment. This shift is carefully studied by Niels Å. Andersen focusing on the semantic history of the public employee in Denmark from 1860 until today (Andersen & Born, 2001;Andersen, 2013, pp. 38-106;(Andersen & Pors, 2016, pp. 149-181;Andersen & Pors, 2014;Andersen & Born, 2007Andersen, 2007). It is a temporality of perpetual emergence.
Furthermore, under conditions described as change on a background of stability, the organization functions as a motor of stability against which 'private' motives show up as relatively variable and unpredictable. Formal membership secures stability of expectations by assigning stable generalized motives to each member, and these motives establish the member's relevance for the organizational communication. In short, the organization is the author of the script defining the member's role (Luhmann, 1996;Andersen, 2013, pp. 40-44). This does not mean that role and person are fully independent. They of course influence one another. On the one side, the person who is taking the role might give it a personal touch, for example, being a teacher known for a particular style, while on the other, the role might end up colouring the personality; for example, we can recognize the doctor role in the person by her habit of always having ready answers. Naturally, there is also always an informal side to formal membership, and generalized motives have probably always been strategically particularized. Dickens (2008), for example, describes a certain Monseigneur as having 'the truly noble idea of general public business, which was, to let everything go on in its own way; of particular public business, Monseigneur had the other truly noble idea that it must all go his way-tend to his own power and pocket. Of his pleasures, general and particular' (p. 101).
The point of formal membership is not that private desires do not exist but that communicative effort is made to draw a boundary that recurrently creates this category of feelings and includes them only on the 'private' side of the constructed divide, as a 'personal' face concealed by the mask of office. It is on this semantic basis that an ideal of public rationality, symbolizing stability, comes to set up contrasts with privatized emotions, symbolizing forms of instability. Historically speaking, such forms of 'emotional' instability have been typically and prejudicially associated with women, children, gays and indeed practically all those categories of humanity perceived to fail to reach the standards of the 'Victorian' white bourgeois male. Mr Stryver, for example, quickly explains to Mr Lorry that he 'cannot control the mincing vanities and giddiness of empty-headed girls; you must not expect to do it, or you will always be disappointed' (Dickens, 2008, p. 141). What matters is that through formal membership, with its distinction between role and person, such disappointed expectations can be avoided. If the role establishes and secures more stable expectations, then the organization need not concern itself with the changeable private motives of persons and need 'say no more about it' (p. 141).
The notion of 'permanent change' challenges the distinction between role and person that constitutes formal membership. When the organization becomes defined by its ability to constantly be something other than what it is, everything in the organization gets cast in this light. It is no longer sufficient for the membership to slowly adapt over time because defined memberships pose an obstacle to change. Even though memberships can be changed over time, changing them still requires a decision. Suddenly, what was once the function of formal membership, that is, to serve as the memory of who is member and on what premises, is now seen as dysfunctional. In times perceived as ever-changing, formal membership decreases the organization's ability to preserve itself by means of change.
Instead, a need develops for a form of membership that can keep changing and that can 'go with' and motivate change, rather than resist it. One solution here is membership of self-enrolment (Andersen, 2013, pp. 46-48;Andersen & Pors, 2016, p. 154). This membership establishes a paradox and contains an inherent state of unrest that prompts a process of discovery in organizations. Basically, this form of membership says, 'We decide to grant you membership on condition that you constantly enroll yourself and define and redefine the conditions of this membership'. In effect, this means that membership is contingent on the member agreeing to write and rewrite the script of their own role.
Giving organizational authorship to members as a condition of membership requires them to draw upon aspects of themselves that were excluded as private under the regime of formal membership, and this in turn requires the organization to take the new form we have described. The membership of self-enrolment copies and re-enters into itself the formal membership distinction between 'role' and 'person' on role side of the difference. That means, on one hand, that the difference between generalized motive and person is maintained. On the other hand, however, the generalized motive becomes personalized. The generalized motive is now defined as a responsibility for personal self-motivation, and this is part of the same process by which the generalized other is transformed into a more personalized 'significant other'. In other words, the generalized motive is now defining an expectation that the person who is introduced into the organization will independently define the generalized motive binding the individual as if the motive were motivated by the organization. The general motive is transformed into a demand for a particular articulation by the individual member and fitted to the immediate situation. At the same time, the individual member commits to justifying the particular articulation using the organization's motive as a point of reference. The person is expected to be personally motivated to motivate herself as if her motivation were that of the organization. The member must still take the perspective of the organization, but now, that attitude has changed.
Many questions might be raised here, but we are concerned with one question: How do you qualify for a membership of self-enrolment? How does the organization distinguish between satisfying and not satisfying self-enrolment? These problems were of minor relevance to formal membership, but a problem of true and authentic self-enrolment emerges as soon as the 'personal' is no longer excluded as a private matter about which the organization must show indifference. That question becomes urgent because the difference between being and not being a member of an organization gets left to each employee's own self-enrolment, and this is a matter that has hitherto exceeded the purview of the organization. This creates a fundamental insecurity in organizations regarding the authenticity of the employee's own selfenrolment. Is she truthful now? Does she mean what she says? Is this a case of genuine engagement? I can see that she is making an effort to work with commitment and engagement, but does she really understand what we are doing here? And the employee is asking similar questions: Am I really happy being here? Is it the work I love or is it my colleagues? Can I be the one I want to be when I am in this organization and in this job? Therefore, am I genuine and authentic in my own membership?
The loss of expectation security that accompanies the fall or transformation of the difference role/person needs somehow to be compensated for. Because it now becomes the responsibility of the single employee to constantly (re)enrol herself and to motivate herself in the image of the organization, the security of expectation can only be established by 'the source' of enrolment. The organization comes to depend on this 'core' of authentic interest that each employee must show in the organization. What is required in this 'transient' context is some means for the organization to observe and manage the emotions of its employees. We will argue in Section 4 that this challenge of observing and managing emotions is being met by the use of a host of new semantic media that we call liminal media of transition. Each of these media is specialized in generating, observing and managing emotions (and affectivity more generally), and, before they featured in work settings, each had its source in contexts of psychosocial transition or becoming. First, however, it is necessary in the next section to present some findings in which we identify these media as semantics within the discourse of those who study and write about organizations.

