Sage Journals: Discover world-class research

Abstract

In the tradition of Interactional Linguistics, we show how speakers make use of Predicate Nominal Constructions (PNCs) in American English conversation. We first provide a grammatical profile of PNCs as a highly frequent type of low-transitive clause in actual everyday use. We then bring evidence to show that, functionally, speakers use PNCs specifically to do two types of more semantic work (to identify or categorize a referent) and two types of more interactional work (to inform about or assess a referent). Finally, we examine the role PNCs play in the social activity of building a characterization.

Keywords

predicate nominal construction identifying categorizing informing assessing characterizing interaction

Introduction

Predicate Nominal Constructions (PNCs) are a prototypical type of ‘low-transitive’ clause (Hopper and Thompson, 1980; Thompson and Hopper, 2001). Following Hengeveld (1992), we consider the predicate in a PNC as one type of ‘non-verbal predicate’ (for an overview, see Haspelmath, 2024). We could schematize this as in Figure 1.

Figure 1.

Schema of low transitive predicate types.

Examples of PNCs from our collection (see below):

1) I was a cheerleader

2) it’s a co-op

The Predicate Nominal itself may consist of an ADJ plus a N:

3) that is a huge thing.

4) °it’s a true story°. ¹

Together with other copular clause types, PNCs are very low in Transitivity. As in most Indo-European languages, English PNCs have a copula ‘linker’, which is a ‘stripped-down’ verb: it inflects as most verbs do, but unlike verbs, it has no semantic content apart from its inflectional ability to show tense, aspect, and modality.

Motivating this study is a substantial body of research questioning traditional claims about constructions or linguistic units and categories, aiming to show how grammar serves as a resource for speakers in actual everyday social interaction (e.g. Fox and Thompson, 2007; Goodwin, 1979, 1981; Hopper, 2004; Keevallik, 2008, Koivisto et al., 2011; Ochs and Schieffelin, 1983; Ono and Thompson, 2003; Ono et al., 2021; Pekarek Doehler et al., 2015; Seppänen and Laury, 2007).

Predicate Nominal Constructions have been noted by many linguists, especially descriptive grammarians and typologists (e.g. Genetti, 2018, Payne, 1997, 2006; Verhaar, 1968), but we find essentially no research studying PNCs in interaction. Based on conversational data, we will show what form PNCs take in English, and what speakers use them to accomplish.

Data

Our American English data come from 40 hours of our own and colleagues’ datasets of mundane conversation, in addition to the CALLFRIEND data archive²; in all of these conversations the participants were simply interacting normally in various informal interactional settings with no instructions as to content from the researchers setting up the recording. All interactions were recorded with permission and transcribed by trained transcribers. From a subset of these interactions we exhaustively collected and analyzed 176 instances of PNCs.

Our research questions

(a) What form do PNCs take in everyday conversation? Are there distributional skewings favoring certain types of PNC formats?

(b) What interactional work do PNCs do in conversation?

What form do PNCs take in everyday conversation?

English PNCs are syntactically simple, with essentially three items in a fixed order: (NP COPULA NP). Strikingly, however, nearly all of the PNCs in our data show the following much more restricted set of properties, as illustrated in the schema in (5):

- a highly restricted subject slot (overwhelmingly non-human, typically it, that, or this)

- a fixed copula slot

- a semi-open Predicate Nominal slot³

- that is, many are formulaic expressions: big deal, problem, trouble, etc.

In our collection of 176 instances of PNCs, 169, or 96%, conform to the format in (5). Here are two instances of the 4% which have ‘heavier’ subjects, and thus do not conform:

(i) [Talking about an upcoming convention]

Monday’ll be the first day that I go.

(ii) [Distinguishing one individual from another]

the one that said twenty was the bo- the branch manager.

We noted above that PNCs in English are highly constrained with respect to their subject Specifically, 79% of PNC subjects in our data are non-human (Table 1).

Table 1.

Humanness of grammatical subjects.

Subject	N	%
Non-human	139	79
Human	37	21
Total	176	100

And a striking 86% of the non-human subjects are it, that, or this (Table 2).

6) th at’s a- that’s a swee:t co:mpliment,

7) it was a waste of time.

8) is this real butter?

Table 2.

Lexical category of grammatical subjects.

Non-human subject	N	%
it	71	51
that	33	24
this	15	11
other pronouns, nouns^a	20	14
total	139	100

Including non-human they, relativizer which, etc.

Although 79% of PNC subjects in our data are non-human, it has also long been known that in mundane conversation humans talk about themselves and other humans most (Du Bois et al., 2003; Givón, 1979). Based on Scheibman (2002: 91), we calculate that some 70% of all subjects in English conversation are human, in sharp contrast with PNC subjects, where only 21% are human. The profile for PNC subjects is thus just the opposite of what has been shown for subjects in general in conversation. We will return to this point in section ‘Conclusions’.

