Warren Buckland. Handbook of Film Studies. Editor: James Donald & Michael Renov. Sage Publications. 2008.
Formalism, both inside and outside Film Studies, is a rich and variegated paradigm. In Film Studies, the formalist tradition covers everything from the Russian formalists (see Eagle, 1981) to Rudolf Arnheim’s pioneering Film as Art (1957), Christian Metz’s film semiotics (1974a; 1974b; Buckland, 1999); Noel Burch’s study of film’s formal principles (1981); Edward Branigan’s description of point of view in narrative cinema (1984); Raymond Bellour’s shot-by-shot analyses of film segments (2000); Barry Salt’s statistical style analysis (1974; 1992; 2004); David Bordwell’s film poetics (1981; 1988a; 1989: ch.11; 1993; 1998; 2000); Kristin Thompson’s neoformalism (1981; 1988); Noel Carroll’s functional analysis of film form (1998); plus the productive tradition of mise en scene and auteur criticism (Cahiers du cinema [Hillier, 1985; Hillier, 1986], Movie magazine [Cameron, 1972; Perkins, 1972], Andrew Sarris [1968]). Some researchers use formalist theory in a purely descriptive manner, constructing meticulously precise taxonomies; others, influenced by poetics, examine how a film is put together; while formalist critics use their formal descriptions to evaluate film style.
Other variations of formal criticism also exist. Art theorist Nick Zangwill (2001: ch. 4) identifies three conceptions of form: extreme formalism, moderate formalism, and antiformalism. For extreme formalists such as the early-twentieth-century art critics Clive Bell (1928) and Roger Fry (1924), form is coterminous with aesthetics. Art appreciation, they argue, only requires a sustained attention to an artwork’s internal properties—its arrangement of lines, shapes, and colours—to the exclusion of external properties such as its making or its representational function. Antiformalists such as art theorists George Dickie (1974) and Arthur Danto (1964) argue from the opposite position: that aesthetic values are not formal, are not internal or intrinsic to a work of art, but are contextual and institutional. Zangwill attempts to defend a moderate aesthetic formalism, an intuitive notion in which some aesthetic properties are internal to a work of art, while others are external. Zangwill argues that form is not autonomous or independent, as extreme formalists argue, but always serves or is dependent on external properties, including an artwork’s representation of content, and its process of making, or its poetics. We shall examine how and why moderate formalism in particular has dominated Film Studies, and consider two fundamental evaluative concepts which have had a strong and decisive influence on film criticism: the moderate formalist concept of organic unity and the extreme formalist concept of significant form.
In the final section, I extend the formalist tradition—especially poetics—by supplementing it with the decision-making processes of film production, as codified in filmmaking manuals. On reading these manuals, I discovered to my surprise that they advocate the same principles found in film criticism: they encourage filmmakers to achieve organic unity and significant form in their films. My extension of the formalist tradition signifies a possible rapprochement between film criticism and filmmaking, and therefore the continuing value of formalism in Film Studies.
Inner Logic
The project of modernity formulated in the 18th century by the philosophers of the Enlightenment consisted in their efforts to develop objective science, universal morality and law, and autonomous art according to their inner logic. (Habermas, 1983: 9)
Formalism emerged from the modernist project of the Enlightenment. The unified, cosmological world-view (informed by metaphysics, religion and myth) that held sway over everyday life and beliefs in pre-modern Western society was challenged during the Renaissance, and was eventually transformed by the time of the Enlightenment and the French Revolution. From the transformed cosmological world-view emerged three differentiated secular world-views (or value spheres)—those of science (study of nature), morality (study of social norms), and aesthetics (study of art, the self, and human freedom). Moreover, each value sphere had its own type of validity—namely, truth, justice, and judgements of taste respectively. This tripartite distinction between value spheres was concretized in the seminal works of Immanuel Kant—his Critique of Pure Reason ([1787] 2003) (the study of scientific knowledge), his Critique of Practical Reason ([1788] 1997) (the study of morality), and his Critique of the Power of Judgement ([1790] 2001) (the study of aesthetics). These three value spheres bring to the fore the autonomy of specialized knowledge; together, they constitute the dominant characteristic of the Enlightenment project and cultural modernity.
For example, aesthetic experience became progressively liberated (or differentiated) from everyday perception and, by the nineteenth century, it had been transformed, in the l’art pour l’art movement, into an autonomous realm of self-validating experience. In the twentieth century, this ‘aesthetic modernity’ (as Jurgen Habermas calls it) has been manifest in various avant-garde movements which, through a series of selfreflexive practices, create a form of radical experience that criticizes the ideology of everyday life—or criticizes the constraints imposed upon everyday life by the other two value spheres, science and morality.
Parallel to this aesthetic modernity is the formalist study of art, an examination of what is unique and irreducible to each medium. For the formalists, although an artwork may be representational, what makes it art is its formal properties, not its representational features (which simply motivate the form) (Carroll, 1999: 110-11). ‘Form’ designates the inner logic of each art, the essential set of features that delimit and shape amorphous matter. Out of form emerges a range of options regarding how to shape matter and create a particular work of art. In pictorial art, ‘form’ refers to the essential features that render something visible. For example, all pictorial art arranges its parts in relation to a surface. Numerous options exist: either the parts can be arranged planimetrically (across the surface) or recessionally (receding into depth), or somewhere in between. The systematic choices made by an individual artist, or school, or period, in the making of artworks, names the preferred style of that artist, school, or period.