| ROMANTIC, PEDAGOGIC, THERAPEUTIC AND LUDIC SEMANTICS
The material in this section comes from an analysis of a rather heterogeneous mix of texts published since 1990 including consultancy books and other prescriptive literature on organization and emotion. Based on a reiterated process of (re)reading and (re)coding, an analytical decomposition of these books resulted in the identification of four distinguishable discursive themes of relevance to our topic. As explicated in full below, each theme implies a distinct semantic grid of intelligibility through which the issue of emotions within organizations shows up in a distinctive manner. We do not claim that no additional themes exist beyond the corpus of books we analysed (an exhaustive analysis was beyond the scope of this project), but we are confident that the themes identified below serve well to characterize the literature we analysed. It will of course be of interest to identify further variants in future research. For a more extensive introduction to this type of discourse analysis, which is grounded in Niklas Luhmann's concept of semantics, see Andersen (2003) and Andersen (2011). Here, it is important to avoid a possible confusion. A main guiding difference in Luhmann's work is semantic/structure. In this distinction, structure marks the differentiation of society (hence, one might say that a society that is functionally differentiated has a different structure to one that is hierarchically differentiated). These societal structures are not in themselves observable. They are only observable through the semantic history representing a kind of trace of past communication.
Here, by contrast, we are using another difference that is also present in Luhmann's work, but in this second difference, semantic is identified with structure and not differentiated from it. Namely, we observe with help of the difference communication/semantic and in this frame semantic is structure representing the stock of repeatable forms of meaning and communication is operations constantly vanishing as they emerge. This difference might not be developed to the same extent by Luhmann, but it is anyway implied in the difference semantic/structure. Social systems are recursive networks of communicative operations. As operations, the communicative elements are things that occur and then perish, giving rise in turn to the next communication. From this perspective, the problem becomes that of how any order or form or structure is possible at all such that one event might connect with another to permit the repetitions needed to compose a system. It is in this theoretical context that the concept of structure as semantics takes on new significance. Where communicative operations constantly form meaning, meaning can, over time, develop into condensed meaning. Communication is thus able to develop structure, which condenses meaning into already actualized forms that are disconnected from the immediate situation of actuality. Condensation is the process of movement from potential to actual whereby a multiplicity of potential meaning is captured into a single form, which-once actualized-becomes available to another (as yet unspecified) communication. Semantics is defined, therefore, as the stock of generalized forms of differences (e.g. concepts, ideas, images, and symbols), which can be used in the selection of meaning within the communication systems (Luhmann, 1993, pp. 9-72).
Below, we discuss four 'semantics' that were distinguishable in our data set, each a relatively coherent 'stock' of meaningful concepts, ideas, images and symbols. Without going into the technicalities here, different analytical traditions would call these 'discourses' or 'themes' or 'narratives' (Andersen, 2003;Stenner, 1993).

| The passionate employee: A romantic semantic
The first semantic about the relation between membership and emotions is the semantic of the passionate employee. This semantic articulates and constitutes a problem of authenticity but does not solve it.
An example of this semantic, introduced earlier, is the book Passion at Work from 1998, written by marketing professional Kevin Thomson whose basic assumption is that 'one of the strongest emotions of all is PASSION. It is a motivator which drives us to incredible limits' (Thomson, 1998a, p. 3). For Thomson (1998a), passion is the single most important quality in an employee because 'Our emotions are what drive us and our organizations to incredible feats. Let them loose!' (p. 6). And Thomson (1998b) links this to a particular trend in the new spirit of capitalism: 'Capitalism as we know it would be better described as "emotional capitalism." (…) The bottom line: emotional capital is the stuff of dreams. It is energy, drive and commitment invested and held in the hearts of everyone connected with business' (p. 13). Essentially, Thomson argues that the basic criterion of being member as an employee of an organization is passion and involvement in the organization. In this semantic, the real organization is observed as emotional relations, and the formal organization becomes the counter concept to the real: 'It is the hearts and minds of everyone in that company that creates its personality' (Thomson, 1998b, p. 22). Here, the organization does not exist outside the emotional relations that constitute it. Passion basically becomes the substance of an organizational ontology. It becomes crucial for an organization to create itself as the significant other of its employees (and customers): 'In today's society employees are no longer willing to be part of a rigid, unfeeling bureaucracy. People want to work in organizations they like, and are like them' (Thomson, 1998b, p. 25). Through this lens, it is important that not only employees but also the organizations are attributed emotions: 'Companies are living entities with emotions just below' (Thomson, 1998b, p. 23).
Often, of course, the phrase 'the organization' is taken to mean 'management' as distinct from the 'labour force', as Fox and Amichai-Hamburger (2001) suggest: 'The Organization must also see itself as having a warm, open relationship between its management and employees' (p. 88). In representing the character of the organization, and hence its expectations, management must engage the positive emotions of its employees. The challenge is: 'How to tap the emotional components of human behavior?' (Fox & Amichai-Hamburger, 2001, p. 87).
Dennis K. Mumby and Linda Putnam articulate a very similar semantic of the passionate employee in several articles from the beginning of the 1990s, though writing from the perspective of feminist management critique rather than marketing. In this account, the old formal (Dickensian/Weberian) counter-concept, which serves as 'other' to their 'caring' organization, is further developed as having hierarchical features expressive of patriarchy, and the organization of the future is observed to be a heterarchy rather than a hierarchy. Heterarchy is seen as a fluid social order in which goals and values are flexible and unpredictability is a fundamental condition. There can be no fixed roles that members can assume. They must therefore take responsibility and care for one another: 'Nested in an environment of caring, members balance the demands of differing values, goals, and relationships to make the group a place where all members feel comfortable and achieve their individual aims' (Mumby & Putnam, 1992, p. 475). Here, the employee is required to work based on a desire to connect and to develop relations. Rigid descriptions of different professions do not work in such an organization: 'The needs of the person or the relational context would guide feeling rules, rather than the occupational identity' (Mumby & Putnam, 1992, p. 477).
