The Linguistic Profile of Horror/Suspense/Mystery Movie Dialogues Over TimeAbstract:
Horror/suspense/mystery movies have been produced since the inception of cinema. However, the literature on the description of their spoken language and how such language contributes to frighten, scare, and leave viewers hanging on the edge of their seat is scarce. Works on the subject tend to focus either on how to produce a good script for the genre or on the genesis of fear and/or suspense. The aim of this paper is to fill this gap by intertwining a section of two studies on the spoken language of movies (Veirano Pinto; Cortes, 2018; Veirano Pinto; Berber Sardinha, 2019), which used a corpus linguistics framework, with their typical narrative plots and considerations on the cognitive dimension of feelings of horror and suspense. Results show that communication, cognitive and speech act verbs, together with adverbials and questions, are responsible for the expression of characters’ emotions and the engagement and frightening of viewers.
Keywords:
horror; suspense; mystery; American movies; corpus linguistics
Resumo:
Filmes de terror/suspense/mistério são produzidos desde o início do cinema. No entanto, a literatura sobre a descrição da sua linguagem falada e sobre como essa linguagem contribui para apavorar, assustar e deixar os espectadores tensos é escassa. Trabalhos sobre o assunto tendem a centrar-se na forma de produzir um bom roteiro para o gênero ou na gênese do medo e/ou do suspense. O objetivo deste artigo é preencher esta lacuna, entrelaçando uma seção de dois estudos sobre a linguagem falada dos filmes (Veirano Pinto; Cortes, 2018; Veirano Pinto; Berber Sardinha, 2019), embasados na linguística de corpus, com suas narrativas típicas e considerações sobre a dimensão cognitiva dos sentimentos de horror e suspense. Os resultados mostram que os verbos de ato de fala, de comunicação e cognitivos, bem como advérbios e perguntas, são responsáveis pela expressão das emoções das personagens e pela tensão e medo gerado nos espectadores.
Palavras-chave:
horror; suspense; mistério; filmes americanos; linguística de corpus
1 Introduction
Horror/suspense/mystery stories have been with us for a long time and in several forms of expression, e.g., cave paintings, tales, short stories, novels, and movies (Schneider, 2009, blurb). “Movie menace has come in many forms and has given rise to some of cinema’s most iconic characters […] Frankenstein, Dracula, Carrie, Candyman, and Hannibal Lecter” (Schneider, 2009, blurb). This paper examines this register to help us understand how the linguistic choices made by scriptwriters and moviemakers contribute to leaving viewers hanging on the edge of their seat.
The movie genres of horror, suspense, and mystery are treated here as a single register because suspense and mystery are key elements in most horror narratives (Carroll, 1990, p. 128), just as the key theme in such narratives is knowledge. That knowledge is explored through the unknown and the unknowable and through conflicts “between humanity and the inhuman and the normal and abnormal” (Carroll, 1990, p. 126).
Although the literature on horror, suspense, and mystery stories is clear concerning plot structures, how these structures build emotions, and how the emotions of characters are received by and relate to those of the viewers (Barthes, 1977; Botting, 2013; Carroll, 1990; Lovecraft, 2013; Ortony; Clore; Collins, 2022; Zillmann, 1980), the literature on how dialogues contribute to triggering a sense of horror, suspense, and mystery is scarce. Our paper will therefore strive to highlight this by discussing what we already know about these movie genres and by entwining such knowledge with the results of previous works that have investigated the lexicogrammatical characteristics of movie dialogues through lexical bundles1 and multi-dimensional analysis2 (Veirano Pinto; Cortes, 2018; Veirano Pinto; Berber Sardinha, 2019). To that end, we will first discuss essential characteristics of horror/suspense/mystery narratives, to then show what we already know about movie speech, followed by the results of two corpus linguistics studies on the verbal language of movies. Finally, we will present dialogues and discuss their linguistic characteristics in view of plot structures, the triggering of viewer emotions, and the exploration of the unknown.
2 Horror/Suspense/Mystery Movies Emotional Aspects and Narrative Plots
2.1 Emotions in Viewers and Characters
For Carroll (1990, p. 14), “the genres of suspense, mystery, and horror derive their very names from the affects they are intended to promote - a sense of suspense, a sense of mystery, and a sense of horror”. The unsettling feelings we have when we watch movies of these genres, or a movie that crosses the three of them, comes from the uncertainty of a situation that may bring benefits or harm to those participating in it (Zillmann, 1980). As such, the main objective of moviemakers that embark on the production of horror/suspense/mystery movies is to lead the viewers to a certain emotional state.
The affects and feelings addressed by Carroll (1990, p. 14) consider the viewer’s awareness that what is being transmitted to them is merely a representation. Like the concept elaborated by Aristotle - the “tragic mimesis” and its relationship with mimicking real life (Aristóteles, 1994) - this filmic representation is based on reality but shows a certain asymmetry in the relationship between the viewers’ emotional state and that of the fictional characters. No matter how close we are to the situations faced by the characters in a cinematographic work, the feeling of suspense that affects the viewer does not reflect the exact emotion of the characters on the screen. As Carroll (1990, p. 23) points out: “although we feel the suspense when the hero rushes to save the heroine tied to the train tracks, he can’t allow himself such emotion”. Unlike horror, suspense is only linked to the viewer and can be characterized as a feeling that involves hope and fear “coupled with the cognitive state of uncertainty” (Ortony; Clore; Collins, 2022). Barthes (1977, p. 120) adds to the views of Carroll (1990) and Ortony, Clore and Collins (2022) by claiming that suspense often emerges from: (1) “emphatic procedures of delay and renewal” of a situation, (2) words such as wait, as they serve as catalyzers of emotions because they both establish a state that can be either rewarded or disappointed, and (3) a “gesture related linguistically”, which can be better explored in the narrative than a gesture that is only visualized. Horror, on the other hand, is symmetric, as during a scene of confrontation with the monster, or menace, the fear, the horror, and the physical manifestations of fear, like shivering, are likely to be felt by characters and viewers alike.
This asymmetry between the feelings of viewers and characters in a suspenseful situation, or the symmetry of feelings in a horrific situation, can be explained by the fact that feelings have physical and cognitive dimensions (Carroll, 1990, p. 24). The physical dimension of feelings of horror may show, for instance, through a state of agitation, shuddering, muscular contraction, and cringing. The cognitive dimension arises from a “cognitive construal and evaluation” of the situation being lived (Carroll, 1990, p. 27). As viewers are privy to knowledge and information that characters are not, it is easy to see why their emotional states differ in suspenseful situations. By contrast, both face the horror of imminent death, a monster or torture, thus having parallel emotions. As this study is concerned with how dialogues contribute to triggering a sense of horror, suspense, and mystery, it will focus on the cognitive dimension of feelings. Example (1) illustrates how dialogues contribute to the asymmetry of feelings between characters and viewers in a scenario of suspense. In this example, the characters Petru, Cazan, Vlad, and other soldiers are too wired to feel either hope or fear, unlike the viewers who may be on the edge of their seats without knowing how the situation will be resolved. In this case, suspense is created for the viewer through prepositional and verbal phrases (through the gates, keep moving along, get these men off their knees, stand up, will be coming from the north, double our sentries, do it now) and the question “Do you believe we can do this?”.
(1) Petru: Through the gates! Keep moving along! This way!
Cazan: You should never have strayed from us, whatever your reasons for leaving were…
Vlad: Quiet. Get these men off their knees. Stand up, do you hear me? Stand up! The Turks are coming! Prayers will not defend these walls!
Petru: Do you believe we can do this?
Vlad: Mehmed’s men will be coming from the north. Double our sentries on the perimeter walls. Go. Do it now.
