Formulaicity and pragmatics of contact-terminating means in cinematic vs. artistic discourse: A corpus study

Full Text

1. Introduction

Over the past three decades, there has been increasing interest in the cognitive-semantic and functional-pragmatic specifics of communication in cinematic discourse, which has given great impetus to the formation of new interdisciplinary areas of linguistics (e.g. Gibbs 2020, Janney 2012, Pavesi & Formentelli 2023, Zykova 2023). One of these areas is represented by the researches aimed at identifying and describing pragmatically relevant language means in contemporary film(ic) speech (or film dialogue, film conversation, television dialogue, television conversation). For instance, Quaglio (2009) explores such means as first- and second person pronouns, hedges, discourse markers (you know, I mean), stance markers (probably, perhaps), copular verbs (seem, feel), modal verbs (could, might), empathic do, lexical bundles (I can’t believe [+ complements]), intensifiers (so, totally), expletives and slang terms, non-minimal responses (sure, fine) and some others (Quaglio 2009). All these units were investigated in the popular American sit-com “Friends” with the aim to compare their distribution and use in the film dialogue and in natural conversation. Quaglio comes to the conclusion that “Friends shares the core linguistic features that characterize <…> face-to-face conversation” although this does not mean that “scripted language of Friends is the same as natural conversation” (Quaglio 2009: 148).

Among pragmatically relevant language units distinguishing conversation in cinematic discourse, formulaic language means (FLM) are of particular interest (Nice to meet you!, You’re never going to believe this, in a nutshell etc.). Nowadays, it is possible to identify at least two main strands of exploring FLM in films.

One of them (most widely-spread) centers on the similarities and differences that FLM have in films and in real-life conversations. For instance, investigating the formulaicity of contemporary film speech, Freddi (2011) carried out a frequency-based analysis and identified the most frequent four-word formulaic clusters used in two American and two British films: What are you doing, What do you mean, I want you to, and some others. To find out how typical they are of spontaneous conversation, the scholar compared these clusters to general spoken corpora, namely the spoken components of the BNC and the COCA. The research findings have shown that most of the clusters identified in scripted film dialogue are common to natural spoken speech (Freddi 2011). This holds true for other pragmatically relevant units. According to Napoli and Tantucci (2022), despite featuring its own norms, film conversation has been claimed in Pragmatics’ research to be a good reflection of naturally occurring speech. Their study of requestive acts (I wonder whether you can …, Can you give me…) in English and Italian films testifies to the fact that they reproduce traits of non-fictional interaction as close as possible. The scholars emphasize that “film speech may be said to involve a hyper-representation of naturalistic interaction, in which pragmatic behaviour is somewhat ‘amplified’ rather than inhibited” (Napoli & Tantucci 2022).

The other main strand concerns the study of the process of penetration of FLM from artistic discourses (cinematic discourse, in particular) into everyday language practice. This approach is based on the assumption that normal conversation exploits language means and strategies elaborated in literature and other kinds of art. In her work, Tannen shows that “ordinary conversation is made up of linguistic strategies that have been thought quintessentially literary” and that the “strategies, which are shaped and elaborated in literary discourse, are pervasive, spontaneous, and functional in ordinary conversation” (Tannen 2007: 1). Analyzing formulaic expressions in a screenplay, “Some Like It Hot”, Van Lancker-Sidtis and Rallon (2004) deduce that “comparisons between constructed and spontaneous conversational talk reveal interesting similarities and differences in presence of formulaic expressions” (Van Lancker-Sidtis & Rallon 2004: 220). The scholars point out that “strategies of repetition are noted in both normal conversation and literature”; “a major source of this practice of repetition is FEs” (Ibid). Their findings are indicative of the fact that “living conversation naturally contains devices seen in the language arts as much as art imitates life” (Ibid).

Thus, exploring various (phonological, semantic, grammatical, functional, pragmatic, creative, etc.) aspects of FLM in films as well as in other types of artistic discourses (novels, poetry, drama, short stories, theatrical performances, etc.) in comparison with natural (casual, everyday, real-life) conversation is a rather steady tendency of their recent investigations. In the majority of cases, researches are based on extensive corpus evidence. Corpus-linguistic studies, as Buerki (2020) notes, focus primarily on conventionality as manifested in language use, considering formulas as “expressions that represent habitual ways of putting things in a community” (Buerki 2020: 106).

 The research undertaken in this paper follows in its key aspects the recent tendencies of exploring FLM in modern linguistics and in its interdisciplinary directions. It addresses formulaicity of cinematic communication in general and more specifically of conversations unfolding in contemporary feature films. The formulaicity will be measured through the analysis of a particular category of language units — multi-word units that are used by interlocutors to terminate communication in films. Hence, the goal of the present paper is to establish the contact-terminating means peculiar to interpersonal interaction in cinematic discourse, estimate them from the point of view of conventionality and pragmatic value. To pursue this goal, the corpus approach is applied.

2. Contact-terminating means

Contact-terminating means (CTM), as understood in this paper, are a sub-category of metacommunicative units (Grigorieva 2006). The increasing interest in their exploration is caused by cultural and pragmatic relevance of CTM in everyday (formal and informal) communication, as was noted in many works.

Pillet-Shore attests to the immense importance of conversational routines, including greeting another person, introducing oneself to someone new, and saying goodbye (Pillet-Shore 2024). As Baranov and Kreidlin (1992) claim, it is impossible to fully describe the axiological and, more broadly, the modal structure of a dialogue without studying the lexical markers of its opening and ending. The latter include different formulaic expressions, such as Blagodaryu za vnimanie!, Nu ladno tebe!, Mozhet, khvatit. According to Tuncer, “closing an interaction is a crucial moment and takes delicate work, because the last words are known to remain effective during the anticipated time of separation and to settle a relationship. One problematic achievement is to open the closing sequence” (Tuncer 2015: 105). As Paltridge (2022) points out, closings are “complex interactional units which are sensitive to the speaker’s orientation to continuing, closing (or not wanting to close) the conversation” (Paltridge 2022: 136). McKeown and Zhang emphasize that “closings present an opportunity to produce a frictionless termination to a communicative exchange” (McKeown & Zhang 2015: 93).