In other words, form is the essential set of features internal to each medium. From these features emerge a set of options concerning how to shape, arrange, or compose an individual artwork. Making choices from the formal options is an expressive activity that creates style. (Form is therefore ‘impersonal’ and exists prior to style and expression.) The analysis of style involves examining the range of formal options available to artists, and the choices they make in constructing an individual artwork. In pictorial art, choosing to organize a visual surface planimetrically, for example, creates an image in the ‘classical’ style, whereas a recessional choice creates a ‘baroque’ style of image.
Organic Unity and Significant Form
The term ‘organic unity’ is central to philosophy (especially to Plato, Aristotle, and G.E. Moore) and aesthetics—including film aesthetics (mise en scene and auteur criticism). Organic unity refers to the way the parts of a whole relate to one another, how they relate to the whole, and how the whole relates to external properties such as content. An organic unity is a whole that is more than the sum of its parts, for the whole possesses an intrinsic value not contained in any of its parts. This new value emerges from or is realized in the parts, but is not contained in any of them. It is not a property that exists by itself, but depends on other properties. It is an emergent value that lies only in the whole, although it is reflected back in the parts.
The concept of organic unity is an evaluative concept aimed at praising one style of art: Classicism. In classical art—meaning Classical Antiquity, the Italian Renaissance, and the various Neoclassical styles that imitate them—each work reaches a state of harmony and equilibrium, for an organic unity is the best possible (the most efficient and optimal) way to organize the parts. All the parts of an organic unity are necessary and sufficient to its status as a unity. Each part has an intrinsic value, and additional intrinsic values arise solely from the combination of all the parts. Any addition or diminution will destroy the intrinsic value of the whole, because the parts are so closely connected and interdependent that it is impossible to think of the whole otherwise. In an organic unity, the parts have reached their highest degree or best possible level of integration. Zangwill comments that ‘[t]he appellation “organic” is fitting in that the unity in question is like that of a living system where parts have a purpose with respect to each other and with respect to the whole’ (2001: 62, n.16).
The extreme formalist Clive Bell came up with the term ‘significant form’ to designate a concept similar to ‘organic unity’:
What quality is shared by all objects that provoke our aesthetic emotions? What quality is common to Sta. Sophia and the windows at Chartres, Mexican sculpture, a Persian bowl, Chinese carpets, Giotto’s frescoes at Padua, and the masterpieces of Poussin, Piero della Francesca, and Cezanne? Only one answer seems possible—significant form. In each, lines and colours combined in a particular way, certain forms and relations of forms, stir our aesthetic emotions. These relations and combinations of lines and colours, these aesthetically moving forms, I shall call ‘Significant Form’; and ‘Significant Form’ is the one quality common to all works of visual art. (1928: 8)
The main difference between organic unity and significant form is that significant form only refers to the relation between the different formal components of an artwork, for it excludes external properties such as the making of an artwork and its representation of content.
The concepts of organic unity and significant form are nonetheless related. To claim that an artwork has significant form similarly argues that the whole is more than the sum of its parts. For Stefan Sharff, who has applied the concept of significant form to film, in some films the parts add up to create a new entity that does not exist in each part: ‘Significant form is the opposite of pedestrian rendition … Images fit together so magnificently that they ascend to a higher level of visual meaning’ (1982: 7). Above I noted that each part of an organic unity has an intrinsic value, and additional intrinsic values arise solely from the combination of all the parts. The value of the individual parts may be very small. What defines an organic unity is the emergent value of the whole. In Sharff’s terms, each individual part of a film may be pedestrian. A film with significant form is one that combines these pedestrian parts into an elevated form, in which the parts are strongly linked in a particular order. In contrast, a pedestrian film is one that is not more than the sum of its parts. The parts of a pedestrian film, when joined together, do not attain ‘a higher level of visual meaning’, but remain an aggregate collection of isolated pedestrian parts. Only a well-made film will manifest ‘significant form’ (and organic unity).
But how do we recognize significant form and organic unity? All filmmakers use the same standard parts—the same formal features, or ‘film language’. The key to their success is the way they combine these features into an expressive style. This is the thinking behind the mise en scene and auteur criticism of Cahiers du cinema, Movie magazine, and Andrew Sarris. In the next section I examine the film criticism of V.F. Perkins, a prominent member of the Movie group, and Noel Carroll’s practice of film analyses.
Stylistic Film Criticism
A fundamental characteristic of moderate formalism, in its argument against the autonomy of form, is that form is both appropriate to and dependent on an external property—namely, content, or what is represented. From a moderate formalist perspective, all aesthetic properties stand in a dependence relationship to non-aesthetic properties—including representational properties.
In Chapter Six of Film as Film (1972) Perkins outlines his analytical criteria for evaluating films from a moderate formalist perspective. Although he does not use the terms, he seeks ‘significant form’ and especially ‘organic unity’. The terms he uses are ‘balance’, ‘coherence’, ‘relatedness’ (the ‘interaction’ and ‘integration’ of filmic elements), ‘productive tension’, and ‘intensity of cohesion’. ‘Coherence’, he writes ‘is the means by which the film-maker creates significance’ and the means by which the spectator ‘recognize[s] meaning at all levels’ (1972:116). Moreover, coherence is not given in advance, is not part of the film’s preexisting content, but is an emergent value formed through the activity of constructing a film, through the creation of relationships that organize the various elements of film into a coherent synthesis:
Useful criteria [for film analysis and evaluation] take account of relatedness by directing us not to single aspects but to the value of their interaction and the extent of their integration. The formal disciplines of balance and coherence embrace the effort to maintain the various elements in productive tension. (1972: 120)
What does Perkins mean by the effort to maintain a productive tension? He means two things—the identification of film’s ‘essence’, and the organization of an individual film’s mise en scene.
