When Jean Rouch died in February 2004, at the age of 86, he had completed around 120 films, with perhaps 20 more awaiting his attention in the editing room. He also left behind an extensive network of associates and collaborators in Europe and Africa, and an even larger number of people who had been inspired by the briefest of meetings with him. The retrospectives and conferences that have taken place this past year to mark Rouch's passing have therefore been heartfelt, not to say emotional gatherings.

As well as commemorating Rouch, they have attempted to assess his contribution to the two distinct, but related, worlds in which he was most active: cinema and ethnography. At the same time they have demonstrated how difficult it can be to get an overview of his creative output. The Comité du Film Ethnographique, part of the Musée de l'Homme in Paris and the guardian of much of Rouch's work, was deluged with requests for copies of his films in the months following his death. Unable to satisfy this demand, it issued an appeal for support in its efforts to preserve and make Rouch's films more widely available. A petition has been launched to support a project which proposes, amongst other things, to digitise Rouch's principal films (1).

The Building Bridges conference and retrospective (2), held in London in October 2004, demonstrated the additional problems faced by English-speaking programmers. The organisers said that they had secured everything that could be projected with English subtitles or voice-over. This amounted to just eight films: Les Maîtres fous (1955), Jaguar (1955/1967) (3), Moi, un noir (1958/1960), La Pyramide Humaine (1959/1961), Chronique d'un été (1961), La Chasse au lion à l'arc (1965), Petit à petit (1970) and Madame l'Eau (1993). In addition, they presented two short films: Cinémafia (1980), with a live “simultaneous translation”, and Tourou et Bitti (1971) with an introduction to explain the gist of Rouch's commentary.

This selection is a mixed blessing. It is by no means representative of Rouch's immense and varied output. But it does illustrate how Rouch has been seen outside his native France by the general public, and by many film historians.

The sample is particularly harsh on Rouch's African ethnographical films, which even within France are rarely screened outside academic circles, despite being edited for public exhibition. There is nothing here to illustrate the 15 or so films he made with the Dogon people, over a 30 year period. And while Les Maîtres fous (known in English as The Mad Masters) and Tourou et Bitti fall within the equally large body of work Rouch made on ritual possession in West Africa, both are exceptional rather than typical. La Chasse au lion à l'arc (The Lion Hunters) would be closer to the mark but for the washed out colour of the print and the replacement of Rouch's lyrical commentary by a vigorous North American narrator (Rouch made no secret of disliking this version of his film).

In the context of a retrospective, the loss of Rouch's voice means the absence of a major theme that developed through his ethnographic work. In the tradition of documentaries about Africa, Rouch's early films have a commentary which describes what is going on for the viewer. In the 1950s, with the aid of the earliest “portable” tape recorders, Rouch was able to transcribe and translate more exactly what the people on screen are saying. His commentaries begin to re-voice this material, reflecting the rhythms used and the poetry of African ritual language, whether it be a mother mourning her son or a hunter praising the poison on his arrows. Later still, Rouch uses poetic commentary to express the full range of narrative in the ceremonies unfolding before the camera – the “oral” tradition in Africa involves movement, costume, even location to transmit meaning as well as words. While intended as a means of representing an external aesthetic, this inevitably reflected Rouch's own “poetry” and becomes a strong feature of his style as a filmmaker. As he started to make more personal films late in life – such as Madame l'Eau - the same lyrical style reappears.

Madame l'Eau

Madame l'Eau belongs to a strand of Rouch's work which is better served by the program available to English-speaking viewers. These are the improvised fictions on African culture that Rouch made with his friends (and collaborators on other films) Damoré Zika, Lam Ibrahima Dia, Illo Gaoudel and Tallou Mouzourane. Jaguar (shot in 1954–55 but only released commercially in 1967) tells of a fictional but plausible journey of three young men from the savannahs to the coastal town of Accra to earn money. Petit à Petit (Little by Little) has Damoré and Lam travel to Paris to investigate how Europeans live in skyscrapers before returning to build one in Niamey. In Madame l'Eau, Damoré, Lam and Tallou travel to the Netherlands to see whether a windmill can be constructed on the Niger to help irrigate their fields, again returning to implement the plan.

