WHAT IS RELATIONAL PRACTICE?

Our analysis of Fletcher's use of the term “relational practice” (henceforth RP) in the workplace suggests that RP has three crucial components:

We discuss each of these briefly.

As its name suggests, doing RP at work involves attending to workplace relationships, including both people's need to feel valued – their “positive face needs,” in Brown & Levinson's (1987) terms – and their “negative face needs,” the requirement that their autonomy be respected. RP may entail being friendly or supportive, as well as being polite and considerate. In the workplace context, RP is often appropriately oriented to people's need that their special skills or distinctive expertise be recognized, and it also crucially involves people's need to feel they are valued and important components in a team or a group.

While this is the most obvious and commonsensical interpretation of the meaning of RP, embracing any aspect of workplace behavior oriented to an individual's relationships with work colleagues, Fletcher 1999 defines it more narrowly. In Fletcher's terms, whatever its interpersonal function, RP is also always relevant to, and often inextricably entwined with, the goal of furthering organizational objectives. In other words, in addition to its core relational function, RP simultaneously serves more instrumental or transactional goals. Since, in the broadest interpretation, any behavior that improves workplace relationships could be regarded as “good for business,” there is clearly room for debate here. Moreover, Fletcher herself often includes behavior that is only very indirectly related to furthering workplace objectives. In our view, however, this criterion needs to be applied stringently if the distinctiveness of the concept of RP is to be maintained, and if RP is to provide insights in the analysis of workplace interaction. We return to and illustrate this point in the analyses below.

Third, despite point (ii), their relevance to workplace objectives, relational practices are typically regarded as dispensable, peripheral, and in some cases even distracting in the workplace. In other words, RP is frequently overlooked, or, if noticed, considered irrelevant if not actually counterproductive. Indeed, this off-record status is sometimes explicitly signaled by the use of discourse markers such as get back to the point, get back on track, enough digressing, enough (of that), and, less obviously but more systematically, by markers associated with digressions, such as so (by far the most frequent), anyway, OK, hey, right, well, or now, as well as by silent pauses and verbal hesitations or filled pauses (e.g., um, mm) (Schiffrin 1987, Lenk 1998, Marra 2003). In one workplace, the phrase moving right along had clearly developed as a routine marker of the end of a digression. Such discourse markers provide clear evidence to support the claim that strategies used in the pursuit of RP are perceived as off-topic digressions that need to be kept under control.

The relationship between these three features is fluid and often difficult to pin down, as the analysis and discussion below demonstrate. All three points are important in distinguishing RP from other, superficially similar types of workplace talk. Subversive humor, for instance, may strengthen solidarity between workmates, but it may also be oriented to undermining rather than furthering the organization's objectives (see Holmes & Marra 2002a). Expressions of approval, in contrast, obviously serve workplace objectives and attend to workplace relationships, but they are often part of the ratified or “official” workplace discourse, and thus not considered irrelevant.¹

On the basis of her interviews, observations, and shadowing of key personnel, Fletcher's analysis of RP identifies four categories or “themes,” which she labels preserving, mutual empowerment, self-achieving, and creating team (1999:48). The first two are somewhat more oriented to transactional or organizational objectives, and the second two to personal and interpersonal goals. Preserving focuses on relational practices that are project-focused; they are primarily aimed at advancing the project's objectives. Preserving activities include doing boring, tedious tasks that simply need to be done, even if they are not, strictly speaking, one's responsibility. Mutual empowering is similarly aimed at furthering the project's goals, but it is essentially other-oriented behavior. It includes such activities as making connections or putting people in touch with others who can assist them achieve their goals, effective networking for the benefit of others, and “empathic teaching” (Fletcher 1999:56) – that is, providing support and guidance in a fashion responsive to the addressees' starting point, and even their emotional state.

