Forms and Functions in Documentary Filmmaking

American Direct Cinema

Thesis (M.A.) 2008 74 Pages

Film Science



1. Introduction

2. Approaching Documentary Film
2.1. Documentary Practice 6
2.2. Reference
2.3. Representation
2.4. Reception

3. The Emergence of Direct Cinema
3.1. Influences
3.2. Technology
3.3. Television

4. Direct Cinema in Practice
4.1. The Drew Associates and the Crisis Structure: PRIMARY
4.2. Nonfiction Features: The Maysles Brothers’ SALESMAN
4.3. Reality Fictions: Frederick Wiseman’s TITICUT FOLLIES
4.4. Beyond Direct Cinema

5. Conclusion

6. References

7. Films


The subject of this paper is documentary film and Direct Cinema as a particular movement in documentary history. Direct Cinema emerged as an innovative form of filmmaking in the United States in the early 1960s, using new technologies and revitalizing documentary in a break with both traditional forms of both documentary and classical Hollywood cinema. Direct Cinema developed anobservationalfilmmaking method that relied on giving up control by minimizing the filmmakers’ intervention before and during the shooting, with no preconceptions of the finished product. The methods the filmmakers employed drew onrealisttechniques such as long takes and free-moving cameras, promoting anuncontrolleddocumentary of immediacy and focusing on thereality effectof the moment of shooting. The result was expressive footage, for which the filmmakers developed a form of representation that relied on the inherent continuity of the filmed event, avoiding the narration and interpretation common to traditional documentary films.

In recent years, documentary films and program formats have become increasingly popular in cinema and on television, while the theoretical discussion of documentary forms has become a growing field of academic research. However, there is an ongoing debate about defining what constitutes documentary, and how it differs from fiction. Like fiction film, documentary is creatively shaped, building on dramatic structures and making use of continuity editing such as cross-cutting and shot/reverse shot employed in classical narrative cinema. Taken alone, the status of the filmic image cannot provide a distinction between documentary and fiction, while strategies of representation in film may also fail to offer a basis for a distinction, since they may be shared by both fiction and documentary. For some, approaching documentary as reception offers a solution to this problem of distinction.

This paper will explore the question of what constitutes documentary – on the basis of documentary material, strategies of representation, and documentary reception – through an analysis of Direct Cinema, and investigate the contribution that Direct Cinema made to the conceptualization of documentary.

The first chapter will introduce and discuss theoretical approaches to documentary concerning the nature of documentary images, forms of narration and representation in documentary, and recent views of documentary as a mode of reception. The chapter looks at how a combination of documentary images and textual strategies (including conventions such as voice-over narration) together constitute documentary in the process of reception by the viewer.

The second part of this paper focuses on the movement of Direct Cinema within documentary history. Looking at the historical and technological factors, along with the aesthetic influences surrounding the emergence of Direct Cinema in the United States in the 1960s, the chapter will provide the basis for a closer look at individual Direct Cinema films.

In the third part, three Direct Cinema films by different directors will be discussed. The analysis of the films will investigate how the filmmakers employed Direct Cinema methods, and what they sought to achieve by doing so. The discussion of the films will show that Direct Cinema changed in its subjects and subject matter during the course of its heyday in the 1960s. The chapter will furthermore look at other filmmakers who criticized the Direct Cinema approach and set out to develop a more assertive kind of documentary that nevertheless built on foundations Direct Cinema had laid.

Situated between the fields of film studies and American cultural history, this paper seeks to place documentary and Direct Cinema in a theoretical, historical and cultural context – showing the contribution the movement made both as an innovative form of filmmaking and as an influential force in American culture.

2. Approaching Documentary Film

Documentary has a long tradition as a distinct form of film, and has established its own canons of filmmakers and films. At the core of documentary stands a distinction between fiction and nonfiction. Whereas fiction involves the viewer in constructed imaginary worlds, nonfiction draws on factual material that is bound to the actual or historical world. The distinction between fiction and nonfiction has pragmatic relevance in film, a fact reflected in the large number of films that are labeled and regarded as documentaries.

