As an Author of ethnographic films, Robert Gardner developed a praxis that was not only very different from that of Jean Rouch but also one that was at odds with some of the central tenets of contemporary anthropology as an academic discipline. The ultimate purpose of his film-making was not so much to discover what other ways of life might mean to those who lived them but rather to explore what those ways of life signified for him personally in existential or aesthetic terms. In sharp contrast to Rouch who returned faithfully to West Africa every year, Gardner travelled all over the world to make his films, rarely lingering long enough in any one location to develop profound personal relationships with the subjects, to learn their language or engage in any extended ethnographic research. Even so, there is much that ethnographic film-makers can learn from the detailed study of his works, in particular with regard to way in which he seeks to communicate both meaning and experience, not through verbal language nor by means of the mimetic reproduction of the world in a direct observational manner, but rather through the carefully wrought juxtaposition of visual symbols.

Over the course of a lengthy film-making career that began in 1951 and which continued until shortly before his death in 2014 (thereby rivalling even that of Rouch in duration, if not in productivity), Gardner shot, directed and edited five major feature-length documentaries on ethnographic topics. The first of these was Dead Birds (shot in 1961, released in 1964), which concerned traditional warfare in highland Papua New Guinea. There then followed two films set in Africa, dealing broadly with issues of gender: Rivers of Sand (mostly shot in 1971, released in 1974), an exploration of relations between women and men among the Hamar of southern Ethiopia, and Deep Hearts (shot in 1978, released in 1981), which concerns a ceremonial display of beauty by the young men of a Nigerien group of Fulani, in West Africa. Gardner's more recent major films dealt with religious topics: Ika Hands (shot in 1981, though not released until 1988), which presents the life of the priestly figures known as mama among the Ika, an Amerindian indigenous people of the Sierra Nevada de Santa Marta, northeastern Colombia, and, finally, Forest of Bliss (shot in 1984, released in 1986), an extended meditation on mortality in the form of a day-in-the-life account, from one sunrise to the next, of the funeral practices in and around the cremation pyres on the ghāts, the stepped embankments of the Ganges as it passes through the Hindu holy city of Varanasi (Benares) (figure 9.1).

9.1 Robert Gardner, aged 58, in Varanasi in 1984, during the shooting of Forest of Bliss (1986).

Gardner also directed four shorter ethnographic documentaries in collaboration with others. Early in his career, in collaboration with William Heick, he made Blunden Harbour and Dances of the Kwakiutl (both released in 1951) in a Kwakwaka’wakw community on Vancouver Island. ¹ Somewhat later, in mid-career, he made two collaborative films in India: Altar of Fire (released in 1976), produced in collaboration with the Sankritist J. F. Staal, which presents the re-enactment of an ancient Vedic ritual in Kerala; while Sons of Shiva (released in 1985), produced in collaboration with anthropologist Ákos Östör, follows a contemporary four-day ritual celebration in West Bengal. Gardner also collaborated in a less central way – as producer, cinematographer or editor – on half-a-dozen further documentaries on a broad variety of ethnographic topics.

Another important strand of Gardner's film work, though not ethnographic in intention, was a series of short portrait films, around ten in total, about North American or European visual artists. This strand began in the earliest period of his career but became increasingly prevalent in later years. He also worked on a number of other films of various kinds, including one feature-length work of fiction, which he produced for the Hungarian director, Miklos Jancso (also the subject of one of his portrait films). In the last decade or so of his life, he dedicated his energies to marshalling his cinematographic legacy, producing two compilation films based on a reworking of what he referred to as ‘forsaken fragments’ drawn from various different points in his sixty years of film-making. He also published two limited edition collections of photographs and three largely autobiographical books related to his film work. During this period, he actively supported the activities of a number of independent film artists through his production company, Studio7Arts. ²

Robert Gardner as ethnographic film-maker

Gardner would often assert that he thought of himself primarily as a poet or an artist rather than as an ethnographic film-maker. At the beginning of his career, while studying anthropology at the University of Washington in Seattle in the early 1950s (a degree that he never completed), he came across the experimental film-maker Sidney Peterson, the painter Mark Tobey and the poet Theodore Roethke, and quickly realised that the poet's view of the world was the one that he wanted to emulate. He found himself drawn to what he called ‘lyrical’ forms of documentary film-making, as represented by such diverse figures as Maya Deren, Stan Brakhage and Leni Riefenstahl, and the kind of truth that they could communicate. By the same token, he was not attracted to ‘the hard-edged cinéma-vérité style of truth every 24th of a second’. He concluded that there was probably no truth in the films produced by this approach and even if there was, he was more interested in the truths that lay beyond those delivered by ‘conventional storytelling or straight observational documentary’.