In most studies, CTM are described as a subcategory that embraces structurally and semantically diverse multifunctional formulaic units and overlaps with other subcategories of metacommunicative units as well as other language units (phraseological units, phrasal discourse markers, speech acts, etc.). Gorodnikova and Dobrovolskij (1998) refer the formulaic expressions that (may) provide ending or breaking a conversation to two subcategories of speech acts: phatic (goodbyes and greetings-at parting) and metacommunicative (evasions). Stribizhev (2005) dwells on the following types of formulas that are used to signal an interlocutor’s intention to end up a conversation: formulas of leave-taking (Good bye!, So long!); formulas of thanking (Many thanks, What are friends for?); formulas of farewell wishes (Have a nice day!, Take care!); formulas of non-replies or evasions (I’ve got nothing to add); closing formulas (I’ve finished), and interruptions of non-formulaic character)¹. Pillet-Shore establishes idiomatic or formulaic expressions that correspond to particular modular components to initiate and constitute the closing phase of interaction: “possible pre-closing” (Things always work out for the best), “announced closing” (Let me get off), “appreciating/reinvoking the reason for the encounter” (Thanks for coming/having us over), “bridging time/arrangement-making” (See you later/soon), “well-wishing” (Have a safe trip), “expressing affection and/or reluctance to separate” (I love you, I’ll miss you), “doing goodbye” (Good night, See ya). The scholar pays special attention to the fact that do-goodbye formulas can be repeated several times by interlocutors during their parting. Besides, closing components are non-linear, and there are multimodal features that can pervade the (pre-)closing phase, e.g.: gathering belongings, deploying stance-marking embodiments that display current personal states (doing “being tired” by yawning and/or producing audible out-breaths/sighs) (Pillet-Shore 2024). Many other aspects of CTM have been also discussed in contemporary linguistic and interdisciplinary researches (e.g., Bladas 2012, Bolden 2017, House & Kádár 2024).

Although much work has been done in the field in question, the use of CTM in artistic communication and more specifically in cinematic discourse still remains under-investigated. This study aims to identify formulaicity of CTM in cinematic discourse and establish its pragmatic specificity in comparison with that in other types of artistic discourse.

In this paper, CTM is defined as a subcategory that brings together multi-word units of different emotive-expressive charge that are exploited to put an end to or interrupt communication. They can be characterized by a varied degree of formulaicity determined by a number of parameters: lexical fixedness, idiomaticity, syntactical variation, relatedness to a particular communicative situation.

3. Data and methodology

This research is of interdisciplinary character. It applies knowledge from a number of disciplines: phraseology, pragmalinguistics, film studies, corpus linguistics, and, in particular, findings from discourse and conversation analysis. The data for this study come from two corpora.

The first corpus is compiled ad hoc. It includes four British feature films released since the beginning of the 21^st century and portraying peculiarities of natural interaction in contemporary settings. The size of this dataset is approximately 41 375 words (see Table 1).

Table 1. The data of the films compiling the first corpus

 All the selected films were manually transcribed from the film soundtrack, marking up the initial and final phases of interpersonal interaction. For ease of reading and computer search, the corpus contains only orthographic transcriptions. All identified CTM were annotated by means of the symbol <ctm>. The dialogue transcriptions were also enriched with the information related to several parameters concerning textual (or contextual) and individual variables: chapter (number and/or title) and scene type/settings, character speaking (including the cases of voice over), linguistic event (e.g., in-person talks, phone calls), an interlocutor’s emotional state (e.g., irritated, bewildered, whispering) and salient non-linguistic behaviour (e.g., waving, nodding). Besides, the corpus houses metadata, such as: genre, year of production, country of production, director, screenwriter, literary source, running time, storyline, awards and nominations, age certificates (e.g. “U”, “PG”, “12A”, etc), popularity rating according to the Internet Movie Database (IMDb). All the films composing the corpus in question are produced in the UK, are of the same genre (drama films), have awards, are popular, and have the rating above 6.0.

The second database applied in the present work is the Written British National Corpus 2014 (the Written BNC2014). It is conceived as a further test case for the findings based on the first corpus. The Written BNC2014 is a major project led by Lancaster University to create a 100-million-word corpus of present day British English. The choice of this corpus is stipulated by two main reasons. First, the data it contains was collected in the time window of 2010–2019, with 2014 being roughly the midpoint. Second, it has two subcorpora that makes it possible to fulfil the research tasks set in the given paper. They are “Fiction” and “Written-to-be-Spoken”. The size of the former is about 20 million words, while the latter contains more than 3 million words. The “Fiction” subcorpus embraces such subgenres as i) poetry; ii) general prose; iii) prose for children and teenagers; iv) science fiction and fantasy; v) crime; vi) romance. The proportion of fiction texts represents their “influential cultural role” (Burnard 2000: 7). The “Written-to-be-Spoken” involves i) television (TV) scripts; ii) modern drama scripts (Brezina et al. 2021). The two mentioned characteristics of the Written BNC2014 are crucial for ensuring the validity of the comparison of the data retrieved from both corpora; they may provide reliable evidence to determine the specificity of CMT in cinematic discourse vs. artistic discourse.

The two corpora will be further referred to as the CF (i.e. the corpus of films) and the BNC2014-F/WBS. The approach taken in this study involves two interrelated case-studies (based on the work with two corpora described above) and six basic steps.

