Empirical testing 1

EMPIRICAL TESTING OF AUTEUR THEORY VIA CONTENT ANALYSIS:

A STUDY OF JANE CAMPION FILMS

Patricie Janstova

Bachelor of Arts and Sciences in Communication

Cleveland State University

May, 2003

Submitted in partial fulfillment of requirements for the degree

MASTERS OF APPLIED COMMUNICATION THEORY AND METHODOLOGY

at the

CLEVELAND STATE UNIVERSITY

December, 2006 Empirical testing 2

CHAPTER I

LITERATURE REVIEW

The film of tomorrow appears to me as even more personal than an individual and autobiographical novel, like a confession, or a diary. The young filmmakers will express themselves in the first person and will relate what has happened to them. It may be the story of their first love or their most recent; of their political awakening; the story of a trip, a sickness, their military service, their marriage, their last vacation . . . and it will be enjoyable because it will be true, and new. . . The film of tomorrow will not be directed by civil servants of the camera, but by artists for whom shooting a film constitutes a wonderful and thrilling adventure. The film of tomorrow will resemble the person who made it, and the number of spectators will be proportional to the number of friends the director has. The film of tomorrow will be an act of love. — François Truffaut (Le Cain, 2001, p. 1).

Introduction

Film is a fascinating phenomenon. It is not a creation of a single person, rather a

creation of a team. However, since the infancy of movies, individuals have been

claiming authorship of a product. It started as early as the beginning of the 20th century,

when German writers claimed authorship to the films created based on their scripts. Who

can claim authorship? Is there such a thing as individual author in film production?

Diving into the literature dealing with and answering such questions, however, leaves one unsatisfied with further questions. Yes, there is such a thing as authorship and it is usually the director who stamps each of his or her films with a unique imprint or signature. But, how do you recognize the director’s signature? Can you rely on the subjectivity of film critics? Are you satisfied when a film critic tells you who is and who is not considered an auteur (the author)? Empirical testing 3

To answer these questions, I am attempting to content analyze a body of work of

one director, Jane Campion, through the auteur theory lens. She has been granted auteur

status by many critics and scholars (Blonski, Creed, & Frieberg, 1987; Delaney, Dupin,

Gee, Gordon, Harrow, Khan, et al. 2002; Mellencamp, 1996; Nelmes, 1999; Polan, 2001;

Schröder, 2004). Jane Campion, Polan (2001) states, is one of the few female directors

who can be considered an auteur. “. . . it is the disturbances in her work – the

divergences; the dispersions; the tensions, for instance, between quirky humour, a making

strange of the familiar, and an interest in the ambiguous, even that which is

uncomfortable and which makes the viewer uncomfortable – that means that to study her

is to study the cinema differently, to rethink the very terms of analysis of the film

director” (2001, p. 167).

An elaborate coding scheme was created to objectively test work of Jane

Campion and compare it to other non-Campion films, work of other directors. Will she be

recognized as an auteur without the subjective eyes of film critics? This is the core

question of this study.

Auteur Theory

The Beginning of Auteur Theory

Today, the vast majority of people would agree that film is an art form. How we Empirical testing 4

think about films, however, is not how it was thought of five decades ago. The auteur

theory changed how film was viewed in terms of authorship, and because of that, it is one

of the most known and most criticized theories in film history.

As mentioned earlier, as early as in the 1920s, the question of authorship arose. In

the 1940s it became a popular debate among the French film critics with Andre Bazin and

Alexandre Astruc’s beliefs that director should be considered the true author of a film

(Bazin, 1967; Bazin, 1971; Thompson & Bordwell, 2003). As quoted in Thompson and

Bordwell (2003), Alexandre Astruc wrote in his essay The Birth of a New Avant-Garde:

La Camera-Stylo: “The filmmaker-author writes with his camera as a writer writes with

his pen” (p. 415).

The Fifties and Romantic – Auteurism

The most influential work on film authorship came out of the early fifties,

particularly the newly founded French journal Cahiers du Cinema. The writers for the

Cahiers formed a polemic – la politique des auteurs, or the policy of looking at films in

terms of authors, as Astruc suggested in his essay. The next step in the development of

auteurism was A Certain Tendency in the French Cinema, written by one of Cahiers’ own members Francois Truffaut. In his article, he attacked the current French industry, which he called “The Tradition of Quality,” for making “essentially scenarists’ films”

(1976, p. 225). Truffaut wrote that the French directors of that time only added visual pictures to the script, with no attempt to add their own creativity, their own talent. He also criticized the scriptwriters for not staying true to the novels they adopted, because they thought it had un-filmable scenes. Of course, he believed there was no such thing as Empirical testing 5

an un-filmable scene. He then continued to critique the scriptwriters of having no

originality:

There are scarcely more than seven or eight scenarists working regularly for the French cinema. Each one of these scenarists has but one story to tell, and since each only aspires to the success of the “two greats” [scenarists], it is not exaggerating to say that the hundred-odd French films made each year tell the same story . . . (Truffaut, 1976, p. 232).

He also attacked the scriptwriters with a question:

What’s keeping [the filmmakers and scriptwriters] from making, from one day to the next, intellectual films, from adapting masterpieces (there are still a few left) and, of course, adding funerals, here, there and everywhere? Well, on that day we will be in the “Tradition of Quality” up to the neck and French cinema, with rivalry among “psychological realism”, “violence”, “strictness”, “ambiguity”, will no longer be anything but one vast funeral that will be able to leave the studio in Billancourt and enter the cemetery directly – it seems to have been placed the next door expressly, in order to get more quickly from the producer to the grave- digger” (Truffaut, 1976, p. 235).

The Cahiers editorial critics followed Truffaut’s critique of the current French filmmakers and argued against the directors’ literary films. The magazine argued that even though film production was an industrial creation, it shouldn’t have stopped the director from introducing his own creativity and adding his personality into the final product. They praised individuality, visual distinction, and consistent thematic focus

(Buscombe, 1981; Caughie, 1981; Cook, 1986; Miller & Stam, 2000; Sarris, 1962; Sarris,

1968; Sarris, 2003; Stam, 2000; Stam & Miller, 2000; Thompson & Bordwell, 2003;

Truffaut, 1976; Wexman, 2003). They argued that even though filmmakers in the United

States are working under strict rules of the studios, individual styles and personalities

were evident in films by directors, such as Alfred Hitchcock, Orson Welles, John Ford,

and Howard Hawks. They argued those directors are true authors of the film, they are the

auteurs (Cook, 1986; Truffaut, 1976). Empirical testing 6

The Romantic auteurism of la politique des auteurs was introduced to the United

States by film critic Andrew Sarris in his 1962 work (Croft, 1982; Dick, 2005; Sarris,

1968; Sarris, 1962; Thompson & Bordwell, 2003; Wexman, 2003; Wollen, 1969). Sarris is responsible for the loose translation of the French term la politique des auteurs as auteur theory. In his article, he said that authorship is “primarily a critical device for recording the history of American cinema, the only cinema in the world worth exploring in depth beneath the frosting of a few great directors at the top” (Sarris, 1962, p. 6).

With the translations of the la politique des auteurs he attached his own ideas.

Sarris took Truffaut’s criticism of the French cinema and admiration of certain American directors a step further. He called the American cinema superior to any other cinema, and believed it was the only one worth studying due to the fact that even though working under strict rules of the Hollywood studios, a director’s style can still be acknowledged

(Sarris, 1962; Sarris, 1976). Sarris, however, failed to support his statements with evidence. No comparisons to other cinemas ware made, or even attempted.

Truffaut, Sarris, and the auteur theorists believed that over time, if there is a recognizable stylistic and thematic personality in a works of a certain director, he or she is to be considered an auteur (Hayward, 1998; Sarris, 1968; Sarris, 1976; Stam, 2000).

Directors’ thematic personalities come alive through recurring themes - the “dominant idea[s] made concrete through its representation by the characters, action, and imagery of the film” (Rabiger, 1997, p. 517).

Sarris believed auteur theory had three principles. First, “the technical competence of a director as a criterion of a value” (1962, p. 7); second, the identifiable personality of a director through his body of work; third, the inner meaning, “the ultimate Empirical testing 7

glory of the cinema as an art… the tension between a director’s personality and his material” (p. 7). It can be thought of as the director’s vision of the world, and his attitude

toward life. Dick (2005) summed Sarris’ three principles of auteurism as technical

competence of the auteur; auteur's personality that brings in recurring themes, which

become his or her signature; and the tension between the auteur and his material, which

may not seem obvious but after analyzing the auteurs work it will come to the surface.

In 1968, Andrew Sarris wrote: “ . . . the notion of quality is difficult to grasp apart from the context of quality. Comprehension becomes a function of comprehensiveness.

As more movies are seen, more cross-references are assembled. Fractional responsibilities are more precisely defined; personal signatures are more clearly discerned” (p. 19). He then ranked the American directors with the most important

(auteurs) in the first category called the “Pantheon Directors” through ten other categories: “The Far Side of Paradise,” “Expressive Esoterica,” Fringe Benefits,” Less than Meets the Eye,” “Lightly Likable,” “Strained Seriousness,” “Oddities, One-Shots, and Newcomers,” “Subjects for Further Research,” “Make Way for the Clowns,” and

“Miscellany.” In the Pantheon he included “ . . . the directors who have transcended their technical problems with a personal vision of the world . . . They were also fortunate enough to find the proper conditions and collaborators for the full expression of their talent” (p. 39). Alfred Hitchcock, Howard Hawks, Charlie Chaplin, D.W. Griffith, Jean

Renoir, Fritz Lang, and Orson Welles were some directors included in the Pantheon. In the Far Side of Paradise category were those directors “. . . who fall short of the Pantheon either because of a fragmentation of their personal vision or because of the disruptive career problems” (p. 83) such as Frank Capra, George Cukor, Cecil B. DeMille, and Empirical testing 8

Samuel Fuller. In the Less than Meets the Eye category he included directors such as

John Ford, David Lean, Billy Wilder, and William Wellman. He wrote “These are directors with reputations in excess of inspirations. In retrospect, it always seems that the personal signatures to their films were written with invisible ink” (p. 155). In the Strained

Seriousness category, Sarris believed “. . . are talented but uneven directors with the mortal sin of pretentiousness. Their ambitious projects tend to inflate rather than expand”

(p. 189). Richard Brooks, Stanley Kubrick, John Sturges, and Robert Wise were listed in that category. Sarris believed that:

. . . the auteur critic does not look to the cinema of completely original artistic experiences. The cinema is both a window and a mirror. The window looks out on the real world both directly (documentation) and vicariously (adaptation). The mirror reflects what the director (or other dominant artist) feels about the spectacle. Modern cinema tends to fog up the window in order to brighten the reflection. It would seem that a theory that honored the personality of a director would endorse a cinema in which a director’s personality was unquestionable supreme (p. 31).

The Late Sixties, Early Seventies and Structuralist - Auteurism

In the sixties, a rethinking of auteur theory was needed due to the historical and political changes in France, with the student uprising, and spreading of Marxistic ideas

(Cook, 1986; Stam, 2000; Thompson & Bordwell, 2003; Wexman, 2003). The British film journals introduced theoretically more evolved and scientific auteurism - structuralist-auteurism (Stam, 2000; Wexman, 2003). In France, the new, radical, more political and intellectual Cahiers critics also moved away from romantic-auteurism by conceptualizing “authorship as an unconscious process . . . [in which an auteur’s] presence could subvert the surface meanings of a given filmic text” (Wexman, 2003, p.

6). Empirical testing 9

The idea of the author being the only and sole creator of meaning has been

attacked. The Marxist movement took the emphasis of the individualism and put it to collectivism. Instead of destroying the theory, it has been remodeled (Cook, 1986) and reorganized with structuralism, a method also used by other disciplines such as linguistics and anthropology. Rather than simply granting the auteur his or her authorship of the

film, structuralist-auteurist theory searched for familiar structures such as characters, or plot that occurred in more films through group and team effort (Corrigan & White, 2004).

Structuralism came out of semiotics, mainly the theories of Swiss scholar

Ferdinand de Saussure, and an American scholar Charles Sanders Peirce. The growing popularity of structuralism was the result of the works of French literary critic Roland

Barthes, and French anthropologist Claude Lèvi-Strauss. The basis for structuralism is that all language can be ordered and understood, according to Saussure (Hayward, 1998;

Stam, 2000; Stam, Burgoyne, & Flitterman-Lewis, 1992; Stam & Miller, 2000; Wollen,

1969). He looked at language and its universal structures or systems shared by a community of speakers (langue), rather than the use of language or individual speech itself (parole) for example utterances made by speakers (Hayward, 1998; Stam, 2000;

Stam, Burgoyne, & Flitterman-Lewis, 1992; Wollen, 1968). According to Saussure, language is made up of signs which have two components: signifier and signified. The signifier being the actual shape of the word, the sequence of the letters, and the signified

being the concept or image that we create in our mind when we hear or read the signifier

(Caughie, 1981; Hayward, 1998; Stam, 2000; Wollen, 1968). Stam, Burgoyne, and

Flitterman-Lewis (1992) defined structuralism as a “ theoretical grid through which

behavior, institutions, and texts are seen as analyzable in terms of an underlying network Empirical testing 10

of relationships, the crucial point being that the elements which constitute the network

gain their meaning from the relations that hold between the elements” (p. 18).

Wollen argued that the work of an auteur is not only formal, but it has semantic

dimensions and that two main schools of auteurs were known: “those who insisted on

revealing a core of meanings, of thematic motifs, and those who stressed style and mise-

en-scène” (1969, p. 78). Auteur-structuralism is looking at the auteur as a critical

construct rather than a human being.

This new approach to auteurism was positively accepted by auteur advocates, as it

was believed that structuralism and its empirical approach would bring “an objective

basis for the concept and counter the romantic subjectivity of auteur theory” (Hayward,

1998, p. 24). Allan Lovell (1977) talked about the possibility of an analytic apparatus that

would force critics to be more objective, and it would uncover the basic structures, and

reoccurring themes in the artist’s work. The Cahiers du Cinema editors applied

structuralism to the work of John Ford and his fixation with President Lincoln in their article “John Ford’s Young Mr. Lincoln” (1976). They wrote that Ford’s films dealt with

“particular aspect[s] either of Lincoln’s synthetic personality or his complex historical

role; he thus appears to be a sort of universal referent which can be activated in all

situations” (p. 501). Peter Wollen (1968) analyzed films of John Ford with the

structuralist method as well. He identified many pairs of opposites in Ford’s films such as

East versus West, civilized versus savage, book versus gun, and garden versus

wilderness, as Ford’s themes. These antinomies and their relationships, he noted, were often reversed in his films, but always present. Wollen wrote, “if the concept of

‘language’ is to be used it must be used scientifically and not simply as a loose, though Empirical testing 11

suggestive, metaphor” (p. 116).

Peter Wollen’s structuralist – auteurism approach to film analyses was moving

toward more systematic and scientific methods to film studies. However, with the rise of

post-structuralism in the 1970s, the interests of the critical theories shifted away from this fascinating work of structuralist – auteurism to debates about ideology, gender and the psychological orientations of the spectator. This study returned to structuralist – auteurism and took it a step further with empirical approach to auteurism. Table 1 illustrates the development of auteur theory as outlined here.

Table 1

Development of Auteur Theory 1950s Romantic-auteurism Author as an expressive Individual - producer of meaning Thematic and stylistic properties read from the films 1960s Auteur-Structuralism Author as constructed from the films Set of structures identifiable within a body of films 1970s Auteur-Post-Structuralism Author as constructed from the films & a subject position within the film

Introduction of spectator - pleasure and ideology

Auteur Theory Revisited

In later articles, Sarris wrote that if he could go back and reformulate the auteur

theory, he would put more stress on the author’s filmic style differences rather than the idealistic suffering of the artists. He believed auteurism is a tendency rather than a theory, Empirical testing 12

“more mystique than a methodology, more an editorial policy than an aesthetic

procedure” (Sarris, 2003, p. 29).

Peter Wollen (1969) believed the survival of auteurism is essential, due to what

Nowell-Smith called a “structural approach” to film criticism. Critics will always seek

the motifs and structures that will differentiate one body of work from another, one

director from another. He also believed that cinema studies, examinations, analyses, and

evaluations have hardly begun. He found it hard to believe that there were only three books on Alfred Hitchcock and only one book on Howard Hawks in 1969. To date, there are at least hundred books on Hitchcock, and dozens of books on Hawks.

The fact that there are many recent studies, examinations, and analyses concerned

with auteur theory, or the authorship of a film (such as Allen & Lincoln, 2004; Buse,

Nuria & Andrew, 2004; Goss, 2004; Gurney, 2006; Hart, 2003; Kathi, 1994; McDonald,

2000; Mosher, 2004; Maule, 2002; Nacashe, 2005; Nystrom, 2004; Petrova, 2006; Polan,

2001; Poynor, 2001; Raphael, 1997; Ray, 1991; Rossenbaum, 1991) leads us to believe

that Wollen’s thoughts on auteurism and film criticism seem to be accurate. Auteurism is

not dead, it is indeed well and alive.

The controversial views on authorship that started almost at the birth of the idea

are continued today. Many scholars, theorists, and critics argue that film is a complex art

form created by a collaboration of director, cinematographer, editor, writer, and/or actor

(Caughie, 1981; Cook, 1986; Dick, 2005; Stam, 2000). Richard Corliss (1974) in his

critique of auteurism advocated the idea of screenwriters being the authors of films, not

just the directors. To criticize Sarris he sarcastically created a screenwriter’s Parthenon,

to oppose Sarris’ Pantheon. He also believed that a screenwriter’s work should be Empirical testing 13

evaluated as a whole, and that his or her style and personality will surface, just as will the

style and personality of the director.

Even after all the disputes, scholarly articles, popular culture articles, and books

that are still being written about the directors as main authors of their films, directors are

still being studied and recognized as the key figures to a success of a film because of the

personal style each auteur/director possesses, the imprints they leave on their films, the

personalities that influence their work.

Many critics, scholars, researchers, and theorists are still classifying films of, for example, Alfred Hitchcock as Hitchcockian films. It is due to the fact that, indeed, a body

of work of one director will uncover and identify the director’s style (Caughie, 1981;

Hayward, 1989; Stam, 2000). To give an example, Schröder (2004) identified a few imprints that could be found in most, if not all, of Hitchcock’s films: First and foremost the suspense theme, then the mistaken identity of a character, female characters with mysterious psychological issues that are cured even more mysteriously, the obsession with blond women, unforgettable point-of-view shots, inventive camera angles, signature cameos, the memorable use of props and many more.

There is an unimaginable amount of different themes, narrative constructions, and

production techniques that could become signatures and identifiers of a director. To illustrate the wide variety of possibilities several contemporary directors and their styles and obsessions were reviewed in Tasker (2002): Tim Burton - love for horror and fantasy, plain weirdness of main characters, technical experimentations with animation and visual effects, the protagonists are villains, outsiders with complex personalities, the

protagonists have almost always very pale skin, themes of exclusion and inclusion, Empirical testing 14 working with an actor (i.e., Johnny Depp); David Cronenberg – the physical and the psychological is usually intertwined, the protagonist is mostly a male character with a bizarre name, themes of science, disease, death in literary metaphor, bodily and psychic mutations, hostility towards women, and womb envy; Allison Anders – multiple protagonists, teen anxiety, gangsters, bio-pictures, emotional intimacy and heterosexual love; and Peter Weir – very nationalistic (Australian) rather than international, avoidance of sexuality, no vivid sex scenes, however erotic desire is shown.

The present study attempted to uncover the personal style and distinctiveness of director Jane Campion and her films through identifying the reoccurring themes, obsessions, personal visions, narrative constructions with characters, and production techniques used throughout her body of work. Campion was chosen for this initial empirical test of auteur theory because of her auteur status, the manageable amount of her body of work, as well as the researcher’s interest in her films.

Jane Campion

Jane Campion was born April 30th, 1954, in Wellington, New Zealand. In the

1960s her father gave her her first camera, a Super 8mm; she started to make home movies. As an undergraduate, she studied Anthropology in her hometown. In 1977, after attending the Sydney film festival, Jane Campion decided to make films on her own.

First, she made a short film Tissues with no close up shots (due to the lack of knowledge about filmmaking). However, after enrolling in the Australian Film and Television

School in 1981 (Redding & Brownworth, 1997), she became the master of astonishing Empirical testing 15 close up shots, meaningful extreme close up shots that speak to audiences even without dialogue.

In 1982 she won the Palme d’Or for best short film (Peel) at Cannes. Redding and

Brownworth (1997) identified some of Campion’s early themes in Peel as study of claustrophobia, family road trips, and family issues; those themes are to be found throughout her body of work. Peel’s simple but brilliant story is: a family takes a short car trip and argues the whole time.

In 1983 she made A Girl’s Own Story – again a short film that dealt with family dysfunctions. It is a story of a girl whose father takes her and his mistress to dinner together. In this film, adolescent sexuality is also depicted. Redding and Brownworth

(1997) also identified Campion’s early themes of “seeing the truth versus stating it, longing to belong, and oppression of children by their families” (p. 181). The same year, she made her third short film, Passionless Moments, a black and white series of vignettes of regular Australian life.

In 1986 Campion made her first feature film (made for Australian television) Two

Friends, a film about two teenage friends – one conservative and one punk rocker, whose paths move away at certain points in their lives and in the story. Not many film analyses or film critiques are written about this film, due to the fact it was a TV based film. In many sources, this film is not listed as Campion’s first feature film.

Jane’s Feature Films

Sweetie (1989). Sweetie is considered Campion’s first feature film. It is, as Walsh

(1994) defined it, “a portrait of a disturbed girl and her family” (http://www.wsws.org.), Empirical testing 16

as well as a story of sibling rivalry between two sisters Kay and Dawn (nicknamed

Sweetie), one of whom is madder than the other. This film was first unsuccessful at the

1990 Cannes Film Festival – Campion even got booed by the audience who stayed to see

the whole film. Later however, it gained a lot of positive reviews and the Village Voice

pronounced it one of the best films of 1990.

Mellencamp (1996) compared the flashbacks in Sweetie to a world according to

Sigmund Freud and the Brothers Grimm, but with a female point of view. Schröder

(2004) identified the experimental feel of Campion’s first feature film with the perfect,

almost seamless but very original fading from reality to dream sequences and vice versa;

he called it “a virtuoso sequencing” (p. 256).

There were many secrets, many ugly secrets present in this film. The father’s sexual abuse of the younger sister, mentally unstable Sweetie, was not deeply exploited; it was shown on the screen, but denied by the family members. McHugh (2001) pointed

out some questions that a viewer might have: “Did Sweeties father seduce her as a child?

How has she come to be ‘mental’? What really happened in this family?” (p. 199). The

film, as Howe (1990) wrote, is about repairing broken relationships among family

members, with the taboo theme of familial dysfunction. Mellencamp (1996) wrote

“Campion is not interested in causes or in explanations” (p. 173), she just raises the

questions, or points to issues without further analysis.

Hinson (1990) saw Sweetie as a horror film with comedy traits, and extraordinary use of “never-before-seen angles” . . . “It’s about family life as Kafka might have viewed

it” (www.washingtonpost.com). Howe admired the use of refreshing camera angles in

this film. He identified five themes – abnormality, love, individuality, family and sisters. Empirical testing 17

He wrote it is a story of irrational sibling relationship, imperfect parenting, and repairing of a romantic relationship. He also recognized Sweetie’s need for psychiatric help, which is not further explored among the family member. Wexman (1999) identified another

Campion’s recurring theme of presence of angels in her films, either visual or mentioned.

Polan (2001) identified a lot of the close up and extreme close up shots used in

Sweetie as a tool to show the isolation of the characters. Dawn’s constantly looking into the camera violated the fictional on-screen world, which made the viewers uncomfortable

in the privacy of their home, or theater seats.

An Angel at my Table (1990). This is a film based on an autobiographical novel from Janet Frame, a New Zealand writer whose oddness and shyness put her to a hospital

misdiagnosed with schizophrenia and hospitalized for eight years. Gillett (2000a) talked about this film as Janet’s story of suffering: “from her childhood onwards she experiences grief, poverty and social stigmatization, deprivation, fear and isolation,

physical pain, manipulation” (www.sensesofcinema.com/contents/00/10/angel.html), told

in a three-part story which is made up of series of anecdotes, Polan (2001) writes.

Mellencamp (1996) talked about this film as being Campion’s fairytale. Her style of storytelling is very close to such experience: The extraordinary images of New

Zealand, the extreme close up shots of children’s eyes and faces, the “loonies”, all the children sleeping in one bed, the children who died.

Henke (2000) stated that Janet Frame’s autobiography was not easily adaptable to the filmic world. Polan (2001), however, believed that the strength of this film was in the visual and stylistic aspect. The frequent use of close up shots and extreme close up shots helped to avoid Campion’s earlier style of strange compositions, or “playful composition Empirical testing 18

patterns” p. 118). Henke (2000) believed that her long scenes with loose structure might

lose readers’ patience easily. She also wrote that there is not enough information on her hospitalization, which would be much more interesting to the reader than the long descriptions of dull farming tasks. In her cinematic translation, Jane Campion, fortunately, chose to put more weight on Janet’s hospitalization.

As Henke noted, Frame herself said that all of the experiences she wrote about might not be true, she might have mixed them up, borrowed from other characters and therefore we don’t know what a real life story is and what a tale is.

This Campion film, as Henke (2000) stated, was created with “visual tableaux, domestic vignettes, lyrical and symbolic fragments, and tacit social or biographical allusions” (p. 653). It dealt with personal isolation and suffering, victimization, and loneliness. Henke believed other filmmakers probably would have simplified her story for the sake of the viewer. Campion, however, chose not to and stayed true to Frame’s autobiography and did not settle for a classic form of bio-picture. She also chose to show, in great detail, the shocking truth about institutional abuse in New Zealand.

Redding and Brownworth (1997) as well as Mellencamp (1996) identified themes of isolation, need to fit in, and family dysfunctions as also seen in her short A Girl’s Own

Story.

“Campion used many close shots to emphasize her [Janet’s] grubby daily existence, punctuating the scenes with breathtaking landscapes that suggest her lyrical vision” (Thompson & Bordwell, 2003, p. 666).

The Piano (1993). In this film the protagonist Ada leaves Scotland to go to New

Zealand with her daughter Flora to fulfill a marriage arranged by her father. Ada is a Empirical testing 19

mute but we never really understand why, she communicates with the world through her

daughter Flora and her piano. She has a hard time fitting in the society and living with /

loving her husband. He makes it harder for her to love him by trading her piano for a

piece of land to Baines, a European living among the natives. Baines, however, wasn’t

after the piano, he was after Ada. Through the trade, he got piano lessons where he only

listened to Ada play, and slowly sells the piano back to Ada for fetishistic favors.

Eventually, Ada and Baines fall in love, causing a clash between Ada and her daughter.

As Redding and Brownworth (1997) indicated, this film was well received as a

feminist piece, due to its message: Women shouldn’t be looked at as men’s accessories and possessions; they are not attached to their husbands, women have their own needs,

feelings, and desires and should be able to act upon them, just as Ada did.