(Dracula Untold, 2014)
Example (2) illustrates the symmetry of feelings in a horrific situation. The dialogue excerpt was taken from a scene (Figure 1) in Anabelle: Creation (2017) and shows the moment when Janice meets the demon inside the doll Anabelle and cries for help. In this scene, Janice starts talking to the doll, who apparently houses the spirit of the dead daughter of the Mullins. Both the viewers and Janice are terrified when they realize that the doll does not house the spirit of the girl, but that of a demon, who reveals itself through the noun phrase “Your soul”.
(2) Janice: Hi. You’re the Mullins’ daughter, aren’t you? What happened to you?
Anabelle: Will you help me?
Janice: What do you need?
Anabelle: Your soul!
Janice: Charlotte! Sister Charlotte! No! No!
(Anabelle: Creation, 2017)
Horror arises from a confrontation between a character and a monster and derives from the emotional state of the character who finds him/herself in such a situation (Carroll, 1990). Such emotional state “provide a set of instructions or, rather, examples about the way in which the audience is to respond to […] monstrous properties” (Carroll, 1990, p. 17). In the case of Example (2), these instructions come in the form of a cry for help: “Charlotte! Sister Charlotte! No! No!”. For the narrative elements to serve as an efficient source of fear, they need to compose an atmosphere (Lovecraft, 2013, p. 7) in which the reader, in our case the viewer, is immersed in the world and situation of horror. In the first decades of the genre, horror draws from its Gothic roots and is characterized by haunted environments, based on the architecture of the Dark Ages, in which the place itself needed to represent a cultural darkness that returns from the past to haunt the present with its superstitions and traditional monsters, such as the vampire and the ghost (Botting, 2013, p. 1). After the genre matured, horror adhered to the intellectual context of the Enlightenment, and its sources of fear began to be based on questions of the mind and the natural sciences, in which readers conventionally believe to this day - laws that supernatural elements will seek to revoke (Carroll, 1990, p. 57).
Of course, the success of horror/suspense/mystery movies, or rather, of any narrative movie, also depends on the identification of the viewers with some of the characters. For Carroll (1990, p. 89), this identification may occur if we: like the characters; recognize the circumstances in which he/she/they find themselves; sympathize with he/she/them; share the same interests, feelings, or principles; see the action unravelling from their perspective; and are entranced enough to fall under the illusion that we became one with the characters. Once the viewers identify with the characters, they respond emotionally to the larger scenario - “a horrifying monster […] threatening a human protagonist” - that they are witnessing from outside (Carroll, 1990, p. 93). In the end, if the monster is punished and the “virtuous” character(s) is/are rewarded, at least by staying alive, the viewer experiments catharsis, i.e., “a feeling of restoration, renewal, and revitalization” (Helm, 2021).
2.2 Narrative Plots
According to Carroll (1990), horror/suspense/mystery narratives are quite predictable, as they use a rather limited repertory of strategies. To him, there are basically two kinds of plots: the complex discovery plot and the overreacher plot (Carroll, 1990, p. 99, 118). The complex discovery plot is structured by four movements: onset, discovery, confirmation, and confrontation. The first movement, onset, happens when the viewers are presented to the monster or to a menace. The second movement, discovery, takes place when one of the characters or a group of characters finds that a monster or menace exists. People in power, such as “the police, eminent scientists, religious leaders, government officials, or the armed forces” do not believe in the existence of the monster or menace (Carroll, 1990, p. 101). It is during this movement that: (1) viewers are primed for the monster’s or menace’s characteristics, such as its invulnerability, strength, and habits, and (2) hypotheses, which will be later confirmed or refuted, are developed, contributing to the cognitive dimension of emotions. The third movement, confirmation, involves the acceptance of the existence of the monster/menace by those in power. Finally, confrontation occurs when the monster/menace is faced by all in a “problem/solution format” (Carroll, 1990, p. 103). During confrontation, we witness the battle between humanity and the inhuman and between normal and abnormal. Carroll (1990, p. 102) also contends that the onset, discovery, and confirmation movements in a complex discovery plot “can be iterated”.
The overreacher plot is characterized by the existence of “forbidden knowledge” related either to scientific experiments or black magic (Carroll, 1990, p. 118). This kind of plot also has four movements: preparation for the scientific experiment or incantation, the carrying out of the scientific or incantation experiment, the gathering of evidence that the experiment/incantation had an adverse effect, and the confrontation of the consequences of the experiment/incantation. In the first movement, we learn about how the experiment/incantation is going to be done and why it should be done. The why often has to do with some personality deviation or mental illness on the part of the character who is going to perform it. The second movement presents the viewer with a sequence of actions that are seen - often through a wide range of special effects - and sometimes reinforced with words. The third movement shows victims, destruction, and death. The final movement can include the villain showing regret and trying to fix the chaos he/she created or being killed and having his/her experiment/incantation reverted or partially reverted by others. Just as with the complex discovery plot, in the overreacher plot, the movements can be iterated (Carroll, 1990).
3 Movie Speech
Movie speech is not only a major part of cinema, but it also reflects a symbiotic relationship between language and culture. In the movies, cathartic narratives, told with the help of dialogues, tend to legitimize the experience of individuals through the representation of a wide range of social archetypes, such as housewives, husbands, lovers, heroes, nerds, jocks, and professionals of all sorts. These archetypes interact in an idealized English language in made up situations, which end up echoing American values, political tendencies, and economic situations, because when moviemakers idealize a story, it becomes tinged with their ideologies and their way of seeing the world.
Since the beginning of sound in the cinema, the aim of moviemakers of narrative American movies (Hollywood) has been to write dialogues that emulate natural conversation, while using communicative strategies to guarantee the development of the narrative (Kozloff, 2000). Such perceptions, regarding screen speech, have led to the understanding that movie dialogues have different functions. These functions are not solely defined by scriptwriters and directors, as they are also influenced by the actors’ performances, the editing and exhibition process, and last but not least, viewer reaction (Kozloff, 2000, p. 30). Kozloff (2000, p. 33) suggested two groups of functions for movie dialogue. The first group is related to how the dialogues communicate the narrative and the second to how they aid in aesthetic effects, ideological persuasion, and commercial appeal. Here, we will name functions in the first group only, as they are the ones that contribute to our better understanding of the verbal language of horror/suspense/mystery movies:
-
Anchorage of the diegesis (story) and characters
-
Communicating narrative causality
-
Enactment of narrative events
-
Character revelation
-
Adherence to the code of realism
-
Control of viewer evaluation and emotions
The first function of dialogue refers to telling the viewers where and when the story is taking place, as well as to presenting the social traits of the characters. Such functions are typically expressed through bundles of referential time, noun phrase expression, adverbial clause fragment, and pronoun/noun phrase + (auxiliary) + copula be (see Tables 1 and 2 for examples). The second function relates to what the characters say that makes the story move forward. In this case, the bundles that often fulfill this function are question fragments (yes-no/wh-) and simple inquiry bundles. The third function puts the action viewers see on the screen into words. As such, these words are often present in (auxiliary) + active verb and imperative attitudinal stance bundles. The fourth function of movie dialogues is to give the viewers a sense of the psychological traits of the characters. A glimpse of these traits tends to be seen in an epistemic personal/impersonal stance, stance (desire, prediction), pronoun/noun phrase + (auxiliary) + copula be, pronoun + verbal phrase (complement clause) bundles. The fifth function has to do with the apparent naturalness of the conversation and situations that seem plausible in a certain plot. Therefore, this function is revealed in 33% of the bundles identified for horror/suspense/mystery movies that had been previously identified in natural language (Biber et al., 1999). Finally, the sixth function is directly connected to the cognitive dimension of feelings and is revealed as shown, for instance, in Example 2, where the feelings of the characters are symmetrical to the feelings of the viewers. The bundles responsible for such symmetry may be (auxiliary) + active verb, noun phrase expression, pronoun/noun phrase + (auxiliary) + copula be and/or imperative attitudinal stance bundles.
