(Im)politeness as object, (im)politeness as perspective

1 Introduction

The distinction between first-order politeness and second-order politeness was proposed as a way of challenging the validity of first wave theoretical accounts of (im)politeness that did not take into account commonsense understandings of (im)politeness (Watts et al. 1992).^[1] The distinction was premised on the argument that if (im)politeness researchers do not consider the perceptions of participants themselves, they lose sight of the fundamental object of their research. While the focus was initially on distinguishing between commonsense and scientific concepts of (im)politeness, the first-second order distinction has since been framed as encompassing a number of different contrasting perspectives over the past two decades (Eelen 2001; Haugh 2012; Kádár and Haugh 2013; Landone 2022). Yet while there has been a lot of emphasis placed on these different perspectives on (im)politeness in the field, what constitutes our object of understanding(s) in the field arguably remains somewhat more elusive.

It is widely acknowledged that (im)politeness is inherently discursive. Not only in the sense that the nature of (im)politeness itself can be disputed by users and observers, but more fundamentally in the sense that without interaction between people there can be no (im)politeness. But it also means that there must be an object for those people to discursively negotiate or dispute in the first place (Haugh 2019). What, then, is our object of research? To say that one is focusing on first-order understandings of participants or the second-order understandings of analysts, or even drawing from both, does not really address this issue. This is because it is a fundamentally ontological question (i.e., the study of what is), not simply an epistemological one (i.e., the study of what it means to know). We know there are different ways of looking at and interpreting our interactions with others, and that these ways of understanding or interpreting these interactions have very real world, material consequences. But what is it about these interactions that we are trying to understand through (im)politeness research? Brown (2017) suggests, quite reasonably in my view, that (im)politeness is fundamentally about attending to the feelings of others with regard to how they expect to be treated. But what then are we examining when we study (im)politeness? Does it encompass feelings, expectations, behaviour, judgements, culturally defined sets of concepts, or all of the above? And if it is all of those things, what then distinguishes (im)politeness research from any other scientific endeavour that is attempting to study the social world?

In this paper, I suggest that one of the reasons for the apparent elusiveness of (im)politeness is that we have inadvertently conflated (im)politeness-as-perspective (an epistemological issue) with (im)politeness-as-object (an ontological issue), and largely ignored the latter in ongoing debates about the first-second order distinction and its relative importance – or not – for (im)politeness researchers. I begin, in the following section, by first briefly reviewing the ways in which the first-second order distinction has been conceptualised and discussed in (im)politeness research. I next introduce two short excerpts from what appear to be fairly unremarkable, mundane interactions between co-participants who behave in ways we might not typically associate with politeness (at least from a broadly middle-class perspective). These observations are then used as a springboard for discussing the multi-layered nature of (im)politeness-as-object, and multi-dimensional nature of (im)politeness-as-perspective, in Sections 4 and 5, respectively. I conclude by considering the implications of these meta-theoretical reflections for the role of the first-second order distinction in (im)politeness research going forward.

2 The first-second order distinction in (im)politeness research

Debates about the role of the first-second order distinction in (im)politeness research have been ongoing for the past two decades since the early 2000s. However, the distinction itself can be traced back further to the introduction to an edited volume by Watts et al. (1992) in which they argued that theories of politeness up until that point had lost sight of their object of analysis. First-order politeness was defined by Watts et al. (1992: 3) as “the various ways in which polite behaviour is perceived and talked about by members of sociocultural groups”, while second-order politeness was defined as “a term within a theory of social behaviour and language usage”. The formulation of the first-second order distinction by Watts et al. (1992) thus echoes the distinction made by Giddens (1987) between “the first-order world of social actors and the second-order viewpoint of the social scientist”, when he cautioned about the problems that arise when we allow those viewpoints to become blurred (cited in Hutton 2017: 95). As Ehlich (1992) also argued:

the conceptualisations of politeness developed in scientific research should be distinguished from the phenomenon of politeness and concepts of politeness that exist in society. Scientific conceptualisations make use of everyday expressions to define politeness, and they very often do so in a completely unquestioning, matter-of-fact way, in other words – unscientifically. In many cases the relationship between these conceptualisations and the phenomenon that they examine remains unexplained. (Ehlich 1992: 73, original emphasis)

In applying the first-second order distinction to politeness research, then, primary emphasis was placed on distinguishing between commonsense and academic understandings of politeness. However, while Watts et al. (1992) framed the first-second order distinction as an epistemological one, they also made reference to the perceptions of members or participants, that is, their judgments or assessments of politeness. In so doing they also implicitly framed the first-order distinction as an ontological one as well, as perceptions, judgments or assessments are interpersonal or social phenomena, and thus an object of research that cannot be simply explained away as a perspective or viewpoint.