Here, again, the membership criterion becomes particular feelings and emotions rather the formal competencies. Putnam and Mumby (1993) call this demanded feeling 'work feeling': 'Work feelings ( …) contribute to the building of community by forming a bond of interrelatedness' (p. 52). Work feelings are seen as the product of a dialogical process in which employees are mutually responsive: 'These work feelings are spontaneous and emergent; they are not directed to particular instrumental goals, but rather are outgrowths of relationships and interpretive schemes. (…) they encourage interrelatedness and mutual understanding. (…) They aid in "bounding" emotions because a person uses these emotions to recognize the other person's subjectivity and to promote responsiveness to others. (…) Employees form a dialogical relationship' (Mumby & Putnam, 1992, pp. 477-78). So work feelings are perceived to be the authentic expression of the employee's real feelings and identity. In this way, the old cold and authoritarian 'generalized other' familiar to Dickens and Weber is replaced by a warm and loving female significant other that is attuned in a caring manner to the situational specificities and cultural contingencies of relational nurturance, and it is the attitude of this significant other that employees are invited to adopt in coordinating their working lives.
But how is the passionate, caring organization able to draw a difference between authentic passions and work feelings on the one side and non-authentic expressions on the other side? In the semantic of both Thomson and Mumby and Putnam, it is crucial. Everything in the organization seems to depend on passionate employees. But when an employee says: 'I am so glad to be here. I really love this place and my work', how can managers or colleagues know whether this expresses true feelings or just expresses that the employee knows what his manager wants to hear? So this semantic of the passionate employee constitutes a significant problem of authenticity, which did not exist in the formal bureaucracy, where the membership criterion simply was to fulfil the formal role. Within the semantic of passion, there are at least three techniques for proving authenticity.
The first technique we might call 'the proof of the opposite': authentic expression must not be a mere play act or game or some other instrument of deception or alienation. Mumby and Putnam especially articulate this. Authentic work feelings are here distinguished from 'emotional labour', which involves emotions that by definition are false (see Hochschild, 1979). Emotional labour 'refers to the way individuals change or manage emotions to make them appropriate or consistent with a situation, a role, or an expected organizational behavior' (Mumby & Putnam, 1992, p. 472). Emotional labour refers to those aspects of care or service work, which prescribe specific emotional behaviour and expression in employees, for example, smiling or speaking in a soft voice. Here, emotions serve 'instrumental goals and task functions' (Mumby & Putnam, 1992, p. 472). In this context, they quote the following statement from Ferguson: 'Like prostitutes, flight attendants often estrange themselves from their work as a defense against being swallowed by it, only to suffer from a sense of being false, mechanical, no longer an integrated self' (Mumby & Putnam, 1992, p. 472). So when the managers hear an employee saying 'I am so engaged. I just love my work', then this statement is considered inauthentic if it can be attributed to roles or instrumental goals or functions in the organization. If we are not able to attribute the statements to the roles and instrumental goals, then we might attribute them to the employee as an authentic person. This contrast, of course, does not fully solve the problem of proving authentic passion but can perhaps cover it up for a while.
The second technique is to turn the difference between true passion/not true passion into a continuum. Richard Chang (2001) tries in his book The Passion Plan at Work to solve the authenticity problem by developing what he refers to as a 'Passion Scale' where organizations can test themselves and their employees and find out if they score high or low on passion on scales like the following (p. 5): 'malaise' -'ambivalence' -'interest' -'enthusiasm'.
Or alternatively: In a variation of this technique, Kevin Thomson follows biblical tradition and distinguishes between 'deadly' and 'dynamic' emotions, identifying 10 of each. His register of dynamic emotions includes things like obsession, commitment, delight, love and trust, and these are contrasted with deadly emotions such as fear, apathy, anxiety, envy and hatred (Thomson, 1998b, pp. 23-24). Again, these techniques can at best cover up the problem of authentic passion, and at worst, they multiply it since now we must distinguish not only true passion from false but also 'low' from 'high' passion and 'deadly' from 'dynamic' passion, since each member of these categories can also be more or less authentic.
The third technique is 'to listen'. Shaul Fox and Yair Amichai-Hamburger argue that the management must actively 'recognize' employee emotions. This is a way to make sure that passion is expected and welcomed. Authentic emotions become more likely when the organization creates a space for listening in a loving manner: 'Listening to people's hopes and dreams, encouraging them to express their fears and doubts, and enabling them to present their visions of the future of the organization may lead them to feel that management is attentive to their concerns and respectful of their feelings' (Fox & Amichai-Hamburger, 2001, p. 90). Once again, while this approach might increase the likelihood of 'authenticity', it can by no means serve as a final test. Indeed, it risks redoubling the problem in a different way: now, it becomes necessary also to establish if the manager is in fact authentically interested in listening to the employees' feelings.
The passion semantics articulate emotion in relation to the question of authentic self-enrolment. The employee's relationship with the organization must spring from within as authentic engagement. Thus, in this semantics, the organization comes to rely entirely on their employees' authentic feelings for the organization. The sought-after engagement is not supposed to be external, instrumental and commodified but spontaneous, relational and expressive of the self-identity of the individual employee. The organization wants to be loved for what it is, and this makes it dependent upon the authentic engagement of its employees. The implicit and explicit expressions of emotion become signs of either the absence or presence of engagement, which is why it becomes so important for employees to be given the opportunity to express their true feelings. Management becomes a continual appeal to emotion, the creation of possibility for engagement, and the recognition of employee emotion.

| Learning to take responsibility for one's emotions: A pedagogic semantic
The second semantic about the relation between membership and emotions is the semantic of sentimental education (with reference to Flaubert's famous novel (2010 [1857]). This semantic inherits the problem of authentic passion and engagement and proposes a solution where the single employee takes responsibility for his emotional expression. As Montgomery (1985) articulates in relation to employees: 'Begin to accept and assume ownership for your anger' (p. 22). As we shall see, this approach also quickly finds itself caught up in paradoxes of authenticity.