As noted above, we have found that many of the PNCs in our collection are formulaic expressions (Corrigan et al., 2009; Erman and Warren, 2000; Wray, 2002; Wray and Perkins, 2000). We follow Wray and Perkins (2000: 1) in characterizing a formulaic expression as:

a string of words . . . which is, or appears to be, prefabricated: that is, stored and retrieved whole from memory at the time of use, rather than being subject to generation or analysis by the language grammar.

While it is widely acknowledged that formulaicity is best understood in terms of a continuum rather than as a clear-cut category, we take the predicate nominals in examples such as these to be good candidates for being ‘prefabricated’:

that’s the problem/trouble it’s a pain in the ass it’s a pleasant surprise he’s a nice guy it’s no big deal it’s a joke it’s like a big deal it’s a Mickey Mouse course

In this section, then, we have shown what form PNCs take in American English conversation; we have described their grammatical properties, and their formulaicity. We turn now to how speakers deploy PNCs to accomplish their social business.

The work that speakers use PNCs to do

The work that PNCs are used for can be usefully understood in terms of two jobs that are more semantic (5.1) and two that are more interactional (5.2).

We hasten to make clear that we are agnostic as to whether speakers orient to a distinction between ‘more semantic’ and ‘more interactional’; we make this distinction for analytic purposes only, aiming for as clear a presentation as possible.

As a rough rule of thumb we can think of the ‘more semantic’ jobs, namely identifying and categorizing, as doing work that is somewhat less dependent on interactional context than are the interactional jobs. That is, whatever social work it might be doing, a PNC like that’s my husband is identifying the referent of its subject that. Similarly a PNC such as I was a cheerleader is categorizing the referent of its subject I, no matter what else it might be doing in its situational context.

In contrast, the job of a PNC doing primarily what we’re calling ‘more interactional’ work can generally only be determined by a close analysis of its position and composition in its context of use. Whether the PNC the mother’s sister is a real bigot, for example, is informing about, or assessing, its referent (or both), can only be determined by a careful examination of its use in the context in which it occurs.⁵

What semantic work do PNCs do? Identifying and categorizing

Our data suggest that PNCs can serve as a resource for speakers to either identify or categorize an entity or situation.⁶ Let us characterize each of these in more detail. We will first show how speakers use PNCs to identify a referent, and then consider how they work to categorize referents.

Identifying

In our conversational data in English, the speaker of a clause in which the Predicate Nominal is marked as ‘definite’ is nearly always working to identify the referent of the subject as a recognizable entity. In this study we take ‘definite’ to include NPs that are grammatically marked with a definite article (the) or determiner (that, this, those, etc.) or that designate a proper noun.

As with the majority of the PNCs in our collection, the subject of such identifying PNCs is generally it, this, or that. Here are four examples from our collection with the PNCs boldfaced:

9) KEL: I mean, ‘cause I really would need a stroller, and that would really be the right price.

Kelli, having heard about strollers for sale, identifies one of the prices her friend has mentioned as the right price.

10) MAR: we’re d ealing with, with this uh (0.7) ˙ tk .hhh these ki ds who’ve go tten out of dr u :gs. (0.3) .hhh this was their l ifesty : le ,

In (10) Marlene’s this refers to being on drugs; her PNC identifies being on drugs as formerly being their lifestyle, that is, the lifestyle of the kids in the theater group she is leading.

11) LIZ: I don’t know (.) I’ve been the:y’ve assigned me an advi:sor, and tomorrow I don’t have to go in til eleven it’s like⁷ the first day I’m going to have a sli:ght reprie:ve,

Here Liz identifies the subject pronoun it, referring to ‘tomorrow’, as the first day I’m going to have a slight reprieve.

12) 01 CAR: who is tha:t. 02 (0.9) 03 CAT: th[at’s: 04 PEN: [ that’s my husba nd .

In response to Carl’s question in line 1, Penny produces a PNC in line 4.

About ¼ of all the PNCs in our data are grammatically definite, serving to identify the subject.

Categorizing

Apart from the ¼ of the PNCs in our data that do identifying work, in the case of the other ¾ of our PNCs, the speaker places the referent of the subject into a category (Edwards, 1991; Mayes, 2024; Mayes and Tao, 2021). Our PNC data suggest that conversational participants assign certain properties to categories according to a ‘commonsense model’ of categoryhood; these include the following:

First, in people’s commonsense model, categories have essences: unobservable properties that make them the way they are and that exist independent 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. (Beekhuizen and Thompson (2022: 152)).

In other words, as we’ll try to demonstrate, to categorize something or someone is to draw on a range of associations based on the community’s shared understanding of what members of that category are like. Categorization can thus be thought of as a kind of ‘shorthand’ way of describing the properties of some referent to convey what that referent is like for speakers of a given community. Invoking the category label does a large part of this descriptive work.