The growing academic discussion of documentary has widened the field and problematized documentary’s status as a form distinct from fiction film. Questions have been posed over the ontological status of images that had been seen as a foundation of documentary film. The view of documentary as a distinct form of representation following exclusive narrational strategies has also been problematized. Some critics, believing that distinctions could no longer effectively be drawn between documentary and fiction on the basis of images and representational strategies, shifted their attention towards documentary reception.

The first part of this chapter examines the emergence of documentary within the history of film, and its relationship to fiction film in historical terms, as well as considering ways of approaching the practice of documentary filmmaking. The second part looks at the semiotic nature of filmic images and discusses questions of authenticity in documentary material. The third part considers forms of representation in documentary, drawing in particular on Bill Nichols’ model of representational modes to provide an overview of different documentary forms and narrative strategies. The fourth part sets out recent views of reception as crucial to approaching documentary.

2.1. Documentary Practice

The documentary idea of recording real events goes back to the beginnings of film. The first films by the Lumière brothers can in fact be regarded as early documentaries, and were referred to by the brothers asactualités. With his invention of thecinématographein 1895, Louis Lumière, together with his brother Auguste, set out to film everyday life, such as the arrival of a train at a station in L’ARRIVÉE D’UN TRAIN EN GARE (1895), or workers leaving a factory in LA SORTIE DES USINES (1895). Some of the films involved performances for the camera but the brothers rejected the use of actors for their short, single-shot, unedited films. These and other early films were long regarded as “primitive;” however, the filmmakers of the time introduced the use of narrative elements such as the opening and closing doors in LA SORTIE DES USINES (Elsässer 1990).

Soon after the first films by the Lumière brothers, the new technology of film was employed for the first fiction films. Georges Méliès made imaginative use of the technology, most famously in A TRIP TO THE MOON (1901), considered by many to be the first science-fiction film. Two years later Edwin S. Porter used new editing techniques such as cross-cutting scenes in parallel action in THE GREAT TRAIN ROBBERY (1903). Another milestone was D.W. Griffith’s BIRTH OF A NATION (1915), one of the first films to establish continuity at feature length. These and other films laid the foundations for classical Hollywood cinema based on the principle of continuity – in storytelling and in editing, creating spatial and temporal coherence (Bordwell 1993).

In the 1920s, a decade after the first fiction films, documentary film developed as a counter to the classical studio productions. Robert Flaherty, considered to be the father of documentary film in the United States, explored remote places and people’s struggle with nature. Flaherty’s NANOOK OF THE NORTH (1922) portrays Nanook, an Inuit, and his family in the Arctic.It is now regarded as the first feature-length nonfiction film. For MOANA (1926) Flaherty went to the Pacific island of Samoa to film its people and their traditional lifestyle. It was for this film that John Grierson, himself a pioneer in the field, first used the worddocumentaryin a review.

In contrast to Flaherty’s poetic films, Grierson, in his work in Britain and later in Canada, promoted a documentary of social responsibility, featuring socio­political issues. Grierson and his team at the documentary film units of the Empire Marketing Boards, and subsequently at the General Post Office in Britain, made hundreds of films between 1928 and 1939, developing a style featuring direct-address narration that mediated an argument to the viewer. Early in the history of nonfiction film, John Grierson famously defined documentary as the “creative treatment of actuality.” With this definition Grierson already drew attention to the fact that documentaries are not pure representations of reality but that the material is selected and shaped – just as in fiction films.

However, in contrast to fiction film, which is treated as a product of the imagination that constructs unique, imaginary worlds, documentary draws on factual material in its reference to the actual, or the historical world. Fictional worlds may closely resemble the actual world – especially in works drawing onrealistrepresentation – while nonfiction films are just as constructed as their fictional counterparts, as Grierson suggested. Nevertheless, the distinction has pragmatic relevance. Bill Nichols, one of the leading documentary theorists, distinguishes between fiction film and documentary as follows:

Although fiction films employ elements of realism in the service of their story, the overall relation of film to the world is metaphorical. Fiction presents the likeness of actual events, motives, appearances, causality, and meaning. Documentary, on the other hand, takes up and uses an indexical relation to the historical world (Nichols 1991, 116).

