Abstract
In this paper, we examine a specific type of Relative Clause (RC). We look at the construction consisting of a ‘light noun’, that is, a noun with highly non-specific lexical content which does not do referential work, plus a relative clause. It has generally been assumed that the functional contribution of RCs is to narrow the set of referents of the head noun to only those for which the predicate of the RC holds true. However, the ‘Light Head RC construction’ (LHRC) has various interactional affordances that mostly revolve around characterizing referents with discourse-relevant properties rather than establishing reference. We argue that these various functions of LHRCs revolve around participants’ orientations to categoryhood. Data are in English and Dutch.
Introduction
Interactional Linguists are interested in seeking explanations of why grammars are the way they are in terms of the ‘contingencies of real-time talk in interaction’ (Fox, 2007). In this paper, we consider the interactional contributions of relative clause constructions (RCs
1
) in Dutch and English, two languages that have similar relative clause constructions. In particular, we consider
‘Light Head-Relative Clause Constructions’ (LHRCs) are of interest because their two components (a light head noun and a relative clause) have been analysed to have opposing functions. It has generally been assumed that the functional contribution of RCs is to narrow the set of referents of the head noun to only those for which the predicate of the RC holds true (Berk, 1999: 265; Biber et al., 1999: 603; Quirk et al., 1985: 1239; Radden and Dirven, 2007: 161), often without much evidence for this position. However, as Fox and Thompson (1990a, 1990b) show, RCs have various interactional affordances: they allow interlocutors to make novel referents relevant by anchoring them to established discourse referents, to introduce novel referents while maintaining membership-category-relevant expectations, and to characterize a referent with discourse-relevant properties. In contrast, the literature on light nouns suggests that nouns like person and thing mostly carry out functions other than reference (Biq, 2004; Mahlberg, 2003; Sinclair, 1999). Mahlberg (2003: 100) suggests that light nouns frequently display a ‘support function’: a light noun ‘helps to present information according to the communicative needs of the speaker/writer and hearer/reader’. Our goal in this paper, then, is to consider LHRCs in conversational language use and to analyse what they can tell us about why speakers use RCs in general. We will show that LHRCs display a range of interactional functions beyond ‘narrowing the set of referents’. These various functions, we argue, revolve around participants’ orientations to
Our data
In considering LHRCs, we restrict our attention here to those ‘light nouns’ that are used to describe people and things (the two most frequently occurring classes in our data):
Our RCs were culled from transcripts of naturally-occurring audio- and/or video-recorded conversations. All of the participants are native speakers of American English or Netherlandic Dutch. The Dutch data were automatically extracted from the syntactically annotated part of the Corpus Gesproken Nederlands (Corpus of Spoken Dutch; henceforth CGN) 3 and manually checked for accuracy and completeness. The English data were collected manually from private corpora, consisting of more than 20 hours of everyday conversations among family and friends. Our collection includes 157 Dutch LHRCs and 77 English LHRCs,
LHRCs invite participants to orient to categoryhood
As opposed to most literature on RCs, our data show that LHRCs are rarely used to establish or narrow down a discourse referent. In many cases it is dubious whether the LHRCs are used referentially at all; in fact, nearly all our examples of LHRCs seem to function instead to establish and present an ‘ad hoc category’, that is, a
For example, consider the Dutch extract in 4):
In lines 3 and 5, Tine reports what she said to the woman she was talking to. Here, the contribution of the relative clause die ik zelf niet lust ‘that I don’t like myself’ (line 5) is not easily seen as a ‘restriction of the referential range’ of the head noun dingen ‘things’. But what
LHRCs, we argue, offer a useful grammatical pattern to express such an evaluative stance, because they present a state of affairs as a
Such interactional affordances, we argue, are made accessible by the way in which interlocutors interactionally treat the property of being a category. In particular, our data suggest that interlocutors are operating under a ‘commonsense model’ of categoryhood that includes the following. First, in people’s commonsense model, categories have essences: unobservable properties that make them the way they are and that exist independently of us as perceivers of those categories. Through these essences, categories also have affordances: ways in which they relate to goals, desires and other subjective, relational states obtaining between the category and us. Finally, through essences again, categories have extensions: any instance of the category has the essential attributes of the category, whereas all non-instances lack the essential attributes. This, in turn, allows us, upon hearing an LHRC, to consider the set of people or things that are members of this category, in other words: the extension.