Classical film theorists in the realist camp, such as Andre Bazin, argued that film is essentially a realist medium due to its photographic capacity, while classical film theorists in the expressionist camp, such as Rudolf Arnheim, argued that film is essentially a new form of expression that defamiliarizes experience by creating a distinct view of the world. Perkins argues that film has no single essence, since it is a hybrid of two conflicting tendencies—realism, which creates credibility, and expressionism, which creates significance. One of the primary skills of being a filmmaker is to find a balance between these two tendencies:
The movie is committed to finding a balance between equally insistent pulls, one towards credibility [realism] and the other towards shape and significance [expressionism]. And it is threatened by collapse on both sides. It may shatter illusion in straining after expression. It may subside into meaningless reproduction presenting a world which is credible but without significance. (1972: 120)
The skilled filmmaker reconciles film’s conflicting tendencies by maintaining a credible world and, at the same time, employing film’s expressive capacities to achieve heightened coherence—or organic unity. An unbalanced, incoherent film is one that either pulls too much towards realism and credibility and does not exploit film’s expressive capacity, or one that overuses its expressive capacity at the expense of realism and credibility: ‘The great film approaches an intensity of cohesion such that its elements do not operate solely to maintain or further the reality of the fictional world, nor solely to decorative, affective or rhetorical effect’ (1972: 131).
A coherent film creates an intensity of cohesion between realism and expressionism. More particularly, at the level of detail, each coherent film creates an intensity of cohesion between the particular elements of mise en scene it combines—action, image, decor, gesture, speech, camera movement and placement, cutting, and lighting. Perkins’ examples throughout Film as Film each demonstrate a tight correspondence, plus novel and inventive relationships, between these elements: for example, speech-gesture-decor in The Courtship of Eddie’s Father (Vincente Minnelli, US, 1963), or action-image-cutting in Carmen Jones (Otto Preminger, US, 1954) (1972: 76, 79-82).
At the other extreme, an imbalance is created in sentimental and pretentious films. A sentimental film, for Perkins, creates an imbalance between ‘pathos asserted (in music, say, or image or gesture) and pathos achieved, in the action’ (1972: 132), while a pretentious film creates imbalance by giving an elevated, unjustified emphasis and significance to particular elements of mise en scene.
Perkins’ analytical criteria are geared towards one type of filmmaking—classical Hollywood films—and just one dimension of those films—mise en scene—and evaluates films positively if they conform to a classical style of mise en scene, negatively if the mise en scene becomes too ornate.
Adrian Martin’s threefold distinction between classical, expressionist, and mannerist mise en scene is valuable in contextualizing Perkins’ preferred style of filmmaking (Martin, 1992). In classical mise en scene the film style is unobtrusive, for it is motivated by the film’s themes and dramatic developments. These films maintain a balance between showing and narrating, since style is linked to function, rather than being autonomous; the mise en scene functions as unobtrusive symbolism that confers upon the film heightened significance. In expressionist mise en scene, a broad or loose fit exists between style and theme. Finally, in mannerist mise en scene, style is autonomous; it is not linked to function, but draws attention to itself. In other words, style is not motivated or justified by the subject matter, but is its own justification.
Perkins argues that mannerist mise en scene is pretentious because it creates a disunity or imbalance between style and theme, and accuses the director of ‘overdirecting’ the film: ‘What happens on the screen must not emerge as a directorial “touch” detached from the dramatic situation; otherwise the spectator’s belief in the action will decrease or disappear. The director’s guiding hand is obvious only when it is too heavy’ (1972: 77). Perkins prefers classical mise en scene because it creates what he calls ‘an intensity of cohesion’, and what we are calling organic unity and significant form. Mannerist mise en scene, by contrast, is simply decoration that lacks form and creates imbalance and disunity.
Additional aspects of classical mise en scene that may lead to organic unity and significant form include: foreground- background relations; productive use of the frame; and the tension between the choice of long takes or cutting. Filmmakers can choose to establish a productive and significant relationship between foreground and background in an image, as in the use of deep focus cinematography in the work of Jean Renoir, Orson Welles, and William Wyler, where several planes of action remain in play and in focus in the same frame. Another common mise en scene strategy involves using the frame to isolate characters in their own shot, or bring characters together in the same shot. If characters appear in the same frame (either a static frame or linked by camera movement), they are united; but if they are separated by cutting, then they are in conflict, or isolated from each other. William Paul writes that: ‘Where the cutting is used to isolate the individual and his responses, the camera movement, as it reintegrates space, reunites the individual with his group to establish a sense of wholeness’ (quoted in Bordwell, 1989: 179). Finally, filmmakers can choose to shoot a scene in one continuous take, where the camera is left rolling while the whole of the action takes place, or shoot the same scene with several shots. The first option involves the filmmaker filming the action as it unfolds, uninterrupted. The second option involves breaking the action down into individual shots. Each new shot will include a change in camera position, camera angle, shot scale, and so on. Filmmakers have to weigh up the advantages and disadvantages in choosing one technique over another for each scene, since the choice of technique will influence whether the resulting scene will manifest unity or disunity.
Significant form and organic unity can be created in film by combining various mise en scene strategies. However, these techniques should not be analyzed in a mechanical, numerical manner (as they typically are in Film Studies and film production classes), not should one automatically assume a film has significant form because it contains some of these techniques. Each technique needs to be evaluated according to how it is used to represent content, and how it works (or does not work) in relation to other mise en scene strategies.
