Confidential Gossip and Organization Studies

This essay sets out the case for regarding confidential gossip as a significant concept in the study of organizations. It develops the more general concept of gossip by combining it with concepts of organizational secrecy in order to propose confidential gossip as a distinctive communicative practice. As a communicative practice, it is to be understood as playing a particular role within the communicative constitution of organizations. That particularity arises from the special nature of any communication regarded as secret, which includes the fact that such communication is liable to be regarded as containing the ‘real truth’ or ‘insider knowledge’. Thus it may be regarded as more than ‘just gossip’ and also as more significant than formal communication. This role is explored, as well as the methodological and ethical challenges of studying confidential gossip empirically.


Introduction
In everyday organizational life people informally tell, are told, or ask for all kinds of information. As is the case in the stylized piece of conversation above, sometimes and probably often, they ask or are asked not to share it. So it's confidential. But the information shared is not very definite, is not official and is based on hearsay. So it's gossip. How can we understand such commonplace interactions and why do they matter for studying organizations? This essay makes the case that the concept of confidential gossip enables us to understand these interactions. It also makes the case that they have an importance because of their particular role in talking organization and organizing into reality.
The particularity of that role arises because, confidential gossip is, itself, a particular kind of gossip. By no means all gossip is confidential (e.g. gossip about celebrities), while by no means all that is confidential takes the form of gossip (e.g. medical records). When it is confidential, it can have more power than gossip in general because, as we will develop, confidentiality is a form of secrecy, and secrecy carries a special charge, for example in being assumed to be 'the real insider's' knowledge.
Gossip in general is an activity that is surprisingly common. Gossip is estimated to account for two thirds of all conversational time (Dunbar, Duncan, & Marriott, 1997;Emler, 1994) and is a major engine for social networks -indeed it is even claimed to be 'at the core of human social relationships' (Waddington, 2016, p. 811; see also Grosser, Lopez-Kidwell, & Labianca, 2010). From this perspective, gossip is far more than just being something that circulates within the confines of the organization. Rather, it is a fundamental part of what Gabriel (1995) calls the unmanaged organization: a kind of organizational hinterland where affects, dreams, fantasies, myths and stories reside and animate organizing and the organization.
In our argument, (confidential) gossip contributes to the construction and constitution of the organization itself. This is built upon the constitutive understanding, rather than the transmission view (i.e. 'sender-message-receiver'), of communication in organization. We suggest that since confidential gossip, like gossip in general, is a form of communication in organizations, then the way to understand its significance is by engagement with the idea that communication is constitutive of organizations, also known as the CCO perspective (e.g. Ashcraft, Kuhn, & Cooren, 2009;Schoeneborn, Kuhn, & Kärreman, 2019). However, since unlike gossip in general confidential gossip is a form of secrecy, its contribution to CCO is of a distinctive sort.
Overall, the essay makes two main contributions. First, it extends existing scholarship on organizational gossip by showing how secret or confidential gossip is a distinctive sub-category of gossip that has particular implications for the communicative constitution of organizations. Second, it provides a methodological platform upon which future empirical studies of confidential gossip may be conducted.
We begin the analysis by discussing gossip as a concept in organization studies. We then combine that with a discussion of secrecy in organization studies in order to generate the concept of confidential gossip. Having done so, we develop the argument that confidential gossip is important for organizations and their study because of its particular role within the communicative constitution of organizations. Finally, we discuss the practical and ethical difficulties entailed in the empirical study of confidential gossip in organizations.

What is Gossip?
Gossip is one of the most basic but perhaps one of the most misunderstood forms of communication. At its core, gossip involves a minimum of two people, engaged in conversation about things that in epistemological terms exists in the space between 'known knowns' and 'known unknowns' -a conversation solidly based on hard facts would not usually be considered as gossip, nor would a conversation about pure fantasy.