For example, in the PNC in this phone call, in describing her schedule this semester to Bee, Ava first categorizes her Wednesday evening course as a ‘Mickey Mouse course’, then as a ‘joke’:

13) Ava: oh I have thee- I have one class in the e:vening. Bee: on M ondays? Ava: y-uh::: W ednesdays.= Bee: =uh-Wednesday,= Ava: =and it’s like a Mickey Mouse course. hh it’s a joke, hh ih speech. (0.2) Ava: hh[hh Bee: [speech,

When a speaker uses a Predicate Nominal to categorize its subject, the Predicate Nominal itself names a category understood to be familiar or identifiable to co-participants. In this way, of course, epistemic access to the category is dependent on culture and social relationships.

In (14), Clacia places Lehigh (that) into the category of ‘good schools’.

14) CLA: I did have lots of fun up at Lehigh. it was- tha - was a good school.

Similarly, in (15), Liz’s PNC in line 2 categorizes her new computer as having a nineteen-inch screen. At the time when this recording was made, the category of 19-inch screens entailed the property of being big for computers, as Liz makes explicit in line 3:

15) 01 ROB: is it bi::g hhh 02 LIZ: it’s a nineteen-i:nch scree::n! 03 (0.5) ⌈pretty big⌉

Summary

In this section, then, we have shown the semantic work that PNCs are used to do in American English conversation. We have analyzed instances of speakers using PNCs either to identify or categorize entities and people.

In section ‘What interactional work do speakers use PNCs to do? Informing and assessing’, we turn to the interactional work of PNCs in American English conversation.

What interactional work do speakers use PNCs to do? Informing and assessing

When we speak of ‘interactional work’, we mean the primary social action being done by a turn unit, what Levinson (2013) has referred to as its ‘main job’. Much has been written about social actions (e.g. Couper-Kuhlen, 2014; Deppermann and Haugh, 2022; Schegloff, 2007),⁸ and it is now clear that categorizing or naming social actions, by either participants or analysts, that is, action ascription, is controversial and fraught with the difficulties that beset any attempt to categorize behavior. We will not attempt to settle such questions here; our data suggest that PNCs in English consistently carry out one of two ‘main jobs’: ‘informing’ and ‘assessing’. This is not to say that a given turn unit cannot carry out both jobs, but, based on analysis of the activity the participants understand themselves to engage in, typically one of these two interactional jobs can reliably be said to be its primary action. We will discuss each in turn and then return to this point.

Informing

We use the term ‘informing’ as a cover term for clauses which deliver (or ask about) information, including what we vernacularly refer to as reports, news, and announcements. In their responses, participants routinely give evidence of their orientation to the action being done by that clause as providing information that recipients had (to a greater or lesser extent) not previously known (cf. Mori, 2006; Golato, 2010; Thompson et al., 2015). About 2/3 of our collection serve to deliver, or ask about, information.

Two examples of speakers using categorizing PNCs to inform are these from our collection:

16) CLA: the most ih - the most a m a:zing thing was to s ee the tu i tion we pai:d, hhh and to go o ver: and I was a cheer leader (when I went) there,

In (16), Clacia informs Diane that she was a member of the category of cheerleader in college.

In (17), Roy, Pete, and Marilyn are fixing dinner.

17) 01 ROY: do you have a par]ticular,um, 02 .. [use for the] red peppers, 03 PETE: [XXX X] 04 ROY: as opposed to the yellow or green pepp[ers]. 05 MARILYN: [no] no, 06 it was all .. salad peppers. 07 ROY: it’s all salad peppers.

Roy seeks information in lines 1 and 2 with a polar interrogative about the bell peppers he’s chopping up, do you have a particular use for the red peppers as opposed to the yellow or green peppers. Marilyn’s response in lines 5 and 6 is an informing, no no, it was all salad peppers, to which Roy uses a near-repeat it’s all salad peppers, displaying that he has receipted Marilyn’s response to his question.

And in informing Ava of her schedule in (18), Bee identifies the class she has on Monday and Wednesday as my linguistics course.

18) AVA: when’s your uh, whe- you have one day you only have one course uh? BEE: mMo[nday and Wednesday:[s right.]= that’s ] my lin guis tics= AVA: [hhhh [ oh. ] that’s-] BEE: =course [hh

We have thus seen that informing can be done with PNCs which are semantically either categorizing or identifying a referent.

Assessing

For about 1/3 of the PNCs in our data, the speaker assesses, or takes a subjective stance toward, an entity or situation (Goodwin and Goodwin, 1987). In our data, speakers assess entities and situations with either of two types of PNCs. In one type, the predicate nominal NP has an assessing adjective in it, as in (19):

19) BEA: it apparently- it was an extre:mely and unusually happy marriage.h

In the other type, the head of the predicate nominal NP itself indexes a stance; these are often fixed expressions, as in this excerpt:

20) LEN: it’s no:t as ba::d as li:ke yeah, be:ing bor:n on Chris:mas: da:y,which is to:tal: gar:bag:e

Summary

In this section, then, we have shown the interactional work that PNCs are used to do in American English conversation. We have analyzed instances of speakers using PNCs either for informing (i.e. delivering or requesting information, reports, news, or announcements) or for assessing (i.e. taking a subjective stance toward an entity or situation).