As a corollary of this self-identification as an artist or poet, Gardner was often ambiguous about the anthropological status of his work. Sometimes he would rather grandly declare an interest in ‘a higher anthropology’ while at others he would deny that his work had any anthropological import or intention. David MacDougall has suggested that we should take this disavowal with a pinch of salt, as analogous to the disavowal of symbolic intent by the poet Robert Frost. But if we consider Gardner's film-making career as a whole, it is clear that there was a marked shift in his relationship to anthropology, at least in its most academic manifestations. ³

In the mid-1950s, Gardner returned to Harvard where he had studied for his first degree and enrolled on a Masters programme in anthropology. It was at this time that Gardner began working with John Marshall on the editing of the latter's Ju/’hoansi material, including the most celebrated of Marshall's films, The Hunters , though his precise role in this process remains a matter of controversy. Later, he also participated in the Marshall expedition to the Kalahari in 1958. ⁴ At this stage of his career, Gardner was actively involved in promoting the use of film in anthropology, in both teaching and research, and played a leading role in setting up the Film Study Center in the Peabody Museum of Archaeology and Ethnology at Harvard in 1956. Gardner became the first director of the Center and remained so in its various institutional guises over the next forty years.

In the early 1960s, when preparing to shoot Dead Birds (1964), the film that marks the beginning of his most productive period as a documentarist, Gardner continued to feel, as he would later put it, ‘bound by anthropological scruples’. But gradually, ‘bewildered by such dismal notions as functionalism and structuralism’ which ‘in some miraculous way overlooked people entirely’, he progressively moved away from academic anthropology. By the 1980s, he claimed that all he needed to prepare himself for shooting Forest of Bliss was some literature that he had read in his student days, in combination with what he had gleaned more recently from a random selection of Indian novels. Under his direction, the interests and activities of the Film Study Center suggest a similar move away from a formal connection with academic anthropology over this period. ⁵

Whether or not Gardner's work can be considered ‘anthropological’ very much depends, of course, on how one defines anthropology, a famously contentious task. But it is certainly very different from the ‘shared anthropology’ of Jean Rouch. The longest that Gardner remained in any one location appears to have been the six months that he spent, when making Dead Birds in 1961, in the Baliem Valley in the highlands of West Papua, in what is now the Indonesian Province of Papua. For his later productions, as his recently published field journals make clear, he would often be on the move. As one would expect under these circumstances, he was not able to learn any local languages, nor develop extensive personal relationships. The journals reveal that he was often unsure of the significance for his subjects of what he was filming, even though he might have his own ideas about its potential meaning within his eventual film. They also reveal a sometimes disturbing detachment, even disdain for his subjects, as well as a number of examples of what many anthropologists would surely consider an unacceptable indifference to the subjects’ objections to the intrusion of his camera. ⁶

In order to overcome difficulties of comprehension or access in the field, Gardner usually worked with collaborators, often anthropologists, who had a more long-standing engagement with the subjects of the film. These collaborators would act as his interpreters and provide him with ethnographic contextualisation. Some of these collaborations appear to have been mutually richly rewarding: this appears to have been the case, for example, in Gardner's collaborations with Karl Heider on Dead Birds , with J. F. Staal on Altar of Fire and with Ákos Östör on Sons of Shiva and Forest of Bliss . All of these collaborators have testified eloquently to the manner in which working with Gardner opened their eyes to certain aspects of the ethnography of the people whom they were studying. ⁷

But in other instances, including his early work with John Marshall, Gardner's collaborations were decidedly fraught. Although personality issues appear also to have intervened, it seems that these tensions would often arise because the collaborators felt that Gardner was too ready to allow his personal vision to blind him to what they considered to be the ethnographic reality. One cause of tension with Marshall, for example, concerned Gardner's proposal to follow up The Hunters with a general film about Ju/’hoansi women, to be entitled ‘The Gatherers’. Marshall was adamantly opposed to this project, believing that Gardner's conception of this film ‘would bend reality too far out of shape’. ⁸ Some years later the anthropologists Jean Lydall and Ivo Strecker, who worked with Gardner on Rivers of Sand , would claim that it presented Hamar life in such a fragmented and inaccurate way that it was no better than ‘an ethnographic farce’. ⁹

