2.1. Face and Politeness

Face—a person’s public self-image—has been a key topic in politeness research since Brown and Levinson’s (1987) seminal work. Brown and Levinson borrowed face, and the concept of politeness as an interpretation of face-work, from Erving Goffman’s (1967, p. 12) work, in which the concept of “facework” refers to “the actions taken by a person to make whatever he is doing consistent with face.” In Goffman’s theory, face-work thus includes a wide variety of practices, including among others, corrective face-work and avoidance face-work. Universalist theories of politeness gave pre-eminence to these (corrective and avoidance) forms of face-work, but in doing so excluded many other aspects of face-work originally noted by Goffman. In addition, the concepts of addressing others’ face needs and protecting one’s own face have been approached primarily through the concept of rationality that is assumed as part of the modus operandi of the CP, even though Goffman himself emphasizes that face-work has a strong emotive element. It is worth noting that emotions are also rational, as social psychologists such as Haidt (2012) argue, but their rationality differs from the calculated nature of ‘rationality’ in Brown and Levinson’s (1987) sense.

After the emergence of the second wave of politeness research, various scholars questioned the straightforward connection between face and politeness. More specifically, while perhaps no researcher has denied that face and politeness are strongly interrelated phenomena, it has been broadly agreed that the operation of face and face-work cannot simply be drawn under the politeness umbrella, as groundbreaking research by Bargiela-Chiappini (2003) has illustrated. In fact, it is even possible to conduct face research without venturing into the realm of politeness, as a recent edited collection by Bargiela-Chiappini and Haugh (2010) has shown in an insightful way. The following example, cited from Kádár and Haugh (2013, pp. 51, 52) illustrates the reason why researchers argue that face and politeness should be treated as different, albeit interrelated, phenomena:

This interaction is cited from the American comedy series Everybody Hates Chris. Here, Chris is trying to convince his parents, Rochelle and Julius, to buy some new clothes for him to wear to the school picture day. After pleading to wear something other than what his mother has picked out, his father indicates that he is not allowed to buy anything new. His mother suggests that he find something else (i.e., that he already owns or can borrow from his brother). It is at this point that the narrator (the grown-up version of Chris) comes up with a strategy, namely, alluding to the potential embarrassment, or threat to his family’s and thus his mother’s face, if people were to think they are too poor to buy new clothes. Rochelle reacts strongly to this potential face threat, and decides they will buy new clothes for Chris in spite of protests from Julius. However, despite the obvious salience of face in this interaction, it is quite clear that evaluations of politeness (or impoliteness) are not at issue here. Rochelle does not decide to get new clothes for Chris because it would be polite to do so, but because she wants to avoid having others think badly of their family. In other words, she wants to protect their face.

It is pertinent to note that academic interest in the concept of face has generated some noteworthy research on equivalents of the English ‘face’ across modern languages and cultures, such as Thai (Ukosakul, 2005), and Chinese and Japanese (Haugh & Hinze, 2003), and also in historical cultures (Ruhi & Kádár, 2011). Such metalexical research helps scholars to tease out culture-specific understandings of this notion. Such explorations have revealed that cultures and times have varying conceptions of face, and these differences also influence the ways in which politeness behavior—which has a strong intersection with face, in particular in popular culture—is conceptualized across language and cultures. In addition, face continues to be in the center of cutting-edge research on the interactional formation of interpersonal relationships (see Arundale, 2006; Spencer-Oatey, 2008). Recent research has also explored face beyond its understanding on the individual level (see e.g., Kádár, 2013).

2.2. Valency and the Moral Order

The rationale for politeness to operate is the existence of valence, which the interactants use as a benchmark for their production and evaluation of politeness. As Haugh argues,

[e]valuations in interpersonal settings […] involve the casting of persons and relationships into particular valenced (i.e., positive-neutral-negative) categories according to some kind of perceived normative scale or frame.

(Haugh, 2014, p. 159)

The notion of valenced categories, which plays a key in the politeness theory of Kádár and Haugh (2013), reveals that (im)politeness as a situated interactional phenomenon cannot exist in a vacuum; its operation presupposes the existence of some common ground between the interactants as regards the value of interactional messages. On the operational level, the enactment of a valenced category showcases the interactants’ polite intentions situated in a particular context. To illustrate the way in which valency operates, let us refer to an interaction, which has been analysed in Kádár (2017):

A couple is arguing in the park. Bystanders overhear the argument but seem conflicted over intervention. An elderly female bystander decides to intervene.