In these and related “ethno-fiction” films, Rouch has serious points to make about the interrelation of European and African culture, and the need for Africa to find appropriate solutions to its problems. In particular, he is a champion of African ingenuity over European patronage. However these are not militant films, and Rouch gives more screen time to the comedy of these situations than any campaigning message.

Finally, Moi, un noir and Chronique d'un été are the most famous of Rouch's films, and the most fully integrated into the history of cinema. In Moi, un noir, a migrant worker from the banks of the Niger describes his life in the port city of Abidjan, the principal “actor” playing himself and voicing his own thoughts over the silent-shot images. This picture of the city's seamier side went down poorly with the colonial government of Côte d'Ivoire, and Rouch made La Pyramide humaine (The Human Pyramid) as an attempt to make amends. This uses a similar approach to examine relations between black and white students in an Abidjan school.

Chronique d'un été (Chronicle of a Summer) is an inquiry by Rouch and the sociologist Edgar Morin into the lives of Parisians in the summer of 1960. It was one of the first to deploy light 16mm cameras and synchronised sound recording, launching a highly mobile style of filming that spread rapidly through both cinema and TV filmmaking. Chronique d'un été also sent the term “cinéma vérité” out into the world, although both Rouch and Morin were quick to abandoned it when it became clear that people in France and beyond were reading this as “truth on the screen” rather than, as they intended, the “truth provoked by filming”.

The organisers of the London conference supplemented this program of Rouch in English on the big screen with three afternoons of video screenings in French. This filled in a lot of the thematic gaps, and included both landmarks from early in his career and rarely seen films from his later years.

Despite this separation of the English-only Rouch from the rest of his work, there was some cross-fertilisation thanks to a series of films on Rouch shown during the conference program. Most touching of all were films made by Bernard Surugue in the days before Rouch's fatal car crash in Niger. Surugue collaborated with Rouch on La Rêve plus fort que la mort (2002), another in the series of ethno-fictions made with Damoré, Lam and Tallou (and which was included in the video screenings).

At the beginning of 2004 Rouch was in Niamey for a festival of Nigerian cinema. Surugue's short film Le Double d'hier a rencontré demain (2004) shows extracts from La Rêve plus fort que la mort, screened in the festival, along with the frail director meeting with friends and receiving plaudits at the screening. At one point his wife, Jocelyne Lamothe, asks if he is OK (“Ça va, Rouch?”). “Rouch isn't here,” he jokes, “he has left.”

Rouch's Gang

The methods behind Rouch's ethno-fictions were exposed in extracts from Jean Rouch and his Camera in the Heart of Africa (1978) by the Dutch writer Philo Bregstein, a film that formed the links with the Netherlands that eventually led to Madame l'Eau. This was complemented by Rouch's Gang (1993), a film by ethnographers from the University of Leiden on the making of Madame l'Eau. Meanwhile Mosso Mosso (1997), by the Cahiers du cinéma critic Jean-André Fieschi, shows Rouch at work on the story that would become La Rêve plus fort que la mort (4).

This concentration of documentary material on Rouch's ethno-fictions, together with their strong showing in the film program, made this strand of Rouch's career the most thoroughly investigated by the conference. It shows the depth of collaboration with Damoré, Lam and Tallou, who would work with Rouch to elaborate the narrative structure of the film. Lam, Tallou and Rouch would scout locations and develop ideas. Damoré, meanwhile, would come upon each situation for the first time during the shoot.

While the framework and sometimes individual actions would be closely planned, the bulk of the action and dialogue would be improvised, with Rouch making suggestions from behind the camera (or rather under the camera - both in the ethnographical films and the fictions Rouch preferred to carry the 16mm camera himself). According to Bregstein, if an idea for an improvisation did not work first time, they would try something completely different.