Whatever the rights and wrongs surrounding the disagreements about the ethnographic probity of these particular films – and Gardner has certainly had his champions ¹⁰ – the general circumstances of filming remain inscribed in the films themselves. From simply looking at the films, it is immediately clear that Gardner's praxis was minimally dialogical, minimally participatory and minimally reflexive. With one or two exceptions, the language of the subjects is almost entirely absent or, if it is present at all, it is not subtitled. Unsurprisingly, since he did not speak their languages, there is little or no engagement between subjects and camera. But not only do they not speak to it, they rarely even look at it: the invisible fourth wall of classical theatre mostly remains firmly in place. The relationship between film-maker and subjects remains completely unacknowledged.

The exceptions to these generalisations are notable, but also only partial. One of these is to be found in Rivers of Sand , which involves some degree of participation in the sense that it features an extended oral testimony by a married woman, whom Gardner names as Omali Inda, in which she describes the manner in which women are controlled and disciplined by men in Hamar society. This is also an unusual film in Gardner's oeuvre in terms of content since whereas most of his ethnographic films are primarily concerned with ritual and ceremonial life, or if not, with warfare, all typically male domains of experience, this film accords greater attention to everyday life and also to women's experience. Made in the immediate aftermath of the collapse of Gardner's first marriage, it offers a particularly bleak vision of the possibility for harmonious relationships between women and men, not just among the Hamar, but anywhere. Indeed, Omali Inda's testimony is intertwined with Gardner's own more personal ruminations on the subject both at the beginning and at the end of the film, in the first case in the form of a rolling title, in the second, as a voice-over comment. However, their two statements remain quite separate: Omali Inda's testimony is delivered as a monologue rather than as part of some form of dialogical exchange with the film-maker (figure 9.2). ¹¹

9.2 Rivers of Sand (1974). Left, the beating of women by men is part of Hamar tradition, ‘so how can it be bad’?, asks Omali Inda. Right, according to Gardner, ‘Men are also afflicted, through their own tyranny, with wasted energies, idle spirits and self-doubt’.

Another partial exception is Ika Hands (1988). This film is unusual within Gardner's oeuvre in that it is structured around a discussion between Gardner and the eminent Colombian anthropologist Gerardo Reichel-Dolmatoff as they watch a preliminary edited version of the film, several years after the filming had been completed. This conversation is reflexive in the sense that it is initially shown in synch on screen and later heard in voice-over, providing not only some ethnographic contextualisation but also reflection on the film-making process. When Reichel-Dolmatoff observes that there are no shots showing social relations in the film, evoking a great sense of loneliness, Gardner wonders, interestingly, whether this says more about him than about the Ika subjects.

These exceptions merely prove the more general rule that Gardner's films arose primarily from a detached and subjective vision rather than from his interactions with the film subjects. But if this limits the ethnographicness of his films, at least as I have defined this concept for the purposes of this book, the most significant divergence concerns the general objectives of his film-making. Over the course of his whole career, very candidly and entirely consistently, Gardner made no bones about the fact that he was more interested in what culturally exotic ways of life meant for him as a film-maker concerned with issues of what he on occasion identified as moral philosophy than he was in what those ways of life meant for those who actually lived them. Thus, early in his career, he observed that while his first responsibility in making Dead Birds , both to himself and to the Dani, had been ‘to document with as much discernment as possible the most telling and important aspects of their life’, what the film represented for him was ‘the opportunity of speaking to certain fundamental issues in human life’. The Dani, he frankly acknowledged, ‘were less important to me than those issues’. ¹²

In this particular case, these ‘issues’ concerned the existential question of how it is possible for human beings to preserve a sense of meaning and value in the face of an awareness of their inevitable physical extinction, or as Gardner puts it, rather more resonantly, ‘how we all, as humans meet our animal fate’. This philosophical concern was a recurrent feature of Gardner's work and also animated the making Forest of Bliss some twenty years later. And as in the film about the Dani, Gardner approached the issue on his own terms, even when these ran against the grain of local attitudes. In the journal recording his initial reactions to the rushes of his material from Varanasi, we find him reminding himself that even though for him, the city is a place of pain and abandonment, for Hindus it is ‘a place of and an opportunity for exultation’. ¹³ Nevertheless, what we are offered in the final film is an unrelenting portrait of pain and abandonment leavened only by the most evanescent prospect of redemption. In this sense, as so often in his films, in Forest of Bliss , ethnographic understanding derived from engagement with the point of view of the subjects has been displaced or submerged by Gardner's more personal subjective vision.

