Printed originally in: Nordicom Review, no. 1, 1994, pp. 65-85. The text has been slightly edited for this version.


“The issue is not whether reality exists, but whether there is only one way to describe it in all cases (Lakoff 1987: 263)


The origin of documentary: document and creativity

The word documentary in connection with visual forms is said to have been used for the first time in 1926 by John Grierson (Corner, John, ed., 1986: vii) in an article on Flaherty’s film Moana. The word was used as a kind of translation of the French word “travelogue”, that is a film documenting a travel, or the ethnographic film, documenting a foreign culture. But as John Corner states: “the project of attempting somehow to “document” real events and circumstances through mechanically recorded images is as old as the technologies themselves.” (Corner, John ed., 1986: vii). In this respect documentary film and television is closely related to the factual forms of journalism (news, current affairs programmes etc.) and the use of film and photo as evidence. In the words of Bill Nichols:

Documentary film, and ethnography in particular, depend heavily upon the indexical nature of the cinematic sign (…) more generally, it is this indexical relation that motivates the use of film footage as courtroom evidence or, with ethnographic film, as cultural evidence” (Nichols, Bill, 1981: 239).


In documentary film and television and in other related genres of a factual nature we expect some kind of documentation with a special and direct reference to reality, and even though factual forms and documentary genres may have heavy emotional impact and may document emotional sides of reality, we expect this to be a fact of the profilmic events and not of a narrative construction of a diegetic universe of the film. In 1948 the World Union of Documentary defined documentary film as:

All methods of recording on celluloid any aspect of reality interpreted either by factual shooting or by sincere and justifiable reconstruction, so as to appeal either to reason or emotion, for the purpose of stimulating the desire for, and the widening of human knowledge and understanding and of truthfully posing problems and their solutions in the spheres of economics, culture and human relations. (quoted from Barsam, R. M., 1973: 1)

The indexical dimension is central here, and the strong referential notion of a more or less direct relation between film and external reality. However, the quotation also talks of interpretation, allows reconstruction, and it is clear that although the words sincere or truthfully has the connotation of objectivity and empirical truth, documentary is also connected with desire and appeal. In a sociological sense the documentary and factual forms have contextual relations to the conventions and codes working in public communication and the creation of public knowledge and public opinion. This is clearly the case for documentary television programs and journalism, but was also a vital part of the strong British tradition in documentary filmmaking from the 1930’s: the aim was to inform and enlighten or even educate the public, but also to create certain attitudes. Basil Wright, a prominent member of this historical movement, in relation to this simply stated that “documentary is not this or that type of film, but simply a method of approach to public information.” (quoted from Barsam, R. M., 1973: 2)

The various historical definitions of documentary film clearly indicate a movement in two directions: the direction of index, document, truth and reality, and creativity, construction, persuasion and propaganda. John Grierson’s famous definition of the documentary genre as “creative treatment of actuality”, seems to combine the two dimensions in one brief form. In his definition of the documentary forms, the concept of story or narrative form and imaginative interpretation is an integrated part of the understanding of documentary as based on raw reality and the real world. Grierson finds documentary film superior to fiction films in the last respect: documentary narratives of reality in his opinion are stronger stories than fictional ones (Grierson in Forsyth Hardy, 1966). American documentarists often make the same combination of narrative and reality. Pare Lorentz talks about documentary as “a factual film which is dramatic” (quoted in Barsam, R. M., 1973), and Willard Van Dyke defines it as a film where “dramatic conflict represents social or political forces rather than individual ones”. Documentary is given epic quality, but at the same time Dyke ties documentary to “real people, real situations – (…) reality” (quoted from Barsam, R. M., 1973: 2).

Some of the characteristics attached to documentary seem to cross the line between what in common sense separates fiction from non-fiction. The American film maker Philip Dunnes defines documentary as a film of “experimental and inventive” nature, which may “even employ actors” and in which “fantasy or fact” may be the stuff, and which “may or may not possess a plot”. Documentary in his opinion is an “idea-weapon (…) an instrument of propaganda” (quoted from Barsam, R. M., 1973: 2). In the light of this, some prefer to make a main-distinction between fiction film and non-fictional films, making the basic stuff on which the film is based and the reference the crucial distinction: non-fictional films are based on reality and fact. Documentary film is then a special form of non-fiction film along with for instance factual films, travel films, educational films and newsreels (see for instance Barsam, R.M., 1973).

In many ways however, the history of definitions of the documentary film seem to indicate the dominance of a much too empirical, simplistic and objectivistic paradigm of how people and films relate to reality. There is too little focus on the central question in focus in this paper: to what extent can the difference between documentary/non-fiction genres and fiction genres be said to rely on differences on a textual level; how can we define the relation on a cognitive level; and to what extent can the difference be said to rest on pragmatic and contextual factors. The pragmatic dimension of documentary is tied to the question of a specific communicative contract of documentary, the different institutional practices, that separates this contract from the fiction contract. The cognitive dimension has to do with the theory of schemas and the cognitive theory of metaphors, the organization of the mental framework, that we use in our actions and interpretations, formed by our body and perceptual capacities, but also acquired through experience. In a cognitive perspective many of our mental activities do not seem to respect any distinction between fiction and non-fiction or between fiction and real life. On the contrary: basic functions of our mental framework are at work in a similar way whether we are confronted with visual cues in fiction or non-fiction film or in real life situations. There is no doubt that the distinction between fiction and non-fiction is essential to our way of relating to the world and to communication, but the question is on what level and basis this distinction is made and is important.

Against objectivism: the cognitive dimension of metaphor and imagination

In his seminal work Frame Analysis (1974), Erwin Goffman is very polemic towards empiricism and the attempt to define reality in the sense of what is real and what is not, or what is true or false. Instead Goffman turns to a more phenomenological and pragmatic tradition in which we “instead of asking what reality is (must) italicize the following question: Under what circumstances do we think things are real? The important thing about reality (…) is our sense of its realness.” (Goffman, Erwin, 1974: 2). Goffman’s point here is not, that the real world in the physical sense of the word is non-existent outside our mind, nor that the distinction between things we regard as real and things we do not accept as real is not important. Goffman’s observation can rather be regarded as an example of a pragmatic and cognitive theory of how we relate to the world and to visual or linguistic constructions and representations of the world. There is no objective reality that we can reach directly, it will always be mediated and formed by the context and mental framework.

One aspect of Goffman’s approach is to dismiss the notion of reality and truth as such and to replace it with what Alfred Schutz in 1945 called “multiple realities.” Instead of reality “an sich” we have reality for somebody or we may even say, that we have different kinds of reality each with their “style of existence” or “cognitive style”. Life is made up of situations and strips of life with a special kind of framing, and it is difficult to point to any primary reality or any clear hierarchy in our sense of reality. However, Goffman’s theory of framing also tells us that in order to interact and communicate we have to have frames, codes, conventions in our interaction with each other, with texts and films and with reality. Otherwise we would not feel any continuity in our daily lives – and we certainly do. Our communication works, we often get intersubjective contracts on a micro or macro level. In our everyday life and communication we have a highly developed ability to shift frames and enter into various institutionalized practices: reading newspapers, reading novels, going to the movies, watching the news, going to work, conversation in the family and so on.