Daniel Goleman who became a major name within this field believes that the modern organization must strive to constitute a partnership among employees and between management and employees. He shares with the semantic of the passionate employee the ideal of an organization kept together by emotional commitment (Goleman, 1998, p. 119). He refers to the organization as 'the organizational marriage' and the relationship between manager and employees as 'the vertical couple' (Goleman, 1998, pp. 212-13). In this way, Goleman starts by acknowledging the emotionalization of both sides of the organization (i.e., the emotionalized employee who takes the attitude of what we have called the organization as 'significant other'). This enables him to argue that, as in every marriage, 'If both do well emotionally-if they form a relationship of trust and rapport, understanding and inspired effort-their performance will shine. But if things go emotionally awry, the relationship can become a nightmare and their performance a series of minor and major disasters' (Goleman, 1998, p. 213).
But Goleman does not simply demand true passion and engagement from the employee. His approach is educational: 'The good news is that emotional intelligence can be learned' (Goleman, 1998, p. 315). The aim is to support positive emotions through the building of emotional competences in managers and employees. This involves a difference between emotion as cultural and feelings as individual. Feelings become separated from their social manifestation, and the social manifestation cannot simply be seen to represent feelings authentically. This distinction makes it possible to view emotions as something that can be acquired and cultivated. Elena Antonacopoulou and Yiannis Gabriel (2001) have elaborated on this difference and their perspective is that employees need to learn positive emotions, which then support organizational change: 'Learning about one's emotions provides a useful starting point for recognising what causes these emotions and how they may be worked on, reconciled with and corrected' (p. 445).
Learning about emotionality liberates employees so that they are no longer victims of their feelings. They go on to argue that 'learning, at its highest, has a liberation quality, defeating ignorance, fear and superstition, unleashing potential and developing new ideas and outlooks. It can stimulate emotions of hope, love and solidarity as well as desire for a better order' (Antonacopoulou & Gabriel, 2001, p. 445). Or as Goleman (1998) frames it: 'At the individual level, elements of emotional intelligence can be identified, assessed, and upgraded (…) Emotional intelligence refers to the capacity for recognizing our own feelings and those of others, for motivating ourselves, and for managing emotions well in ourselves and in our relationships ' (pp. 315, 317). Thus, the basic assumption is that the authentic expression of one's feelings does not necessarily effect hope, love and solidarity. Knowing about emotionality and its effects creates better conditions for passion and engagement. They argue that managers of organizational change typically focus on negative feelings such as opposition and fear but that organizational change can be better supported by teaching employees positive emotions.
The basic premise is that if emotions are educated, then this will do away with the problem of distinguishing authentic passion by turning it into a matter of staff development. The semantics of sentimental education contain a notion of a self, capable of splitting itself first into feelings (internal operations) and emotions (external expressions) and then into desirable and undesirable emotions. One has to learn to assume responsibility for the effects of one's emotional expressions. Positive feelings will likely follow positive emotions (for a Foucaultian analysis of the discourse on emotional intelligence, see Hughes, 2010, and for a critique of positive psychology, see Greco & Stenner, 2013). This semantic does not supply an answer to how to be authentic and honest and how to distinguish honesty and dishonesty in someone else. In fact, it complicates the question of authenticity by constructing emotions as anyway a public matter whose effects must be controlled regardless of their relationship to internal feelings. Its solution seems to be training in how to manage and manipulate emotions as an organization medium, and from there, feelings too are expected to fall into place. It seems clear that a developed proficiency with operating a 'language of emotions' on the part of employees will not in itself solve the problem of authenticity. Indeed, from one angle, this shows up as training for professional dishonesty.

| When pain is considered more authentic than passion: A therapeutic semantic
The third semantic about the relation between membership and emotions is the semantic of care focusing on pain. This therapeutic semantics relates to the negative emotions that are produced in organizations as the side effect of passionate work.
Peter Frost contributes to this development. His starting point is the passionate organization in which self-enrolment is codified in a highly intimate way: 'And as people are increasingly invested in their work rather than in their personal lives or communities, organizations more and more become the stages where people's hopes and expectations play out. Indeed, the modern work organization has become a zone where life is increasingly lived for full meaning-encompassing the full spectrum of human emotions and experiences' (Frost, 2003, p. 102). Given this expansion of life investment in the work place, the organization is under an obligation to care for the pain that can arise when employees invest themselves fully. This is not mere altruism, since Frost (2003) sees the emotional pain of employees and managers as a form of toxification of the organization: 'We live in times where there is much pain and suffering in and around organizations. There is much to be learned about toxicity in organizations and how best to handle it' (p. 226). Frost writes himself up against Goleman like this: 'While Goleman's early work, at least, focuses on positive effects of high emotional intelligence, what became most salient to me (…) was that contagion could be positive or negative, and the emotions experienced by one person might also be absorbed by the person who attempts to help' (p. 4).
The individual employee can become toxic and thus become a toxin for the organization: 'Over time, a single employee can do any or all of these kinds of things, fostering a workplace that is unhappy indeed, because employees who are uncivil are toxic' (…) 'Toxicity is produced when an individual's attitude or an organization's policies, or both, fail to take into account the emotional attachment people have to their contributions to work' (Frost, 2003, pp. 51, 56). The solution to the growing toxicity problem is the compassionate organization, equipped with specialist 'toxin handlers': 'The compassionate organization: -sees a clear link between the emotional health of employees and the organization's bottom line -recognizes and rewards mangers who are good toxin handlers -hires for attitude as well as technical skill -maintains a fair-minded workplace, recognizing the direct connection between consistent values such as loyalty, responsibility, and initiative, and the health of the organization overall -has intervention strategies in place for times of distress or change (…). -Build a company culture that values compassion and community as beneficial to productivity and to people' (Frost, 2003, p. 28).
Compassion represents the central form of care: 'We identify compassion as comprised of three interrelated elements: noticing another's suffering, feeling empathy for the other's pain, and responding to the suffering in some way' (Frost, Dutton, Maitlis, Lilius, & Worline, 2006, p. 846). Compassion is seen as interpersonal work, which includes listening and the creation of a space for pain. Furthermore, it is not just management that needs to listen. Rather, management is responsible for helping employees listen to one another: 'Pain in organizations can often be overlooked or misinterpreted unless the listener actively engages empathically and commits to listen for emotions in the message that those in pain allow themselves to send' (Frost et al., 2006, p. 850).