While all of these properties have been theorized in philosophy and linguistics as relevant aspects of what categories are like (e.g. Aristotle’s Metaphysics for essences (Ross, 1924), Gibson (1979) for affordances, and Frege (1884) for extensions, here we instead consider them primarily as aspects of a commonsense model: a culturally shared ‘organized body of considered thought’ (Geertz, 1983: 75). On this view, our oriented-to conception of categoryhood and the use of this conception in interaction are co-constitutive: essences allow us to talk in terms of timeless, subjectless truths, affordances allow us to negotiate goals and desires as they pertain to referents that have such essences, and extensions help joint reasoning about the existence of category instantiations. It is to an exploration of these aspects of categoryhood and their various interactional functions that we turn next.
Category formulations can be motivated by Essences
Essences are properties that are stable across time and that are definitional for the category (Gelman, 2003; Medin, 1989). As such, they deal with the criteria or conditions under which we can call something a member of a category. In its simplest form, we see the employment of LHRCs to orient to a category’s essence in definitions. In fragment 5), Tom explains the difference between a teaching assistant (in a university context) and a co-assistent (resident in a medical context):
As noted in Sidnell (2016), the definitional formulation ‘NP is RC’ is a grammatical format that allows interlocutors to resolve trouble concerning lexical knowledge, as is the case in 5). We add to this analysis that predicated LHRCs are also well-suited for doing definitions because they present the lexical knowledge as a case of category assignment, with the LHRC
Extract 6) is analogous: again, in response to a question-word question, Virginia’s LHRC (line 8) is designed to embody the criterial property of ‘gwaffs’, that they are ‘really immature’. In this case, the formulation as a category can be shown to be sequentially relevant when comparing the formulation in lines 8–9 with an alternative formulation as a predicated adjective (a gwaff is just really immature). This alternative formulation, by naming just one characteristic of gwaffs, fails to respond appropriately to the question in line 6, which asks for a definition. In line 9, then, we see that the formulation as a category, but not the formulation as a predicate adjective, allows Virginia to refer back to the categorical status of ‘gwaff’ with the anaphoric like that (shaded).
In extracts 5) and 6), speakers support their formulations of definitions by orienting to the conception of categories as immutable and invariant. We find further evidence for the fact that the orientation to categoryhood is ‘real’ for interlocutors in extract 7) below. Here, a small family group has been discussing running marathons. Betsy, an American who has been living in Europe, offers her impression of the activity of running marathons.
After about 4 minutes of listening to the discussion about running marathons, Betsy’s LHRC in line 4 suggests that for her, an essential and invariant characteristic of training for, and running in, marathons is how very American the whole activity seems to be. Betsy orients to the status of ‘things that are very American’ as a category, not just by means of the predicated LHRC, but also by explicitly charactering it as this whole concept in line 13, that is, by choosing a noun that categorizes the topic under discussion as a category. This act of classification reorients the conversation to one jointly characterizing a
Could Betsy have offered her opinion without an LHRC, that is, with the clausal predicate adjective construction seems to be very American, instead of seems to be something that’s very American (line 4)? We suggest that with the LHRC in line 4, she invites grouping running marathons with other ‘very American’ activities, like celebrating Thanksgiving, for example. Formulating the assessment of marathons as something that has the quality of being ‘very American’ with an LHRC doesn’t just assert that marathons have that quality, but also presents the quality as being immutably present for this objectively existing category, similar to the ‘things I don’t like myself’ in extract 4). A predicate adjective does not contribute this characterization, and is therefore less well suited to discuss marathon racing on a ‘conceptual’ level (which the speakers seem to orient to, as fragment 8) shows.
The orientation to a commonsense understanding of category essences also supports other discourse functions, as fragment 9) illustrates. This fragment presents the upshot of a story the couple Fred and Marit are telling their friends Sanne and Jos. When Fred and Marit were moving, they had left their washing machine unattended on the sidewalk. Coming down to check on it, Fred noticed a man loading it into a van. The man is reported as saying that he regularly looks through the trash for valuable things, and thinking the washing machine was trash. Fred and Marit wrap up the telling by jointly characterizing the man, and in response to a challenge from Marit, Fred incrementally characterizes the situation as a ‘crime of opportunity’.