Apart from the identification of functions 1 to 6, by Kozloff (2000), previous works (Forchini, 2012; Veirano Pinto; Berber Sardinha, 2019; Zago, 2016) have shown that movie dialogues are complex and marked by a series of linguistic and oral discourse features, such as modals, adverbs, adjectives, interjections, pause fillers, verbs, wh- questions, and pronouns. When combined, these features show the different degrees of spontaneity, planning, argumentation, expression of stance, persuasion, and assessment present in a movie dialogue as well as a tendency to focus on sequences of events, interpersonal relationships, or the situations lived by the characters (Veirano Pinto; Berber Sardinha, 2019). We also know that socioeconomic aspects and production circumstances affect movie speech (Berber Sardinha; Veirano Pinto, 2016). The stringent censure of the Hays Code coupled with the “assembly line” production process adopted in the Golden Age of Hollywood (1930s to late 1940s) had a strong influence on the verbal language of movies, making it very standardized. From the 1950s onwards, with the end of the Hollywood Studio System and the profound changes in society, including the emergence of hippies, free love, and rebellion against social standards and governmental rules, movie language started to be less standardized and more similar to natural conversation (Berber Sardinha; Veirano Pinto, 2016).
However, all the characteristics mentioned above are of movie speech in general. To the best of our knowledge, not much has been reported about the role of dialogues in horror/suspense/mystery movies and how they help build a feeling of tension, fear, and horror. An exception seems to be Kozloff (2000), but even in her work the focus is on dialogues of action, drama, and comedy movies. To fill this gap, in Sections 4 and 5, we will discuss two studies (Veirano Pinto; Cortes, 2018; Veirano Pinto; Berber Sardinha, 2019) that were carried out using two versions of the North American Movie Corpus (NAMC), focusing solely on horror/suspense/mystery movies. The first version of the NAMC consists of 640 movies of the genres of action/adventure, comedy, drama, horror/suspense/mystery, produced between the 1930s and the early 2000s, totaling 5,836,283 tokens and 325,000 types. The second version extended the corpus to include another 20 movies of each of these genres produced in the 2010s, thus including 720 movies, totaling 6,418,029 types and 60,873 tokens. The objective is to explore the linguistic profile of horror/suspense/mystery movies, pointing out how it has possibly contributed to viewers’ affects and to the plot structure, in an attempt to help us better understand the role of dialogues in such movies and to inform the training of scriptwriters.
4 The Lexical Bundles of Horror/Suspense/Mystery Movies
Veirano Pinto and Cortes (2018) used the first version of the NAMC, including movies from the 1930s to the early 2000s - 640 movies in total - to identify lexical bundles for the four different movie genres within the NAMC, classifying them according to their discourse functions and the movie dialogue functions mentioned above. In this version of the NAMC, there are 160 horror/suspense/mystery movies, totaling 1,118,467 tokens and 24,879 types. The bundles had to occur 40 times per million words in five texts or more and were extracted with the help of the Cluster tool in the toolkit AntConc, series 3. This frequency threshold follows a tested and tried tradition in lexical bundles studies (Biber, 2006; Biber; Barbieri, 2007; Biber et al., 2004).
Results showed that 33% of the bundles identified for horror/suspense/mystery movies had been previously identified in natural language (Biber et al., 1999). This result suggests that these three genres are the closest to natural oral language, as only 20% of the bundles of action and adventure movies, 17% of the bundles present in comedies, and 9% of the bundles in dramas had been previously identified in Biber et al. (1999).
Veirano Pinto and Cortes (2018) also revealed that, structurally (Table 1), horror/suspense/mystery movie bundles can be classified in six structural types, comprised of verbs (got to do with), pronouns and copula be or verb phrases (it’s going to be, you want to see), adverbial clause fragments (as soon as I), noun phrases (a matter of fact), and questions (how long have you). This result shows how these bundle structures contribute to: (1) the discovery, confirmation, and confrontation movements in the complex discovery plot (didn’t know you were, got to do with, as far as I, a long time ago, what have you got, I’d like you to see, get the hell out of, son of a bitch, can I do for), (2) prime the viewers for the monster’s or menace’s characteristics (it’s going to be, a matter of fact, a bit of a), and (3) the building of hypotheses (I don’t think so, I don’t know if, it’s got to be, it will be all right, but I don’t know). In the overreacher plot, bundles like take a look at, the name of the, want to talk to, where are you going, what are you trying, I’ve got to get, are likely to inform the viewer of the preparation of the experiment/incantation, while the bundles the middle of the, the end of the, nothing to do with, let me tell you, in front of the, I’ll be right back, may be present in the carrying out of the scientific or incantation experiment. The gathering of evidence that the experiment had an adverse effect and the confrontation of the consequences of the experiment, can involve the use of bundles like what happened to the, it was an accident, get the hell out of, are you trying to, can you hear me, I didn’t mean to, I don’t care what.
Another finding of Veirano Pinto and Cortes (2018) was that the discourse functions of the bundles in horror/suspense/mystery movies (Table 2) help perform personal and impersonal epistemic stances (know what to say; it was an accident), imperative attitudinal stances (get the hell out), desires (I want to show), prediction stances (it’s got to be), simple inquiries (what have you got), and referential/time (a soon as I). As such they add to the cognitive dimension of emotions by revealing characters’ feelings, such as insecurities (but I don’t want, I don’t know anything about), concerns (are you all right, do you mind if, I wish I could), agitation (get the hell out of, a hell of a, get him out of), anger (how did you get, I don’t care what, son of a bitch), fear (got to get out, get me out of), and incredulity (I don’t believe it, the hell are you, where did you get).
When the bundles in Veirano Pinto and Cortes (2018) were analyzed in terms of the six functions of movie dialogues identified by Kozloff (2000), the authors found that most of the bundles identified for horror/suspense/mystery movies are used to communicate narrative causality, to enact narrative events, and to reveal the nature of the characters. Only 8% of the bundles are used in the anchorage of the diegesis and characters. In view of the considerations made about the emotions in viewers and characters, as well as the plot structure of such movies, this comes as no surprise.
In short, we can say that when directors and scriptwriters produce movie dialogues, they take into consideration all the other filmic elements that are part of a genre to select language that will help them achieve a desired result. They seem to be aware, for instance, that to promote a sense of suspense, a sense of mystery, and a sense of horror, which comes from the uncertainty of a situation that may bring benefits or harm to those participating in it (Zillmann, 1980), they can resort to language that may be less varied structurally, but that is rich in terms of emotion and action. Such language will help them reveal the psychological nature of characters (stance bundles), anchor the story (referential bundles), communicate, and enact narrative events (question and adverbial clause fragment bundles; verb phrases). The findings presented in this section tally with the results found in Veirano Pinto and Berber Sardinha (2019), who identified the multidimensional linguistic profile of American movies in general, and in so doing, the multidimensional linguistic profile of horror/suspense/mystery movies. These results, together with a review of their work, will be presented in the next section.
5 Dimensions of Movie Variation: The Linguistic Profile of Horror, Suspense, and Mystery Movies
As mentioned in the introduction, the linguistic profile for these genres was obtained from Veirano Pinto and Berber Sardinha (2019). The purpose of this study was to describe the linguistic profile of American movies by means of identifying their dimensions of variation. To that end, Veirano Pinto and Berber Sardinha (2019) extended the first version of the NAMC, described in Section 4, to include movies produced in the 2010s. As a result of this extension, the NAMC in this study included 180 horror/suspense/mystery movies, totaling 1,244,397 tokens and 38,626 types. The 2010s movies were selected based on the criteria used when the corpus was first built in 2012. Those criteria included the opinion of film critics - found in the different editions of “1001 films to see before you die” and on the website Filmsite LLC (1996-2025) - and the nominations and awards they received from the following Academies of cinema: AMPAS (Academy of Motion Pictures Arts and Sciences; Oscar), BAFTA (British Academy of Films and Television Arts) and Horror, Fantasy and Science Fiction Films (Science Fiction Fantasy and Horror Films). The films were collected in their entirety, manually, by extracting subtitles from some DVDs and largely through the websites Subscene.com and Opensubtitles.org (2006-2025). Table 3 shows the composition of the second version of the North American Movie Corpus (NAMC).