This interaction occurs in the reality show Primetime: What Would You Do? The scene features public abuse, which triggers intervention from an elderly woman who draws the attention of the wrongdoer with “Hey buddy! Cool it!”, the inclusion of “buddy” with an accentuated and ironic prosody is clearly conflictive in that it conveys the opposite of its literal meaning: the person addressed is not a friend of the intervener. While “buddy” is not necessarily used to belittle, it is pragmatically appropriate to signal disagreement or opposition, and this meaning is even stronger in this interaction due to the emotive context and also to the age gap between the wrongdoer and the intervening person. In order for the elderly woman’s evaluative utterance “That’s not how you treat someone” to take place, she needs to have a valenced category of the fair treatment of others; reference to this category implies that a) the wrongdoer is behaving in an unacceptable (and, as such, impolite) way, and b) the intervening person has the right to intervene, and so her interruption is not impolite. Observers of this interaction may understand clearly why the woman makes this evaluative utterance—which illustrates that valence tends to entail a common ground either between the participants of an interaction, or a participant and an observer of an interaction, or both.

Valenced categories are not only shared by certain individuals: their operation assumes that they reflect the interactants’ undelaying perceptions of what counts as (in)appropriate in an event in a communal sense. Such perceptions can be defined through the concept of the “moral order.” The moral order is often approached in the field by following Garfinkel’s (1964) study of routinized activities (see an overview in Kádár & Haugh, 2013). That is, the moral order tends to be interpreted as a set of conventions in the form of valenced categories that become visible if and when they are violated. For example, if someone’s greeting is not responded to, this person is likely to notice this lack and voice it in terms that reflect his understanding of the other’s behavior as inappropriate. Along with this definition of moral order, recent research by Kádár and Marquez Reiter (2015) and by Kádár (2017) use the moral order in Douglas’ (1999; see this concept also in Douglas, 1968, 1986, 1991) and Whutnow’s (1989) social anthropological and sociological sense, as a collective term for the normative flow of events and perceived social good, which are often animated and maintained by communal actions (see rituals in Section 3.5). As Douglas (1999, p. 299) argues, “people all over the world contrive to incorporate nature into the moral order”; in terms of interaction this implies that any individual is surrounded by a cluster of perceived of moral orders, and uses or evaluates language according to the moral order that a given context or interpersonal relationship triggers. These moral orders count as normative from the perspective of the language user, and they are moral in the most common sense of the word: if someone violates the moral order, this violation triggers the feeling that something is inappropriate, and this sense of inappropriateness tends to be voiced in some form on the evaluative level. For example, in the interaction studied above, the intervening person’s valenced category reflects the belief that the on-going event would be perceived as immoral by the broader society (which provides the right for this person to intervene), and as such the act of intervention is needed in order to restore the normative flow of things. Importantly, politeness often co-occurs with moralizing comments even in non-conflict scenarios, as moral orders underlie norms of politeness (Terkourafi, 2011); that is, the moral order tends to be referenced even in cases when it is not breached.

Note that moral orders reflect situated (inter)personal values, and the studying of this notion helps researchers to examine culture-specific politeness values. For example, as Kádár and Marquez-Reiter (2015) argue, the moral order in the interaction above reflects the Judeo-Christian importance of being a good Samaritan and treating others fairly.

2.3. Politeness as a Social Action and Practice

Politeness is a social action (Goodwin, 2000), which embodies a social group’s practice. The operation of politeness involves evaluations prompted by social actions and meanings that are recognizable to participants (Haugh, 2013). Being associated with a certain practice does not imply that a certain social action is predestined to be interpreted in a certain way. Rather, it suggests that the act’s contextually situated evaluation may be influenced by the recipient’s perception of the social practice that the given action embodies, and the relationship between this social practice and the perceived moral order.

Approaching politeness as a social action helps analysts to go beyond the boundaries of language, which is key when it comes to politeness, as politeness phenomena often come into existence in interaction through a combination of linguistic, paralinguistic and nonlinguistic behavior (Arndt & Janney, 1985). The need to avoid limiting politeness to the boundaries of language has been emphasized since the 1980s; for instance, Ambady, Koo, Lee, and Rosenthal (1996) illustrates this point by the following narrative example:

Consider the following scenario (familiar to some): Mary, a graduate student getting ready to face a dismal job market, receives a letter informing her that a paper coauthored with her advisor, a fellow graduate student, and a senior undergraduate has just been accepted by a prestigious journal with very few revisions. Mary rushes to share the good news with her coauthors.