Within this space, what 'counts' as gossip is not simply about the content, but more importantly the context in which it occurs. As Hannerz (1967, p. 36) points out, 'the communication that Mrs A's child is illegitimate is not gossip if it is occurring between two social workers acting in that capacity, whilst it is gossip if Mrs A's neighbours talk about it'. It is the relational dimensions of gossip (Bergmann, 1993, p. 48) that enable us to determine whether it is gossip or not. Contexts of gossip are, thus, themselves products of accumulated communication, memories and relations, sedimented into patterns of interpretations and presuppositions that we employ to understand social realities (see Stewart & Strathern, 2004).
Gossip has long been understood as part of informal communication (Paine, 1967, p. 293; see also Bok, 1982). Organizationally it is linked to the 'informal communication network' (Noon & Delbridge, 1993, p. 23) and more generally to the 'informal structures of organizations' (p. 24). Importantly, Mills (2010) points out that gossip is nevertheless embedded in, affects, and is affected by, formal organizational processes. This does not mean that gossip itself is not an informal practice ('formal gossip' would be an oxymoron) but that, as organization theorists have long known, formal and informal organization are not discrete domains (see for example Diefenbach & Sillince, 2011). This is significant for our argument here because it means that (confidential) gossip may have a role in the constitution of formal as well as informal organization.
Because gossip involves at a minimum two people, it is a social process that can be understood at interpersonal, group and organizational levels and, importantly, as operating at the conjunction of and across these levels (Michelson, van Iterson, & Waddington, 2010). That matters, here, because it relates to the capacity of gossip to play a role in the communicative constitution of organizations. Much of the literature on gossip emphasizes social comparison (e.g. Suls, 1977;Wert & Salovey, 2004) as one aspect of the wider issue of how gossip helps organizational members make sense of 'how things are done around here' (e.g. Baumeister, Zhang, & Vohs, 2004;Grosser et al., 2010), and to understand appropriate ways of behaving within a particular social setting (e.g. Wert & Salovey, 2004). Several studies have also shown another aspect of such constitution where gossip acts as a vehicle to transmit group norms, values and moral principles (e.g. Paine, 1967;van Iterson, Waddington, & Michelson, 2011). In particular, gossip can mark out a group's social boundaries (e.g. Hannerz, 1967;Rosnow & Foster, 2005), constructing the awareness that '"we" do not gossip with any "they" but among ourselves only' (Paine, 1967, p. 282). In this sense, gossip is intimately linked with the socialization and construction of groups (which might be groups within organizations, or the organization as a group in its own right), and mark those who are insiders from those who are outsiders.
While these studies are concerned with what might broadly be thought of as the 'integrative' possibilities of gossip (i.e. its role in bringing people together), other studies have shown its potentials as a manipulative tool (Rosnow, 1977, p. 159). The former shows how gossip can play a role within organizational politics and power play (e.g. Feldman, 1988), including as a way of exposing or resisting power and inequality (e.g. Meyer Spacks, 1985). The latter relates to how gossip can shape others' impression of oneself (e.g. Suls, 1977) or to construct and/or reconstruct the image of a third party (e.g. van Iterson et al., 2011). Finally, it should not be forgotten that gossip in organizations can also provide pleasure and entertainment, generating excitement, enjoyment or relaxation (e.g. Foster, 2004;van Iterson et al., 2011). It may also create displeasure, anger, or a sense of exclusion for those who are the target of, or are not privy to, gossip.