Case study-1. In a first step, a film genre, topic, and size matched corpus of English films is compiled (the CF). In the second step, dialogical profiles of the films are established and compared in terms of their similarity and differences. This comparison allows assessing the balance, validity, and representativeness of the analyzed empirical material. In the third step, the identification and comprehensive manual extractions of CTM from the CF are carried out; all identified CTM are classified according to their pragmatic types; the established pragmatic types of CTM are counted and percentages for each type are determined. In the fourth step, the formulaicity of CTM of each pragmatic type is measured through a number of parameters: syntactic and lexical fixedness, idiomaticity, and distribution across the feature films in the CF.

Case study-2. In the fifth step, the formulaic CTM that prove to be most characteristic of cinematic discourse, are tested in two subcorpora of the Written BNC2014: “Fiction” and “Written-to-be-Spoken” (the BNC2014-F/WBS), assessing their frequency, variability, and distribution. In the sixth and final step, the research findings from two corpora (the CF and the BNC2014-F/WS) are compared and estimated in terms of whether formulaic CTM identified in the feature films are typical of other types of artistic discourse and have similar functions in them.

4. Research results

4.1. Case study-1. CTM in cinematic discourse

Case study-1 is corpus-driven. The dialogical profiles of the drama films in the CF have been created according to several parameters. One of them is the number of dialogues that constitute the verbal structure of the films in question. Another important parameter that may influence the contact-termination phase is whether a dialogue takes place in person or over the telephone. In the latter case, the participants are not co-present on the screen and the remarks of only one of the participants are accessible for the analysis. The number of interlocutors involved in the interaction is also of high relevance (if there are two or more than two participants). This parameter influences the frequency of CTM as they are usually used by each participant of a dialogue. And last but not least is the parameter of the interlocutor who terminates the contact: if it is the participant who initiates the dialogue or it is his/her partner(s). All the data obtained in the course of the analysis are summarized in table 2 (see table 2).

Table 2. Dialogical profiles of the drama films in contrast 

Note: *Person dialogues include cases of inner dialogues and talks with pets; **All verbal and non-verbal reactions are regarded as indicators of contact-termination (words, silence, various hand, head, eye, and other body gestures).
Resource: The author’s research output.

The formation and comparison of the dialogical profiles allow assessing the validity and objectivity of the data about the contact-termination phase that are extracted from the CF. As is seen in table 2, the total number of dialogues used in the films is 205, which can be considered a rather representative sample for the analysis of CTM. The dialogical profiles of the analyzed films have both differences and similarities. One of the differences is their “dialogical density” determined by the number of dialogues used in them. Interestingly, although the film “Driving Lessons” has the shortest duration compared to the other three films and in terms of its size (i.e. the number of words) it ranks only third, this film has the highest dialogical density (61 dialogues). It implies a broader range of situational contexts of using CTM. In contrast to the other three films, in “Another Year” the communication is only in-person, telephone calls are not characteristic of its dialogical profile. Besides, the number of dialogues that involve several participants is greater than the number of two-participant dialogues in this film. In the other films, two-participant dialogues predominate. The feature that all the dialogical profiles have in common is that in the majority of cases the character who opens the conversation (i.e. initiator) does not close it.

The study of the dialogical profiles leads to the following conclusion. Their differences provide us with the possibility of taking into account most of the structural, topical, and deictic (participants, locations/settings, and time) diversity of dialogues (in particular, their CTM) that are peculiar to natural communication. Their similarities are indicative of the specifics of the interpersonal interactions depicted in the films as a certain type of artistic communication.

In 205 dialogues, there were established 318 occurrences of CTM (repeated units were included), out of which 292 cases are verbal CTM and 26 cases are non-verbal CTM, cf. e.g.:

(1)   Laura: Then I’ll expect you whenever you can get back. Good night, Ben. (“Driving Lessons”)

(2)   Claire: Do you realize how mad you sound? …To people like me. …To normal people. You sound mad.

Joe: [silent, his gaze and face expressions signal the end of the conversation]. (“Enduring Love”)

Due to the aim and tasks of the present paper, non-verbal CTM and single words exploited to end conversational interactions in the films under consideration were excluded from further analysis. The quantity of single-word CTM is 25 (e.g., Oh!, Yeah, Mum, Another). Their removing from the material yields 267 occurrences of multi-word CTM, which were analyzed according to the following parameters: 1) a level of functioning (intradiegetic and extradiegetic); 2) pragmatic value, and 3) formulaicity.

The research carried out has shown that CTM can function at both intradiegetic and extradiegetic levels. The level of functioning is determined by a set of factors: whether they are addressed to a film character or a film viewer; whether their use intends to terminate a topic of a film dialogue or a film scene (a film episode); whether they are uttered on screen or in a voice over. The CTM, which are addressed to the film character, used to end one of the topics of a film dialogue or a whole dialogue, and pronounced on screen, are intradiegetic. It means that they intend to imitate the naturalness or spontaneity of everyday interpersonal interactions (example 3). The CTM, which are addressed to the film viewer, exploited mainly to shift to another film episode, and/or pronounced by a voice over, are extradiegetic. It implies that they serve primarily to develop a film narration and contribute to the conflict representation (example 4).

(3)   Mary (says on-screen to Ken): Look at the food in this fridge. I haven’t got anything in mine. I’ll see you later, all right? (“Another Year”)

(4)   Jamie (voice-over): When I was a child, we would go to this cove under the cliffs called Hope Gap. <…> My mother would sit on the rocks and wait for me while I explored. I never asked myself what she was thinking or if she was happy. You don’t, do you? (“Hope Gap”)

The overwhelming majority of the CTM under study are intradiegetic units that make the filmic dialogues akin to natural communication. However, they also provide the plot advancement and, consequently, function at an extradiegetic level as well. Just a few CTM of an exclusively extradiegetic nature were found in the CF (e.g. Edward’s voice over narrating the events of the War of 1812 in the film “Hope Gap”). As a whole, the use of CTM at both levels of functioning (intra- and extradiegetic) can be regarded as an important feature of this subcategory of metacommunicative units in films that distinguishes the latter from natural communication and other types of artistic discourse.