Consider another alternative: Mary reads that the article has been rejected. She knows that the paper will need major revisions if it is to be accepted in any journal. She has the unpleasant task of conveying the news to her coauthors. How will Mary convey the good and bad news? And how will she convey the news differently to her advisor, to her peer, and to the undergraduate?

[…]

She might say, “Well, guess what? The editors said ‘no’—looks like it’s back to square one!” with a confident vocal tone, directly gaze at the listener, shrug her shoulders, and smile, or she could say the same thing with a downcast gaze, a hesitant tone of voice, and no smile. Even though the linguistic content is the same, the two scenarios will be interpreted quite differently.

(1996, pp. 996, 997)

As this example shows, language is just one, albeit important, element of politeness behavior.

2.4. Interactional Co-Construction

Although politeness is a social action that embodies social practices, hence animating the perceived moral order(s) of the interactants, this does not mean that the production and evaluation of politeness always follow regular and predictable patterns—interactants may agree or disagree about what counts as polite, and interpersonal politeness may come into existence in the form of interactional negotiations, as a co-constructed outcome of an interaction. The study of co-construction has gained momentum in the second wave of politeness research, as a criticism of the universalistic frameworks that operate with straightforward and invented utterances (e.g., Mills, 2003). Yet, it is pertinent to note that the concept of co-construction is present in third wave frameworks that do not limit their focus to the micro-level of interpersonal behavior. On the one hand, experts of interactional style, perhaps most notably the works of Cook (2006, 2008) have pointed out that switches between interactional styles, which are associated with different types of politeness behavior, tend to follow the dynamics of interactions. For example, in a Japanese academic consultation, lecturers and students may continuously make switches between formal (honorific) and informal styles, in order to index distance and sympathy at the same time; according to Cook, this kind of behavior is the norm rather than the exception; that is, the interactional co-construction of politeness is not necessarily an idiosyncratic form of behavior. On the other hand, third wave theories, such as Kádár’s (2013) recent framework, argue that co-construction can be observed even in recurrent and seemingly straightforward practices associated with politeness—it is a phenomenon that should be incorporated into theories that aim to capture politeness on both the macro and the micro levels. Politeness, as it unfolds in interaction, tends to operate with the interactional features of incrementality and sequentiality. Incrementality refers to the way in which speakers’ adjust or modify their talk in light of how the progressive uttering of units of talk is received by other participants. In other words, the fact that social actions and meanings are produced incrementally in interaction means they are inevitably subject to ongoing evaluation as they are produced, and so can be adjusted accordingly in real time. Sequentiality, on the other hand, refers to the way in which current turns or utterances are always understood relative to prior and subsequent talk, particularly talk that is contiguous (i.e., immediately prior to or subsequent to the current utterance). This means that next turns are a critical resource for participants in reaching understandings of the evaluations of others, including inferences of one’s interactant’s understandings of one’s own evaluations (see also Kádár and Haugh, 2013). Another aspect of sequentiality is that certain recurrent form of interaction are expected to follow strict sequential characteristics, and deviations from these characteristics tend to be sanctioned by default. This does not imply that incrementality does not operate in such interactions, but rather that it is more constrained than in punctuated and relatively free-flowing interactions (see Kádár, 2017).

Note that the first, second, and third waves do not necessarily follow a temporal order. Ideas of the second wave have been present within the first wave of politeness research, and the same applies to the third wave.

2.5. Understandings of Politeness

A key criticism that second wave politeness research has made about the first wave is that researchers impose their own understandings of politeness onto the data studied. As Eelen (2001) has pointed out, politeness can be divided into first-order and second-order types, the first including the language users’ understandings of politeness, with the latter covering theoretical/the theoretician’s understandings of it. This essential distinction—which recurs in various forms in various politeness theories, such as Watts (2003), Locher (2004), and Locher and Watts (2005), just to mention a few representative examples—is key to disentangling the interactants’ evaluative moments from the theoretician’s own evaluations. This distinction becomes particularly important if researchers focus on longer chunks of interaction. In the second-order conceptualization of politeness, researchers have used various technical terms, such as politic behaviour (Watts, 2003) and rapport management (Spencer-Oatey, 2000) to distinguish their own academic definitions and understandings from that of popular ones.