Campion gave us a scene where Ada used her husband for sexual pleasures, she

didn’t let him touch her, the situation reversed – Ada was the one in control, she was the

woman of power (Sharp & Gillard, 2004). Female desire and female lust were explored

in this film as Hopgood noted (2002). Sharp and Gillard wrote that Ada’s muteness

categorized her as the obedient sexual object in the patriarchal society, causing her self-

imprisonment, but Ada gained her power and respect . Kelly (1996) also identified the

themes of isolation and imprisonment as well as themes of family dysfunction and

sexuality, power, and passion.

Campion introduced the clash between sexual love and maternal love in this

gothic fairytale (Bentley 2002; Mellencamp, 1996). Bruzzi (1999) posed the question of

sexuality and gender roles by describing the “masculine/feminine; distance/nearness,

looking/touching” (p. 98) roles of Ada and her daughter, and her husband and Baines. Empirical testing 20

Dalton and Fatzinger (2003) quoted Colbert “The Piano is not a simple women’s

film about woman’s past, but rather a cryptic and evocative exploration of how women’s

sexuality, clothes, and lives interconnect” (p. 34).

Campion’s first feature films were funded with government grants that guaranteed

her full creative control in pre-production, production, and post-production phases of

filmmaking. Rather than giving up full creative control with The Piano, Jane Campion

turned down the offers of American investors who wanted Hollywood to have the final

control, and turned to a French production firm that assured her full control over the

production (Redding & Brownworth, 1997).

Portrait of a Lady (1996). It is a historical story, a story of young Isabel Archer,

an American woman who tries to find her identity in Europe. However she finds herself

falling in love with shady and dishonest Osmond, who used his ex-lover to trap Isabel,

successfully. Isabel marries Osmond, and loses her independence and freedom.

Osmond, the masculine and over sexed character awakens Isabel’s sexuality but at

the same time, he forces her to lose her spirit, and freedom (Gordon, 2002). Hopgood

(2002) wrote about the control Osmond exercised over the main female characters in

Portrait of a Lady, controlling not only his daughter Pansy, but also Isabel and Madame

Merle. Gordon (2002) wrote that to illustrate his power over the women and especially

Isabel, Campion shot the disturbing scenes with overhead shots and canted angles, such

as the domestic violence scene which made us feel like peeping Toms, as we felt earlier

in her “bedroom fantasy” scene. Empirical testing 21

Hopgood (2002) noted a similarity of the perverse father in Sweetie with the

perversity of the scene where Pansy sits on her father’s lap in this film. The feeling one

gets watching both scenes is very similar.

This film has a similar message to The Piano; it warns women to protect their

independence, and illustrates their quest for identities (Corrigan & White, 2004). Redding

and Brownworth (1997) wrote: “ . . . it is a parable for modern women, telling them to guard the independence that was so hard won and so easily lost” (p. 184). Keough (1999), as well, identified the theme of isolation and loss of freedom of the female protagonist.

Polan (2001) wrote about the beautifully shot surrealism scene of the talking beans in this film being very original. She also talked about the opening scene where modern women are interacting with each other and the camera. As found in Polan,

Shriver believed that this film, as well as other Campion films, is giving more importance

to woman-to-woman relationships, rather than woman-to-man. He believed that the much talked about opening scene of contemporary women talking about first kisses, and romantic relationships, indicated the significance of the complexity of female-to-female relationships, which would be presented in the film. Polan (2001) added that Campion’s recurring theme is the dysfunctional nature of woman-to-woman relationships. Shaw

(2002) identified the loss of power and independence when Isabel married Osmond willingly and argued that Isabel from the novel is a different Isabel from the film.

Holy Smoke! (1999). This film is a story of Ruth a young Australian who is on the hunt for the truth. Ruth, the female protagonist, as Isabel from Portrait of a Lady, is seeking her purpose abroad (Gillett, 2000b). She travels to India where she thinks she has found the truth in an Indian guru named Baba. Her family finds out, and tricks her to Empirical testing 22

come back home just to deprogram her from her newly found religion. They hire an

American professional “cult exiter” named PJ to help their daughter. PJ promises that in

three days she will be back to normal. After Ruth’s and PJ’s deprogramming, Ruth

returns to India.

Holy Smoke! is filmed in the Australian outback, as Sweetie. However as Gillett

(2000b) noted, Holy Smoke! reveals more of the mythic land and pays attention to the

beauty of the land. The landscape, nonetheless, is still secondary to the characters. Sarris

(1999) in his review describes Holy Smoke! as “witty, messy, perceptive, outrageous, incoherent, satirical, bombastic, cruel, lecherous, compassionate, violent, reflective and terminally unstable from one shot to the next” (www.members.tripod.com/filmlover

2/id50.htm).

On visual level, Fisher (1999) suggests that Holy Smoke! is:

this year's richest work, a sublime conglomeration of vivid colours, which beautifully represent the divided cultures that Campion strikingly explores . . . At the same time, as visually hypnotic as the film is, that aspect of Holy Smoke doesn't detract from her detailed sense of character and theme. The film is intrinsically a two-hander, and Campion's skillful casting of Kate Winslet and Harvey Keitel, ensures that her comment on sexual power is compelling. And it is. The characters are both fascinating to watch and listen, and what one sees is never what one gets, which makes Holy Smoke a persuasive film to sit through. The film's exploration of sexuality and the role of men and women become both sexy and funny, erotic and intense, and always intriguing. These aspects of the film are in stark contrast to Campion's droll satire of Australian suburbia and the nature of family, themes she initiated with Sweetie. (www.mem bers.tripod.com/filmlover2/id50.htm).

Some of Holy Smoke!’s themes are “relationship between the sexes, and the problems faced by women who are considered – by themselves and/or by Campion – to

be at odds with or standing apart from society” (Nichols & Walsh, 2000, p. 1). Those

themes are to be found in Campion’s larger body of work as well. Empirical testing 23

In the Cut (2003). According to Fuller (2003), In the Cut feels like a fairytale full

of dreams, hopes, and fantasies as did A Portrait of a Lady. It is Campion’s latest feature

film about Frannie, the female protagonist, who is a strong independent woman, whose

sexual desires are fulfilled by a strong male character – detective Malloy. In the Cut is a

sexually provocative New York noir with a series of murders happening in Frannie’s neighborhood. What makes it so provoking is that the more clues point to Malloy being

the murderer, the more our protagonist is fascinated by him.

Felperin (2003) compared it to the relationship that Ada and Baines had in The

Piano with “a sensuous study of desire and masochism . . .” (p. 38).

Hopgood (2004) identified other Campion trademarks present in In the Cut:

Sibling relationships, whether functional or dysfunctional; the attention to female body; the camera attention to “. . . feather like touch of fingers” (p. 30); the insertion of sepia- toned footage similar to the black and white travelogue from Portrait of a Lady; and the vibrant storytelling from The Piano.

The surreal fantasy sequences where her father courts her mother on ice-skates are very similar to those in Portrait of a Lady and the black-and-white sequences in Sweetie

(Calhoun, 2003; Felperin, 2003). We learn later about the courtship of her four times married father, when she tells her half-sister about the “fairy-tale” proposal of her father to her mother.

Calhoun (2003) talked about the production techniques used in the film such as the opening scene with looser camera style and a lot of hand held camera, which was very similar to the opening scene in Holy Smoke! in India’s New Delhi. Fuller (2003a) pointed out the extreme close up shot at the beginning of the film of a “woman’s head Empirical testing 24 bobbing over a man’s loins . . .” (p. 16), the extreme “pore-revealing” shot of Pauline’s

(Frannie’s sister) face, and again unflattering extreme close up shots of Pauline’s face where we could see the detail of her grubby make up, as well as close up and extreme close up shots of her sweaty body.

A gender switch in In the Cut was similar to one in Holy Smoke!. Frannie cuffed her detective and was sexually aggressive and whereas her lover actually admitted to feeling like a woman (Fuller, 2003a) in Holy Smoke!. Ruth actually transformed PJ into a female, putting make up and dressing him in a red dress.

In an interview, Jane Campion admits to relating to emotional masochism. She also admits to living in a male dominated society, where not very many sexual scenes in films are supposed to be pleasurable to women (Frencke, 2003); she works towards the opposite.

As with the non-traditional bio-picture An Angel at My Table, Campion made another genre rule breaking film with In the Cut. It is not the usual “slasher” film where the killer’s hatred for women stays undisclosed to the audience. It is a hybrid of

Hollywood love stories and thrillers, giving it a film noir feel (Fuller, 2003a). Wexman

(1999) identified recurring theme of the presence of an angel in her films.

Jane Campion’s Trademarks

Across the seven Campion films produced to date, we may identify a number of consistent thematic and formal trademarks. Empirical testing 25

Sexual obsession and female lust. Gordon (2002) called Campion’s theme of sexuality as a “clinical exploration of sexuality” (p. 14); the presence of a masculine, strong, secretive male arouses Isabel’s sexuality, as well as the young American for Janet,

Baynes for Ada, PJ for Ruth, and Malloy for Frannie; the erotic impulses in The Piano

(Sharp & Gillard, 2004) and the exploration of 19th century sexuality (Coombs &

Gemmell, 1999). As found in Mellencamp (1996), Freiberg identified a recurring theme of sexual obsession and perversity. Gordon (2002) also identified the theme of perversity.

Hopgood (2002) identified “desire” as one of Campion’s main themes. Nelmes (1999) identified the recurring theme of exploration of female sexuality as well.

Male dominated society. Taubin (2003) ties in Campion’s theme of female sexuality with power struggle while choosing the wrong man. Gillett (2004) identified the patriarchal society and male dominance in the police force in In the Cut. In The Piano the male dominated society is portrayed through Ada’s husband’s decision to sell the piano, by ordering Ada and identifying himself as the head of the family, as well as the exploitation of patriarchal alliance in the film (Hendershot, 1998; Kelly, 1996; Sharp &

Gillard, 2004). Hopgood (2002) identified Jane Campion’s disempowerment of women and gender politics as her main theme. Hopgood wrote about the control Osmond exercised over the main female characters in Portrait of a Lady.

Female protagonist. Polan (2001) pointed out that Campion’s lead characters are always women, Schröder (2004) added they are younger women looking for their identities, and fighting against the stereotypical gender roles. Hopgood (2002) wrote that

Campion’s lead characters are women rebellious toward gender roles. Campion stated Empirical testing 26

that her being a woman makes her automatically more interested in female protagonists,

and how other women live their lives is very appealing to her (Wexman, 1999).

“Madness” of a female character. In Sweetie, Mellencamp (1996) identified

Sweetie as being mentally unstable; Howe (1990) also pointed out that even though

Sweetie did not, she should have been taking medication for her state of mind, for her

strangeness (Polan, 2001). Janet Frame in An Angel at my Table was misdiagnosed and

kept in a psychiatric hospital for eight-years (Henke, 2000; Hopgood, 2002). In The

Piano, Ada’s husband thought she might be “soft in the head” due to the fact she does not

speak. Ruth’s family and friends in Holy Smoke! thought she was crazy or brainwashed,

and hired a professional to help her be normal (Hopgood, 2002). In In the Cut, Pauline

(our female protagonist’s sister) calls herself a “lunatic” and a “freak” (Gillett, 2004).

Hopgood (2002) identified “madness” as one of Campion’s main themes. She believed that Campion protagonists were labeled as “crazy” or “mad” in the films because they refused to conform to their gender roles; they did not fulfill their feminine roles in society. The only character not labeled as crazy was Isabel Archer; however, she was a nonconformist with her stubbornness, as well as leading her life as she pleased.

Female masochism. Examples of this theme include: In In the Cut, Pauline’s unhealthy love affair, and Frannie’s involvement with an unknown suspicious man

(Fuller, 2003a; Hopgood, 2004; Taubin, 2003); Isabel’s masochistic tendencies in

Portrait of a Lady (Gordon, 2002).

Women on the move, displaced heroines. Gillett (2000b) pointed out that most of

Campion’s heroines are in some ways displaced women such as world travelers, or migrants. Ruth in Holy Smoke! traveled twice to India looking for the truth; Isabelle in Empirical testing 27

The Portrait of a Lady is a young American who lived and traveled in Europe; Ada in

The Piano, was “shipped” from Scotland to New Zealand almost as “a mail order bride”;

Janet in An Angel at My Table was traveling to Europe (England, France and Spain) to gain some experience for her writing carrier; Sweetie was one heroine who is not really traveling; and was actually forbidden to take the trip the whole family is taking, she was tricked into staying home alone (Gillett, 2000a). Another character that seemed to be stationary is Frannie in In the Cut. However, she was on a quest of finding the murderer of her sister and two other women.

Heroines in quest of their identities. Many scholars and critics identified the theme of woman’s identity and belonging (Gillett, 2000a; Gillett, 2004; Mellencamp,

1996; Polan, 2001; Schröder, 2004).

Entrapment, imprisonment. Sweetie was almost imprisoned by her family – not wanting to include her in a trip, leaving her behind in the house (Polan, 2001). Janet was hospitalized for eight years, wrongfully; Ada was a prisoner in an arranged marriage and imprisoned by her husband in the house (Gordon, 2002; Polan 2001); Isabel was a prisoner in a marriage she chose herself, she became a virtual prisoner (Gillett, 2000b;

Gordon, 2002; Polan, 2001); Ruth was imprisoned by her family – locked in the secluded hut in the middle of nowhere (Gordon, 2002; Polan, 2001); Frannie was imprisoned in her own way, in her own head, she “holds her prison bars in front of her while floating freely in space” (Fuller, 2003a, p. 17), and later in the film Frannie was imprisoned by the killer in the lighthouse.

Issues with family and sibling. The Piano dealt with restructuring of a family

(Sharp & Gillard, 2004) and the difference between real and mythical family (Chumo II, Empirical testing 28

1997). Sweetie was a story of familial chaos, neurotic siblings, and confused parenting

(Hinson, 1990; Howe, 1990; McHugh, 2001). As found in Mellencamp (1996), Freiberg pointed out the theme of somewhat dangerous domestic situations. Hopgood (2002) also identified the recurring theme of clash between the female protagonist and their families,

significant others, or other characters. Fuller (2003b) identified parental abandonment in

In the Cut.

The fantasy, or dream images. Critics and scholars identified Campion’s theme of

fantasy and dream sequences in her films such as the black and white fantasy images in

Sweetie (Polan, 2001), Flora’s vivid story telling in The Piano (Hopgood, 2004), black

and white surreal travelogue (Hopgood, 2004; Polan, 2001) and the “Buñuelian

surrealism as beans on Isabel’s plate with flies buzzing above them . . . ” in Portrait of a

Lady (Polan, 2001, p. 127), Ruth’s and PJ’s colorful hallucinations in Holy Smoke!

(Polan, 2001), and the sepia toned dream sequences and stories about Frannie’s parents’ engagement in In the Cut (Hopgood, 2004).

Production techniques. As found in Mellencamp (1996), Freiberg found her mise- en-scène to be surreal, with unusual cuts and framing. Keough (1999) pointed out

Campion’s production techniques: The unpredictable angles and elliptical cuts, which she developed in her early filmmaking. Wexman (1999) wrote about the use of beautiful wide shots of images of vegetation and scenery shown in Campion films. Polan (2001) pointed out Campion’s frequent use of close up shots and extreme close up shots.

Schröder (2004) identified the perfect reality-to-dream-to-reality sequences in Sweetie, as well as the perfect wide shots of New Zealand scenery in An Angel at my Table. He writes “the bright green hills, the blue ocean, the always present wind makes it seem like Empirical testing 29

those are the shadows of Janet Frame” (p. 257). Thompson and Bordwell (2003) and

Sharp and Gillard (2004) also admired the beautiful shots of New Zealand, as well as the

close up shots of Janet’s face.

Schröder (2004) admired the visual complexity in The Piano as well as the sound

complexity: The constant humming of the wind, floor creaking, the whispering of the

ocean’s waves, we could even hear the fingers on the piano keys. McHugh (2001)

pointed out the popularity of voiceovers among women filmmakers, as well as

Campion’s distinct selection of diegetic and non-diegetic music and sounds.

Corrigan and White (2004) pointed out the perfect use of angles in Campion’s

films such as the high angles, extreme low angles, and overhead shots in The Piano.

They also wrote about the original use of voiceover that connected the viewer with the protagonist on a subjective level. Nelmes (1999) identified the use of tight close up shots of Ada and her daughter Flora, and Fuller (2003a; 2003b) pointed out the unflattering close up shots of Pauline’s face in In the Cut.

Never a true genre. Much contemporary genre mixing was identified in the literature such as: Holy Smoke! being a mix of drama and comedy (Hopgood, 2002;

McKew, 1999; Polan, 2001), The Piano with traits of woman’s film, melodrama, romance, gothic thriller (Coombs & Gemmell, 1999; Sharp & Gillard, 2004), Sweetie a blend of horror and comedy (Hinson, 1990) and In the Cut, slasher film, film noir and detective/murder mystery traits (Hopgood, 2004).

Research Questions Empirical testing 30

As summarized above, there were many recurring themes, narrative constructions,

and production techniques identified by scholars, critics, and researchers in Jane

Campion’s films.

The literature also uncovered three phases of auteur theory evolution, but no history of empirical testing. The focus of the present study was on the first two phases:

the romantic-auteurism, and the structuralist-auteurism which helped to uncover the

recurring ideas, creative personalities, visual distinctions and thematic fixations of the

films’ director as well as the narrative construction with familiar characters. Based on the

literature, three questions have been posed:

RQ1: Do Jane Campion films differ from non-Campion films by themes and

motifs?

RQ2: Do Jane Campion films differ from non-Campion films by narrative

construction with characters?

RQ3: Do Jane Campion films differ from non-Campion films by production

techniques? Empirical testing 31

CHAPTER II

METHOD

Content Analysis

There have been many definitions for content analysis, such as “Content analysis

is a research method that uses a set of procedures to make valid inferences from text”

(Weber, 1990, p. 9); “Content analysis is a research technique for making replicable and

valid inferences from data to their context” (Krippendorff, 1980, p. 21); however

Neuendorf (2002) provides a more recent and certainly more comprehensive definition.

She states, “Content analysis is a summarizing, quantitative analysis of messages that

relies on the scientific method (including attention to objectivity-intersubjectivity, a priori design, reliability, validity, generalizability, replicability, and hypothesis testing) and is

not limited as to the types of variables that may be measured or the context in which the

messages are created or presented” (Neuendorf, 2002, p. 10). The present study was

conducted according to the definition stated by Neuendorf by applying the proper scientific techniques. The a priori design ensured objectivity or at least intersubjectivity,

reliability was measured among coders to ensure the consistency and validity of the results, the proper documentation ensured replicability, and good sampling assured generalizability. Empirical testing 32

This study conducted a content analysis of films to provide answers to the three

research questions. Human coding was performed to collect data from Campion and non-

Campion films. Jane Campion is an author of seven feature films – all dramas. To be able

to make concrete conclusions as to whether she falls into the category of an auteur, her

films were compared to non-Campion films.

Sampling Frames

Campion films. The census of seven Jane Campion films was examined. However,

Campion’s first feature Two Friends (1986) was made exclusively for television with a

TV crew. Certain production rules were agreed upon before the shoot, such as no close

up shots (Wexman, 1999). This information stated by Campion in her interview helped to

decide on exclusion of the film from this analysis.

The remaining six films made by Jane Campion (Sweetie, 1989; An Angel at My

Table, 1990; The Piano, 1993; Portrait of a Lady, 1996; Holy Smoke!, 1999; and In the

Cut, 2003) were compared to eighteen non-Campion films.

Non-Campion films. The sampling frame for the non-Campion films is defined as

English language feature length drama films made in 1989, 1990, 1993, 1996, 1999, and

2003 (matching the years of release of Campion films) that are listed in the Internet

Movie Database (IMDb), an on-line source (www.imdb.com).

There are thousands of films made every year, with different styles and different technological advances for each time period. To avoid the possible differences, only films made in the six years identified above were included in the sample. The sample was limited to English language feature films; reading subtitles while coding would be Empirical testing 33

excessively distracting and ineffective. To make sure that genre theory is not interfering

with the test of auteur theory, only drama films were selected, enabling for a stricter test.

Random Sampling. Systematic random sampling was conducted in the IMDb

environment. Lists for each of the targeted six years were generated. Each list was then

further limited to English language, and the drama genre – a function of the IMDb web

site. It must be noted that even though some films in the IMDb were categorized as

dramas, several were eliminated due to cross listing. If a film was listed as an

action/drama or a horror/drama, such a film was not selected. There are too many

production, plot, and theme differences among action, horror, and drama films.

Three films for each year were selected. A total of 18 non-Campion films with availability on DVD or VHS were selected. IMDb is a very extensive database, with

some limited options. For example, it does not allow for a search of feature films only, and it offers information on films not recognized by the public. If a short film or a film that is not available on DVD or VHS was randomly selected, the next eligible film was selected instead. Out of the 1,335 English language drama films made in 1989 these three were randomly selected: Communion; Lean on Me; and Speaking Parts. Out of the 1,336

English language drama films made in 1990, these three were randomly selected: The

Comfort of Strangers; The Juniper Tree; and Stanley & Iris. Out of the 1,369 English language drama films made in 1993, these three were selected: The Crush; Guilty as Sin; and Public Access. Out of the 1,688 English language drama films made in 1996, these three were selected: City Hall; Kissed; and The War at Home. Out of the 2,120 English language drama films made in 1999, these three were selected: The Cider House Rules;

Liberty Heights; and Stonebrook. Finally, out of the 3,463 English language drama films Empirical testing 34 made in 2003, these three were selected: Big Fish; It Runs in the Family; and Runaway

Jury.

Even though these films were identified as available on DVD or VHS, Juniper

Tree (1990) and Kissed (1996) had to be replaced due to non-availability. Their replacements were China Moon (1990) and Lilies (1996). A complete list of all movies content analyzed can be found in Table 2.

Table 2

List of Analyzed Films

ID Film Title Film Director Country Year

1 Communion Philippe Mora USA 1989

2 Lean On Me John G. Avildsen USA 1989

3 Speaking Parts Atom Egoyan Canada 1989

4 Sweetie Jane Campion Australia 1989

5 The Comfort of Strangers Paul Schrader Italy/USA 1990

6 China Moon John Bailey USA 1990

7 Stanley and Iris Martin Ritt USA 1990

8 An Angel at My Table Jane Campion New Zealand 1990

9 The Crush Alan Shapiro USA 1993

10 Guilty as Sin Sidney Lumet USA 1993

11 Public Access Bryan Singer USA 1993

12 The Piano Jane Campion New Zealand - France 1993

13 City Hall Harold Becker USA 1996 Empirical testing 35

14 Lilies John Greyson Canada 1996

15 The War at Home Emilio Estevez USA 1996

16 The Portrait of a Lady Jane Campion GB 1996

17 The Cider House Rules Lasse Hallstrom USA 1999

18 Liberty Hights Barry Levinson USA 1999

19 Stonebrook B. W. Thompson USA 1999

20 Holy Smoke! Jane Campion USA/Australia 1999

21 Big Fish Tim Burton USA 2003

22 It Runs in The Family Fred Schepisi USA 2003

23 Runaway Jury Gary Fleder USA 2003

24 In The Cut Jane Campion Australia/USA/GB 2003

Concept Sets

The present study measured variables representing three main concept sets: (A)

Themes and motifs, (B) Characters in films, and (C) Production techniques. More detail

can be found in Appendix A.

Themes and Motifs. Theme, as mentioned earlier, is defined as “a dominant idea

made concrete through its representation by the characters, action, and imagery of the film” (Rabiger, 1997, p. 517). Motif, as mentioned earlier, is defined as “any formal element repeated from film history or from the film itself whose repetition draws attention to an unfolding thematic statement” (Rabiger, 1997, p. 513). Themes and motifs were operationalized with thirty-four variables such as how many lead, major, and medium characters were present in the film, what kind of loss might be dealt with in the Empirical testing 36

film, the degree of nudity shown in the film, and the degree of sexual encounters shown

in the film.

Narrative Constructions of Characters. Characters that were going to be analyzed

in this study were the lead and major role characters. Characters were operationalized with forty-nine variables such as demographics; love situation throughout the film; identity change; physical, psychological or sexual abuse; family issues.

Production Techniques. Production techniques are defined as elements of the filmmaking, camera movement and placement, lighting, sound, and editing. Production techniques were operationalized with forty-five variables. We were measuring camera movement (such as tracking shots, hand held camera), camera positioning (such as angles, close up shots, depth of field, lens focus), editing techniques (such as use of transitions, slow motion, time lapse), lighting techniques (such as use of color filters), and sound (such as use of diegetic or non-diegetic music, superhearing sensation).

Units of Data Collection

Data were collected separately for each of the three concept types. Each films has been watched three times with separate data collections. For Section A, a full length feature film (sixty minutes or longer) was coded as a whole. For Section B, data was collected on lead and major characters that were identified in the film. For Section C, data was collected for five-minute intervals of the films, which is a norm for television and film data collection (Lombard, 1997; Neuendorf, 2003).

Codebook and Coding Forms

A twenty-eight page codebook was developed to measure the variables. It may be found in Appendix A, followed by the coding forms in Appendix B. Many variables Empirical testing 37 were identified by reviewing literature on Jane Campion’s films. To be objective, non-

Campion variables had to be included in the study. After consulting many internet sources and publications, a book by Y. Tasker, Fifty Contemporary Filmmakers (2002), has been selected and numerous other variables having nothing to do with Jane Campion were taken from it. Every third auteur included in the book was reviewed. Selected themes, motifs, narrative constructions of characters, and production techniques from the selected auteurs were embedded in the codebook. All the identified, “signatures” by those auteurs can be found in Appendix C. However, not all were selected and embedded in the codebook. The fifteen directors examined were: Gregg Araki, Luc Besson, Tim Burton,

Jackie Chan, David Cronenberg, David Fincher, Jim Jarmusch, Abbas Kiarostami, Ang

Lee, Michael Man, John Sayles, Steven Soderbergh, Oliver Stone, Peter Weir, and

Allison Anders. The traits and signatures of some auteurs integrated in the codebook include: The absence of parents in Luc Besson’s films, characters being outsiders in Tim

Burton’s films, the theme of modernity versus tradition in Ang Lee’s films, wide angles in David Fincher films, and Peter Weir’s avoidance of sexuality.

Section A. The unit of data collection for section A was a full-length feature film.

Section A measured the general themes and motifs in each film as a whole to uncover general patterns in the film. Thirty-four variables were developed to operationalize general themes and motifs of each film and ultimately give answers to research question

1. This section identified the number of lead, major, and medium female characters as well as number of lead, major, and medium male characters present in the film. It also uncovered whether the father or mother of a lead, major, medium or other character was physically present or talked about. Clashes between tradition and modernism, as well as Empirical testing 38 clashes between cultures were recognized. A set of variables measured unwanted losses:

Loss of a child, spouse, lover, and/or family member due to break up or death. Loss of freedom, job, money, or physical beauty were also measured. If a major loss that was not included in the codebook happened in the film, the coder was to identify it as an “other” loss variable. Coders were asked to identify who the losses happened to: Lead/major or other characters. Coders counted each shot of frontal nudity (partial, top, bottom and full nudity) as well as back side nudity (partial, top, bottom and full nudity) of male and female characters. Coders also counted each shot of romantic kissing, French kissing, oral sex, sexual intercourse, or masturbation. Coders were asked to document any taboos that were violated in the film, if the film had a primary point of view of a female, and if angels were present in the film.