In this sense gossip, as a form of communication, is not merely passed on within an organization or organizational realities, rather, it contributes to the processes of how organizing and organization are 'coming into being' through communicative practices (e.g. Ashcraft et al., 2009). Such constitution is distinctively formed through the paradox of particular gossip where confidentiality plays a significant role at its organization: it is not meant to be shared (e.g. 'among ourselves only') and yet leakage is inevitable. This reflects the fundamental question of CCO as to how language and communication are part of the creation of the social world: the ontogenesis of organization and organizing as such (Schoeneborn et al., 2019). Among other things, CCO research empirically look at how different devices -such as tropes, lists, figures, models and others -in different modespersuading, informing, bullshitting, gossiping -are used in actual conversations and unfold their constitutive and formative role in organizing collectives (Schoeneborn et al., 2019). Schoeneborn and colleagues (2019) suggest that this is performed in three specific ways: communication as constitutive of (a) organization as a social entity or actor; (b) organizing as a social practice or process; and (c) organizationality as an attribute or degree. The first way engages with the tension between communication as a process and organization as an entity. Organization is understood as an emergent and ever-fluctuating network of interlocking communication processes, rather than merely a container of communication. The second way takes an interest in the intersection between communication as a process and organizing as flows of practice, focusing on how for example initiatives, rituals and policies are continuously talked and written into being, edited and erased. The third way attempts to move beyond standard forms of organization and organizing, widening the gamut into networks, markets, social movements, communities, and so on. It is animated by the question of what makes these phenomena more or less 'organizational' (Dobusch & Schoeneborn, 2015; see also Blagoev, Costas, & Kärreman, 2019). Gossip may be involved in all of these processes, but, as we will argue later, may play a particular role for organizationality.

Introducing Confidential Gossip
Our fundamental definition of confidential gossip is that it is that activity where gossip and secrecy overlap. This means it has some characteristics that are not present for all kinds of gossip (as well as lacking the characteristics of some kinds of secrecy, e.g. it has no legal enforcement). Regarding the former, we will argue that this characteristic renders it especially potent because secrecy has significant effects in terms of bestowing a sense of belonging to an in-group, and also in generating a sense that the information being shared has a particular veracity. Gossip is, manifestly, a form of communication. Confidential gossip, for all that it entails secrecy, is also a form of communication, undertaken according to particular rules concerning who may and may not be communicated with, and about what.
Recent scholarship in organization studies has identified secrecy as an important aspect of organizational life (Costas & Grey, 2014Parker, 2016;Scott, 2015) and yet an underdeveloped area of analytical investigation (e.g. Anand & Rosen, 2008;Jones, 2008). The philosopher of secrecy, Sissela Bok, points to the etymology of secrecy and secrets as deriving from secretum and ultimately secernere, meaning to separate or keep apart, and for this reason regards concealment as the defining trait of secrecy. Yet Bellman (1984, p. 10) stresses a paradox surrounding practices of secrecy as its essential contradiction that 'the informant who is telling a secret either directly or tacitly makes the claim that the information he or she speaks is not to be spoken'. In this sense, secrecy is embedded with the inextricable dialectics between the withdrawal and the communication of knowledge (Horn, 2011). It is both known and unknown, both silent and communicable, to different groups and identities of people.
This dual character points to the way that the formation and maintenance of secrecy entails the construction and reconstruction of social relations and behaviour. In this way, as Costas and Grey (2016) argue, secrecy makes and shapes organizations by creating 'compartments' within which there is shared knowledge and around which there are boundaries, so that there are insiders and outsiders of secrecy according to who is in possession of secrets and who is excluded from them. In doing so, secrecy constitutes organization by constructing a 'hidden architecture of organizations' (Costas & Grey, 2016; see also Parker, 2016).
Unlike formal secrecy, enforced by laws and official organizational rules, informal secrecy is unofficially operated and maintained by social rules and norms (Costas & Grey, 2016). Informal secrecy has audience and contexts with 'logics of making and unmaking particular to these contexts' (Hardon & Posel, 2012, p. S3). The 'logics' are constituted by and constitute the complex rendering of social relations such as membership, allowing different ways of selecting, presenting, interpreting and identifying 'us' and 'them'. Therefore, informal secrecy is able to form, maintain and/or split cliques and networks between and within the units of an organization that may cut across the formal organizational structure (Parker, 2000). Participating in informal secrecy can offer opportunities and tensions to explore networking and social capital in organizations by being part of the 'right network', which can provide an individual with first-hand information about upcoming projects and career opportunities. However, when informal secrecy brings social differentiation, it simultaneously requires the differentiation to be maintained, despite the 'seductive temptation [of breaking] through barriers by gossip or confession' (Simmel, 1950, p. 466). Hence, when informal secrecy is created to serve certain purposes and to protect vulnerabilities, it is itself vulnerable to betrayal.