With this additional information it is possible to go some way in making a case for Rouch's status as maker of genuinely African films. And yet, they are also remarkably personal to the group of friends making them. None of the speakers in London attempted to locate Rouch's ethno-fictions within the recent history of African cinema, and it seems that even those of his collaborators who went on to make films independently were inspired by the spirit rather than the substance of Rouch's cinema.

The legacy of Rouch's ethno-fictions is even harder to detect in European film. The conference included papers and film excerpts from two people working in this area: Julian Henriques, who has made films with residents of a housing estate in London (for example, We the Ragamuffin [1992]), and is currently working in the dance halls of Jamaica; and Andy Porter, who has made community-based fiction films for TV (for instance Wingnut and the Sprog [1994], in Protestant east Belfast) and also filmed in settings such as prisons. While there are common issues, such as self-representation and the power relationship between the filmmaker and the film's participants, there is no clear debt to Rouch.

Both speakers came to Rouch only after they had worked in film for some time, finding their way largely from improvised or community theatre. This results in differences, for instance, in the use of improvisation as a way of preparing a filmed fiction. So, Porter was able to show video clips of the workshop improvisation alongside the resulting external shots from Wingnut and the Sprog, whereas Rouch simply filmed the improvisation and made his film from that.

However, there are films that can be seen in this part of Rouch's tradition. Joram ten Brink, of the University of Westminster, suggested that the Michael Winterbottom film In This World (2002) may be “the perfect Jean Rouch film”. It follows a young Afghan boy as he journeys across central Asia and Europe to get to Britain, following its non-professional actors with video cameras.

The influence of Rouch was no clearer in much of the discussion of documentary film, in particular where the work of Anglo-Saxon filmmakers is concerned. Brian Winston, of the University of Lincoln, argued that this problem had two causes: in general, British and American filmmakers had little experience of Rouch beyond Chronique d'un été; and even then they have had little sympathy with what Rouch and Morin were trying to achieve.

Hence, Winston said, all that Rouch was seen to be adding was to put the filmmaker in the frame, a practice that was taken up in a tradition of reflexive documentary that leads to Michael Moore and Nick Broomfield. However, the presence of these directors in their own films is disingenuous, since the naiveté or foolish characters they present are far from reality. As Winston put it more bluntly, while Rouch puts himself in the frame as “an earnest of a type of cinema truth”, with Moore and Broomfield it is “a marker of a lying narrator”.

La Pyramide humaine

In contrast, Winston argued, Rouch's cinema seems in perfect accord with the current global trend in reality TV, a theme picked up by Jon Dovey of the University of Bristol. These programs construct unreal situations in which people reveal things about their personalities in playing themselves for the benefit of the camera. This is exactly the sort of “truth through cinema” that Rouch and Morin had in mind in Chronique d'un été, and which Rouch had practised in bringing together black and white students in La Pyramide humaine. While dismissing any direct link between Rouch and reality TV, Dovey said that perhaps Rouch's methods were becoming more appropriate for understanding the sophisticated forms of performance in this “mediated world”.

The debt to Rouch is clearer outside the Anglo-Saxon world, as demonstrated by the work of Les Ateliers Varan (5). This is a documentary workshop program which Rouch helped to establish, initially in Mozambique, and then in Paris. It trains people to make films, particularly across cultural boundaries, although in a documentary sense rather than in strict ethnographical terms. Séverin Blanchet, one of those responsible for running Les Ateliers Varan, insisted that its debt was to Rouch's spirit rather than his practice. However, it was the excerpts of Varan films that recalled Rouch more insistently than others shown.

For instance, there were encounters between three filmmakers from Papua New Guinea and the European French on the streets of Paris, between North African and European children in a school yard, and an examination of the pre-match rituals of French rugby players. Equally the “cinéma de contact” that Rouch strove for, through using hand-held cameras and wide-angle lenses, was apparent in a film made in a home for people with senile dementia.