The CTM identified in the CF were classified into the following twelve pragmatic types: 1) ‘farewells or partings’; 2) ‘personal opinion or stance-taking’; 3) ‘common knowledge or a widely-spread opinion’; 4) ‘informing about something’ (some event, the weather, immediate-future actions, planned actions, one’s attitude and feelings, among other); 5) ‘information request or information verification’; 6) ‘agreement or approval’; 7) ‘promise or warning’; 8) ‘offer or invitation’ (to do something); 9) ‘request or command’; 10) ‘apology’; 11) ‘expletive or insult’; 12) ‘thanking’. The percentage ratio of these pragmatic types of CTM are presented in Figure 1.

Figure 1. The percentage ratio of 12 pragmatic types of CTM extracted from the CF
Resource: The author’s research output.

As was established, 45% of all the CTM are formulas that are traditionally and regularly used in real-life situations when interlocutors are taking leave. These formulas are conventional means of saying goodbye, expressing good wishes at parting, appreciating the encounter.

Among the CTM-farewells, the formulas with the most frequency of occurrence are Good(-)bye (35%) and Hope to see you soon (26%). The former one occurs in all the four films under consideration, which means that it has an even distribution across the corpus material. This formula is used predominantly in a contracted form: Bye, Bye-bye. The latter formula is not found in the film “Hope Gap”. In the three other films, Hope to see you soon has a (creatively) modified or a contracted form and intends to arrange next possible future contact. Its variable elements range from the general (later, soon, etc.) to more specific deictic (temporal, spatial) units (on Thursday, there, next week, etc), cf., e.g.:

(5)   Tom: Right, we’ll see you when we see you.

Katie: Soon, hopefully. (“Another Year”)

(6)   Jed: Right. Well, bye. Bye, then.

Joe: Okay. See you, bye. (“Enduring Love”)

To bring the conversational interaction to an end, the characters also exploit such formulaic phrases as: I’d best be off, We’d best get going, I’ve got to go, I might head off in a minute, It’s time to go, Let’s go. This formulaic group ranks third in frequency of occurrence (8%), e.g.:

(7)   Laura: Ben!

Ben (on the phone): I have to go now. (“Driving Lessons”)

The occurrence of the other CTM of this pragmatic type ranges from 6 times to 1 time. These are the following formulas: Good night, Take care, Ta-ta [for now], Lovely/nice to see/meet you, Thank you for coming or Thanks for coming, I’ll give you a ring, Look after yourself, Give/send my love to, Good luck, Safe journey, Keep in touch, e.g.:

(8)   Jamie: I should be getting... going. She’ll be waiting for me so...

Edward: Well, then keep in touch. (“Hope Gap”)

Also, in the CF four formulaic CTM are found to convey a rather conventional way of ending an encounter in a special communication situation, i.e. at a church service, the funeral, the office: Have mercy on us; These things are of no consequence, be you Christian or atheist, unless in your heart you are true; We’ll remember them; If you’d like to follow me.

The CTM-farewells are formulas characterized by lexical fixedness (to some degree, as they have variable slots in their structure) and/or idiomaticity (to some degree) [about these and other criteria for identifying formulaic sequences see in (Bladas 2012, Buerki 2020, Namba 2010, Wood 2015)].

The CTM of other pragmatic groups can be divided into non-formulaic and (to some extent) formulaic. The formulaicity may manifest itself in a number of ways: syntactically, lexically, and both syntactically and lexically.

As Figure 1 demonstrates, the CTM-statements informing about something (13%), the CTM-opinions (7%), and the CTM-requests/commands (7%), rank second and third in frequency of occurrence, correspondently, e.g.:

(9)     Jamie: I can’t bring him back.

Grace: You don’t know that. Have you tried? You go on seeing him. That means you let him think he has your approval.

Jamie: Mom, he’s in love! (“Hope Gap”)

(10)    Joe: Well, you know, we’ll cross that bridge when we come to it.

Claire: I think we have come to it.

Joe: Not now. (“Enduring Love”)

(11)    Evie: You do not touch my things!

Ben: Yes. I’m sorry. I’m very sorry.

Evie: You tidy around them. (“Driving Lessons”)

The statements that a character uses to close a conversation by informing his/her partner about something are qualified as more non-formulaic, cf., e.g.: It’s going to rain again; He’s all right; As well as can be expected; I’m in the park across the street; I love you; I need him here; Straight on, second on the right; You’re hurting me; I’m on my way. Their syntactical structure and lexical composition are diverse and not recurrent. Only some CTM of this pragmatic type have formulaic elements, e.g.: God only knows where I’d be without you!; So you see, I’m not really coping after all.

The formulaicity of the CTM-opinions can be observed in the recurrence of the utterances beginning with I think, e.g.: I think you’ll find that we men are; I think you probably have, old son. Even if this formulaic opinion-classifier is omitted, it is implied, e.g.: They have no feelings at all = [I think] They [men] have no feelings at all. Among CTM-requests (direct or indirect) there are a few phrases which are recognized as regularly used means of conversation termination: Would you stop saying that?; Don’t say anything; Mind your own business; Come on; Don’t even talk about it.

A certain degree of formulaicity is observed in other pragmatic groups of CTM. Dialogues in the films under consideration are closed by means of formulas of thanking. The formula Thank you is used 12 times, the formula Thank you/thanks for smth/doing smth — 3 times, and the formula Thanks a lot — one time (the total frequency rate is 6%), e.g:

(12)    Evie: Sadly, your prices are beyond us, but for your aid we thank you. (“Driving Lessons”)

These formulas have lexical fixedness and are considered rather conventional means of contact-termination, especially in informal situations.

A few interrogative utterances are found in the CF as CTM, e.g.: Was he?; Am I?; Mum, did she come looking for me? What do you mean?; Who’s gonna forgive me?. They are not idiomatic and differ in their lexical composition. From the syntactical point of view, they are different types of questions used in the forms that are quite typical of dialogical speech. All these features are indicative of their more non-formulaicity.