Like Perkins, Noel Carroll is also a moderate formalist critic. In Interpreting the Moving Image, he carries out a series of film analyses—which he calls interpretations—and privileges organic unity. He aims to explain the presence of features (or parts) in a film and their interrelationships:
[M]y practice of interpretation … tends to be holistic or organic or functional. I interpret features of films, for the most part, in light of their relation to hypotheses about the unity of the works in question. In this, I do not imagine that my interpretations account for every detail of the films I discuss (I am talking about relative unity, not totalized unity), nor do I claim that there may be other (compatible) interpretations of the works I examine. (1998: 10)
In this statement Carroll does not sufficiently distinguish a functional analysis from a holistic-organic analysis, even though such a distinction is necessary to understand the difference between a totalized unity and a relative unity. In an organic unity, as we have already seen, all the parts interrelate to create the whole. This is a totalized unity, and involves examining all the relations between the parts, regardless of function. For Carroll, only those parts that perform the artwork’s intended function need to be examined (relative unity). Carroll develops the concept of relative unity in Chapter Three of Philosophy of Art, in which he advocates examining only those relations that carry out the artwork’s intended purpose or function: ‘According to the functional account, the form of an artwork is correlative to its purpose’ (1999:146). In a functional analysis, one examines the purpose an artwork is intended to fulfil. If it successfully fulfils its function, it is designated as good art, and if it is unsuccessful, it is labelled bad art. The most important functions are representational and expressive. Carroll’s emphasis on functionalism aligns him to a moderate formalist position, since he examines formal, expressive, and representational features in individual films.
Formalist Film Theory
In this section I review key descriptive studies of film form, those that give minimal attention to style or evaluation. The work of these researchers (Christian Metz, Noel Burch, Edward Branigan, Raymond Bellour, Barry Salt, David Bordwell, and Kristin Thompson) is marked by the same qualities: it is systematic, rigorous, methodical, and meticulous in its description of film form. It raises Film Studies to a higher level of precision and clarity—sometimes attempting to reach an axiomatic level, where it endeavours to present its findings as a series of unambiguous postulates.
Christian Metz’s Film Semiotics
In ‘Problems of Denotation in the Fiction Film’ (1974a) Christian Metz identified a finite set of sequence (or syntagmatic) types operative in classical narrative cinema, a paradigm of syntagmas from which a filmmaker can choose to represent events in a particular order. Metz identifies each syntagma from a moderate formalist perspective, that is, by the particular way film form organizes the spatio-temporal relationship between the events it depicts. Syntagmas are distinct because the same events depicted by means of a different syntagma will have a different meaning. These spatio-temporal relationships between the images constitute cinematic language for Metz because they articulate the filmed events in terms of a specific cinematic space and time. In other words, this cinematic space and time confers upon these events a meaning that transcends their analogical relationship to the image.
Metz detected eight different spatio- temporal relationships in total, which constitute eight different forms of image ordering (syntagmas). Metz called the resulting ‘paradigm of syntagmas’ the ‘grande syntagmatique’ of the image track. The grande syntagmatique identifies syntagmatic units only when a change in shot produces a change in meaning—that is, when a spatio- temporal transition (the cut, etc.) on the level of the filmic signifier correlates with a change in meaning on the level of the signified (= the spatio-temporal relationship between the events). Each filmic syntagma is constituted by the same spatio-temporal relationship between its images. As long as the same relationship holds across cuts, there is no change in meaning. A change in meaning therefore occurs when a spatio-temporal transition on the level of the filmic signifier is correlated with a new spatio-temporal relationship between filmed events, for a new relationship signals the end of one syntagma and the beginning of another.
Noel Burch’s Study of Film’s Formal Principles
Noel Burch’s formalist theory of film is presented in his book Theory of Film Practice. In a fashion similar to Metz, although edging towards extreme formalism, Burch created a formal taxonomy of all the possible spatio- temporal relationship that can exist between two shots (1981: 3-16). He identified five different temporal relationship (temporal continuity, measurable ellipsis, indefinite ellipsis, reversal in time, and flashback) and three types of spatial relationship (spatial continuity, discontinuity, and proximity). He then studied the techniques that work to create the last type of spatial relationship (spatial proximity), such as the eye-line match.
Burch supplemented this taxonomy of spatio-temporal relationship by examining the opposition between on-screen and off-screen space (1981: 17-31). On-screen space names the space inside the film frame, and offscreen space lies beyond the film frame. Offscreen space is divided into six segments: the four spaces beyond each frame line, a fifth space—the space behind the camera, and a sixth space—space hidden within the film frame. With the exception of the analytical cut-in, which shows a detail of the previous shot, a cut materializes one area of off-screen space, and consigns the on-screen space to the status of off-screen space. Furthermore, with any given shot, attention can be drawn to any off-screen space via an entrance to and exit from on-screen space, or via the use of off-screen sound. Burch notes that ‘off-screen space has only an intermittent or, rather, fluctuating existence during any film, and structuring this fluctuation can become a powerful tool in a film-maker’s hands’ (1981: 21).
Edward Branigan’s Formal Description of the Point-of-View Structure
Edward Branigan delineates the formal permutations of the point-of-view (POV) structure, which he defines as one ‘in which the camera assumes the position of a subject in order to show us what the subject sees’ (1984: 103). Taking his cue from Burch’s study of the techniques that create spatial proximity between two shots, Branigan identifies the necessary and sufficient conditions for defining the POV structure: two shots in proximate relation, and six elements. Shot A (Point/Glance) is made up of (1) Origin: a point in space, and (2) Vision: the glance from the point. Between shots A and B there is (3) Time: a transition that suggests temporal continuity or simultaneity. Shot B (Point/Object) consists of (4) Frame: from the point in space defined in (1), the camera locates (5) the Object of the glance. Lastly, the viewer links shots A and B through the construction of (6) Mind: the presence of a character or subject (1984: 103).