Rouch's ideas could also be seen in the work of a video collective called Vision Machine (6), which is carrying out a filmed investigation of the Indonesian massacres of 1965–66. According to Michael Uwemedimo, a member of the collective, this can be thought of as an “archaeological performance”, in which people recover and reconstruct the events in which they were involved. This involves tactics familiar from Rouch, such as getting people to narrate their own stories over images of themselves, or re-presenting their stories on screen, either as fiction or documentary reconstruction. However, Uwemedimo also denied a direct influence from Rouch, preferring to talk of “moments of resonance”.

Meanwhile, Canadian filmmaker Peter Wintonick talked about the use of video technology, by human rights and other activists, in particular to gather evidence for political campaigns. What this work takes from Rouch is the reversal of the gaze, he said, the return of visual or media power to the viewed. However reasonable, it is hard to see such a strong link when Rouch was adamant that he had no right to get involved in political issues outside his own culture and criticised those of his compatriots who did so.

Aside from Rouch in the present, much conference time was given to Rouch's place in the history of post-war European cinema. Alain Bergala, formerly with Cahiers du cinéma, examined the connections between Rouch, Roberto Rossellini and the directors of the French Nouvelle Vague. Both Rouch and Rossellini could be considered “uncles” of the Nouvelle Vague, Bergala said, older than the young Cahiers critics who were in the process of becoming directors, but both respected and influential. Rouch was a familiar face at the Cinémathèque Française (at least, when he was in Paris rather than Africa) in the mid-1950s, and would screen rough cuts of his films there. Rossellini was also part of this circle, having relocated to Paris following the poor reception in Italy of his films with Ingrid Bergman.

The question that concerned Bergala was why the Nouvelle Vague directors had resisted the 16mm camera technology that would have allowed them to film freely and cheaply. The case for 16mm cinema had been made in 1949 by Jean Cocteau, whose opinion carried a lot of weight with the Cahiers critics. Rouch's films of the 1940s and 1950s gave a practical demonstration of what could be done with colour 16mm, and films such as Moi, un noir and Les Maîtres fous drew plaudits from the Nouvelle Vague, in particular from Jean-Luc Godard. They liked the fact that Rouch's fiction emerged from an encounter between the actor (professional or non-professional) and the camera, and his willingness to break the rules of cinema.

The classic example of rule breaking, cited elsewhere in the conference, is Rouch's construction of a sequence shot with “jumps”, as the central character in Moi, un noir acts out his experience as a soldier in French Indochina, while walking next to the Abidjan lagoon. For Rouch this was a way of resolving the technical problem of getting a long sequence shot with a wind-up camera capable of shooting for only a short period of time. It can be argued that this was taken up as a stylistic device by the Nouvelle Vague, as early as François Truffaut's Les Quatre cent coups (1959) and Godard's A Bout de souffle (1959). Jean-André Fieschi added it was significant that the same editor, Marie-Josèphe Yoyotte, worked on Moi, un noir, Les Quatre cents coups and Cocteau's Testament d'Orphée (1960).

But while the Nouvelle Vague directors may have respected Rouch and been inspired by his attitude, they did not adopt his techniques, despite their suitability for realising the aesthetic they were developing. Bergala argued that the reason for this was ideological. The Nouvelle Vague wanted to be part of the mainstream cinema, not (like Rouch) on its margins. This meant filming in 35mm, which in turn meant black and white (colour was just too expensive) and post-synchronised sound. If they used 16mm at all, it was for apprentice works, although Godard's Les Carabiniers (1963) appears to be an exception.

Instead, it was Rouch and Rossellini who influenced each other in a practical sense. On seeing the rushes of Jaguar in 1955, it was Rossellini who persuaded Rouch to have the actors narrate their own story. In return, Rossellini took up Rouch's approach and filmed in 16mm when he went to India in 1957–58. These colour films, with a hand-held camera and no set script, were used by Rossellini as a means of developing the film he would subsequently shoot in 35mm. There were even instances when Rossellini used shots from the 16mm “notebooks” in the final feature, India, Matri Buhmi (1958). The 16mm films were also screened on French and Italian TV, as J'ai fait un beau voyage (1958), with Rossellini providing his own commentary, very much in the style of Rouch's cinema, Bergala said.