Compassion is the central concern of toxin handlers: 'The work of toxin handlers is about responding compassionately to pain in their organization in order to either minimize or prevent it, identify it, contain it, remove it, or find ways for people to live with it constructively' (Frost, 2003, p. 62). Frost (2003)asserts that a toxin handler is responsible for 'listening, holding space for healing, buffering pain extricating others from painful situations, transforming pain' (p. 63). Listening to pain also means to create human bonds within the organization: 'Listening with compassion to someone else's pain, providing a moment of human connection' (Frost, 2003, p. 63). As a listener, the toxin handler also becomes a pain manager who provides the 'attention that allows the person in pain to feel heard, respected, and helped' (Frost, 2003, p. 66).
This notion of listening as the management of toxins creates a particular variation of the question of inclusion/exclusion. When the toxin handler listens to the toxic and pained employee, the employee is considered emotionally excluded, perhaps self-excluded. Painful emotions signify that the employee does not feel appreciated and respected. Listening does not simply lead to inclusion but rather to the organization's ability to contain the excluded. The organization needs to be able to contain pain. This produces a re-entry of the distinction inclusion/exclusion so the excluded becomes included as excluded. So while the pain-stricken employee has not successfully self-enrolled, the effort on the part of the organization to 'contain' the pain indicates a recognition of the troubled enrolment effort. The one who does not feel loved is recognized, and her love pains observed, and she thus becomes included in a compassionate collective.
Toxin handlers/pain managers need to be able to read and anticipate the other's emotions: 'She can visualize and feel the anger, fear, or demoralization that derogatory statements or actions would trigger if they were directed at people personally. The handler is often particularly quick at recognizing such situations and stepping in to serve as a buffer' (Frost, 2003, p. 71). Compassion can help those in pain reconnect with their work and recreate their sense of worth. Compassion might even transform the individual's self-relation: compassionate acts 'can help transform people's sense of themselves, change the way they relate to their colleagues, and shape the way they view their organizations' (Frost et al., 2006, p. 850). The inclusion of the excluded, in short, can lead to the employee's self-inclusion.
However, this work with toxins is not without its dangers. Toxin handlers can be infected by the pain they see and allow it to overwhelm them (Frost, 2003, p. 90): 'The people who handle the emotional pain of others might themselves be vulnerable to that very same pain. In effect, handling emotional toxins can be as hazardous as working with physical toxins' (Frost, 2003, p. 4). Again, we see the metaphor of contagion, which indicates that the individual's pain is relevant to the organization. Feelings of pain cannot be isolated. Therefore, toxin handlers have to be treated with compassion themselves, not least because toxin handlers can be affected by the suffering because they come to depend upon the emotions generated through the work of toxin: 'Handlers for their part, bring some of this suffering on themselves. They get careless about their own well-being. They pay more attention to others than to themselves. They become consumed with worry about whether they are doing the right thing. They become addicted to the "fix" that helping other people can deliver' (Frost, 2003, p. 105). Ultimately, a toxin handler can become a toxic employee and a toxin for the organization: 'Handlers of toxins can become so infected with other's pain that they, in a real sense, become "toxic" themselves, and begin inflicting pain on others' (Frost, 2003, p. 8).
Authenticity becomes an important question but not in the same way as in the semantics of passion. It is not a question of finding proof of love. In fact, misrecognized or troubled engagement is observed as more likely, and indeed more authentic than 'true engagement'. What is articulated here is a question of listening to what cannot be heard. The toxin handler is assigned the role of listening to the suffering of employees, allowing space for this suffering, feeling the pain of employees and showing it through compassion. The task of the toxin handler is therefore to 'listen for emotions in the message', that is, to hear non-communicated emotions and pain. Not only this, the toxin handler must communicate the emotions of non-communicated emotion back to the employee so that the employee feels listened to, respected and helped. This is so even though, or perhaps especially when, they have not explicitly asked for help and indeed may have offered something more like toxic frustration or even aggression. Communicating one's listening constitutes the compassionate and detoxifying act. But this challenge of communicating the non-communicable doubles the problem of authenticity, particularly when the empathy of the toxin handler is itself inherently questionable, prone as it is to inhabitation by inauthentic motives to be perceived as a 'saviour'. What if the pain manager were calling out pain even where it did not previously exist? What if the alleged toxins, spreading throughout the organization, were the projection of a toxin handler whose mode of operation is always to expect toxicity especially where it does not appear to exist because it is expressed otherwise.
This semantics, in short, generates organizational expectations about pain. The perspective of pain expects pain. Pain becomes the flipside of emotional engagement. Pain does not constitute engagement but bears witness to it, that is, as rejected or misrecognized engagement. Thus, in a strange way, pain is assigned value. Pain becomes a sign of authenticity, which stands in for love. And which is perhaps considered more authentic and direct than more explicit love declarations.

| The emotional meanings of play: A ludic semantic
The fourth semantic about the relation between membership and emotions is the semantic of play. We see a ludic semantics about emotions, which suggests that emotion is something we should play with.
An organizational semantics of play has developed during the last two decades or so, and its focus is on the relationship between play and emotion. In this semantics, the problem of authenticity becomes ubiquitous and relates to the notion of total dedication. In 2000, Pat Kane, who wrote the 2005 book about the ethics of play, writes: 'So to call yourself a 'player', rather than a 'worker', is to immediately widen your conception of who you are and what you might be capable of doing. It is to dedicate yourself to realising your full human potential; to be active, not passive'. We see a vitalist idea about emotional self-realization through a playful approach to the world. If one is seriously playing, playful feelings are not only required in specific instances of play, but at all times as a general approach, a worldview, which should apply everywhere. According to Kane (2000), to call oneself a player implies being constantly playing and dedicated to play: 'Play as the exercise of human freedom and self-fulfillment, from birth to death'.