Dimensions of variation are linguistic constructs derived from multidimensional analysis, a methodological framework that uses the statistical procedure known as factor analysis to map the existing associations between linguistic variables, grouping them into factors. Once the factors are interpreted communicatively and functionally, the dimensions emerge. The interpretation of the dimensions involves looking at the features that co-occur in a pole, to find an underlying communicative function for them. As an example, a careful look at the features of Dimension 1 positive pole (Simulated spontaneity) in Table 4, reveals that these features are quite common in face-to-face conversations. As lines in movies are written to be spoken and there is very little spontaneity in them, if any at all, as the lines are planned, rehearsed, and edited in the final cut of the movie, the positive pole of factor 1 was interpreted as Simulated spontaneity. Veirano Pinto and Berber Sardinha (2019) followed this procedure for all the factors that resulted from the factor analysis carried out in their study and identified seven dimensions for North American movies, namely:
-
Simulated spontaneity versus planned discourse
-
Expression of event sequences versus argumentative discourse
-
Expression of stance
-
Interpersonal focus
-
Expression of persuasion
-
Situational focus
-
Expression of assessment
These dimensions describe the language we find in 90 years of American spoken movies. In other words, we know that each of the movies in the NAMC has a greater or lesser degree of spontaneity, planning, verbalization of sequences of events, argumentation, stance, persuasion, assessment, and focus on interpersonal relationships or the situation, which are the functional communicative parameters present in movies. The lexicogrammatical characteristics that make up each of these dimensions are given in Table 4.
The average dimension score for horror/suspense/mystery movies - which are the focus of this study - in these dimensions are 1.51 (Dimension 1), -0.18 (Dimension 2), 1.05 (Dimension 3), 1.19 (Dimension 4), 1.74 (Dimension 5), -0.17 (Dimension 6), and -0.56 (Dimension 7). These average scores mean that the dialogues of such movies are slightly marked by simulated spontaneity - a fact confirmed by the fact that 33% of the bundles found for the genre had been previously identified in natural conversation (Biber et al., 1999) -, argumentation, stance, interpersonal relationships, and persuasion. Examples 1 to 7 will illustrate some of the movie functions proposed by Kozloff (2000), movements of the complex discovery plot and of the overreacher plot, the structures used to reveal characters’ emotions, as well as how the linguistic features present in each of these dimensions are combined to perform the communicative functions of each dimension (these will be underlined).
Example (3) was taken from the movie The Hitcher (1986). It tells the story of a boy who escapes a serial-killer hitchhiker and is later stalked by him and framed for his crimes. The score of The Hitcher in Dimension 1 is 28, well above the average score of Dimension 1 (1.51). The excerpt below is the opening scene, and we can clearly see the onset movement of the plot. The first lines are coming out of the radio in the car and anchoring the diegesis and the characters, telling us the time (“It’s 4:15 on a Saturday morning”) and where the characters are (“El Paso town”). The dialogue also tries to give the viewers a sense of the personality of the boy (“My mother told me never to do this”), who seems to be a bit rebellious, but who has had an apparently caring mother. The man whom we later learn is a serial-killer seems, at first, to be a good, religious, thoughtful man (“Bless you. My name’s Jim Halsey”, “I’m getting your car wet”). We only start to think there is something wrong with him when the boy starts to feel nervous (“You gonna tell me where you’re going?”, “What are you looking at me like that for?”, “What are you doing? What’d you do that for? I asked you a question”, “Look, I think you better get out now. The ride’s over. Goodbye”) because of the evasive manner of the hiker (“You got a smoke?”, Just looking.”) and his sarcasm (“Scared you?”).
(3) “It’s 4:15 on a Saturday morning in El Paso town. This is Uncle Bruce here going on with the farm and market report on the station all El Paso listens to.”
“My mother told me never to do this.”
“Bless you. My name’s Jim Halsey.”
“John Ryder. So, you want me to drop you off somewhere?”
“I’m getting your car wet.”
“Oh, this isn’t my car. It’s a drive-away. The owner’s in San Diego.”
“So, where you headed?”
“You got a smoke?”
“Yeah, sure. You gonna tell me where you’re going?”
“Yeah, sure.”
“What are you looking at me like that for?”
“Just looking.”
“What are you doing? What’d you do that for? I asked you a question.”
“Scared you?” “Look, I think you better get out now. The ride’s over. Goodbye.”
(The Hitcher, 1986)
Example (3) also has several marks of simulated conversation (Dimension 1) like contractions (name’s, I’m, isn’t, you’re, gonna, what’d, ride’s), wh- questions (where you headed, what are you looking at me like that for, what are you doing, what’d you do that for), first- and second-person pronouns (I, you, me, my, your), demonstrative pronouns (this, that), pronoun it, to complement clauses controlled by verbs of desire, intention, and decision (want me to drop you off), and discourse markers (look).
In sum, Example 3 clearly shows how dialogues contribute to creating suspense by reinforcing the actions we see on the screen and conveying how the characters feel. When the scriptwriter gives the line “What are you looking at me like that for” to the boy, he/she is putting into practice what Barthes (1977, p. 120) contends as an effective way of creating suspense: “a gesture related linguistically” can be better explored in the narrative than a gesture that is only visualized. The suspense is further enhanced when the boy voices his nervousness and agitation by saying: “What are you doing? What’d you do that for? I asked you a question”. We can suggest that both these questions are rather rhetorical and are there mostly to start giving the viewer a feeling of discomfort and to build suspense, as suspense is created when the viewer is privy to information the characters do not have (Carroll, 1990).
Example (4) shows an excerpt from The Ring (2002). The movie is about a journalist who is investigating a videotape that seems to precipitate the death of those who watch it. The Ring score is 14.7, also well above the mean score for Dimension 1 (1.51). The excerpt illustrates one of the discovery movements of the plot, i.e., when a character finds that a monster or menace exists, but people do not believe him/her (“I’m sorry, but I’m not going to get all worked up over some high school rumor”). It is also in this movement that the viewer learns about some characteristics of the menacing videotape (“The numbers are all screwed up”; “That’s impossible. The numbers are the control track. They’re put on the tape whenever it’s recorded. Which means, theoretically, there shouldn’t be any images”; “So, the control track could tell us where it came from. But to not have one… that’s like being born without fingerprints”).
Thus, the function of these lines is to reveal some of the characters’ characteristics to the viewers. They establish Noah as a grounded video and digital image expert, who does not let himself get easily influenced or disturbed by what people say (“I’m sorry, but I’m not going to get all worked up over some high school rumor”). They show Rachel as a person who is considerate (“I don’t want her to see it”; “I did have a problem”), perseverant (“I’m an obsessive bitch, needs to lighten up”), and pragmatic (“So why don’t we just cut to that part right now and save ourselves a lot of grief?”).
(4) “You sure this is a copy?”
“Yeah, why?”
“The numbers are all screwed up.”
“The same problem got copied, I guess.”
“That’s impossible. The numbers are the control track. They’re put on the tape whenever it’s recorded. Which means, theoretically, there shouldn’t be any images.”
“Hey, Noah. Can you pretend for one minute that I don’t read Video Geek Magazine?”