Pragmatic and cognitive approaches reject what you might call “objectivist theories of meaning and rationality” (Johnson, Mark, 1987: xxii). Objectivism is closely related to the representation theory of how to define fiction vs. factual forms and documentary forms. In objectivism and representational theories, the argument goes something like this: “Meaning is an abstract relation between symbolic representations (either words or mental representations) and objective (i.e. mind independent) reality. These symbols get their meaning solely by virtue of their capacity to correspond to things, properties and relations existing objectively in the world” (Johnson, Mark: 1987: xxii). In such a theory fiction is equal to pure imagination, illusion or even lie, and even documentary and factual forms, defined in principle as objective representation of reality, may have problems with the correct reference or mirroring of the mind independent reality. Against the objectivistic and representational theory Johnson points to the fact, that:

Whether it be for human events or for words and sentences (including visual forms, IB) meaning is always meaning for some person or community. Words do not have meaning in themselves; they have meaning only for people who use them to mean something (…) The Meaning of the symbol stems from the imposition upon it of a certain intentionality, which is always a matter of human understanding. Intentionality is the capacity of a mental state or of a representation of some kind (concept, image, word, sentence) to be about, or directed at some dimension or aspect of one experience(…)meaning is always a matter of relatedness (as a form of intentionality). An event becomes meaningful by pointing beyond itself to prior event structures in experience or toward possible future structures.” (Johnson, Mark, 1987: 177)

The quotation may be said to support a semio-pragmatic approach to communication, as it has been proposed by Roger Odin (1983) and Francesco Casetti (1990), and social semiotics (Dines Johansen 1993). This is related to the pragmatic dimension of the communication contract, but it also supports the cognitive dimension and the importance of our dynamic use of schemas in relation to real world events and to communication in either words or moving images.

It is often claimed about the distinction between fiction and factual forms, that the factual sign/symbol is metonymic/indexical and that the fictional sign is metaphorical/symbolic, or it may even on a different level be said that fictional forms work with things related more to feelings and the imaginary, whereas factual forms address the rational citizen. But this is a question of convention and degree, rather than of nature. Mark Johnson’s argument in The Body in the Mind and in Johnson and Lakoff’s book on Metaphors we live by is that metaphorical ways of reasoning and the use of embodied image-schemata play a major role in our whole way of constructing meaning and experiencing the world. In so far as this penetrates the division between science and art, between rational and emotional ways of thinking, and since this means the elevation of imagination to a very important and general aspect in our mental model- and meaning-building, this is also a factor to be considering in dealing with documentary forms and fictional forms.

In Lakoff’s and Johnson’s theory of metaphors, they try to define a third way, experientalism, between objectivism and subjectivism. Metaphoric ways of understanding pervade our whole reality and our conceptual system is grounded in basically three domains of experience: body-experience, interaction with physical environment and other people, and communications in culture. Each of these basic domains form experiential gestalts, coherent organizations of experience, that form the basis of structured, recurrent concepts and metaphors. The theory of experiental gestalts rejects the objectivist theory. According to objectivism, objects, experiences, texts, and other forms of communication are characterized by inherent properties, that may be categorized in taxonomic, logical, structural sets of properties. But for Johnson and Lakoff this is only a minor part of the experience of reality, language and communication. Most of our experience is most likely organized as interactional properties, where metaphorical connections play an important role, and where we form categories by using prototypes, rather than logically defined categories (Lakoff and Johnson, 1980: 119-120).

This kind of approach has important consequences not only for linguistic analysis, which is the main area for Lakoff and Johnson, but also for the general approach to genre and to visual communication. Genre in the form of basic genres like fiction and non-fiction, and subgenres like the different forms of fiction-films or documentary films, should be regarded as prototypes rather than fixed categories, and communication of course basically is an interactional activity, where inherent properties are less important than experienced ones. In other words:

Concepts are defined by prototypes and by types of relations to prototypes. Rather than being rigidly defined, concepts arising from our experience are open-ended. Metaphors and hedges are systematic devices for further defining a concept and for changing its range of applicability.” (Lakoff and Johnson, 1980: 125).

Documentary forms are without doubt in our prototypical understanding of communication related to concepts like truth and knowledge. In his review of Metz’ book L’énonciation impersonelle ou le site du film (1991), Roger Odin defines different modes of reception: the private mode, the fictional mode, the documentary mode, the aesthetic mode, and the artistic mode. He points to the fact that the context of a given film to a large degree determines the mode of reception, that is the framing of the reception is important (Odin, Roger, 1992: 207). Following that he says, that these different modes result in a very different construction of the enunciative structure of the film:

To watch a film in a fictional mode means constructing a form narrative enunciation where you are absorbed and forget the normal reality. To watch a film in the documentary mode implies to construct an enunciation and discourse in which you relate to questions regarding the films relation to a direct reality and to what is true or false

(Odin, 1992: 207, my translation and italics).


There is a good correspondence between this semio-pragmatic point of view and the cognitive, semantic approach to the category of truth. What Odin and the whole pragmatic theory makes clear is that there is no essential or objective way from structures of signs to the referent in reality or to the meaning or truth of a given genre. There are different prototypical ways of constructing genres, that are related to prototypical constructions of reception. Truth and reference are not inherent properties in communication, but aspects of interactions with the world based on our experience. However, going from an objectivistic concept of truth to an experientalist theory of truth does not mean to exclude the importance of truth from our life or our communication, it means defining the cognitive dimensions that determine how we approach reality and questions of truth. We base our daily life on what we consider to be true and certain, we act, interact and communicate in god faith and in accordance to principles of sensemaking and cooperation. But the basis of this truth and trust is so natural to us, that it is difficult to be aware of it. However, basically we have different forms of truth, because truth is dependent on categorisation. According to Lakoff and Johnson, this dependency can be defined in at least four ways:

  • truth in any communicated sense is only true relative to some understanding of it
  • understanding involves human categorization of an interactional and functional way based on experiences
  • truth in any communicated sense is relative to the highlighted dimensions, a shift in focus may change the truth or meaning communicated
  • categories used to define or frame communication are neither fixed nor uniform, they are based on prototypical resemblance, and prototypes are adjustable to both contexts and new aspects and experiences (Lakoff and Johnson, 1980: 165-66)

One of the basic aspects of visual communication, compared to linguistic forms of communication, is that both fictional forms and documentary forms have iconic features, they resemble our perception of daily reality. This means that in film and television, more than in other forms of communication, the mental processes in constructing and decoding a film to a large degree make use of every day perception and mental image schemas. One of the main points in the cognitive-pragmatic and experientalist theory is that the traditional concept of correspondence and realism is problematic. Resemblance and correspondence are of course relevant aspects in the sense that communication is about something which we presume exists or corresponds to something in our known world.

Communication must make sense to our experience of reality in the broadest sense – whether we talk of fiction or non-fiction. But this meaning and correspondence cannot be defined in any objective or absolute sense: it is relative, specified and partial and dependent on our framing, understanding and experience. The same goes for our sense of realism. Realism can only in a superficial way be defined as a match between signs and reality, because it is always mediated through culture and through our built in or acquired mental forms. Johnson and Lakoff’s attack on objectivism and the classical realism is in fact an attack on the deep-rooted western dualism between rationality, emotion, and imagination, a dualism often repeated in the definition of fiction vs. non-fiction. However, if metaphorical, metonymic and image-based concepts are basic to our whole way of reasoning and communicating, then this dualism is false.

Lakoff and Johnson’s definition of metaphor as “imaginative rationality” is in itself a metaphor for a way of defining fiction vs. non-fiction. It rejects the traditional dualistic approach, for instance in psycho-semiotic theories of the imaginary signifier, or the classical Freudian definition of dreams and art. Perhaps it is no coincidence then, that documentary forms have been defined – by those making documentaries – as “creative treatment of actuality”, signalling this metaphorical relation to documented reality. The analysis of how documentary genres work is one way of focusing on communicative and cognitive dimensions that transcends the traditional dualism.