Of particular interest are the CTM that express common knowledge or a widely-spread (commonly-accepted) opinion. Some of them are utterances that are distinguished by an aphoristic character and/or idiomaticity. These traits make them akin to proverbs. Hence, the CTM of this pragmatic type can be considered proverbial prototypes and can be qualified as potential formulas, cf., e.g.:

(13)    Edward: He’s got his own life to live. (“Hope Gap”)

(14)    Evie: Life is confusing. Just when we think it’s all over, it throws a view like this at us and we don’t know where we are. (“Driving Lessons”)

The CTM-agreements/approvals are lexically and syntactically diverse utterances lacking idiomaticity: I know!; Oh, well; Here goes; That’s the spirit; Not gay, apparently; Grace is right. However, some of them contain formulaic elements: Bloody hell right; Whatever you want, darling; All right; Yes, I see.

The CTM-offers/invitations and the CTM-apologies have the same frequency of occurrence in the CF. Both types of CTM involve phrases of a formulaic character due to their recognizable syntactical and/or lexical peculiarities. To end a dialogue, the following formulas of offering or inviting to do something are exploited in the films: Well, you must come again; Think about it; Come and sit yourself down, Mary, and have a cup of tea; Come on and Let’s-phrases. The latter two formulas signal the transition to another action or activity of the character in the films. The CTM-apologies are rather conventional phrases formed with the help of two words: excuse (used when the dialogue is interrupted) and sorry (used when the character expresses sorrow, sympathy or regret for a misdeed or a mistake), e.g.:

(15)    Evie: I…I thought I’d begin today’s recital by reading William Shakespeare’s sonnet number twenty nine, “When in Disgrace with Fortune.” Sorry. Excuse me. Sorry. (“Driving Lessons”)

The cases of terminating a conversation by means of promises or warnings are very rare in the CF. The formulaic character of such CTM is determined by their syntactic patterns rather than lexical constituents, cf., e.g.: I promise; We’ll have a proper game next time; I’ll kill her. As was established, there are only five hits of expletives used in the films under consideration to interrupt or end an interpersonal communication.

Importantly, the CTM with the low frequency of occurrence (i.e. from 6 to 1%) are characterized by a varied distribution in the CF. Five of the eight pragmatic types do not have distribution in all the films constituting this corpus. These are the following types: 3) common knowledge/widely-spread opinion; 7) promises/warnings; 10) apologies; 11) expletives/insults; 12) thanking. The data of the CTM distribution in the CF is given in Figure 2.

Figure 2. The distribution of CTM of different pragmatic types in the CF
Resource: The author’s research output.

To sum up, Case study-1 resulted in establishing that the most frequent means of terminating a conversation in the cinematic discourse are farewells. This pragmatic group is characterized by an absolute degree of formulaicity, bringing together corresponding types of formulas. These formulas are characterized by lexical fixedness, idiomaticity, and a steady association with the final phase of interpersonal interaction, reflecting the conventions of formal (task-oriented) and informal (or casual) conversational speech in everyday life.

Their relatively high frequency of occurrence and even distribution across the films in the CF testify the development of regular associations between the contact-terminating phase and the following three pragmatic groups of language means: (i) statements informing about something, and utterances that express (ii) personal opinions and (iii) requests (orders). CTM-statements are more non-formulaic than CTM-opinions and CTM-requests that contain formulaic elements and therefore exhibit more formulaicity. A degree of formulaicity is also characteristic of other pragmatic types of CTM identified in the present research (apologies, invitations, etc.).

4.2. Case study-2. CTM in artistic discourse

Case sudy-2 is corpus-based. It was carried out on the basis of two subcorpora of the Written BNC2014 that contain written texts of different (sub)genres: “Fiction” and “Written-to-be-Spoken” (the BNC2014-F/WBS). Three formulas have been selected from the pragmatic group of CTM-farewells as this group has the highest frequency and distribution rates in the CF. These formulas are (1) See you soon/later/tomorrow, (2) Thank you/thanks for coming, and (3) Keep in touch/contact. To find and compare their frequency in the written artistic discourse, the KWIC tool was applied. Table 3 contains the results of their analysis in the BNC2014-F/WBS (see Table 3).

 Table 3. Rates of the three formulas in the BNC2014-F/WBS

 Note: *in this section, rates represent absolute and relative frequency of hits (hits per 1M tokens in the corpus).
Resource: Retrieved by the author from the Written BNC2014 July 20, 2025.

According to the corpus-based findings, the first formula with the variable element later has the highest rate of hits in both subcorpora (see Table 3). In the “Fiction” subcorpus, the absolute frequency of See you later is 132 with the relative frequency being 6,46. It appears in 107 texts out of 1069 texts in this subcorpus, with the highest rate of relative frequency in such a category of texts as “women’s” (15,05). In the “Written-to-be-Spoken” subcorpus, the absolute frequency of this formula is 116 with the relative frequency being 36,66. See you later occurs in 84 texts out of 726 texts in this subcorpus, having the highest rate of relative frequency in the category of “TV scripts” (64,62). See you tomorrow occurs with the highest rate of relative frequency in the subgenre categories “women’s” (8,78) and “TV scripts” (7,53) in the two subcorpora, correspondingly. Unlike these two variants of the formula under consideration, the highest rate of relative frequency of See you soon is observed in the subgenre categories “humour” (18,80) and “drama scripts” (8,28).

As the formula See you soon/later/tomorrow is supposed to be a contracted form, “probably short for Hope to see you soon” (OEtD 2025), the left-context concordances were also taken into account. According to the number of hits in both subcorpora, the top two of two-word patterns that proceed this formula are the first-person singular and plural pronouns (in L2 position) used in the majority of cases with the contracted form of the auxiliary verb will/shall (in L1 position) — I’ll (in a few cases I will), we’ll, e.g.:

(16)    Finally, at the point of departure, she said, ‘I’ll see you soon.’ (“Fiction”).