Robert Gardner as cinematographer

Whatever reservations one might have about the ethnographic status of his films, Gardner was undeniably a highly accomplished cinematographer. His camera style was predominantly realist, though within this stylistic register he mixed a great variety of shots, from hand-held wide-angle tracking shots to tripod-based shots on the end of a long lens, all executed with consummate skill. Intermediate midshots or close-ups exploring intimate details of human bodies or material objects are a particularly important element within this repertoire. All his feature-length documentaries were shot in a richly saturated 16 mm colour stock, seemingly Kodachrome. This serves to give his films a suffused, poetic aura, though in considering his rushes from Varanasi, Gardner himself worries that this stock can also have the effect of aestheticising rubbish and decay in an overly saccharine manner. ¹⁴

Somewhat at odds with the predominantly realist tenor of his camerawork, Gardner had a tendency, particularly in his later films, to insert special effects, such as slow motion, time-lapses, unusual camera angles or distortions arising from the use of extreme lenses. The significance of these special effects is often either obscure or rather self-consciously artistic: as the screen studies scholar Scott MacDonald has put it, they are ‘little more than affectations – “arty” rather than artful’. MacDonald suggests, plausibly, that they can be put down to Gardner's ‘decades-long wrestle with the idea of being both documentarian and poetic film-maker’. ¹⁵

These special effects are most evident in the third of Gardner's feature-length documentaries, Deep Hearts (1981). This is a film about the geerewol, a ceremony involving the competitive display of song, dance and beauty by the men of a particular subgroup of the Fulani pastoralists of West Africa. Most commonly referred to in the English-language literature as the Wodaabe, this subgroup is also known as the Bororo or Borroro, which is the term preferred by Gardner. ¹⁶ The prevalence of unusual camera angles and other effects in this film is due to the fact that, owing to technical problems with Gardner's own camera, some of the cinematography was performed by his long-term friend and collaborator Robert Fulton, a specialist in using the Bolex, a non-synchronous 16 mm camera that allows one to achieve a range of special effects while still on location. At one point in his field journal, Gardner remarks that he is hoping to persuade Fulton ‘to look for the meaning in gestures and not be content with optical allure alone’. The final film suggests that he had only limited success in this matter. ¹⁷

Not only in this respect but in others too, Deep Hearts is certainly the weakest of Gardner's feature-length documentaries. In terms of substantive content, there really is not a great deal to this film other than a description of one part of the performance of the geerewol itself (figure 9.3). Although this is indeed a truly remarkable event, there is little exploration of the social or cultural complexities associated with it, other than some portentous and rather vacuous generalisations delivered through the narration. Rather than suggesting some further level of analysis or insight, the special effects appear to be merely making up for the shortcomings of content that one might expect in a film shot in just under four weeks, in what were clearly very difficult circumstances. Gardner's witnessing of the climactic events of the geerewol itself was confined to no more than a few days, and his various attempts to settle down with a group of Bororo to film everyday life met with repeated rejections. Although he did have an anthropological adviser with him on this trip, one Patrick Paris, introduced to him by Jean Rouch, Gardner frequently expresses concern in his field journal about Paris's shortcomings and his name does not appear in the credits to the film. ¹⁸

9.3 Deep Hearts (1981). Left, a young man prepares for the geerewol ceremony; right, a young woman selects her favourite from amongst the line of male dancers.

In both the variety and complexity of the shots employed, as well as in the occasional use of special effects, Gardner's cinematographic signature was quite unlike that of Jean Rouch. In general, Rouch's camerawork was much more rough and ready, not necessarily because he had a lesser cinematographic ‘eye’ than Gardner (though this may indeed have been true), but also because, through his engagement with this subjects and his stress on spontaneity, Rouch did not have the time and detachment in the moment of shooting to achieve the disciplined cinematographic effects that Gardner was able to do. This difference relates to their respective attitudes regarding the direction of the subjects of their films. Neither Gardner nor Rouch worked with anything resembling a formal script, but whereas Rouch might have actively sought to provoke his subjects into performances for his camera, he was reluctant – at least in principle, not always honoured in practice – to instruct them specifically in what they should then do. Gardner, by contrast, was prepared to direct his subjects, particularly in the early period of his career.