The conflation of (im)politeness-as-perspective (an epistemological distinction) and (im)politeness-as-object (an ontological distinction) that was introduced in the original formulation of the first-second order distinction has arguably persisted in subsequent second wave renditions of it.^[2] Mills (2003: 110), for instance, characterises first-order politeness in clearly ontological terms in arguing that “politeness is a matter of judgement and assessment”, although invokes epistemological concerns when discussing problems that arise when analysts impose their viewpoints in (im)politeness research (Mills 2005, 2011, 2017).

Eelen (2001) takes a similar, although somewhat more nuanced position in characterising first-order (im)politeness as simultaneously a commonsense, spontaneous concept involving “lay assessments about politeness” that emerges when “people talk about and provide accounts of politeness” (Eelen 2001: 32, original emphasis), and a practice of ordinary people engaging in “lay assessments of (one’s own or other speakers’) politeness in action” (Eelen 2001: 32, original emphasis). The latter, politeness-in-practice, is further broken down by Eelen (2001: 35) into expressive politeness1 (i.e., “politeness encoded in speech … where the speaker aims at ‘polite’ behaviour”), classificatory politeness1 (i.e., “hearer’s judgement [in actual interaction] of other people’s interactional behaviour as ‘polite’ or ‘impolite’”), and metapragmatic politeness1 (i.e., “talk about politeness as a concept, about what people perceive politeness to be all about” in interaction). However, while first-order (im)politeness is characterised in largely ontological terms, second-order (im)politeness is framed in epistemological terms as “the scientific conceptualisation of the social phenomenon of politeness in the form of a theory of politeness1” (Eelen 2001: 43). In other words, second-order (im)politeness involves analysis and theorisation by scientists (an epistemological viewpoint) about a first-order phenomenon (an ontological object).

Watts (2005), on the other hand, subtly nudges the first-second order distinction towards a reframing of it as an ontological one by suggesting that first-order politeness concerns “how participants in social interaction perceive politeness” (Watts 2005: xix), and “how people use the terms that are available to them in their own languages and … the discursive struggle over those terms” (Watts 2005: xxii). However, he also persists in characterising second-order politeness in epistemological terms as “a theoretical construct, a term within a theory of social behaviour and language usage” (Watts 2005: xx, citing Watts et al. 1992: 3). This equivocality is echoed in his subsequent characterisation of first-order politeness as “a quality of emergent social practice in a constructionist theory of human behaviour … assigned to interactants involved in that practice by co-interactants”, and second-order politeness as “a category to be defined, explained and operationalised in a rational theory of human behaviour” (Watts 2010: 55, emphasis added). On this account, first-order (im)politeness is both a phenomenon (“a quality”) and a viewpoint (“a constructionist theory of human behaviour”), and second-order (im)politeness is also simultaneously a phenomenon (“a category”) and a viewpoint (“a rational theory of human behaviour”).

Equivocality as to whether the distinction between first-order and second-order (im)politeness is an epistemological or ontological distinction is also apparent in a joint work with Locher that attempts to resituate (im)politeness within a broader category of relational work (Locher and Watts 2005). In their model, first-order (im)politeness is defined as “a discursive concept arising out of interactants’ perceptions and judgments of their own and others verbal behaviour” (Locher and Watts 2005: 10), while second-order (im)politeness (or what they term ‘relational work’) is defined as assessments of behaviour as ‘impolite’ (non-politic, inappropriate, negatively marked), ‘non-polite’ (politic, appropriate, unmarked), ‘polite’ (politic, appropriate, marked), or ‘over-polite’ (non-politic, inappropriate, negatively marked) (see Locher and Watts 2005: 12). However, the latter assessments are presumably made by analysts since such terms are not used by participants themselves.^[3]

Third wave theorists, in contrast, have tended to foreground epistemological over ontological concerns in their characterisations of the first-second order distinction. Terkourafi (2005, 2011 treats (im)politeness1 as involving spontaneous concepts of participants, while (im)politeness2 is characterised as a technical term used by social scientists, a formulation that is echoed by Christie (2005), who ties politeness1 to “participants’ conceptualisations of what counts as politeness”, while politeness2 is defined as the “analysts’ application of theories of politeness that have been generated by scholarship” (Christie 2005: 5). Haugh, in work with Kádár (Haugh 2012; Kádár and Haugh 2013), argues that the first-second order distinction involves multiple user-observer perspectives, including emic-etic, participant-metaparticipant, lay observer-scientific analyst, and folk theoretic-scientific theoretic understandings. Jucker (2020), on the other hand, characterises first-order (im)politeness as an emic perspective that is language specific, generally involving the “analysis of speakers’ talk about concepts” (Jucker 2020: 15), while second-order (im)politeness is characterised as an etic perspective that is language independent involving “language independent categories delineated and labelled by the researcher” (Jucker 2020: 15).