Branigan identifies variations of this taxonomy that may produce deviant POV structures (for example, no point or several points may be given; there may be uncertainty as to whether a glance has actually occurred, etc.). Also, shot A may be repeated to create a closed, stable structure. Or shot B may occur before shot A, creating a retrospective POV structure. A and B can be separated, delaying the object of the glance; or the object may never be revealed, creating an open POV structure. A continuing POV, in which one character looks at several objects, rendered in several shot Bs or in a subjective travelling shot. A cheated POV occurs when the camera in shot B does not take the point in space identified in shot A (the camera may be placed closer to the object to give the spectator a better view). A multiple POV structure refers to more than one character glancing at the same object (the characters may be in the same shot, or shot B is linked to two separate A shots, one before and one after). An embedded POV structure names the nesting of one POV structure within another character’s POV: ‘For example, in Psycho we see Marion inside her car glance (shot A) at a policeman outside the car who then glances (shot B) at her licence plate (shot C)’ (Branigan, 1984: 117). Finally, the reciprocal POV structure names moments when the object revealed by the glance is another person who returns the glance, which is common in conversation sequences.
Raymond Bellour’s Shot-by-Shot Analyses
Raymond Bellour is well known for his meticulous shot-by-shot analyses of key segments from classical Hollywood films. He usually combines his moderate formalist perspective with interpretation, in which he identifies a film’s symptomatic meanings using psychoanalysis. In his predominantly formalist analysis of The Big Sleep (Howard Hawks, US, 1946) in ‘The Obvious and the Code’ (2000: 69-76), Bellour takes a seemingly innocuous, self-contained scene towards the end of the film, in which Marlowe (Humphrey Bogart) and Vivian (Lauren Bacall) drive from Eddy Mars’ (John Ridgely) country home to Geiger’s (Theodore von Eltz) home. Hawks uses 12 shots in this classically constructed economical scene that simply represents the characters’ transition between two spaces. Bellour examines the formal relations that exist between these 12 shots to determine how they are integrated into the film, how their obviousness is constructed by codes. His analysis is based on the semiotic hierarchy between the finite, underlying (latent, nonobservable) reality and potentially infinite surface (manifest, observable) reality. The underlying reality is ‘an imperceptible content lending structure to the perceptible insofar as it signifies and conveys precisely the historical experience of the individual and group’ (Deely et al., 1986: xiv). Semioticians call this finite, non-perceptible, underlying system, which lends structure to the perceptible, a ‘system of codes’. Bellour incorporates this opposition into his analysis—and, indeed, into the title of his essay—in which the surface, perceptible reality is designated the ‘obvious’ and the underlying system that confers meaning on the surface reality is called a system of ‘codes’. Bellour analyzes the action of six codes in this scene: differences in framing, static/moving camera, camera angle, characters in the frame, presence/absence of speech, and the relative duration of each shot.
Using a table summarizing his analysis, plus 12 frame enlargements, Bellour enumerates the similarities and differences between the six codes from shot to shot. For example, between shots 1 and 2 he notes the following similarities: identity of duration, presence of the same two characters, and same camera angle; and the following differences: shot 1—moving camera, shot 2—static camera; no dialogue/dialogue; outside the car/inside the car. Between shots 2 and 3 he notes the following differences: shot 2—long take, shot 3—short take; two characters in the frame/one character (Marlowe); medium shot/close up; centring of dialogue on one character (Marlowe, on screen). From shots 3 to 4 he notices the following differences: in shot 4 Vivian (rather than Marlowe) is depicted; the angle has changed; and dialogue comes from both characters (that is, it comes from on- and off-screen).
At this stage of the analysis Bellour comments on the distribution of characters, dialogue, and time. In terms of characters, shots 3, 4, 5, 6, 8, 9, and 11 focus on individual characters, while 1, 2, 7, 10, and 12 unite the couple in the same frame. In the transition from the two-shot to the one- shot (which happens 3 times), Vivian is privileged twice and Marlowe once. Bellour concludes that Vivian is privileged in terms of the image (she appears on her own 4 times, Marlowe only 3 times). In terms of dialogue, Vivian speaks only when on-screen, whereas Marlowe speaks both when on- and off-screen. Bellour concludes that Marlowe is privileged in terms of dialogue. In terms of duration, the middle of the scene (shot 7) lasts almost as long as the other shots combined.
From this analysis Bellour draws a number of general conclusions regarding the formal structure of classical Hollywood films. His main conclusion is that a balance is maintained between symmetry (a repetition) and dissymmetry (variation) of codes. That is, the high number of shots allows for a huge variation of changes, but this potential is limited by means of the repetition and alternation of codes.
Barry Salt’s Statistical Style Analysis
Statistical style analysis (or stylometry) belongs to the extreme formalist tradition (and does not maintain a strict distinction between form and style). It analyses style numerically, by measuring and quantifying a film’s formal parameters. At its simplest, the process of measuring involves counting elements, or variables, that reflect a text’s style, and applying statistical tests to the data.