But surely the greatest difference between Gardner and Rouch in terms of cinematographic practice concerns their respective attitudes to sound synchronicity. Throughout the early part of his career, Rouch was actively seeking to overcome the technical difficulties of achieving synchronous sound on location. By contrast, even when the technical means to shoot in synchronous sound were readily available, which would certainly have been the case by the time that he shot Rivers of Sand in 1971, Gardner continued to shoot and record sound separately, and he went on doing so until at least as late as 1984 when he made Forest of Bliss . ¹⁹ This technical choice was symptomatic of a belief, shared by many film-makers who learned their craft in the era prior to the development of lip-synchronous sound, that the introduction of synchronous speech threatened to reduce non-fiction film-making from a form of cinematographic art to no more than a banal form of current affairs journalism.

Robert Gardner in the edit suite

Although Robert Gardner's skills as a cinematographer were undoubtedly remarkable, his skills as an editor were, arguably, even greater. Indeed, it was in the edit suite that his authorship as a film-maker took on its most distinctive form, considerably exceeding in sophistication the editorial praxis of Jean Rouch. It is for this reason that I shall give preponderant attention to this aspect of his work in the remainder of this chapter.

A privileged insight into Gardner's editorial praxis is afforded by the release on DVD of two of his major films, Forest of Bliss , in 2001, and Dead Birds , in 2004. In both cases, the films are accompanied by commentaries by Gardner on the postproduction process, delivered in informal conversation with colleagues. The conversation about Dead Birds is available as additional material on the DVD itself, and took place some forty years after the film was released. In the case of Forest of Bliss , it takes the form of a free-standing book with the DVD inserted into the back sleeve. The book reproduces a conversation between Gardner and the anthropologist with whom he worked on this film, Ákos Östör. Although this text was not published until 2001, the conversation took place as they watched the film together in 1987, when the memory of shooting the film was still fresh in their minds. ²⁰

In effect, Gardner began editing his films as he was shooting them, not only, as many documentarists do, at the level of the sequence or the scene, but even at the level of the individual shot. As his field journals make clear, while still on location, Gardner was always on the look-out for iconic shots that, through metaphorical association, would point to some domain of meaning lying beyond mere surface appearances. Contrasting his approach with that of anthropologists who are interested in social change, he comments in his Hamar field diary that, ‘My own interests are to look for that which is an apt symbol or sign and, at the same time, is distinctive in and of itself.’  ²¹ This was also the key to his praxis once he was back in the edit suite. Rather than take Gardner's images at face value as descriptive registrations of the world as it is, one should be constantly reading them as signifiers of some more transcendent meaning, even while bearing in mind that these images may be burdened with a phenomenal surface meaning that acts as an obstacle to grasping their true ulterior significance.

In order to help his audiences to get past surface meanings, Gardner often makes use of montage, though not in the most common present-day sense of the term, that is, a rapid sequence of images intended merely to truncate time or to summarise an event. It is a form of montage more akin to the so-called ‘intellectual montage’ developed by Soviet directors in the 1920s in which the aim was to generate, from the juxtaposition of two or more shots, a meaning that goes beyond the sum of the parts. But whereas Soviet montage was typically based on a sequence of brief shots, often self-consciously non-realist in character (in order to produce ostranenie, to defamiliarise), in Gardner's films these montages are typically composed of stylistically realist shots, often of relatively long duration.

This montage technique may also be applied to whole sequences based on a progressive, normal chronology that are intercut in the manner of parallel editing. But whereas conventional parallel editing is normally intended only to suggest temporal simultaneity, in Gardner's use of the device it often also carries some additional symbolic significance. Precisely because of the cinematographic realism and the often-normal chronology, a viewer who is not alert to Gardner's semiotic purposes may be lulled into interpreting these juxtapositionings as being merely descriptive. This is one of the reasons, I suspect, why some critics have been unable to read the carefully thought out analyses that are embedded in Gardner's films.