Overall, then, while there are important differences across these different ways of conceptualising the first-second order distinction, a broad consensus seems to have been reached that first-order (im)politeness refers to “the conception of the lay speaker in practice, both in defining politeness in an abstract way and evaluating it in the situated interactions in which she is involved”, while second-order (im)politeness has come to be characterised as “the understanding of politeness from the point of view of an external observer [that] is aimed at some form of theorising” (Landone 2022: 70). In short, “the relationship between doing and knowing is theorised in the distinction between first-order politeness (politeness1) and second-order politeness (politeness2)” (Landone 2022: 70, original emphasis). A key challenge for the field, however, is that “doing” and “knowing” are not easily disentangled. It is now widely acknowledged, for instance, that it is unproductive for (im)politeness researchers to impose “an analytic model of the researcher onto the psychological reality of the speakers” (Eelen 2001: 81), as that leads to ontological ambiguity regarding the status of concepts and practices, and epistemological ambiguity regarding legitimate practices of scientific theorising (Landone 2022: 73).

However, as we have also seen, much of the more recent debate to date has focused on epistemological issues with respect to the first-second order distinction in analysing and theorising (im)politeness. In such debates, the presumption has been that (im)politeness itself is an interpersonal attitude, judgement, evaluation or assessment by participants. One consequence of this assumption, however, is that while fairly nuanced accounts of different epistemological viewpoints on (im)politeness have emerged, there has been less considered debate about the ontological status of (im)politeness itself. I would suggest that this is because in debates about the first-second order distinction, and its relevance to (im)politeness research, we have inadvertently conflated (im)politeness-as-perspective (an epistemological issue) with (im)politeness-as-object (an ontological issue), and have not given sufficient weight to the latter in those debates.

Some scholars may contend that it is difficult, if not impossible, to disentangle object from perspective when studying (im)politeness. Jucker (2020), for instance, argues that while Eelen (2001) makes “a distinction between the thing itself and people’s perception of the thing”, one cannot study a practice without first having some concept in mind:

Scholars can either observe how people talk about behaviour for which they choose to use terms like polite, impolite, rude, civil and so on (politeness1 studies), or they may study a range of phenomena for which they introduce a technical term (such as ‘politeness’, i.e., politeness2). But they cannot study behaviour of a particular kind unless they specify what particular of behaviour they want to study. (Jucker 2020: 13)

It is indeed important to acknowledge that our object of study in (im)politeness research is inevitably both afforded and constrained by how we observe that object (Haugh and Watanabe 2017). What (im)politeness is taken to be depends very much on who is involved. However, in what follows it will be argued that since (im)politeness inevitably involves evaluation of some kind or another, we can formally distinguish between (im)politeness-as-object (an evaluation), and (im)politeness-as-perspective (how we come to understand that evaluation), even if in the final analysis (im)politeness-as-object and (im)politeness-as-perspective are invariably seen as dialectically entwined.

3 Data

In this section I introduce and make some passing observations about two pieces of data that illustrate not only the importance of carefully considering what kind of perspective we take on (im)politeness, but also press us to reflect more deeply on what constitutes our object(s) of analysis in (im)politeness research. The excerpts are taken from a larger collection of transcribed recordings of various interactions between six housemates who live together (Paul, Jak, Dazza, Darren, Nate, and Gab), and a friend who comes over regularly (Jason, otherwise known as “Bunger”), which were video-recorded by a non-participant observer (Roberta), who was a girlfriend of one of the housemates.^[4] The collection of 43 recordings in the Aussie Housemates (AH) dataset, which are 3 h and 40 min in total, primarily involve mundane, everyday interactions among male, broadly “blue collar” (or working class) co-participants.^[5] What is interesting about these interactions is that, on the whole, they hardly seem – from a broadly middle-class perspective at least – to be the place one might look for politeness. Indeed, perhaps quite the opposite.

Consider the following excerpt in which Jak is “requesting” Gab to move over and give him some space on the couch to watch television.^[6] The excerpt begins with Jak and Daz arguing about whether they should watch a movie or watch a football game (here “Broncos” refers to the Brisbane Broncos, a Queensland-based rugby league football team).