Section B. The unit of data collection for Section B was the major or lead character. Major character has been defined as one that appears in over 50% of the film but does not appear to be lead character. If this character was crucial to the plot line

(meaning if the character was not there, the story would change) but did not appear in over 50% of the film, the character was coded as a major role anyways. Lead character has been defined as major plus the whole story revolves around that character, and his or her presence is essential to the story. Fifty variables were used to measure features of each character to help discover the distinctiveness of the use of lead and major role characters by Jane Campion. This section has helped to answer research question 2.

General demographic and role characteristics were recorded for each character: Role

(major or lead), whether the character appeared in the opening scene, social age of the character (child, adolescent, young adult, mature adult, or elderly), estimated age of the Empirical testing 39 character, gender (male, female), race (Caucasian, African, Asian, Middle Eastern,

Native-American, Bi-racial, or Other), attractiveness (extremely attractive, attractive, average, unattractive, extremely unattractive), and occupation (27 types). The coders also documented if the character traveled (outside of the country or within the country), and if the character lived abroad, if the character was happy throughout majority of the film.

The character’s love situation at the beginning, middle and end of the film with six different choices was identified, as was if the character was looking to change his or her identity by gender change, job/career change, getting married, religion change, relocating, or by other factors. If other identity change was identified, coders were asked to explain it.

Abusive behavior was measured by identification of physical, psychological, or sexual abuse by the same or opposite gender characters, as well as if the character was physically, psychologically, or sexually abusive to the same or opposite gender characters. It was documented only if it happened or not (Yes or No). Characters’ personalities were assessed by Eysenck and Eysenck’s (1985) definitions of introversion and extroversion, loneliness of the character (sometimes, most of the time, or never), if the character was an outsider, and if he/she seemed realistic. Coders were asked to recognize if the character engaged in substance abuse (alcohol, drugs, both or none), if the character tried to commit suicide, and if he/she died in the film.

A set of variables measured family issues and dysfunctions with siblings, parents, children, grandparents, grandchildren, spouse, and other family members. Coders documented if such issues occurred and also were asked to record number of siblings the character had. Empirical testing 40

Section C. The unit of data collection for Section C was a five-minute interval.

The reason for breaking the full-length movies into five-minute long intervals was to make it easier and more manageable to code. Otherwise it would be nearly impossible to keep track of all the counts of the variables. Forty-five variables were developed to operationalize the production techniques and patterns of each film and to provide data to answer research questions 3. Definitions and examples for all variables were provided in the codebook (Appendix A). For each interval, coders recorded occurrence of a voiceover, presence of diegetic or non-diegetic/background music, color picture, black and white picture, sepia, or animation/cartooning. Coders were then asked to count occurrences of extreme close up shots of the motion of touching, close up and extreme close up shots of faces, feet/legs, hands/arms, other body parts, furniture, windows, mirrors, trees, flowers/plants, food, and animals. Counts of point-of-view shots, long tracking shots, special visual effects/blurred image shots, distorted image shots, slow motion, time-lapse, and jump cut shots were also recorded. Use of color filters (blue, red, green, other color) was recorded as a count of each scene/sequence. A count of each unusual transition was recorded, and the use of simultaneous live-action and animation was identified by counting each scene/sequence. If a color and black and white picture appeared on the screen at the same time, coders were asked to record counts of each occurrence. Handheld camera technique was recorded by counts of each scene/sequence.

Different camera angle shots were counted and recorded (low angle, high angle, overview, underview and canted). More counts of shots were recorded: The character breaking the fourth wall, super hearing sensation, extremely long shot, and medium shot revealing a secondary body element arm/hand or leg/foot. Empirical testing 41

Coders became very intimate with the content, due to such an immense coding scheme. Very little room was left for vague or general coding. Choices such as “other” or “unable to determine” are minimally present in the coding scheme. Such a forced coding technique is appropriate for this study.

Training of Coders and Inter-Coder Reliabilities

A total of three coders, including the principal researcher, participated in this study, two females and one male. Training took place over several meetings where several non-Campion movies were viewed, followed by interactive discussions and clarifications of definitions using various examples. Several revisions of the codebook were made. During the meetings, variable families (variables that belong to a similar group) were also identified. In section A, four families were identified: Counts of characters (A1-A6), losses (A11.1-A11.22), nudity (A12-A25) and sexual acts (A26-

A31). In section B, four variable families were recognized: Love situation (B12-B14), identity change (B16-B22), abuse (B23-B34) and family dysfunctions (B43-B49). In section C, seven variable families were identified: Music (C2, C3), picture (C4-C7), close up and extreme close up shots (C8-C19), motion (C21 & C24-C26), color filters (C27-

C30), angles (C35-C39), and medium shots of arms, hands, legs, feet (C43-C44).

Identification of the variable families became crucial after coding the full length of the reliability movie River’s Edge (Daly, Gibson, Liuzzi, Pillsbury, Sanford, & Streit,

1986). Due to a large number of variables in the codebook, it was nearly impossible for all of them to be present in the above mentioned reliability movie. Therefore, to keep as many variables as possible, at least one variable from the variable family had to be present and coded, and an acceptable level of reliability had to be reached, in order to Empirical testing 42

keep the whole variable family. For example, in River’s Edge, all the variables belonging to the family dysfunctions (Section B) were kept in the dataset even though all were not tested via the reliability check. Four out of seven family dysfunction variables were coded with high reliability. It was assumed that if the rest of the variables had occurred, they would be coded with a high reliability as well.

To be able to calculate reliability for Section A, additional movies had to be coded (i.e., reliability coefficients ay not be calculated from single data point). Due to coder unavailability, it was decided to code three eleven-minute segments of Sweetie

(MacKinnion & Maynard, 1989), Cider House Rules (Blomquist et al., 1999), and Big

Fish (Cohen, Frauenfelder, Jinks, Schmidt, & Zanuck, 2003), movies already coded by the unavailable coder.

To keep some key variables from Sections B and C, another additional coding had to be conducted. Due to two coders’ unavailability; this coding has been done by the principal coder only for content already coded by the unavailable coder(s). As a result of this supplementary coding, fourteen variables (six from Section B, and eight from Section

C) were recovered.

PRAM, a Windows-based computer program, was used to calculate inter-coder

reliability coefficients among the three coders. The program calculates the following

reliability coefficients: Percent Agreement, Scott’s Pi, Cohen’s Kappa, Krippendorff’s

Alpha, Spearman Rho, Pearson Correlation, Lin’s Concordance, and Holsti’s Reliability.

In the present study, Multiple-coder kappa has been used to assess nominal variables,

Spearman Rho for ordinal variables, and Lin’s Concordance coefficient for interval and

ratio variables. The recommended criterion for Cohen’s kappa is above .40, according to Empirical testing 43

Banerjee, Capozzoli, McSweeney, and Sinha (1999). As found in Neuendorf (2002),

Ellis identified rules of thumb for correlation coefficients over .75 being of high reliability. It is also agreed among some scholars, that 70% and higher agreement is considered reliable.

All acceptable inter-coder reliabilities for nominal variables are presented in

Table 3.1.

Table 3.1

Acceptable Inter-Coder Reliabilities for Sections A, B, & C

Multiple-coder Kappa

Multiple-coder %

Kappa Agreement

Avg.

between

Coders: 1 - 3 coders

A11.5 Loss of a spouse due to death 1.00 100

A11.7 Loss of lover due to death 1.00 100

A11.9 Loss of family mem. due to death 1.00 100

A11.11 Loss of a friend due to death 0.46 67

A11.16 Loss of health 0.70 83

A11.17 Loss of material things 0.68 83

A11.18 Loss of freedom 0.70 100

B2 Role 0.70 87 Empirical testing 44

B5 Gender 1.00 100

B8 Occupation 0.58 100

B9 Travel 1.00 100

B10 Abroad 1.00 100

B12 Love at beginning 0.42 87

B13 Love in the middle 0.78 87

B14* Love in the end 0.40 60

B15 Sexual orientation 0.70 87

B35 Introvert vs. Extrovert 0.78 87

B39 Substance abuse 0.73 87

B42 Char. dies in the film? 1.00 87

B43 Issue w/ sibling 1.00 100

B44 Issue w/ parent 1.00 100

B45 Issue w/ child 1.00 100

B48 Issue w/ spouse 1.00 100

B50 Psych Disorder 1.00 100

* Variables that need to be analyzed with caution due to low reliability

Table 3.2 shows adequate inter-coder reliabilities for ordinal variables. There

were two variables, appearance and loneliness, that had low reliability coefficients and

the percent agreement did not meet the acceptable level as well. Therefore results linked with these two variables must be examined with caution.

Empirical testing 45

Table 3.2

Acceptable Inter-Coder Reliabilities for Sections A, B, & C

Spearman's Rho

%

Spearman's Rho Agreement

Avg.

Coder Coder Coder among

pair: 1, 2 pair: 1, 4 pair: 2, 4 coders

B4 Age 1.00 1.00 1.00 100

B7* Appearance 0.79 0.53 0.81 60

B36* Appears lonely 0.75 0.75 0.50 60

* Variables that need to be analyzed with caution due to low reliability

Table 3.3 presents acceptable inter-coder reliabilities for interval and ratio variables. There were two variables (low angle shot, and high angle shot) and two calculated variables (overview shot and underview shot) that have low reliability scores

and results linked with those variables must be analyzed with caution.

Empirical testing 46

Table 3.3

Acceptable Inter-Coder Reliabilities for Sections A, B, & C

Lin’s Concordance

%

Lin’s Agreement

Coder Coder Coder Avg.

pair: 1, pair: 1, pair: 2, among

2 3 3 coders

A1 # of lead females 1 1 1 83

A2 # of major females 1 1 1 100

A3 # of medium females 1 0.92 0.92 83

A4 # of lead males 1 1 1 100

A5 # of major males 1 0.67 0.67 70

A6 # of medium males 0.96 0.87 0.86 50 A father present VA7 A7.1+A7.2+A7.3+A7.4+A7.5 0.5 1 0.5 83

A7.1 Father - of lead char. 1 1 1 100 A mother present VA8 A8.1+A8.2+A8.3+A8.4+A8.5 1 1 1 100

A8.1 Mother - of lead char. 1 1 1 100

A10 Clash between cultures 1 1 1 100

A12 Partial nude front female 0.97 0.93 0.85 100

A13 Top nude front female 0.75 0.96 0.61 75

A14 Bottom nude front female 1 1 1 100 Empirical testing 47

A15 Full nude frontal female 0.98 0.95 0.99 100

A19 Full nude back female 1 1 1 100

A20 Top nude frontal male 0.98 0.98 0.93 100

A22 Full nude frontal male 1 1 1 100

A23 Top nude back male 1 0.75 0.75 83 Kissing VA28.1 A26+A27+A28 0.75 0.75 1 83

A29 Oral sex 1 1 1 100

A31 Sexual Intercourse 1 1 1 100 All Sex VA31 VA28.1+29+A30+A31 0.75 0.75 1 83

A32 Taboo 1 1 1 100

A33 Primary POV of female 1 1 1 100

A34 Presence of angels 1 1 1 100

B3 Opening scene 1 1 1 100

B4.1 Estimated age 0.99 0.98 1 100

B18 Identity change: marriage 1 NC NC 100

B23 Phys abused by same gender 1 0.55 0.55 75 Psych abused by opposite B26 gender 1 NC NC 100

B28 Sex abuse by opposite gender 1 NC NC 100

B29 Phys abusive to same gender 0.55 0.55 1 75

B30 Phys abusive to other gender 1 NC NC 100

B32 Psych abusive to other gender 1 1 1 100

B34 Sex abusive to other gender 1 1 1 100 Empirical testing 48

B37 An outsider 1 1 1 100

B40 Tries to hurt self 1 NC NC 100

B41 How many siblings? 1 1 1 100

C1 Voiceover 0.5 1 0.5 93

C2 Diegetic music 0.79 0.78 0.57 86

C3 Non-diegetic music 0.64 1 0.64 96

C8 XCU - Motion of touch 0.72 NC NC 100

C9 Face: CU 0.98 0.85 0.92 100

C9.1 Face: XCU 0.92 NC NC 93

C11 Hands: CU 1 1 1 100

C13.1 Furniture: XCU 1 NC NC 100

C14 Window: CU 0.65 0.64 0.84 93

C15 Mirror: CU 0.44 0.64 0.64 93

C21 Long tracking shots 0.6 0.9 0.69 83

C24 Slow motion 0.85 NC NC 96

C27 Blue color filter 1 NC NC 100

C34 Handheld camera 1 NC NC 100

C35* Low angle 0.4 0.88 NC 63

C36* High angle 0.62 0.89 0.5 63

C37 Overview 1 1 1 100

C38 Underview 0.84 0.78 0.84 100

C39 Canted angle NC 1 NC 97 Empirical testing 49

C41 Superhearing 0.47 0.64 0.63 87

C42 Scenery XL Shot 0.47 0.36 0.81 87

C43 Medium shot arm/hand 0.46 1 0.46 93

C44 Medium shot leg/foot 0.64 0.64 0.05 93

C45 Focus pulls 1 NC NC 96 Shots of legs/feet VC50 (C10+C10.1+C44) 0.44 0.64 0.64 93 Shots of arms/hands VC51 (C11+C11.1+C43) 0.62 1 0.62 93 Shots of looking up VC52* (C35+C38) 0.4 0.87 0.42 63 Shots of looking down VC53* (C36+C37) 0.88 0.62 0.5 63 Note: Variable starting with V (for example VC53) it is a computed variable

* Variables that need to be analyzed with caution

Calculated variables. There were eight calculated variables. They were

constructed from several original variables that could not be kept due to infrequent

occurrence and corresponding low reliability. “A father present” variable was calculated by adding father of lead (A7.1), major (A7.2), medium (A7.3), minor/other (A7.4) characters, and father that was not present but talked about (A7.5). “A mother present” variable was calculated by adding mother of lead (A8.1), major (A8.2), medium (A8.3), minor/other (A8.4) characters, and mother that was not present but talked about (A8.5). A

“kissing” variable was constructed by adding romantic kissing (A26), French kissing

(A27) and kissing other body parts (A28). An “all sex” variable was constructed by adding kissing (VA28.1), oral sex (A29), masturbation (A30) and sexual intercourse

(A31). Empirical testing 50

Even though additional coding was done, several variables that did not occur in the reliability coding did not belong to any variable family. Unfortunately, those variables had to be discarded. They are presented in Table 4.

Table 4

List of Variables with Un-Measurable Reliability

Multiple

Kappa Lin's Concordance

Coder Coder Coder

Coders: 1 - 3 pair: 1, 2 pair: 1, 3 pair: 2, 3

A11.14 Loss of faith in self NC

A11.20 Loss of money NC

A11.21 Loss of physical beauty NC

A7.2 Father -of major char. * NC NC NC

A7.4 Father -of minor/other char. * NC NC NC

A8.3 Mother -of medium char. * NC NC NC

A8.4 Mother -of minor/other char. * NC NC NC

C22 Blurred image NC NC NC

C31 Unusual transition NC NC NC

C32 Live-action& animation NC NC NC

C33 Color and B&W NC NC NC

C40 Breaking the fourth wall NC NC NC

Note: * Variables were used to calculate new variable Empirical testing 51

There were also several variables that did not meet the suggested reliability level; therefore those had to be discarded. Those variables are presented in Table 5.

Table 5

List of Variables Dropped Due to Low Reliability Coefficient

Multiple

Kappa Lin's Concordance

Coder Coder

Coders: Coder pair: pair: pair:

1 - 3 1, 2 1, 3 2, 3

A11.14 Loss of faith in others 0.14

A11.19 Loss of a job -0.09

A11.20 Other loss (Identify): -0.09

A7.3 Father - of medium char. * 0 1 0

A8.2 Mother - of major char.* 0 0 1

A26 Romantic kissing* 0 NC 0

A27 French kissing* 0 0 NC

A28 Kissing other body parts* 0 1 0

B38 Is very realistic 0 0 0.62

C23 Distorted image 0 NC 0

Note: * Variables were used to calculate new variable

Empirical testing 52

A complete list of the original and calculated variables and inter-coder reliabilities can be found in Appendix D. Variables that were part of a variable family but could not be measured due to lack of occurrence are included in Appendix D.

Empirical testing 53

CHAPTER III

RESULTS

Descriptives

Section A. As previously discussed, a total of 24 films were coded for this study,

six Campion films (25%) and 18 non-Campion films (75%). Four films per year (1989,

1990, 1993, 1996, 1999, and 2003) were coded. A list is presented in Table 2. The

principal researcher coded sixteen movies (66.7% of the sample); coders 2 and 3 each coded four movies (16.7%).

Section B. A total of 101 lead or major characters were coded, nineteen Campion characters (18.8%) and 82 non-Campion characters (81.2%). One film had only one lead or major character, one film had only two lead or major characters, five films had three lead or major characters, nine films had four lead or major characters, three films had five lead or major characters, three films had six lead or major characters, and two films had seven lead or major characters. Out of the total of 101 coded characters, 37 were coded as lead (36.6%) and 64 (63.4%) as major characters. The estimated range of the characters’ ages was 6-89 years old (M=36.65, SD=16.41). There were 42 female characters (41.6%) outnumbered by 59 male characters (58.4%), which is consistent with past studies

(Powers et al., 1993; Smith, 1999). Empirical testing 54

Section C. Each of the twenty-four films was divided into five-minute long intervals. There were 530 intervals; 167 Campion intervals (31.5%) and 363 non-

Campion intervals (68.5%) were coded.

The length of the films ranged from 85 minutes (17 codeable intervals) to 155 minutes (31 codeable intervals). Table 6 illustrates the difference in proportions for the

number of Campion versus non-Campion films and the five-minute intervals. It is clear

that Jane Campion’s films run much longer than non-Campion films. Twenty-five percent

of all the films analyzed were directed by Jane Campion, compared to 31.5% of all five-

minute intervals directed by Jane Campion.

Table 6

Length of Films and Intervals, and Number of Characters

Films Characters Intervals

N % N % N %

Campion 6 25 19 18.8 167 31.5

Non-Campion 18 75 82 81.2 363 68.5

Total 24 100 101 100 530 100

Research Questions.

Research Question 1 asked if Jane Campion films differed from non-Jane

Campion films by themes and motifs. To answer the question, many one-way ANOVAs

for interval or ratio variables, and chi square analyses for nominal and ordinal variables Empirical testing 55

were conducted using the director (Campion or non-Campion) as the independent

variable.

Count of Characters. One variable in this variable family was found significant for Campion versus non-Campion films: Number of lead males, F(1, 22) = 5.02, p =.04,

indicating Campion rarely used a lead male character. One variable was found to be

approaching significance: Number of lead females, F(1, 22) = 3.04, p =.10. The rest of the

variables in this variable family were found to be not significant: Number of major

females, F(1, 22) = 0.82, p =.37, number of medium females, F(1, 22) = 2.52, p =.13, number

of major males, F(1, 22) = 2.84, p =0.11, and number of medium males, F(1, 22) = 0.00, p

=.98. See Table 7.

Table 7

Analyses of Variance for Section A: Themes and Motifs

Non-

Campion Campion Total

(n=6) (n=18) (n=24)

Variable Mean sd Mean sd Mean sd F(1, 22) Sig.

A1: Number of lead females 1.00 0.00 0.56 0.62 0.67 0.56 3.04a 0.10 A2: Number of major females 0.83 0.75 1.17 0.79 1.08 0.78 0.82 0.37 A3: Number of medium females 3.83 4.17 1.83 2.04 2.33 2.76 2.52 0.13 A4: Number of lead males 0.33 0.52 1.06 0.73 0.88 0.74 5.02* 0.04 Empirical testing 56

A5: Number of major males 1.00 0.63 1.78 1.06 1.58 1.02 2.84 0.11 A6: Number of medium males 3.83 2.56 3.89 4.27 3.88 3.86 0.00 0.98 VA7: Father character present 0.83 0.41 0.67 0.49 0.71 0.46 0.57 0.46 VA8: Mother character present 1.00 0.00 0.83 0.38 0.88 0.34 1.10 0.31

A9: Tradition versus modernism 0.17 0.41 0.33 0.49 0.29 0.46 0.57 0.46 A10: Clash between cultures 0.67 0.52 0.50 0.51 0.54 0.51 0.47 0.50

A12: Partial nude front female 9.17 8.08 2.00 3.63 3.79 5.83 9.23** 0.01 A13: Top nude front female 6.83 7.70 0.50 1.29 2.08 4.69 12.21** <0.01 A14: Bottom nude front female 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 A15: Full nude front female 2.50 2.51 0.28 0.67 0.83 1.63 12.50** <0.01 A16: Partial nude back female 0.50 0.84 0.11 0.32 0.21 0.51 2.84 0.11 A17: Top nude back female 1.50 2.74 0.72 1.93 0.92 2.12 0.59 0.45 A18: Bottom nude back female 0.50 1.22 0.11 0.32 0.21 0.66 1.61 0.22 A19: Full nude back female 1.83 2.14 0.33 0.77 0.71 1.37 9.78* 0.02 A20: Top nude front male 10.17 9.20 6.94 14.10 7.75 12.94 0.27 0.61 A21: Bottom nude front male 0.33 0.82 0.00 0.00 0.08 0.41 3.3a 0.08 A22: Full nude front male 1.17 1.17 0.33 1.03 0.54 1.10 2.77 0.11 A23: Top nude back male 2.00 1.79 1.22 2.34 1.42 2.21 0.55 0.47 Empirical testing 57

A24: Bottom nude back male 0.33 0.82 0.06 0.24 0.13 0.45 1.79 0.20 A25: Full nude back male 1.50 1.76 0.50 0.79 0.75 1.15 3.81a 0.06

VA28: Kissing 8.50 3.02 3.50 3.28 4.75 3.85 10.86** <0.01

A29: Oral sex 0.83 0.98 0.11 0.32 0.29 0.62 7.81** 0.01

A30: Masturbation 0.67 1.21 0.17 0.51 0.29 0.75 2.09 0.16

A3: Sexual intercourse 0.83 0.98 0.39 0.98 0.50 0.98 0.93 0.35

VA31: All sex 10.83 4.54 4.17 4.08 5.83 5.04 11.42** <0.01

A32: Taboo 0.83 0.41 0.11 0.32 0.29 0.46 19.78** <0.01

A33: POV female 1.00 0.00 0.06 0.24 0.29 0.46 93.50** <0.01

A34: Presence of angels 0.67 0.52 0.00 0.00 0.17 0.38 33.00** <0.01 Note: Variables starting with V (e.g., VA31: All sex) are constructed variables

a .05< p < .10

* p < .05

** p <.01

Presence of father/mother. The constructed variable presence of father was found to be not significant, F(1, 22) = 0.57, p = 0.46, as was presence of mother, F(1, 22) = 1.10, p

= 0.31. The clash between tradition and modernism variable was also found to be not Empirical testing 58

significant, F(1, 22) = 0.57, p = 0.46, as well as clash between cultures, F(1, 22) = 0.47, p

=0.50.

Nudity. Four nudity variables were found to be significant for Campion versus

non-Campion films: Partial nude frontal female, F(1, 22) = 9.23, p = 0.01, top nude frontal

female, F(1, 22) = 12.21, p < 0.01, full nude frontal female, F(1, 22) = 12.50, p < 0.01, and full nude back female, F(1, 22) = 9.78, p = 0.02. These results indicated that Campion films

showed significantly more female frontal nudity than non-Campion films. Two nudity

variables were approaching significance: Bottom nude front male, F(1, 22) = 3.30, p = 0.08,

and full nude back male, F(1, 22) = 3.81, p = 0.06, indicating Campion films showed more male nudity than non-Campion films. The rest of the nudity variables were found to be

not significant: Partial nude back female, F(1, 22) = 2.84, p = 0.11, top nude back female,

F(1, 22) = 0.59, p = 0.45, bottom nude back female, F(1, 22) = 1.61, p = 0.22, top nude front male, F(1, 22) = 0.27, p = 0.61, full nude front male, F(1, 22) = 2.77, p = 0.11, top nude back

male, F(1, 22) = 0.55, p = 0.47, and bottom nude back male, F(1, 22) = 1.79, p = 0.20. The

bottom nude frontal female variable could not be tested due to lack of occurrence.

Sexual acts. Three sexual variables were found to be significantly different

between Campion and non-Campion films: The constructed variable of kissing, F(1, 22) =

10.86, p < 0.01, oral sex F(1, 22) = 7.81, p = 0.01, and the constructed variable all sexual

activity, F(1, 22) = 11.42, p < 0.01, demonstrating Campion films showed more sexual

activity than non-Campion films. Only two sexual variables were found to be not

significant: Masturbation, F(1, 22) = 2.09, p = 0.16, and sexual intercourse, F(1, 22) = 0.93, p

= 0.35. Empirical testing 59

Analyses of variance found three other variables to be significant: Portrayal of taboos in the films, F(1, 22) = 19.78, p < 0.01, female point-of-view, F(1, 22) = 93.50, p <

0.01, and presence of angels, F(1, 22) = 33.00, p < 0.01. All three were found to be more prevalent in Campion films. The ANOVA statistics for all variables demonstrating the difference between Campion and non-Campion films in regards to main themes and motifs can be found in Table 7.

Loss. Chi-square tests were used to determine whether there were significant differences between occurrences of different kind of losses in Campion and non-Campion films. Only one variable was found significant: Loss of a spouse due to break up. Only

16.7% of Campion films did not deal with a marital break up, compared to 94.4% of non-

Campion films. This difference was statistically significant, χ2(2) = 15.41, p < 0.01, and is presented in Table 8. Due to small numbers of cases in some cells, results of the chi- square test should be examined with caution.

Table 8.

Loss of a Spouse Due to Break up in Campion versus Non-Campion Films

Non-

Campion Campion Total

No

count 1 17 18

row % 5.6% 94.4% 100%

column % 16.7% 94.4% 75%

Yes - lead/major character(s) Empirical testing 60

count 2 1 3

row % 66.7% 33.3% 100%

column % 33.3% 5.6% 12.5%

Yes - Other character (s)

count 3 0 3

row % 100% 0% 100%

column % 50% 0% 12.5%

Total

count 6 18 24

row % 25% 75% 100%

column % 100% 100% 100%

Note: 5 cells have expected count of less than 5

Pearson Chi-Square = 15.41

df = 2

p < 0.01

Three additional variables reached near significance: Character’s break up with his/her lover, character’s loss of health, and loss of freedom. The majority of Campion films, 66.7%, dealt with a lead/major character's break up with his/her lover, compared to only 22.2% of non-Campion films. This difference was approaching statistical significance, χ2(1) = 4.00, p = 0.07, and shall be reviewed with caution. Table 9 shows the statistics.