Semantic and interactional work done together

As we pointed out above, PNCs are often used to do both semantic and interactional work at the same time.

A PNC which both identifies a referent and informs a recipient about it can be seen in (12), repeated here as (21):

21) 01 CAR: who is tha:t. 02 (0.9) 03 CAT: th[at’s: 04 PEN: [ that’s my husba nd .

In response to Carl’s question in line 1, Penny informs him in line 4 that that is her husband, by identifying the referent of Carl’s that in line 1.

Similarly, in (22), two nurses are discussing a full-time position involving the care of a patient. Bea, in response to Rosalind’s query in line 1, uses a PNC to do two kinds of work: Bea informs Rosalind what kind of case it is by identifying the specific patient metonymically as the post brain surgery.

22) 1 ROS: = hhh oh uh what ^kind of a case is it. 2 (0.7) 3 BEA: uh-hum (0.3) ‘tk (0.6) <it’s the uh (.) post brain surgery,

And in (16), repeated here as (23), Clacia informs Diane that she was a member of the category of cheerleaders.

23) CLA: the most ih - the most a m a:zing thing was to s ee the tu i tion we pai:d, hhh and to go o ver: and I was a cheer leader (when I went) there,

Extract (13), repeated here as (24), illustrates two PNCs which together implement three of the four semantic and interactional jobs that we discussed above:

24) 01 AVA: oh I have thee- I have one class in the e:vening. 02 BEE: on M ondays? 03 AVA: y-uh::: W ednesdays.= 04 BEE: =uh-Wednesday,= 05 AVA: =and it’s like a M ickey M ouse course. 06 hh it’s a joke, hh 07 ih-speech. 08 (0.2) 09 AVA: hh[hh 10 BEE: [speech,

In lines 5 and 6, Ava informs Bee about how easy she finds her Wednesday evening course by categorizing and assessing it as a Mickey Mouse course and a joke.

And in (14), repeated here as (25), we see a single PNC performing the same three jobs: Clacia is both assessing and categorizing Lehigh as a ‘good school’, as well as informing Bee that it was a good school.

25) CLA: I did have lots of fun up at Lehigh. it was- tha - was a good school .

Summary

We have seen that for English speakers, PNCs are used to accomplish two types of jobs, one more semantic, the other more interactional. Speakers use them to identify or categorize the subject referent, and to inform about, and/or assess, the subject referent, respectively.

The fact that the work that PNCs do is limited to these two kinds of work is worth noting because such a simple clause structure might be expected to carry out a large range of interactional actions, and yet the number of uses to which English speakers put this clausal format is small. We argue that this is because, to understand what speakers are doing, we need to recognize that they are not using the full range of possible PNCs in the language. Rather, 96% of their PNCs are of the form shown in (5) above, most with it/that/this (and other pronominal) subjects, which can be understood as construction involving formulaicity. It thus might not be surprising that we find a relatively small number of functions for it.

Characterization sequences featuring PNCs

We turn now to briefly consider a favored environment for PNCs, extended sequences in which participants are engaged in the activity of characterizing someone or something. In the study of talk in interaction, the notion of ‘activities’ is generally ‘used to refer to larger action complexes, or “big packages” (Sacks, 1992[1971]), such as storytellings, argumentations, descriptions, . . .’ (Couper-Kuhlen and Selting, 2018: 28). In line with this understanding, we use the term ‘activity’ as in Robinson (2012) and Mazeland (2019); in mundane talk, ‘activity’ can include a goal-defined (i.e. completable) task, a reason for getting together (making breakfast), or identifiable phases of a get-together (cutting vegetables). For Robinson (2012), activities are sequences of action that are organized into clusters that ‘hang together’, including these social activities, among others: opening a birthday present in a group, telling a trouble, or telling ‘what happened’ (pp. 257–258). Like the text-linguistic notion of ‘genre’, activity types can be seen as constituted by language that makes the activity types recognizable, and conversely as constitutive of the interpretation of the language used in them.

One of the key activity contexts in which PNCs are found in our English conversational data is when interlocutors are working together to build a characterization. When participants engage in doing characterization, they are assigning attributes, states, and habitual activities to a referent in the service of specifying ‘what someone or something is like’. We will show how characterization activities involve recurrent low-transitive grammatical formats and lexical choices, including especially the specific PNC shown in (5) above (repeated here as 26), and the implications of the use of these formats for how sequences play out.

Characterization sequences preferentially aim toward affiliation and accountability, and interlocutors orient to this preference through their lexical and grammatical formulations. We aim, then, to answer a question of turn design (cf. Drew, 2012): how are lexico-grammatical choices responsive to the unfolding of this activity so far, and how are they consequential for the subsequent development of the activity? As we shall see, characterizations demonstrate the role of PNCs in longer sequences, revealing participants’ orientation to the aims of the activity, including affiliation and accomplishing intersubjectivity.