Leslie Yerke's book, Fun Work, presents a somewhat less radical perspective although the demand for dedication and authenticity is the same. The preface reads: 'Fun-loving, passionate people are just going to make better leaders than nonsense, unenthusiastic types who have no place for a bit of foolishness in their lives' (Yerkes, 2007, p. xi). Yerkes (2007)goes on to say: 'True fun is not something you choose to do, it is something you choose to be' (p. 11). There is no possible exteriority in a playful approach to work since to be playful is a mode of being something one IS. Playing allows the individual to become identical with itself; it abolishes every split in the self and creates wholeness: 'To bring your full, fun self to your work relationships, remove the layers of grudges and betrayals that insulate your heart. When your heart joins your head and hands in work, you will have released one of the most powerful forces in your life-the energy of your whole being' (Yerkes, 2007, p. 43). 'Fun is losing your self in the work' (Yerkes, 2007, p. 224).
Accordingly, play becomes the form that fulfils the passionate membership and solves love's problem of authenticity: 'The best conduit to our heart is fun. Fun makes work enjoyable, it makes us love what we do. Fun connects to love through the heart. When we enjoy what we do, we say we love our job. When our job is a "labor of love," it's less work and more fun. The more our work is fun, the more we love our work. (...) Be authentic. Bring your heart' (Yerkes, 2007, p. 175). Fun eliminates the doubt that relates to passion. Whereas passion always creates uncertainty about whether I am loved for who I am and whether I love the other for who they are and therefore creates doubt as to the authenticity of expressions such as 'I love you ', Yerkes' (2007) claim is that play demands authenticity without creating such doubts: 'Authenticity cannot be learned, it cannot be faked' (p. 109). Yerkes quotes a midlevel manager: 'Fun is having passion for your work and linking up with your colleagues' (p. 142). Yerkes also talks about loving management: 'Offer love and encouragement. The more you give the more you get. Make it your priority to support; seek out those who need it' (p. 218).
Moreover, authenticity is defined not only in terms of the social dimension but also in terms of the temporal dimension. Authenticity is perceived as pure presence in the now: 'Capitalize on the spontaneous. (…) Fun doesn't happen according to schedule. It isn't something we plan. Fun grows in a culture that fosters its existence; it springs automatically from the proper environment. Don't inhibit its existence by scheduling too tight; allow room for it to breathe and grow' (Yerkes, 2007, p. 45). Fun is spontaneous, but at the same time, the argument is that fun performs: 'We suffer from the lack of integration of fun and work' (Yerkes, 2007, p. xiii). 'Work needs fun' (Yerkes, 2007, p. 5). 'We are beginning to discover that fun belongs with work. It is my premise that fun and work naturally go together. That fun works and work pays off better when it is fun (…) When fun is integrated with work instead of segmented from work, the resultant fusion creates energy; it cements relationships between co-workers and between workers and company. When fun is integrated into work, it fosters creativity and results in improved performance' (Yerkes, 2007, p. 8). This creates a strange paradoxical figure, which subjects itself to the demand for performativity and thus directs itself purposefully towards the future while insisting that performance is enhanced through absolute presence, that is, by disregarding and suspending future performance demands.
Certain emotions, however, function as a threat to play. Feelings of stress are considered a threat to play and authenticity: 'Stress is the enemy. It reduces your capacity for engaging your fun self' (Yerkes, 2007, p. 28). Fear is another emotion, which is seen as a threat to play: 'Fear and fun are opposite ends of a continuum. Fear comes from low trust; fun comes from high trust. When we have high trust, we have fun; when we have fun, there is high trust. (…) Fear is often the thing that stands between us and what we want most. Fear creates a reaction that makes our desire elusive. What we want most is coated with our own fear. (…) To have trust, you have to believe in the future and have fun in the present' (Yerkes, 2007, p. 75). It is important to note that stress and fear are emotions that point to a dependency upon an environment.
Loyds Sandelands (2010) formulates a particular relationship between organizational change, love and play: 'Change in organizations is best taken in the spirit of love that is play. As change calls to love, the greatest changes call to the greatest love of the divine in which all things are possible ' (p. 71). This means that change is contingent upon certain emotions: 'Play is the form that love takes at the boundary between fantasy and reality where new social arrangements arise to take the place of old social arrangements (…) play is to know not by analysis via mind and reason, but by intuition via body and feeling ' (p. 72). Feelings for play become constitutive for organizational change, and this perspective draws in particular on Csikszentmihalyi's concept of flow: 'Flow denotes the holistic sensation present when we act with total involvement. It is the kind of feeling after which one nostalgically says "that was fun," or "that was enjoyable."' (Csikszentmihalyi in Sandelands, 2010, p. 74). The feeling of total engagement is thus emphasized as that which creates play.
In total engagement, the self is abolished and merges with the collective (which is why play is seen as the most profound and authentic form of love). Sandelands (2010) speaks of 'selflessness': 'In play, the boundaries that usually isolate one person from another-the identities that distinguish them as individuals-are overcome by the life of community' (p. 76). Play is simply founded on love: 'Born of love, play is the second moment of social life. It is love's bloom of creation at the boundary of unseriousness (fantasy) and seriousness (reality)' (Sandelands, 2010, p. 77). Furthermore, play's feeling of total engagement is seen as equivalent with feelings of being human and fully alive: 'In play we come to community fully human and fully alive' (Sandelands, 2010, p. 78). Sandelands continues his chain of equivalencies and speaks of 'the feelings of play' as the feeling of 'being in it with others', 'feeling of growth', 'feeling of rhythm' and about 'feeling of undergoing, of movement on the way to an unknown and undecided resolution' (p. 79). He sums up play as 'a feeling of social life' and points out that 'although the feelings are felt in person, they are not personal' (Sandelands, 2010, p. 79). Thus, feelings relate not only to the psychic system but also to the social. Play becomes the identification of the feeling of 'the deepest vitality of human community' (Sandelands, 2010, p. 81) and thus also effects 'the deepest and most lasting change' (Sandelands, 2010, p. 82). This is truly a radical figure of authenticity since it insists on the simultaneous identity with oneself and with the significant other. The individual has become one with the social.