Starting the concession that he thought the man knew he was doing something wrong in line 1, Fred notices trouble with his acknowledgement of the man’s bad intentions: the characterization of the man as ‘knowing damn well that it wasn’t right’ makes accessible the possible inference that he was a thief. It is that inference that Fred seems to repair in line 3, by characterizing him as niet iemand die echt wasmachines stal ‘not someone who really stole washing machines’. Here, the formulation with a plural indefinite and a simple past, wasmachines stal, licences an interpretation as a habitual statement, but at the same time, an interpretation as a description of a past activity is not ruled out. The ambiguity of the formulation as having either a habitual or a simple assertion interpretation then leads him to self-repair the characterization in line 9 with another LHRC iemand die leefde van de wasmachine stelen ‘someone who was making a living by stealing the washing machine’. Here, the habitual interpretation is made explicit in the formulation leefde van ‘made a living by’ and the generic understanding of the plural wasmachines ‘washing machines’ is underscored by a reformulation as a (somewhat unconventional) definite noun phrase.
This way, Fred achieves a formulation that ascribes bad intentions to the man, but by negating membership of the category ‘someone who steals washing machines’, presents it as accidental rather than essential. The LHRCs in line 3 and 9 orient the speakers away from the single act of taking the washing machine and towards a category of acts (‘stealing washing machines’) and an accompanying social category (‘people who steal washing machines for a living’), which can then be negated. Looking at alternative grammatical formulations, we notice that a main clause formulation (‘I don’t think he really stole washing machines’) further biases an interpretation towards an interpretation of a single sanctionable event. On the other hand, reformulating it as a predicated noun compound (‘I don’t think he really was a washing machine thief’) presupposes that the lexical category of ‘washing machine thief’ is shared and accessible, which may not be the case.
Category formulations can be motivated by affordances
A second important aspect of the commonsense conception of categoryhood is that categories have characteristics that allow us to interact with them in certain ways, and for certain purposes. Adopting Gibson’s terminology, we can say that interlocutors orient to the affordances of category membership in talk. Relatedly, and applied to social categories, such affordances have been observed as interactionally relevant in the framework of membership categorization analysis (MCA: Sacks, 1972; Schegloff, 2007). In MCA, social categories are understood to have
We find that category affordances are frequently oriented to through LHRCs. In a case such as extract 4) above, we saw that the LHRC dat zijn dan
Assumed inferences in fragment 9) can also be seen to motivate Jet’s use of an LHRC:
In this fragment, the LHRC in line 1 categorizes the referent in a way that is relevant for the goals of Jet, who is nervous about her job interview. The newly introduced referent is characterized as having already had such an interview, allowing for the commonsense inference that the referent of ‘somebody’ has the epistemic authority to talk about it (Heritage, 2012a, 2012b; Raymond and Heritage, 2006). The relevance of this authority is oriented to in the second part of the coordinated clause in line 2: die zei dat het heel erg ontspannen was en gezellig. ‘They said it was very relaxed and convivial’; the reported assesser has the right kind of experience to make the attributed assessment presented in the reported speech.
In terms of categoryhood, what we see here is that a novel referent 4 is introduced as a member of a category (‘people who have had a job interview with this particular company’), and that the category membership makes inferences accessible (that this person can speak to the issue of what the job interview is like). An alternative formulation, in which the informational content of the LHRC is split into two main clauses is possible as well in both Dutch and English (as in ‘I spoke to someone. They had already had such an interview and they said that it was very relaxed and convivial’), but in this case, the introduction of the referent with iemand/someone without a relative clause, is less ‘useful’ to Jet for two reasons. First, drawing on Sacks and Schegloff’s (1979) argument that speakers design their turns so that recipients can recognize who they are referring to, we see that iemand ‘someone’ will not do for the work that Jet needs this clause to do, since it is a ‘non-recognitional personal reference’ (Sacks, 1995, vol 2). This being the case, Jet’s clause ‘k heb iemand gesproken ‘I spoke to someone’, upon its production, would fail both to reveal its relevance to Jet’s previously stated concern about the upcoming job interview, or to project possibly relevant talk concerning the referent of iemand. Second, this non-LHRC clause would simply state that this individual has already had a job interview and then report on what they had said. Such a juxtaposition would of course allow for the inference of the epistemic authority from the statement that they have already had the job interview. But it would not situate that person as a member of a category of people whose characteristics grant them the epistemic authority to speak to Jet’s concern, thereby facilitating that inference, which is understood to follow from the new referent’s category membership.