Self-achieving, as the label suggests, identifies relational practices aimed at enhancing the professional effectiveness of the individual. Fletcher's use of the term covers such behaviors as reestablishing relations with someone after a disagreement in a meeting (1999:65), and it includes reflection aimed at improving understanding of one's own and other's motivations, and of the emotional complexity of situations. Finally, “creating team” is the term Fletcher uses to discuss activities aimed at “creating the background conditions in which group life [can] flourish” (1999:74). It includes all the typically unobserved behind-the-scenes behaviors that foster group life and the development of team esprit de corps – activities such as taking the time to listen and respond empathically to non-work-related information, creating opportunities for collaboration and cooperation, interfacing or facilitating productive interaction, and defusing potentially confrontational situations. In this article, we focus on just two of these categories: the ways in which people “create team” and “preserve the workplace enterprise” at work.

Although Fletcher attempts to treat these four categories as mutually exclusive, it is clear that they are often difficult to distinguish. Our experience in analyzing workplace discourse indicates that people are very skilled in exploiting the multifunctional aspects of human communication systems, including language. One utterance typically serves several functions – it may simultaneously advance the project objectives (preserving) and pay attention to the interpersonal dimension of team relationships (creating team), for instance – as, indeed, the definition of RP presented above suggests. In the next section, we use our data to illustrate Fletcher's claims regarding the complex ways, and particularly the subtle ways, in which people manage and reconcile the competing demands of transactional and interpersonal objectives in face-to-face workplace interaction.

HOW IS RELATIONAL PRACTICE MANIFESTED IN WORKPLACE DISCOURSE?

Our analyses of New Zealand workplace discourse suggest that two fundamental dimensions underlie the many and varied realizations of RP in interaction. First, there is RP that is oriented to constructing and nurturing good workplace relationships, to establishing and maintaining solidarity between team members, and to networking and creating new work relationships. In Fletcher's terms, these are obvious ways of “creating team.” Second, there is RP that is, to a greater or lesser extent, concerned with damage control: RP oriented to constructing and maintaining workers' dignity, to saving face and reducing the likelihood of offense being taken, to mitigating potentially threatening behavior, and to minimizing conflict and negotiating consensus. The parallels with Brown & Levinson's (1987) concepts of positive and negative face, crucial components of politeness theory, are obvious. Our purpose here is to elaborate these analytical dimensions in relation to workplace discourse. While damage control strategies are most obviously oriented to “preserving” and protecting the workplace enterprise, our conception of RP emphasizes how both sets of strategies may contribute to this goal. In what follows, we provide examples of some of the diverse manifestations of each of these dimensions of RP in workplace discourse.

Creating team

People at work use a wide variety of discourse strategies to construct and maintain good relations with their co-workers. These include engaging in small talk and social talk (Holmes 2000b), introducing humor into workplace discourse (Holmes 2000c), telling entertaining stories or anecdotes (Holmes & Marra 2001), and paying compliments or giving approval. In this section, we provide examples of these varied manifestations of RP, indicating how they fulfill the three criteria described above. We focus on strategies that are stereotypically associated with “women's language” or “girltalk” in order to lay the foundation for a discussion of the “gendered” nature of RP.

Small talk and social talk

Small talk and social talk at work clearly serve the function of establishing and nurturing workplace relationships; the label “small talk” itself explicitly signals the perceived status of this type of talk as trivial and irrelevant to serious workplace business.²

Holmes 2000b explores in some detail the distribution, content, and complex functions of small talk and social talk at work.

In fact, most social and interpersonal talk in the workplace is typically discounted, or in Fletcher's terms “disappeared,” from the organizational record. Thus, small talk clearly meets two of the criteria for RP. Not all social talk at work, however, meets the third criterion – serving to advance the primary objectives of the workplace – except in the rather indirect respect, mentioned above, that fostering good relationships at work generally contributes to achieving workplace goals. Nevertheless, we did identify a number of interesting instances of social talk in our data that met this third criterion.

This conversation moves very clearly from social talk to work talk, from a discussion of nonwork topics, Peg's baby's health and gender (lines 1–5) and Peg's own health (line 6–11), to the discussion of the impact of her pregnancy on her contribution to the organization (lines 12–16). Peg's manager is clearly engaging in RP; the expressions of interest about the baby and concern for Peg's general condition are classic (positive) face attention strategies, constructing and nurturing good workplace relationships. However, the discussion also addresses the implications of this information for the project team's objectives. Although the content of line 12 (you're hoping you're gonna work through) could be simply a further expression of interest, Peg's response (lines 13,15–16) indicates that she orients to her manager's comment as transactional rather than interpersonal in intent. This is classic RP, and we have many similar examples in our data from a wide range of different workplaces (see, for example, the conversation between a personal assistant and manager in a government organization concerning taking time off in the school holidays, reproduced in Holmes 2000b:53).