Thus, in addition to a semantics of love, pedagogy and care that relate to emotions, we also see a semantics of play about emotions. Like the semantics of love, this semantics centres around the question of authenticity. However, as we have seen, it too is threatened by the spectre of emotions like fear, anger and anxiety, which militate against a generalized playful mindset, threatening to tip it over into something less enjoyable, like a fairground ride, which spins out of control and turns into a real disaster. It seems that these negative affects at the edge of play (especially the 'serious play' that is valorized in this semantic) can never quite be eliminated from playful situations, which perhaps function precisely to flirt with them. Furthermore, the radicality of the play semantics centres on its association with situations of transformation in which, as we have seen, 'new social arrangements arise to take the place of old social arrangements'. Such situations of becoming are not just inherently ambivalent in an emotional sense (Greco & Stenner, 2017); they are also prone to manipulation by those who have concrete interests in the outcome. Simply affirming the value of play does not resolve the problem of authenticity but, again, redoubles its significance (see Stenner, 2017b).

| THEORETICAL DISCUSSION: ROMANTIC, PEDAGOGIC, THERAPEUTIC AND LUDIC SEMANTICS AS LIMINAL MEDIA OF TRANSITION
We began our article with the concept of formal membership arguing that the ideal of constant change and transformation challenged formal membership forcing it to evolve into self-enrolment. Formal membership is the unity of role and person, defining everything personal as organizationally irrelevant. Chester Barnard and Niklas Luhmann talk about a zone of indifference implying both the organization's indifference to personal motives for being in the organization, as well as the individual's indifference to the organization's instructions and distribution of responsibilities as long as they fall within the framework of the general motive (Barnard, 1968;Luhmann, 2018, pp. 82-83, 232). The zone of indifference helps to transform the excluded problem of personal authenticity into a formal problem of authorization. An employee can talk with authority as long as she refers to the generalized motive (the role) decided by the organization. The four semantics we have analysed above show no sign of such a zone of indifference. Instead, we might talk about a zone of intense investment in which the employees and the managers mutually try to prove their authentic engagement in work and organization (Andersen, 2013, pp. 48, 171).
This shift is profound. Authority and authenticity are almost antagonistic concepts. Authority can be proved and tested. It concerns representation based on formal concepts. The employee has authority when her actions demonstrably reflect the role. The organization can observe whether the employee's action represents the role or not. Authenticity, in whatever form it takes (Stenner, 1999) might be performed but cannot be represented and so is incommunicable. It is a 'virtue' that operates, as it were, below the level of the speech act (Williams, 1971). In order for a self to represent herself, she must draw a distinction and therefore split herself, being both herself, and not herself. Authenticity is basically a paradox (Langraf, 2002, p. 160). Since membership by self-enrolment nevertheless demands proof of authenticity, this creates a particular puzzle: What communicative medium is appropriate for being shaped into the form of intense personal investment?
We have shown how the 'emotionalization' of organizations is associated with a shift in organizational membership, and that the issues of authenticity raised by this shift are being articulated in relation to semantics of love, education, therapy and play. We have thus distinguished four media of communication that have emerged in contemporary organizations as means for communicating these new forms of intense personal investment and vital subjectivity. We have opted to call these romantic, pedagogic, therapeutic and ludic media of transition. Why 'media of transition'?
We use the word 'media' because each of these semantics serves as medium for the forms proper to the new conditions of membership by self-enrolment. They are, in other words, the means by which people involved in organizations can talk about and work upon the authenticity of their feelings in a work context. In this sense, they are distinct from the communicational medium proper to formal membership, which we showed to be a discursive medium characterized by means-ends rationality.
We use the word 'transition' for two main reasons. First, because the shift from formal membership to membership by self-enrolment has taken place under conditions characterized by a new premise of permanent change as distinct from change on a background of stability. Under these paradoxical new conditions of permanent liminality (Szakolczai, 2017), organizations have come to value and expect flexibility, innovation and openness to possibility. Second, because the semantics of love, education, therapy and play that we have identified in our analysis have each been derived from a source domain that is directly associated with liminal transitions, before being newly applied in the work domain. First, the romantic medium of transition has its source in relationship transformations entailed by a concern with love and passion. Second, the pedagogic medium has its source in the education of children and hence in practices that facilitate personal developmental transitions, along with the emotional issues these raise. Third, the therapeutic medium has its source in those practices that facilitate movement beyond 'stuck' existential circumstances, and the emotional fears and phobias these entail. Finally, the ludic medium of transition has its source in spontaneous playfulness and the games that build upon this, both of which have long been associated with ontogenetic transformations and cultural rites of passage (and evoke the emotions of joy and excitement associated with play). In sum, before they featured in work settings, each medium had its source in contexts of psychosocial transition or becoming. For this reason, the four media of transition share in common a clear concern with transitional phenomena combined with a concern with the affectivity that is inevitably associated with transition (Stenner, 2017a). These are the transitions involved in new relationships, in new learning, in situations of life difficulty, rupture and transformation, and in ontogenetic development and ritual. Together, when adapted and applied with the new domain of organizations, they comprise a communicative medium appropriate for being shaped into the form of intense personal investment that is required under conditions of membership by selfenrolment.
Our final theoretical contribution is the suggestion that as liminal media of transition, these organizational semantics are special in at least two other ways: (1) As semantic structures they are anti-structural, resisting any fixed general other, and (2) as a stock of generalized forms, they are loaded more with symbols than with concepts. Both pose challenges for organizational communication.

| Antistructure
!?A3B2 twb=.29w?>At the start of Section 3, we articulated Luhmann's concept of semantics and showed how semantics are the form that structure takes within social systems: meaning is condensed into a form that lends coherence and predictability to communication.