“Okay, when you record a tape, the makeup of the tracks is like a signature for whatever did the recording, like a camcorder, VCR, whatever. So, the control track could tell us where it came from. But to not have one… that’s like being born without fingerprints.”
“Then, how did this get recorded?”
“I don’t know. But I’m going to find out.”
[…]
“You’re taking the tape?”
“I don’t want her to see it.”
“You don’t want her to see it? What about me seeing it? You had no problem showing it to me.”
“I did have a problem.”
“You’re angry.”
“I’m not angry.”
“Oh, no? What are you then?”
“In a hurry.”
“I’m sorry, but I’m not going to get all worked up over some high school rumor. What are you thinking?”
“I’m thinking this is the part where I say, ‘you’re a flake, never finishes anything’. Then you say, ‘I’m an obsessive bitch, needs to lighten up.’ Then I storm out, all pissed off, kicking myself for ever thinking that you could just… grow up. So why don’t we just cut to that part right now and save ourselves a lot of grief?”
“All right. Bye.”
(The Ring, 2002)
Just as Example (3), Example (4) has several occurrences of the language present in simulated conversation (Dimension 1). There are many occurrences of verbs (is, are, got copied, guess, put on, recorded, means, there shouldn’t be, pretend, tell us, came, being born, know, find out, taking, want, seeing, showing, worked up, thinking, finishes, needs, grow up, cut, save), contractions (that’s, they’re, it’s, shouldn’t, don’t, you’re, I’m), first- and second-person pronouns (I, you, us, we, myself, her, me, ourselves) to complement clauses controlled by verbs of desire, intention, and decision (I’m going to find out, I’m not going to get all worked up over some high school rumor, needs to lighten up), pronouns it, wh- clauses (how did this get recorded?, What about me seeing it?, What are you then?, What are you thinking, why don’t we just cut to that part right now and save us a lot of grief?), demonstrative pronouns (this, that), and discourse markers (Hey, okay, all right).
On watching the scene in Example (4), shown 35 minutes after the beginning of the movie, the viewer might start feeling sympathy and/or concern for Rachel - the journalist who is investigating the menacing videotape - as she will clearly have trouble convincing people that the videotape kills those who watch it. The scene also helps build a sense of mystery (“Which means, theoretically, there shouldn’t be any images”; “So, the control track could tell us where it came from. But to not have one… that’s like being born without fingerprints”; “Then, how did this get recorded?”) and suspense, as the viewer is uncertain about what she will do next. This scene is probably going to be followed by a few more iterated discovery and confrontation movements, to give the viewer some more information about the videotape and to cause horror, fear, and shuddering - when those who watch the videotape die and help Rachel gather information to convince others of the menace the videotape poses. Next, Example (5) will illustrate the linguistic features that are typical of Dimension 2 (Argumentative discourse) and a confirmation movement in the plot.
Example (5) was extracted from the movie Bram Stoker’s Dracula (1992). In Bram Stoker’s Dracula the vampire Count Dracula goes to England to seduce Mina Murray, who is to be married to Jonathan Harker. The score of the movie in the negative pole Dimension 2, Argumentative discourse, is -13.62, which is above the mean score for the genre: -0.18. The dialogue that was extracted takes place between Mina and Prof. Van Helsin. The function of this dialogue is to communicate what is next to the viewer (narrative causality), in this case, how they will find Dracula again (“I want you to help me find him before it is too late. Please, help me find him”). It also reveals Van Helsin’s impressions of Dracula, showing that he admires and respects Count Dracula (“He was, in life, a most remarkable man and his mind was great and powerful”), despite him being a monster (“but greater is the necessity to stamp him out… and destroy him utterly”). The viewer, who by now knows how much Dracula loves Mina, and might identify him/herself with the monster on some level and may be feeling sorry for him, upon hearing this dialogue, is likely to be reminded of the need to end such evil (“Your salvation is his destruction”).
(5) “You admire him.”
“Ja. He was, in life, a most remarkable man and his mind was great and powerful… but greater is the necessity to stamp him out… and destroy him utterly.”
“Doctor?”
“Ja?”
“I know that I am becoming like him.”
“Your salvation is his destruction. That is why I want to hypnotize you. I want you to help me find him before it is too late. Please, help me find him. Please. Look into this light… the light of all light, into this flame. Your eyes are heavy. You want to sleep. Sleep now. Sleep.”
(Bram Stoker’s Dracula, 1992)
The movement of confirmation in this scene takes place through Mina’s and Van Helsin’s acceptance that she is infected by Dracula’s blood, and by Van Helsin presenting Mina with the reasons for the need to find and kill Dracula. Among the features that are typical of Dimension 2, we have coordinating conjunctions as phrasal connectors (and), predicative adjectives (great, powerful, greater, late, heavy), amplifiers (most, too, all), and wh- relative clause on subject position (why I want to hypnotize you). Next, Example (6) brings another discovery movement that was likely designed to build suspense that will lead to the movement of confrontation.
Example (6) was taken from the movie Demonic (2015). In this movie a police officer and a psychologist are looking into the death of five friends who were holding a séance in an abandoned house, where people had been murdered during another séance years earlier. The murders had never been solved. The mean score for Dimension 3, Expression of stance, is 1.5 and the score of Demonic in this dimension is 15.5. The dialogue in Example 6 is an interview that Brian is conducting with John, who knows about the history of the house. The function of the dialogue is to reveal their personality (“I think we actually do the things that we can say we can do, right?”; “I just want to know why this guy always acts he’s too cool for all of us”; “I think you’re actually curious about this house. You think there’s some kind of connection that you can’t fully explain”; “Because we’re all full of shit. And here you are. So, why don’t you admit that deep down you hope that we can make sense of this for you”), while communicating to the viewer that John may have ulterior motives to go with the others to the house (“I think you’re actually curious about this house. You think there’s some kind of connection that you can’t fully explain”). Thus, adding a sense of mystery that will likely contribute to the cognitive dimension of emotions by triggering curiosity.
(6) “I think we actually do the things that we can say we can do, right?”
“Bryan, easy up man. This is just an interview.”
“Shut the hell up.”
“I just want to know why this guy always acts he’s too cool for all of us.”
“Not all of you, Bryan.”
“Fuck! You know what I think. I think you’re actually curious about this house. You think there’s some kind of connection that you can’t fully explain.”
“You don’t believe any of this, of course.”
“Because we’re all full of shit. And here you are. So, why don’t you admit that deep down you hope that we can make sense of this for you.”
(Demonic, 2015)
Example (6) also shows how the linguistic features in Dimension 3 combine to help the viewer understand who the characters are. There are mental and cognition verbs (think, know, believe, admit, hope, make sense), emphasizers (the hell, too, fully, deep, full), factive adverbs (actually, of course), hedges (just, kind of), adverbs (always, just), and conjunctions (because, so). Next, Example (7) shows a confrontation plot movement that depicts the linguistic features that are typical of Dimension 4, Interpersonal focus.
Example (7) was taken from the movie Psycho (1960). Psycho has a score of 6.89 in Dimension 4 (mean score 0.20) and is a classic horror movie that tells the story of a secretary, Marion, who steals US$ 40,000 from her company. While running away, Marion stops at a motel run by a psychopath, Norman Bates. Her sister, Lila, and her lover, Sam, start searching for her and end up in Bates Motel, where Norman has already killed Marion and a private detective that also went to the motel in search of Marion. The scene in Example (7) is a confrontation between Sam, and Norman Bates, just before Norman goes after Lila to kill her. By this time, Lila and Sam have already convinced the local police that there is something odd going on at Bates Motel. The dialogue reinforces the abnormality of Norman’s character and reveals his feelings of fright (“You look frightened. Have I been saying something frightening?”), denial (“I don’t know what you’ve been saying”), confusion (“Do what?”), and anger (“Why don’t you just get in your car and drive away from here, okay?”; “Shut up!”). This dialogue also voices several unanswered questions to the viewer and prepares he/she for what comes next: Norman going after Lila to kill her. The features that are typical of a focus on interpersonal relationships are underlined. They are communication verbs (saying, talking, tell), infinitive verbs (hide, get, do), and past tense verbs (socked, came).