This bipartite approach has been further elaborated by Kádár and Haugh (2013), who propose a more complex approach to various understandings of politeness. As they argue, from a user perspective, there are four inter-related perspectives from which the nature of politeness, as an assumed part of our social reality, can be understood:

The notion of ‘meta-participant’ includes participants who do not actively engage in an interaction, but who may contribute to evaluations of (im)politeness behavior. The concept of ‘emic’ refers to insider understandings, while ‘etic’ refers to outsider understandings of (im)politeness.

From an observer perspective, there are four inter-related ways by which we can account for how we evaluate something to be polite, not polite, impolite, and so on in the first place.

Thus, there are four important loci, not just two as commonly thought, that constitute the first-second-order distinction, namely, participation (participant/meta-participant) and expectancies (emic/etic), which are first-order loci of understanding, and observation (analyst/lay observer) and conceptualization (theoretical/folk theoretic), which are second-order loci of understanding.

3.1. Politeness and Impoliteness

In first wave approaches, impoliteness plays only a small role, supposedly due to the focus of these works on rational behavior and conflict avoidance through facework. Research on impoliteness in pragmatics was started by Culpeper’s (1996) groundbreaking paper, which models impoliteness behavior through pragmatic lenses. With the emergence of second wave approaches to politeness, impoliteness has gained momentum in the field, and a number of high-impact studies have been published on this area, including Culpeper, Bousfield, and Whichmann (2003), Culpeper (2005), Bousfield (2008), Bousfield and Locher (2008), and Culpeper (2011b).

Impoliteness research has brought a large number of key innovations into the field. For example, it has brought fresh blood into academic discussions on the concept of intentionality, which has been a recurring theme in politeness research. It is clear that a range of impoliteness behavior comes into existence when someone intends to offend the other; however, as Culpeper (2011b) points out, (full) intentionality is not necessarily a precondition for impoliteness to operate. It is possible to be impolite unintentionally. Impoliteness has also contributed to research on emotions (see Kienpointer, 2008; Locher & Langlotz, 2008)—an emerging and important area within the field—due to researchers’ focus on the reactions (usually: the feeling of being upset) triggered by impoliteness behavior (see Işık-Güler & Ruhi, 2008). Impoliteness has also generated interest in culture-specific behavior: at the moment, researchers know relatively little about impoliteness behavior in certain languages and cultures, such as Chinese and Arabic, and it can be rightly supposed that a variety of high-impact studies are yet to appear on this area in the near future. The study of impoliteness has also triggered research on phenomena, like abuse and bullying, that have been relatively ignored in pragmatics in spite of their importance in other fields such as social psychology (see Kádár, 2013). Finally, impoliteness research has generated some noteworthy interest in metapragmatics: on the one hand, researchers have pursued interest in metapragmatic behavior triggered by impoliteness (see e.g., Ferenčík, 2015), and on the other hand they have undertaken thought-provoking research on metalexemes, words that are used about politeness (e.g., rudeness vs. impoliteness; see Culpeper, 2011b).

It can be argued that impoliteness research is an area with importance beyond its own borders because it is challenging to study politeness without discussing impoliteness, and vice versa. Due to this fact, researchers in the field often use the label (im)politeness when they discuss politeness phenomena in general.

3.2. Intercultural Politeness

When it comes to the concept of culture, it is not far-fetched to argue that politeness research at the moment is a heavily cross-cultural rather than intercultural field: while a large number of studies have explored politeness across cultures in a comparative way, relatively few studies have been engaged in the examination of how people from different cultural backgrounds interact with each other. In addition, since Eelen’s (2001) seminal study, culture has been treated in the field as a problematic term, due to the fact that first wave theories tend to associate culture with national culture (see a detailed discussion in Spencer-Oatey & Franklin, 2009). Yet, the exploration of intercultural politeness is a key task, considering the importance of politeness behavior in intercultural interactions.

The existing key studies within this area mostly focus on politeness in interaction between native speakers of English and speakers of exotic languages such as Chinese (Chang & Haugh, 2011; Pan & Kádár, 2011), Japanese (Nakane, 2006), and Korean (Murphy & Levy, 2006). A characteristic of these studies is that analysts usually focus on the difficulties that arise from intercultural communication, that is, they implicitly interpret culture on the national level, as a potential barrier of communication. While works such as Sifianou (2013) have shown in a powerful way that this is not necessarily the case, and culture is often an addition to many factors that influence interpersonal interaction in the globalized world, it remains a task for future research to integrate intercultural communication—in which culture tends to be interpreted in more complex ways than in pragmatics—with politeness research. The recent project by Spencer-Oatey & Kádár (2016) aims to address this knowledge gap. It is hoped that future studies will contribute to the development of this area by examining intercultural politeness in a wide range of naturally occurring interaction types across various languages and cultures.