“Night Mail” – information and metaphor

Basil Wrights classical documentary Night Mail (1936) is an excellent example of a documentary, where a creative treatment of actuality takes place, and where documentation of a piece of reality is embedded in a metaphorical and lyrical use of image, montage and sound (both music and words). The opening sequences of the film give us a number of pragmatic cues indicating a reading-context and genre-frame of documentary nature. The sender is identified as GPO (General Post Office) and instead of a list of characters we are simply told, that this film is made by Basil Wright with “Workers of the travelling Post Office” and “Workers of the Railway”. The understanding of these institutional signs depends on both our pragmatic, contextual knowledge of the world and our textual schemas.

The film – which lasts 25 minutes – cannot in any way be said to follow a canonical, narrative structure, but on the other hand it cannot be said to follow the classical form of an argument either. It is not an instructional film or straightforward informational film. The macrostructure of the text is the travel, and the identification between the film and its spectator in a cognitive sense rest both on our ability to use textual, documentary schemas and our ability to apply our schematic knowledge of travelling by train and the associations attached to that in the decoding of the film. At the same time however, the spectator will probably also try to establish assumptions about the intentions of the sender to make the film: what kind of message is imbedded? Included in the macro-frame of the film we also find reference to schematic knowledge of typical work-routines and the form of communication in modern life. But on the more local levels of the film text we can identify at least three prototypic structures with distinct visual style and meaning, that may activate different forms of mental schemas.

First of all, we have scenes and situations where workers are filmed on actual locations performing their daily work routines. The scenes have bits of dialogue and natural reality sound. Given the framing of the film as non-fiction, we easily identify the scenes as filmed reality or reconstructed reality, but there are no textual indication allowing us to judge the fictional or non-fictional character of the scene. They might just as well be scenes from a narrative, English fiction film, in fact they have the nature of small pieces of a narrative, diegetic universe. Cognitive processes in relation to these parts of the film may thus in part resemble processes characterizing fiction.

Secondly, we have prototypical scenes with the same kind of structure as the first, but with a more direct informational and factual intention. Here a male voice-over explains the visual scenes as part of a social process of cooperation and communication. We are informed about time and place and we are told about the functions and processes attached to “the travelling post office”. In a strict sense this can be identified with the structure of an argument, where the words explain and the pictures document. This kind of direct address in various forms – the anchor person in news or the journalist speaking directly to us from the field – is prototypical for factual presentations. The mere existence of a voice-over narrator however is not in itself a sign of non-fiction, as Sarah Kozloff has showed in her study of “voice over narration in American fiction film” (Kozloff, Sarah, 1988). But the linguistic nature of the voice over and the direct relation between images and voice over will normally – given the right framing – make us assume, that this is a documentary presentation.

Finally, we have extended prototypical examples of metaphorical and lyrical montage. The filmed elements of reality are still the same, they have the indexical nature of documented reality. But from the start of the film we experience small strips of another kind of rhythmic voice-over spoken in verse (by the English poet W.H. Auden). The spoken verse is underlined by music in adapted rhythm and with changing images of the train, the workers, the nature, the cities, the industries and the telephone wires cut in the same rhythm. This third layer of textuality is increasing in importance and extension during the film, reaching a sort of crescendo approximately 20 minutes into the film. In this way it is given the status of conclusion. The spectator is therefore confronted with a gradual movement away from a basically direct and commented form of a constructed representation of reality to a metaphorical form, a movement from a seemingly direct and determinate reference to a more indirect and indeterminate.

The translation of the metaphor is dependent on the spectator’s stock of metaphorical and schematic knowledge of the world. The text helps and builds up cues, that finally interact in a message concerning the importance of communication, human cooperation and technology in modern society. The train, the mail, the telephone wires interact with the images of nature, city and industry. But without schematic basis in both extra-textual experience and inter- textual experience, this textual construction of meaning does not take place. The example demonstrates on the one hand the pragmatic and cognitive dimensions of interpretation of meaning and reality status, and on the other hand the fact that metaphoric dimensions are clearly not only at work in fiction – they are a vital part of our whole understanding of reality.

Reference in fiction and non-fiction

It follows from this then, that the difference between a fiction film and a documentary film is not whether it is a true representation of reality or not, or whether it is about our external reality as such or not. In almost any visual communication there is some kind of resemblance with our known everyday reality, and fiction certainly gets its fascination from the fact, that it is about our reality, though in a special way. The American film theorist Edward Branigan therefore claims that sometimes you may read a given film in a meaningful way both as fiction and non-fiction. Both responses may be considered real in as far as they make us relate what we see to our daily life, our experience of the world.

Normally we will not perform fictional readings of documentary films, but under certain contextual circumstances we may switch between modes. As Odin (Odin, Roger, 1983 and Stam, Robert, et al., 1992: 214) has argued, fictionalization is a semio-pragmatic process with a number of interacting operations: figurativisation, diegeticisation, narrativisation, demonstration, belief, mise-en-phase and fictivisation. But this fictionalisation process may be blocked because we come to the film without sufficient context and knowledge, or because we may zap into a programme, that is already half finished. Non-fictionalization is only in principle the opposite of fictionalization, but as Odin points out more likely a sort of blocking of some of the usual processes of fictionalization – seldom all of them. In any case – apart from the semio-pragmatic aspect – the cognitive dimension tells us, that the mental process of decoding fiction and non-fiction pictures may have much in common on a certain basic level – for instance character-reading and identification.

Metaphorical mechanisms thus have pervasive influence on our mental and communicative practices, but narrative structures are of equally basic importance to all areas of communication and understanding. Not only the semiotic tradition but also the cognitive tradition tells us, that narrative structures and narrative comprehension mechanisms play an important role in natural language, in fiction, in news, in documentaries and in the reading of real-life situations. We construct stories all the time, and we look for narrative clues in order to be able to make sense of things.

To look for the distinctive difference between fiction and non-fiction in specific textual features related to either narrative structure vs. non-narrative structure, in the use of narrator or in stylistic features, therefore will lead nowhere. Obviously, we can define textual elements that will usually be more dominant in non-fiction than in fiction, for instance the use of interview or a narrator with direct address to the spectator. But it will always be possible to find examples to contradict the general tendency. As we have seen the question of truth and reference also create difficulties, when used in too abstract and objectivist sense. In conclusion Branigan says:

Thus, neither truth-claims nor rhetoric can be taken as features that distinguish between fiction and non-fiction. Rather my argument will be, that the method or procedure for making decisions about assigning reference is different in each case even if the results are the same (i.e., knowledge about some condition in the world)” (Branigan, 1992: 193).

Branigan then defines the difference not in essence and aesthetics, but in the context and mental procedures determining the way the reference to a meaningful reality is made:

Fictional terms denote real things, though not determinate ones (…) A fiction does not determine exactly which object it represents, and this openness is what distinguishes fictional reference from other sorts of reference (…) to interpret a symbol fictionally is to operate in a precarious, intermediate zone between sets of possible references (open functions) and a specific reference (…) Considered as a cognitive activity, fiction is a complex way of comprehending the world in which one is first required to hold open sets of variables while searching for a reasonable fit between language and lived experience, between sets of symbols and acts of the body (…) fiction is a partially determined reference which is initially neither true nor false, its usefulness must be found and determined” (Branigan, 1992: 194 -196).