In terms of gender analysis, it is worth considering whether such talk instantiates stereotypically feminine discourse. While the nonwork topic of an expected baby in (1) undoubtedly comprises stereotypical women's talk, it also provides entrée to explore the consequences for the organization of Peg's potential unavailability, a concern that is much less obviously stereotypically feminine. Is the manager skillfully managing the discourse for this purpose? We return to this gender issue in the discussion below.

In a similar fashion, discussions of leave and holidays frequently move subtly from plans for how the recreational time will be spent to discussion of how the organization will cope with the person's absence, skillfully serving both interpersonal and transactional goals. Off-topic social talk functions as a means of addressing an issue of direct concern to the progress of a project or the smooth running of a department. One interesting fact about such conversations is that they are consistently located at the boundaries of workplace interaction. They typically occur at the ends of meetings, often as people are walking out of a room, and even during social breaks. In other words, these conversations usually occur “off the record”; the transactional goal is achieved “by the way” during a conversation that both parties would typically consider as irrelevant small talk or social talk. They clearly fit the criteria for RP outlined above, and they illustrate the subtle ways in which apparently irrelevant workplace discourse can serve organizational goals.

Positive humor

Humor in the workplace is an obvious candidate for inclusion in a review of strategies that qualify as manifestations of RP. Humorous comments are typically intended to entertain and amuse colleagues, to “create team” by constructing and strengthening workplace solidarity and collegial relationships.⁴

See Holmes 2000c for a detailed discussion of the functions of humor in the workplace data we recorded.

And, like social talk, of which it is often a component, humor qualifies as typically irrelevant to the main objectives of the workplace. It is always “de trop” or dispensable in the strictest sense, since the contribution made by humorous behavior or discourse at work falls squarely in the area of interpersonal relationships. We have discussed these characteristics of workplace humor at length elsewhere, providing many and varied examples from a range of workplaces (see Holmes 2000c; Holmes & Marra 2002a, 2002b).

Example (2) illustrates humor that is tightly integrated into talk which contributes to the goals of the team. It centers on the need to coordinate the taking of annual leave to ensure minimum negative impact on the work project.⁵

This example is discussed in more detail in Holmes 2000c.

(2) Panic early. Context: Planning meeting of a group of colleagues. S is the section manager. At this point they are discussing leave plans.

The manager suggests it may be useful to panic early (line 4), a humorous way of indicating the importance of planning ahead, and something of a contradiction in terms, since the notion of panic is almost inextricably tied to last-minute pressures. The group members clearly share a common reaction to the notion, and this is a good example of them “doing collegiality” through humor. The humorous scenario is interactively achieved or jointly constructed: H's comment (line 6) and T's contribution (lines 10–11) both build on the manager's humorous suggestion, which is elaborated in lines 8–9.

This example illustrates how humor provides an ideal means of doing RP at work. First, the very collegial and cooperative structure of the talk, as well as its function of providing amusement, supports the argument that it is contributing to the development of workplace solidarity. Second, the discussion is obviously relevant to achieving the group's workplace objectives. And third, it is clear that the content of this exchange could have been communicated without any humorous component; in other words, the humor is, strictly speaking, irrelevant and superfluous.

Example (3) illustrates a more sophisticated instance of RP manifested in a humorous interchange. Three colleagues are discussing the fact that eliciting a response to a proposal or request for advice is a slow process in their organization:

(3) The Len factor. Context: A meeting of three work colleagues in a government department.