Our data show that PNCs that do categorizing work play a key role in the organization of characterization activities. Speakers use PNCs, together with other Low Transitive clauses, including habituals and especially Predicate-Adjective clauses, to collaboratively build a characterization of a person or situation with their interlocutors. In this excerpt, for example, Meg has asked Florence to tell her about the recent birth of the baby that Florence and her partner have adopted.

27) 1 FLO: but then fi:ve twenty-five Mae was born, 2 MEG: right, 3 FLO: and (.) the birth mother’s mo ther was with her through the whole 4 thing, 5 MEG: uh-huh. 6 FLO: and she was fan[tastic . 7 MEG: mhm. 8 FLO: she was ↑just great . she was so nice to us, and so solid there 9 for her daughter, (.) 10 and just a wonderful - (.) 11 MEG: woman (( )) 12 FLO: person 13 and um she was great.

In line 3, Florence introduces the birth mother’s mother, whom she assesses positively in lines 6 and 8 in a series of Predicate Adjective clauses. In line 10, Florence then begins to build a Predicate Nominal (and just a wonderful -), to which Meg, in line 11, affiliatively proposes a candidate head noun woman. Florence immediately provides her own head noun, person, in line 12 as an alternative completion for this NP.⁹

This entire sequence is done with extremely low-transitive clauses, further evidence for Thompson and Hopper’s (2001: 53) argument that:

the low Transitivity in our conversational data is to a considerable extent determined by the kinds of things we are doing when we talk with friends and acquaintances . . . our talk is mostly about ‘how things are from our perspective’. Our data show that we describe states, reveal our attitudes, ascribe properties to people and situations, and give our assessments of situations and behavior . . . these are the ways in which we display our identities, convey who we are to others, express our feelings and attitudes, and check our views of the world with our community-mates.

Three features of the design of Florence’s turn in lines 8–10 and 12 are notable. First, this sequential format, where Florence draws on a series of Predicate Adjective clauses assessing the birth mother’s mother, followed by a categorizing predicate nominal, conforms to a robust pattern in our data for talking about a non-present third party. Florence’s Predicate Adjectives do not evoke a category, but they help build Florence’s characterization; her ‘summarizing’ predicate nominal just a wonderful person, then, not only does evoke a category, that of ‘wonderful people’, but it also assesses the woman. Sequentially it completes Florence’s characterization of the birth mother’s mother. Second, Florence formats this clause with the word just, often found in PNC assessing characterizations, as in line 10.¹⁰ Third, when Meg in line 11 produces a candidate completion woman to Florence’s emerging PNC in line 10, just a wonderful –, she not only aligns with Florence’s characterization, she does so by symbiotically building on Florence’s PNC with a candidate predicate nominal of her own to complete the Predicate Nominal that Florence had just started.

Here is another example.

(28) [Three women friends (in their forties) talking about lactose intolerance] 1. MAY: .hhh I mean I take the::= 2. LIS: mhm 3. (1.8) 4. MAY: La:ctai::d= 5. LIS: =oh me too:. (In fact the) mi::lk, with lactaid. 6. (1.0) 7. LIS: ( [) 8. MAY: [oh ↑we’re ( ) the ↑s a me. 9. (.) 10. MAY: that’s right. .hhh so, right.=Before (0.8).hh I’ve just started to 11. buy that type of mi:lk.= 12. LIS: =yeah it’s something that happens to us as we get older. 13. (0.3) 14. LIS: .hh ((laugh))

In line 1, Maya begins a habitual mode low-transitive clause, but she leaves the patient NP incomplete; her ‘thinking face’ indicates that she is searching for the appropriate head noun (Goodwin and Goodwin, 1986; Hayashi, 2003), which her friends do not offer to help with. She finishes the clause with Lactaid¹¹ in line 4. With a further low-transitive clause in line 8, Maya establishes that she and Lisa are both using the same strategies to deal with their lactose intolerance. In the target turn in line 12, Lisa then characterizes lactose intolerance with a PNC doing informing, it’s something that happens to us as we get older. This complex predicate nominal, a (light head + relative clause) construction (cf. Beekhuizen and Thompson, 2022), allows Lisa to both name lactose intolerance as an ‘entity’ and characterize it in terms of a category of conditions that accompany aging. With this characterization, she thereby closes the sequence by categorizing lactose intolerance as a ‘thing’ to which the three of them are vulnerable by virtue of ‘getting older’.

To support this analysis, we note that if Lisa had formulated her assertion with a non-PNC, for example, ‘it happens to us as we get older’, that is, without the heavy NP ‘something that happens to us as we get older’, the clause would not have reified lactose intolerance as a ‘thing’ in the same way.

Finally, let us consider a collaborative characterization of an entity. Sophie, a Canadian, and Jason and Mary, who are American, have been talking about Americans’ attitudes toward Canada. In line 1, Sophie introduces a Canadian TV special.