In fiction we have an open kind of reference, which has to be decided and worked out, and in fiction the process of relating the visual events to the profilmic events is not the normal procedure. Rather the viewer tends to construct a post-filmic reality in which structures of the film are discovered and translated. Branigan – just like Lakoff and Johnson and Odin – defines reference, not as a product of objective processes and features of the text in itself, but as a product of human design and use, “based on the rules, habits and conventions of a community of individuals.“ (Branigan, 1992: 197). In all forms of reception there is therefore more than one translation to a meaningful world and more than one kind of reference, but perhaps this is more the case for fiction than for non-fiction. However, there are not endlessly many translations and receptions, since basic mental processes and mental schemas tend to be in function also. Variations occur, but on the basis of often common and shared views and perceptions, otherwise communication and understanding would be impossible.

In the typical documentary film on the other hand, we assume that there is a more direct and determined and causal relation between the film pictures and the profilmic event. More important, although a documentary film is also constructed and not a copy of reality, the mental activities in the reception of a documentary film may depend more easily on what Branigan calls the “social conventions and categories of causality in a community.” (Branigan, 1992: 204). There seems to be a more direct way of reference in documentary and other non-fiction forms, because documentary films address us, as already pointed out, as members of a more defined community, and usually with a more precisely defined public theme. As a consequence of this difference, the degree of openness and indeterminable character built into a fictional text might be said to demand a more dynamic use of prior knowledge structures and mental frames and schemas, it puts a delay and expansion on the degree of instant framing and schema-use (Branigan, 1992: 195).

The indexical and iconic nature of documentary film makes it more common to suppose, that spectators will try to make direct assumptions about the profilmic reality and its relation not only to the filmed version, but also to the public and private reality of daily life. In fact, one of the main points of documentary fascination, besides the desire to know, probably is the authenticity-effect. When confronted with strong scenes in fiction film, war and murder for instance, we probably use the same kind of mental and emotional procedures in our reading, as in documentary film. But in the case of fiction we add a frame of as-if, whereas in documentary we establish another kind of real-link reference.

Schema-theory and documentary

In cognitive schema-theory the concept of “default assumptions” play an important role. In his book on cognitive psychology The Society of Mind (Minsky, 1985), Marvin Minsky deals with the concept of frame. His definition of frames has much in common with the concept of prototype and Goffman’s concept of frame: “A frame is a sort of skeleton, somewhat like an application form with many blanks or slots to be filled. We’ll call these blanks its terminals; we use them as connection points to which we can attach other kinds of information.” (Minsky, Marvin, 1985: 245). A frame then connects our general knowledge and experience with particular instances of representations, objects or situations. When we see a particular sequence or scene in a movie or enter a room in real life, we instantly recognize and interpret the whole set of visual cues and objects: we make sense of or recognize what we see and translate it to familiar terms. This process in cognitive terms is based on the fact that “each perceptual experience activates (…) frames – structures we’ve acquired in the course of previous experience. We all remember millions of frames each representing some stereotyped situation.” (Minsky, 1985: 244).

Given the impact and quantity of impulses we get, in reality and in watching tv and film, a lot of our perception of the world has to be done on a more or less non-conscious way based on firmly established schemas or scripts. We use the phrase to watch film and television, but in fact visual communication is only one aspect of many: visual cues can be found in the system of montage and the composition of the single shot, and audio signals are received as both sounds, music and linguistic information, and besides that interpersonal and interactional information is very important together with the reading of body language and facial impressions. Default assumptions then is what helps us quickly to process these multiple forms of information. They fill out what is missing to form a typical representation or meaning and they help us select and focus on what is important and what is less important. Through default assignments our mental frames are constantly set to work when we interact with the world and with communication. Frames and default assumptions allow us to reason, generalize and to predict, foresee or deduct what may happen or has happened. Since frames are fuzzy categories dominated by default assumptions, we often have to adjust them or reorganize them, when we are confronted with new experiences or new forms of communication (Minsky, 1985: 245 and 246).

Minsky’s concept of frame resembles the concept of schema very much. This concept has been used especially by Schank and Mandler (Mandler, 1984 and Schank, 1990) in the analysis of how narrative structures works in our memory. Mandler works with a basic story schema and with event schemas and scene schemas. Experiments show that people have mental and cognitive structures that are made up by a number of basic experiences with stories and how stories are structured. Very early in life we are able to recognize a basic story structure and to use it in both production and reception. The same goes for a number of events and scenes organized as thousands of experiences with specific and generalized events and scenes. We carry a catalogue of event- and scene-schemas around in our minds, schemas that allow us quickly to respond to and understand situations in daily life. We interpret the world and communication on the basis of frames and schemas and we tend to choose those that seem to match most readily with what we see. Our schemas and frames may vary according to our experience and daily routines and our prior knowledge of communication forms, so that basic functions may differ from person to person. But it is also clear, that within a given community or a wider culture, story, event- and scene-schemas are shared structures.

We use schemas and frames on a default basis and on the basis of almost automatic matching. It seems that if our first match does not make sense we start again and make use of more expanded networks of assumptions. Fiction (and probably even more complicated forms of documentary genres, that mix fictional and factual forms) activate large networks of assumptions, and tend in some cases to challenge our default assumptions and deeply imbedded schemas and frames. The concept of genre can be defined as a special combination of expected story structure and scene- and event-schemas and a number of other frames or schemas from experience with communication and real life. The viewing of a given film may be characterized as a constant negotiation between frames and schemas and the cues from the text. Negotiation takes place both on the macrolevel of the text and on a more local microlevel. There are indications in cognitive research, that the establishing of a contract on a macro-level between the program and the viewer is strong in the beginning. Mandler (Mandler, 1984: 55 f) refers to the fact that a number of studies on the pace of reading at the beginning of different kinds of stories show, that people read very slowly at the beginning of a text and then speed up. This may indicate, that it takes some time to establish the main-contract and to get frames and text to work together. But once the main- direction of the story is established, we read faster because we can use our adjusted default assumptions more unconsciously.

Reception of fiction and non-fiction

There has been some empirical testing done on the use of schemas in the reception of both fiction and non-fiction. The non-fiction tested is both television news and documentaries. It is a well stated fact, that when asked about their interest in watching factual tv- programs and films, people will refer to cognitive benefits in a more rational sense of this term: the need for knowledge, for being updated on public matters and public conversation and for information making it possible to act in and evaluate social, political and cultural life (Gunter, 1991). However, when people are asked to retell news-stories or to answer simple questions about information given in news, they often fail or make mistakes. The results show variations depending on peoples acquired knowledge from other sources and their general schemas, and there are also indications that the use of for instance narrative schemas and strong cooperation between visual and linguistic information may improve reception-output. The empirical testing may indicate, that other schemata, than simply knowledge-schemata are at work in even the most factual form of all, television news.

One researcher (Bruhn Jensen, 1987) has formulated the hypothesis, that people often tend to respond to particular news stories through different super- themes, a form of culturally established event- or story-schemas. Along the same line, empirical testing through interviewing (Höijer, 1992 and 1992a) shows, that people use schemata of the same sort when responding to fiction and non- fiction. Höijer talks about our cognitive structures as organised in a complex network related to our experiences: universal experiences (basic to all humans), cultural experiences (specific for members of a more or less defined community) and private experiences (unique for each individual). In one study (Höijer, 1992), she tested the reception of a tv-science magazine program on AIDS and she found, that viewers related to the program with schemata often related to the psychologically and concrete experience on a both universal and personal level.