Empirical testing 61

Table 9.

Loss of a Lover due to Break up in Campion versus non-Campion Films

Non-

Campion Campion Total

No

count 2 14 16

row % 12.5% 87.5% 100%

column % 33.3% 77.8% 66.7%

Yes - lead/major character(s)

count 4 4 8

row % 50% 50% 100%

column % 66.7% 22.2% 33.3%

Total

count 6 18 24

row % 25% 75% 100%

column % 100% 100% 100%

Note: 2 cells have expected count of less than 5

Pearson Chi-Square = 4.0 df =1 p = .07

Empirical testing 62

Statistical significance was found to be approaching for loss the of health

variable, χ2(1) = 4.00, p = 0.07, between Campion and non-Campion films. Only 22.2% of non-Campion films included characters experiencing any health issues, compared with an overwhelming 66.7% of Campion films. Due to small numbers of cases in some cells, results of the chi-square test should be examined with caution. Table 10 shows the statistics. Empirical testing 63

Table 10

Loss of Health of Lead/Major Characters in Campion versus Non-Campion Films

Non-

Campion Campion Total

No

count 2 14 16

row % 12.5% 87.5% 100%

column % 33.3% 77.8% 66.7%

Yes - lead/major character(s)

count 4 4 8

row % 50% 50% 100%

column % 66.7% 22.2% 33.3%

Total

count 6 18 24

row % 25% 75% 100%

column % 100% 100% 100%

Note: 2 cells have expected count of less than 5

Pearson Chi-Square = 4.00 df = 1 p = .069

Empirical testing 64

The third variable that was found to be approaching a significant difference in the chi-square test was loss of freedom of lead/major character, χ2(1) = 4.13, p = 0.08. Half

of Campion films (50%) included character(s) that experienced loss of freedom compared

to only 11.1% non-Campion films. This difference was approaching statistical

significance and shall be reviewed with caution. Due to small numbers of cases in some

cells, results of the chi-square test should be examined with caution. The statistics are

presented in Table 11. Empirical testing 65

Table 11

Loss of Freedom of Lead/Major Characters in Campion versus Non-Campion Films

Non-

Campion Campion Total

No

count 3 16 19

row % 15.8% 84.2% 100%

column % 50% 88.9% 79.2%

Yes - lead/major character(s)

count 3 2 5

row % 60% 40% 100%

column % 50% 11.1% 20.8%

Total

count 6 18 24

row % 25% 75% 100%

column % 100% 100% 100%

Note: 2 cells have expected count of less than 5

Pearson Chi-Square = 4.13 df = 1 p = .078

Empirical testing 66

The remaining variables from the loss variable family were found to be not significant. The frequencies for all nominal and ordinal variables for Section A are presented in Table 12.

Table 12

Frequencies for Nominal Variables - Section A

% of films with Losses

(n=24)

Variable Never Lead/Major OtherBoth

A11.1 Loss of a young child due to death 83.3 12.5 4.2

A11.2 Loss of a young child due to break up 95.8 4.2

A11.3 Loss of an old child due to death 87.5 8.3 4.2

A11.4 Loss of an old child due to break up 87.5 12.5

A11.5 Loss of a spouse due to death 70.8 25 4.2

A11.6** Loss of a spouse due to break up 79.2 16.7 4.2

A11.7 Loss of lover due to death 66.7 29.2 4.2

A11.8a Loss of lover due to break up 66.7 33.3

A11.9 Loss of fam. member due to death 45.8 45.8 4.2 4.2

A11.10 Loss of fam. member due to break up 87.5 12.5

A11.11 Loss of a friend due to death 62.5 29.2 8.3

A11.12 Loss of a friend due to break up 91.7 8.3

A11.13 Loss of faith - religion 95.8 4.2

A11.16a Loss of health 66.7 33.3 Empirical testing 67

A11.17 Loss of material things 91.7 8.3

A11.18a Loss of freedom 79.2 20.8

a chi-square .05 < p < .10

* chi-square p < .05

** chi-square p < .01

Research Question 2 asked if Jane Campion films differed from non-Campion

films by narrative construction of characters. To answer the question, many one-way

ANOVAs for interval or ratio variables, and chi square analyses for nominal and ordinal variables were conducted using the director (Campion or non-Campion) as the independent variable.

The presence of a lead or major character in the opening scene was found to be significant for Campion versus non-Campion films, F(1, 99) = 19.78, p = 0.04, in a way that

Campion films showed lead and major characters more often in the opening scenes than did non-Campion films. Characters’ happiness was also found to be significant, F(1, 99) =

9.66, p < 0.01, where none of the Campion characters were found to be happy throughout the majority of the films.

Identity Change. In this variable family, two variables were found significant for

Campion versus non-Campion films: Getting or wanting to get married, F(1, 99) = 6.01, p =

0.02, and change of religion, F(1, 99) = 9.46, p < 0.01, indicating Campion characters were

getting or wanting to get married as well as changing their religion more than non-

Campion characters. Other identity change variables were found to be not significant: Empirical testing 68

Change of gender, F(1, 99) = 0.23, p = 0.63, change of a job, F(1, 99) = 0.12, p = 0.74, and moving to a different city, state or country, F(1, 99) = 1.14, p = 0.29.

Physical, Psychological and Sexual Abuse. Three variables from the abuse

variable family were found to be significantly different for Campion versus non-Campion

films: Character physically abused by opposite gender character, F(1, 99) = 13.98, p < 0.01,

character physically abusive to opposite gender character, F(1, 99) = 9.30, p = 0.04, and

character sexually abusive to opposite gender character, F(1, 99) = 9.23, p < 0.01,

concluding Campion characters were more likely to be physically abused, they were

more likely to be physically abusive, and also sexually abusive to opposite gender

characters, than non-Campion characters. The rest of the abuse variables were found to

be not significant: Physically abused by same gender character, F(1, 99) = 0.38, p = 0.54,

psychologically abused by same gender, F(1, 99) = 0.34, p = 0.56, sexually abused by

opposite gender character, F(1, 99) = 2.67, p = 0.11, physically abusive to same gender

character, F(1, 99) = 1.44, p = 0.23, psychologically abusive to same gender character, F(1,

99) = 0.75, p = 0.39, and psychologically abusive to opposite gender character, F(1, 99) =

1.98, p = 0.16.

Character being an outsider was found to be not significant between Campion and non-Campion films, F(1, 99) = 0.96, p = 0.33, as well as substance abuse, F(1, 99) = 1.62, p =

0.21, character being suicidal, F(1, 99) = 2.67, p = 0.11, and how many siblings a character

had, F(1, 99) = 0.39, p = 0.53. Psychological disorder of the character was found to be

significant, F(1, 99) = 15.84, p < 0.01, indicating that Campion characters were more likely

to suffer with a psychological disorder than were non-Campion characters. Empirical testing 69

The statistics for all interval/ratio variables demonstrating the comparisons between Campion and non-Campion films in regards to characters used in films can be found in Table 13.

Table 13

Analyses of Variance for Section B: Narrative Construction

Non-

Campion Campion Total

(n=19) (n=82) (n=101)

Variable Mean sd Mean sd Mean sd F(1, 99) Sig.

B3: Opening scene 0.32 0.48 0.12 0.33 0.16 0.37 4.45* 0.04

B4.1: Estimated age 31.58 10.38 37.83 17.35 36.65 16.41 2.27 0.14

B11: Happiness 0.00 0.00 0.34 0.48 0.28 0.45 9.66** <0.01

B16: Identity change - gender 0.00 0.00 0.01 0.11 0.01 0.10 0.23 0.63 B17: Identity change - job 0.11 0.32 0.13 0.34 0.13 0.34 0.12 0.74 B18: Identity change - marriage 0.37 0.50 0.13 0.34 0.18 0.38 6.01* 0.02 B19: Identity change - religion 0.11 0.32 0.00 0.00 0.02 0.14 9.46** <0.01 B20: Identity change - move 0.26 0.45 0.16 0.37 0.18 0.38 1.14 0.29 Empirical testing 70

B23: Physically abused by same gender 0.05 0.23 0.10 0.30 0.09 0.29 0.38 0.54 B27: Sexually abused by same gender 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 B26: Psych. abused by opposite gender 0.16 0.37 0.11 0.31 0.12 0.33 0.34 0.56 B28: Sexually abused by opposite gender 0.11 0.32 0.02 0.16 0.04 0.20 2.67 0.11 B29: Physically abusive to same gender 0.05 0.23 0.16 0.37 0.14 0.35 1.44 0.23 B30: Physically abusive to opposite gender 0.32 0.48 0.07 0.26 0.12 0.33 9.30** <0.01 B31: Psychologically abusive to same gender 0.05 0.23 0.12 0.33 0.11 0.31 0.75 0.39 B32: Psychologically abusive to opposite gender 0.16 0.37 0.06 0.24 0.08 0.27 1.98 0.16 B33: Sexually abusive to same gender 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 B34: Sexually abusive to opposite gender 0.16 0.37 0.01 0.11 0.04 0.20 9.23** <0.01

B37: An outsider 0.37 0.50 0.26 0.44 0.28 0.45 0.96 0.33

B39: Substance abuse 0.00 0.00 0.15 0.50 0.12 0.45 1.62 0.21

B40: Suicidal 0.11 0.32 0.02 0.16 0.04 0.20 2.67 0.11

B41: # of siblings 52.79 50.06 60.51 48.41 59.06 48.56 0.39 0.53 Empirical testing 71

B50: Psychological disorder 0.37 0.50 0.06 0.24 0.12 0.33 15.84**<0.01

a .05< p < .10

* p < .05.

** p <.01

Chi-square tests were used to determine whether there were other significant differences between characters in Campion and non-Campion films. A statistically significant difference between Campion and non-Campion films was found for the traveling variable, χ2(3) = 13.74, p = 0.003. Only 36.8% of Campion characters did not travel at all compared to 73.2% of non-Campion characters, 21.1% of Campion characters traveled within the country of the character’s origin compared to 15.9% of non-Campion characters, and 42.1% of Campion characters traveled outside the country of the character’s origin compared to 9.8% of non-Campion characters. Due to small numbers of cases in some cells, results of the chi-square test should be examined with caution. Table 14 presents the Chi-square test. Empirical testing 72

Table 14

Traveling by Characters in Campion versus Non-Campion Films

Non-

Campion Campion Total

No

count 7 60 67

row % 10.4% 89.6% 100%

column % 36.8% 73.2% 66.3%

Yes - within country of origin

count 4 13 17

row % 23.5% 76.5% 100%

column % 21.1% 15.9% 16.8%

Yes - outside country of origin

count 8 8 16

row % 50% 50% 100%

column % 42.1% 9.8% 15.8%

Yes - within and outside country of origin

count 0 1 1

row % 0% 100% 100%

column % 0% 1.2% 81.2%

Total

count 19 82 101 Empirical testing 73

row % 18.8% 81.2% 100%

column % 100% 100% 100%

Note: 4 cells have expected count of less than 5

Pearson Chi-Square =13.74

df = 3

p = .003

The characters living abroad variable was found to be statistically significant,

χ2(2) = 15.64, p < 0.01, where only 57.9% of Campion characters did not live abroad at

all compared to 92.7% of non-Campion characters, 21.1% of Campion characters lived

abroad at least part of the film compared to 3.7% of non-Campion characters, and 21.1%

Campion characters lived abroad the whole length of the feature film, whereas only 3.7% of non-Campion characters have done so. Due to small numbers of cases in some cells, results of the chi-square test should be examined with caution. The statistics are presented

in Table 15. Empirical testing 74

Table 15

Characters Living Abroad in Campion versus Non-Campion Films

Non-

Campion Campion Total

No

count 11 76 87

row % 12.6% 87.4% 100%

column % 57.9% 92.7% 86.1%

Yes - part of the film

count 4 3 7

row % 57.1% 42.9% 100%

column % 21.1% 3.7% 6.9%

Yes - full length of film

count 4 3 7

row % 57.1% 42.9% 100%

column % 21.1% 3.7% 6.9%

Total

count 19 82 101

row % 18.8% 81.2% 100%

column % 100% 100% 100%

Note: 4 cells have expected count of less than 5

Empirical testing 75

Pearson Chi-Square = 15.64

df = 2

p < 0.01

Characters’ love situation in the middle of film was found to differ statistically significantly, χ2(5) = 26.50, p < 0.01, indicating that nearly half of Campion characters

(47.4%) were in an unhappy relationship in the middle of the film, but were not looking

for new one, comparing to only 4.9% of non-Campion characters. Due to small numbers

of cases in some cells, results of the chi-square test should be examined with caution.

Table 16 presents the statistics. Empirical testing 76

Table 16

Characters Love Situation in the Middle of the Film in Campion versus Non-Campion

Films

Non-

Campion Campion Total

Not in relationship, not looking

count 4 15 19

row % 21.1% 78.9% 100%

column % 21.1% 18.3% 18.8%

Not in relationship, looking

count 1 11 12

row % 8.3% 91.7% 100%

column % 5.3% 13.4% 11.9%

In happy relationship

count 2 26 28

row % 7.1% 25% 100%

column % 10.5% 31.7% 27.7%

In unhappy relationship, not looking

count 9 4 13

row % 69.2% 30.8% 100%

column % 47.4% 4.9% 12.9%

In unhappy relationship, looking

count 0 1 1 Empirical testing 77

row % 0% 100% 100%

column % 0% 1.2% 1%

Love situation not apparent

count 3 25 28

row % 10.9% 89.3% 100%

column % 15.8% 30.5% 27.7%

Total

count 19 82 101

row % 18.8% 81.2% 100%

column % 100% 100% 100%

Note: 5 cells have expected count of less than 5

Pearson Chi-Square = 29.50

df = 5

p < 0.01

A Chi-square test found statistical significance between Campion and non-

Campion films and how lonely the lead/major characters appeared, χ2(2) = 6.44, p = .04.

The results indicated that Campion characters tended to be lonelier than non-Campion characters. Only 15.8 % of Campion characters were not lonely, comparing to 47.6% of

non-Campion characters. Due to small numbers of cases in some cells, results of the chi-

square test should be examined with caution. The statistics are presented in Table 17.

Empirical testing 78

Table 17

Characters Appear to be Lonely in Campion versus Non-Campion Films

Non-

Campion Campion Total

No

count 3 39 42

row % 7.1% 92.9% 100%

column % 15.8% 47.6% 41.6%

Sometimes

count 12 33 45

row % 26.7% 73.3% 100%

column % 63.2% 40.2% 44.6%

Most of the time

count 4 10 14

row % 28.6% 71.4% 100%

column % 21.1% 12.2% 13.9%

Total

count 19 82 101

row % 18.8% 81.2% 100%

column % 100% 100% 100%

Note: 1 cells have expected count of less than 5

Pearson Chi-Square = 6.44 df = 2 Empirical testing 79

p = .04

It was also found that there was a statistically significant difference between

Campion and non-Campion characters having issues with siblings, χ2(2) = 15.93, p <

.001. Over half of Campion characters (63.2%) did not have their siblings present in the films, however, 90.2% of non-Campion characters did not have siblings present as well.

Only 21.1% of Campion lead or major characters had siblings present and did not have issues with them compared with 9.8% of non-Campion characters. None of the non-

Campion characters had to deal with issues among their siblings, compared to 15.8%

Campion characters. Due to small numbers of cases in some cells, results of the chi- square test should be examined with caution. The chi-square statistics are presented in

Table 18.

Table 18

Characters with Issues with Siblings in Campion versus Non-Campion Films

Non-

Campion Campion Total

No Siblings present

count 12 74 86

row % 14% 86% 100%

column % 63.2% 90.2% 85.1%

Has siblings, but no issues

count 4 8 12 Empirical testing 80

row % 33.3% 66.7% 100%

column % 21.1% 9.8% 11.9%

Has siblings with issues

count 3 0 3

row % 100% 0% 100%

column % 15.8% 0% 3%

Total

count 19 82 101

row % 18.8% 81.2% 100%

column % 100% 100% 100%

Note: 3 cells have expected count of less than 5

Pearson Chi-Square = 15.93

df = 2

p < 0.01

A chi-square test was also used to determine whether there was a significant difference between Campion and non-Campion characters with regard to issues with their

spouse. The difference was found to be statistically significant, χ2(2) = 19.08, p < 0.01.

Only 68.4% of Campion characters did not experience issues with a spouse, compared to

87.8% of non-Campion films. Also 9.8% of the non-Campion characters had a spouse present and had no issues compared to none of the Campion characters. Only 2.4% non-

Campion characters dealt with issues with a spouse compared to 31.6% of Campion Empirical testing 81 characters. Due to small numbers of cases in some cells, results of the chi-square test should be examined with caution. Table 19 shows the statistics.

Table 19

Characters with Issues with a Spouse in Campion versus Non-Campion Films Non-

Campion Campion Total

No spouse present

count 13 72 85

row % 15.3% 84.7% 100%

column % 68.4% 87.8% 84.2%

Has spouse, but no issues

count 0 8 8

row % 0% 100% 100%

column % 0% 9.8% 11.9%

Has spouse with issues

count 6 2 8

row % 75% 25% 100%

column % 31.6% 2.4% 7.9%

Total

count 19 82 101

row % 18.8% 81.2% 100%

column % 100% 100% 100% Empirical testing 82

Note: 2 cells have expected count of less than 5

df = 2

p < 0.01

Three additional variables were approaching significance: The age of characters, gender, and issues with parents experienced by the lead/major characters. The age of

characters, χ2(2) = 8.47, p = 0.07, shows that the majority (73.7%) of Campion characters

were young adults, compared to 39% of non-Campion characters. Campion characters

tended to be younger than non-Campion characters. Due to small numbers of cases in

some cells, results of the chi-square test should be examined with caution. The results are

presented in Table 20.

Table 20

Age of the Characters in Campion versus Non-Campion Films

Non-

Campion Campion Total

Child

count 1 3 4

row % 25% 75% 100%

column % 5.3% 3.7% 4%

Adolescent

count 1 14 15

row % 6.7% 93.3% 100% Empirical testing 83

column % 5.3% 17.1% 14.9%

Young Adult

count 14 32 46

row % 30.4% 69.6% 100%

column % 73.7% 39% 45.5%

Mature Adult

count 3 28 31

row % 9.7% 90.3% 100%

column % 15.8% 34.1% 30.7%

Elderly

count 0 5 5

row % 0% 100% 100%

column % 0% 6.1 5%

Total

count 19 82 101

row % 18.8% 81.2% 100%

column % 100% 100% 100%

Note: 5 cells have expected count of less than 5

Pearson Chi-Square = 8.47 df = 4 p = .07

Empirical testing 84

Gender of characters was also approaching significance, χ2(1) = 2.56, p = 0.09.

The results indicate that 57.9% of Campion characters were females and only 37.8% of non-Campion characters were female. Results of the chi-square test are presented in

Table 21 and should be examined with caution due to small numbers of cases in some cells. Empirical testing 85

Table 21

Gender of the Characters in Campion versus Non-Campion Films

Non-

Campion Campion Total

Female

count 11 31 42

row % 26.2% 73.8% 100%

column % 57.9% 37.8% 41.6%

Male

count 8 51 59

row % 13.6% 86.4% 100%

column % 42.1% 62.2% 58.9%

Total

count 19 82 101

row % 18.8% 81.2% 100%

column % 100% 100% 100%

Note: 2 cells have expected count of less than 5

Pearson Chi-Square = 2.56 df =1 p = 0.09

Empirical testing 86

Statistical significance was found to be approaching for the issues with parent

variable, χ2(2) = 4.71, p = 0.10. Only 6.1% of non-Campion characters were shown to be having issues with their parents, compared with 21.1% of Campion characters. Due to small numbers of cases in some cells, results of the chi-square test should be examined with caution. Table 22 shows the statistics.

Table 22

Characters with Issues with Parents in Campion versus Non-Campion Films

Non-

Campion Campion Total

No parents present

count 12 67 79

row % 15.2% 84.8% 100%

column % 63.2% 81.7% 78.2%

Has parents, but no issues

count 3 10 13

row % 23.1% 76.9% 100%

column % 15.8% 12.2% 12.9%

Has parents with issues

count 4 5 9

row % 44.9% 55.6% 100%

column % 21.1% 6.1% 8.9%

Total Empirical testing 87

count 19 82 101

row % 18.8% 81.2% 100%

column % 100% 100% 100%

Note: 2 cells have expected count of less than 5

Pearson Chi-Square = 4.71

df = 2

p = 0.10

There were a number of nominal variables that were found to be not significantly different between Campion and non-Campion characters. They are: Role, race, appearance, occupation, love at the beginning of film, love at the end of film, sexual orientation, introvert versus extrovert, substance abuse, death of a character, issues with child, issues with grandparent, issues with grandchild, and issues with other family member. Their frequencies are presented in Table 23, along with those for which statistical significance was found. Empirical testing 88

Table 23

Frequencies for Nominal Variables - Section B (Characters)

(n = 101)

Variable %

B2 Role

Lead 63.4

Major 36.6

a B4 Age

Child 4

Adolescent 14.9

Young Adult 45.5

Mature Adult 30.7

Elderly 5

B5a Gender

Female 41.6

Male 58.4

B6 Race

Caucasian 86.1

African, African-American 7.9

Asian 3.9

Middle Eastern 1 Empirical testing 89

Native-American 2

Bi-racial 0

Other 0

B7 Appearance

Extremely Attractive 8.9

Attractive 25.7

Average 60.4

Unattractive 5

Extremely Unattractive

B8 Occupation

(There were 28 categories)

Student (most frequent) 8.9

B9** Travel

No 66.3

Yes - within country of origin 16.8

Yes - outside country of origin 15.8

Yes- within and outside 1

B10** Live abroad

No 86.1

Yes - part of the film 6.9

Yes - full length of the film 6.9

B12 Love at beginning of film

Not in relationship - not looking 31.7 Empirical testing 90

Not in relationship - looking 11.9

In happy relationship 15.8

In unhappy rel. - not looking 5

In unhappy rel. - looking 34.7

Love situation not apparent 1

B13** Love at middle of film

Not in relationship - not looking 18.8

Not in relationship - looking 11.9

In happy relationship 27.7

In unhappy relationship - not

looking 12.9

In unhappy relationship - looking 1

Love situation not apparent 27.7

B14 Love at end of film 4

Not in relationship - not looking 21.8

Not in relationship - looking 28.7

In happy relationship 2

In unhappy relationship - not

looking 1

In unhappy rel. - looking 30.7

Love situation not apparent 11.9

B15 Sexual Orientation

Heterosexual 80.2 Empirical testing 91

Homosexual 4

Bisexual 3

Unable to determine 12.9

B35 Introvert vs. Extrovert

Introvert 11.9

Extrovert 23.8

In the middle 64.4

B36* Appears lonely

No 41.6

Sometimes 44.6

Most of the time 13.9

B39 Substance abuse

No 92.1

Yes - alcohol 5

Yes - drugs 2

Yes - alcohol and drugs 1

B42 Character dies in film

No 81.2

Natural death - old age 3

Natural death - health issues 3

Accident 2

Murder 8.9

Suicide 2 Empirical testing 92

Unable to determine 0

B43** Issue w/sibling

None present 85.1

Present but no issues 11.9

Present with issues 3

Unknown 0

B44a Issue w/ parent

None present 78.2

Present but no issues 12.9

Present with issues 8.9

Unknown 0

B45 Issue w/ child

None present 87.1

Present but no issues 6.9

Present with issues 5.9

Unknown 0

B46 Issue w/ grandparent

None present 0

Present but no issues 0

Present with issues 0

Unknown 0

B47 Issue w/ grandchild

None present 0 Empirical testing 93

Present but no issues 0

Present with issues 0

Unknown 0

B48** Issue w/ spouse

None present 84.2

Present but no issues 7.9

Present with issues 7.9

Unknown 0

B49 Issue w/ other fam. Member

None present 98

Present but no issues 1

Present with issues 1

Unknown 0

a chi-square .05 < p < .10

* chi-square p < .05

** chi-square p< .01

Research Question 3 asked if Jane Campion films differed from non-Campion films by production techniques. Many one-way ANOVAs for interval or ratio variables were conducted using the director as the independent variable.

Music. Use of non-diegetic music was found to be significant, F(1, 528) = 8.05, p <

0.01, in such a way that Campion content played less non-diegetic music than did non- Empirical testing 94

Campion segments. Use of diegetic music was found to be not significant, F(1, 528) = 0.10, p = 0.75. See Table 24.

Table 24

Analysis of Variance for Section C: Production Techniques in Five-Minute Segments

Non-

Campion Campion Total

(n=167) (n=363) (n=530)

Variable Mean sd Mean sd Mean sd F(1, 528) Sig.