29) 1 SOP: there’s a special, 2 MAR: [oh] 3 SOP: [that was] about like, 4 (1.3) 5 the guy went, 6 to talk to, 7 MAR: [mhm] 8 SOP: [Ame]ricans, 9 from all over the States. 10 (.) but it’s not like, 11 MAR: (.) oh, 12 it’s not like a re[gular pro-]- 13 SOP: [a weekly ]- 14 MAR: [oh] 15 SOP: [a regular] [program ] 16 MAR: [I– I thought it] was- 17 SOP: no. 18 MAR: [@@@@@ ] 19 SOP: [it was a special], 20 it was like , 21 MAR: [@(h) ] 22 SOP: [a one] hou-- 23 one-hour special , 24 something like that.

Note how Sophie’s attempts to characterize the TV special, as well as Mary’s attempts to grasp what kind of TV show it is, are done almost entirely with PNCs or PNC fragments doing informing (Sophie) or seeking confirmation of an informing (Mary).

So in lines 10–17, Sophie and Mary use negative PNCs to collaborate in establishing what the TV show is not: it’s not like a regular or weekly program: In line 12, Mary proposes a completion to Sophie’s line 10, it’s not like, by offering it’s not like a regular program. In line 16, Mary identifies a discrepancy in their respective second- and first-hand knowledge with a truncated clause projecting another PNC I thought it was – (cf. Smith, 2013).

Finally, in line 19, Sophie’s PNC continues the characterization by categorizing the program it was a special, and in lines 20, 22–24, she narrows that categorization with the PNC it was like a one-hour—a one-hour special, something like that, closing the sequence.

In this section, then, we have sketched the way in which turn design is implicated in one of the interactional activities in which PNCs are particularly prominent, that of collaboratively constructing a characterization of an entity or situation.

Conclusions

So what have our data revealed about Predicate Nominal Constructions? Let us examine this question in terms of our research questions. Our first research question was:

(1) What do Predicate Nominal Constructions look like in conversation? Are there distributional skewings favoring certain types of PNC formats?

To address this question, we provided evidence that the PNCs that English speakers recurrently use do have particular structural characteristics. First, they are constructions in the sense of Bybee (2010), constructions typically having some open slots and some fixed slots.

As illustrated in (5), repeated below as (30), the English PNC fits this understanding; it has:

- a highly restricted subject slot

- a fixed copula slot

- a semi-open Predicate Nominal

Second, many PNCs are formulaic expressions: it was a waste of time, it’s no big deal, that’s the problem, that’s the trouble, etc.

Third, 79% of the PNCs in our collection have non-human subjects, and of these, 86% are it/that/this. As we noted in section ‘What form do PNCs take in everyday conversation?’, the profile of PNC subjects is just the opposite of what’s been shown for subjects in general in conversation, where roughly 70% of all subjects in conversation are human. We suggest that while speakers in everyday conversations indeed tend to talk about themselves and other humans, a strong trend is to use a PNC to categorize situations rather than themselves and other humans.

Our second research question was:

(2) What social work do PNCs do in conversation?

We found that English speakers use PNCs to do two cross-cutting kinds of work, one more semantic, the other more interactional. Semantically, PNCs serve as a resource for speakers to either identify or categorize an entity or situation. Interactionally, PNCs are used to inform about and/or assess an entity or a situation.

Finally, we illustrated the way speakers exploit PNCs in longer sequences where participants are working together to build a characterization of a person, event, or situation.

We close this investigation of PNCs in English with a further, crucial question: how typical are our findings cross-linguistically? Are PNCs as readily identified in interactional data from other languages as they are in English? What about languages in which neither ‘clause’ nor ‘subject’ are apparent, or languages whose arguments are mostly unexpressed, or ‘allusive’ (Djenar et al., 2018: 112–130; Ewing, 2024)? Do PNCs do similar kinds of work in other languages as in English? To answer these questions, we support and advocate more research on PNCs and low transitivity in general in the conversational interactions of many more languages.

Footnotes

Appendix

Acknowledgements

We are grateful to Mira Ariel, Michael Ewing, and Ryoko Suzuki for valuable comments on this article; we alone, however, are responsible for the way we have incorporated their input.

ORCID iDs

Sandra A. Thompson

Tsuyoshi Ono

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Notes

Author biographies

Sandra A. Thompson is Research Professor Emerita of Linguistics at the University of California, Santa Barbara. Her work focuses on Interactional Linguistics, especially on how recurrent patterns in everyday embodied conversational interactions shapes patterns of grammar. She has published nearly two hundred scientific articles, drawing on interactional data from English, Mandarin, Japanese, and Dutch. She is the co-author with Charles Li of Mandarin Chinese: A Functional Reference Grammar. She has co-edited Interaction and Grammar with Elinor Ochs and Emanuel Schegloff, and published Grammar in Everyday Talk: Building Responsive Actions with Elizabeth Couper-Kuhlen and Barbara Fox.