While studies of workplace gossip have indicated its 'non-triviality' in organizational processes, they do not have a specific focus upon, although they do sometimes touch on, gossip as a form of secrecy (e.g. Kurland & Pelled, 2000, p. 432;Michelson et al., 2010, p. 380). Confidential gossip can be regarded as having all of the various characteristics of gossip identified in the literature discussed above, but with the additional feature of being shared as a form of informal secrecy. Costas and Grey (2016, pp. 93-7) suggest that this entails the use of particular verbal and nonverbal cues which highlight the confidentiality of a particular exchange of gossip. Phrases such as 'you must keep this to yourself', 'this is just between us' or 'within these four walls' are all common examples of such cues. Non-verbal cues might include facial gestures or the lowering of the voice. Confidential gossip might be exchanged with a particular injunction about who it can and cannot be shared with (e.g. 'don't tell X') but more generally participants may be assumed to know with whom it is appropriate or inappropriate to share the gossip (Costas & Grey, 2016, p. 97).
The distinction between gossip and confidential gossip is sometimes a hazy one, but it is meaningful. It is true that many studies link gossip to discussion of absent third parties, which might denote some degree of confidentiality (i.e. with respect to that third party). But even that is not necessarily the case: it is quite possible conceptually, and no doubt common empirically, for X and Y to gossip about Z ('have you heard that Z is getting married?') and for Y to then discuss this with Z ('I hear you are getting married'). Gossip becomes confidential when it involves marked boundaries surrounding the shared knowledge and the processes of guarding the boundaries ('don't tell Z, but s/he is going to get the sack'). Equally, gossip about third parties might be confidential not with respect to the third party but to a fourth party ('have you heard, Z is getting promoted but don't tell A, because s/he will be envious').
It is therefore most certainly not the case that gossip always carries a connotation of secrecy or confidentiality. For it is easy to identify some cases of gossip which are not confidential at all (e.g. workmates gossiping about celebrities) and other cases where there is a strong expectation that the information will be rigorously guarded (e.g. gossip about deeply personal or criminal matters). But between these extremes there may well be a lack of clarity about exactly who can and cannot be included in gossip. Very often gossip is passed on in the full knowledge that it will be spread further. Even gossip accompanied with the injunction 'don't tell anyone' might within context contain a tacit understanding that this means 'you can tell X but don't tell Y'. This means that gossip may become 'secret' depending on who is being spoken to (Michelson et al., 2010, p. 380), and so the line between confidential gossip and gossip is both context-dependent and mutable. It is, nevertheless, a line the strength of which will be indicated by the wording or way the gossip is communicated and the extent of its overtness.
The use of verbal markers to denote confidentiality does not in itself tell us why this is being done (i.e. what it is about the information being shared that makes it sensitive), only that, for whatever reason, it is. Horn (2011, pp. 108-9) identifies the three 'logics of secrecy' of which the first is, indeed, that of secretum, or locking away; the second is arcanum, also meaning locking away but connoting the hiding of something special, and the third is mysterium, an almost supernatural sense of what is hidden such that has the capacity to 'elicit awe' (Luhrmann, 1989, p. 138).
Thus making gossip secret itself communicates something over and above whatever information may be shared: the very fact of it being secret (secretum) communicates that this is some 'special' sort of information (arcanum), perhaps even something 'magical' (mysterium). By extension, it may suggest to those who are (and who are not) included in the secret that they are (or are not) special sorts of people. The consequence of this is that although confidential gossip has all of the characteristics of gossip with the addition of secrecy, it does not follow that those characteristics are unchanged as a result of secrecy being added. In other words, confidential gossip is not just 'gossip plus confidentiality'. It becomes qualitatively different from gossip in general. Put another way, confidentiality meaningfully changes the character of the content of the gossip. The cloaking of secrecy makes it more meaningful, intense, important or otherwise consequential. Simmel's (1906Simmel's ( /1950 central insight that secrecy powerfully shapes social relations by creating insiders and outsiders who share (or are excluded from) very strong bonds is crucial here (see also Costas & Grey, 2014Dalton, 1959;Schein, 1985). Secrecy has a mystique about it, generating a sense of exclusiveness among those who share secrets but also a temptation to hint at knowledge of secrets in order to boost a sense of importance. When there is the 'thrill' of secrecy, the pleasure of gossiping can be intensified. Conversely, when secrets are revealed, the sense of anger, loss or betrayal can be significant.