(17)    Right, I better get off. I’ll see you later. (“Written-to-be-Spoken”)

In the form I hope to see you [again] soon, the formula is found to occur two times only in the “Fiction” subcorpus, e.g.:

(18)    ‘Thank you, Miss Clark,’ he said, sitting back down before I’d even left the office. ‘I hope to see you again soon.’

These left-context concordances provide the formula with extra pragmatic value of making a promise and expressing uncertainty.

The analysis of 432 contexts from two subcorpora under study has revealed that the formula See you soon/later/tomorrow is exploited by the characters in all the cases with the intention to end (or interrupt) the interpersonal interaction, i.e. as CTM, e.g.:

(19)    ‘See you tomorrow. Give my regards to your mum,’ he said, clearly enough to be heard. (“Fiction”)

(20)    See you later. Are you going out? Yes, my friend Bella’s. (“Written-to-be-Spoken”)

In the “Fiction” subcorpus, the formula Thank you for coming is characterized by a higher rate of hits than its variant Thanks for coming, cf.: 46 (2,25) vs. 40 (1,96), correspondingly. The relative frequency of the former is highest in such a category of subgenre of fiction as “humor” (9,40), while the latter occurs most frequently in the subcategory “women’s” (9,75). Interestingly, according to the data retrieved from the subcorpus “Written-to-be-Spoken”, there is a very slight difference in frequency of occurrence between Thank you for coming and Thanks for coming, cf.: 27 (8,53) vs. 28 (8,85). Moreover, the rate of their distribution across the texts in this subcorpus is the same (see Table 3). Both formulas appear in the subgenre of “TV scrips” more frequently than in the subgenre “drama scripts”. The rates of relative frequency in these subgenres correlate as follows: Thank you for coming is 11,92 vs. 5,10; Thanks for coming is 13,18 vs. 4,46.

Importantly, only 26 out of 141 hits (18,4%) in these formulas are used as CTM, for example:

(21)    ‘See you, Granddad,’ said Jamie, leaning in. ‘What lovely grandchildren! Thank you for coming. You must have more important things to do.’ (“Fiction”)

(22)    Good luck with that. And thanks for coming over. It always helps to talk things through.’ She walked him to the door. (“Fiction”)

(23)    Good night, Sarah. Night. Thank you for coming. (“Written-to-be-Spoken”)

(24)    Thank you so much for coming. Thanks for coming. Dave will take you over. (“Written-to-be-Spoken”)

The analysis has shown that in the majority of cases — 88 hits (62,4%) — the formulas under consideration are used by the characters at the opening phase of their interaction to (greatly/much) appreciate encountering interlocutors and to establish an appropriate connection with them. Thus, these formulas perform a different metacommunicative function. They contribute to the conversation opening and serve as contact-establishing means, for example:

(25)    Seated at right angles to each other, they began the business in hand. ‘Thank you for coming to us,’ Thea said. ‘We do appreciate it.’ (“Fiction”)

(26)    They’re here. Thank you for coming. Please have a seat. (“Written-to-be-Spoken”)

In 27 cases (19,2%), the formulas in question are not pragmatically linked with either the end or the beginning of the conversation in the artistic discourse. They have more semantic value and are used to depict the speaker’s attitude to the partner’s conduct or the relationships between the characters, contribute to the conflict-developing or dramatizing the situation, e.g.:

(27) I was embarrassed, and cross that you’d left me stranded. But I’m OK now. Thanks for coming back for me.’ (“Fiction”)

(28) It’s done now, anyway. Thanks Jenny. Thanks for coming over. Thanks. (“Written-to-be-Spoken”)

As far as the third formula is concerned, the data retrieved from the BNC2014-F/WBS testify to the fact that it occurs much more frequent with the element touch in the artistic discourse than with the element contact, cf.: 59 hits in 51 texts vs. 6 hits in 5 texts (absolute frequency) (see Table 3). The highest rate of distribution of Keep in touch is in such subgenres of fiction and written-to-be-spoken texts as “women’s” (6,69) and “drama scripts (3,18), correspondingly. Keep in contact appears with the highest rate of frequency distribution (0,89) in the subgenre of “fantasy”.

The qualitative analysis of all the hits from the BNC2014-F/WBS has resulted in establishing 26 cases (40%) out of 65, in which the formulas under consideration are exploited by the speakers with the intention to terminate a conversation, i.e. as informal parting phrases, e.g.:

(29)    Okay, Ma, it’s late here, I’m getting my head down now.’ ‘Keep in touch, won’t you, CeCe?’ (“Fiction”)

(30)    — Oh! Look after yourself, Katy. — I will. Keep in touch, OK? (“Written-to-be-Spoken”)

Notably, as CTM the formula Keep in touch occurs not only in dialogues, but also in the narration to depict the traditional way of ending a conversation, e.g.:

(31)    They exchange numbers and make promises to keep in touch, and he offers to pick her up on his way back to London. (“Fiction”)

The study of left contexts (a window span of two words on the left, L2 position) reveals the following. The co-occurrence with the first-person singular and plural pronouns and the auxiliary verb will (I’ll, We’ll, I will), as well as with the verbs try, do (empathic), the word please, the form Let’s imparts the formulas in question extra pragmatic value of making a promise and expressing a request, an offer or a wish, e.g.:

(32)    ‘Thanks for ringing, Dan. I’ll keep in touch, mate.’ Then he dropped the phone and turned to her, his face a picture of disbelief. (“Fiction”)

(33)    ‘Let’s keep in touch, old fellow,’ said Hugo as he climbed into his car. (“Fiction”)

In 39 cases (60%), keep in touch/contact is used in the artistic discourse in its idiomatic meaning ‘communicate with someone regularly’, e.g.:

(34)    I do all right in sixth-form college, though Samantha Hogan doesn’t keep in touch like she said she would and I find that all I really do is sit in my pyjamas in Auntie Cheryl’s spare room and play on my iPhone. (“Written-to-be-Spoken”)

Thus, in various genres of texts in two subcorpora of the Written BNC2014 — “Fiction” and “Written-to-be-Spoken”, the formulas — (1) See you soon/later/tomorrow, (2) Thank you/thanks for coming, and (3) Keep in touch/contact — are exploited as CTM. Unlike the first formula, whose pragmatic value is most salient and sustainable, the second and third formulas can fulfill other (meta)communicative functions (e.g., opening a conversation) and additional pragmatic tasks (such as promises, offers, etc), and can be used as semantic units rather than pragmatic ones.