(1)

While it appears Jak has conceded they should watch the game in order to potentially have the chance to score one over Gab (who is a supporter of the NSW Dragon’s team), it is Jak’s subsequent imperative-formatted directive (line 11), Gab’s imperative-formatted response (line 13), and Jak’s reiteration of that directive (line 15), and accompanying use of the canonical insult, “cunt”, by both Jak and Gab that is of interest here. It would appear, on the surface at least, that what we have here are instances of impoliteness, as the directives are delivered through bald-on-record imperatives combined with a personalised negative vocative (Culpeper 1996, 2010). They are also reciprocated in a kind of tit-for-tat exchange, which is another typical feature of impoliteness (Culpeper and Tantucci 2021).

However, these instances are hardly unusual in this dataset. Indeed, they are so common it seems somewhat at odds to consider them examples of genuine impoliteness. It is also not clear that they constitute instances of mock impoliteness here either, as there is little to suggest that Jak does not want Gab to make room for him on the couch (i.e., the directive is indeed meant seriously). While I doubt very much that colleagues who have worked on impoliteness would consider these genuine examples of impoliteness, the potential risk here is that if we strictly adhere to our current models of impoliteness, we may inadvertently be invoking what Allan (2016) refers to as the “middle class politeness criterion”, or are tacitly bringing in middle class ideologies of (im)politeness to bear on the data (Mills 2017). What is at issue here is the role of our own subjectivities as researchers, and acknowledging the impact they can potentially have on our analysis and theorisation of (im)politeness.

Let us consider another excerpt in which a “request” that looks more canonically “polite” arises. Daz, Jak and Gab have been talking about working in high areas on building sites, and at the point this excerpt begins, they are talking about the use of harnesses. While Jak remains engaged in this talk (lines 1–6), he also at looks up a number on his mobile phone at the same time (Figure 1a), and then proceeds to make a call (Figure 1b), which is subsequently answered (Figure 1c).^[7]

(2)

What is important to first note here is how Jak formulates his request to speak to Bunger. He starts with a standard greeting (“hello”), followed by a modal form of a conventionally indirect request (“would X be there?”), which includes his friend’s name (“Jason”) rather than nickname (“Bunger”), followed by turn-final “please” (line 7). He also expresses gratitude using a standard form (“thank you”) when the person answering the call presumably goes to find Bunger (line 9). Jak’s voice quality is also clearer and higher in pitch when making this request, especially when compared to the much lower and more compressed voice quality he uses, in line 20, when Bunger eventually comes to answer the phone. It appears, then, that while Jak does not always use these kinds of canonically “polite” forms to make requests, he can indeed do so. Or at least he has chosen to do so here.

However, what is perhaps even more interesting to note is how Daz, who is an overhearer to this phone call, appears to subsequently respond to the way in which Jak has asked to speak to Bunger. Daz first lifts up his head gazing upward and starts laughing (line 11 and Figure 1d), before going on to urge Jak to ask for “Bunger” (rather than “Jason”) (line 13), and then provides an alternative way of formulating a request to speak to him (line 15). In so doing, Daz is hearably teasing Jak about the way in which he asked to speak to Bunger. Drew (1987) argues that teasing treats a prior action by the target as somehow “overdone”. In this case, it appears that Jak’s request is being construed as overly “formal” or “polite”, because Daz’s alternative formulation clearly involves a downgrading of the register or degree of formality of that request.

In this case, we see a participant treating what appears to be a canonically “polite” request – at least according to research to date on (im)politeness – as “overdone” and thus a legitimate target for teasing. In other words, what appears at first glance to be a typically “polite” request is itself treated as an object of negative evaluation by a co-participant. It appears, then, that can observe at least two forms or layers of evaluation here. It also appears that we are encountering an ideologically-charged stance with respect to (im)politeness that diverges from that typically associated with “middle class” politeness ideologies (Allan 2016; Mills 2017). Where does this leave Jak? Is he being “polite”, “over polite” or something else entirely? Or is this even the right question to be asking?

In the next two sections, I go on to suggest that if we are to start to unravel cases such as these, we not only need to traverse beyond binary perspectives involving “participants” versus “analysts” or “lay” versus “scientific” accounts of (im)politeness, but to also consider more carefully what kind of object(s) of analysis we are dealing with here in the first place.