In the course of his career, Gardner's use of symbolic association in his editorial juxtapositionings became increasingly sophisticated and correspondingly more difficult to read. In the opening sequence of Dead Birds , his semiotic intentions are almost too obvious. Over the opening image of a gliding hawk shot from above, a sonorous voice-over, delivered by Gardner himself, relates a Dani legend that the Bird, on winning a race with the Snake in mythological times, became destined thereafter to die. This is immediately followed by a shot of a dead young man being taken down from a funeral chair prior to his cremation. This direct association of humans and birds as sharing the common condition of mortality is heavily reinforced by the title graphics. These come up in two stages, in reverse order, with ‘birds’ over the end of the outgoing shot of the hawk, with ‘dead’ only coming in with the cut to the human corpse. Throughout the remainder of Dead Birds , there are recurrent juxtapositions of images of men with images of birds or sounds of birds. There are also many shots of men with feathers in their hair or with the elaborate feather head-dresses which they wear when engaged in warfare (figure 9.4).

9.4 Dead Birds (1964). Left, a Dani man injured by an enemy arrow; right, a warrior holding a feather head-dress looks over the battleground.

Somewhat more difficult to read is the intertwining at certain points of the symbolic interplay between birds and humans with the contrast between men and women. For example, in an early sequence, Gardner intercuts between Weyak, the principal male subject, looking out for enemies from a watchtower silhouetted against the sky and Lakha, his wife working in the gardens nearby, apparently in the same late afternoon. But as Gardner explains in his discussion with fellow film-maker Ross McElwee on the DVD, the intercutting of their activities was intended to indicate more than mere simultaneity. What he also wants us to understand from this juxtaposition is not merely the contrast between men's work (seemingly light) and women's work (seemingly heavy), but between men's symbolic identification with the sky (and hence with birds) and women's identification with the earth. This association of humans and birds is also invoked in the contrast between Weyak and Pua, a young ‘swineherd’ who anxiously anticipates his own manhood. Towards the end of the film, these contrasts are elided as the women too become symbolic birds when they dance to celebrate the revenge killing of an enemy, while Pua kills and eats a bird and, as sign of his developing maturity, puts the feathers in his hair.

In Forest of Bliss , released some twenty years later, the symbolic elements are not only more numerous, but their juxtapositions are more complex and varied. Nor does Gardner provide any helpful keys to interpretation through voice-over narration. In Dead Birds , there is very substantial narration, written and delivered in a somewhat literary style. In addition to alerting the audience to the significance of what is happening on the screen, it is also used for retailing ethnographic information. Looking back at the film forty years later, Gardner regrets this, saying that it resulted from a ‘weird’ sense of the need to be a responsible witness. After Dead Birds , voice-over narration progressively diminishes in Gardner's work so that by the time of Forest of Bliss , it has disappeared entirely. Instead, Gardner seeks to communicate both significance and context by a series of non-verbal means, some of which involve metaphorical association, while others rely rather on formal properties of the filmic text, notably the transitions between sequences and the overall narrative structure.

Material metaphors

In his conversation with Östör about Forest of Bliss , Gardner gives a number of different examples of how he sought to use non-verbal means to impose some order on the ‘endless possibilities for confusion’ that Varanasi represented for him. In the first instance, he did so by approaching the reality of the city through a number of particular elements which, although very simple and very material, he saw as being laden with potentially illuminating metaphorical meanings: wood, marigolds, dogs, birds, children's kites. Rather than using doctrinal exposition by local ritual specialists or voice-over narration, Gardner takes on the extraordinarily bold challenge of evoking the principles of Hindu eschatology through the images and sounds of these simple material features of the world within which the disposal of the dead takes place in Varanasi.

Central to this eschatology, as described by the Indian sociologist Radihka Chopra in her review of the film, is the idea that life is to death as creation is to destruction, but that even while being opposed, these binary oppositions are connected to one another cyclically. Far from being kept apart in everyday life, they exist side by side, in a state of constant interpenetration with one another. ²² In Forest of Bliss , this philosophical dialectic is played out by the recurrent juxtaposition of opposites that are simultaneously both solidly material and highly symbolic. Thus the snarling dogs, scavengers of filth and gnawers of human corpses and, as such, symbols of extreme pollution as well as of the boundaries between human and animal, and beyond that, between life and death, are recurrently opposed in the film to the beautiful and bright marigolds, gathered in the peaceful countryside. Marigolds are also used as ritual markers of transition in a variety of contexts: adorning images of deities at the threshold of the human and the divine, garlanding a newly blessed boat about to be launched, as well as newly dead corpses, newly born babies and even the neck of a young puppy.