(7) “You look frightened. Have I been saying something frightening?”
“I don’t know what you’ve been saying.”
“I’ve been talking about your mother, about your motel. How you gonna do it?”
“Do what?”
“Buy a new one in a new town where you won’t have to hide your mother.”
“Why don’t you just get in your car and drive away from here, okay?”
“Where will you get the money to do that, Bates? Or do you already have it socked away?”
“Shut up!”
“A lot of it. $40,000. I bet your mother knows where the money is and what you did to get it. I think she’ll tell us.”
“Where’s that girl you came here with? Where is she?”
(Psycho, 1960)
In another excerpt that follows the physical confrontation between Norman, Sam, and Lila (Example 8), the local sheriff, Al Chambers, explains - to the characters and viewers alike - what drove Norman to keep the body of his dead mother in the cellar and to become the psychopath that he is. Thus, satisfying viewers’ curiosity and giving further details of Norman’s personality through the emotions that he felt. Example (8) brings further features that are characteristic of Dimension 4, like past tense verbs (was, assumed, felt, met, wanted), that complement clause controlled by verbs (assumed that she was as jealous of him).
(8) “He was never all Norman, but he was often only Mother, and because he was so pathologically jealous of her, he assumed that she was as jealous of him. Therefore, if he felt a strong attraction to any other woman, the mother side of him would go wild. When he met your sister, he was touched by her, aroused by her. He wanted her.”
(Psycho, 1960)
Finally, Example (9), illustrates the expression of persuasion in movies (Dimension 5; mean score 1.74). This was taken from the movie Dr. Jekyll and Mr. Hyde (1931; score 14.4), which tells the story of a scientist who becomes a monster after conducting an experiment to split his good from his bad personality. In Example (9), we have the conversation between the scientist, Dr. Jekyll, and his friend Dr. Lanyon. This example represents the first movement of the overreaching plot: the preparation for the scientific experiment. The dialogue informs the viewer of what the experiment entails (“Separate the two natures in us”) and why Dr. Jekyll wants to risk his life performing it (“I don’t want to accept. I want to be clean, not only in my conduct, but in my innermost thoughts and desires”). As proposed by Carroll (1990), the why often has to do with some personality deviation or mental illness on the part of the character who is going to perform it (“Why aren’t you frank enough to admit that other indecent self in you”). The dialogue in Example (9) also reveals the characters’ personality - the tenaciousness of Dr. Jekyll (“I don’t want to accept”; “it’s unscientific not to admit the possibility of anything”) and the compliant nature of Dr. Lanyon (“You have to accept certain things”) - and what will happen next (“We’ll see”).
(9) “Perhaps you’ve forgotten you’re engaged to Muriel.”
“Forgotten it?”
“Can a man dying of thirst forget water? Did you know what would happen to that thirst if it were denied water?”
“If I understand you correctly, you sound almost indecent.”
“What names you give things.”
“Why aren’t you frank enough to admit that other indecent self in you?”
“No. You prefer to hide it, pretend it isn’t there.”
“You have to accept certain things.”
“I don’t want to accept. I want to be clean, not only in my conduct, but in my innermost thoughts and desires. There’s only one way to do it.”
“How?”
“Separate the two natures in us.”
“That absurd theory of yours.”
“I tell you, it’s unscientific not to admit the possibility of anything.”
“You’re mad.”
“Mad, Lanyon? We’ll see.”
(Dr. Jekyll and Mr. Hyde, 1931)
Example (9) presents some of the linguistic features that characterize the expression of persuasion, such as modals of prediction (would; We’ll), verb be (were, aren’t, isn’t, be), perfect aspect verbs (have forgotten), sum stance to complement clauses controlled by adjectives (frank enough to admit that other indecent self in you; unscientific not to admit the possibility of anything), conditional subordination (if), downtoners (almost), to complement clause controlled by verbs of probability and simple fact (prefer to hide it; happen to that thirst), and existence verbs (There’s). Now that we have commented and illustrated the movements of the two plots that are common in horror/suspense/mystery movies, analyzed the words used to reveal characters’ emotions, as well as shown how linguistic features present in each of these dimensions combine to help create horror, suspense, and mystery, we will make some final considerations regarding how the linguistic choices of moviemakers develop the narratives and express characters’ emotions to involve and frighten viewers.
6 Final Considerations
The studies on which this paper was based (Veirano Pinto; Cortes, 2018; Veirano Pinto; Berber Sardinha, 2019) described the language of movies as a whole and scratched the surface of the language present in the different movie genres they examined, namely, action/adventure, comedy, drama, and horror/suspense mystery movies. Thus, to have a deeper understanding of which lexicogrammatical features contribute to frightening, scaring, and leaving viewers hanging on the edge of their seat, it was necessary to consider the functions of dialogues in movies, the typical plots of the genre, as well as the cognitive dimension of horror.
Examples 1 to 9 have shown that, to a greater or lesser extent, horror/suspense/mystery movies have dialogues that use simulated spontaneity, argumentation, stance, interpersonal relationships, and persuasion, but do not focus on the horrific or mysterious situation or the assessment of such situations by the characters. Rather, horror/suspense/mystery stories seem to be told through interpersonal relationships, stance, argumentation, and persuasion. Such a finding matches Carroll’s (1990, p. 126) viewpoint that the key theme in such narratives is knowledge, which seems to be conveyed to the viewer through lines that: (1) focus on the relationships between the characters, (2) reveal characters’ stance on the situations they face, (3) expose characters’ arguments in favor or against a horrific situation, and (4) try to persuade interlocutors to believe/disbelieve a threat. This refers to knowledge that is explored through the unknown and the unknowable, as well as through conflicts “between humanity and the inhuman and the normal and abnormal” (Caroll, 1990, p. 126). Knowledge in its several manifestations - from the perspective of the characters and the viewers - prompt the viewers for what is coming next, reinforce the actions being shown on the screen to enhance suspense (Barthes, 1977), and help viewers gain a sense of both characters’ personality and what is in store for them. The five discourse functions of the bundles, identified for an earlier version of the NAMC, corroborate such characteristics, as three of these are related to stance and attitude and one with simple inquiry (Veirano Pinto; Cortes, 2018). Thus, the conveying of characters’ emotions in movies seems to be much more linked to structures that combine: (1) pronouns and verbal phrases/copula be, (2) auxiliary verbs and action verbs, (3) adverbials, and (4) noun phrase expressions of time and place, as well as question fragments referring to words and expressions that describe feelings. As such, the speech act verbs, communication, and cognitive verbs that are part of the structures described in Section 4, together with the adverbials and questions, seemed to be, more often than not, the tools used by scriptwriters and moviemakers to trigger - in a symmetric or asymmetric way - the cognitive dimension of emotions in viewers.
These results show that horror/suspense/mystery movies tend to voice actions in their lines, have very few marks of planned discourse - especially those from the 1940s onwards -, and advance their plots through stances marked as argumentative discourse since their inception in the 1930s. Therefore, it is interesting to see that the language of horror/suspense/mystery movies has not been significantly affected by socioeconomic aspects and production circumstances, such as the “assembly line” production process adopted in the Golden Age of Hollywood (1930s to late 1940s), the lifting of the Hays Code, or the end of the Hollywood Studio System at the end of the 1950s, the latter being followed by profound changes in society - like the emergence of hippies, free love, and rebellion against social standards and governmental rules.