4 (Im)politeness-as-object

The first-second order distinction can be understood as either an analytical, and thus epistemological, or an ontological distinction. An example of the former is the distinction in sociology that Giddens (1987) makes between “the first-order world of social actors and the second-order viewpoint of the social scientist” (Hutton 2017: 95). This epistemological distinction lies at the heart of the second, discursive wave of (im)politeness research, as we have already discussed. An example of the latter is the type-token distinction, first proposed by Peirce (1906), in which it is maintained that there is an ontological difference between identifying a “class” or “category” (type) of objects, and identifying individual “instances” or “cases” (tokens) of that class or category. The (second-order) type versus (first-order) token distinction arguably lies at the heart of the first wave of (im)politeness research, in which the driving aim was to identify culture-general sets of principles (Lakoff 1973), maxims (Leech 1983), or strategies (Brown and Levinson 1987) that underpin choices of interpersonally and situationally appropriate language across languages and cultures.

While it has been applied in various ways across disciplines (Hutton 2017), what underpins these various renditions of the first-second order distinction are processes of abstraction. These processes of abstraction can either involve generating progressively more abstract viewpoint(s), or reflexive object(s) of greater degrees of abstraction, or both. Abstraction, of course, lies at the core of any scientific endeavour, because through it we can study not only “non-recurrent, non-discrete events which can be located spatio-temporally” (Hutton 2017: 93, citing Love 1990) in all the rich and nuanced detail they demand, but we can also isolate “units or entities that are identifiable by virtue of their repetition or iteration” (Hutton 2017: 93, citing Hutton 1990). Notably, abstraction is not limited to professional analysts or scientists, but is also the means by which local participants or members themselves make sense of their worlds, if we allow that these units or entities can include those methods or procedures by which members themselves recognise and account for their conduct as recognisable and accountable (Garfinkel 1967). In other words, abstraction enables us to understand and explain the world around us whether we are lay or scientific observers, as well as, on some scientific accounts, provide for valid and reliable generalisations and predictions to be made about our object(s) of study.

What kind of object, then, is (im)politeness, and what kind of ontological distinctions can be brought to bear on it? While it lies beyond the scope of a single article to adequately address such a complex question, I would tentatively suggest that if we agree that what lies at the heart of (im)politeness research are evaluations of self and other, then (im)politeness-as-object should not be reduced to a simple either/or binary. Instead, I propose that (im)politeness-as-object is more productively understood as involving a complex, multi-layered set of ontologically distinct, but nevertheless reflexively-related evaluations. In short, my claim is that (im)politeness-as-object is not a first-order or second-order concept, but should be reconceptualised as a first-order and second-order, and in some cases, third-order evaluation.

(Im)politeness as first-order evaluation: While (im)politeness is nowadays frequently linked to evaluations of the social actions or conduct of speakers by hearers or recipients, evaluations are arguably immanent to social action at a more fundamental, first-order level. Early first-wave research on (im)politeness attempted to demonstrate that the forms and strategies by which “speech acts” are accomplished are sensitive to evaluations of “power”, “distance”, “imposition”, and the like (Brown and Levinson 1987). Subsequent third wave studies of the way in which social actions are interactionally accomplished has focused on the ways that interactants are demonstrably sensitive to evaluations of affective and deontic entitlement, epistemic authority, benefactive distributions, and spatio-material contingencies, among other things (Clift and Haugh 2021; Haugh 2013, 2015; Karafoti 2021; Mitchell 2022; Mitchell and Haugh 2015; Ogiermann 2019; Reichl 2021, 2024). Such work has demonstrated that evaluations are arguably immanent to social action in the sense that these evaluative sensitivities are built into the turn and response design of social action itself.

We can observe the way in which evaluations these kinds of considerations are built into the interactional design of the two instances of “requests” we discussed in the prior section. It is quite apparent from these two examples that the speaker in question, Jak, is orienting to different interpersonal and situational sensitivities in accomplishing those respective “requests.” (Im)politeness as a first-order phenomenon thus encompasses the way in which evaluations are immanent to the interactional accomplishment of social action itself.

(Im)politeness as second order evaluation: As (im)politeness researchers have repeatedly observed, however, social actions may also themselves become the object of evaluation. The judgments or assessments of (im)politeness that have occupied the attention of second wave approaches to (im)politeness are thus evaluations that are occasioned by social action (Haugh 2013; Kádár and Haugh 2013). Evaluations that are occasioned by social action do not, on this view, constitute first-order phenomena, however, but rather are more formally understood as second-order phenomena, in the sense that there can be no evaluation without there being an object to evaluate in the first place. As scholars of stance more generally have argued (e.g., Du Bois 2007; Jaffe 2009; Kiesling 2022), evaluations involve subjects positioning themselves with respect to other subjects through evaluating objects. In the case of (im)politeness, what lies at the heart of the “stance triangle” is arguably social actions, as illustrated in Figure 2.