In relation to confidential gossip this means that the various things that gossip 'does' in organizations are likely to be inflected differently and perhaps heightened when confidentiality is added to the mix. For example, gossip about 'the way things are done around here' might be taken especially seriously if it is shared with an imprecation to confidentiality because the mystique of secrecy means that it is more likely to be regarded as more real and true as 'the "truth" behind the "truth"' (Stewart & Strathern, 2004, p. 38): if the things being kept concealed were not important or special, why would they be confidential? In this way, even though confidential gossip may be perceived as more invasive or destructive than gossip, people might accept or perhaps even encourage it as it can be a form of 'triangulation' of what is open knowledge, exploring what is going on 'behind the scenes' and supplementing certain understanding of organizational life. For example, a new joiner to an organization might be told one thing in a formal induction session but then told, in confidence, that the reality is rather different, and would be likely to take this as the 'real truth'.
While confidential gossip constructs a strong and 'aristocratic' sense of 'us' as insiders, the formation of groups is built alongside the possibilities of their 'de-formation' or collapse. The emotional pleasure of belonging also carries with it a vulnerability. For the individual, that includes the possibility of themselves being the subject of confidential gossip among others, or the things that they have confided being repeated without their knowledge or agreement. In this sense, confidential gossip circles, similar to other secretive groups, are 'emotionally tightly knit. . . [partially as] the direct result of the cost of betrayal' (Luhrmann, 1989, p. 160).
With its emotional connectivity and tensions, confidential gossip enables gossip producers along with other participants to manage communication topics and trajectories and to influence (individual and) group identification. Such management might be achieved through forming 'strategic ambiguity' (Eisenberg, 1984) that facilitates indirectness and avoidance of communication (e.g. Hallett, Harger, & Eder, 2009) and leaves room for ongoing adjustment of evaluation that shapes communication trajectories. In this way, evaluation could -at first -be constructed implicitly (Hallett et al., 2009), for as Bergmann (1993, p. 154) writes, 'the delicate character of an event is constituted by the very act of talking about it cautiously'. In the process of developing and shaping directions of evaluations (e.g. praise or criticism of particular behaviour), confidential gossip generates 'organizing properties' (Christensen, Morsing, & Thyssen, 2017) that stimulate participants' ongoing discussions and learning about what 'appropriateness' means to them. Because of the mystique of secrecy, knowledge imparted as secrets about what 'we' are and what 'they' are is more likely to be taken as the reality and therefore to have an especially strong impact upon group (and individual) identity recognition. By talking a specific sense of 'appropriateness' into existence, confidential gossiping regulates group behaviour, which in turn reinforces a sense of characterization and identification for being 'us' and not like 'them'. This process might also create identification tensions for some participants whose situated understanding of 'appropriateness' are competing with the shared understanding (Winkler, Etter, & Castello, 2020), leading to the weakening or even the breaking of social bonds. In this sense, confidential gossip can be a source of (generating and expressing) dissent and simultaneously a way to manage it.
This means that information which is particularly controversial, or resistant to power structures, or sensitive in some other way, can be shared among participants while being kept from others. In this sense, we might expect confidential gossip to be especially prevalent in particular organizational contexts where open communication is for some reason difficult or discouraged. Examples might include organizations which have a culture of bullying, or where there are high levels of conflict and politicization. This is both because confidential gossip can protect participants from bullying and conflict, and also because it can be a vehicle for these things to occur (e.g. Crothers, Lipinski, & Minutolo, 2009;Feldman, 1988).