For Nichols, images and sounds recorded in the actual or historical world form the evidential basis of a documentary’s authenticity. Like fiction films, documentaries tell stories but, according to Nichols, it is less thestorythat is crucial in a documentary but anargumentabout the historical world. Nichols draws parallels with story and plot in fictional texts; in his view, evidence and argument in documentary have a similar relationship to that between plot and story in fiction (Nichols 1991).

The documentary practice of filmmaking may be employed for different purposes. Michael Renov identifies four fundamental tendencies in documentary:

1. To record, reveal, or preserve
2. To persuade or promote
3. To analyze or interrogate
4. To express

According to Renov, these tendencies operate asmodalities of desire. The first mode is the most elemental function and can be understood as the mimetic drive of cinema. The second mode of persuasion can be found in documentaries in the Grierson tradition. Documentaries in the third mode build on questions of representation and processes of interrogation, while the fourth mode encompasses the primarily aesthetic function of making something beautiful out of authentic footage (Renov 1993).

In recent years, boundaries have been blurred by media productions ofreality TV,drama-documentariesand other hybrid genres. Nevertheless, large numbers of films are produced and labeled as documentaries. Nichols considers documentary from the point of view of the filmmaker, including the institutional framework of production, the text, and the audience. He offers three different, but not contradictory definitions. First, one simple way of defining documentary is that documentaries are what the filmmakers, organizations and institutions that produce them make. Documentary filmmakers share assumptions about their work, although their subject matter and style may be different. Producers and institutions, such as broadcasters, also share ideas about their output. Furthermore, all these practitioners are accustomed to labeling their films. Carroll introduced the concept ofindexto draw attention to labels by filmmakers, producers, and distributors, which signify the forms and genres to which a film belongs (Carroll 1996).

Second, documentaries can include characteristic features. For example, voice-over commentary is widespread in documentary films; also common are interviews and location footage, as opposed to studio recording. The people seen in a documentary are generally not actors as in fiction films (although fiction films may also use amateurs) but may be calledsocial actors(Nichols 2001).

A third approach to documentary can be considered from the point of view of the viewer, who holds certain assumptions and expectations – recognizing a representation of the actual world that forms the basis for the reception of the film as documentary.

2.2. Reference

Reference may be defined as the relation of asignto itsreferent. The semiotician C.S. Pierce differentiated signs betweenicons,indices, andsymbols. Icons, according to Pierce, are “likenesses... which serve to convey ideas of the things they represent simply by imitating them.” Indices “show something about things, on account of their being physically connected with them.” Finally there are symbols, “which have become associated with their meaning by usage” (Pierce 1998, 5).

Writing around the turn of the 20th century, Pierce did not include film in his discussions. But he did consider the medium of photography. According to Pierce, at first glance one might associate photographs with the class of icons, representing the likeness with an object. However, as Pierce suggests, a photograph not only represents the imitation of an object but carries the physical trace and thus belongs to the class of indices:

Photographs, especially instantaneous photographs, are very instructive, because we know that they are in certain respects exactly like the objects they represent. But this resemblance is due to the photographs having been produced under such circumstances that they were physically forced to correspond point by point to nature. In that aspect, then, they belong to the second class of signs, those by physical connection (ibid.).

The medium of photography was seen as a breakthrough in the history of the arts for its ability to capture a moment in time and space and record it. André Bazin, the realist film theorist and critic, like Pierce believed in the indexical quality of photographic images. In his influential essayThe Ontology of the Photographic Imagehe writes: “For the first time, between the originating object and its reproduction there intervenes only the instrumentality of a nonliving agent. For the first time an image of the world is formed automatically, without the creative intervention of man” (Bazin 1967, 13). Free of human intervention and relying solely on technology, photography can thus be considered objective in Bazin’s view, lending it the quality of credibility. As such, the viewer of a photograph will accept the object that it represents as existent in reality.