The two strands finally come together in the “sketch” film Paris vu par... (1964). Known in English as Six in Paris, this arose from an offer by Rossellini, in the late 1950s, to help find funds to produce films by the Cahiers circle. According to Bergala, Rossellini received an enthusiastic response from the young critics but did not like the films proposed. He let the proposal drop, but it persisted and finally emerged as a collection of sketches, filmed in 16mm, by Godard, Claude Chabrol, Jean Douchet, Jean-Daniel Pollet, Eric Rohmer and Rouch himself.

Chronique d'un été

Further connections between Rouch and French mainstream cinema were made by Michael Uwemedimo, of Birkbeck College, University of London, as well as the Vision Machine collective. He considered the rise of the sociological survey in the French press in the late 1950s, and its adoption as a subject by filmmakers in the early 1960s. Rouch and Morin's Chronique d'un été is framed as a sociological inquiry into Parisian life, complete with a street survey on whether or not people are happy and in-depth interviews. However, as Uwemedimo pointed, the film raises many questions about the practice of survey and interview, by now common on French TV. At the very beginning of the film, for instance, Morin and Rouch talk with Marceline, one of the “characters”, about what they are about to do in the film. At the time Marceline was working as a poll-taker. So, here are the professional questioners discussing questions with a professional questioner. At the end of the film the “characters” discuss whether or not they were acting when being interviewed, and whether it is “decent” to reveal such things in public.

Uwemedimo went on to trace the use of such sociology in the cinema, from Truffaut's Les Quatre cent coups; to Chris Marker's Le Joli mai (1963) and Le Mystère Koumiko (1965). Godard's Masculin-féminin (1966) completes the sequence, drawing on Chronique d'un été, Le Joli mai and Georges Perec's novel Les Choses (1965), in which the principal characters are pollsters.

In a final paper considering Rouch's place in post-war European cinema, Anna Grimshaw of Emory University compared and contrasted Jaguar with Federico Fellini's La Strada (1954). Her aim was to free Rouch from his label as an ethnographer and “recontextualise” him with other leading directors. This was a more academic exercise than the other papers, dealing, for instance, with the relations in each film between the characters and the city, and with the journey they make towards personal fulfilment or redemption. During the discussion, Fieschi was asked whether such links were ever discussed at the time by the Cahiers critics. “We did some crazy things,” he replied, “but never that!” This was subsequently explained to be a reflection of Fellini's low stock at Cahiers (“He is a director for people who don't like cinema”) rather than an attack on Grimshaw's particular work.

The conference's examination of Rouch's relationship with Surrealism was more problematic. The papers in this area set the scene rather than giving a detailed exploration of Rouch's own work. Elizabeth Cowie, of the University of Kent, discussed the “Surrealist impulse” in documentary, and its ability both to raise questions about what is real at the same time as it strives to depict what is real. Meanwhile David Bate, of the University of Westminster, described the close links between Surrealism and ethnography in France in the 1930s, epitomised by the presence of “ex-Surrealist” Michel Leiris on the Dakar-Djibouti mission of 1931–33 and the accounts of the mission in the Surrealist-dominated journal Minotaure. Rouch has acknowledged the impact on him of this edition of Minotaure - the Dogon masks followed by De Chirico's landscapes - and he subsequently studied under the mission's leader, Marcel Griaule.

But it was clear from the subsequent discussion that Rouch cannot be considered a card-carrying Surrealist. His interest in African culture is clearly deeper than the Parisian infatuation with masks and other flea-market discoveries. Several anecdotes were put forward to illustrate this gap, from André Breton's repulsion at the possession rituals he witnessed in Haiti and in Rouch's Les Maîtres Fous, to Luis Buñuel's decision not to join the Dakar-Djibouti expedition, on the grounds that there was enough exotica at home to be going on with.