Statistical style analysis has three standard aims: (1) to offer a quantitative analysis of style, usually for the purpose of recognizing patterns, a task now made feasible by the use of computer technology. In language texts, the quantitative analysis of style and pattern recognition is usually conducted in the numerical analysis of the following variables: word length, or syllables per word, sentence length, the distribution of parts of speech (the different percentage of nouns, pronouns, verbs, adjectives, and so on in a text), calculating the ratio of parts of speech (for example, the ratio of verbs to adjectives), or by analyzing word order, syntax, rhythm, or metre; (2) for the purposes of authorship attribution, in cases of disputed authorship of anonymous or pseudonymous texts; and (3) for purposes of identifying the chronology of works, when the sequence of composition is unknown or disputed (for example, Plato, Shakespeare’s plays).
The first aim, the quantitative analysis of style, involves descriptive statistics. The second and third aims (authorship attribution and chronology) involve both descriptive and inferential statistics. As its name implies, descriptive statistics simply describes a text as it is, by measuring and quantifying it in terms of its numerical characteristics. The result is a detailed, internal, molecular description of a text’s (or group of texts’) formal variables.
Inferential statistics then employs this formal description to make predictions—it uses this data as an index, primarily an index of an author’s style, or to put the author’s work into chronological order on the basis of measured changes in style of their work over time. Whereas descriptive statistics produces data with complete certainty, inferential statistics is based on assumptions the statistician makes on the basis of the descriptive data. The assumptions the inferential statistician extrapolates only have degrees of probability rather than certainty.
In film analysis, the descriptive statistics can be used to quantify the formal parameters of the shot in a film or group of films. It is primarily a systematic version of mise en scene criticism, and is more credible and valid because it downplays the critic’s subjective impressions of a film in favour of a more detached, systematic, and explicit mode of analysis. One disadvantage is that it borders on extreme formalism. In ‘Statistical Style Analysis of Motion Pictures’ (1974), in his book Film Style and Technology (1992), and his study of the films of 1999 (2004), Salt describes the individual style of directors by systematically collecting data on the formal parameters of their films, particularly those formal parameters that are most directly under the director’s control, including: duration of the shot (which involves the calculation of average shot length); shot scale; camera movement; angle of shot; and strength of the cut.
Salt then represents the quantity and frequency of these formal parameters in bar graphs, percentages, and ‘Average Shot Lengths’. When he compares and contrasts the form of the films of different directors, he moves into the realm of stylistic analysis rather than simply formal description. Style in this sense designates a set of measurable patterns that significantly deviate from contextual norms. As just one example, Salt calculated that the average shot length of a Hollywood film between 1940-5 is 9.5 seconds (1992: 231). An early 1940s film with an average shot significantly higher than 9.5 seconds deviates from the norm, and is therefore a significant indicator of style. Salt (1992:231) mentions Vincente Minnelli’s The Clock (US, 1945), which has an average shot length of 19 seconds.
Historical Poetics and Neoformalism
For more than two decades David Bordwell and Kristin Thompson (the latter under the name ‘neoformalism’) have been formulating and practicing historical poetics of film.
Since Aristotle, the term ‘poetics’ has designated the process of ‘making’, the activities and techniques involved in producing a (well-made) work of art. In its turn, poetics presupposes ‘know-how’, a set of skills or procedures in making a work of art. As such, poetics is sharply distinguished from aesthetics. Whereas poetics names the activity of making, aesthetics refers to the reception of art, including its contemplation and evaluation.
The process of making is an external property of an artwork because it refers to the decision-making activity—the choices an artist makes—prior to the artwork’s completion, or final form. The form of an artwork is not autonomous but the end result of purposeful activity. As the art historian Michael Baxandall argues:
The maker of a picture or other historical artefact is a man [s/c] addressing a problem of which his product is a finished and concrete solution. To understand it we try to reconstruct both the specific problem it was designed to solve and the specific circumstances out of which he was addressing it. (1985: 14-15)
For Baxandall, artists use their know-how and knowledge of a medium to solve a problem. To analyze an artwork from this perspective entails explaining why the artwork has the form it does, rather than simply describing form as an autonomous entity. Explanation identifies the function formal parts play in relation to one another and in relation to the whole artwork.
Only the solution (the artwork) is pregiven. The art historian needs to work backwards from the solution and reconstruct the problems it is addressing, together with the artist’s intentions. Baxandall is well aware of the problems surrounding the concept of intentionality, together with the difficulties in reconstructing the particular problem an artwork was designed to solve and the specific circumstances out of which the artist addressed the problem: ‘We cannot reconstruct the serial action, the thinking and manipulation of pigments that ended in Piero della Francesca’s Baptism of Christ, with sufficient precision to explain it as an action. We address the finished deposit of an activity we are not in a position to narrate’ (1985: 13-14). Referring to the philosopher Karl Popper, Baxandall continues: ‘we do not re-construct the actor’s thought but produce an “idealized and reasoned reconstruction” of an objective problem and objective situation on a level different from [the artist’s] actual reasoning’ (1985: 14). The art historian’s account of an artist’s activity in making a work of art is not an actual representation of that activity, but is a simplified, albeit rational reconstruction of it.