An interesting property of the category presented by the LHRC in fragment 9) is that the category is ‘ad-hoc’ and ‘goal-derived’ (Barsalou, 1991), as opposed to, for example, extract 6), where Virginia’s gwaff is presented as a category with characteristics that do not overtly relate the category to a goal. Barsalou characterizes
John’s point is that the course emphasizes how to organize one’s life around other things besides smoking. John’s LHRC in line 6 designates a category that is made relevant in terms of not only the short-term goal of having something to do each day when the urge to smoke strikes, but of the long-term goal as well of quitting smoking and sustainably dealing with this urge.
Category formulations can be motivated by Extensions
The final property of categoryhood that is frequently made relevant by an LHRC is the fact that categories have extensions. In a commonsense conception, categories define sets of people and things that can appropriately be referred to with the category’s label. Whereas an orientation to category essences and affordances selects the conceptual, or ‘idea-like’, character of a category, orientations to their extensions focus on the referential potential: are there entities in the world that this category can refer to, and if so, how many are there (compared to some other category)? This function seems similar to the referential function offered in the literature on RCs. However, we argue that the referential potential of an LHRC is instead derived from its formulation as a category rather than being its primary function. We find such orientations frequently in LHRCs that are subjects in existential clauses (and in related constructions such as the direct object of to have). Often, the extension of the category is oriented to as a way of introducing the (kinds of) people and things that fit the extension, as in fragment 10):
In line 1, Sophia asserts that her parents ‘are still trying to pursue legal action against that guy’, and in line 7, she starts to provide the motivation for her parents’ case, as a main clause, after which she reformulates it as an existential clause with an LHRC in line 8. We attribute her reformulation to the fact that if she had continued with this main clause, ‘you’re not supposed to lie about some things’, she would have failed to introduce those ‘things’ as the novel category ‘things you’re not supposed to lie about’, on which her parents’ legal case rests. Critically, she goes on to exemplify the novel category by asserting that one doesn’t have to disclose deaths in the house but one has to tell the truth if asked. As line 12 indicates, this is indeed the kind of instantiation of the novel category that Sophia wants to talk about, so we can interpret the orientation to the existence of the members of a category afforded by an LHRC formulation as providing the background knowledge for the discussion of the elaboration of the legal action Sophia’s parents are pursuing.
In other examples, the formulation of a category having or not having members can be a vehicle for establishing local contrast. Consider fragment 11) below:
In extract 11), we see two LHRCs. With the question in line 1 asking about instantiations of the well-established category of ‘complaining neighbours’, Sanne concedes in line 4 that another, similar, category might exist, namely mensen die wel blij zullen zijn dat de kat straks weg is ‘people who are likely happy that the cat will soon be gone’. This LHRC introduces a novel, highly particular category (involving reference to a specific cat and a specific event of moving), which supports an interpretation of contrast between the two categories: the ‘complaining neighbours’, who are said not to exist, and the ‘people who will be happy that the cat will soon be gone’, who might exist. 5 The two categories are similar in inviting the inference that the neighbours are unhappy about the cat, but they differ in whether the neighbours express their unhappiness about the cat’s behaviour. A reformulation without an LHRC (sommige mensen zullen wel blij zijn als de kat straks weg is ‘some people will be happy that the cat will soon be gone’) would fail to support this contrastive interpretation, as the two important elements of the contrast (the kinds of neighbours and their existence) are not presented as being at issue: the existence of the category is relegated to a nominal quantifier like ‘some’ and the future happiness of the neighbours is not formulated as a category-defining property, but as an accidental state of affairs.
Sanne’s concession involves the assertion of the existence of an adjacent category (cf. Mazeland et al., 1995), which allows the interlocutors to then orient to the members of this adjacent group, as Jos does in his telling of their neighbour Marga in lines 7–17, who was awakened by the cat, but who was also understanding about the situation. Marga is paraphrased as ‘not wanting to complain about it’, a formulation directly contrasting with the original formulation of the category of ‘complaining neighbours’ in line 1. Sanne then introduces a second neighbour in lines 18–22 as another concession to the weaker category of ‘unhappy neighbours’ This neighbour did express a less-than-positive assessment of the cat, but the quoted formulation of the assessment is not an overt complaint; the cat is just humorously said to be een schuinsmarcheerder ‘a rascal’. Again, this specific case is formulated as an LHRC, thereby presenting a specific communicative event as an instantiation of another adjacent category (as indicated by the existential-like formulation ik heb wel ‘ns iemand gehad die. . . ‘I’ve had someone who’).