From the perspective developed here, confidential gossip with its situational embeddedness in local settings concerns ongoing social processes in which secrecy is initiated, sustained and enforced through social interaction, and in turn shapes social interaction. When we gossip about others who are known to us, we are interacting with the structural feature of social relations that presuppose a measure of ignorance and reciprocal secrecy. Thus, confidential gossip is a genre of informal communication which selectively circulates intentionally and informally concealed knowledge within a particular social network. It is shared among selected members of the network who are both privy and relevant to the circulated knowledge and have shared interests. At the heart of confidential gossip lie social relations that are the cause and consequence of confidential gossip. At the centre of its constitution of organizationality lies its processes and influence as a metaphorical crayon that draws and redraws a map of socialization and social relations at work, being powerfully persuasive that it communicates the 'inside truth'.
Schoeneborn and colleagues (2019) make the point that CCO implies a 'low threshold' view of what an organization, or organizing, may be. From a CCO point of view the organizationality, as Schoeneborn et al. (2019) label it, of collectives can emerge in almost any fashion, whenever collective action is called upon (see also Taylor & Van Every, 2000). Importantly, this means that the 'organizationality' of a social phenomenon is not a yes/no question but rather a matter of degree (Blagoev et al., 2019;Dobusch & Schoeneborn, 2015).
In this sense, confidential gossip has particular potential to cast light on the emergence of organizationality. Confidential gossip in organizations is not only a means of knowledge exchange with semantic understanding, but also a way to render organizational members' social recognition, perception and awareness of organizational life. The collaborative production of confidential gossip through its participants shapes individual construction processes of social realities at work, influences desires and motivations of individuals and groups, and affects power structures in both preferred and unintentional ways that in turn 'breeds' the confidential gossip predominantly. Confidential gossip is in this way a ploy for insiders to achieve individual-and/or group-level interests. Processes of constructing, maintaining and breaching confidential gossip may not be identical across organizations, and hence are engaged by and embedded within a certain specialness and uniqueness of an organization. Thus, by partaking in confidential gossip, participants reinforce their recognition of organizational existence.

How Can We Study Confidential Gossip?
So far, we have put forward the idea of confidential gossip as a purely theoretical concept. Ultimately, it would require empirical exploration, to which there are formidable methodological barriers. Research on confidential gossip is by definition methodologically challenging, given its character as a subset of both informal secrecy and gossip and therefore its social ephemerality, contextuality, sensitivity and embeddedness. Confidential gossip, similarly to gossip, is an ephemeral activity that is difficult to 'catch in the act' of being perpetrated (van Iterson et al., 2011). It emerges, submerges and re-emerges in between social connections and tensions, between interpersonal attachments and detachments.
Given the methodological challenges of confidential gossip and the similarities between general and confidential gossip, we draw in the first instance on the empirical studies on gossip as a reference point to consider how to study confidential gossip. Empirical studies on gossip are primarily ethnography focused and mainly conducted in the domains of anthropology, sociology and psychology with some attention paid by gender and socio-linguistic studies. Gossip has also been explored through several other approaches across disciplinary domains, including archive studies (e.g. Besnier, 1989), diary studies that utilize a structured diary record with seven-point scale questions and open questions (Waddington, 2005), experimental research (e.g. Cole & Scrivener, 2013), online research that focuses on conversations on electronic bulletin boards (Harrington & Bielby, 1995), questionnaire studies (e.g. Baumeister et al., 2004;Grosser et al., 2010) and interview studies (e.g. Hafen, 2004;Mills, 2010).
The ephemeral, contextual, sensitive and embedded characters of confidential gossip distinguish it from other ways of speaking. It may primarily be accessed through exploring meanings and social interaction. Qualitative methods are particularly appropriate as they can be used to capture the narratives and actors' interpretations of situations and people (Filstead, 1970). This points to ethnographic methods as a possible way to access confidential gossip for, as Edgar Schein notes in relation to his work on secrecy and in-groups, 'you had to be a real insider to know' (Schein, 1985, p. 100). It is a topic eminently suited to what Anselm Strauss apocryphally referred to as 'hanging around and listening in' by residing at the heart of social life in an organization. It brings the particularities and 'irrationalities' of off-the-stage life (Goffman, 1959) to the fore by giving a view of the variability of forms of organizing, the multiplicity of social relations that constitute such various forms, the diverse connectivity that ties the social relations in particular ways, the rules that are developed to maintain the connectivity, and the sanctions that are introduced to prevent the maintenance from being hindered.