C1: Voiceover 0.17 0.38 0.15 0.36 0.16 0.36 0.54 0.47 C2: Diagetic music 0.35 0.48 0.33 0.47 0.34 0.47 0.10 0.75 C3: Non-diegetic music 0.58 0.49 0.71 0.46 0.67 0.47 8.05** <0.01

C4: Color picture 1.00 0.00 1.00 0.05 1.00 0.04 0.46 0.50

C5: B&W picture 0.04 0.19 0.02 0.14 0.02 0.15 1.32 0.25

C6: Sepia picture 0.03 0.17 0.01 0.09 0.02 0.12 3.63a 0.06

C7: animation 0.02 0.13 0.00 0.05 0.01 0.09 3.54a 0.06

C8: Xcu of motion of touching 0.25 0.65 0.00 0.05 0.08 0.39 49.48** <0.01

C9: Face cu 8.10 6.21 6.85 8.77 7.24 8.07 2.73a 0.10

C9.1: Face xcu 0.68 1.19 0.21 1.00 0.36 1.09 21.47** <0.01 Empirical testing 95

C10: Feet cu 0.12 0.45 0.01 0.12 0.05 0.27 17.60** <0.01

C10.1: Feet xcu 0.05 0.31 0.00 0.00 0.02 0.17 8.88** <0.01

C11: Hand cu 0.59 0.97 0.14 0.49 0.28 0.71 49.24** <0.01

C11.1: Hand xcu 0.22 0.61 0.00 0.05 0.07 0.36 46.78** <0.01

C12: Other cu 0.01 0.08 0.01 0.07 0.01 0.08 0.00 0.95 C13.1: Furniture xcu 0.02 0.13 0.00 0.00 0.01 0.08 6.62** 0.01

C14: Window cu 0.04 0.19 0.02 0.13 0.02 0.15 1.94 0.16 C14.1: Window xcu 0.04 0.19 0.00 0.00 0.01 0.11 13.48** <0.01

C15: Mirror cu 0.10 0.36 0.02 0.16 0.05 0.24 13.64** <0.01 C15.1: Mirror xcu 0.04 0.30 0.00 0.05 0.02 0.17 5.90* 0.02

C16: Tree cu 0.02 0.19 0.02 0.15 0.02 0.16 0.24 0.62

C16.1: Tree xcu 0.01 0.08 0.00 0.05 0.00 0.06 0.32 0.57 C17: Flowers/plants cu 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 C17.1: Flower/plants xcu 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00

C18: Food cu 0.03 0.39 0.02 0.14 0.02 0.24 0.22 0.64

C18.1: Food xcu 0.00 0.00 0.00 0.05 0.00 0.04 0.46 0.50

C19: Animal cu 0.03 0.17 0.04 0.33 0.04 0.29 0.10 0.75 C19.1: Animal xcu 0.02 0.17 0.03 0.37 0.03 0.32 0.25 0.62

C21: Tracking shot 0.08 0.32 0.10 0.39 0.10 0.37 0.37 0.54 Empirical testing 96

C24: Slow motion 0.16 0.68 0.05 0.29 0.08 0.45 6.74** 0.01

C25: Time-lapse 0.04 0.22 0.02 0.27 0.02 0.26 0.48 0.49

C26: Jump cut 0.01 0.08 0.02 0.33 0.02 0.28 0.38 0.54

C27: Blue filter 0.69 3.34 0.01 0.09 0.22 1.90 15.06** <0.01

C28: Red filter 0.05 0.23 0.01 0.09 0.02 0.15 10.94** <0.01

C29: Green filter 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00

C30: Other filter 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00

C34: Hand held camera 0.20 0.68 0.07 0.33 0.11 0.47 8.93** <0.01

C35: Low angle 0.10 0.57 0.21 0.79 0.18 0.73 2.61 0.11

C36: High angle 0.38 1.30 0.52 1.37 0.47 1.35 1.10 0.30

C37: Overview 0.08 0.27 0.10 0.85 0.09 0.72 0.08 0.78

C38: Underview 0.04 0.22 0.02 0.24 0.03 0.23 0.26 0.61 C39: Canted angle 0.06 0.26 0.03 0.20 0.04 0.22 2.03 0.16

C41: Superhearing 0.04 0.20 0.11 0.61 0.09 0.52 1.83 0.18

C42: Extremely long shot 0.17 0.42 0.15 0.52 0.16 0.49 0.29 0.59 Empirical testing 97

C43: Arm - medium shot 0.88 1.15 0.17 0.53 0.40 0.84 94.35** <0.01 C44: Leg - medium shot 0.67 1.15 0.06 0.29 0.25 0.74 91.61** <0.01

C45: Focus pull 0.14 0.43 0.11 0.39 0.12 0.41 0.66 0.42

VC50: Legs 0.84 1.36 0.07 0.31 0.31 0.88 104.58** <0.01

VC51: Arms 1.69 1.74 0.32 0.70 0.75 1.30 168.03** <0.01

VC 52: Up 0.14 0.61 0.24 0.84 0.21 0.77 1.88 0.17

VC53: Down 0.46 1.33 0.61 1.75 0.56 1.63 0.97 0.32 Note: Variables starting with V (e.g., VcC0: legs) are new/constructed

For variables C1 through C6 the mean score is also a proportion

a .05< p < .10

** p <.01

* p < .05.

Picture format. Two variables were found to be near significance: Sepia picture

F(1, 528) = 3.63, p = 0.06, and animated picture, F(1, 528) = 3.54, p = 0.06, indicating that

Campion films include sepia sequences or animated sequences more than do non-

Campion films. Color picture, F(1, 528) = 0.46, p = 0.50, and black and white picture, F(1,

528) = 1.32, p = 0.25, were found to be not significant. Any results connected with these two variables (color picture, and black and white picture) should be examined with Empirical testing 98

caution due to non-calculable reliabilities. The researcher decided to keep these variables

due to the easy nature of the variables. It was assumed, that coders would recognize color

picture format from black and white, sepia and animation easily.

Close up and extreme close up shots. There were ten variables in this variable family found to be significantly different between Campion and non-Campion segments:

Extreme close up shots of motion of touching, F(1, 528) = 49.48, p < 0.01, extreme close up

shots of face, F(1, 528) = 21.47, p < 0.01, close up shots of feet, F(1, 528) = 49.48, p < 0.01,

extreme close up shots of feet, F(1, 528) = 8.88, p < 0.01, close up shots of hand, F(1, 528) =

49.24, p = 0.00, extreme close up shots of hand, F(1, 528) = 46.78, p < 0.01, extreme close

up shots of furniture, F(1, 528) = 6.62, p = 0.01, extreme close up shots of window, F(1, 528) =

13.48, p = 0.00, close up shots of mirror, F(1, 528) = 13.64, p < 0.01, and extreme close up shots of mirror, F(1, 528) = 5.90, p = 0.02, indicating Campion segments include higher

number of the close up and extreme close up shots listed above. The rest of the variables were fount not to be significant when comparing Campion to non-Campion segments:

Close up shots of face, F(1, 528) = 2.73, p = 0.10, close up shots of furniture, F(1, 528) = 1.37, p = 0.24, close up shots of window, F(1, 528) = 1.94, p = 0.16, close up shots of tree, F(1, 528)

= 0.24, p = 0.62, extreme close up shots of tree, F(1, 9528) = 0.32, p = 0.57, close up shots

of food, F(1, 528) = 0.22, p = 0.64, extreme close up shots of food, F(1, 528) = 0.46, p = 0.50,

close up shots of animal, F(1, 528) = 0.10, p = 0.75, and extreme close up shots of animal,

F(1, 528) = 0.25, p = 0.62.

Motion shots. One variable found a significant difference between Campion and

non-Campion films: Slow motion, F(1, 528) = 6.74, p = 0.01, indicating Campion segments included more slow motion effect than did non-Campion segments. The rest of the Empirical testing 99

motion shots variable family were found to be not significant: Tracking shot, F(1, 528) =

0.37, p = 0.54, time-lapse, F(1, 528) = 0.48, p = 0.49, and jump cut, F(1, 528) = 0.38, p = 0.54.

Color filters. Two variables were found to be significant: Blue color filter, F(1, 528)

= 15.06, p < 0.01, and red color filter, F(1, 528) = 10.94, p < 0.01. The other two variables

belonging to this variable family (green color filter, and other color filter) could not be

tested due to lack of variance (i.e., lack of occurrence).

Angle shots. Variables in this family were found to be not significant: Low angle

shots, F(1, 528) = 2.61, p = 0.11, high angle shots, F(1, 528) = 1.10, p = 0.30, overview shots,

F(1, 528) = 0.08, p = 0.78, underview shots, F(1, 528) = 0.26, p = 0.61, and canted angle shots,

F(1, 528) = 2.03, p = 0.16. A computed variable, looking up, combined low angle and

underview shots but was found to be not significant as well, F(1, 528) = 1.88, p = 0.17.

Another combined variable down angles, combined high angle and overview shots and was found to be not significant as well, F(1, 528) = 0.97, p = 0.32. Any results in this

variable family should be looked at with caution due to low reliabilities.

Medium shots of arms, hands, legs and feet. All four variables in this variable

family were found to be significant: Medium shots of arm/hand, F(1, 528) = 94.35, p < .01,

medium shots of leg/foot, F(1, 528) = 91.61, p < .01, all shots of leg/foot, F(1, 528) = 104.58, p

<.01, and all shots of arm/hands, F(1, 528) = 168.03, p < .01. These results indicate that

Campion segments paid more attention and showed more shots of arms, hands, legs, and

feet than did non-Campion. However, due to low reliabilities, these results ought to be

analyzed with caution.

There were four more variables that did not belong to any variable family. One

variable, hand held camera, was found to be significantly different, F(1, 528) = 8.93, p < Empirical testing 100

0.01, indicating Campion segments showed more hand held camera shots than did non-

Campion segments. The rest of the variables found no significant differences between

Campion and non-Campion segments: Superhearing, F(1, 528) = 1.83, p = 0.18, extremely long shots, F(1, 528) = 0.29, p = 0.59, and focus pull, F(1, 528) = 0.66, p = 0.42.

Empirical testing 101

CHAPTER IV

DISCUSSION

The purpose of this study was to empirically test auteur theory on the body of

work of Jane Campion, a female director from New Zealand. The intention of this study was to objectify the subjective nature of auteur theory with evidence collected via an empirical study.

Through the literature, we have learnt that a more scientific approach to auteurism was proposed with the introduction of structuralism. Many of the film critics and scholars such as the Cahier’s editors, Peter Wollen and Allan Lovell, were awaiting such movement with great anticipation. They wrote their critiques of directors and analyses of

their films with structuralist approach, but never took it to the next level of scientific testing of the cinema.

On the other hand, mass communication scholars were, and still are, conducting enormous numbers of empirical studies concerning video games, television programs, advertisements, billboards, music, and film, but failed to include film theories and chose

to ignore film literature. It seems that these two fields of mass media and film studies

were not related, or purposely never joined together. Empirical testing 102

This study is attempting to bring the critical film theories and empirical mass communication traditions together by testing the auteur theory, a critical film theory, with a content analysis method.

At the beginning of this study, I posed a few questions. Can you rely on the subjectivity of film critics? Are you satisfied when a film scholar tells you who is and who is not considered an auteur (the author)? We shouldn’t have to rely on the bias of film scholars, they are after all human beings with different tastes, different views, and

different ideas and who one thinks should be considered an auteur another might not. The

present study introduced an objective way of studying film via auteur theory.

Due to this research, auteur theory can be used not only as a tool for critical

studies, but as a tool for scientific studies as well. It illustrated that auteur theory is

indeed quantifiable, and that directors can be identified not only through the subjective

eye of a film critic, but also through an empirical and scientific study. It is however a first

attempt to do so, therefore many studies need to be conducted to further confirm and extend these results.

As mentioned earlier, many analyses have been written, for example, on John

Ford by Peter Wollen, Cahiers editors, and Andrew Sarris. They identified many themes

such as the love for the West, American history, military and army, the antinomies of

garden versus wilderness, East versus West, and civilized versus savage (Wollen, 1968).

Those themes and his signatures have been identified by the critics, but have never been

tested empirically.

This study reviewed literature and themes identified by film scholars and critics

on Jane Campion, and took it a step forward by testing via the scientific method of Empirical testing 103

content analysis. Detailed discussion of the findings are reported in the next sections.

Research Question 1

The first research question asked whether Jane Campion films differed from non-

Campion films by themes and motifs. Many recurring themes were identified in the literature, pointing to a significant difference between Campion and non-Campion themes, recognizing many imprints Campion leaves on her films.

It was found that there were not as many lead male characters present in Campion films as in non-Campion films. It was also found that in Campion films there were

always lead female characters present difference, this fact however, was only

approaching significance, but supported the findings from the literature that Campion’s

lead characters were always women (Hopgood, 2002; Polan, 2001; Schröder, 2004): Kay

and Sweetie in Sweetie, Janet Frame in An Angel at my Table, Isabel Archer in Portrait of

a Lady, Ada in The Piano, Ruth in Holy Smoke!, and Frannie in In the Cut. It was found

significant, that all Campion’s films had a point-of-view of a female as predicted by the

literature (Gillett, 2004; Gordon, 2002; Hopgood, 2002; Hopgood, 2004; Mellencamp,

1996; Polan, 2001; Schröder, 2004; Sharp & Gillard, 2004).

Campion films revealed more nudity of women and men than non-Campion films.

In her films, female bodies were shown fully naked from both front and back,

significantly more frequently than in non-Campion films as well as full nude frontal

male. In the films of Jane Campion partial nude frontal female body was shown 4.5 times

more than in non-Campion films, top nude frontal female was shown 13.5 times more

than in non-Campion films, full nude frontal female 8 times more than in non-Campion

films, full nude back female 2.5 times more than in non-Campion films, full nude frontal Empirical testing 104 male 3.5 times more than in non-Campion films, full nude back male 3 times more than non-Campion films than in non-Campion films, and bottom nude frontal male body was actually never shown in non-Campion films.

Even though one film included in the random sample, Lilies, was dealing with homosexuality and suppression of gay lovers (this film showed a lot of male nudity), it was still found that Campion films revealed significantly more full frontal nudity of the male body than non-Campion films and almost significantly more full nudity of the back male body than non-Campion films. Campion films showed significantly more kissing, oral sex, and sexual content than non-Campion films supporting the identified themes of exploration of female sexuality (Coomps & Gemmel, 1999; Gillett, 2004; Gordon, 2002;

Hopgood, 2002; Hopgood, 2004; Nelmes, 1999; Polan, 2001; Sharp & Gillard, 2004), sexual obsession and perversity (Gordon, 2002; Mellencamp, 1996). Even though literature did not uncover the fact about the large amount of nudity that Campion films portrayed, it may be closely connected with her obsession with sexuality, female desire, and a hint of perversity as mentioned earlier. In an interview, Campion said that not very many sexual scenes today are supposed to be shown as pleasurable to women (Frencke,

2003). Males were, and still are, dominating the film industry and the world for that matter, but Campion was daring enough to actually show on screen that sexuality is and should be enjoyable for both sexes starting with sexual explorations of Ada and Isabel, and sexually aware and experienced Ruth and Frannie.

Polan identified Campion’s obsession of showing disturbing images and taboos

(2001) which Campion admitted to in an interview by stating she is being drawn to what is not spoken about, what is cruel (Wexman, 1999). This study revealed that Campion Empirical testing 105

films dealt with taboos (Howe, 1990) significantly more than non-Campion films. Such

instances were: the sexual harassment scene in Sweetie, where Sweetie was helping her

dad take a bath dropping the soap and “fishing” for it under the water, or in Portrait of a

Lady, where Osmond’s daughter sat on his lap, him caressing her oh so gently, which

made us, the viewers, extremely uncomfortable thinking he was sexually abusing her (it

is however, only an assumption). Other scenes that can be viewed as shocking, or

perverse are: In An Angel at my Table, where Janet is menstruating and hiding the dirty

cloths from her aunt; the almost rape scenes in The Piano; the naked and urinating on

herself Ruth in Holy Smoke!; or the oral sex scene in In the Cut, to mention a few. As

found in Mellencamp (1996), Freiberg described Campion’s theme as dealing with

“everyday, domestic, and trivial scenes and situations with an edge of menace” (p. 174).

Theme of presence of angels (physical or verbal) was identified in the literature

(Wexman, 1999) and found to be significant in Campion films. No angels were present or talked about in the non-Campion films.

Campion films revealed a more realistic nature of human relationships by engaging her characters in ups and downs of their relations. An overwhelming 88% of her films dealt with spousal break up and 66.7% dealt with break up of lovers compared to non-Campion films where only 5.6% dealt with spousal break up, and 22.2% non-

Campion films dealt with break up of lovers. The “and they lived happily ever after” kind of relationships of the non-Campion films are not the nature of real relationships as we know it, and Campion knows it as well. In an interview, she said, “I am not committed to niceness, I am committed to seeing what’s there . . .” (Wexman, 1999, p. 9). Gordon Empirical testing 106

(2002), Polan (2001), Mellencamp (1996), and Lewis (200o) identified her interest in real-life male and female relationships in her films.

The true reflection of real life in Campion films was also shown with characters having or dealing with health issues. Over half of her films dealt with health problems, compared to barely one fourth of non-Campion films. This finding however was not predicted by the literature. The critics and scholars seemed to either ignore this fact, or did not find it exciting as one of her recurring themes.

Last but not least, her signature identified in literature and partially supported by this study was the recurring theme of loss of freedom. Half of Campion films, as found in this study, dealt with loss of freedom. However, in the literature all of Campion’s lead female characters experienced loss of freedom. Polan (2001) summed it as: Sweetie’s exclusion on a family trip leaving her trapped at home alone, Janet’s physical hospitalization in a mental institution, Ada’s imprisonment in an arrange marriage,

Isabel’s captivity in her marriage, Ruth’s physical imprisonment by her family in a secluded hut, and Frannie’s mental state imprisoning herself as well as being locked up in the light house (Fuller, 2003a). It is evident that all of Campion’s films dealt with imprisonment; however the lack of a complete definition in the code book used in this study allowed for only half of the films being identified as dealing with such phenomenon.

The themes that differentiated films directed by Jane Campion from non-Campion films were: Stories told from a female point of view, with a female lead character, and minimal use of lead male character with an exposure to a vast amount of female and male nudity as well as sexual activity. Campion was also not afraid of showing disturbing Empirical testing 107 images and references to angels. The relationships that her characters engage in are similar to real life, with many break ups, and health issues, as well as power struggles causing the female characters being or feeling imprisoned or entrapped.

Research Question 2

The second research question asked if films of Jane Campion differed from non-

Campion films by narrative construction with characters. Many character traits were found in the literature and supported with the present study.

Interestingly, it was found that Campion’s lead or major characters were more likely to be present in the opening scene than non-Campion characters, and that not one of Campion’s characters was actually happy through majority of the film. It was interesting to find that an overwhelming three-quarters of Campion’s characters were young adults, compared to not even close to half of non-Campion characters, and that

Campion’s female characters outnumbered non-Campion female characters with a ratio of 6:4.

Campion’s lead characters were looking for their identities (Corrigan & White,

2004; Gillett, 2000b; Gillett, 2004; Polan, 2001). In this study, the quest for the character’s identity was broken down into several categories, with a finding that over a third of Campion’s characters searched for their identities by getting married, and two characters by changing religion. It was interesting to find that even thought a majority of

Campion’s films dealt with break ups of romantic relationships, characters were still getting married. This can be compared to many real life experiences.

It was also found that Campion characters experienced more physical abuse from opposite gender characters than non-Campion characters. It may be explained with the Empirical testing 108 fact that a majority of Campion characters were independent, strong women living in and dealing with male dominated societies (Gillett, 2004; Hopgood, 2002; Kelly, 1996; Sharp

& Gillard, 2004). Therefore the women’s independence had to be suppressed by exercising their male power with physical abuse.

It was also found that Campion’s characters were found to be sexually more abusive to opposite gender characters than non-Campion characters, and also a tendency

(even though not significantly different between Campion and non-Campion films) that

Campion characters experienced more of a sexual abuse by opposite gender characters than non-Campion characters. This again can be explained with the male dominance theme, assuming the sexual abuse done to females by males. From the films and literature, we know that Sweetie was most probably sexually molested by her father, Ada was almost twice raped by her husband, Isabel was physically abused by her husband,

Frannie was attacked and sexually assaulted, and Pauline was raped and killed.

As the literature and findings of this study suggested, most of the female Campion protagonists were labeled or viewed as mad, or crazy, identifying another recurring construction of the narrative with characters. As mentioned earlier, it may be explained with the unconformity of Campion’s female characters, their rebellion against the traditional gender roles, and disobedience of the patriarchal order (Hopgood, 2002), hence viewed and labeled as crazy.

It was also found that Campion constructs her narrative with characters that are on the go, traveling or living abroad. In the literature, Campion’s lead characters were described as women on the move, or displaced heroines (Gillett, 2000b, Redding &

Brownworth, 1997; Schröder, 2004). Overwhelmingly 63% of Campion’s characters Empirical testing 109

engaged in traveling, compared to 26% of non-Campion, and 42% of her character lived

abroad for at least some time if not the full length of the film, compared with 7% of non-

Campion characters. It may be connected with the quest for identity, looking for oneself.

The majority of Campion’s female protagonists did travel and live abroad: Janet and

Isabel to Europe, Ruth to India. Sweetie and Frannie however, seemed to stay where they

were, even thought Sweetie was forbidden to go on a family trip, and Frannie was on a

virtual voyage of her mind– going back and forth in time. She was also on a poetic quest,

finding new words, new meanings, as well as on a hunt for the murderer of her sister.

As mentioned earlier, Campion films dealt with real life relationships, where the characters dealt with ups and downs or romantic relationships. It was found that in the middle of Campion films almost half of her characters were not happy in the relationship they were in, but were not looking for a new one, compared with almost a third of non-

Campion characters being actually happy. This fact can be tied in the fact that the overwhelming majority of Campion’s characters were lonely, compared to only half of non-Campion characters.

The theme of dysfunctional families (Hinson, 1990; Hopgood, 2002; Howe, 1990;

Mellencamp, 1996; Redding & Brownworth, 1997; Sharp & Gillard, 2004) was measured by narrative construction of characters experiences with issues with siblings, parents, children, grandparents, grandchildren, a spouse or other family members. It was found that Campion’s characters engaged in significantly more issues with siblings (none of the non-Campion characters dealt with such issues), parents, and spouse than non-Campion characters. One-third of Campion’s characters dealt with spousal issues, compared to the overwhelming majority of non-Campion never dealing with such a problem. Again this Empirical testing 110

can be indicating the fabrication of on-screen relationships with the fairytale “and the

lived happily ever after” notion prevalent in non-Campion films.

A main trademark of Jane Campion’s narrative construction is the hint of sadness

and loneliness of her characters, due to the experience of some physical and sexual abuse by opposite gender characters, assuming that her female characters are receivers of such

abuse by her main male characters. The real nature of her constructed relationships can

be considered another trademark of Jane Campion’s characters: The ups and downs of the

romantic relationships, and the everyday family issues that they went through. Campion

once said in an interview, that in a way family issues are funny and tragic at the same

time, and that one can not escape his/her family, they carry family ties through their

whole lives (Wexman, 1999). More trademarks were identified as quests for identities by

finding life partners, by traveling and living abroad.

Research Question 3

The third research question asked if Jane Campion films differed from non-

Campion films by production techniques. In the literature, not very many specific

techniques were identified.

It was found that Campion films did not include as much non-diegetic music as

non-Campion films. In Campion films on the other hand, more segments included hand

held camera movement, use of blue and red color filter as well as segments with slow

motion shots.

One of Jane Campion’s production techniques identified by the literature was use

of unpredictable angles (Keough, 1999), brilliant use of high, as well as low angles

(Corrigan & White, 2004). This study found that Campion’s segments did not include Empirical testing 111

nearly as many low angle shots as non-Campion segments, therefore not supporting those

statements. However, it may be assumed that the measurement of the low and high angle

variables did not correspond with what Keough (1999) called unpredictable, this study

simply measured number of occurrences, not the uniqueness. Schröder (2004) also

identified the complexity of sound, the constant humming, wind, or squeaking of piano

keys, that was measured in the study (with a superhearing variable) but it was not found

to be her signature trademark as well.

Many critics and scholars pointed out Campion’s use of wide shots of New

Zealand scenery and vegetations (Schröder, 2004; Thompson & Bordwell, 2003;

Wexman, 1999). This study measured the amount of extreme wide shots of scenery, but

no real difference was found between Campion and non-Campion films. Again, the

beauty or originality of such shots was not measured.

It was found that Campion included more animation and sepia format shots in her

films than non-Campion, probably because of her surrealistic images of talking beans,

vivid story telling, animated travelogue, hallucinating images, or dream sequences

(Hopgood, 2004; Polan, 2001).

Campion’s obsession with and interest in the human body is seen in one of her

production technique signatures- the use of close up and extreme close up shots of

people, be it their faces, arms, hands, legs, or feet. Polan (2001) pointed out Campion’s frequent use of close up or extreme close up shots, and Thomson and Bordwell (2003) identified the close up shots of faces. Campion’s segments contained significantly more

close up and extreme close up shots of human bodies, as well as mirrors, and extreme

close up shots of furniture and windows than non-Campion segments. Empirical testing 112

From pilot work, the researcher identified a possible Campion trademark of medium shots of arms and hands as well as medium shots of legs or feet as secondary elements included in the frame. It was observed that Campion liked to include body limbs in the picture. This study supported this observation, with a finding that Campion’s films contained more medium shots of arms, hands, legs, and feet than non-Campion films.

It can be concluded that a main trademark of Jane Campion’s production technique is her obsession of showing images of human body in close up or extreme close up shots. The images of furniture, mirrors, and windows are also close to her as well as slow motion technique, hand held camera shots, and color filters. Campion’s signature is also the use of animation, and other color formats such as sepia.

Even though most of the production techniques were not a priori identified via reviewing analyses of Jane Campion’s films, we can still identify those as her unique imprints and trademarks. When watching her films, one can not forget the extreme close up shots of little Janet’s face, the unflattering but yet gorgeous shots of Pauline’s face,

Baines’ erotic but romantic touches of Ada’s body, the poetic slow motion shots, or her fascination with human hands and feet. Campion is not afraid to challenge or shock the viewers by exposing them to some day-to-day routines, such as menstruation and urinating, that women experience but when shown on screen are embarrassing, and/or shocking. So is the exploration of female lust and sexuality or exposure of fully naked bodies on screen. We are not used to seeing fully naked bodies on screen, we are not used to seeing much sex scenes in regular films, but in real life it is a different story. Campion is interested in real life; her films mirror real life situations, relationships, abuse, loneliness, and family issues. Empirical testing 113

This study was based on the already recognized signatures and trademarks of Jane

Campion’s films, as well as selected signatures of other directors/auteurs with three

research questions in mind. All the trademarks are presented in the Appendix E

(Campion’s signatures collected from the literature, Campion’s signatures identified by

the researcher from pilot work, the production techniques collected from the literature,

and non-Campion trademarks collected from literature).

The three research questions and their answers combined give an overall answer to a broader spectrum question: Will the results of this scientific study indicate that Jane

Campion is an auteur without the subjective eyes of film critics? This study generally supported the subjective views of films critics and film scholars that Jane Campion is an

auteur by uncovering many recurring themes, narrative constructions, and production

techniques that empirically clarifies her precise trademarks and signatures through the

scientific method. Most of the predicted themes and trademarks were confirmed, but

some were not, and others were newly discovered. The full list of the trademarks can be

found in Appendix E.

It must be noted that few elements might be carving the personality of Campion

films that are not being taken into an account in this study. The fact that Jane Campion is a female director from New Zealand living away from her home land, but sometimes returning to shoot a film there, might be a great distinction among the non-New Zealand male directors randomly selected in the sample. Because there were no variables testing for such differences, we can only speculate. This however brings ideas for future studies.

Empirical testing 114

Limitations and Future Investigation

As mentioned earlier, this is a first attempt at scientific testing of the auteur theory

via content analysis. One issue in this study was with some variables such as mix of

genres. The present study limited the random sample to the drama genre to avoid genre

theory issues, therefore not including a variable that would identify the genre of each

film. The literature however identified the multi-genre trend of Campion films that could

not be supported empirically due to the missing genre variable.

If the nudity variables were coded in Section B by lead and major character such

results would have given us more information. Even though it was found that the

majority of the nudity variables were occurring significantly more in Campion films than

non-Campion, in a study of a different director, it might not have been the case.

One of the major criticisms of auteur theory lies in the heart of the collaboration

tendency. The auteur theory resistance argues that there is not one author of a film, there

are many people who work on the film such as the director, director of photography,

editor, writer, etc. For the next study, I am proposing that the IMDb random sample be

eliminated; instead the population for the sample should be the films that the main artists

such as the director, cinematographer, editor and writer worked on. To use Jane

Campion films as an example, she worked with three cinematographers: Dion Beebe,

Stewart Dryburgh, and Sally Bongers. Films that those three worked on as

cinematographers could constitute the non-Campion films, ensuring a tighter, more precise auteur analysis.