An additional area of interest is the relationship between politeness, intercultural communication, and English as a lingua franca. For instance, in recent work on politeness in English as a lingua franca, House (2008) found convincing evidence that challenges the view that intercultural interactions inevitably give rise to perceptions of impoliteness, as participants of interactions tend to strategically reinterpret utterances that may cause misunderstandings and that would thus trigger impolite inferences.

3.3. Cross-Cultural Politeness

Cross-cultural politeness has played a central role in the field since its foundation: in Brown and Levinson (1987) and other high-impact theories, the notion of culture has been present as a starting point for conducting comparative analyses of politeness behavior. Various frameworks, perhaps most importantly Sifianou (1992), have tested the applicability of Brown and Levinson on culture-specific data. Cross-cultural politeness has become one of the most high-profile areas in the field due to Blum-Kulka, House, and Kasper (1989) Cross-Cultural Speech Act Realization Project, which compares requests and apologies across cultures with the aid of discourse completion tasks. Along with publications of pragmatic theoretical scope, a number of works have engaged in the empirical mapping of politeness across cultures; a most representative example is Hickey and Stewart (2005).

Cross-cultural research has also played some role in second wave politeness research. A mainstream publication is Kádár and Mills (2011), which compares politeness behavior across East Asian cultures by using a strictly second wave approach to politeness. Yet, culture has remained a relatively low-key issue in the second wave of the field, due to researchers’ focusing on politeness as a co-constructed and idiosyncratic form of behavior: as the volume of Bargiela-Chiappini and Kádár (2010) has found, it is challenging to conduct a large-scale cross-cultural politeness research project by rigorously following the methodologies of second wave politeness research. This might be the reason why the bulk of theoretical research that was published in the first decade of the 2000s is heavily English-centred (although this has always been the case in the field to some extent).

Conceptualizing culture as a rigorous notion to be used in cross-cultural analysis remains key in politeness research, including third wave approaches such as Culpeper (2011b) and Kádár and Haugh (2013). It is pertinent to note that the examination of culturally-situated politeness is significant beyond the realm of politeness research, namely, that the examination of culturally situated understandings of politeness can contribute to attempts to extend pragmatic theory beyond Anglo-academic conceptualizations and debates; such attempts tend to be defined within the scope of emancipatory pragmatics (see Hanks, 2014). It is pertinent to note that, while in-depth pragmatic research on politeness has been carried out in various languages such as Spanish and Chinese, it is very often the case that research in these languages still utilizes notions that come from the Anglo-academia.

3.4. Historical Politeness

If culture is a suitable testing ground for politeness theory, the same can be argued about historical data: by exploring temporally (and spatially) different cultures, researchers can explore the validity of frameworks that reflect modern understandings of interpersonal politeness. History has been present in the field throughout its development. Some works, such as Gu (1990), have attempted to link modern norms of politeness behavior with historical ones, while others such as Watts (1999) have examined historical data in order to illustrate the validity of politeness frameworks. Furthermore, in historical pragmatics, historical politeness has been studied for its own sake: key topics of such inquiries include historical terms of address (see e.g., Taavitsainen & Jucker, 2003), ideological understandings of politeness (see e.g., Klein, 1994), the historical sociology of politeness (see e.g., Carter, 2001), and so on. In addition, Ehlich’s (1992) noteworthy study has examined historicity—the concept that all actions and understandings of politeness are situated in time—as a philosophical concept in politeness theory.

Historical politeness research has recently been established as an independent field, with the appearance of the two books that Culpeper and Kádár (2010), and Bax and Kádár (2012) dedicated to this theme. These books attempt to promote the use of historical data in politeness theorization, and they propose methodologies by means of which researchers can bridge the seeming gaps between modern and historical politeness research (in particular, in terms of methodology). Since many modern practices of politeness have roots in history, it is perhaps not too ambitious to argue that no theory of politeness can be complete without engaging in some form of historical research. In addition, the concept of historicity also plays a key role in the examination of politeness in modern narrated and mediated data, such as news reports on peoples’ (im)politeness behavior.