We here witness, or rather overhear, the creation of a piece of workplace jargon, the Len factor, a humorous instance of in-house shorthand which becomes well established in this workplace. In lines 1–4, A describes how the same problem frequently elicits completely different advice from different individuals. Then V informs the others of a phrase, the Len factor, coined to describe the unavoidable delay that must be built into any estimate of how long obtaining a response will take. Len is a colleague who is well known for always seeing difficulties and identifying problems, rather than facilitating the speedy resolution of an issue or the smooth passage of a proposal. The Len factor is an amusing, succinct means of referring to the inevitable delays such behavior generates.

This is canonical RP: The three colleagues are clearly on the same wavelength; the humor contributes to their sense of belonging to the same team; they are discussing an issue of obvious relevance to their workplace objectives; and the humor is, strictly speaking, dispensable. It is a particularly telling example because the jargon phrase thus created continues to generate (strictly irrelevant) humor whenever it is used, while succinctly summarizing an important step that needs to be taken into account in any planning process.

Is humor gendered discourse – discourse that contributes to the construction of a person as male or female? And if so, with which gender is humor associated? On the one hand, there is research evidence that women at work are regarded as lacking a sense of humor (Lakoff 1975, Cox et al. 1990, Crawford & Gressley 1991, Crawford 1995). On the other hand, whenever we discuss such research, especially with working New Zealand women, they are at least skeptical and often incredulous of such claims. Our own analyses of humor in the workplace provide substantial evidence that, in a range of workplace contexts, women contribute at least as much humor as men (Holmes, Marra & Burns 2001, Stubbe, Holmes, Vine & Marra 2000). Again, this is an issue we will return to below.

Off-record approval

A third strategy for creating team or constructing good work relationships is the obvious one of giving approval. There were many examples of this strategy in our data, especially from superiors to subordinates and between colleagues of equal status, though in line with the egalitarian values of New Zealand culture (in which “tall poppies” are not tolerated, and praise is often regarded with suspicion), they are typically brief and low-key (e.g. great, good work, fine, nice one). The instances we are concerned with here, however, are instances where approval was provided indirectly or off-record, where it was in a sense “unofficial.” These are instances of approval that qualify as RP; they are invisible, overlooked, and perhaps undervalued examples of individuals providing positive feedback to others in the workplace.

Like the instances of social talk discussed above, instances of approval that function as RP tended to occur at the peripheries or boundaries of workplace interaction. They were components of interaction that were literally “off-record” in meetings – for instance, not recorded in the minutes, and often occurring before a meeting started or after it finished. Throwaway comments such as nice job, given by one colleague to another in the corridor after a presentation, for instance, or you nailed them following a confrontational meeting, illustrate remarks that go formally unrecorded but contribute to team spirit.

Example (4) illustrates the rather more complex and challenging situation of a subordinate giving positive feedback to a superior. Examples of this kind of RP in our data demonstrate great skill on the part of the subordinate, since, as noted above, it is crucially important to avoiding accusations of flattery or “crawling.”⁶

Unsurprisingly, there are many further, unprintable terms for this important concept.

(4) J's work experience. Context: Beginning of a meeting at a big commercial organization. J is the Project leader and Chair and has arrived first. B is next to arrive, and they are chatting as they wait for the rest of the team members to arrive.

In this standard position for small talk, B introduces the topic of J's management skills and proceeds to pay a compliment: you seem really good at it and awfully confident (lines 3, 5). In this position, before the start of a meeting, their talk is clearly off-record. Indeed, it is unlikely B would have made these comments once others had arrived. This is a nice example, then, of RP: an off-record, supportive comment that clearly relates to the goals of the organization. Appreciation such as this, even from a subordinate, is likely to encourage J to continue to perform well and to maintain good management skills. Yet, by virtue of its position and off-record status, the approval has no formal or official weight.

Similar examples occur elsewhere throughout our data. In one workplace, a team member who had been engaged in a relatively minor aspect of the organization's work for some time expressed appreciation to the CEO for explicitly recognizing the value of this work by including reference to it in the agenda of a meeting of high-status managers. In another organization, a wry comment, we missed you, from a subordinate to their manager returning from leave, served primarily as a way of introducing a jocular complaint about the deputy who had temporarily replaced the manager. In all these cases, even though the expression of appreciation is indirectly achieved, it nevertheless functions to attend to workplace relationships, thus maintaining the team's momentum on the project.