The identification with characters in the program is strong, and the retelling of the program is detailed and filled with associative processes between program and viewer. The more general, informational aspects of the program on the other hand was not captured in a very clear manner. Despite the factual character of the program, it was clearly a more associative and metaphorical approach that dominated. However, variations in the use of schemas could also be found in the use of schemas, that showed cultural variations, for instance related to professional experiences or very private experiences. But the main result is the strong dominance of universal schemas in the emotional relation to the program.

Another more complicated study (Höijer, 1992a) tests the use of schemas in relation to both serial-fiction and tv-news, and is supplemented by a study (Höijer, 1992b), that also includes documentaries. Höijer defines a number of so-called social schemas among which we find not only story schema, event schema and scene schema, but also person schema, role schema and self-schema, the last three relating to how we interpret and process information about characters in real life and media. She then defines different experience spheres such as the private sphere, the occupational sphere and the media sphere, in which we can define both more general schemas and domain-related schemas. In her conclusion

she says:

The viewers had a tendency to use a mix of cognitive schemas from various experience spheres, and this tendency is valid across genres. At the same time, there were differences between different viewers as well as between different genres, in how predominant the schemas originating from different experience spheres were (…) In order to make sense of a television program, the viewer must find connections between the text and her or his own inner world. One characteristic of the inner world is that it is cognitively organized in fuzzy schemas representing generic social experiences, cultural knowledge and specific personal experiences (…) Which mix of schemas is activated depends partly on central conceptions in the viewers thinking, partly on the text. (Höijer, 1992a: 294 and 299-300).

One of her findings were, that a national tv-serial of a “realistic” nature activated more personal schemas and opened up for referential interpretations, whereas an American soap seemed to activate more general and intertextual schemas and less referential interpretation. Interestingly enough, her findings at the same time indicate, that documentaries activated very emotional and personal patterns of interpretation and schemas, and it was very clear, that people related to the programs as to real life persons and situations, there was no distancing or reference to textual schemas. In fact, realistic national fiction and documentaries seemed to activate a number of equivalent schemas, a fact that underlines one of the fundamentals in the cognitive theory of metaphor, namely that it crosses the line of fiction and non-fiction.

This is also underlined in an empirical study on childrens way of reading either in a fictional or a factual mode (Steffensen, Bo, 1991). The children were given three fictional texts (on a scale from realism to fantasy) and were asked to identify them either as non-fiction or fiction and to give arguments for their choice. The children frequently mistook the most realistic of the fictional stories for factual, and they typically used arguments like:

  • the text gives direct information and enlightenment about the world
  • the text can be used for guidance in how to behave or what to think
  • the text has direct reference to reality and a high degree of truth-value
  • the interaction between text and reader is based on identification with the reality described and its alikeness with the readers reality.

Reality, imagination and identification: cinema verité narration

A number of the problems related to the more philosophical definition of fiction vs. non-fiction lie, as already pointed out in the tendency to create a principal distinction between reality/fact/ logic and fiction/illusion/lie. As pointed out in the cognitive tradition (see Johnson 1987 and Grodal, 1993: 26 f) this distinction is problematic since processes like imagining, playing, simulation and metaphoric structuring are all a vital part of our way of relating to and thinking about the real world and our normal, daily, mental processes and our reasoning.

As Torben Grodal has pointed out all “higher animals are able to perform in both an ‘actual’ and a ‘hypothetical-playful mode’” (Grodal, 1993: 27), and this seems to indicate, that the fictional mode is part of reality and also, that in relating to fiction we use schemas and other mental processes, which are also found in our relation to real life situations. When we use our memory or when we make plans and imagine things we want to do or are in fact going to do, or when the scientist is trying to work out a new theory, in all these cases we simulate and make pretend, we create ‘fictions’.

It is of course important for us to realize when we are in the actual mode, and when we are in the hypothetical mode. In the same manner it is important to know the reality- status of a given visual product, the mode in which it addresses us. However, this does not mean that there is a huge difference in our way of relating to the world of fictional genres and the world of non-fiction genres. The empirical testing has indicated a number of similar schemas and a number of similar ways of emotional identification and referential processes. To say that fiction appeals to imagination and emotion through a direct identification with the story and the characters, and that non-fiction mainly addresses the rational side of reality, truth and arguments is only partly true. In fact, documentaries may possess strong emotional identification possibilities, and strong narrative structure, and the reality-status may even increase the emotional impact.

This is for instance often the case in documentaries of the ‘cinema verité-type’, where the use of experts and direct address is minimalized and characters perform in their actual environment or address us more indirectly. In his analysis of Frederick Wisemans documentaries, Bill Nichols (1981: 208 f) stresses the mosaic, situational, narrative structure of Wisemans films and their metaphorical and associative nature. Though he basically defines documentaries as an expositional form, where arguments and rhetoric dominate, he defines the documentary nature of Wisemans films as a sort of fiction-like mosaic of strips of life characterized by a diegetic unit where spatial and temporal unity prevails. The profilmic reality is framed as a series of institutionally and socially coded local narratives from which the viewer will have to construct the overall meaning and message, without help from the authoritative voice of the narrator.

This tendency is strongly represented in modern tv-documentaries. On Danish television Lars Engels recently finished a series of 5 documentaries on life at Vesterbro, one of the old inner-city milieus of Copenhagen, dominated by working class people, poverty, drugs, violence and prostitution. All the programs begin with a birds eye view over the city by night. The sound of the city is dramatically raised to an unnatural degree: we hear at a close distance but see things from a distant and elevated perspective. Then after some time the title of the program and the credits are announced and the camera is slowly lowered down into the reality, which is to be portrayed. There is no voice over or explanatory narration, but simply the observation of situations, persons and scenes, sometimes dramatic, sometimes just documenting routines of daily life. The camera is a silent witness, an ethnographic eye on the local subculture, and the director only visible through the editing and selection, and in glimpses also as an interviewer. What we get is an impressionistic mosaic of voices, characters, scenes, situations, a fragmented narrative of reality. The film may resemble a social realistic fiction film, and certainly we relate to it much the same way we would if it was a fiction film reconstructing actual life or a real-life observation of that same reality. But given the context we do of course use the frame documentary for it, a documentary with strong elements of emotional identification and use of imagination to fill out the blanks in the glimpses of a story and a personal life we see.

In many ways this example, as well as Night Mail shows, that the documentary genre has a special position inside the general frame of non-fiction. Probably the common sense notion of non-fiction or the prototypical concept of a factual, visual program is the live-reportage and the news-program. None of these programs however can be said to mirror reality or present objective reality and truth. In a common sense perspective people might respond to programs of that genre as telling the truth or reporting reality. But in fact, this is a product of cultural, institutional and contextual processes. The national news represents the mental model of trustworthiness, authority and knowledge, and in the case of live- reportage our direct mental model will be one, that creates the feeling of direct and un-mediated interaction with the reality communicated. Most of us know, that live-transmissions represent a perspective on reality that is framed and often beyond normal perception (in football you see the same action from different angles and repeated in slow motion), and that the presentation of news is a selection according to a specific news-code. But given the culturally defined “spaces of communication” in Odin’s sense (Odin, 1983) we have what we may define as a graded categorization (Lakoff, 1987: 287) of what is factual, defined by our institutionalized communicative genres and acts. And in the case of mass communication genres, news, live-reportage, factual programs and to a certain degree documentaries come close to our understanding of the prototype of factual information to be found also in classroom interaction, courtroom interaction and the like.