All in all, such findings show the embarrassment of riches that corpus linguistics brings to the extant literature on horror/suspense/mystery movies. Without its tools and methodologies, it would be much harder to see how the language choices made by scriptwriters and moviemakers contribute to engender horror, suspense, and mystery in stories that frighten, scare, and leave viewers hanging on the edge of their seat. The results presented here have the potential to inform courses on the scriptwriting of horror/suspense/mystery movies and provide some food for thought for moviemakers of the genre.
Data Availability Statement
The data used in this study is available upon request to Marcia Veirano Pinto, who does not have access to any safe repository to house the corpus or the analyses spreadsheets for free.
References
- ANABELLE: Creation. Directed by David F. Sandberg. [S. l]: New Line Cinema, 2017.
- ARISTÓTELES. Poética: tradução, prefácio, introdução, comentário e apêndices de Eudoro de Sousa. Trad. Eudoro de Sousa. 4. ed. [S. l]: Imprensa Nacional, Casa da Moeda, 1994. 319p.
- BARTHES, R. Image Music Text London: Fontana Press, 1977. 220p.
-
BERBER SARDINHA, T.; VEIRANO PINTO, M. Predicting American Movie Genre Categories from Linguistic Characteristics. Journal of Research Design and Statistics in Linguistics and Communication Science, v. 2, n. 1, p. 75-102, 2016. DOI: https://doi.org/10.1558/jrds.v2i1.27515.
» https://doi.org/10.1558/jrds.v2i1.27515 -
BIBER, D. A Typology of English Texts. Linguistics, v. 27, p. 3-43, 1989. DOI: https://doi.org/10.1515/ling.1989.27.1.3.
» https://doi.org/10.1515/ling.1989.27.1.3 -
BIBER, D. et al If You Look at…: Lexical Bundles in University Teaching and Textbooks. Applied Linguistics, v. 25, n. 3, p. 371-405, 2004. DOI: https://doi.org/10.1093/applin/25.3.371.
» https://doi.org/10.1093/applin/25.3.371 - BIBER, D. et al Longman Grammar of Spoken and Written English Harlow: Pearson, 1999. 487p.
- BIBER, D. University Language: A Corpus-Based Study of Spoken and Written Registers. Amsterdam; Philadelphia: John Benjamins, 2006. 261p.
-
BIBER, D.; BARBIERI, F. Lexical Bundles in University Spoken and Written Registers. English for Specific Purposes, v. 26, n. 3, p. 263-86, 2007. DOI: https://doi.org/10.1016/j.esp.2006.08.003.
» https://doi.org/10.1016/j.esp.2006.08.003 - BOTTING, F. Gothic 2. ed. London; New York: Routledge, 2013. 240p.
- BRAM Stoker’s Dracula. Directed by Francis Ford Coppola. [S. l]: Columbia Pictures, 1992.
- CARROL, N. The Philosophy of Horror, or Paradoxes of the Heart New York; London: Routledge, 1990. 256p.
- DEMONIC. Directed by Will Canon. [S. l.]: Dimension Films, 2015.
- DRACULA Untold. Directed by Gary Shore. [S. l]: Universal Pictures, 2014.
- DR. Jekyll and Mr. Hyde. Directed by Rouben Mamoulian. [S. l]: Paramount Pictures, 1931.
- DUTRA, D. P.; BERBER SARDINHA, T. The Role of Corpus Linguistics in EAP. In: SARMENTO, S.; REBECHI, R.; MATTE M. L. (ed.). English for Academic Purposes: Reflections, Description and Pedagogy. Porto Alegre: Editora Zouk, 2024. p. 14-54.
-
FILMSITE LLC. Homepage [S. l.: s. n], 1996-2025. Available at: Available at: http://www.filmsite.org Accessed on: 09 Sept. 2025.
» http://www.filmsite.org - FORCHINI, P. Movie Language Revisited: Evidence from Multi-Dimensional Analysis and Corpora. Bern: Peter Lang AG, 2012. 142p.
-
HELM, R. Complete Catharsis [S. l.], 2021. Available at: Available at: https://www.jtdschool.org/news-detail?pk=1161687 Accessed on: 16 Aug. 2024.
» https://www.jtdschool.org/news-detail?pk=1161687 - KOZLOFF, S. Overhearing Film Dialogue Berkeley: University of California Press, 2000. 323p.
- LOVECRAFT, H. P. Supernatural Horror in Literature Abergele: The Palingenesis Project (Wermod and Wermod Publishing Group), 2013. 212p.
-
OPENSUBTITLES.ORG. Homepage [S. l.: s. n], 2006-2025. Available at: Available at: http://www.opensubtitles.org Accessed on: 09 Sep. 2025.
» http://www.opensubtitles.org - ORTONY, A.; CLORE, G.; COLLINS, A. The Cognitive Structure of Emotion Cambridge: Cambridge University Press, 1988. 293p.
- PSYCHO. Directed by Alfred Hitchcock. [S. l.]: Alfred J. Hitchcock Productions, 1960.
- SCHNEIDER, S. J. (ed.). 101 Horror Movies You Must See Before You Die London: Barron’s, 2009. 415p.
- THE HITCHER. Directed by Robert Harmon. [S. l.]: HBO Pictures, 1986.
- THE RING. Directed by Gore Verbinski. [S. l.]: DreamWorks Pictures, 2002.
- VEIRANO PINTO, M.; BERBER SARDINHA, T. A Text Typology of American Movie Dialogues. In: INTERNATIONAL CORPUS LINGUISTICS CONFERENCE, 10, 2019, Cardiff.
- VEIRANO PINTO, M.; CORTES, V. Lexical Bundles Across North American Movie Genres: Discursive Functions and Levels of Routinization. In: AMERICAN ASSOCIATION FOR CORPUS LINGUISTICS, 14., 2018, Atlanta.
- ZAGO, R. From Originals to Remakes: Colloquiality in English Film Dialogue Over Time. Roma: Bonanno Editore, 2016. 195p.
- ZILLMANN, D. Anatomy of Suspense. In: TANNENBAUM, P. H. (ed.). The Entertainment Functions of Television London: Psychology Press, 1980. p. 133-157.
-
1
“Lexical bundles are sequences of words that most commonly co-occur in a register” (Biber et al., 1999, p. 989). However, the concept of lexical bundles should not be confused with those of idioms, as although the words that make them up also have a statistical tendency to co-occur, the meaning of their lexical components is transparent. In addition, the words that construct them do not allow for any variation, as variation in one of their words gives rise to a new lexical bundle. Other criteria for a sequence of words to be considered a bundle are that the sequence occurs in at least five texts in the corpus and with a minimum frequency of 40 times per million words (Biber, 2006; Biber; Barbieri, 2007; Biber et al., 2004).
-
2
“Multi-dimensional analysis is a framework used to identify sets of correlated linguistic features shared across many different texts in a corpus. These correlated sets, which are statistically identified through factor analysis, are communicatively interpreted as dimensions, the underlying parameters of variation in language use” (Dutra; Berber Sardinha, 2024, p. 36). “It is important to note that it is common for dimensions to comprise two ‘poles’, that is, two different sets of features in complementary distribution in the texts, such that when the features in one pole occur in the text, the features in the other pole are generally absent, and vice-versa” (Dutra; Berber Sardinha, 2024, p. 37).
-
Use of AI:
This study did not make use of AI.
-
Reviews:
As part of the commitment made by the Brazilian Journal of Applied Linguistics to Open Science, the journal publishes the reviews issued regarding its published works, when authorized by all parties involved.
-
Funding:
The authors acknowledge the financial support of the São Paulo Research Foundation (FAPESP), Grant #2022/05848-7.