Figure 2:

Evaluative stance triangle (adapted from Du Bois 2007: 163).

Given evaluation is immanent to the object in question itself, (im)politeness as a second-order phenomenon can be understood as evaluations that are occasioned by the evaluations that are built into, and thus immanent to, social actions. This takes us squarely into the territory of second wave approaches in which (im)politeness is characterised as “a quality of emergent social practice” (Watts 2010: 55) that is instantiated with respect to attitudinal or evaluative fields (Watts 2011; see also Kádár and Haugh 2013).

We can observe the way in which social actions can themselves become the object of evaluation by co-participants in the example of “teasing” about the formulation of another participant’s “request”, which we discussed in the prior section. The evaluation that is implemented through that teasing by Daz is ontologically distinct from the evaluations that are immanent to the design of Jak’s request. It follows, then, that while (im)politeness has been more commonly studied as an evaluation occasioned by social actions in second wave approaches, in so doing researchers have arguably neglected the way in which evaluation is also immanent to the interactional accomplishment of social action. In other words, the dual role played by evaluation has not been fully recognised in second wave critiques of first wave approaches to (im)politeness. On that view, the first and second waves of politeness research should not be set in contra-distinction or opposition, but rather address different sides of the same coin as it were. A useful way of bridging the gap between these two approaches to (im)politeness is thus to systematically analyse those social actions that deliver evaluations, that is, assessments, complaints, criticisms, compliments, and so on. This is because in these cases evaluation is not only immanent to their design, but they also constitute a form of social action in their own right (Haugh 2013).

It is important to note, then, that the way in which I define (im)politeness as a first-order evaluation does not, therefore, align with what lay participants might call “(im)politeness”, which is in my view more properly considered a second-order evaluation. In the case of first-order evaluations, participants are attending to the feelings of others with respect to how they expect to be treated through the way in which the social action(s) in question are themselves interactionally accomplished. In the case of second-order evaluations, participants are making assessments or judgments about the extent to which the social action(s) in question are attending to the feelings of others with respect to how they expect to be treated. Evaluative social actions thus lie at the intersection of these two layers, as they simultaneously implement first-order and second-order evaluations.

(Im)politeness as third-order evaluation: (Im)politeness researchers have also drawn attention to the ways in which evaluations occasioned by social actions may themselves become reflexive objects of talk or discourse. Evaluations that are accomplished through this kind of metapragmatic talk or discourse are invariably ideologically-charged (Eelen 2001; Kádár and Haugh 2013; Mills 2009, 2017). The model of evaluations developed by Davies (2018), in which she distinguishes between classifications, assessments and rationales, offers a very useful framework for studying these kinds of third-order evaluations. As Davies (2018) argues, by examining the implicit and explicit rationales that are used to justify (or contest) metapragmatic evaluations of (im)politeness, we can systematically draw out their ideological roots (Chang et al. 2021; Haugh and Chang 2019a). We can observe glimmers of different ideologies at play in the excerpts we discussed in the prior section. Indeed, such examples indicate more work needs to be done if we are to go beyond the largely “middle class” ideologies of (im)politeness that have dominated the field to date (Allan 2016; Mills 2017).

A multi-layered model of (im)politeness-as-object: The account of (im)politeness-as-object that has been sketched here is one in which (im)politeness is characterised as a first-order evaluation that is open to second and third order evaluations.^[8] On that view, (im)politeness is not a first or second order object as argued by first and second wave theorists, but is arguably simultaneously a first, second and third order phenomenon. The picture that emerges is one in which (im)politeness forms a complex, multi-layered set of first, second and third order evaluations that are reflexively-interrelated, but nevertheless ontologically distinct, as summarised in Figure 3.

Figure 3:

A multi-layered model of (im)politeness-as-object.

In this attempt to sketch out a model of (im)politeness-as-object, it has been proposed that evaluation may be studied as immanent to social action (first-order object), or as occasioned by social action (second-order object), with evaluation as a form of social action lying at the intersection of these two layers of evaluation. Finally, evaluations that surface in the conversational record through evaluative social actions, as well as evaluations that are occasioned by social actions (which may remain private or surface in various ways in interaction), can themselves become the reflexive objects of evaluation (third-order object). In other words, we can analyse the “doing” of (im)politeness as an evaluation of the situation accomplished through social actions (first-order evaluation), the evaluation of those social actions as (im)polite (second-order evaluation), or as evaluative metadiscourse about interactions (third-order evaluation). My view is that (im)politeness studies can benefit from studying all of these different orders of evaluation, and that this kind of multi-layered approach to (im)politeness-as-object should lie at the heart of third wave approaches to (im)politeness.