Our understanding of genres: schematics of categorization

We can visualize the grading of visual genres on a line reaching from fiction in one end to non-fiction in the other end, or we may visualize the relation as a radial categorization, where all genres can be seen as some kind of overlapping network structure based on the principle of centre – periphery (Lakoff, George, 1987: 287). It is much more difficult to see visual genres categorized in logical hierarchies, since we have basically defined genres as fuzzy prototypes. Documentary genres generally are placed somewhere round the centre of the radial line. They are allowed more rhetorical freedom of expression and “creative treatment of” reality/actuality that those genres at the factual point of the line. Documentary genres have often challenged and moved the frames of normal factual presentation and journalism.

This has also lately been the case with the television-documentary, where a new kind of norm for public communication has changed the line between the public sphere and “front stage” and the access to and importance of “back stage” information (Meyrowitz, 1985). In this kind of modern television documentary, fiction modes and documentary modes are mixed together with basic generic prototypes such as for instance crime, melodrama and romance. In 1990 the Danish tv-documentary director Ulrik Holmstrup was given the TV-price for the program De voksne børn (“The grown-up children”). It is a 55 min. long documentary dealing with the social and psychological problems arising from family life with an often single, unemployed and alcoholic mother, where children are forced to take on the role of parent. The program will no doubt and without problems be decoded and categorized as a documentary by the average Danish viewer. But if we study the operations, particularly in the beginning of the program, we can identify very different processes of this documentization (as opposed to the above-mentioned Odin-notion of fictionalization) and more or less central aspects of the documentary contract. These elements are differently situated on the graded line from fiction to non-fiction.

First of all we have presentation and signature representing what you might call pragmatic and contextual cues. They are part of the borderline-ritual that is an important part of all communicative genres and have been defined in speech act theory as signs of pragmatic macrostructures (Dijk, 1980: 175 f). Is this case the presentation defines the topic and social problem addressed in the program and in fact the genre. The signature identifies both the institution and the genre in typed letters on a blue background, accompanied by the sound of a typewriter and a single musical tone rising in loudness. The signature is metaphoric and combines the sign of the working journalist and the authority of writing/logos/the word with the sign of the institution and its image of objectivity.

Then we have interview-sequences, that is sequences where we see a speaking person in the picture, either an “expert-witness” (in the program for instance a psychologist and a social worker), a “victim-witness” ( the grown up children or their parents) and the “milieu-witness” (persons related to or with knowledge about the cases and the victims). It is a characteristic feature of the program, that the interview-questions are left out, so that the interviews almost all the time appear as statements or personal life-stories narrated directly to the camera and the viewer – seldom though with direct, frontal eye-contact. The use of expert- witnesses is also very limited, and in the very first sequence of the program, an expert-witness, the psychologist briefly identifies the problem, seen from the systems point of view, but she is not identified (name, title on the screen) as such.

So, although the first sequences of the program signal fact and documentary, the discourse is not arranged as a clear hierarchy with a factual voice-over on top. The tendency is to let reality speak out and narrate itself. Another type of sequences is prelude– and interlude– or concluding sequences. It is a series of very complex statements in words, images and music functioning as a kind of metaphoric scripts for the overall meaning of the program. They structure the thematic universe and create an imaginative, lyrical and emotional background for the different stories narrated and showed in the program. Scripts in the cognitive theory (Mandler, 1984) are mental models of prototypical actions and events, and in this case the scripts that are evoked have to do with expectations of family life and the schemata related to roles and processes involving children, parents and growing up. Our normal schemas are challenged and reversed in this program, and a lot of our automatic default assumptions are denied normal functioning. The program has an impact on both universal and personal schemas that will elicit strong emotional identification and memories.

The metaphoric sequences are used for the first time right after the sequence with the psychologist. We see a peaceful, idyllic landscape with almost unnaturally green pastures, yellow flowers, trees, a blue sky and a landscape with one road. The pictures are underlined by classical, soft guitar- music. After a short while a female voice-over, identified on the screen as “Dorte, 23 years” starts telling the story of her former life as a grown-up child of only 8 years of age. As she speaks a little girl trying to master a much to big lady-bike comes into the picture, following the road she continues with great difficulty towards the top of a little hill. Dorte concludes her story with a remark about how the authorities came into her life, and as she says “that was the first time anyone ever bothered to listen to my story”, the little girl on the bike reaches the top of the hill, and where she disappears the title of the program rises like the sun.

The rhetorical richness and the potential frames and schemas evoked in this little piece of visual narrative is as complex as a piece of fiction, although the macro-frame is undoubtedly documentary. The relation between words, music and visual metaphors can be defined both as a typical, factual relation (words determine visuals) or a typical fictional (a diegetic universe of an action to be translated to real life). And these kinds of sequences are used through the program as interludes, whenever there is a shift in discourse. The sequences may vary slightly in style and content, but in the concluding metaphorical sequence there is a more fundamental change. Here the girl of the much to big bicycle returns, and the prelude is repeated, with one big change: at the end the girl throws away the bicycle and continues on her own two feet. A metaphorical conclusion containing the morale and conclusion of the program.

The main part of the program however is based on two case-stories, the story of the girl Moni and the story of the boy Dennis. Both stories follow a kind of narrative and dramatic structure, and they are told with a form and content pretty close to a social melodrama or a soap. The actual rhetorical and aesthetic form of the two stories vary. If we take Moni’s story as example, then the first sequence of this sub-narrative starts right after the metaphoric prelude described above. First, we start with an extreme close up at the eye of a child (Moni) lying awake in her bed at night. In the background we hear the noises of the big city, we return to the girl, then to the city at night, cars driving, people running, police sirens and flashing lights, a squeaking and swinging signpost in front of a night café. Then we return to the bed with the child, we see another child next to her, we return to Moni’s face and then Moni’s voice starts as an off-screen narrator: “Sometimes she doesn’t come home at night” – referring to the mother. Then we cut to a new scene, Moni, her sister and her mother sitting in their living room, watching tv together. Then finally we hear a journalistic voice-over narrator, telling us about the situation and the persons on the screen. After that we return to Moni’s voice-over narration of her own story, visualized in typical everyday situations. At a certain point the off-screen narration is suddenly changed and we see her as speaking person in the picture.

If a spectator zapped into the program from the beginning of this case-story, he/she would probably have difficulties identifying part of the sequences as either fiction or documentary. However, the specific use of off-screen narrator, combined with journalistic voice-over would work as signs of the documentary frame. But to a large degree this special form of cinema verité narration uses “internal focalization” normally expected of a fictional discourse. The film makes use of identification, metaphor and image schemas in order to create a richness of both internal and external reference to reality, private memories, universal categories and public debate.

Crime-fiction and investigative journalism: Errol Morris’ documentary “The Thin Blue Line.”

In Bill Nichols analysis of film genre and film narration (Nichols, Bill, 1981), he makes a distinction between narrative, exposition and poetics, a division he roughly makes equivalent with fiction, documentary and experimental film. Documentary in Nichol’s sense is tied to non-narrative, to arguments and rhetorical textual strategies. The film addresses us directly and indirectly, trying to persuade or convince us. Relating to the more textual and enunciative level this may lead to the conclusion, that the classical documentary normally presents itself through what Branigan calls “non-focalized or externally focalized narration” in order to stress the public or intersubjective aspects of meaning and reference (205-206). Normally then, says Branigan, internal focalization through a character is not common, and this in turn makes the use of dream sequences, subjective flashbacks, point of view shots rare in documentary. In fact, Branigan sees the process of constructing meaning and making reference in fiction or non-fiction as a difference also of levels in relation to the narrated world and the assigning of authority.