The film theorist Brian Winston sees the camera as an instrument of science comparable with a thermometer, giving photographs what Winston calls “scientific inscription” (Winston 1993).1 For example, photographs taken at crime scenes are treated as legitimate documents of evidence in court. Likewise, film recordings are used in the legal system. Video surveillance has become a common technique for public security and is used to investigate crimes and identify suspects.2

Documentary draws on this inscription of authenticity in its footage. This authenticity may be emphasized in different ways. Filmmakers of Direct Cinema, for example, employed filmmaking techniques of long unedited sequences filmed with a minimum of intervention for an increased impression of authenticity, seeking to generate a reality effect experienced by the viewer.3 However, just as photographs can be manipulated, filmic images may have the impression of authenticity that can in fact be constructed:

Certain technologies and styles encourage us to believe in a tight, if not perfect correspondence between image and reality, but the effects of lenses, focus, contrast depth of field, color (...) seem to guarantee the authenticity of what we see. They can all be used, however, to give the impression of authenticity to what has actually been fabricated or constructed (Nichols 2001, xxii).

As Nichols points out, aesthetic effects lending the impression of authenticity can be employed in fiction filmmaking. For example, BLAIR WITCH PROJECT (1999) draws on hand-held cinematography, suggesting the footage was shot by the protagonists in the film. But on the other hand, a documentary filmmaker may also use fictional reenactments to construct visuals of an event in the past of which there is no footage, as in Errol Morris’ THE THIN BLUE LINE (1988).

This shows that the indexical nature of filmic images as such cannot provide evidence for the fictional or nonfictional status either of the film, or of the events they represent – that is, whether they are constructed or authentic. According to the film theorist Phillip Rosen, “In cinema, indexicality designates the presence of camera and sound-recording machinery at the profilmic event, which, in turn, guarantees that the profilmic event really did exist in the past” (Rosen 1993).4 Therefore, the filmic image authenticates the existence of the profilmic event but it cannot confirm its status.

In this light, the images in a film cannot serve as the basis of its definition as a documentary. But what, then, authenticates documentary? As Nichols points out, documentary authenticity can only be inferred by the viewer as it can otherwise not be guaranteed:

There is no other guarantee than the inference we ourselves make, based, often, on very good evidence such as the similarities between the photographic image we see and others of the same subject (for public figures and well-known places and events), on explicit assurances of authenticity by the film itself, and on our familiarity with everyday conduct and how it differs from fictional representations (Nichols 1991, 151).

Thus, documentary images do play a role but have to be seen in the context of the film in which they are presented. The film’s organization of the material according to documentary forms of representation provides the basis for what constitutes documentary.

2.3. Representation

In the course of documentary history, different strategies of narration and representation have been employed. Bill Nichols developed a model of representational modes, which, for him, are the basic ways of organizing documentary material according to certain recurrent features or conventions. Nichols identifies four dominant modes: expository, observational, interactive, and reflexive (Nichols 1991).

Theexpositorymode features direct-address commentary that often dominates the images and is highly didactic.5 Theobservationalmode seeks to capture events by observing without intruding, enabled by lightweight camera equipment and synchronous sound recording. Theinteractivemode of representation relies on interviews and often focuses on the act of filmmaking through the participation of the filmmaker in the film. The more complexreflexivemode mixes observation with interviews and alienates other modes. Nichols’ representational modes for documentary film resemble David Bordwell’s modes of narration in fiction film: the classical, art cinema, parametric, and historical materialist. Bordwell defines a mode of narration as a “historically distinct set of norms of narrational construction and comprehension” (Bordwell 1985, 150). Nichols’ model also follows the historical dominance of each mode at certain times and in certain regions. His model is based on the idea that representational modes belong to a “dialectic in which new forms arise from the limitations and constraints of previous forms and in which the credibility of the impression of documentary reality changes historically” (Nichols 1991, 33). The modes are, therefore, both formal and historical. The boundaries between the different modes are, however, rarely clear-cut – and many documentaries feature characteristics of different modes.

Nichols’ model begins with documentaries in theexpositorymode. The classic expository documentaries were made in the 1920s and 1930s as vehicles for information, as a response to fiction films that merely served as entertainment. Films in the expository mode focus on a rhetorical continuity of narration rather than on visual continuity. The authoritarian style of direct address emphasizes the impression of objectivity, laying claim to a thorough basis in fact and well-grounded argument, giving the films a didactic character. The expository mode often builds a dramatic structure towards a solution or answers to addressed problems or questions.