Like small talk, compliments and praise are strongly associated with women's talk; and there is evidence that in many contexts women do indeed provide more positive comments than men (e.g. Holmes 1988, 1995; Herbert 1990; Johnson & Roen 1992). Our workplace data of actual recordings in a number of very different organizations throw light on how such patterns may develop so that stereotypical gendered identities are constructed and fostered within particular communities of practice. We return to this point in the discussion of gender and relational practice below. We turn now to examples of strategies that preserve the workplace enterprise and further workplace goals predominantly through their orientation to face protection; one could regard them as “damage control” strategies.

Damage control

Strategies that nurture positive relationships among workmates and colleagues illustrate one dimension of RP. Their converse, and just as important in furthering workplace goals, are strategies that minimize sources of conflict between workmates and attend to the maintenance of workers' dignity. A good deal of the discourse in some workplaces is oriented to avoiding conflict, to minimizing the chances of a breakdown in workplace relationships, and to negotiating agreement. This kind of discursive behavior contributes to preserving the workplace relationships that are essential to the advancement of the organization's objectives. In our data, some of the most obvious ways in which these goals were achieved were strategies of mitigation. Individuals concerned with achieving their workplace objectives would seek ways of mitigating negatively affective speech acts such as disagreements, refusals, complaints, and criticisms.

We now illustrate just two such strategies, indicating how they fulfill the three criteria described above. First, we consider skillful, off-record, facilitative work, and second, the use of humor to mitigate face-threatening or even face-attacking acts. Both are used in the interest of avoiding conflict, nurturing important team relationships, and thus furthering the organization's goals.

Covert facilitation

It is possible to identify a range of different instances in our workplace data where individuals take steps to manage potentially conflictual situations, to pour oil on troubled waters, or to anticipate problems and head off possible challenges (see, for example, Holmes & Stubbe 2001). Two brief examples must suffice here. In (5), S reports eliciting information that suggests the project team should proceed gently with the next stage of their project:

(5) Corridor work. Context: project team meeting in large commercial organization.

S here describes picking up rumbles of concern among those responsible for actually implementing the project this team is managing. This is precisely the kind of invisible off-record discussion that Fletcher 1999 emphasizes as canonical RP: the relational work involved in anticipating problems, having antennae tuned for potential disaster or failure, and thus “preserving” the project and increasing its chances of success. S's observations lead to a discussion of what can be done to reassure and support the implementation group.

Such “discussions in the corridor” are classic instances of RP: They provide a means for those most directly involved to express their worries about the project timetable. S's covertly facilitative behavior thus serves to warn those higher up the authority structure of potential implementation problems, so that preventive action can be taken.

The second instance, (6), illustrates a project manager setting up the department manager to provide approval for a valuable contribution to the discussion and the project. The project manager is providing a report on exactly what the various team members have accomplished since the last meeting:

(6) Vita merits praise. Context: Large project team meeting in commercial organization. Project manager is reporting on project progress to department manager.

In lines 1–3, the project manager reports on what the team agreed Vita should do by this meeting, and in line 4, Vita confirms that she has indeed accomplished the specified task. Since the department manager makes no immediate response, J proceeds in lines 5–6 (Vita's done a work plan just for that implementation) to “prime” C to provide positive feedback to Vita. C responds appropriately in line 7 with a positive and appreciative comment, and J then continues with the next item. The facilitative move is made extremely discreetly, and the manager picks up J's cue without missing a beat. This is excellent RP – subtle, backgrounded, relational work, attending to workplace relationships in the interests of the project's progress.

In another organization where we recorded extensively, the manager of one section regularly engaged in a good deal of off-record RP around the edges of the more on-record transactional interactions that dominated the official record of the department's output. In addition, the skillful negotiation of consensus documented in detail in Holmes 2000d includes many instances of covert facilitative strategies used to minimize conflict.

Closely related are examples where one team member defends another from face attack, often in the form of teasing or jocular abuse, by other team members. Such protective behavior may function to underline subtly the importance of good team relations in furthering project goals; the covert message is “Don't undermine team members – we all need each other.” Indeed, humor can itself be used as a damage control strategy, as we illustrate in the next section.