Ethnographic studies on gossip have been predominately conducted by anthropologists. Researchers often work as participant observers such as staying residents (e.g. Colson, 1953;Hannerz, 1967;Haviland, 1977) and team members in an organization (e.g. Kniffin & Wilson, 2005). The importance of participant observation is noted by Hannerz (1967, p. 45) as that 'probably there is no other way of acquiring knowledge about gossip', which can make the disconnected connected, make sense of the nonsense, make the obscure clear, make the silly funny (Roy, 1959). Somewhat in contrast, Colson (1953) acts as both participant observer and non-participant observer for particular age groups of informants. Studies also utilize two types of invisibility when researchers are non-participant observers and are absent during the occurrence of (certain) empirical evidence, including the use of electronic devices to record participants in a particular setting (e.g. Thornborrow & Morris, 2004 videotaped a TV show), and participants recording/taping their own social interaction (e.g. Guendouzi, 2001).
Linking these studies on gossip to confidential gossip, the methodological character of participant observation enables researchers to conduct a 'curious kind' of empirical work on confidential gossip through forming an 'inbetweener position' with both closeness and distance in the field. Being a relative outsider could help to capture wider socialization contexts and processes in which confidential gossip might be embedded. Being a relative insider could enable researchers to be a member of the social system being studied and to interpret and 'translate' the subtleties of meanings and implications surrounding confidential gossip. Hence the 'inbetweener position' hopefully is close enough to the quotidian experiences and yet distant enough to not turn such experiences into the taken-for-granted parts of organizational life. In such scenarios, which might not be a 'fly on the wall' method, researchers are neither complete participant nor complete observer. And, of course, fieldwork practices are contextually and biographically varied (Van Maanen, 2011), so how we engage in and with the field is largely shaped by what we want to know. Therefore the specific design of fieldwork on confidential gossip will be shaped by particular purposes and focuses as well as situated access negotiation(s).
Each of the possible methods requires the confirmation of ethical approval from relevant institution(s) and of research consent from its participants prior to the conduct of research. Among the discussed methods, ethics for ethnographic methods on confidential gossip might be the most complicated. The social embeddedness and contextuality of confidential gossip indicate that if it is to be found, it would be interwoven in the wider socialization. Therefore observational studies on confidential gossip might involve a broader focus on a wider range of interaction than confidential gossip itself. Moreover, given the undesirability of being identified as (confidential) gossipers (e.g. Bergmann, 1993), making confidential gossip a direct and explicit focus to the observed might make it difficult, if not impossible, for researchers to be actually included in confidential gossip. This points to the particular ethical complexities that attend the research on confidential gossip which, in its nature, might be sensitive and which was to some degree being covertly studied. Roulet, Gill, Stenger and Gill (2017) emphasize that ethical consideration of any (degrees of) covert study should be context dependent, as 'it is in the particular cases of the here and now with participants that ethics are situationally accomplished' (Calvey, 2008, p. 908). Thus ethical considerations of research conduct on confidential gossip should draw on a 'situated ethics perspective' (Roulet et al., 2017;see also Calvey, 2008) that pays more attention to ethics 'as an ongoing social practice' (Roulet et al., 2017, p. 16) and 'contingent, dynamic, temporal, occasioned and situated affairs' (Calvey, 2008, p. 912). The characteristics of confidential gossip as a genre of informal, evaluative and sensitive communication require researchers to conduct the research within ethical boundaries without triggering the dangers of hidden agendas, especially regarding the issues of privacy invasion to both the researched and the researchers, and sufficient protection of participants' identities both during and after fieldwork.