In a way non-fiction places the authority of interpretation at a very high level: the non-fiction mode starts with the assumption, that this is the registration of a specific part of real life, and from there the spectator works his way into the text. In the fiction mode, we are absorbed at the lower level of the diegetic universe and from there we try to construct meaning and story on a higher level. Documentary forms tend to limit the “range of interpretation” through minimizing the amount of diegetic narration or at least to motivate it in a very specific way, whereas fictional forms tend to expand the range of interpretation through extended use of diegetic narration and more different forms of focalization. (Branigan, 1992: 204-205). Again, this is not a difference in nature, but a difference in convention and degree.

In Errol Morris’ documentary film The Thin Blue Line (1988) – dealing with a man probably innocently accused of police murder – the clear line between documentary mode and fictional mode is in fact very thin. Although Morris’ film, just as Holmstrup’s Danish film, clearly guides the spectator into a documentary space of communication, they also both rely heavily on our ability to use basic story schemas and genre-frames most often used in fiction. In Errol Morris’ case it is the combination of investigative journalism and the crime series or police-story genre format that form the overall macrostructure of the text. The total structure of the text can be seen as a very repetitive, arguing story, where different versions and interpretations of a case are confronted, and where the spectator symbolically is given a place as a member of a jury. But the film also has a very narrative and dramatic structure, partly divided in chapters, following events or leaping back and forth in time. One element is also the use of old crime-film- footage of a fictional nature in the explanation of the psychology of some of the witnesses. So, the spectator is left with a lot of narrative clues and has to work hard to construct a coherent story.

In his book Representing Reality (1991) Bill Nichols distinguishes between four basic modes of documentary: the expository mode (the classical documentary with direct address and authoritative message and comment), the observational mode (where the authoritative voice is removed in favor of the mere representation of a piece of reality), the interactive mode (where the maker of the film directly in production and on the film interacts with his object and the characters) and the reflexive mode (where the film and its status as a documentary somehow is involved as a meta-dimension). Often these modes are mixed in concrete films, and Morris’ film is at least a mixture of the observational mode and the reflexive.

The authoritative voice of the documentary producer, or the hierarchy of interviewed voices and “experts” is abandoned, and in fact the film seems to deal with the fragile concept of fact, truth and reality in itself. In one comment on the film Nichols characterizes the stylistic use of oblique framings and angles in the film, the use of extreme close up on persons and objects and the decontextualization and fragmentation of objects and scenes in the film as a clear indication of reflexivity and undermining of the indexical authenticity of the film as argument and evidence (Nichols, 1991: 270 n. 18).

Again, if we look at the first part of the film, the establishing of the contract between film and spectator, the documentation-process is complex. The sender is identified on the screen as an “American Playhouse Presentation” and An Errol Morris film, and the title is presented through white letters on a black background, where the word blue is red and with a horizontal blue line dividing the letters in two. At the end of the film the title is explained as a quotation from one of the police witnesses in court: “The thin blue line of police, that separate the public from anarchy.” The following credits indicate documentary, no characters and players are identified, only the production team. However, this textual presentation, underlined by Philip Glass original, disturbing, circular music also clearly has metaphoric elements, and the play with colours and the title-quotation raises the question, what blue line we are talking about – not police/anarchy perhaps, but rather reality/fiction, truth/fabrication.

The rest of the eight-minute long establishing sequence is divided into the following prototypical forms

  • purely audiovisual interludes ( for instance pictures of cityscapes and buildings, flashing police light, city maps and birds eye view of landscapes, signposts etc.) that have both indexical character in so far as they are related to the case, but also function as symbolic signs of change in discourse or tone or metaphoric indicators of the programs whole questioning of truth and justice in American society. The city where the crime takes place is Dallas, and the Dallas-pictures used are clearly intertextually related to the soap by that name, just as the Kennedy- shooting is mentioned several times.
  • voice over narration by the two men involved as suspects in the case or the police officers involved, with pictures illustrating parts of the story or documenting details of events and scenes.
  • person in picture narration with the same persons looking almost directly into the camera.
  • reconstructions /e-enactments of the supposed events and the crime, either with voice over narration by the to suspects, or the police officers, with different possible versions confronted in reconstructed scenes or just in words, or with only pictorial narration of for instance the shooting of the police officer, where the actual body of the dead, photos and newspaper articles form the basis of the reconstruction.

The re-enactments are repeated again and again in different forms and versions, and both the visual cues and narrative scenes, the statements represented, the supposed pieces of evidence and quotations from press and courtroom- transcripts, all this makes it very likely, that the cognitive activity of the spectator takes the form of a constant testing and re-evaluation of activated schemas and default assumptions. This particular example of a documentary mode does not try to limit the range of possible interpretations, the indexical quality of the images is doubtful and the relation between signs and referent is very indeterminate.

Besides that, we actually see reality presented in several internally focalized forms, there is no last instance of control and no easy way to intersubjective coherence on the level of either narrative or argument. Not until the final sequence of the film – where the second suspect, who is now condemned to death for another murder, and who may have done the police killing, for which the other is serving a life time sentence, seem to admit, that he and the police framed an innocent man – the spectator is given fulfilment for the desire to know. Thus, narrative desire and fictional construction-processes combine with the documentary desire to know and documentary construction – a detective story in documentary disguise or vice versa.

The documentary contract: Pragmatic dimensions of the documentary

It is possible to see the cognitive dimension of audio-visual decoding processes as a part of the pragmatic dimension. What the cognitive dimension tells us is how big a role our acquired mental framework and schemas play in the interaction with texts, and through the cognitive dimension the institutional context gets a psychological foundation. However, the pragmatic dimension of both the documentary contract and fictional contract is not just the result of psychological and mental forms, but also of socially and culturally defined institutional practices. Such institutional contexts have mental representation, because our schemas are based on previous communication experiences. We have prototypes of contextual situations in our mental toolbox. But nevertheless, pragmatic context and contract is also a vital part of textual and communicative practices.

The cognitive dimension points to processes often unconscious or automatic and to more or less universal structures and processes based on the way our brain and body works. The pragmatic dimension deals with general processes of a more conscious kind and defined by more specific textual, social and cultural elements. When we speak of communication in a more sociological sense and genres in communication it is also important to see this in the image of an ongoing negotiation and cooperation between agents and the text in a given context.

The specific form of this communicative process of interaction between a sender, a text and a receiver is what defines a contract in a given genre in a sociological and textual sense. The contract for a talk show on television for instance is both similar to and very different from a gameshow, and a similar relation could be defined for the overall fictional contract and documentary contract. In pragmatic theory communication is a game based on specific rules and conventions in some kind of institutional context, where meaning is the result of interaction, construction and some basic form of generic contract between sender-”text”-receiver (Rorty, 1982: 110).

This pragmatic point of view can be defined on a macrolevel for instance through speech act theory (Searle, 1969). This is the case in attempts to define the fictional speech act (Kjørup, 1978, Lanser, 1981, Pratt, Mary Louise, 1977) or the so-called “Akte des fingieren” in Isers theory of reading (Iser, 1978, and 1983). Also in Teun van Dijk’s book on Macrostructures (Dijk, 1980) we find the definition of textual superstructures, such as narrative and argument, and we find an attempt to define pragmatic macrostructures, that organize different forms of speech acts. However, these pragmatic theories are still very general and mostly used in relation to linguistics and not visual communication.