Mitigating humor

Humor is a very flexible communication strategy. As illustrated above, it is an effective means of creating team, but it can also be used to attenuate the force of negatively affective speech acts such as disagreements, refusals, complaints, and criticisms. Humor can soften the impact of a face-threatening act (FTA) such as a directive (Brown & Levinson 1987:124, Holmes 2000c), and it can take the sting out of a refusal (Holmes & Stubbe 2001) or a criticism (Holmes & Marra 2002a). Humor is used extensively in this way in our data. There are, for example, critical comments about the time it takes someone to read a document or take notes, or the length of something they have written, all of which are attenuated with ironic humor. There is a humorous exchange about the “final” status of a document that has been back for revision many times (this is THE final version). And there are many jocular insults that serve to express irritation or dissatisfaction in an acceptable manner.⁷

Not all jocular insults serve this function, of course (see Hay 1994, 1995; Stubbe 1999).

Here we analyze just one example to illustrate how humor may be employed as a core RP strategy:

(7) No screendumps.⁸

This example is taken from Holmes & Stubbe 2001, where it is analyzed from a different perspective.

Context: regular weekly meeting of project team in large white-collar organization.

The team has received requests to allow people to print from the computer screen (to “screendump”). The section manager clearly opposes this proposal, with an explicit prohibition, no screendumps (line 4). The rest of the team are unhappy with this, and they proceed to provide reasons why screendumps should be permitted, but the manager overrides their opposition by simply repeating no screendumps. Humor is then coopted as a damage-control strategy to preserve good working relations in the face of the peremptory veto. P's sarcastic thank you Chris (line 7) provides an initial tension-breaker, and S's suggestion that Chris may be wavering – so that's a clear well maybe no (line 12) – is humorously internally contradictory. It leads Chris to restate the position quite explicitly – it's a no (line 13) – and again S defuses the tension with a humorous hyperbolic comment, it's a no a royal no (line 14), echoing a reference to an earlier humorous episode in which Chris's status had been satirized as a member of the royal family. Finally, Chris too contributes to the defusing of the tension with a humorous comment which draws explicit attention to feelings people usually conceal in a business context: did people feel disempowered by that decision (line 15).

The humor in this short excerpt nicely illustrates all three aspects of RP. Its function is to help manage the conflict and to preserve the group relationships, as well as maintain the longer-term project goals. While P and S use light satirical comments to defuse the sting of Chris's refusal, the content of their contributions (a sarcastic expression of appreciation, and a satirical request for clarification of something already crystal clear) also imply acceptance of Chris's intent. These humorous comments could, strictly speaking, be regarded as a digression from the team's core business, since they delay progress with the next agenda item. They are certainly off-record and technically irrelevant, though, of course, like all effective RP, the invisible work they do is crucial to the long-term health of the project and the project team's relationships.

Damage control is certainly considered a distinctive feature of women's talk in the language and gender literature.⁹

This claim is extensively documented in many of the articles in Holmes & Meyerhoff 2003b.

There is extensive evidence that women tend to use a wide range of both facilitative strategies and mitigating devices to attenuate negatively affective speech acts and maintain good relations (see, for example, Holmes 1995, Coates 1996). In the next section, we consider the extent to which the use of such RP strategies in the workplace can be considered to be predominantly aspects of “women's work.”

RELATIONAL PRACTICE, GENDER, AND WORKPLACE CULTURE

Fletcher's analysis of RP focuses on women. First, she argues that RP is “women's work,” and second, her subjects were six professional women in a predominantly male workplace, an engineering firm. The data she describes from her observations and interviews were collected from her women subjects, and they clearly support her view that RP is undervalued and invisible, and even “disappeared” – that is, discounted as unimportant and irrelevant to the organization's core business.