5 (Im)politeness-as-perspective

The first-second distinction in (im)politeness research was initially premised, as we previously discussed, on a distinction between commonsense and scientific concepts of (im)politeness (Watts et al. 1992). It was also argued that the purpose of any second-order (im)politeness theory is to explain the first-order concepts and perceptions of (im)politeness by members of different sociocultural groups (Eelen 2001). This binary distinction was subsequently characterised as also involving a distinction between (lay) participant and (professional) analyst understandings of (im)politeness (e.g., Holmes 2005; Ogiermann and Garces-Conejos Blitvich 2019), or between emic and etic perspectives on (im)politeness (e.g., Haugh 2012; Spencer-Oatey and Kádár 2016). Culpeper and Haugh (2021) suggest that it is these three ways of framing the first-second order distinction that have largely dominated understandings of (im)politeness-as-perspective to date, although they reframe the participant-analyst distinction in more general terms as one between users and observers. They go on to argue that:

[t]he first-second order distinction thus encompasses at least three distinct sets of contrasting perspectives: (1) commonsense, ordinary, non-academic versus academic, technical or scientific ways of talking and thinking about (im)politeness; (2) the understandings of participants themselves versus the understandings of observers of interactions; and (3) emic understandings of (cultural) members versus etic understandings of non-members. (Culpeper and Haugh 2021: 320)

One problem with such accounts of the different first-second order perspectives that can be brought to bear on (im)politeness is that they can give the (unfortunate) impression that what we are dealing with here is still a two-dimensional, binary distinction. However, a much more nuanced model of first-second order perspectives on (im)politeness has emerged over the past decade. As Landone (2022) has recently pointed out, Haugh (2012) and Kádár and Haugh (2013) refined the original two-dimensional first-second distinction into a multi-dimensional prismatic model in which “one person has multiple understandings of politeness at the same time, deriving from different perspectives, and not always carrying the same interpretative relevance” (Landone 2022: 70). On that view, then, first and second order perspectives on (im)politeness are not binaries; that is, they should not be set in contra-distinction or opposition, but rather form a complex, multidimensional prism of perspectives that can be flexibly and dynamically navigated by users and observers.

In revisiting this model, I would suggest that there are arguably four distinct loci of understanding (participant-metaparticipant; emic-etic; lay observer-professional analyst; folk theoretic-scientific), which are arrayed across three dimensions (user-observer; situated-abstract; footing-rationale). These can be rotated to foreground different perspectives on (im)politeness, as illustrated in Figure 4.

Figure 4:

A multi-dimensional, prismatic model of (im)politeness-as-perspective (adapted from Haugh [2012: 123], Kádár and Haugh [2013: 87] and Haugh [2018: 161]).

For instance, from a user’s perspective there are four epistemologically distinct understandings of (im)politeness available: participant understandings (those who are taking part in the interaction in question),^[9] metaparticipant understandings (those who take part in the interaction vicariously, such a television audiences or viewers of comments on social media platforms), emic understandings (cultural insiders or members), and etic understandings (cultural outsiders or non-members).^[10] From an observer’s perspective there are also four epistemologically distinct understandings of (im)politeness available: understandings of lay observers, understandings of professional analysts, folk theoretic understandings, and scientific theoretic understandings.

However, if we rotate the model, we can observe there are four epistemologically distinct footings (participant, metaparticipant, lay observer, professional analyst), and four epistemologically distinct rationales (emic, etic, folk theoretic, scientific theoretic) that can underpin understandings of (im)politeness. Rotating the prismatic model once again allows us to observe that there are also four relatively situated perspectives (participant, lay observer, emic, folk theoretic), and four relatively abstract perspectives (metaparticipant, professional analyst, etic, theoretic) on (im)politeness.

This does not mean, of course, that all of these different perspectives on (im)politeness are necessarily at play all the time. Participants are able to navigate the interconnected loci of (im)politeness interpretation quite naturally in social interaction. However, it is also abundantly clear that there are multiple ways in which different understandings of (im)politeness may arise, whether it involves participants from culturally similar or dissimilar backgrounds (Haugh and Chang 2019a). An important point to note then, is that according to this model, understandings of (im)politeness are more often than not rooted in multiple perspectives. Explaining variability in understandings of (im)politeness thus lies at the core of this multi-dimensional, prismatic model.