Discussion
In the previous sections, we have shown how LHRCs are frequently used to allow speakers to make the category-like status of a state of affairs interactionally relevant. We presented three facets of interactants’ commonsense conception of categories and demonstrated how they recruit them for interactional purposes. Our analysis of LHRCs contrasts with the function typically assigned to RCs in general, namely to restrict the referential scope of the head noun’s referents. In our interactional data, the properties of LHRCs are better explained by interactional and intersubjective functions, which we argue are derived from the way interactants perceive categories as being ‘out there’. This should not come as a surprise: light nouns more generally are known to be used for such functions (Biq, 2004; Mahlberg, 2003; Sinclair, 1999). That a lexico-grammatical construction predominantly fulfils such a role raises some further questions which we explore in this section.
Further motivations for LHRCs?
The starting point of our work was a reconsideration of the function of relative clause constructions, in particular those with light nouns as their heads. A first question is how complete the account we propose is. In other words: are all the LHRCs we encounter in our data explained by an orientation to categoryhood? Since the interactional functions we discuss in this paper are non-exclusive, non-determinate, and hard to classify into discrete classes, quantifying the coverage of our account does not seem to be an appropriate method to answer this question. We consider instead other possible functions of the LHRC and discuss their occurrence in our data.
We did not find cases of LHRCs fulfilling exclusively a referential function. A plausible case can be seen in extract 12) :
Bert and Fien are trying to establish what a poer ‘foundation block’ is. Bert gives a definition by ostension by referring to a specific object he and Fien both know; the LHRC thus actually restricts the referential range of ‘things’ to only those ‘we have in the living room’, which, combined with Fien’s background knowledge of the kind of things they are trying to establish (where TVs, furniture, etc., have already been ruled out) should suffice for her to identify Bert’s referent. Such cases are, however, vanishingly rare in our data.
However, we occasionally do find LHRCs being motivated by other functions as well. In particular, the fact that LHRCs contain clauses, and the information-structural affordances that being a clause offers may motivate formulation as LHRC as well. An NP with a relative clause facilitates more lexical material, adverbials in particular, which in turn allows LHRCs to fulfil functions such as contrasting and expressing epistemic and evaluative stance in a way that non-clausal NP modifiers, such as adjectives and prepositional phrases, do not. In fragment 13), we see a good example of a speaker finding the clausality of an LHRC working to her advantage:
In line 11, Donna might well have formulated her question with a noun compound like ‘snake people’. But since the compound snake person in line 2 is already committed to the interpretation of ‘liking/not being freaked out by snakes’, this would not be a useful option. Somewhat more restrictive would be a formulation of this concept as snake owners. But this compound NP would have missed two features that the LHRC provides: first, snake owners could denote people who own snakes for research or to supply pet shops, whereas the clausality of the LHRC allows for both verbs like keep and prepositional-phrase adjuncts like for pets. We note that the function of orienting to the categoryhood of LHRCs is co-present with the function of providing a clausal platform for adjuncts, so the clausality is not a function that competes with the categoryhood orientation.
We can see the clausality of an LHRC working in a speaker’s favour in various of our examples. If we return to fragment 12), where John says they have (just) specific things to do each da:y, for instance, the clausality of this LHRC provides John with the opportunity to specify the ‘daily-ness’ of the activities with an adverbial each day. Similarly, in extract 8), the adjunct echt ‘really’ in iemand die echt wasmachines stal ‘someone who really stole washing machines’ is readily accommodated by the clausal nature of the formulation.
Implications and further questions about the ‘orientation to categories’ view
In our account, LHRCs contribute an orientation to the categoryhood of a state of affairs to the interaction. What we haven’t addressed yet is why LHRCs carry out this function in the first place. One possible answer derives from the fact that LHRCs are nominal expressions which, unlike non-complex nominals,
Regardless of precisely what motivates the orientation to categoryhood, our data show that the issue of categoryhood is of importance to speakers. The recruitment of the different facets of categoryhood for various interactional goals sheds light on a commonsense conception of what it is to be a category. What our data suggest, moreover, is that the commonsense conception of categoryhood is strongly enmeshed in interaction; it is a way of creating an interactionally-motivated shared conception of a state of affairs.