The material we have collected, and from which we have drawn the excerpts in this article, is taken from a wide range of workplaces, and the recordings involve both women and men. Our analysis clearly demonstrates that RP is by no means the sole prerogative of women. Readers may have noticed that we used a variety of strategies to avoid indicating the gender of the participants in the material analyzed above. We avoided names and singular pronouns, and where names were used, unisex pseudonyms were selected.¹⁰

So is RP undertaken more often by women than men in the workplace? Or is it the domain of the powerless and those with less status in the organizational hierarchy? This “essentialist” question raises a number of issues, some of which lie right at the heart of gender analysis. One is a simple practical matter: It is impossible to document all instances of RP. Many are nonverbal, and, by their very nature, most instances go undocumented and unseen. As we have repeatedly noted, RP is invisible, off-record, behind-the-scenes support work, which is typically regarded as irrelevant and dispensable. Counting instances is not a serious option. But, more important, as noted and extensively illustrated by Holmes & Meyerhoff, “The field has moved well beyond descriptions of (perceived or actual) differences between men's and women's speech, or finger-pointing that maps power hierarchies with gender hierarchies” (2003a:8). Current research is rather more concerned with analyzing the complex and skillful ways in which people at work accomplish different aspects of their workplace identity (Holmes 2000d, Holmes & Stubbe 2003, Kendall 2003). Doing RP is clearly directly relevant here. Regardless of the sex of the speaker, as a strategy for performing gender, RP typically contributes to the construction of a more “feminine” or other-oriented social identity.

However, while recognizing the inadequacy or invidiousness of essentialized, dichotomous conceptions of gender, we must acknowledge that in everyday life it really is often the case that gender is “essential.” As Holmes & Meyerhoff (2003a:9) point out,

This raises the important theoretical issue of why and how RP has become associated with women. RP is clearly not work that is undertaken only by women, but nonetheless it is work strongly associated with women rather than men. As described by Fletcher, and illustrated in our analysis above, RP is “gendered” work, stereotypically “feminine” behavior. It is support work, oriented to people's face needs, to preserving their relationships with others and their dignity and self-esteem. As Fletcher notes, “Relational practice is not gender-neutral behavior. It is behavior that engages deeply held gender identities and beliefs” (1999:133). Eckert & McConnell-Ginet make a similar point: “Traditional women's jobs are in the service sector, and generally involve nurturing and support roles …Wherever they are, women are expected more than men to remember birthdays, soothe hurt children, offer intimate understanding” (2003:11). So, even when undertaken by men, RP is perceived as feminine behavior because of the gender stereotypes that pervade our culture (see Cameron 2003, Talbot 2003).

This is what we mean by saying that RP involves “gendered” discourse. Gender is a sociocultural category, and, regardless of whether we are biologically perceived and categorized as women or men, we are continuously constructing relatively masculine or feminine social identities in our ongoing social interactions. The discursive strategies that Fletcher identifies as components of RP are strategies stereotypically associated with feminine rather than masculine ideals. Whether used by women or by men in the workplace, they are generally associated with feminine stereotypes and perceived as contributing to a more feminine than masculine social identity. As Eckert & McConnell-Ginet 2003 make clear in their extensive discussion of the deeply embedded nature of the ideological beliefs associated with gender, it is impossible to escape the constraints of such ingrained social practices. Our discourse is drenched in gender. Assumptions about what constitutes more “feminine” as opposed to more “masculine” ways of talking are constantly being reinforced in everyday interaction, and the process of “gendering” individuals is clearly an ongoing, dynamic one. And, undoubtedly, the instances of RP exemplified in the previous sections illustrate the kind of supportive, facilitative, interpersonally oriented talk, which is typically perceived as women's talk and assigned to the category “feminine” rather than “masculine.”

But this is clearly not the whole story. Our analyses suggest that there may be other, distinctly unfeminine ways of doing RP. Focusing on the functions of RP, as identified in the three criteria specified at the outset of this essay, our research suggests that there are alternative off-record means of “creating team,” for example, while also furthering workplace objectives. Although some communities of practice (those identified as more “feminine” in interactional style) tend to adopt the kinds of strategies discussed by Fletcher and illustrated in detail in this article, other, more “masculine” communities of practice appear to prefer different ways of doing RP. These orientations emerged in particular in our studies of humor, an indisputably off-record discourse strategy in the workplace, but they were also evident in other aspects of workplace interaction.