According to Bate, Rouch's interest in African myth concerned the poetics of the pre-conscious, not the unconscious. And while Rouch shared the Surrealist fascination with dreams, according to Cowie it was for their magic rather than as a way into the unconscious. “I saw that he was not a dreamer himself,” Rouch said of Freud, “but was rather exploiting dreams - like Karl Marx.”

According to Christopher Thompson, formerly of the University of Warwick and editor of a recent study on Surrealism, cinema and ethnography (7), Rouch's “sympathy with Surrealism” is best seen in his insistence on the adventure inherent in ethnography and the value of chance encounter. This is one reason why Rouch was drawn to film hunts, he said, which are at once the hunt for an animal and, for Rouch, the hunt for a film, its story and its images. This doubles back on itself when the hunters themselves tell the story of the hunt on their return to the village. The same can be said of Rouch's interest in migrations and the journeys made by his characters in films such as Cocorico! Monsieur poulet (1974), an ethno-fiction that follows the adventures of an itinerant chicken trader working the bush markets around Niamey.

Similarly the importance of chance encounter is shown in two of Rouch's Parisian films, neither screened here but which Thompson argued were key in Rouch's development as a filmmaker. La Punition (1962) shows a young woman leaving school one day, and meeting first with a French student, then an engineer, then an African student. It was in this film that Rouch fully applied the mobile camera work and synchronised sound that had been used towards the end of making Chronique d'un été.

Gare du Nord

Gare du Nord was Rouch's contribution to the “sketch” film Paris vu par... . Made of two 10-minute takes, the film involves a woman arguing with her husband, leaving their apartment, and encountering a stranger in the street outside. He asks her to run away with him, she refuses and he commits suicide. Both films also demonstrate Rouch's desire, at the time, to film without interruption, constructing his films within sequence shots rather than on the editing table. In discussion, Fieschi agreed that Gare du Nord demonstrated Rouch's sympathy with Surrealist poetics and openness to encounter, and was a more profound link with the movement than the “self-proclaimed Surrealism” of Fellini.

Finally, Thompson cited Liberté, égalité, fraternité, et puis après (1990), shown in the video screenings, as an example of Rouch's sympathy with another Surrealist trait - provocation, or as he put it “releasing disquieting objects” into the wider world. Responding to a commission from the committee charged with celebrating France's revolutionary bicentenary, Rouch sent two black actors out in costume to infiltrate the official pageant and interact with other actors portraying historical figures. This is followed (among other things) by a Haitian ritual in front of the Hotel des Invalides in Paris, home of Napoleon's tomb, to reconcile the Emperor's spirit with that of his one-time captive, Pierre Toussaint l'Ouverture, the black liberator of San Domingo.

The conference illustrates that while the Jean Rouch “classics” are already assured, there is much in his vast body of work that deserves to be brought into the light of day. The difficulty is in finding the thread or the theme which allows one to navigate through his phenomenal output. The conference illustrated one thread well - the ethno-fictions - but could only hint at others. The links that Bergala and other speakers picked out between Rouch, Rossellini and the Nouvelle Vague demand film illustration - Moi, un Noir followed by Godard's A Bout de souffle; Chronique d'un été followed by Le Joli mai and Masculin-féminin; the whole of Paris vu par...; Rossellini's Indian films.

But other, less obvious, suggestions were made, raising the most tantalising ideas for ordering and understanding Rouch's legacy. Look at Rouch's work in Niger this way, said Brice Ahounou, an anthropologist at the Comité du Film Ethnographique. In 1942 Rouch “met” and subsequently established a friendship with Dongo, the Songhay spirit of thunder. Over many years, Rouch filmed a series of cinematic portraits of this spirit, possessing men and women, young and old. These little-seen films contain a dialogue between Rouch, with the camera on his shoulder, and this divinity (8). Can there be anything to compare in film history?