Recent key areas of historical politeness research include historical metapragmatic research on politeness related terms (see Kádár & Paternoster, 2015) and the examination of politeness in ancient languages (see Ridealgh, 2016).

3.5. Politeness, Convention, and Ritual

Politeness often comes into existence in recurrent forms of behavior, which are referred to in the field as convention and ritual. As Terkourafi and Kádár (2017) argue, convention and ritual differ in a number of ways:

At the same time, the two phenomena have a number of shared characteristics, including the following:

Due to these shared characteristics, convention and ritual play key roles in politeness behavior.

Convention has been broadly studied in the field since Grice’s work, while ritual has only been examined by a limited number of researchers, perhaps the most outstanding one being Bax (2010) (see an overview in Terkourafi & Kádár, 2017). The recent monograph of Kádár (2017) models the relationship between interpersonal (im)politeness and ritual from the perspective of ‘mainstream’ politeness theory.

3.6. Politeness and Society

As politeness is a socially situated phenomenon, it may not be surprising that social variables have received significant attention in the field. Such variables include, for example, age (He, 2012) and social class (Deutschmann, 2003; Mills, 2003). However, the most broadly studied social variable in the field has been gender, which has been studied since the formation of politeness research. First wave research tends to approach gender and politeness in categorical ways, by clearly distinguishing masculine and feminine forms of politeness behavior (see e.g., Holmes, 1988; Ide, 1982); while such research reflects a stereotypical and overgeneralized view of gendered language, it is important to emphasize that politeness research at the time followed broader trends in gender and language research (see e.g., Tannen, 1993). In the second wave of politeness research, gendered politeness received significant attention, following Mills’ (2003) monograph: a number of studies, including Mullany (2004), Mills (2005), and others have pointed out that gendered language is subject to interactional negotiations, instead of being a pre-existing phenomenon. Yet, various researchers such as Holmes (2005, 2006) have maintained that, while gendered language is subject to negotiations and interactional co-construction to some extent, the gender of the interactants tends to define politeness behavior at least to some extent. Along with mapping gendered language, research on gender has also contributed to understandings of the relationship between language and ideology in terms of politeness (see e.g., Okamoto, 2016).

4.1. Methods of Data Collection and Data Types

The first wave of politeness research has devoted much attention to the utterance level of interaction. This focus brought along with it certain preferred methods of data collection: in order to obtain data that is as illustrative as possible, various studies use carefully selected pieces of either (allegedly) naturally occurring or, more often than not, elicited data; the former refers to utterances that arise in spontaneous interaction, and the latter refers to utterances that arise in discourse or interaction facilitated through intervention by the researcher. Careful selection of data refers to the fact that many researchers set out with the methodological assumption that certain naturally occurring utterances can and must be excluded from the analysis on theoretical grounds. In practice this means, for example, that the analyst can ignore an utterance that deviates from what is defined as the standard usage of politeness. This methodology of data collection also presupposes reliance on a certain analytical stance—namely, observer (analyst) coding of linguistic politeness. It is pertinent to note that elicited data continues to be regarded as important in certain areas of politeness research, such as research on speech acts (see Marti, 2006), even though in theoretical research on politeness this methodology is generally regarded as problematic (Eelen, 2001).

Since the 2000s, a large body of politeness research has explored politeness in naturally occurring interaction, by focusing on both the production and the evaluation of politeness. Data types include face-to-face interaction (e.g., Pan, 2000; Watts, 2003), computer-mediated communication (e.g., Graham, 2015; Locher, 2010), and written texts such as news items (e.g., Neurauter-Kessels, 2011); note that current scholarship does not usually set up strict borderlines between written and spoken communications, and so this listing primarily aims to indicate that politeness research examines a wide variety of data types. It can be argued that, in terms of data, politeness is a broad church: politeness researchers with discourse analytic backgrounds have examined various datasets, spanning family data (e.g., Locher, 2004), through reality shows (Blitvich, Bou-Franch, & Lorenzo-Dus, 2013), to business letters (e.g., Pilegaard, 1997); critical discourse analysts have studied politeness in a range of institutional scenarios, such as political speeches (e.g., Harris, 2001), care homes (Backhaus, 2009), and police interviews (e.g., Thornborrow, 2002); conversation analysts such as Hutchby (2008) tend to examine politeness in data types preferred in their field such as counselling sessions and phone calls. While it is beyond the scope of this article to overview all data types studied in the field, the present discussion might have illustrated the variety of data involved in politeness research. One concept that keeps all these various datasets and research methodologies together is the interpretation of politeness as an interactional phenomenon. It is pertinent to note that the interactional view of politeness can be extended to the analysis of monologic texts, such as letters, by setting claiming that a monologue is a dialogue in a broader context (see several studies in Culpeper & Kádár, 2010). Thus, the interactional analysis of politeness can, in principle, include any text type.