For example, in a detailed analysis of the management style of two women in contrastingly “gendered” workplaces – a government department and a soap factory – Holmes and Stubbe 2003 show that, while both women used relational strategies in their interactions (as well as more assertive and directive strategies where appropriate), there were nevertheless noticeable differences in their preferred interactional styles. In the workplace regarded as more “feminine” in its overall style of interaction, there was “a marked orientation towards collaborative styles and processes of interaction, together with a high level of attention to the interpersonal dimension” and a preference by the manager for “less direct, more linguistically polite strategies to achieve her goals in a consensual way” (Holmes & Stubbe 2003:587–88). Self-deprecating humor was used to reestablish good relations after the manager had been particularly directive. In the factory team, a more “masculine” community of practice, the team leader generally adopted more direct, authoritarian, and forceful strategies to communicate with her team as a whole, including unvarnished criticism when the team failed to meet its targets. The humor in this workplace was very different from that in the “feminine” workplace: It was very frequent and much more “sparky,” contestive, and aggressive, with a good deal of jocular abuse, good-humored insult, and sarcastic comment (see Stubbe 2000).

These analyses suggest an alternative perspective on the issue of RP as a gendered phenomenon. It seems at least plausible that different communities of practice develop not only distinctive ways of handling humor in the workplace, but also different ways of doing RP. We are not referring here to differences in the distribution of particular RP strategies – such differences are inevitable – but rather to the more fundamental issue of possible differences in what the researcher “counts” as RP in different workplaces.

Most of the teams in the workplaces in which we recorded provided evidence of the feminized discourse of RP that we have exemplified above. People skillfully made use of subtle and off-record strategies to “create team” and to pay attention to face needs, to reduce face threat, and to manage potential conflict, while furthering their work team's objectives. The teams in some workplaces, however, presented rather different patterns. In one team (which we will label “Maxiteam”) there was scarcely any conventional small talk, for instance, and the humor was predominantly aggressive and sarcastic, directly face-threatening rather than supportive or attenuating in its effect. And as mentioned, in the factory team the humor consisted predominantly of aggressive jocular insults and verbal abuse. Both these communities of practice were perceived as very “masculine” in their style of interaction. Their workplace discourse was often confrontational and aggressive, with frequent challenges and disagreements.¹¹

This question provoked us to reconsider the notion of RP, as defined by Fletcher. The “small talk” of the Maxiteam members, for example, which occurred at the beginning and end of their meetings, was consistently work-focused. Though it was off-record and off-topic in terms of the agenda of the current meeting (see Holmes 2000b), it was generally related to some other aspect of the team's work. In the six meetings of this team that we videotaped in full, there is scarcely a single topic that is not directly related to some aspect of their work. And the humor of this group was predominantly aggressive and face-threatening. Over 90% of the humorous comments that occurred in one meeting, for instance, were sarcastic and negative jibes, intended to put down the addressees or deflate them.¹²

Clearly, what happens in such communities of practice is that those who “can't hack it” leave. One man commented, for instance, that the interactional style of Maxiteam was so uncomfortable for him that he applied to move to another team. In her study of an Australian bank, Beck notes that “a considerable number of senior women managers despaired at the ongoing masculinised culture … one woman referred to the Bank's ‘macho’ culture which had driven her to leave her post after only eight months” (1999:205). It seems worth considering, then, how discursive strategies that appear so alienating to some employees may form the underpinning for or constitute the bonds of strong working relationships between others. In Fletcher's account, RP fosters workplace relationships through behaviors associated with women, behaviors conventionally regarded as polite, facilitative, and feminine. Consideration of ways of creating team, while also furthering workplace goals, in more “masculine” workplaces suggests that the term RP may need to be “degendered.” This proposal also has interesting implications for relational theory, which assumes that the primary function of RP is to foster interpersonal relations in order to promote personal growth (e.g., Gilligan 1982, Miller 1986, Miller & Stiver 1997). If alternative forms of RP are recognized, the answer to the question of whether they have the same positive effects on the individual is by no means obvious.