The comparison of all the quantitative and qualitative data from the two corpora — the CF and the BNC2014-F/WS — identifies both similarities and differences in the use of the three CTM-formulas. A most important similarity consists in the same quantitative (or statistical) trend of their use in cinematic discourse and in different types of artistic discourse. In both corpora, the formula See you soon/later/tomorrow ranks first in frequency of occurrence and distribution, the formula Thank you/thanks for coming ranks second, and the formula Keep in touch/contact — third. Besides, like in fiction and written-to-be-spoken texts, in the feature films under analysis the use of See you soon/later/tomorrow can imply a kind of promise (especially when it co-occurs with I’ll, we’ll), and the formula Thank you for coming is characterized by a higher frequency of occurrence than Thanks for coming.

As far as differences are concerned, the first formula with the variable element later (i.e. See you later) has the highest rate of hits in frequency and distribution in the BNC2014-F/WS, while in the CF the number of its hits is smallest in contrast to the use with the variable elements soon and tomorrow. In the studied films, this formula exhibits a significant variability of the third element. The components soon, later, tomorrow can be replaced by other deictic words denoting time or space (on Thursday, next week, next Sunday, eleven o’clock, there). In the CF, the formula Thank you/thanks for coming functions only as CTM, the cases with its performing other (meta)communicative functions or pragmatic tasks that were detected in the BNC2014-F/WS were not found. Keep in touch has one hit in the CF which means that as CTM it is not typical of cinematic discourse. However, the data from the BNC2014-F/WS (concerning its frequency, distribution, structural variability) makes it possible to qualify Keep in touch/contact as quite common (conventional, usual) CTM for the artistic discourse.

5. Discussion

The results of this study fall into three general areas of significance. First, the comparison of the data from two corpora has shown that artistic discourse (including cinema) mirrors the communicative and pragmatic peculiarities of spontaneous talk in its formulaicity aspect with regard to CTM. The highest rates of occurrence and distribution of such pragmatic type of CTM as CTM-farewells in the CF and the data of the occurrence and distribution of three formulas of farewells retrieved from the BNC2014-F/WS verify the assumption that the phase of terminating a conversation in films and other types of artistic discourse is characterized by formulaicity that makes filmic (in particular) and artistic (in general) speech is quite similar with natural speech. The dialogical profiles of the films under consideration also testify to the “pragmatic authenticity” of closing an interpersonal interaction in cinematic discourse. They reveal that CTM are used in the filmic dialogues that are devoted to the discussion of a substantial range of topics with the number of participants varying from two to more-than-two. These findings agree with the results from earlier works comparing the functioning of formulas in films and in everyday speech and exploring pragmatic aspects of artistic discourse (Grant & Starks 2001, Lancker-Sidtis & Rallon 2004, Соколова & Фещенко 2024). Thus, formulaicity can be a reliable indicator of some general principles of artistic representation of real-life (everyday) interpersonal interaction in different types of artistic discourse.

Second, the results also reveal that cinematic discourse does not actually follow the generally accepted patterns of natural communication as it modifies them, adjusting their choice and use to the achievement of certain aesthetic goals of this art form. In accord with our findings in the CF, Taylor reports that the language of film “must be regarded as an entity in itself”, film dialogue differs from purely written and purely spoken discourse in terms of many parameters referring to the characteristics of language use (Taylor 2006). The comparative analysis carried out by the scholar showed significant differences in the use of discourse markers typical of the spoken language (e.g., right, OK, now) between film texts and spontaneous oral language taken from the Cobuild ‘Bank of English’ spoken corpus (Taylor 2004). Also, analyzing closings of telephone calls in 20 popular English-language films, Ryan and Granville suggested that films provide inauthentic models of conversation (Ryan & Granville 2020).

In the present research, a remarkable modification of the so-called standardized ways of terminating a conversation is a (relatively) high frequency of occurrence of such pragmatic types of CTM as CTM-statements informing about something, CTM-opinions and CTM-requests/commands. To put it differently, to inform someone about something in order to close a dialogue is not something ordinary or predictable, it deviates from a discursive norm of everyday talk. Although the so-called “non-conventional” means of ending a talk can also be found in natural communication, they are spontaneous rather than regular or systemic and are of different pragmatic nature compared to those in cinematic discourse. These “non-conventionalities” in real-life communication are described in House and Kádár (2014). Dividing 25 speech acts into two major types — ‘Substantive’ (that include attitudinal and informative speech acts) and ‘Ritual’ (that involve opening and closing speech acts), the scholars argue that “this typology represents the default function of speech acts, and any speech act can ‘migrate’ into other slots”, e.g. “in certain contexts, a Substantive Attitudinal speech act can take on a Ritual function” (House & Kádár 2024: 1699). Focusing on English-speaking conventions of extracting oneself from the interaction, House and Kádár attempted to systematize all the speech acts through which closing an interaction can be when it comes to a situation where extracting is needed. The analysis revealed that in English the closing phase is fulfilled not only by the Extractor, but also by such speech acts as Excuse/Justify, Apologise, Opine, and Thank. The authors emphasize the fact that in English the speech acts in the closing phase tend to be realized by routine formulae. They argue that the easy availability of routine formulae results in a more strongly ritualized pragmatic convention of extracting oneself in English, much less interactional work needs to be done to realize extracting (House & Kádár 2024: 1709). Also importantly, as House’s research showed, the reliance on routine formulas in the realization of many speech acts is much more typical of English speakers than of speakers of various other languages, e.g. German and Chinese (House 2006). According to Larina, while taking a leave, English speakers use more speech formulas than Russian speakers; the communicative actions of the former are more ritualized and conventionalized (Larina 2009, 2025).