4.5. Politeness and Metapragmatics

Politeness is not only important on the level of production, but also in the way in which interactants reflect on it; language that reflects upon language use is metapragmatics (see an overview in Lucy, 2004). There are three metapragmatic areas that are particularly relevant to politeness research:

Focusing on metapragmatics helps researchers to tease out perceptions and understandings that underlie politeness behavior.

Metapragmatic research on politeness started relatively early (see Blum-Kulka, 1992), and is one of the key research methodologies (see e.g., Meyer, 1995; Spencer-Oatey, 2011).

4.6. Units of Analysis

As Eelen (2001) explains, first wave approaches tend to use the individual as a unit of analysis to make projected descriptions of politeness in languages, societies, or cultures. This is a top-down approach, in the sense that it does not analyze politeness behavior on the level of localized individuals and smaller groups, and then build up their macro-views on cultures and societies on the basis of this. Instead, it is usually assumed that politeness phenomena associated with the given culture or society manifest themselves in the language use of the individuals. Most commonly, first wave theories have adopted culture as the key notion for explaining differences in politeness forms and strategies. As Eelen notes,

in Brown and Levinson’s discussion of ‘cultural variation’… the terms ‘culture,’ ‘society,’ and ‘group’ are used interchangeably. Sometimes the term ‘subculture’ is also encountered, although it is not clear how it relates to the other three.

(2001, pp. 159–160)

The problem with this approach is its normative characteristics, that is, an essentialist approach to culture and politeness presupposes that members of a certain culture tend to share these claimed values. This concept has often been criticized by second wave politeness scholars.

Due to this problem, various alternative units of analysis have been proposed. Perhaps most important among these units is the concept of “community of practice” (see Wenger, 1998); this refers to a group of people, who are brought together through engagement in a joint (often but not always professional) activity or task. A recurrent issue in the field, however, is that community of practice is a concept that has been created for the analysis of organizational discourse, and so it cannot be used to describe politeness behavior in all kinds of interaction. An alternative unit has been relational networks. This refers to sets of intersecting social links between persons who collectively form the basis of an identifiable group, such as pupils at a certain school, or residents of a certain area (see Milroy and Milroy, 1992).

In current politeness research, particularly in third wave approaches, politeness tends to be approached in a bottom-up manner, as researchers attempt to capture the regularities of politeness production and evaluation by examining large datasets of interpersonal interaction. Focus on individual productive and evaluative moves in interaction does not entail that the object of politeness research can only be individualistic behavior: as politeness is often situated in organizations and institutions (Harris, 2003), by examining politeness behavior, researchers can gain insight into norms of power and other sociopragmatic factors that motivate situated interactional behavior, and, as a next step, the cultural and social understandings that underlie the operation of such norms (see e.g., Schnurr, Marra, & Holmes, 2007).

Focusing on politeness situated in relational networks, organizations, and other situated settings also motivates research to go beyond analyzing politeness within the speaker–hearer dyad. Certain interactions operate within complex participation frameworks (Goffman, 1981); because of this, the production and evaluation of politeness comes into existence through a) the involvement of participants beyond the dyad of the speaker and the hearer, and b) with awareness of the presence of such participants (see an overview in Kádár, 2017). For example, someone’s interpersonal behavior tends to change significantly if this person is aware of the presence of bystanders, eavesdroppers, etc.

4.7. Politeness Beyond Pragmatics

While politeness is predominantly a pragmatics-related field, it has emerged in other fields as well, and it can be argued that politeness research has an essentially multidisciplinary nature. To mention a few key examples, politeness has been studied through the lenses of language acquisition and socialization (see e.g., Ochs & Schieffelin, 2009), cognition (see e.g., Escandell-Vidal, 1996), social psychology (see e.g., Holtgraves, 2005), and sociolinguistics (see e.g., Morand, 1996).