For Bordwell, film poetics answers two basic questions: ‘First, how are particular films put together? Call this the problem of films’ composition. Second, what effects and functions do particular films have?’ (1989: 263). Bordwell adds: ‘At this moment I believe that the most promising avenues for poetic analysis are those opening onto compositional processes of form and style’ (1989: 270-1). The study of composition is central to film poetics because it entails examining both the concrete activities involved in filmmaking plus the fundamental principles and conventions which guide that activity. The poetician looks for the options available to filmmakers at a specific historical period, and studies the preferred set of choices they make in putting together an individual film. As with Baxandall’s model of art history, Bordwell’s film poetics conceives a film as the end result of purposeful activity, and attempts to rationally reconstruct that activity from historical data. In studying the aesthetic effects achieved by a particular choice of options, Bordwell (1985) has primarily examined perceptual- cognitive effects, and has concluded that spectators are active when watching a film. Finally, in relation to determining how films serve specific functions in particular historical contexts, Bordwell writes: ‘To analyze a film’s composition and function requires us to consider what processes brought it into being (for example, to what problems does its composition represent an attempted solution?) and what forces have mobilized it for various purposes’ (1989: 265). Again, the first part mirrors Baxandall’s study of artworks as concrete solutions to problems.
Kristin Thompson makes explicit and clarifies many of the assumptions implicit in formalism and poetics. Neoformalist film analysis, she argues, closely adheres to and is guided by the film. It does not impose theoretical doctrines onto a film simply to illustrate the theory in a self-confirming manner. Instead, it treats each film individually, viewing it as an artificial construct formed by filmmakers choosing from and reinventing filmmaking strategies, norms, conventions and techniques. Two key terms for recognizing a film’s individuality include ‘function’ and the ‘dominant’. Thompson uses the concept of ‘function’ differently from Bordwell. For Thompson, function ‘is crucial to understanding the unique qualities of a given artwork, for, while many works may use the same device, that device’s function may be different in each work’ (1988: 15). The neoformalist does not assume a particular device or technique has a fixed function from film to film. For example: ‘Bar-like shadows do not always symbolize that a character is “imprisoned”, and verticals in a composition do not automatically suggest that characters on either side are isolated from each other’ (1988: 15). Given devices do not always or automatically signify a fixed meaning, for ‘they can serve different functions according to the context of the work, and one of the analyst’s main jobs is to find the device’s function in this or that context’ (1988: 15).
The dominant names ‘the main formal principle a work or group of works uses to organize devices into a whole … The dominant will pervade the work, governing and linking small-scale ones; through the dominant, the stylistic, narrative, and thematic levels will relate to each other’ (Thompson, 1988: 43). In a standard film, the dominant names the most typical devices and techniques, together with their organization, while in a highly original film, it names the most unusual devices.
The dominant is an important concept for rethinking organic unity. It is a principle of organization that does not simply involve all the elements of a film co-existing and mutually relating to one another. Instead, it involves a small number of elements organizing the remaining elements of a film. A hierarchy is set up between these more important organizing elements (the dominant) and the less important subordinate elements. Thompson notes that, without a concept such as the dominant, ‘we would be condemned to study every device in a film with equal attention, for we would have no way of deciding which were the more relevant’ (1988: 92). The concept of the dominant is closely related to Carroll’s concept of relative unity, for both avoid the need to study a film as a totalized unity by setting up a hierarchy between each film’s more important and less important elements. One of the skills of film analysis is to identify and privilege these dominant elements.
Filmmaking Manuals
At present, Film Studies only engages superficially (if at all) with filmmaking procedures. In this section I aim to overcome this superficial engagement by grounding Film Studies in filmmaking, but without compromising the integrity of Film Studies. In fact, I aim to reinvigorate Film Studies by expanding its disciplinary boundaries, and by making it more relevant to both Film Studies and film production students.
The essence of this expansion involves the combination of two types of knowledge—the declarative knowledge of Film Studies (the propositions of film aesthetics and film theory), and the procedural knowledge (technical know-how) of filmmaking. To bring into effect this expansion, Film Studies needs to be supplemented with the technical knowledge of filmmaking as embodied in filmmaking manuals.
Filmmaking manuals identify at least three fundamental skills at the core of filmmaking:
- Visualization and shot flow (shot plans and storyboarding)
- Blocking the action (staging the ‘zones of action’)
- Filming the action (not simply recording the action, but using visual rhetoric to create dramatic emphasis of action points).
1. Visualization and shot flow. Steven D. Katz has codified the director’s tacit knowledge in his two books Film Directing: Shot by Shot (1991) and Film Directing: Cinematic Motion (1992). He argues that the two key terms for the working director are ‘visualization’ and ‘shot flow’. Visualization involves hands-on pictorial design to work out the content of the shots, plus their sequencing; and a moment-by-moment evaluation of the images as they are shaped, making different choices from the many options available.
According to Katz, the visualization of options available and revising the choices made should be enabled and constrained by the content of the screenplay. The problem for the director to solve is the representation of the screenplay’s content in visual and narrational terms, and his or her solutions are initially worked out in storyboards. This stage of filmmaking can be creative and imaginative, because ‘the filmmaker is confronted with a variety of visual decisions that the screenplay does not address’ (1991: 5).
An unimaginative or novice director may see his or her task as simply filling the frame with the content of the screenplay, and will not try out the most effective and efficient options for narrating the story in a particular scene. Such a film will be merely pedestrian and under-directed, and will create what Perkins calls an imbalance between credibility and coherence, by privileging credibility and not exploiting film’s expressive capacity. Such a film will lack organic unity and significant form. Other directors may over-direct by imposing pretentious decorative flourishes on the screenplay, leading to a mannerist mise en scene. The ideal balance, as practitioners like Katz and critics like Perkins argue, is for the director to choose to intensify part of the script’s possibilities, to visualize it in terms of significant form or organic unity.