In contrast to these observations from authentic speech practice, in the British films under study an interpersonal interaction is terminated by utterances of another pragmatic value (as pointed out above): statements informing about something, opinions, and requests/commands. It is worth special mentioning that their regularity of occurrence as CTM and frequency in the CF signals the development of fixed associations of their use with a certain “standardized communication situation” (according to Coulmas) — the closing phase of a filmic dialogue. In the majority of cases, they are utterances of non-formulaic nature from the point of view of their lexical fixedness and idiomaticity (Bladas 2012). Only some CTM of these pragmatic types involve formulaic elements. Thus, by modifying customary patterns of natural speech, cinematic discourse is elaborating its own pragmatic “conventions” of conversation closure, creatively synthesizing the use of formulaic and non-formulaic CTM, as well as trigger the emergence of new formulas. The identified pragmatic innovations can be explained by CTM functioning at simultaneously two levels in films — intradiegetic and extradiegetic. They terminate not only a dialogue between characters but also the whole scene, point to a move to the next episode. According to our findings, the majority of CTM-statements and CTM expressing opinions and requests/commands intensify the dramatic way of perceiving the events in the films.

Third, the formulaic component of the feature films exhibits not only pragmatic commonalities with other types of artistic discourse in the Written BNC2014 (i.e. fiction and scripts), but also divergences. These divergences in exploiting formulas to terminate interpersonal communication are both linguistic and metalinguistic. The linguistic ones concern the discovered quantitative and qualitative data of using formulas in the drama films on the one hand, and in various kinds of written texts of fiction and scrips, — on the other hand. The metalinguistic divergences consist in the stated dependency of distribution of formulas under analysis on a subgenre (of a fiction work, a script, or a film). In the films, the formulas undergo more syntactic and lexical variation, and are therefore more susceptible to the formation of slots in their structure — the process that Dąbrowska defines as “the transition from formula to schema”. As the scholar claims, this process “does not require translation into a different representational format, but merely loss of detail” (Dąbrowska 2014: 619). The variation of formulas in cinematic discourse can be accounted for by, as we would call it, the “spoken-to-be-written” format of films in comparison to the written format of fiction texts and scripts. As our material is limited to four drama films, to verify whether the established peculiarities are of regular character and can be defined as typical of feature films rather than of artistic discourse in general, it is necessary to conduct a research on larger empirical dataset. Within the field of pragmalinguistics, the comparative study of using formulas as well as other pragmatically relevant units in different kinds of artistic communication and art forms still remains a rather neglected area of inquiry. However, the findings of the present corpus analysis as well as of some earlier explorations of various pragmatic phenomena (including formulas) in artistic discourse make evident the fact that the issue of the so-called “pragmatic authenticity or pragmatic artificiality” of cinematic discourse and other types of artistic discourse is still unsolved and has many routes for further prolific scientific searches.

6. Conclusion

The present paper aimed to identify formulaicity of contact-terminating means in cinematic discourse and establish its pragmatic specificity in comparison with that in other types of artistic discourse. It outlined the results of two interrelated case studies of language means used to terminate a conversation in cinematic discourse compared to artistic discourse. To identify the formulaicity and pragmatic value of contact-terminating means (CTM), a complex methodology was elaborated, based on the application of two kinds of corpus approach: corpus-driven (Case study-1) and corpus-based (Case study-2). For Case study-1, a corpus of four British drama films was compiled ad hoc (the CF); whereas for Case study-2, the two subcorpora (“Fiction” and “Written-to-be-Spoken”) of the Written BNC2014 was applied (the BNC2014-F/WS).

The quantitative and qualitative data let us draw the general conclusion that the CTM used in cinematic and artistic discourse exhibit certain pragmatic specificity that consists both in modifying or deviating from conventional pragmatic patterns of conversation closure (due to particular artistic tasks) and in creatively exploiting formulas as well as producing new ones. All the modifications and innovations described in the present research facilitate not only to assess “the pragmatic authenticity” of cinematic discourse, but to trace recent trends of changes in real-life interpersonal interaction. The findings contribute to the development of pragmalinguistics of cinema and artistic communication by giving new knowledge of using formulaic means in artistic dialogues.

As a further step in identifying the formulaicity of the language of films, other types of pragmatically relevant language units (idioms, collocations, constructions, etc.) will be analyzed in a larger number of films and will be compared to particular types of artistic discourse.

Abbreviations

BNC2014-F/WBS — the “Fiction” and “Written-to-be-Spoken” subcorpora of the Written BNC2014 corpus

CF — the corpus of films

CTM — contact-terminating means

FML — formulaic language means

_{¹ Stribizhev, Viktor V. 2005. Speech cliches in Modern English: Meta-communicative function: Abstract of dissertation ... Candidate of Philology. Belgorod. 20 p. (In Russ.)}

About the authors

Irina V. Zykova

Author for correspondence.
Email: irina_zykova@iling-ran.ru
ORCID iD: 0000-0002-0430-7769

Doctor Habil. of Philology and a leading researcher at the Yuri Stepanov Department of Theory and Practice of Communication

References

Supplementary files

Supplementary Files

Action

1. JATS XML

Download

2. Figure 1. The percentage ratio of 12 pragmatic types of CTM extracted from the CF
Resource: The author’s research output.

Download (126KB)

Indexing metadata

3. Figure 2. The distribution of CTM of different pragmatic types in the CF
Resource: The author’s research output.

Download (108KB)

Indexing metadata