In Pratt’s and Lanser’s theory of literature and fiction as a specific speech act they start out by trying to define the fictional speech act as either a specific type of illocutionary act not described in Searles typology (representatives, directives, commissives, expressives and declarations). They describe fiction speech acts as “hypotheticals” (Lanser, 1981: 289), or as a kind of “quasi speech act”, that uses the form of normal elocutionary acts, but at the same time suspends the normal rules for such acts. In Wolfgang Isers definition of “Akte des fingierens/pretending acts” he defines this act as the relation between “das reale/the real”, “das fiktive(the fictional” and “das imaginäre/the imaginary”. That is the fictional discourse is a special kind of speech act in a kind of “as-if-mode”, where the normal procedures for the reader/viewers construction of the referent are suspended, and the reference instead is made through the fictional interaction between the real and the imaginary. But where Iser places the fictional, one might add the documentary, in so far as the documentary mode can be defined as a kind of interaction where the difference to the fictional act is one aspect, but where nevertheless, the interplay between the real and the imaginary is also an important aspect. All in all, the results of speech act theory in defining fictional speech acts so far has not resulted in a clear definition of the codes and rules involved. All theorists in the end turn to more fuzzy, prototypical pragmatic and contextual rules and to specific institutional formations of genres, that activates specific decoding activities given the right textual signal:

The fictional signal indicated in the text will not work as such if not specific variations of historical conventions are shared by the author and the audience and understood as such. The fictional signal does not define fiction in itself, but the contract between author and writer, and the conventions of this contract, which the text does not carry as discourse, but as a staged discourse. (Iser,1983: 135, my translation).

The context forming the basis of a communicative contract has a number of important dimensions. The first dimension is the institutional, that is the context defining the sociological and cultural structure of the media and communication situation in general (public television, commercial television, cinema etc.). The second dimension is the intertextual, that is the more specific textual, cultural and psychologic aspects of genres and the relation between genres, that are activated, when we interact with a text. We interact by prototypical expectation, recognition and comparison, and we respond to signals not only in the text but also signals surrounding texts. The third dimension is what we could call experiental, following Johnson and Lakoff’s definition of how we acquire our knowledge, attitudes, concepts etc. from our body experience and our interaction with the world. This dimension, combines our social and cultural experiences and the cognitive structuring of our mind and emotion and thus the frames and schemas we are likely to apply on texts. The fourth and final dimension is the situational, that is the very concrete time, place and circumstances of a given communicative interaction.

What this article has tried to work out in a more detailed manner is the intertextual and experimental dimension of the documentary contract in close comparison to the fictional contract. Both the intertextual and experimental dimension tells us, that a number of textual elements by convention more often are found in documentary, than in fiction, but also that documentary is placed on a very flexible, graded line of generic prototypes ranging from fiction to non-fiction. There are numerous examples of mixing of textual formats, but it is also clear, that our cognitive ability to act in either an “actual” or a “hypothetical” mode constitutes different forms of “realness” and different forms of reference in fiction and documentary. Our mental activity in relation to real life situations and visual information in either documentary and fictional form are often very alike on a basic level, and we use imagination, emotion and metaphoric understanding across these lines. But since it is still important to distinguish fact from fiction, the ability to decide this is part of our mental “hardware”, and as a socially and historically acquired competence it is both a product of cognitive and semio-pragmatic dimensions.



Barnouw, Eric, 1983: Documentary. A history of the Non-Fiction Film. (Oxford University Press). Barsam, R. M., 1973: Nonfiction Film. A Critical History. (Dutton & Co, New York).

Brannigan, Edward, 1992: Narrative Comprehension and Film (Routledge, London). Bruhn Jensen, Klaus, 1987: Seernes tv-avis. (DR-report, 2B/87)

Bryant, J. & Zillmann, D., 1991 ed.: Responding to the Screen. (Lawrence Erlbaum, Hillsdale New Jersey).

Casetti, Francesco, 1990: D’un Regard L’Autre. Le film et son spectateur. (Presses Universitaires de Lyon).

Corner, John, ed., 1986: Documentary and the Mass Media. (Arnold, London).

Dijk, Teun van, 1980: Macrostructures. (Lawrence Erlbaum, Hillsdale New Jersey). Forsyth Hardy (ed), 1979: Grierson on Documentary. (Faber&Faber, London).

Giddens, Anthony, 1984: The Constitution of Society. (Polity Press, London). Goffmann, Erwin, 1974: Frame Analysis. (Harper & Row, New York).

Grodal, Torben, 1993: Cognition, Emotion and Visual Fiction. (KU, Institut for Film & Medievidenskab).

Gunter, Barry, 1991: “Responding to News and Public Affairs.” (in Bryant, J. & Zillmann, D., 1991) Höijer, Birgit, 1992: “Socio-cognitive structures and television reception.” (in Media, Culture and Society. vol. 14)

Höijer, Birgit, 1992a: “Reception of television narration as a socio-cognitive process: A schema- theoretical outline.” ( in Poetics,, no. 21).

Iser, Wolfgang, 1978: The Act of Reading. (John Hopkins University Press, Baltimore/London). Iser, Wolfgang, 1983, ed.: Funktionen des Fiktiven. (Fink Verlag München).

Johansen, Jørgen Dines, 1993: Dialogic Semiosis. (Indiana University Press). Johnson, Mark 1987: The Body in the Mind. (Chicago University Press).

Johnson & Lakoff, 1980: Metaphors We Live By. (Chicago University Press). Kjørup, Søren, 1992: “Faktion – en farlig blanding.” (i Mediekultur, nr. 19, 1992).

Kjørup, Søren, 1978: “Pictorial Speech Acts.” (i Erkenntnis, vol. 12, no. 1, jan. 1978). Kozloff, Sarah, 1988: Invisible Storytellers. (University of California Press).

Lakoff, George, 1987: Women, Fire and Dangerous Things. (University of Chicago Press).

Lanser, Susan Sniader 1981: The Narrative Act. (Princeton University Press).

Mandler, Jean Matter, 1984: Stories, Scripts and Scenes. Aspects of Schema Theory. (LEA, Hillsdale, New Jersey).

Meyrowitz, Joshua 1985: No sense of Place. The Impact of Electronic media on Social Behavior. (Oxford University Press).

Minsky, Marvin, 1985: The Society of Mind. (Simon & Schuster, New York). Nichols, Bill, 1981: ideology and the Image. (Indiana University Press).

Metz, Christian (1991). L’ énunciation impersonelle ou le site du film. Paris: Meridiens Klincksieck

Nichols, Bill, 1991: Representing Reality. (Indiana University Press).

Odin, Roger, 1983: “Pour une sémio-pragmatique du cinema”. (in Iris, vol. 1, 1). Odin, Roger, 1992: Book review, Metz. (in Iris, no. 14-15).

Pratt, Mary Louise, 1977: Towards a Speech Act Theory of Literary Discourse. (Indiana University Press, Bloomington,London).

Rorty, Richard 1982: Consequences of Pragmatism. (University of Minnesota Press, Mineapolis). Schank, Roger, 1990: Tell me a story, (Scribner, New York).

Searle, John, 1969: Speech Acts: An essay in the Philosophy of Language. (Cambridge University Press) Steffensen, Bo 1991: Fiktion som teksthandling. (Akademisk forlag, København).

Stam, Robert: New Vocabularies in Film Semiotics.  (Routledge, London).