Theobservationalmode is characterized by indirect address, avoiding voice-over commentary and interviews. This mode is linked to ethnographic filmmaking, which attempts to record the practices and rituals of other cultures and societies. But according to Nichols’ model, it was Direct Cinema that established the observational mode in documentary filmmaking at the beginning of the 1960s. Direct Cinema was made possible by lightweight new cameras and sound technology, enabling filmmakers to shoot in long takes and to follow their subjects freely while recording synchronous sound. Rather than posing problems and finding solutions as common in expository documentary, the films offer the audience a new experience ofbeing there. In the observational mode the intervention of the filmmaker in the profilmic event is minimized. In contrast to the expository mode, in which a rhetorical continuity of argument is established, the observational mode seeks to follow the continuity of the shooting, maintaining it as far as possible in the editing. In contrast to the cause and effect logic in expository documentary and in classical narratives, observational modes of representation can include sequences that Nichols callsdeadoremptytime, where nothing of narrative importance occurs but where time is depictedas it is lived(ibid. 40).

The observational ideal of not intervening is reversed ininteractiveorparticipatorydocumentary, in which the filmmaker actively intervenes in the profilmic events and interacts with the subjects. Common features of interactive documentaries are interviews or verbal statements given by subjects speaking directly into the camera, or speaking parallel to the images asvoices of witness. These statements and interviews by social actors serve to construct authority. Interactive documentary can take on many different forms but, for Nichols, all draw their social actors into direct relationship with the filmmaker. As Nichols notes, this raises questions over the power relationship between the interviewer and the interviewee. Continuity in the interactive mode is built around presenting individual viewpoints or the conversation between social actor and filmmaker, establishing apresent tense qualityof encounter without rhetorical arguments in commentary (ibid. 44). Forms of interactive documentary like historical reconstructions make use of archival material. A distinct feature of the interactive mode is the reference to the historical world through witnesses with information about a certain time and place, raising the viewer’s expectation of first-hand information and knowledge (ibid.).

In thereflexivemode, representation itself becomes the focus. Reflexive documentary texts draw attention to their constructedness, and thus raise questions over accepted forms of representation. Filmmakers might directly address the issue of representation by exposing and alienating other modes. The reflexive mode, according to Nichols, developed in the 1980s in a context of postmodern skepticism where epistemological doubt was emphasized and problems of representation were addressed. Whereas documentaries in the observational or interactive mode focus on their subjects, the viewer and the text’s reception become central in the reflexive mode: “In its most paradigmatic form the reflexive documentary prompts the viewer to a heightened consciousness of his or her relation to the text and of the text’s problematic relationship to that which it represents” (ibid. 60).

Nichols saw the reflexive mode as a “final stage” and expected it to develop into a modified form of the expository mode. However, he has later proposed aperformativemode as an alternative to the reflexive mode. Nichols defines the performative mode as a mode that “does not draw our attention to the formal qualities or political context of the film directly so much as to deflect our attention from the referential quality of documentary altogether” (Nichols 1994, 93). Performative documentary might include other modes of representation.

Its reference to the historical world, however, is reduced and moves towards subjectivity, as the construction of textual authority becomes less crucial. For Nichols the “excessive use of style” puts films of the performative mode close to avant-garde filmmaking, blurring the boundaries between documentary and fiction film (Nichols 1994; 2001).

The expository and interactive modes of representation, as Nichols acknowledges, mainly serve historical reconstruction or investigation, while the observational mode addresses the present tense of shooting. Each mode, however, presents its own difficulties. Whereas expository documentary may appear authoritarian, observational documentary lacks a broader context. The passiveness of observational film is reversed in the interactive mode, which may seem intrusive and problematic in terms of hierarchy. The reflexive mode draws attention to these problems but may be too abstract in style, while the stylized nature of the performative mode may blur the boundaries with fictional forms.