Tsuyoshi Ono is Professor of Japanese Linguistics at the University of Alberta in Canada, where he directs the Spoken Discourse Research Studio. In collaboration with many colleagues, he has published extensively on the structure and usage of language in everyday speech, adopting a cross-linguistic and emergent perspective. Currently, he is involved in two large-scale collaborative projects, where he examines clause chaining in everyday spoken Japanese and explores ways to maintain the Ikema dialect of Miyako, an endangered language spoken on remote Japanese islands near Taiwan.

References

Beekhuizen

Thompson

(2022) ‘Something that’s very American’: The interactional role of Light-Head Relative Clauses. Discourse Studies 24(2): 149–167.

Bybee

(2010) Language, Usage and Cognition. Cambridge: Cambridge University Press.

Corrigan

Moravcsik

Ouali

, et al. (eds) (2009) Formulaic Language. John Benjamins Publishing Company.

Couper-Kuhlen

(2014) What does grammar tell us about action? Pragmatics 24(3): 623–647.

Couper-Kuhlen

Selting

(2018) Interactional Linguistics: Studying Language in Social Interaction. Cambridge: Cambridge University Press.

Deppermann

Haugh

(2022) Action ascription in interaction. In: Deppermann

Haugh

(eds) Action Ascription in Interaction. Cambridge University Press, pp.3–28.

Djenar

Ewing

Manns

(2018) Style and Intersubjectivity in Youth Interaction. Mouton De Gruyter.

Drew

(1992) Contested evidence in courtroom cross-examination: The case of a trial for rape. In: Drew

Heritage

(eds) Talk at Work. Cambridge University Press, pp.470–520.

Drew

(2012) Turn design. In: Sidnell

Stivers

(eds) The Handbook of Conversation Analysis. Wiley, pp.131–149.

10.

Du Bois

Kumpf

Ashby

(eds) (2003) Preferred Argument Structure: Grammar as Architecture for Function. John Benjamins Publishing Company.

11.

Edwards

(1991) Categories Are for Talking: On the Cognitive and Discursive Bases of Categorization. Theory & Psychology 1(4): 515–542.

12.

Erman

Warren

(2000) The idiom principle and the open choice principle. Text 20(1): 29–62.

13.

Ewing

(2024) Elusive referentiality and allusive reference in Indonesian conversation. In: Ewing

Laury

(eds) (Non)Referentiality in Conversation. John Benjamins Publishing Company, pp.11–34.

14.

Fox

Thompson

(2007) Relative clauses in English conversation: Relativizers, frequency and the notion of construction. Studies in Language 31(2): 293–326.

15.

French

Simpson

Harvey

(2006) Downplaying leadership: Researching how leaders talk about themselves. Leadership 2(4): 469–479.

16.

Fuller

(2003) Use of the discourse marker like in interviews. Journal of Sociolinguistics 7(3): 365–377.

17.

Genetti

(2018) How Languages Work: An Introduction to Language and Linguistics. Cambridge University Press.

18.

Givón

(1979) On Understanding Grammar. Academic Press.

19.

Golato

(2010) Marking understanding versus receipting information in talk: Achso and ach in German interaction. Discourse Studies 12(2): 147–176.

20.

Goodwin

(1979) The interactive construction of a sentence in natural conversation. In: Psathas

(ed.) Everyday Language: Studies in Ethnomethodology. Irvington, pp.97–121.

21.

Goodwin

(1981) Conversational Organization: Interaction between Speakers and Hearers. Academic Press. Available at: http://www.sscnet.ucla.edu/clic/cgoodwin/publish.htm

22.

Goodwin

(1986) Gesture and coparticipation in the activity of searching for a word. Semiotica 62(1/2): 51–75.

23.

Goodwin

(1987) Concurrent operations on talk: notes on the interactive organization of assessments. IPRA Papers in Pragmatics 1(1): 1–54.

24.

Haspelmath

(2024) Nonverbal clause constructions. Language and Linguistics Compass 19(2): e70007.

25.

Hayashi

(2003) Language and the body as resources for collaborative action: A study of word searches in Japanese conversation. Research on Language and Social Interaction 36(2): 109–141.

26.

Hengeveld

(1992) Non-Verbal Predication: Theory, Typology, Diachrony. Mouton.

27.

Holmes

EJB

Toerien

Jackson

(2017) The interactional bind of “Just [Do X]”. Research on Language and Social Interaction 50(4): 419–434.

28.

Hopper

(2004) The openness of grammatical constructions. Chicago Linguistic Society 40: 239–256.

29.

Hopper

Thompson

(1980) Transitivity in grammar and discourse. Language 56: 251–299.

30.

Keevallik

(2008) Clause combining and sequenced actions: The Estonian complementizer and evidential particle et. In: Laury

(ed.) Crosslinguistic Studies of Clause Combining: The Multifunctionality of Conjunctions. John Benjamins Publishing Company, pp.125–152.

31.

Koivisto

Laury

Seppanen

E-L

(2011) Syntactic and actional characteristics of Finnish ‘että’-clauses. In: Laury

Suzuki

(eds) Subordination in Conversation: A Cross-linguistic Perspective. John Benjamins Publishing Company, pp.69–102.