Similarly, heavy piles of wood destined for the funeral pyres are associated with dead bodies, while both are juxtaposed and contrasted, on the one hand with the healthy bodies of those who load the wood onto barges and row it downstream to the cremation grounds and, on the other, with the kites that are children's playthings and whose progress across the sky, dancing on the uplifting thermals from the cremation fires, evokes lightness, vitality and vulnerability all at the same time. In the striking montage of shots with which the film begins, it is suggested, by editorial juxtaposition, that a young boy, in seeking to launch his kite, is also somehow pulling up the sun. Much later in the film, shortly before dusk, two of these kites happen to fall down into the river in the background of a shot just at the moment that a child's corpse is being committed to the river. At this moment, the dialectical connection between the symbolically opposed values of sky and river, of spirit and body, of life and death is rendered in a form that is, as Gardner puts it, ‘powerfully actual’. ²³

This last shot is one of the many ways in which the spatial relationship between the sky and the river are exploited symbolically in the film. There are frequent cuts from dogs or corpses, in the river or on its banks, to shots of birds flocking noisily in an azure sky. Sometimes the two come together, as when we hear the sound of carrion birds cawing in the background while dogs gnaw on something, perhaps a corpse, or as the birds flock around the boat bearing wood for the funeral pyres upstream. Some of these birds are kites, others vultures, others sparrows, but all of them, although symbolic of the vitality of life in the moment, are also, for Gardner, harbingers of mortality, just as birds of other species had been for him in New Guinea. ²⁴

Nor is this vertical connection between sky and river the only form of movement through space that is used to symbolic effect. Equally important is the horizontal relationship between upstream and downstream, as manifest in the movement of boats through the frame of the film: at the very beginning of the film – as noted by Michael Oppitz in a perceptive early review – they move from left to right, while at the very end, they move in the opposite direction. At the beginning, the boats come out of a haze and gradually take shape, while the boat in the final scene gets wrapped up again in this haze before it disappears into the void. Oppitz concludes that ‘the former may be taken as movements into life and the later ones as movements out of life’. ²⁵ Gardner's comments, although not published until some fifteen years later, indicate that this change of direction was certainly intended and although he does not quite confirm the interpretation offered by Oppitz, it seems very likely that this is what he had in mind. ²⁶

On the other hand, Gardner is entirely explicit in identifying the symbolic significance for him of the diagonal movements in space from one side of the river to the other – that is, from the active western bank, site of the city and the cremation ghāts, to the sandy and unpopulated eastern bank. Even before he began filming, Gardner formed an analogy in his mind between the Ganges and the Styx, with the eastern bank being that ‘far shore’ from which, proverbially, no traveller returns, while the feral dogs that ranged there were, for him, none other than real-world embodiments of Cerberus, the hellhound who guarded the entrance to the Underworld of Greek mythology (figure 9.5). Although the people of Varanasi might consider the eastern bank no more than a place of recreation, and while Östör thought of it as a peaceful refuge from the confusion of the city, for Gardner it was a place of ‘quite forbidding mystery’, representing the world of death as opposed to the world of life on the more populated shore. ²⁷

9.5 Forest of Bliss (1986), material metaphors. Feral dogs eating carrion left, evoke Cerberus, the guardian of the Underworld in Greek mythology, while the wood transported upriver for the cremation pyres, right, suggests the presence of death.

In this spatial distinction, Gardner saw yet another way to organise the ‘chaos’ of Varanasi. Thus the film opens with a series of shots from the ‘far shore’, with ‘spectral galleons’ passing through the mist and Cerberus-like dogs on patrol, before crossing to the inhabited shore with all its confusion and vitality, for the main body of the film. There the action remains, apart from one or two interpolated shots from the ‘far shore’ that act as a sort of memento mori, before seemingly returning there in the final shot in which a rowing boat disappears slowly and inexorably into the mist once again.

Aural metaphors

Throughout his career, Gardner generally acted as his own picture editor but, as a sign of the importance that he attributed to the soundtrack of his films, he often worked hand in hand with highly skilled sound editors. In his first film, Blunden Harbour (1951), Gardner used asynchronous drumming and shamanic chanting to suggest transcendent significance. In his later ethnographic films, the sound editing becomes progressively more sophisticated. In Dead Birds , the soundtrack is greatly enriched but largely in a straightforward realist manner. In Rivers of Sand (1974), however, it begins to take on a more metaphorical function. Early in the film, the sound from a shot of a donkey braying in synch is carried over to cover the next shot, which shows a young man whipping a woman, a ritual practice that forms part of the Hamar male initiation ceremony. The clear implication of this sound overlay is that in Hamar society, men treat women as if they were no more than beasts of burden. Though this view of gender relations in Hamar society has been vigorously disputed by Jean Lydall and Ivo Strecker, the anthropologists who advised Gardner during the shooting of this film, the editorial technique itself nevertheless remains very interesting. ²⁸