The film director Edward Dmytryk agrees: ‘If there is one rule that should hold for film it is that the techniques of filmmaking must be at the service of the material filmed rather than the other way around’ (1988: 62). Michael Rabiger holds the same opinion in his manual on film directing: ‘Good camerawork, composition, and blocking [are] always trying to show relatedness. This helps to intensify meanings and ironies, and reduces obvious signification through editing’ (1997: 53). His comments parallel those of Perkins, who also argues that films should show ‘relatedness’, and that film techniques need to intensify meanings without drawing attention to themselves.
Katz recommends visualization via story- boarding because the storyboard conveys the shot flow of a scene. ‘Shot flow’ is the name Katz gives to the kinetic effect of a sequence of shots, which is largely determined by the physical transformations brought about by shot size, camera angle, and the timing of cutting points. He uses the metaphor of a river to characterize shot flow, for an effective shot flow links together shots in a single, uninterrupted process. With the concept of ‘shot flow’ Katz has invented a powerful metaphor because it suggests a series of shots forming an organic unity.
2. Blocking the action. Much of Daniel Arijon’s directing manual Grammar of the Film Language (1991) is taken up with outlining an elaborate set of options for blocking action and rendering it on film. His overwhelming number of countless variations can be reduced to a simple series of variables:
Blocking:
- Change the position of actors within the six different sectors of the image
- Change the position of actors from zone to zone
- Change the actor’s body level (standing, sitting, lying down).
Filming:
- Render on film using a static or moving camera
- Render on film using cutting or a long take.
In relation to zone changing, he writes:
With changes of zone the group can move from zone to zone, expand to several zones or contract from several to only one zone. There is no limit to the number of areas that can be employed but three to five is generally enough since each area can be used several times if the development of the story so requires. (1991: 542)
Arijon develops his last point (‘if the story so requires’) further ahead when he writes: ‘A pattern of movement that expands and contracts periodically during the sequence should be at the service of the story and not arbitrarily imposed on a scene’ (1991: 554). Arijon clearly favours classical over mannerist mise en scene.
3. Filming the action. Finally, guided by visualization and blocking decisions, a director commits the scene to film. Directors who use the camera simply to record a scene are ‘under-directing’, for they end up producing a static, passive film that does not exploit the expressive capacities of film. Directors who use flashy camera movements and an excessive number of shot transitions (that is, unmotivated by and not serving the story) can be accused of ‘over-directing’ their film.
For Katz, a well-directed film uses film techniques to add dramatic emphasis to the story. Katz’s comments parallel Perkins’ discussion of classical mise en scene, in which the director uses the expressive elements of film to intensify the significance of the credible image without drawing attention to the expressive elements themselves.
Of course, a director can decide to film the entire scene with a fixed master shot keeping both subjects immobile while talking to one another. This is an instance of underdirecting, although as an option it cannot be ruled out entirely, for its under-stated nature may reinforce the nature of the conversation, or the dramatic relationship between the characters, which is the case for example with Jim Jarmusch’s early films Stranger than Paradise (US/West Germany, 1984) and Down by Law (US/West Germany, 1986). In analyzing and evaluating a film, each scene needs to be considered on its own terms. Film analysis and evaluation needs to be inductive, bottom up, and empirical, privileging perception of the work itself as a material object, rather than deductive, top down, and idealistic, which is dogmatic, reductionist and makes generalizations that are easily refutable through the close analysis of individual shots and scenes.
All three stages of directing—visualization, blocking, and filming—should be planned together to create a coherent film, a film unified by a clear vision and design. Katz’s codification of the director’s tacit knowledge is premised on the same principles as Perkins’ guideline for film evaluation: a classically constructed film demonstrating organic unity and significant form.
In Directed by Steven Spielberg (2006), I analyze Spielberg’s filmmaking practice through this combination of film aesthetics and technical know-how. I examine Spielberg’s day-to-day decision-making process in solving filmmaking problems, and analyze the blocking and filming options available to him, especially those relating to style and narration. From the available filming options, Spielberg frequently makes a small number of habitual choices in the visualization and technical execution of each shot and scene in his films. I argue that a formalist analysis, informed by filmmaking procedures, is ideally equipped to examine in detail the structure and unity of a filmmaker’s oeuvre.
Conclusion
In his defence of modernity, Habermas takes to task those who conflate cultural modernity with what Habermas calls ‘societal modernization’ (including Daniel Bell, Friedrich Nietzsche, Martin Heidegger, Gilles Deleuze, Michel Foucault, Jacques Derrida and other neoconservative, postmodern theorists). Although the idea of cultural modernity emerged from the eighteenth-century Enlightenment project, during the nineteenth century, modernity became equated with industrialism and the social reorganization it entails—particularly reification and alienation (the defining characteristics of societal modernization). In their discussion of modernity, neoconservative theorists concentrate exclusively on societal modernization, which accounts for their negative evaluation of the whole of the modernist/Enlightenment project. They argue that modernity necessarily leads to reification and alienation because it has simply resulted in the production of technologically exploitable knowledge. Stephen White succinctly expresses this negative side of modernity when he writes: ‘Modernization in the West has thus generated a pathology: an unbalanced development of its potential’ (1995: 8). However, following Habermas, White argues that this imbalance can be resisted, that modernity is not an ‘iron cage’, as Weber suggested. For Habermas, the rationality of cultural modernity (which emphasizes the process of communicative rationalism and the concrete realization of universal freedom) must be kept separate from the Zweckrationalitat (bureaucratic rationality and instrumental reason) that merely informs the economy and polity. However, postmodernists fail to make this distinction, and thereby fail to understand the value of modernity, including the achievements resulting from the formalist analysis of film’s inner logic.