My future goal, with the help of many auteur content analysis studies to come, is to build a master codebook, which would serve the purpose of an objective auteur Empirical testing 115

analysis. With this comprehensive codebook a researcher would be able to choose a

director, and with a little training and coding, auteur status would or would not be

uncovered.

One more limitation is noteworthy: What about the post-structuralist-auteurism?

The introduction of psychoanalysis to auteurism in the 1970s brought a larger issue with

objective analysis – an interpretation of messages. Different people can interpret and

understand the same situation or the same image in many different ways. Is there a way

to empirically test such content? Obviously, the empirical extension of this theoretic

approach would involve a series of audience studies, tapping interpretations and reactions

to an auteur’s content.

Conclusion

This study presented an empirical analysis of Jane Campion films through the

auteur theory lens, which until now, no study has done. The findings of the content

analysis have recognized Jane Campion to be an auteur as did many film critics and

scholars with their opinions. This study also recognized many trademarks, themes,

narrative constructions, and production techniques that distinguish Jane Campion’s style.

It has also demonstrated the potential for empirical assessment of auteurism by revisiting the three stages: romantic – auteurism, structuralist – auteurism, and post – structuralist auteurism. This thesis, uncovered a potential for a fourth stage, an empirical – auteurism.

The proposed fourth stage opened the door to future content analyses of film via the auteur theory lens. Will it be optimistically recognized and welcomed among film Empirical testing 116

scholars, or will it be booed as Jane Campion’s Sweetie at Cannes? That is a question awaiting an answer. Empirical testing 117

REFERENCES

Allen, M. P., & Lincoln, A. E. (2004). Critical discourse and the cultural consecration of American films. Social Forces, 82, 871-893.

Angelic, R., Brown, E., Ivernel, F., Kidman, N., Parker, L. (Producers), & Campion, J. (Director). (2003). In the cut. [Motion Picture]. USA: Pathe Productions.

Allingham, D., Barber, G., Mora, P., Redshaw, P., Simons, E., Strieber, R., Strieber, W. (Producers), & Mora, P. (Director). (1989). Communion. [Motion Picture]. Great Britain: Allied Vision Ltd.

Avildsen, J., Schiffer, M., Seelig, D., Twain, N. (Producers), & Avildsen, J. (Director). (1989). Lean on me. [Motion Picture]. USA: Warner Bros. Pictures.

Banerjee, M., Capozzoli, M., McSweeney, L., & Sinha, D. (1999). Beyond kappa: A review of interrater agreement measures. Canadian Journal of Statistics, 27(1), 3- 23.

Barber, G., Liroff, M., MacDonald, M., Robinson, J., Segal, J. (Producers), & Shapiro, A. (Director). (1993). The crush. [Motion Picture]. USA: Morgan Creek Productions.

Baywater, T., & Sobchack, T. (1989). An introduction to film criticism: Major critical approaches to narrative film. New York, NY: Longman.

Bazin, A. (1967). What is cinema? Vol. 1. Berkeley, CA: University of California Press.

Bazin, A. (1971). What is cinema? Vol. 2. Berkeley, CA: University of California Press.

Becker, H., Carroll, E., Lipper, K., Mack, T., Mulvehill, C., Pressman, E. (Producers), & Becker, H. (Director). (1996). City hall. [Motion Picture]. USA: Columbia Pictures Corporation. Empirical testing 118

Bentley, G. (2002). Mothers, daughters, and (absent) fathers in Jane Campion’s The Piano. Literature Film Quarterly, 30, 46-59.

Betz, M. (2001). The name above the (sub) title: Internationalism, coproduction, and polyglot European art cinema. Camera Obscura, 16, 1-46.

Bishop, C., Chapman, J., Goldstein, J., Turnbull, M., Weinstein, B., Weinstein, H. (Producers), & Campion, J. (Director). (1999). Holy smoke!. [Motion Picture].USA: Miramax.

Blomquist, A., Cohen, B., Gladstein, R., Holleran, L., Platt, M., Poster, M., Weistein, B., Weistein, H., Yacoub, L. (Producers), & Hallström, L. (Director). ( 1999). The cider house rules. [Motion Picture]. USA: Miramax.

Blonski, A., Creed, B., & Frieberg, F. (1987). Don’t shoot darling!: Women’s independent filmmaking in Australia. Richmond, Australia: Greenhouse.

Brandstein, J., Duff, J., Estevez, E., Graham-Rice, T., Ivers, J., Krevoy, B., McDermontt, M., Oman, C., Richards, P., Stabler, S. (Producers), & Estevez, E. (Director). (1996). The war at home. [Motion Picture]. USA: Touchstone Pictures.

Brown, S., Downer, J., Fleder, G., Mankiewicz, C., Milchan, A. (Producers), & Fleder, G. (Director). (2003). Runaway jury. [Motion Picture]. USA: New Regency Pictures.

Bruzzi, S. (1999). Tempestuous petticoats. In F. Coombs & S. Gemmell (Eds.), Piano lessons: Approaches to The Piano (pp. 97-107). Sydney, Australia: John Libbey & Company.

Buscombe, E. (1981). Ideas of authorship. In J. Caughie (Ed.), Theories of authorship (pp. 22-34). London: Routledge.

Buse, P., Nuria, T., & Andrew, W. (2004). The Spanish “popular auteur” Alex de la Iglesia as polemical tool. New Cinemas: Journal of Contemporary Film, 2, 139- 148.

Cahiers du Cinema editors. (1976). John Ford’s young Mr. Lincoln. In W. Nichols (Ed.), Movies and methods: An anthology (pp. 493-529). Berkeley, CA: University of California Press.

Calhoun, J. (2003). Interior landscapes. American Cinematographer, 11, 76-85.

Campion, J., Chapman, J. (Producers), & Campion, J. (Director). (1986). Two friends. [Motion Picture]. USA: Milestone. Empirical testing 119

Carlson, R., Kim, C., Osborne, B., (Producers), & Bailey, J. (Director). (1990). China moon. [Motion Picture]. USA: Tig Productions.

Carmody, D., Jacobs, L., Moroney, J., Ransohoff, M., Robinson, B. (Producers), & Lumet, S. (Director). (1993). Guilty as sin. [Motion Picture]. USA: Hollywood Pictures.

Cass, R., Gelbart, A., Stratton, A. (Producers), & Greyson, J. (Director). (1996). Lilies. [Motion Picture]. Canada: Triptych Media Inc.

Caughie, J. (1981). Theories of authorship: A reader. London: Routledge & Kegan Paul Ltd.

Chapman, J., Depardieu, A., Turnbull, M. (Producers), & Campion, J. (Director). (1993). The piano. [Motion Picture]. USA: Miramax Films.

Chumo, P., II (1997). Keys to the imagination: Jane Campion’s The Piano. Literature Film Quarterly, 23(3), 173-177.

Cohen, B., Frauenfelder, K., Jinks, D., Schmidt, A., Zanuck, R. (Producers), & Burton, T. (Director). Big fish. [Motion Picture]. USA: Columbia Pictures Corporation.

Combs, R. (1990). Cinema’s vision thing. Listener, 124, 36-37.

Cook, P. (1986). The cinema book. New York, NY: Pantheon Books.

Coombs, F., & Gemmell, S. (Eds.). (1999). Piano lessons: Approaches to The Piano. Sydney: John Libbey.

Corrigan, T. (1991). A cinema without walls: Movies and culture after Vietnam. New Brunswick, NJ: Rutgers University Press.

Corrigan, T., & White, P. (2004). The film experience: An introduction. Boston, MA: Bedford/ St. Martin’s.

Corliss, R. (1974). The Hollywood screenwriter. In G. Mast & M. Cohen (Eds.), Film theory and criticism: Introductory readings (pp. 541-550). New York, NY: Oxford University Press.

Cotone, M., Reisman, L., Rizzoli Jr., A., Thompson, J. (Producers), & Schrader, P. (Director). (1990). The comfort of strangers. [Motion Picture]. Great Britain: Sovereign Pictures.

Croft, S. (1982). Authorship and Hollywood. Wide Angle, 5(3), 16-22. Empirical testing 120

Dalton, M., Fatzinger, K. (2003). Choosing silence: Defiance and resistance without voice in Jane Campion’s The Piano. Women and Language, 26, 34-39.

Daly, J., Gibson, D., Guest, D., Lewitt, E. (Producers), & Folley, J. (Director). (1986). At close range. [Motion Picture]. USA: Hemdale Film Corporation.

Daly, J., Gibson, D., Liuzzi, G., Pillsbury, S., Sanford, M., Streit, D. (Producers), & Hunter, T. (Director). (1986). River’s edge. [Motion Picture]. USA: Hemdale Film Corporation.

Delaney S., Dupin, C., Gee, S., Gordon, S., Harrow, A., Khan, A., et al. (2002). 16+ source guides: Auteur theory/ Auteurs. Retrieved November 1, 2005 from www.bfi.org.uk

Dick, B. (2005). Anatomy of film. Boston, MA: Bedford/St.Martin.

Douglas, J., Douglas, M., Drogin, M., Orent, K., Schepisi, F., Wigutow, J., (Producers), & Schepisi, F. (Director). (2003). It runs in the family. [Motion Picture]. USA: Buena Vista Pictures.

Egoyan, A., Frieberg, C., Ranvaud, D. (Producers), & Egoyan, A. (Director). (1989). Speaking parts. [Motion Picture]. Canada: Ego Film Arts.

Eysenck, H., & Eysenck, M. (1985). Personality and individual differences: A natural science approach. New York, NY: Plenum Publishing.

Felperin, L. (2003). In the Cut. Sight and Sound, 12, 37-38.

Fisher, P. (1999). Reviews/commentary: Holy Smoke. Retrieved November 11, 2005 from http://members.tripod.com/filmlover2/id50.htm

Frencke, L. (2003). Jane Campion: Dangerous liaisons. Sight and Sound, 11, 19.

Fuller, G. (2003a). Sex and self-danger. Sight and Sound, 11, 16-19.

Fuller, G. (2003b). Shots in the dark. Interview, 33(10), 70.

Garcia, M., Thompson, B., Thompson, W. (Producers), & Thompson, B. (Director). (1999). Stonebrook. [Motion Picture]. USA: First Ignite.

Gillett, S. (2000a). Angel from the mirror city: Jane Campion’s Jane Frame. Senses of cinema. Retrieved November 11, 2005 from http://www.sensesofcinema. com/contents/00/10/angel.html

Empirical testing 121

Gillett, S. (2000b). Never a native: Deconstructing home and heart in Holy Smoke. Senses of cinema. Retrieved November 11, 2005 from http:// www.sensesofcinema.com/ contents/00/5/holy.html

Gillett, S. (2001). A pleasure to watch: Jane Campion’s narrative cinema. Screening the past. Retrieved from http://www.screeningthepast.media.latrobe.edu/ archives

Gillett, S. (2004). Engaging Medusa: Competing myths and fairytales in In the Cut. Senses of cinema. Retrieved November 11, 2005 from http://www.sensesof cinema.com/contents/04/31/in_the_cut.html

Golin, S., Leonardt, U., Montgomery, M., Turnbull, M., Wingate, A. (Producers), & Campion, J. (Director). (1996). Portrait of a lady. [Motion Picture]. USA: Polygram Filmed Entertainment.

Gordon, R. (2002). Portraits perversely framed. Film Quarterly, 56(2), 14-25.

Goss, B. M. (2004). Steven Soderbergh’s The Limey: Implications for the auteur theory and industry structure. Popular Communication, 2, 231-255.

Grant, C. (2000). www.auteur.com?. Screen, 41, 101-108.

Gurney, D. (2006). Film and authorship [Review].Velvet Light Trap, 57, 101-104.

Haley, M., Levinson, B., McCormick, P., Solan, A., Weinstein, P. (Producers), & Levinson, B. (Director). (1999). Liberty Heights. [Motion Picture]. USA: Baltimore Pictures.

Hart, K. (2003). Auteur/brioleur/provocateur: Gregg Araki and postpunk style in The Doom Generation. Journal of Film & Video, 55(1), 30-38.

Hayward, S. (1998). Cinema studies: The key concepts. New York, NY: Routledge.

Hendershot, C. (1998). (Re) visioning the gothic: Jane Campion’s The Piano. Literature Film Quarterly, 26(2), 97-109.

Henke, S. (2000). Jane Campion frames Janet Frame: A portrait of the artist as a young New Zealand poet. Biography, 23, 651-669.

Hinson, H. (1990, March 2). Sweetie. Washington Post. Retrieved November 11, 2005, from http:www.washingtonpost.com

Hopgood, F. (2002). Jane Campion. Senses of Cinema. Retrieved November 11, 2005 from http://www.sensesofcinema.com/contents/directors/02/campion.html

Empirical testing 122

Hopgood, F. (2004). Inspiring passion and hatred. Metro, 139, 28-32.

Howe, D. (1990, March 2). Sweetie: Critical Review. Washington Post. Retrieved November 11, 2005, from http:www.washingtonpost.com

Ikin, B., Major, G., Maynard, J. (Producers), & Campion, J. (Director). (1990). An Angel at my table. [Motion Picture]. Australia: ABC.

Kathi, M. (1994). Sick puppy auteur. Fantasy & Science Fiction, 86, 121-127.

Kelly, D. (1996). The lady in the frame: Two portraits by Henry James and Jane Campion. Senses of Cinema. Retrieved November 11, 2005 from http://www. sensesofcinema.com/contents/01/18/lady_frame.html

Keough, P. (1999). Piano lessons. Retrieved November 11, 2005 from http://www. bostonphoenix.com/archive/movies/99/01/28/JANE_CAMPION_A_COMPLETE RE.html

Kokin, K., Ripp, A., Singer, B. (Producers), & Singer, B. (Director). (1993). Public access. [Motion Picture]. USA: Cinemabeam.

Krippendorff, K. (1980). Content analysis: An introduction to its methodology. Beverly Hills, CA: Sage.

Le Cain, M. (2001). Voyeurism of the soul: The films of Philippe Garrel. Senses of Cinema. Retrieved November 17, 2006 from http://www.sensesofcinema.com/ contents /01/12/garrel.html

Lewis, J. (2000). Wholly Jane: Jane Campion on her new movie and other mysteries. LA Weekly. Retrieved November 11, 2005 from http://www.laweekly.com

Lombard, M. (1997). Structural features content analysis project. Retrieved April, 2006 from http://astro.temple.edu/~lombard/research/sf_man2p1.html

Lovell, A. (1977). Don Siegel: American cinema. London, England: C. F. Hodgson & Son.

MacKinnion, W., Maynard, J. (Producers), & Campion, J. (Director). (1989). Sweetie. [Motion Picture]. Australia: Arenafilm.

Maule, R. (2002). The importance of being a film author: Germain Dulac and female authorship. Senses of Cinema. Retrieved October 12, 2005 from http://www.sensesof cinema.com

McDonald, N. (2000). Writers, directors, and photographers. Quadrant, 44, 66-71.

Empirical testing 123

McHugh, K. (2001). “Sounds that creep inside you”: Female narration and voiceover in the films of Jane Campion. Style, 35, 193-218.

McKew, M. (1999). Jane Campion and Holy Smoke. Retrieved November 11, 2005 from www.abc.net.au/7.30/stories/s73088.htm.

Mellencamp, P. (1996). A fine romance: Five ages of film feminism. Philadelphia, PA: Temple University Press.

Miller, T., & Stam, R. (2000). A companion to film theory. Malden, MA: Blackwell Publishing.

Mosher, M. (2004). The visual turn: Classical film theory and art history. Leonardo, 37, 164.

Nacache, J. (2005). Group portrait with a star: Jeanne Balibar and French “jeune” cinema. Studies in French Cinema, 5, 49-60.

Nelmes, J. (1999). Women and film. In J. Nelmes (Ed.), An introduction to film studies (pp. 268-305). London: Routledge.

Neuendorf, K. A. (2002). The content analysis guidebook. Thousand Oaks, CA: Sage.

Nichols, J., & Walsh, D. (2000). The filmmakers can’t help themselves. Retrieved November 11, 2005 from http://www.wsws.org.

Nystrom, D. (2004). Hard hats and movie brats: Auteurism and the class politics of the New Hollywood. Cinema Journal, 43 (3), 18-41.

Palmer, P., Sellers, A., Van Wyck, J., Winitksy, A. (Producers), & Ritt, M. (Director). (1990). Stanley & Iris. [Motion Picture]. USA: Metro-Goldwyn- Mayer.

Paul, M. (1999). Her side of the story: Reading of Mander, Mansfield & Hyde. Dunedin, New Zealand: University of Ontago Press.

Petrova, V. (2006). Auteur discourse and the cultural consecration of American films. Poetics, 34, 180-203.

Polan, D. (2001). Jane Campion. London: BFI Publishing.

Powers, S., Rothman, D., & Rothman, S. (1993). Transformation of gender roles in Hollywood movies: 1946-1990. Political Communication, 10, 259-283.

Poynor, R. (2001). Auteur theory. Print, 55(1), 38-41.

Empirical testing 124

Pryor, R., Wilson, J. (Producers), & Pryor, R. (Director). (1986). Jo Jo Dancer, your life is calling. [Motion Picture]. USA: Columbia Pictures Corporation.

Pullinger, K. (1999). Soul survivor – Women directors’ special. Sight and Sound. Retrieved November 11, 2005 from http://www.sightandsound.com.

Rabiger, M. (1997). Directing, film techniques and aesthetics. Newton, MA: Focal Press.

Raphael, F. (1997). The auteurs behind the auteurs. The Times Literary Supplement, 1, 4- 7.

Ray, N. (1991). Ray’s world: According to Ray. Film Comment, 27 (5), 47-49.

Redding, J. M., & Brownworth, V. A. (1997). Film fatales: Independent women directors. Seattle, WA: Seal Press.

Rossenbaum, J. (1991). Guilty by omission. Film Comment, 27 (5), 42-46, 50-57.

Sarris, A. (1962). Andrew Sarris: Notes on the auteur theory in 1962. Film Culture, 27, 1-8.

Sarris, A. (1968). American cinema: Directors and directions 1929 – 1968. New York, NY: Dutton.

Sarris, A. (1976). Towards a theory of film history. In W. Nichols (Ed.), Movies and methods: An anthology (pp. 237-251). Berkeley, CA: University of California Press.

Sarris, A. (1990). Auteurism is alive and well. Film Comment, 26 (4), 19-22.

Sarris, A. (1999). Winslet joins Keitel in Campion sequel. Retrieved November 11, 2005 from http://members.tripod.com/filmlover2/id50.htm

Sarris, A. (2003). The auteur theory revisited. In V. W. Wexman (Ed.), Film and authorship (pp. 21-29). New Brunswick, NJ: Rutgers University Press.

Schröder, N. (2004). Slavni filmovy reziseri: 50 nejvyznamejsich reziseru, od Chaplina az po Almodovara [Famous film directors: 50 most important directors, from Chaplin to Almodovar]. Czech Republic: Slovart.

Sharp, H., & Gillard, G. (2004). A path of great courage: The Piano. Australian Screen Education, 35, 109-112.

Shaw, D. (2002). Isabel Archer: Tragic protagonist or pitiable victim. Literature/Film Quarterly,2, 249-255. Empirical testing 125

Smith, A. (1999). Girls on film: Analysis of women’s images in contemporary American and “Golden Age” Hollywood films. Unpublished master’s thesis, Cleveland State University, Cleveland, OH.

Smith, G. (2004). Tough love. Film Comment, 40(4), 60-62.

Stam, R. (2000). Film theory: An introduction. Malden, MA: Blackwell Publishing.

Stam, R., Burgoyne, R., & Fliterman-Lewis, S. (1992). New vocabularies in film semiotics: Structuralism, post-structuralism, and beyond. London: Routledge.

Stam, R., & Miller, T. (2000). Film and theory: An anthology. Malden, MA: Blackwell Publishing.

Tasker, Y. (2002). Fifty contemporary filmmakers. London: Routledge.

Taubin, A. (2003). The wrong man. Film Comment, 39(6), 51-52.

Thompson, K., & Bordwell, D. (2003). Film history: An introduction (2nd ed.). New York, NY: McGraw-Hill.

Truffaut, F. (1976). A certain tendency of the French cinema. In W. Nichols (Ed.), Movies and methods: An anthology (pp. 224-237). Berkeley, CA: University of California Press.

Walsh, D. (1994). Jane Campion’s The Piano: A sensitive touch to a fairly selfish theme. Retrieved November 11, 2005 from http://www.wsws.org

Weber, R. P. (1990). Basic content analysis (2nd ed.). Newbury Park, CA: Sage.

Weiss, M. W. (1998). “a film by…”. The Independent Film and Video, 21 (2), 30-31, 60-62.

Wexman, V. W. (1999). Jane Campion interviews. Jackson, MS: University Press of Mississippi.

Wexman, V. W. (2003). Film and authorship. New Brunswick, NJ: Rutgers University Press.

Wollen, P. (1969). Signs and meaning in the cinema. London: BFI. Empirical testing 126

APPENDICES

Empirical testing 127

APPENDIX A

CODEBOOK

SECTION A., B., & C.

Coding Instructions: ƒ First code section C – it will help with coding Section A. & B. ƒ If you want, it is possible to code Section A. & B. at once ƒ If you are asked to GIVE COUNTS, it will make it easier to record if you make a check mark for every occurrence, then report the total ƒ Code from the beginning of credits except when credits are shown over abstract images, graphics, or plain background ƒ Code each film with closed captioning (if available) ƒ All coding needs to be done with the information provided in the movie, and only in the movie, assume no prior knowledge.

Please fill in all the information required before coding each film:

1. Coder ID: Indicate the individual who coded the film, according to the coder ID list:

Coder ID Coder Name 1 Patrika Janstova 2 Kim Neuendorf 3 Benjamin Simon

2. Film ID: Identify each film with a number (1 through 28), according to the list below.

3. Director ID: Identify film’s director (1= Jane Campion, 2 = non-Jane Campion) according to the list below.

4. Year ID: Identify the year film was made according to the list below.

Film Director Year ID Film Title Film Director ID Year ID 1 Communion Philippe Mora 2 1989 2 2 Lean On Me John G. Avildsen 2 1989 2 3 Speaking Parts Atom Egoyan 2 1989 2 4 Sweetie Jane Campion 1 1989 2 Empirical testing 128

Film Director Year ID Film Title Film Director ID Year ID 5 The Comfort of Strangers Paul Schrader 2 1990 3 6 China Moon John Bailey 2 1990 3 7 Stanley and Iris Martin Ritt 2 1990 3 8 An Angel at My Table Jane Campion 1 1990 3 9 The Crush Alan Shapiro 2 1993 4 10 Guilty as Sin Sidney Lumet 2 1993 4 11 Public Access Bryan Singer 2 1993 4 12 The Piano Jane Campion 1 1993 4 13 City Hall Harold Becker 2 1996 5 14 Lilies John Greyson 2 1996 5 15 The War at Home Emilio Estevez 2 1996 5 16 The Portrait of a Lady Jane Campion 1 1996 5 17 The Cider House Rules Lasse Hallstrom 2 1999 6 18 Liberty Hights Barry Levinson 2 1999 6 19 Stonebrook Byron W. Thompson 2 1999 6 20 Holy Smoke! Jane Campion 1 1999 6 21 Big Fish Tim Burton 2 2003 7 22 It Runs in the family Fred Schepisi 2 2003 7 23 Runaway Jury Gary Fleder 2 2003 7 24 In the Cut Jane Campion 1 2003 7

Fill in the date that the coding was completed (mm/dd/yyyy)

6. Closed captioning: Identify if you were able to watch the movie with or without it. 0. OFF 1. ON

SECTION A.

Unit of data collection: Full length feature film – 60 minutes or longer film

Unit of analysis: Full length feature film ______

Empirical testing 129

DEFINITIONS AND EXAMPLES:

LEAD ROLE CHARACTER = the whole story revolves around this character, and his or her presence is essential to the story. (Ex. It is possible that the film will have only a male lead character, not a female lead.) MAJOR ROLE CHARACTER = is one that appears in over 50% of the film but does not appear to be lead character. If this his character is crucial to the plot line ( meaning if the character was not there, the story would change) but did not appear in over 50% of the film, code the character as a major role character. MEDIUM ROLE CHARACTER = is a character that speaks more than 5 sentences, but does not appear to be a major character. This character appears in less than 50% the film’s speaking scenes and is not crucial to the plot line.

Note: To help with the count of the characters, please use the IMDB worksheet attached to each movie.

A1. Number of lead female role characters: __ (count)

A2. Number of major female role characters: __ (count)

A3. Number of medium female role characters: __(count)

A4. Number of lead male role characters: __(count)

A5. Number of major male role characters: __(count)

A6. Number of medium male role characters: __ (count)

DEFINITIONS AND EXAMPLES:

FATHER/MOTHER = a person known to be a father or a mother of any minor, medium, major or lead character visible in the film PHYSICALLY PRESENT= character’s existence is visible or audible to the viewer or to their child (a character) in the film. (ex. A phone conversation between a character and his mother, even though the mother is not visible on the screen.

A7.1. A father figure of one or more lead characters was physically present in the film. 0. No 1. Yes

A7.2. A father figure of one or more major characters was physically present in the film. 0. No 1. Yes Empirical testing 130

A7.3. A father figure of one or more medium characters was physically present in the film. 0. No 1. Yes

A7.4. A father figure of minor/other character was physically present in the film. 0. No 1. Yes

A7.5. A father figure that was not physically present in the film was talked about in the film. 0. No 1. Yes

A8.1. A mother figure of the lead character/s was physically present in the film. 0. No 1. Yes

A8.2. A mother figure of major characters was physically present in the film. 0. No 1. Yes

A8.3. A mother figure of medium character/s was physically present in the film. 0. No 1. Yes

A8.4. A mother figure of minor/other character/s was physically present in the film. 0. No 1. Yes

A8.5. A mother figure that was not physically present in the film, was talked about. 0. No 1. Yes

DEFINITIONS AND EXAMPLES:

TRADITION = a custom or practice that is passed down from generation to generation – be it cultural customs, religious practices, etc… (Ex. An Easter tradition I have learned from my mother, who learned it from her mother, etc…) MODERNISM = rebellion towards the traditions of the family, culture, religion, etc… (Ex. Westernization of the kids in traditional Asian family) CLASH = a shown, talked, or argued about contrast, or a conflict CULTURE = national, ethnic, religious…. Empirical testing 131

A9. A clash between tradition and modernism was present in the film. 0. No 1. Yes

A10. A clash between different cultures was present in the film. 0. No 1. Yes

DEFINITIONS AND EXAMPLES:

LOSS = unwanted loss, being deprived OCCURRED = an event that took place on the screen or was talked about “BREAK UP” = when two people go separate ways. Be it literal break up among the characters, or a loss of a relationship, they become estranged, alienated. YOUNG CHILD = the character behaves and speaks as one who is 12 years old or younger. OLDER CHILD: The character behaves and speaks as one who is 13 years old or oler – this includes adult children. LOVER = a person who has a romantic relationship with other person, to whom he/she is or is not married to.