Edited by
-
Responsible Editor:
Andréa Machado de Almeida Mattos, Universidade Federal de Minas Gerais (UFMG), Belo Horizonte, Minas Gerais/MG, Brasil. Lattes: http://lattes.cnpq.br/7749222257907067, ORCID: https://orcid.org/0000-0003-3190-7329, e-mail: andreamattos@ufmg.br.
Review 1
About the reviewerReview 1
The article “The Linguistic Profile of Horror/Suspense/Mystery Movie Dialogues Across Time” provides an exploration into the intricate relationship between language and genre-specific cinematic storytelling. Through a robust corpus linguistics approach and nuanced narrative analysis, the authors bridge significant gaps in the literature, offering a rich understanding of how linguistic choices shape viewer experience in horror, suspense, and mystery films.
Strengths:
1. Comprehensive Corpus Analysis:
The study uses the North American Movie Corpus (NAMC), comprising over 90 years of cinematic dialogues across 720 films. The longitudinal scope, from the 1930s to the 2010s, allows for an insightful diachronic analysis, making the findings robust and relevant.
2. Genre-Specific Linguistic Features:
The identification of lexical bundles and their narrative functions stands out as a major contribution. For instance, speech act verbs, cognitive verbs, and adverbials emerge as critical tools in crafting suspense and horror. These linguistic features demonstrate how dialogues engage viewers cognitively and emotionally, aligning with plot structures.
3. Integration of Narrative and Emotional Dimensions:
The study adeptly connects linguistic findings with narrative theories, highlighting the symmetry or asymmetry between character emotions and viewer responses. The emphasis on the cognitive and physical dimensions of emotions, as described by Carroll, adds depth to the analysis.
4. Practical Implications for Filmmakers:
The article not only advances theoretical understanding but also provides actionable insights for scriptwriters and directors. By detailing how linguistic structures contribute to suspense, mystery, and horror, the study offers a toolkit for crafting compelling dialogues that resonate with audiences.
5. Illustrative Examples:
The detailed analysis of iconic scenes from films such as *Psycho* and *The Ring* enriches the discussion, providing concrete illustrations of how linguistic elements interact with plot and character development to evoke fear, suspense, and curiosity.
Areas for Further Exploration:
1. Cross-Cultural and Multi-Language Comparisons:
While the study focuses on American cinema, expanding the analysis to non-English films (Brazilian ones, for instance) could uncover universal versus culture-specific linguistic strategies in these genres.
2. Interplay Between Dialogue and Non-Verbal Elements:
Incorporating an analysis of non-verbal cues such as intonation, pauses, and visual elements could provide a more holistic understanding of how suspense and horror are linguistically and audiovisually constructed.
3. The Role of Production Contexts:
Exploring how historical production constraints, such as censorship, shaped the evolution of dialogue across decades could offer additional layers of insight.
Conclusion:
“The Linguistic Profile of Horror/Suspense/Mystery Movie Dialogues Across Time” is a study that skillfully integrates corpus linguistics, narrative theory, and genre analysis. Its findings are both academically rigorous and practically valuable, making it a pivotal resource for scholars and filmmakers alike. The article’s contribution to understanding the role of language in cinematic storytelling is both timely and transformative, inviting further exploration into the intersection of linguistics and film studies.
- peer review recommendation: accept
Review 2
About the reviewerReview 2
Theoretical and Conceptual Coherence in the Abstract and Full Text
The abstract presents a clear summary of the study’s objectives, methods, and findings. It follows an academic structure that aligns with expectations in corpus linguistics, and the research questions are connected to the broader discussion on linguistic patterns in cinematic dialogue. However, the abstract does not make it explicit how this research builds upon or differentiates itself from previous studies on film dialogue and corpus linguistics. Stating the distinct contribution of this study would help clarify its originality.
The full text is grounded in a well-referenced theoretical framework, particularly in discussions of linguistic features in movie scripts. Nevertheless, the connection between theory and empirical analysis could be stronger. Some sections introduce extended explanations that could be refined for better readability. Additionally, technical terms such as “lexical bundles,” “multi-dimensional analysis,” and “collocational patterns” are appropriate within the field but assume prior knowledge. Brief explanations or references to foundational corpus linguistics studies would make the text more accessible to a broader academic audience.
Coherence Between Sections
The article follows a logical structure, with each section contributing to the study’s progression. The introduction effectively sets up the research problem, though its alignment with the study’s objectives could be clearer. The literature review provides important context, but does not always link back to the research questions outlined in the abstract. Strengthening these connections would improve cohesion.
The methodology section includes detailed descriptions of data selection and analysis, yet the reasoning behind some methodological choices is not always explained. For instance, the frequency threshold for identifying lexical bundles could be justified more explicitly. Additionally, a summary that connects specific methods to key findings would strengthen the articulation between sections.
The conclusion synthesizes the main results well, but it could more directly address the research questions and theoretical framework introduced earlier. A short discussion on how the findings contribute to corpus linguistics and the study of cinematic discourse would reinforce the study’s broader significance.
Consistency Between Methodological Procedures and Research Objectives
The methods are well-defined and appropriate for the study’s objectives. The use of corpus-based analysis to examine linguistic variation in movie scripts is well justified. However, the selection criteria for some analytical categories, such as the classification of linguistic dimensions, could be clarified. A brief explanation of how these criteria were established-whether based on prior research or empirical observations-would enhance transparency.
While the findings are systematically connected to the research questions, the discussion could establish clearer links between the results and the theoretical framework. Showing how these findings compare to previous studies on scripted dialogue or corpus-based genre analysis would provide a stronger foundation for interpretation. The tables and figures contribute to the study’s clarity, but their relevance should be discussed more explicitly in the text.
Relevance and Originality of the Study
The research contributes to an important area by examining linguistic patterns in movie scripts using a corpus-based approach. The focus on horror, suspense, and mystery films offers a perspective that is not frequently explored in linguistic studies. However, the abstract and discussion sections could more clearly highlight the study’s unique contributions. Stating explicitly how this research advances prior studies on corpus linguistics and movie dialogue would strengthen its originality.
Additionally, while the study focuses on linguistic variation, a discussion of how these findings might be applied-whether in scriptwriting, audiovisual translation, or AI-generated dialogue-would add practical relevance.
Contribution to the Field
This research adds to corpus linguistics by applying quantitative methods to analyze scripted dialogue, an area that remains underexamined. The integration of lexical bundles and multi-dimensional analysis offers useful insights into how film dialogue differs from spontaneous conversation. However, the discussion could further explore the potential impact of these findings beyond descriptive analysis. Addressing how this research could inform audiovisual studies, computational linguistics, or language modeling would enhance its relevance.
Additionally, while the study presents well-structured data and analysis, it could provide more discussion on how linguistic variation in movie scripts has evolved over time. Engaging with prior studies on historical shifts in cinematic language would add depth to the discussion.
Recommendations for Improvement
-
Clarify the study’s distinct contribution - Specify how this research extends or refines previous work on corpus linguistics and cinematic discourse.
-
Refine the methodology section - Justify key methodological choices, particularly the frequency threshold for lexical bundles and the classification of linguistic dimensions.
-
Strengthen connections between theory and results - Ensure that the discussion explicitly ties back to the theoretical framework introduced in the literature review.
-
Improve coherence between sections - Establish stronger links between the research questions, methods, and findings.
-
Discuss broader applications - Expand on how these findings might inform applied fields such as scriptwriting, language technology, or audiovisual translation.
-
Integrate tables and figures more effectively - Provide clear references to visual data within the text to guide interpretation.
Final Recommendation:
Accepted with mandatory revisions
The study presents a relevant and well-structured analysis of linguistic variation in cinematic dialogue, contributing to corpus linguistics and discourse studies. However, revisions are necessary to clarify the study’s contribution, streamline methodological descriptions, and reinforce connections between theory and findings. Addressing these points will improve the overall coherence and impact of the paper.
- peer review recommendation: accept