32.

Lee

(1987) The semantics of just. Journal of Pragmatics 11(3): 377–398.

33.

Lee

(1991) Categories in the description of just. Lingua 83(1): 43–66.

34.

Levinson

(2013) Action formation and ascription. In: Sidnell

Stivers

(eds) Handbook of Conversation Analysis. Wiley-Blackwell, pp.103–130.

35.

Mayes

(2024) Low transitive constructions as typical clauses in English: A case study of the functions of clauses with the nonverbal predicate be in stance displays. Languages 9(12): 372.

36.

Mayes

Tao

(2021) Referring expressions in categorizing activities: Rethinking the nature of linguistic units for the study of interaction. In: Ono

Laury

Suzuki

(eds) Usage-based and Typological Approaches to Linguistic Units. John Benjamins Publishing Company, pp.87–121.

37.

Mazeland

(2019) Activities as discrete organizational domains. In: Reber

Gerhardt

(eds) Embodied Activities in Face-to-Face and Mediated Settings. Palgrave Macmillan, pp.29–61.

38.

Miller

Weinert

(1995) The functions of ‘like’ in dialogue. Journal of Pragmatics 23: 365–393.

39.

Mori

(2006) The workings of the Japanese token hee in informing sequences: An analysis of sequential context, turn shape, and prosody. Journal of Pragmatics 38: 1175–1205.

40.

Ochs

Schieffelin

(1983) Foregrounding referents: A reconsideration of left dislocation in discourse. In: Ochs

Schieffelin

(eds) Acquisition of Conversational Competence. Routledge, pp.158–174.

41.

Ono

Laury

Suzuki

(eds) (2021) Usage-Based and Typological Approaches to Linguistic Units. John Benjamins Publishing Company.

42.

Ono

Thompson

(2003) Japanese (w)atashi/ore/boku: They’re not just pronouns. Cognitive Linguistics 14(4): 321–347.

43.

Payne

(1997) Describing Morphosyntax: A Guide for Field Linguists. Cambridge University Press.

44.

Payne

(2006) Exploring Language Structure: A Student’s Guide. Cambridge University Press.

45.

Pekarek Doehler

De Stefani

Horlacher

A-S

(2015) Time and Emergence in Grammar: Dislocation, Topicalization and Hanging Topic in French Talk-in-Interaction. John Benjamins Publishing Company.

46.

Pomerantz

Fehr

(1997) Conversation analysis: An approach to the analysis of social interaction. Discourse Studies: A Multidisciplinary Introduction 2: 165–190.

47.

Rymes

(1995) The construction of moral agency in the narratives of high-school drop-outs. Discourse & Society 6(4): 495–516.

48.

Robinson

(2012) Overall structural organization. In: Stivers

Sidnell

(eds) The Handbook of Conversation Analysis. Wiley-Blackwell, pp.257–280.

49.

Sacks

(1992[1971]) April 12: Long sequences. In: Jefferson

(ed.) Lectures on Conversation, vol. 2. Blackwell, pp.354–359.

50.

Schegloff

(2007) Sequence Organization. Cambridge: Cambridge University Press.

51.

Scheibman

(2002) Point of View and Grammar: Structural Patterns of Subjectivity in American English Conversation. John Benjamins Publishing Company.

52.

Seppänen

E-L

Laury

(2007) Complement clauses as turn continuations: The Finnish /että/-clause. Pragmatics 17(4): 553–572.

53.

Smith

(2013) ‘I thought’-initiated turns: Addressing discrepancies in first-hand and second-hand knowledge. Journal of Pragmatics 57: 318–330.

54.

Thompson

Couper-Kuhlen

Fox

(2015) Grammar and Everyday Talk: Building Responsive Actions. Cambridge University Press.

55.

Thompson

Hopper

(2001) Transitivity, clause structure, and argument structure: Evidence from conversation. In: Bybee

Hopper

(eds) Frequency and the Emergence of Linguistic Structure. John Benjamins Publishing Company, pp.27–60.

56.

Verhaar

(1968) The Verb ‘Be’ and Its Synonyms: Philosophical and Grammatical Studies. Springer.

57.

Wray

(2002) Formulaic Language and the Lexicon. Cambridge University Press.

58.

Wray

Perkins

(2000) The functions of formulaic language: An integrated model. Language and Communication 20: 1–28.

What do speakers do with predicate nominal constructions in English conversation?

Abstract

Keywords

Introduction

Data

Our research questions

What form do PNCs take in everyday conversation?

The work that speakers use PNCs to do

What semantic work do PNCs do? Identifying and categorizing

Identifying

Categorizing

Summary

What interactional work do speakers use PNCs to do? Informing and assessing

Informing

Assessing

Summary

Semantic and interactional work done together

Summary

Characterization sequences featuring PNCs

Conclusions

Footnotes

Appendix

Acknowledgements

ORCID iDs

Funding

Declaration of conflicting interests

Notes

Author biographies

References