A11.1. Loss of a young child due to death (or miscarriage or abortion) occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.2. Loss of a young child due to ‘break up’ (or miscarriage or abortion) occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.3. Loss of an older child due to death occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.4. Loss of an older child due to ‘break up’ occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) Empirical testing 132

3. Yes – Both Lead/Major character and other characters

A11.5. Loss of a spouse due to death occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.6. Loss of a spouse due to ‘break up’ occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.7. Loss of a lover due to death occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.8. Loss of a lover due to ‘break up’ occurred to: 0. No 1. Yes –Lead/Major Character 2. Yes – Other character 3. Yes – Both Lead/Major character and other characters

A11.9 Loss of a family member due to death occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.10. Loss of a family member due to ‘break up’ occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.11 Loss of a friend due to death occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.12. Loss of a friend due to ‘break up’ occurred to: 0. No-one Empirical testing 133

1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.13. Loss of faith (religion) occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.14. Loss of faith (trust) in character’s own abilities occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.15. Loss of faith (trust) in other character’s abilities occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.16. Loss of health (permanent and life changing) occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.17. Loss of material thing(s) occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.18. Loss of freedom – imprisonment in a facility against ones will. 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.19. Loss of a job occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

Empirical testing 134

A11.20. Loss of money occurred to: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.21. Loss of physical beauty: 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

A11.22. Other major loss occurred to (Write in): 0. No-one 1. Yes –Lead/Major character(s) 2. Yes – Other character(s) 3. Yes – Both Lead/Major character and other characters

DEFINITIONS AND EXAMPLES:

PARTIAL NUDE (FEMALE) = when a female character is shown unclothed, except for the area that would typically be covered by bikini area – top and bottom; or she is in bed assumed to be naked BUT her bikini areas are covered under the sheet; or if the camera chooses to show only the “safe”/ “appropriate” parts of the naked body. (Ex. Female is standing – naked, but a bouquet is covering her bikini area - female’s breast and sex organ are covered).

TOP NUDE (FEMALE) = topless = when a female character is shown unclothed, except the bikini bottom area is covered; or if a character is fully naked but the camera chooses to show only the “safe”/ “appropriate” parts of the body; or if a character is assumed to be naked, but his boxer area is covered under the sheets. (Ex. Female without a bra, but wearing panties or pants female’s chest is visible to the viewer).

TOP NUDE (MALE) = when a male character is shown unclothed, except the boxers / pants area is covered; or if a character is assumed to be naked, but his boxer area is covered under the sheets or if a character is fully naked but the camera chooses to show only the “safe”/ “appropriate” parts of the body. (Ex. Male in briefs/boxer shorts/pant male’s chest is visible to the viewer).

BOTTOM NUDE (FEMALE AND MALE) = bottomless = when a male or female character is shown unclothed, except the bikini top /T-shirt. Or if a character is assumed to be naked, but his boxer area is covered under the sheets. (Ex. Male in T-shirt, but nothing on the bottom, the character’s sex organ is visible to the viewer).

FULL NUDE (FEMALE AND MALE) = topless and bottomless = you can clearly see that the character is fully naked, and 80% or more of the body is visible on the screen. Empirical testing 135

Note: Count each shot and each different ADULT character. Also, if a character is in swimming suite – and uses it for the purpose of swimming or suntaning – do not count it as partial nudity. However – if it is NOT used for such activities – code it appropriately. (Ex. A girl at a bar in bikini top and jeans = partial nude female – frontal or back side).

Female:

A12. Partial nude frontal female body was exposed: __(count)

A13. Top nude frontal female body was exposed: __(count)

A14. Bottom nude frontal female body was exposed: __(count)

A15. Full nude frontal female body was exposed: __(count)

A16. Partial nude back side of a female body was exposed: __(count)

A17. Top nude back side of a female body was exposed: __(count)

A18. Bottom nude back side of a female body was exposed: __(count)

A19. Full nude back side of a female body was exposed: __(count)

Male:

A20. Top nude frontal male body was exposed: __(count)

A21. Bottom nude frontal male body was exposed: __(count)

A22. Full nude frontal male body was exposed: __(count)

A23.Top nude back side of a male body was exposed: __(count)

A24. Bottom nude back side male body was exposed: __(count)

A25. Full nude back side of a male body was exposed: __(count)

Empirical testing 136

DEFINITIONS AND EXAMPLES:

ROMANTIC KISSING = A romantic kissing activity which does not lead to sexual intercourse. (Ex. Kissing ones lips but no tongue is involved) FRENCH KISSING = A romantic or sexual kissing involving tongues. KISSING OF OTHER PART OF A BODY = A romantic kissing elsewhere (excluding lips, and sex organs) ORAL SEX = A sexual activity that involves use of mouth and tongue to stimulate genitalia. SEXUAL INTERCOURSE = A sexual activity that involves genital-to-genital/anal contact. MASTURBATION = a stimulation of ones sexual organs

Note: Count each scene (not shot) and each pair characters. (Ex. A man is kissing a woman, she walks away and another woman walks in and kisses the same man. You should code 2 instances in the same scene).

A26. Romantic kissing was shown in the film: __(count)

A27. French kissing was shown in the film: __(count)

A28. Kissing of other parts of a body was shown in the film: __(count)

A29. Oral sex was shown in the film: __(count)

A30. Masturbation was shown in the film: __ (count)

A31. Sexual intercourse was shown in the film: __(count)

DEFINITIONS AND EXAMPLES:

TABOO = the “forbidden” or unthinkable by American standards is shown or dealt with in the film. (Ex. One of Jane Campion short film’s shows a woman urinating by the highway.)

A32. Film deals and exposes the audience to societal Taboo’s, shocking the audience with such acts: 0. No 1. Yes

DEFINITIONS AND EXAMPLES:

FEMALE POINT OF VIEW = film deals with female issues, female roles in the world, female thinking. The plot revolves around a female.

Empirical testing 137

A33. A film has a primary point of view of a female: 0. No 1. Yes

A34. Presence of an angel(s) in the film: 0. No 1. Yes –mentioned, read about, talked about 2. Yes – visible (Ex. Seeing of an angel, angel costume, angel paintings)

SECTION B.

Unit of data collection: Major and Lead characters. MAJOR ROLE CHARACTER = is one that appears in over 50% of the film but does not appear to be lead character. If this his character is crucial to the plot line ( meaning if the character was not there, the story would change) but did not appear in over 50% of the film, code the character as a major role character. LEAD ROLE CHARACTER = the whole story revolves around this character, and his or her presence is essential to the story.

Unit of analysis: Full length feature film ______

B0. Character ID: Give each codeable character a unique number (First character that becomes codeable, the ID should be 2; Second character should have ID# 2, etc…) B1. Name or description of the character:

B2. Role: 1. Major 2. Lead

B3. Character appears in the opening scene: 0. No 1. Yes

DEFINITIONS AND EXAMPLES

CHILD: The character behaves and speaks as one who is 12 years old or younger. ADOLESCENT: The character behaves and speaks as one who is 13 to 19 years old. YOUNG ADULT: The character behaves and speaks as one who is 20 to 39 years old. MATURE ADULT: The character behaves and speaks as one who is 40 to 64 years old. ELDERLY: The character behaves and speaks as one who is 65 or older.

Empirical testing 138

B4. Social Age of the character: 1. Child 2. Adolescent 3. Young adult 4. Mature adult 5. Elderly

B4.1. Estimated age of the character (not actor): __

B5. Gender of the character: 1. Female 2. Male

B6. Race of the character: 1. Caucasian 2. African, African-American 3. Asian 4. Middle Eastern 5. Native-American (including Eskimo and those of Native South American, Native Central American, and Native Mexican extraction) 6. Bi-racial 7. Other (Identify):

DEFINITIONS AND EXAMPLES

EXTREMELY ATTRACTIVE = professional model status ATTRACTIVE = very pleasant looking, above average AVERAGE = Average in looks, but still attractive (Ex. Someone who is not model like, but clean and good looking) UNATTRACTIVE = Unpleasant to one’s eye. EXTREMELY UNATTRACTIVE = extremely non-pleasant looking, or considered ugly.

B7. The character is: 1. Extremely Attractive 2. Attractive 3. Average 4. Unattractive 5. Extremely Unattractive

B8. Occupation of the character at the beginning of the film: 1. Artist 2. Athlete 3. Attorney 4. Businessperson 5. Criminal/Convict Empirical testing 139

6. Educator 7. Entertainment Industry 8. Factory Worker 9. Farmer 10. Homemaker 11. Hotel/Hospitality Management 12. Independently Wealthy (no need to work) 13. Law Enforcement 14. Military 15. Physician/Medical 16. Prostitute 17. Religious professional – Minister, Nun, Priest 18. Restaurant Business 19. Sales 20. Secretarial/Clerical 21. Service Oriented (such as hairdresser, flight attendant. Please fill in the occupation.): 22. Spy/Government Agent 23. Student 24. Unemployed 25. Writer 26. Retired 27. None identified 28. Other (Please identify):

DEFINITIONS AND EXAMPLES

TRAVELING = the character is spending at least one night away from home

B9. The character is traveling: 0. No 1. Yes – within his/her country of origin, outside his/h town 2. Yes – outside his/her country of origin 3. Yes – within AND outside his/her country of origin

B10. The character lives abroad (in a foreign country to the character): 0. No 1. Yes – part of the film 2. Yes – Full lengths of the film

B11. The character is happy throughout 70+% of the character’s screen time: 0. No 1. Yes

B12. The character’s love situation at the beginning of the film (first 20min of the film): Empirical testing 140

1. The character is not in a romantic relationship and is not apparently looking for one. 2. The character is not in a romantic relationship and is looking for one. 3. The character is in a happy romantic relationship. 4. The character is in an unhappy romantic relationship, not looking. 5. The character is in an unhappy romantic relationship and is looking for new one. 6. Character’s love situation was not apparent.

B13. The character’s love situation in the middle of the film 1. The character is not in a romantic relationship and is not apparently looking for one. 2. The character is not in a romantic relationship and is looking for one. 3. The character is in a happy romantic relationship. 4. The character is in an unhappy romantic relationship, not looking. 5. The character is in an unhappy romantic relationship and is looking for new one. 6. Character’s love situation was not apparent.

B14. The character’s love situation at the end of the film: 1. The character is not in a romantic relationship and is not apparently looking for one. 2. The character is not in a romantic relationship and is looking for one. 3. The character is in a happy romantic relationship. 4. The character is in an unhappy romantic relationship, not looking. 5. The character is in an unhappy romantic relationship and is looking for new one. 6. Character’s love situation was not apparent.

B15. Sexual orientation of the character: 1. Heterosexual 2. Homosexual 3. Bisexual 4. Unable to determine

DEFINITIONS AND EXAMPLES

SEARCH FOR IDENTITY = the character is not comfortable in his or her role in the world and is trying to change by different actions, the character does not understand self- concept and is trying to understand his/her role by making long term and major life changes. (Ex. A female character is struggling with her role as a housewife by searching for a job outside her home)

B16. The character is looking for his / her identity by changing or trying to change his or her gender: 0. No 1. Yes

B17. The character is looking for his / her identity by changing or trying to change a job/career: Empirical testing 141

0. No 1. Yes

B18. The character is looking for his / her identity by getting married, or wanting to get married: 0. No 1. Yes

B19. The character is looking for his / her identity by changing religion: 0. No 1. Yes

B20. The character is looking for his / her identity by moving to another city, state, or country: 0. No 1. Yes

B21. The character is looking for his / her identity by OTHER factors: 0. No 1. Yes - Please explain:

B22. The character is looking for his / her identity by OTHER factors: 0. No 1. Yes - Please explain:

DEFINITIONS AND EXAMPLES

ABUSE = “a general term for the use or treatment of something (person, thing, idea, etc.) that causes some kind of harm (to the abused person or thing, to the abusers themselves, or to someone else) or is unlawful or wrongful” (http://en.wikipedia.org/wiki/Abuse). PHYSICAL ABUSE = is abuse involving contact intended to cause pain, injury, or other physical suffering or harm” (http://en.wikipedia.org/wiki/Physical_abuse). (Ex. Striking, punching, pushing, pinching, kicking, strangling, drowning, exposure to cold, exposure to heat, cutting, infecting with a disease.) PSYCHOLOGICAL ABUSE = emotional and/or verbal abuse, “refers to the humiliation or intimidation of another person, but is also used to refer to the long-term effects of emotional shock. Psychological abuse can take the form of physical intimidation, controlling through scare tactics and oppression. It is often associated with situations of power imbalance, such perhaps as the situations of abusive relationships and child abuse (http://en.wikipedia.org/wiki/Psychological_abuse). SEXUAL ABUSE = “the improper use of another person for sexual purposes, generally without their consent or under physical or psychological pressure” (http://en.wikipedia.org/wiki/Abuse).

Empirical testing 142

Note: If a character harms someone in self-defense, it is not to be coded as physical abuse.

B23. The character was physically abused by same gender character: 0. No 1. Yes

B24. The character was physically abused by opposite gender character: 0. No 1. Yes

B25. The character was psychologically abused by same gender character: 0. No 1. Yes

B26. The character was psychologically abused by opposite gender character: 0. No 1. Yes

B27. The character was sexually abused by same gender character: 0. No 1. Yes

B28. The character was sexually abused by opposite gender character: 0. No 1. Yes

B29. The character is being physically abusive to others of same gender character: 0. No 1. Yes

B30. The character is being physically abusive to others of opposite gender character: 0. No 1. Yes

B31. The character is being psychologically abusive to others of same gender character: 0. No 1. Yes

B32. The character is being psychologically abusive to others of opposite gender character: 0. No 1. Yes

Empirical testing 143

B33. The character is being sexually abusive to others of same gender character : 0. No 1. Yes

B34. The character is being sexually abusive to others of opposite gender character: 0. No 1. Yes

DEFINITIONS AND EXAMPLES:

INTROVERT = “The typical introvert is a quiet, retiring sort of person, introspective, fond of books, rather than people, he is reserved and distant except to intimate friends. He tends to plan ahead, looks before he leaps, and distrusts the impulse of the moment. He does not like excitement, takes matters of everyday life with proper seriousness, and likes a well-ordered mode of life. He keeps his feelings under close control; seldom behaves in an aggressive manner, and does not lose his temper easily. He is reliable, somewhat pessimistic, and places great value on ethical standards” (Eysenck, & Eysenck, 1968, p.6).

EXTROVERT = “The typical extrovert is sociable, likes parties, has many friends, needs to have people to talk to, an does not like reading or studying by himself. He craves excitement, takes chances, often sticks his neck out, acts on the spur of the moment and is generally an impulsive individual. He is fond of practical jokes, always has a ready answer, and generally likes change. He is care-free, easygoing, optimistic, and likes to laugh and be merry. He prefers to keep moving and doing things, tends to be aggressive and to lose his temper quickly. His feelings are not kept under tight control, and he is not always a reliable person” (Eysenck, & Eysenck, 1968, p.6).

B35. The character appears to be an: 1. Introvert 2. Extrovert 3. In the middle

B36. Character appears to be lonely. 0. No 1. Yes – sometimes in the film 2. Yes – most/all of the film

DEFINITIONS AND EXAMPLES:

OUTSIDER = Being alienated, not fitting to their family, religious group, community, town, country, society. Note: Code if the major theme of the film is the character being an outsider.

Empirical testing 144

B37. Character appears to be an outsider: 0. No 1. Yes

DEFINITIONS AND EXAMPLES:

REALISTIC CHARACTER = one that can easily be found in a real world. The character appears as if he/she could be a classmate, a co-worker, a neighbor. He/ she is an ordinary person, not eccentric, cartoonish, non-human character.

B38. Character seems to be very realistic. 0. No 1. Yes

DEFINITIONS AND EXAMPLES:

SUBSTANCE = alcohol, illegal drugs or prescription drugs if used for other than medical treatment.

B39. Character engages in substance abuse: 0. No 1. Yes – alcohol 2. Yes – drugs 3. Yes – alcohol and drugs

B40. Character tries to or does commit suicide: 0. No 1. Yes

B41. How many siblings does the character have? __

B42. Character dies in the film: 0. No 1. Yes – natural death – old age 2. Yes – natural death - health related issues 3. Yes – Accident 4. Yes – Murder 5. Yes – Suicide 6. Yes – Unable to determine

DEFINITIONS AND EXAMPLES:

A DYSFUNCTIONAL FAMILY = “is a family in which conflict, misbehaviour and even abuse on the part of individual members of the family occur continually, leading other Empirical testing 145 members to accommodate such actions…” (http://en.wikipedia.org/wiki/Family_dysfunction). (Ex. One or more members of the family have problems with other members. It appears one or more members of the family can not communicate with others. Family members are not able to agree or solve problems together. Frustration is taking over the environment.)

B43. Character deals with family issues and dysfunctions with sibling/s. 0. No siblings present 1. Has sibling/s, but no issues 2. Has sibling/s with issues 99. Unknown

B44. Character deals with family issues and dysfunctions with parent/s. 0. No parents present 1. Has parent/s, but no issues 2. Has parent/s with issues 99. Unknown

B45. Character deals with family issues and dysfunctions with child/children. 0. No children 1. Has child/children, but no issues 2. Has child/children with issues 99. Unknown

B46. Character deals with family issues and dysfunctions with grandparent/s. 0. No grandparents 1. Has grandparent/s, but no issues 2. Has grandparent/s with issues 99. Unknown

B47. Character deals with family issues and dysfunctions with grandchild/grandchildren. 0. No grandchildren 1. Has grandchild/grandchildren, but no issues 2. Has grandchild/grandchildren with issues 99. Unknown

B48. Character deals with family issues and dysfunctions with spouse. 0. No spouse 1. Has spouse, but no issues 2. Has spouse with issues 99. Unknown

B49. Character deals with family issues and dysfunctions with other family member. 0. No other family member Empirical testing 146

1. Has other family member, but no issues 2. Has other family member with issues 99. Unknown

DEFINITIONS AND EXAMPLES:

PSYCHOLOGICAL DISORDER = Mental illness = a broad generic label for a category of illnesses that may include affective or emotional instability, behavioral dysregulation, and/or cognitive dysfunction or impairment, such as major depression, generalized anxiety disorder, bipolar disorder, schizophrenia, and attention deficit hyperactivity disorder (http://en.wikipedia.org/wiki/Psychological_disorder). Such condition has been made evident in the film by: a) other characters spoke about it, b) the character has been seeing a psychologist or psychiatrist; c) the character has been already diagnosed with such condition, d) the character self declares to such condition, e) other characters reactions – non verbal.

B50. Character suffers with a psychological disorder. 0. No 1. Yes

SECTION C.

Unit of analysis: 5 minute intervals. (Meaning, imagine chopping the whole movie into 5 minute sections, and coding each section as a separate film. For example, if a movie is 93 minutes long, you will have 18 entries, with the last one 8 minutes long instead of 5.) Coding Instructions: Start the first interval at the actual beginning of the film, not when “feature presentation” or other information is running on the screen. (Ex. Given this, you might start coding with 43 seconds on the running time. The first interval will then end at 5mins.43secs.)

Interval ID: Each 5 minute interval will have a number (1st 5 min. of the film = 1, 2nd 5 min. interval = 2, etc…) ______

DEFINITIONS AND EXAMPLES:

VOICEOVER = a voice that doesn’t match the picture.

C1. The 5 min. interval included voiceover: 0. No voice over Empirical testing 147

1. Lead Character’s voice 2. Other character’s voice that appears in the movie 3. A narrator – a voice belonging to someone who is not introduced in the film 4. More than one character

DEFINITIONS AND EXAMPLES:

MUSIC = vocal or instrumental sounds that have rhythm, melody, OR harmony. A single instrumental tone also is considered music. INCLUDE music that is part of program content (Ex: A radio being played in the scene on a fictional program, a live band performing).

DIAGETIC MUSIC = music that is a part of the film, meaning characters can hear it and interact with it. (Ex. Characters listening to a record player)

NON-DIAGETIC (BACKGROUND) MUSIC- music that is not part of the film (e.g., a radio being played in the scene on a fictional program or a live band performing is part of the program content and is NOT background music; most music is background music that accentuates the emotions or actions in the program without having a specific identifiable source, that is, the viewer can’t tell who is playing the music). The fact that the source of the music is not visible in the image is not enough on its own to code the music as background music – there must be no indication that the music being heard is part of the scene or program content.

*C2. The 5 min. interval included diagetic music present? 0. No 1. Yes

*C3. The 5 min. interval included non-diagetic background music present? 0. No 1. Yes

C4. The 5 min. interval included picture in color format: 0. No 1. Yes

C5. The 5 min. interval included picture in black and white format: 0. No 1. Yes

C6. The 5 min. interval included picture in sepia format: 0. No 1. Yes

C7. The 5 min. interval included picture in animation/cartooning format: Empirical testing 148

0. No 1. Yes

DEFINITIONS AND EXAMPLES:

DOMINANT ELEMENT = the object(s)/entity(ies)/subjects that is/are the primary focus of attention on the screen; it appears to be most important, central, emphasized.

CLOSE-UP SHOT (CUs) = a dominant element fills up majority of the screen; there are no other dominant elements in the frame. (Ex. Hands playing the piano. Or piano keyboard.)

CUs FACE = head and shoulders. When deciding on a close up, use this as a guideline. If an arm pit/ arm crease is visible in the shot, it is not to be coded as a CUs. Only head and top part of shoulder counts as a CUs of face.

EXTREME CLOSE-UP SHOT (XCUs) = a dominant element fills up majority of the screen, the shot reveals extreme detail of the element – it does not fit on the screen causing it spilling off the screen.

MOTION OF TOUCHING = (Ex. A woman is stroking her lover’s hair; the camera focuses on that action - a shot of her hand doing so is shown in a CU or XCU. Therefore you should code appropriately to the image – CU of hand (C7.), CU of Motion (A32.)

C8. Motion of touching is shown: __XCU (count)

C9. Face (Including ears, nose, lips, eyes, cheeks…) shown in: __ CU (count) __XCU (count)

C10. Feet (Including toes, heal, shoes…) shown in: __ CU (count) __XCU (count)

C11. Hands are shown in: __ CU (count) __XCU (count)

C12. Other body parts (PLEASE IDENTIFY) shown in:______CU (count) __XCU (count)

C13. Furniture shown in: __ CU (count) Empirical testing 149

__XCU (count)

C14. Window shown in: __ CU (count) __XCU (count)

C15. Mirror shown in: __ CU (count) __XCU (count)

C16. Trees shown in: __ CU (count) __XCU (count)

C17. Flowers/Plants shown in: __ CU (count) __XCU (count)

C18. Food (Ex: Bread, Fruit, Meat) shown in: __ CU (count) __XCU (count)

C19. Animals shown in: __ CU (count) __XCU (count)

DEFINITIONS AND EXAMPLES:

POINT-OF-VIEW SHOT (POVs) = shows the character’s point of view. Camera takes us inside the character, we - the audience become the eyes of the character.

C20. Point-of-view shots: ___(count)

DEFINITIONS AND EXAMPLES:

LONG TRACKING SHOT = moving camera shot in which the camera dolly often runs on tracks like a miniature railroad for more than 20 seconds following a moving element or revealing information (Ex. Truffaut's famous shot of the back up car traffic due to accident. Ex. Camera travels through the woods to find a dead body).

C21. Long tracking shots: ___(count of each shot)

Empirical testing 150

DEFINITIONS AND EXAMPLES:

SPECIAL VISUAL EFFECTS = A production technique other than text or graphics that adds or modifies any part of the image that would be seen in a simple recording of the film. It is likely to DECREASE viewers' suspension of disbelief or illusion of non- mediation by reminding them of the artificial/mediated nature of the viewing experience.

BLURRED IMAGE - an image that contains vague, indistinct, or fuzzy representations of object(s)/entity(ies). An example is the subjective view of a person about to pass out.

*C22. In the 5 min interval, blurred image was identified: ___(count of each shot)

DEFINITIONS AND EXAMPLES:

DISTORTED IMAGE - an image that contains representations of object(s)/entity(ies) that are twisted out of shape, scrambled, mangled or otherwise changed. An example is a "scrambled" section of the screen when the producers are keeping a person's identity anonymous (as on Cops).

*C23. In the 5 min interval, distorted image was identified: ___(count of each shot)

DEFINITIONS AND EXAMPLES:

SLOW MOTION - a representation of an action that is made to appear slower than it actually occurred.

*C24. In the 5 min interval, slow motion was identified: ___(count of each scene=sequence)

DEFINITIONS AND EXAMPLES:

TIME-LAPSE - a representation of an action made to appear speeded up. The effect is created by having the camera take a picture at regular intervals such as every X minutes. The action may appear fluid or "jumpy". Examples include a sequence that shows a flower blooming and the commercial in which brownies quickly are stacked up on a plate.

*C25. In the 5 min interval, time-lapse was identified: ___(count each scene/sequence)

DEFINITIONS AND EXAMPLES:

JUMP CUT(s) - in this type of cut the position of object(s)/entity(ies) on the screen in one frame is suddenly shifted in the next. The result is that the object(s)/entity(ies) seem Empirical testing 151

to jump from one screen location to another for no apparent reason. A sequence of jump cuts that mimics the faster-than-normal passage of time is a time-lapse effect.

*C26. In the 5 min interval, jump cut was identified: ___(count each scene/equence)

DEFINITIONS AND EXAMPLES:

COLOR FILTER - one or more colors are added to the image, which causes at least part of the scene to appear distinctly tinted.

*C27. In the 5 min interval, use of blue color filter was identified: ___(count each scene/sequence)

C28. In the 5 min interval, use of red color filter was identified: ___(count each scene/sequence)

C29. In the 5 min interval, use of green color filter was identified: ___(count each scene/sequence)

C30. In the 5 min interval, use of other color filter was identified: ___(count each scene/sequence) DEFINITIONS AND EXAMPLES:

UNUSUAL TRANSITIONS - any transition between shots that is not a simple cut, dissolve, or fade. Examples are a wipe in which the new image slides into the screen and a transition in which the image evolves from a geometric shape (these and others are often used in Home Improvement).

*C31. In the 5 min interval, use of unusual transitions was identified: ___(count each transition)

DEFINITIONS AND EXAMPLES:

SIMULTANOUS LIVE-ACTION AND ANIMATION - both animation and live action appear on the screen at the same time. (Ex. In a film FRIDA, Frida travels to America – the trip is built from photographs and animation, but you can see the actress on the screen also.)

*C32. In the 5 min interval, use of simultaneous live-action and animation was identified: ___(count each scene/sequence)

DEFINITIONS AND EXAMPLES: Empirical testing 152

SIMULTANOUS COLOR AND BLACK&WHITE - both color and black & white appear on the screen at the same time.

*C33. In the 5 min interval simultaneous color and black and white picture was identified: ___(count)

DEFINITIONS AND EXAMPLES:

SHAKY CAMERA TECHNIQUE = HANDHELD =A series of at least 2 camera movements in which the frame of the image moves quickly and unpredictably in any direction; this technique makes it seem that the camera does not have a steady form of support such as a tripod (i.e. it seems to be handheld); the movement or movements must continue for at least 30 frames (1 second). Examples include home movies, COPS, NYPD Blue, and Homicide: Life on the Street.