Returning to the four semantic media of transition, it is striking that they have a quite different relation to structure. Elsewhere, we have argued that semantics do not always and only function structurally to enhance the expectability-and hence 'connectability'-of communication (Andersen & Stenner, 2020). Sometimes, the expectation structures facilitated by the historical sedimentation of semantics can prove counter-productive to social systems. In situations of transition, for example, it becomes necessary to suspend and perhaps even deconstruct expectations that have become redundant and counterproductive. In such circumstances, semantics are required that can operate in a manner that is not just structural but antistructural or 'antiform' (Luhmann, 2000, p. 34). It becomes necessary to suspend or disrupt ordinary expectations, to challenge outdated meanings and assumptions and to create new possibilities for meaningful connection. This is indeed the case in self-enrolment. As a form of membership, self-enrolment is not simply a new form but a fold or a re-entry of formal membership, where enrolling yourself essentially consists in resisting any temptation to accept a predefined role. Self-enrolment involves a continuous work of suspending formal roles while always being open to new needs of the organization as significant other. Here, transition has become a permanent state.

| Symbols
The second special characteristic also relates to Luhmann's definition of semantics as the available stock of generalized forms of differences. He is open to the idea that meaning need not be solely conceptual (and hence representational) but can be condensed into a variety of forms such as images and symbols, which are certainly meaningful, but not necessarily in the representational sense discussed above. For the most part, however, the focus in Luhmann's (1993) semantic analysis is in fact upon the condensation of meaning into concepts. A concept, put succinctly, is a unity of the difference between concept and counter-concept (Andersen, 2011). It is a condensation and generalization of a multiplicity of meanings and expectations, which can in turn be contrasted with a counter concept. Compared with more stable situations, circumstances of liminal transition tend to be loaded with presentational symbols (often embedded in rituals) and are often less conducive to standard conceptual discursive thought. This is because they entail experiences of becoming in which those involved have, as it were, left behind one relatively clear universe of discourse but have yet to acquire the conceptual framework of the new world still in process of emergence (Stenner, 2017a, pp. 151-195). It is hard to 'represent' when that which is to be represented has yet to clearly emerge, and such circumstances precisely call for new forms of sensemaking. Our family of four semantics are indeed well described as stocks of symbols and rituals rather than of concepts. These media are rich with non-conceptual, nonrepresentational forms of presentational symbolism (Stenner, 2017a, pp. 76-80), including ritual, art, theatre and games (Stenner, 2017b), and these in turn are tightly connected with questions of affect and emotion. Where concepts stress a difference and a distance between signifier and signified, symbols, in the sense we are using them here (following Luhmann, 2000), stress unity. Symbols, in this usage, are a very different form of condensed meaning than concepts. We see a dominance of such symbols in early modern semantics, and not least in religion (Luhmann, 2013, p. 245). In the middle ages, symbols were not just signifying but offered access to something otherwise inaccessible. Symbols were, so to say, unities of the visible and the invisible, the present and the absent, and not only do they signify these unities, they also perform them (Luhmann, 2012, p. 141). As Luhmann (2000) writes: 'The symbol marks the inaccessible within the realm of the accessible; we are therefore dealing with a form of reentry of a distinction into what it distinguishes. The symbol contains a reference to its own origin, which grounds the representation in its "given" form' (p. 169).
We have argued that the key problem faced in the emotionalized organization is the problem of the 'inaccessibility' of the authenticity of self-enrolment. This problem refers ultimately to a fundamental gap between communication and consciousness. Only communication can communicate. Consciousness cannot communicate itself 'because every communication is always already autopoietic components of a social system' (Luhmann, 2001, p. 21). Organizations need some kind of proof of self-enrolment and its authenticity, and the effect is a problem of incommunicability that cannot be solved by concepts. Symbolization might here be seen as a functional equivalent to concepts in situations of obvious incommunicability. In this way, we might understand why the romantic, pedagogic, therapeutic or ludic media of transition are loaded with emotionally salient symbols. They are forms of meaning that promise a kind of bridge to link what can be represented within discursive communication to what cannot. Some of these symbols (and indeed the affectivity associated with them) might even promise a kind of symbiotic function regarding the relation between communication and consciousness (Luhmann, 1998, p. 27, Stenner, 2004. 'Play', to give an example from our analysis, is a presentational symbol more than a representational concept. Play symbolizes authenticity. It does not simply represent authenticity but invokes it. Play as a symbol does not communicate about authenticity. It is a communication that avoids communication claiming to give immediate access to what it symbolizes (i.e., it is 'presentational'). Play claims to coproduce authentic playfulness, obviating the need to test authenticity. If Morten is playing, then of course, he is authentic. But there is a paradox here because, by the same token, play is at odds with the idea of authenticity since to be play it must ongoingly distinguish itself from 'the real thing', and hence is always open to being identified as inauthentic 'mere' play, lacking true commitment. Similarly, declarations of love by employees to the organization as their significant other provide another example of a semantic that is more symbolic than conceptual. The symbols present or perform the love relation, making it more than indicating it. Employees declaring their engagement, taking initiatives, showing involvement make use of this symbolic semantic in a manner that does away with the problem of authentic self-enrolment. And yet at the same time, just like romantic love, one is always left wondering if this is 'the real thing', and paradox once again looms. As Landgraf (2002) writes about romantic love communication: 'Tears, speechlessness, fainting or, in writing, exclamation marks, ellipses, and hyphens come to communicate incommunicability. Unfortunately, all the communicative strategies are also open to imitation and simulation. Communicative speechlessness quickly becomes another commonplace, another rhetorical strategy that cannot guarantee authenticity and inwardness' (p. 171).
This symbolic semantic, in sum, offers only a partial resolution to the problem of incommunicable authenticity that is simultaneously 'antistructural' since it can always be questioned, withdrawn and doubted, and-like Sisyphus with his rolling stone-must be recurrently and endlessly re-enacted in a flexible manner. In this sense, each medium of transition can be grasped as a means for the management of liminality (Stenner & Moreno, 2013;Greco & Stenner, 2017), including the management of the new forms of uncertainty and emotional volatility that come with membership by self-enrolment and its insoluble problem of authenticity. The semantic movement from concepts to symbols can be observed as an attempt to deal flexibly with the problem of authenticity. But this movement does not solve the problems but pushes them around, ultimately contributing to the construction of a trembling organization (Andersen, 2013, p. 160), which grows increasingly uncertain because what is most important to it is that which is not-and cannot be-communicated.