Nichols’ typology of representational modes may seem problematic in that it aspires both to identify basic formal patterns and to describe historical developments in documentary filmmaking (Decker 1995). The fact that Nichols felt compelled to expand his model with an alternative, fifth mode reflects this weakness. Expanding the model contradicted its formal aspirations, yet the four original modes alone were not enough to describe historical developments. Although acknowledging that most documentaries are mixed forms of representation and that modes can exist in parallel, Nichols’ model suggests a teleological logic towards more complexity, making afinal stagedifficult. Christof Decker questions whether the reflexive mode can be treated as an independent mode since it, according to him, concerns all other modes and functions as a textual strategy rather than a formal pattern. He suggests that reflexive documentaries should be summarized in anintertextualmode (ibid.).

A broader model based on degrees of narrational authority was developed by Carl Plantinga as not all documentaries, in his view, could be easily identified with Nichols’ modes of representation. Plantinga takes up the metaphor of voice to identify theformal, theopen, and thepoeticvoice in documentary. Similar to Nichols’ model, Plantinga’s approach is formal, focusing on the documentary text itself. He describes his model as aheuristic typologyintended to draw attention to the textual means by which documentary films perform their functions (Plantinga 1997, 106).

According to Plantinga, documentaries in theformalvoice have a high degree of narrational authority, capable of delivering information to the viewer about the historical world. For Plantinga, documentary films of the formal voice are similar to classical fiction films because of two major features: 1) they pose a clear question or a relevant and coherent set of questions and 2) they answer every salient question they pose (ibid. 107). The formal voice is closest to Nichols’ expository mode of representation but could also include documentaries of historical reconstruction in the interactive mode.

Documentaries in theopenvoice, according to Plantinga, observe and explore. He describes the degree of authority in open voice documentaries asepistemically hesitant. In contrast to the formal voice, documentaries in the open voice are less likely to pose overarching questions and to offer clear answers. Rather than explaining, the open voice shows and explores, making it less explicitly rhetorical. Plantinga associates the open voice with Direct Cinema, resembling Nichols’ mode of observational representation although the interactive mode, and perhaps even the reflexive mode, might be included in a classification of open voice documentary. Comparing the open voice with fiction film, Plantinga sees similarities with Bordwell’s art cinema mode of narration since both often lack a clear causal structure (ibid. 108).

Finally, documentaries in thepoeticvoice lack narrational authority, according to Plantinga, because of their aesthetic concerns. Poetic voice documentaries rarely exhibit what Plantinga calls a central function of documentary, namely providing information through explanation, observation, or exploration, which films of the formal and the open voice share (ibid. 171).

Nichols’ and Plantinga’s models are both useful tools for approaching documentary in terms of narrational and representational strategies that,together with the image, form part of the basis of what constitutes documentary.


1 Winston notes that scientific devices can, however, be manipulated.

2 However, video evidence is not always unambiguous. Bill Nichols’ analysis of the Rodney King case shows how videotape was used in different ways by the defense and by the prosecution, each side presenting the material in a way that appeared to back up their arguments (Nichols 1994, 17-41).

3 The term reality effect was introduced by Roland Barthes. He sees the descriptions of seemingly unimportant details in a literary text as a strategy to increase the reality effect – filling the text with an increased ‘measure’ of reality and thus authenticating the text as realistic (Barthes 1989, 141-148).

4 The concept of the profilmic event was introduced by Etienne Souriau and is commonly used in film theory. It refers to everything before the camera, as well as the camera itself.

5 For direct address Nichols identifies four basic types: 1. Narrator or orator speaking as a “voice of authority”, 2. “Voice of God”, omniscient source of information, 3. Social actors in interviews talking directly into the camera, 4. “Voice of witness” speaking parallel to images (Nichols 1981, 183).


ISBN (eBook)
ISBN (Book)
File size
1 MB
Catalog Number
Institution / College
Humboldt-University of Berlin – Anglistik und Amerikanistik
Direct Cinema Dokumentarfilm Beobachtende Methode Frederick Wiseman Albert Maysles David Maysles Drew Associates Emile De Antonio Robert Drew Cinema Verite Observational Cinema Documentary



Title: Forms and Functions in Documentary Filmmaking