Despite all of these challenges, we should not give up on the attempt to study confidential gossip empirically. Going back to the stylized example of an interaction with which we began this essay, readers might consider how often they, themselves, take part in interactions of this sort. Our own experience is that they are commonplace and, to that extent, there is a face validity in the claim that they are an important part of organizational life which we, as researchers, should seek to understand. If that poses difficulties then we need to find ways to overcome them rather than shy away from studying this aspect of organizations.

Conclusion
In this essay we have brought forward the concept of confidential gossip to denote the practices where gossip and secrecy overlap. As a form of gossip, it is a special form of communication; as secrecy, it is a special form of gossip. One might therefore envisage confidential gossip as a subset of gossip and gossip as a subset of communication.
Drawing on CCO that has identified the key role of communication in the constitution of organization, we argue that confidential gossip plays a role within that. What we have urged is that not only is that so, but its role is an important and distinctive one. What at first sight may seem quite trivial is decidedly nontrivial. Consider even the very simple hypothetical example with which we began, and imagine how it might play out.
In the hypothetical example, perhaps A does not apply for the job because of what B has said, taking it especially seriously because B has said they shouldn't be saying it, and so A is persuaded that it is most likely true. That has a direct effect on who ends up doing the role, a small contribution to the making of organization. Perhaps, because A doesn't feel able to check B's information is true as it was given in confidence, it turns out to be false and A is cheated of the chance of the job. Perhaps, because both have committed to keep each other's confidence they go on to share further confidences, with spiralling effects each time. Perhaps one or both of them break the confidence, leading to A being marginalized in their department, or B being told off by their boss, or both; and one or both, angered by the betrayal of confidence, embarks on a feud which may have all kinds of repercussions for the organization.
These and many other imaginable developments of that basic scenario are all organizationally impactful and potentially organizationally important, and are all primarily attributable to the particular way in which information was communicated, namely with the promise of confidentiality. If, as students of organizations, we take seriously the idea that organizations are communicatively constructed then we simply could not understand those developments if we did not both know about the interaction and understand its specific quality as an instance of confidential gossip, with all the charge that secrecy brings to it. And, not just as students of organizations but as members of them, imagine if you were A or B, and your confidence had been betrayed. How would you feel? That will perhaps disclose the emotional texture of what is at stake, but that, again, only exists because it is about something that you thought was confidential, and was not just 'routine' gossip.
So we need the concept. But we also need to know about the interaction, so we have to have a way, or ways, of accessing it empirically. Hence we have also explored how studying confidential gossip is a complex matter, posing particular empirical challenges and ethical concerns. The methodological difficulty is to capture the ephemeral phenomenon and elusive practice of confidential gossip, and to encapsulate its fluid development and circulation in a given social context, while not violating the rights of research participants. We have suggested that ethnography or participant observation is the most feasible method as it 'gets inside' these interactions, whereas methods relying on retrospective accounts (e.g. interviews) or diary-keeping may fail, precisely because of the ephemeral nature of the phenomenon. Yet participant observation involves the greater ethical challenge, since it entails reporting on things said in confidence. One aspect of confidential gossip, and a possible method of studying it which we have not touched on, is how it may occur not in face-to-face conversation but be technologically mediated (e.g. WhatsApp groups) and this could be the focus of future work.
In pointing to its often ephemeral nature, and despite highlighting the ethical issues in studying it, we should nevertheless stress that confidential gossip is not necessarily or even often about uncovering 'juicy facts' or the hidden darkness of organization. It does not need to be scandalous or thrilling in order to be important for organization and understanding organization, though it may sometimes be so. Confidential gossip is an ordinary part, yet an extraordinary reflection, of our everyday life. As Crewe (2015) notes in her ethnography of parliamentarians at work, 'this [book] has theatre, conflict and secrets at its heart. The secrets are neither scandalous nor even shocking -they are everyday revelations about how our parliament really works, seen through the eyes of its main protagonists' (Crewe, 2015, p. 9).
Researching confidential gossip is therefore a way to notice the 'unnoticed source(s) of beauty' (Parker, 2017(Parker, , p. 1002) that many if not all of us see and experience in our everyday organizational life.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.