*C34. Noticeable use of handheld camera (shaky camera technique): ___(count each scene/sequence)

DEFINITIONS AND EXAMPLES:

CAMERA ANGLE - the position from which the camera views the subject in an image.

SUBJECT - the object(s)/entity(ies) that is/are the primary focus of attention in an image; the part of the image that appears to be most important, central, emphasized.

LOW (looking upward)- a view from below, looking up at the subject/action at least 45 degrees up.

HIGH (looking down) - a view from above, looking down on the subject/action at least 45 degrees down.

OVERVIEW (looking straight down)- a view from directly above looking straight down.

UNDERVIEW (looking straight up) - a view from directly below looking straight up.

CANTED – camera is tilted to the right or to the left on the horizontal axis.

*C35. Low angle: __ (count each shot)

*C36. High angle: __ (count each shot)

*C37. Overview: __ (count each shot)

Empirical testing 153

*C38. Under view: __ (count each shot)

C39. Canted angle: __ (count each shot)

DEFINITIONS AND EXAMPLES:

BREAKING FOURTH WALL - the "fourth wall" is the implied wall through which the viewer is able to see objects/entities on television (i.e. the wall is the camera lens or glass viewing screen). When a character suddenly or unexpectedly looks at the camera and makes some statement (verbal or otherwise) to the viewers, this is considered "breaking the fourth wall." This must occur in either a fictional program (sitcom, drama, etc. -- in this genre the viewer's presence is never acknowledged except with this technique) or in a commercial; if it occurs in a commercial watch the entire segment and make sure that the viewer's presence is not initially acknowledged.

*C40. In the 5 min. interval, a character broke the fourth wall: __ (count shot)

DEFINITIONS AND EXAMPLES:

SUPERHEARING SENSATION = Viewer can hear sounds that s-he wouldn’t be able to hear within 6 feet. (Ex. You can hear a shaving razor sliding on the face, clicking finger nails on the piano keys, or inhaling and exhaling when talking…)

C41. In the 5 min. interval, a superhearing sensation was experienced: __ (count each shot)

DEFINITIONS AND EXAMPLES:

EXTREMELY LONG SHOT = an outdoor shot that reveals scenery/surroundings with no foreground dominant element. If a person was shown in this shot, it would occupy no more than 10% of the screen.

C42. XLS reveals the scenery/surroundings: __ (count each shot)

DEFINITIONS AND EXAMPLES:

SECONDARY BODY ELEMENT = legs, feet, or hands that are shown on the screen without the characters full body. MEDIUM SHOT (MS) = a secondary element does not fill up majority of the screen; there are other elements in the frame. (Ex. Shot of a person’s legs walking. Ex. Shot of a hand pointing at something.)

Empirical testing 154

C43. Medium shot reveals arm/hand : __ (count each shot)

C44. Medium shot reveals leg/foot: __ (count each shot)

DEFINITIONS AND EXAMPLES:

FOCUS PULL = The focus pull is a creative camera technique in which you change focus during a shot. Usually this means adjusting the focus from one subject to another.

C45. In the 5 min. interval, focus pulls were used to change focus: __ (count each shot).

______Note: * Definitions for these questions are taken from Lombard’s Structural Features Content Analysis Project, 1997.

Empirical testing 155

APPENDIX B

CODING FORM – SECTION A

*Note: If variable in bold - give count 1 CODER ID 2 FILM ID 3 DIRECTOR ID 4 YEAR ID 5 CLOSED CAPTION A page 1 A1 # of lead females A2 # of major females A3 # of medium females A4 # of lead males A5 # of major males A6 # of medium males A7.1 Father - of lead char. physically present A7.2 Father - of major char. physically present A7.3 Father - of medium char. physically present A7.4 Father - of minor/other char. physically present A7.5 Father figure that was NOT present was talked about A8.1 Mother - of lead char. physically present A8.2 Mother - of major char. physically present A8.3 Mother - of medium char. physically present A8.4 Mother - of minor/other char. physically present A8.5 Mother figure that was NOT present was talked about A9 Clash between tradition and modernism A10 Clash between cultures A11.1 Loss of a young child due to death A11.2 Loss of a young child due to 'break up' A11.3 Loss of an older child due to death A11.4 Loss of an older child due to 'break up' A11.5 Loss of a spouse due to death A11.6 Loss of a spouse due to 'break up' A11.7 Loss of lover due to death A11.8 Loss of lover due to break up A11.9 Loss of family member due to death A11.10 Loss of family member due to break up A11.11 Loss of a friend due to death A11.12 Loss of a friend due to break up A11.13 Loss of faith - religion Empirical testing 156

1 CODER ID 2 FILM ID 3 DIRECTOR ID 4 YEAR ID 5 CLOSED CAPTION A page 2 A11.14 Loss of faith in self A11.15 Loss of faith in others A11.16 Loss of health A11.17 Loss of material things A11.18 Loss of freedom A11.19 Loss of a job A11.20 Loss of money A11.21 Loss of physical beauty A11.22 Other loss (Identify): A12 Partial nude frontal female: each shot A13 Top nude frontal female: each shot A14 Bottom nude frontal female: each shot A15 Full nude frontal female: each shot A16 Partial nude back female: each shot A17 Top nude back female: each shot A18 Bottom nude back female: each shot A19 Full nude back female: each shot A20 Top nude frontal male: each shot A21 Bottom nude frontal male: each shot A22 Full nude frontal male: each shot A23 Top nude back male: each shot A24 Bottom nude back male: each shot A25 Full nude back male: each shot A26 Romantic kissing A27 French kissing A28 Kissing other body parts A29 Oral sex A30 Masturbation A31 Sexual Intercourse A32 TABOO A33 Primary POV of female A34 Presence of angels

Empirical testing 157

CODING FORM – SECTION B

1 CODER ID 2 FILM ID 3 DIRECTOR ID 4 YEAR ID 5 CLOSED CAPTION B page 1 6 CHARACTER'S ID 1 2 3 4 5 6 7 (etc…) B1 Character's name B2 Role B3 Opening scene B4 Age B4.1 Estimated age B5 Gender B6 Race B7 Appearance B8 Occupation B9 Travel B10 Abroad B11 Happy B12 Love at beginning B13 Love in the middle B14 Love in the end B15 Sexual orientation B16 Identity change: gender B17 Identity change: career B18 Identity change: marriage B19 Identity change: religion B20 Identity change: move B21 Identity - other: (identify) B22 Identity - other: (identify) B23 Phys abused by same B24 Phys abused by opposite B25 Psych abused by same B26 Psych abused by opposite B27 Sex abuse by same B28 Sex abuse by opposite B29 Phys abusive to same B30 Phys abusive to other B31 Psych abusive to same B32 Psych abusive to other B33 Sex abusive to same B34 Sex abusive to other B35 Introvert vs. Extrovert B36 Appears lonely B37 Appears to be an outsider B38 Is very realistic Empirical testing 158

1 CODER ID 2 FILM ID 3 DIRECTOR ID 4 YEAR ID 5 CLOSED CAPTION B page 2 6 CHARACTER'S ID 1 2 3 4 5 6 7 (etc…) B39 Substance abuse B40 Tries to hurt self B41 How many siblings? B42 Char. dies in the film? B43 Issue w/ sibling B44 Issue w/ parent B45 Issue w/ child B46 Issue w/ grandparent B47 Issue w/grandchildren B48 Issue w/ spouse B49 Issue w/other fam.member B50 Psych Disorder

Empirical testing 159

CODING FORM – SECTION C

*Note: If variable in bold - give count 1 CODER ID 2 FILM ID 3 DIRECTOR ID 4 YEAR ID C page 1 5 CLOSED CAPTION 0-5min 5-10min 10-15min 15-20min 6 Interval ID 1 2 3 4 (etc…) C1 Voiceover C2 Diagetic music C3 Non-diagetic music C4 Color picture C5 Black&White picture C6 Sepia picture C7 Animation/cartooning C8 Motion of touching - XCU C9 Face: CU C9.1 Face: XCU C10 Feet: CU C10.1 Feet: XCU C11 Hands: CU C11.1 Hands: XCU C12 Other body parts (identify):CU C12.1 Other body parts (identify):XCU C13 Furniture: CU C13.1 Furniture:XCU C14 Window: CU C14.1 Window: XCU C15 Mirror:CU C15.1 Mirror:XCU C16 Trees: CU C16.1 Trees: XCU C17 Flowers/Plants: CU C17.1 Flowers/Plants: XCU C18 Food: CU C18.1 Food XCU C19 Animal: CU C19.1 Animal: XCU C20 POV shots C21 Long tracking shots C22 Blurred image C23 Distorted image C24 Slow motion C25 Time lapse (fast motion)

Empirical testing 160

1 CODER ID 2 FILM ID 3 DIRECTOR ID 4 YEAR ID C page 2 5 CLOSED CAPTION 0-5min 5-10min 10-15min 15-20min 6 Interval ID 1 2 3 4 (etc…) C26 Jump cut C27 Blue color filter C28 Red color filter C29 Green color filter C30 Other color filter C31 Unusual transition C32 Simult. live-action&animation C33 Simult. color and B&W C34 Handheld camera C35 Low angle C36 High angle C37 Overview C38 Underview C39 Canted angle C40 Breaking the fourth wall C41 Superhearing C42 XLS/wide shot of scenery C43 Medium shot arm/hand C44 Medium shot leg/foot C45 Focus pulls

Empirical testing 161

APPENDIX C

THEMES AND MOTIFS OF NON-CAMPION AUTEUR - DIRECTORS

Gregg Araki 1 Intensifies lighting and color of the city (LA) 2 Love, sex and lust 3 Queer characters 4 Tries to show being gay doesn’t make a difference among people 5 Characters are obsessed with video cameras 6 Adolescents and homosexuality Luc Besson 1 Lack of dialog – comic strip influences? 2 Stories of young people who have difficulties adapting to a society 3 Society lets the family down - the young people become emotionally deprived 4 Youth in crisis 5 Dysfunctional families 6 Parents are absent 7 If mother figure appears – as evil force 8 Protagonists are lonely and suffering 9 Self-erasure of the protagonists 10 Literal portrayals of underworlds – metro, sea, lower regions of cities 11 Characters are “larger than life” 12 Bricolage of genres – more genres in one Tim Burton 1 Love for horror and fantasy 2 Plain weirdness of main characters 3 Technical experimentation with possibilities of animation and visual effects 4 Main characters are outsiders villains with extremely complex personalities 5 Main characters have very pale skin 6 Themes of exclusion and inclusion 7 Johnny Depp Jackie Chan 1 Kung-Fu stunts 2 Master-pupil theme (comical) 3 Silent cinema feel to some of his scenes (i.e., chasing scenes, bar fights) 4 Female roles are always over the top girlish Empirical testing 162

David Cronenberg 1 The physical and the psychological are usually intertwined 2 Main character is often male scientist – with a bizarre name 3 Themes: science, disease, aging, invasion, death in literary metaphors 4 Bodily and psychic mutations 5 Hostility towards women – men fear and are jealous of females 6 Womb envy David Fincher 1 Critiques of today’s society 2 A man alone in the universe 3 Visual translation of feelings of isolation of the characters 4 Wide angles, shallow focus = isolation 5 Muted light and color 6 Psychological and emotional state of American males of 20th century 7 Decay of American family 8 Females are barely present Jim Jarmusch 1 Alienation of characters 2 Use of jump cuts 3 Characters have no real direction in life 4 Not interested in What? and Why? but How did one character get where he is now... 5 Main characters are outsiders Abbas Kiarostami 1 Challenges the audience 2 Poetic and philosophical style 3 Self-referentiality 4 Social realist traits/themes 5 Bleakness and loss - main themes 6 Often hopeful and enthusiasm accompanies the results 7 Likes to involve the audience as much as possible Ang Lee 1 Modernity vs. tradition 2 Deep secrets and internal torment come to surface (http://en.wikipedia.org/wiki/Ang_Lee) 3 Conflict between freedom and societal traditions

Empirical testing 163

Michael Mann 1 Interested in real events 2 Fading centrality of a male 3 Visual: soft blues, and sterile and harsh whites John Sayles 1 Social concerns 2 Political awareness 3 Complex characters 4 No central protagonist, usually group of characters 5 Stretches the boundaries of genres 6 Witty dialog 7 Realistic characters Steven Soderbergh 1 Character based films 2 Ironic humor 3 Sense of isolation and loneliness 4 Characters are harshly detached, emotionally frozen 5 Characters are unable to fit with the world around them – lonely 6 Ultimately, his films are character studies of populated with strangers in a strange land Oliver Stone 1 Lonely and isolated male protagonist in quasi-spiritual journey 2 Character goes through innocence, suffering and experiences liberation or destruction 3 Testosterone-driven and often didactic dialog 4 Marginalization of women 5 The lure of homo-social is evident Peter Weir 1 Very national – Australian – rather than international 2 Avoidance of sexuality 3 No vivid sex scenes 4 However he shows erotic desire Allison Anders 1 Multiple protagonists 2 Teen angst (anxiety, fear, worry) 3 Gangsters 4 Biopic 5 Emotional intimacy and heterosexual love atavistic Empirical testing 164

APPENDIX D

INTER-CODER RELIABILITIES SECTION A Lin's Multiple Kappa Concord. Coder Coder Coder Variable Coders: pair: pair: pair: code Variable name 1 - 3 1, 2 1, 3 2, 3 A1 # of lead females 1 1 1 A2 # of major females 1 1 1 A3 # of medium females 1 0.92 0.92 A4 # of lead males 1 1 1 A5 # of major males 1 0.67 0.67 A6 # of medium males 0.96 0.87 0.86

A father present Va7* (A7.1+A7.2+A7.3+A7.4+A7.5) 0.5 1 0.5 A7.1** Father - of lead char. 1 1 1 A7.2** Father - of major char. NC NC NC A7.3** Father - of medium char. 0 1 0 A7.4** Father - of minor/other char. NC NC NC A7.5** Father figure that was NOT present 1 1 1 A mother present Va8* (A8.1+A8.2+A8.3+A8.4+A8.5) 1 1 1 A8.1 Mother - of lead char. 1 1 1 A8.2** Mother - of major char. 0 0 1 A8.3** Mother - of medium char. NC NC NC A8.4** Mother - of minor/other char. NC NC NC A8.5** Mother figure that was NOT present 1 1 1

A9 Clash between tradition & modernism NC NC NC A10 Clash between cultures 1 1 1

A11.1 Loss of a young child due to death NC A11.2 Loss of a young child due to 'break up' NC A11.3 Loss of an older child due to death NC A11.4 Loss of an older child due to 'break up' NC A11.5 Loss of a spouse due to death 1 A11.6 Loss of a spouse due to 'break up' NC A11.7 Loss of lover due to death 1 A11.8 Loss of lover due to break up NC Empirical testing 165

Lin's Multiple Kappa Concord. Coder Coder Coder Variable Coders: pair: pair: pair: code Variable name 1 - 3 1, 2 1, 3 2, 3 A11.9 Loss of family member due to death 1 A11.10 Loss of family member due to break up NC A11.11 Loss of a friend due to death 0.46 A11.12 Loss of a friend due to break up NC A11.13 Loss of faith - religion NC A11.14** Loss of faith in self NC A11.15** Loss of faith in others 0.14 A11.16 Loss of health 0.7 A11.17 Loss of material things 0.68 A11.18 Loss of freedom 0.7 A11.19** Loss of a job -0.09 A11.20** Loss of money NC A11.21** Loss of physical beauty NC A11.22** Other loss (Identify): -0.09

A12 Partial nude frontal female: each shot 0.97 0.93 0.85 A13 Top nude frontal female: each shot 0.75 0.96 0.61 A14 Bottom nude frontal female: each shot 1 1 1 A15 Full nude frontal female: each shot 0.98 0.95 0.97 A16 Partial nude back female: each shot NC NC NC A17 Top nude back female: each shot NC NC NC A18 Bottom nude back female: each shot NC NC NC A19 Full nude back female: each shot 1 1 1 A20 Top nude frontal male: each shot 0.98 0.98 0.93 A21 Bottom nude frontal male: each shot NC NC NC A22 Full nude frontal male: each shot 1 1 1 A23 Top nude back male: each shot 1 0.75 0.75 A24 Bottom nude back male: each shot NC NC NC A25 Full nude back male: each shot NC NC NC

A26** Romantic kissing 0 NC 0 A27** French kissing 0 0 NC A28** Kissing other body parts 0 1 0 Va28.1* Kissing (A26+A27+A28) 0.75 0.75 1 Empirical testing 166

Lin's Multiple Kappa Concord. Coder Coder Coder Variable Coders: pair: pair: pair: code Variable name 1 - 3 1, 2 1, 3 2, 3 A29 Oral sex 1 1 1 A30 Masturbation NC NC NC A31 Sexual Intercourse 1 1 1 Va31* All Sex (VA28.1+29+A30+A31) 0.75 0.75 1 A32 Taboo 1 1 1 A33 Primary POV of female 1 1 1 A34 Presence of angels 1 1 1

NC= Not Calculable due to lack of variance

*New / computed variables

**These variables have been dropped due to low reliability and/or lack of occurrence

Empirical testing 167

INTER-CODER RELIABILITIES

SECTION B

Multiple Kappa Spearman's Rho Lin's Concordance Coder Coder Coder Coder Coder Coder Variable Coders: pair: pair: pair: pair: pair: pair: code Variable name 1 - 3 1, 2 1, 3 2, 3 1, 2 1, 3 2, 3 B1 Character's name B2 Role 0.7 B3 Opening scene 1 1 1 B4 Age 1 1 1 B4.1 Estimated age 0.99 0.98 1 B5 Gender 1 B6 Race NC B7 Appearance 0.79 0.53 0.81 B8 Occupation 0.58

B9 Travel 1 B10 Abroad 1

B11 Happy NC NC NC

B12 Love at beginning 0.42 B13 Love in the middle 0.78 B14 Love in the end 0.4

B15 Sexual orientation 0.7

B16 Identity change: gender NC NC NC B17 Identity change: career NC NC NC Identity change: B18 marriage 1 NC NC B19 Identity change: religion NC NC NC B20 Identity change: move 0.5 NC NC Identity - other: B21 (identify) NC NC NC Identity - other: B22 (identify) NC NC NC

Empirical testing 168

Multiple Kappa Spearman's Rho Lin's Concordance Coder Coder Coder Coder Coder Coder Variable Coders: pair: pair: pair: pair: pair: pair: code Variable name 1 - 3 1, 2 1, 3 2, 3 1, 2 1, 3 2, 3 B23 Phys abused by same 1 0.55 0.55 B24 Phys abused by opposite 0.7 NC NC B25 Psych abused by same NC NC NC Psych abused by B26 opposite 1 NC NC B27 Sex abuse by same NC NC NC B28 Sex abuse by opposite 1 NC NC B29 Phys abusive to same 0.55 0.55 1 B30 Phys abusive to other 1 NC NC B31 Psych abusive to same NC NC NC B32 Psych abusive to other 1 1 1 B33 Sex abusive to same NC NC NC B34 Sex abusive to other 1 1 1

B35 Introvert vs. Extrovert 0.78 B36 Appears lonely 0.75 0.75 0.5 Appears to be an B37 outsider 1 1 1 B38** Is very realistic 0 0 0.62 B39 Substance abuse 0.73 B40 Tries to hurt self 1 NC NC B41 How many siblings? 1 1 1 B42 Char. dies in the film? 1

B43 Issue w/ sibling 1 B44 Issue w/ parent 1 B45 Issue w/ child 1 B46 Issue w/ grandparent NC B47 Issue w/grandchildren NC B48 Issue w/ spouse 1 Issue w/other B49 fam.member NC

B50 Psych Disorder 1 NC= Not Calculable due to lack of variance **These variables have been dropped due to low reliability and/or lack of occurrence

Empirical testing 169

INTER-CODER RELIABILITIES

SECTION C

Spearman's Rho Variable Coder pair: Coder Coder code Variable name 1, 2 pair: 1, 3 pair: 2, 3 C1 Voiceover 0.5 1 0.5

C2 Diegetic music 0.79 0.78 0.57 C3 Non-diegetic music 0.64 1 0.64

C4 Color picture NC NC NC C5 Black & White picture NC NC NC C6 Sepia picture NC NC NC C7 Animation/cartooning NC NC NC C8 Motion of touching - XCU 0.72 NC NC C9 Face: CU 0.98 0.85 0.92 C9.1 Face: XCU 0.92 NC NC C10 Feet: CU NC NC NC C10.1 Feet: XCU NC NC NC C11 Hands: CU 1 1 1 C11.1 Hands: XCU NC NC NC C12 Other body parts (identify):CU NC NC NC C12.1 Other body parts (identify):XCU NC NC NC C13 Furniture: CU NC NC NC C13.1 Furniture:XCU 1 NC NC C14 Window: CU 0.65 0.64 0.84 C14.1 Window: XCU NC NC NC C15 Mirror:CU 0.44 0.64 0.64 C15.1 Mirror:XCU NC NC NC C16 Trees: CU NC NC NC C16.1 Trees: XCU NC NC NC C17 Flowers/Plants: CU NC NC NC C17.1 Flowers/Plants: XCU NC NC NC C18 Food: CU NC NC NC C18.1 Food XCU NC NC NC C19 Animal: CU NC NC NC C19.1 Animal: XCU NC NC NC

C20** POV shots -0.03 1 -0.03 Empirical testing 170

Spearman's Rho Variable Coder pair: Coder Coder code Variable name 1, 2 pair: 1, 3 pair: 2, 3 C21 Long tracking shots 0.7 0.9 0.69

C22** Blurred image NC NC NC C23** Distorted image 0 0 0

C24 Slow motion 0.85 NC NC C25 Time lapse (fast motion) NC NC NC C26 Jump cut NC NC NC

C27 Blue color filter 1 NC NC C28 Red color filter NC NC NC C29 Green color filter NC NC NC C30 Other color filter NC NC NC

C31** Unusual transition NC NC NC

C32** Simult. live-action & animation NC NC NC C33** Simult. color and B&W NC NC NC

C34 Handheld camera 1 NC NC

C35 Low angle 0.3 0.88 0.26 C36 High angle 0.62 0.89 0.5 C37 Overview 1 1 1 C38 Underview 0.84 0.78 0.84 C39 Canted angle 1 NC NC VC52* Shots of looking up (C35+C38) 0.38 0.87 0.42 VC53* Shots of looking down (C36+C37) 0.88 0.62 0.5

C40** Breaking the fourth wall NC NC NC

C41 Superhearing 0.47 0.64 0.63

C42 XLS/wide shot of scenery 0.81 0.36 0.57

C43 Medium shot arm/hand 1 1 0.56 C44 Medium shot leg/foot 0.84 0.64 0.55 VC50* Shots of legs/feet (C10+C10.1+C44) 0.84 0.64 0.74 Empirical testing 171

Spearman's Rho Variable Coder pair: Coder Coder code Variable name 1, 2 pair: 1, 3 pair: 2, 3 Shots of arms/hands VC51* (C11+C11.1+C43) 0.62 1 0.62

C45 Focus pulls 1 NC NC NC= Not Calculable due to lack of variance

*New / computed variables

**These variables have been dropped due to low reliability and/or lack of occurrence

Empirical testing 172

APPENDIX E

AUTEUR TRADEMARKS

JANE CAMPION – AUTEUR TRADEMARKS (FROM LITERATURE)

SIGNIFICANT NON SIGNIFICANT # of lead females Masturbation # of lead males Love at beginning Loss of freedom Love in the end Loss of a spouse due to 'break up' Identity change: gender Loss of lover due to break up Identity change: career Loss of family member due to break up Identity change: religion Kissing (A26+A27+A28) Identity change: move Oral sex Phys abused by same All Sex (VA28.1+29+A30+A31) Psych abused by same Taboo Psych abused by opposite Primary POV of female Sex abuse by same Presence of angels Sex abuse by opposite Age of the character Phys abusive to same Gender Phys abusive to other Travel Psych abusive to same Abroad Psych abusive to other Happy Sex abusive to same Love in the middle Black &White picture Identity change: marriage High angle Phys abused by opposite Overview Sex abusive to other Underview Appears lonely Canted angle Issue w/ sibling Superhearing Issue w/ parent XLS/wide shot of scenery Issue w/ spouse Issue w/ child Issue w/ grandparent Issue w/grandchildren Issue w/other fam.member Psych Disorder Sepia picture Animation/cartooning Motion of touching - XCU

Empirical testing 173

SIGNIFICANT NON-SIGNIFICANT Face: CU Face: XCU Feet: CU Feet: XCU Hands: CU Hands: XCU Low angle Shots of looking up (C35+C38) Shots of looking down (C36+C37)

Empirical testing 174

JANE CAMPION – AUTEUR TRADEMARKS (FROM PILOT WORK)

SIGNIFICANT NON-SIGNIFICANT Partial nude frontal female Top nude back female Top nude frontal female Bottom nude back female Full nude frontal female Top nude frontal male Partial nude back female Top nude back male Full nude back female Bottom nude back male Bottom nude frontal male Full nude frontal male Full nude back male (Character appears) in Opening scene Medium shot arm/hand Medium shot leg/foot Shots of legs/feet (C10+C10.1+C44) Shots of arms/hands (C11+C11.1+C43)

Empirical testing 175

PRODUCTION TECHNIQUES (FROM LITERATURE)

SIGNIF NON-SIGNIFICANT Non-diegetic music Voiceover Furniture: XCU Diegetic music Window: XCU Color picture Mirror: CU Furniture: CU Mirror: XCU Window: CU Slow motion Trees: CU Blue color filter Trees: XCU Red color filter Flowers/Plants: CU Handheld camera Flowers/Plants: XCU Focus pulls Food: CU Food XCU Animal: CU Animal: XCU Long tracking shots Time lapse (fast motion) Jump cut Green color filter Other color filter

Empirical testing 176

NON-CAMPION AUTEUR TRADEMARKS (FROM LITERATURE)

SIGNIFICANT NON-SIGNIFICANT Loss of health # of major females # of medium females # of major males # of medium males Clash between tradition & modernism Clash between cultures A father present Father - of lead char. Father - of major char. Father - of medium char. Father - of minor/other char. Father figure that was NOT present A mother present Mother - of lead char. Mother - of major char. Mother - of medium char. Mother - of minor/other char. Mother figure that was NOT present Loss of a young child due to death Loss of a young child due to 'break up' Loss of an older child due to death Loss of an older child due to 'break up' Loss of a spouse due to death Loss of a spouse due to 'break up' Loss of lover due to death Loss of lover due to break up Loss of family member due to death Loss of family member due to break up Loss of a friend due to death Loss of a friend due to break up Loss of faith - religion Loss of faith in self

Empirical testing 177

SIGNIFICANT NON-SIGNIFICANT Loss of faith in others Loss of material things Loss of a job Loss of money Loss of physical beauty Other loss (Identify): Introvert vs. Extrovert Appears to be an outsider Is very realistic Substance abuse Tries to hurt self Char. dies in the film?