This characterization of how speakers consider categoryhood suggests ‘being a category’ is something that is first and foremost of relevance for interaction. Here, our analysis aligns with Edwards (1991) that categories exist for talking, as well as Silverstein’s (2004) analysis of cultural concepts as primarily existing in an interactional practice, but we crucially draw attention to the observation that the ‘metalanguage’ of our commonsense models is itself a commonsense model that exists in and for interaction. In other words, the category of category is ‘for talking’ too. This conception furthermore generalizes the insights of Membership Category Analysis: it is not just social categories such as immigrant, child and guy that should be seen as sequentially regulated interactional contributions, but categories more generally. Such a view suggests that the categories associated with our words, while constrained by language-external reality, may emerge to a large extent in and through interaction, in other words, that interactional requirements causally ground categories (Enfield, 2014).
Commonsense conceptions of categoryhood and grammar
If categoryhood is an active concern of speakers, it is likely that there are other lexico-grammatical patterns that conventionally allow speakers to recruit their commonsense knowledge of categoryhood as well. In the nominal domain, other constructions exist (sometimes called ‘occasionalisms’ (Crystal, 1995)) for creating categories ‘for the nonce’ (Clark, 1983). For example, Downing (1977) argues that Noun-Noun compounds are prominent members of this family, productively created ‘as ad-hoc names for entities or categories deemed name-worthy’ at particular interactional moments (p. 841). Ariel and Mauri (2018) find that the most common function of or constructions is actually to introduce higher-level categories, quite often ad hoc ones. Interestingly, whereas LHRCs illustrate speakers adopting a top-down approach, by explicitly defining the conceptual essence of the relevant category and deriving its extensions from it, or-construction categories reveal speakers applying a bottom-up procedure. These categories are ad hoc abstractions over the explicit or alternatives, each construed as a category member and together pointing to the higher-level category.
Creative agentive derivation in English and other languages might also be cited. With the perspective presented in this paper, it would be of interest to revisit the interactional motivations for choosing to formulate a state of affairs with a Noun-Noun compound, an or construction, or an agentive derivation, in particular when they are coined for the nonce. A more distant cousin of these patterns can be found in the verbal domain, where light verb constructions with proper nouns as the direct object (pull a Rumsfeld, do a Reagan, have a Chernobyl) are used to take a ‘similative’ meaning (‘behaving/being like X’) and make it into a category of events, that is, present it as a kind of event instead of as an event that is similar to another event.
The cases discussed here open the intriguing possibility that some parts of our grammar may be dedicated to orienting to the categoryhood of certain states of affairs that we are talking about. In doing so, these grammatical structures conventionally recruit our socially-shared commonsense models, so that their formulations as relative clauses, noun compounds, etc., eventually come to carry culturally grounded meaning (as opposed to meaning grounded in individual cognition).
Cross-language comparisons
Our data, in addition to the considerations presented in this discussion, suggest that it would be of great interest to consider ways of orienting to categoryhood in languages that are more distant from each other than the two considered here. There are two reasons why we might expect there to be noticeable cross-linguistic differences in the way relative-clause-like constructions are put to use in interaction. First, relative-clause-like constructions may be conventially associated with other functions than orienting to the categoryhood of a state of affairs. They may, indeed, be used more for the referential function that has been thought to be dominant for the English relative clauses, or they may may used more for information presented as secondary (like non-restrictive relative clauses). Of particular interest here are languages with grammatical patterns that deviate substantially from English and Dutch, such as the ‘Noun-Modifying Constructions’ discussed for Japanese by Matsumoto (1997), Matsumoto et al. (2017) and Takara (2012), and for Mandarin by Tao (2002) and Huang (2013: chap. 7). Secondly, it may be that relative-clause-like constructions are conventionally associated with an orientation to categoryhood, but the commonsense conception of categoryhood differs from the conception we presented for English and Dutch.
We hope to have shown, then, that English and Dutch speakers’ use of LHRCs sheds light on the ways in which recurrent linguistic patterns can invoke categories that matter at specific moments to participants engaged in everyday interaction.
Footnotes
Appendix A
Acknowledgements
We gratefully acknowledge the valuable comments we have received from Mira Ariel and Tsuyoshi Ono on this paper; neither of them is responsible for the way we have incorporated their input.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: “Barend Beekhuizen is supported by an NSERC Discovery grant (RGPIN-2019-06917)”.