Haugh and Chang (2019b), for instance, argue that:

an emic understanding is properly grounded in the perspective of a cultural insider or member (in the ethnomethodological sense), and can be contrasted with the etic perspective of cultural outsiders or non-members, while a participant understanding is properly grounded in the footings of users in their instantiation of turns at talk (and in responses to prior talk and conduct), and can be contrasted with the perspective of observers of interaction. (Haugh and Chang 2019b: 924, original emphasis)

They thus propose a four-way model for the analysis of (im)politeness in interactions involving participants who orient to coming from different cultural backgrounds, as summarised in Figure 5 below.

Figure 5:

Emic-etic versus participant-analyst perspectives on talk and conduct (Haugh and Chang 2019b: 924).

As Haugh and Chang (2019b: 925) point out, however, “in practice participants are typically treated as members (of a particular cultural network), and so these two perspectives are, for all intents and purposes, very often laminated or fused together”.

Indeed, this is where the data we briefly examined in Section 3 raises serious questions about how we justify the perspective(s) we take, as analysts, on (im)politeness. It is obvious that in analysing these recordings that I am taking up an observer footing, albeit attempting to ground my inferences about the understandings of those participants based on their uptake of prior turns along with details of turn and response design (Clift and Haugh 2021). However, my footing as a cultural insider is much less clear-cut. While general familiarity with mainstream Australian English speaker norms offers me some degree of emic (i.e., cultural insider) insights into what is treated as marked or unmarked conduct by these participants, perhaps more so than what might be available to speakers of other varieties of English, it would be going a step too far to consider myself fully in tune with the interactional practices arising in these examples. In short, while I might be able to offer plausible interpretations of what is going on in these cases, the indexical value of these practices remains somewhat more opaque to me. Does that mean I am taking up an etic, cultural outsider perspective here? That seems a step too far as well. Perhaps my understandings lie somewhere between.^[11]

Indeed, closer examination of these four distinct loci of understanding suggests once again that we are not dealing with simple dualistic binaries, but rather perspectives that lie on clines. An understanding can be more emic, more etic or somewhere between. While some understandings of (im)politeness may be based on scientific theories, others may be grounded in folk theoretic accounts, and yet others may lie somewhere between (Kádár and Haugh 2013; Terkourafi 2011).^[12] And as Sifianou (2023) has recently argued, not only do lay understandings of (im)politeness enter into academic discourse, but academic understandings of (im)politeness may, in turn, enter into popular discourse. The boundaries between commonsense and academic understandings of (im)politeness are by no means as clear-cut as they might appear at first glance. The picture that emerges, then, is not strict adherence to a first or second order perspective (as argued by second wave scholars), but a commitment to drawing from both first and second order perspectives in varying combinations to varying degrees. What is critical for (im)politeness researchers to do going forward, therefore, is to recognise and acknowledge these shifts in perspectives in their work.

6 Concluding remarks

The first-second order distinction has dominated theoretical discussions about (im)politeness for the past two decades. However, while it has proven hugely influential, I have argued here that it has sometimes been characterised and implemented in ways that conflate politeness-as-object (its ontological status) with politeness-as-perspective (its epistemological status). I have argued that what lies at the heart of (im)politeness is evaluation, and have proposed that (im)politeness-as-object encompasses first-order evaluation (as immanent to social action), second-order evaluation (as occasioned by social action), and third-order evaluation (as reflexive ideological discourse). However, since (im)politeness is an inherently discursive phenomenon, what we take to be our object of analysis and theorisation is inevitably both afforded and constrained by how we go about observing it. This calls, in turn, for a more nuanced account of (im)politeness-as-perspective, which encompasses a range of different first order (commonsense, emic, user, folk theoretic) and second order (academic, etic, observer, scientific theoretic) understandings. (Im)politeness-as-perspective can thus be modelled as a multi-dimensional prism in which researchers inevitably draw on various combinations of these different perspectives to varying degrees. The latter point, as we have seen, is also crucial. These different epistemological perspectives do not constitute mutually exclusive categories, but almost invariably involve clines.

The set of epistemological and ontological distinctions that have emerged through this largely meta-theoretical discussion is admittedly complex and multifaceted. However, this should not really surprise us. Studying (im)politeness is to study an essential part of what makes us human, and human beings are quite obviously very complex and multifaceted. And so, therefore, is (im)politeness. There is still, of course, much work to be done to further refine some of the ideas that have been sketched out here. Some of the distinctions being made here may prove more productive than others. Yet other distinctions drawn here may ultimately fail Occam’s razor and the preference for theoretical simplicity where at all possible. However, no matter what the fate of specific elements in this proposal, I suspect that the days of grand theories of (im)politeness in which it can be reduced to a few key variables or set of simple binary distinctions are well and truly behind us.