The principal sound editor on Rivers of Sand was Michel Chalufour who would later work on a number of Gardner's other films, including Forest of Bliss . ²⁹ In the latter, the manipulation of sounds takes a variety of forms but one of its primary metaphorical purposes is to provide an aural complement to the many visual memento mori. The combined effect is to suggest that even in life one is surrounded, if not by death itself, then at least by potentially fatal suffering or menace. In some instances, the manipulation of sound consists merely of enhancing the diegetic synchronous sound. A simple example occurs when a carpenter takes a break from constructing the ladder-like bamboo biers that are used to transport corpses around the city. He lights up a cigarette and then exhales very loudly. This exhalation has been much augmented, Gardner explains, in order to suggest the final expiration of a dying person. ³⁰

There are many similar examples of sound augmentation in the film, some of which are intradiegetic, that is, they come from within the world of the film, while others appear to be entirely extradiegetic, that is, they come from outside the material recorded on location. The latter appears to be the case, for example, with the sharp, staccato footsteps of a dog trotting along a sandy riverbank that feature in the opening shot of the film. The crisp, almost percussive effect of the footsteps adds a pronounced sense of menace to the image, anticipating the violence of the snarling dog-fight that follows immediately afterwards.

The mere augmentation of sounds can be distinguished from a more elaborate intradiegetic use of aural metaphors. A recurrent example in Forest of Bliss involves the sound of chopping of wood. Throughout the film, this functions as a memento mori on account of its association with the preparation of wood for the funeral pyres. Often we see this in synch, but at other points the sound is featured in scenes in which the wood chopping itself is not visible. The first example occurs under the main title of the film, where one hears the sounds of trees being felled. Gardner was influenced here by a comment made by Baidyanath Saraswati, an Indian ethnographer who, along with Östör, worked as an anthropological adviser on the film. Saraswati said that as a child growing up in a rural village, whenever he heard the sound of mango trees being chopped down, he knew that a death had occurred. In their discussion, Gardner and Östör reflect on how they went to a great deal of trouble to film tree felling at some distance from the city. However, in the end, Gardner decided that the visual image of the woodsmen actually doing the felling would be ‘too puny’ to carry such a weighty metaphorical charge. ³¹ Yet without the aid of the contextualising comment from Saraswati, it is surely optimistic to think that any viewer will be able to understand this metaphorical reference, especially as it occurs quite so early in the film.

Another acoustic effect used in this way is the sound of bells tolling. Although tolling bells do occur in synch in the sequences set in temples and shrines, they are also used in an intradiegetic way, with ambiguous effects since, as Östör suggests with regard to a usage very early in the film, one is not quite sure whether they are tolling to announce a death or are simply liturgical bells marking the time of day. Gardner confirms that he wanted this ambiguous effect to encourage the audience to ask questions:

Throughout the film, a number of sounds are used recurrently to emphasise and reinforce the juxtaposition of the symbolic opposites of life and death, of sky and water, purity and pollution. These include funeral chanting (rama nama satya he – ‘God's Name is Truth’), the swishing of water, the barking of dogs and the squawking of birds. But of all the aural metaphors used in the film, undoubtedly the most striking is one that we hear for the first time about seven minutes into the film. Here it features over a shot of the prow of a boat which, together with a brief and difficult-to-discern shot of a dog gnawing at a half-submerged corpse, is inserted non-sequentially into the middle of an otherwise largely observational sequence of one of the main characters taking an early-morning bathe in the river. It is a strange, creaking, and somehow ominous sound, which, like the sound of the tolling bells, is clearly intended to provoke a question in the mind of the viewer.

This sound occurs on two further occasions over shots of a boatload of wood being rowed upstream for the funeral pyres, but it is not until some 37 minutes into the film that we finally discover what it is. It turns out to be the sound produced by the grating of bamboo oars in the small rope lassos that serve as rowlocks on these boats. Oppitz describes this acoustic device as a ‘coup de maître’ and suggests that it acts as ‘the musical leitmotiv of the entire film, a sound metaphor for terrestrial suffering, pain, labour, and disharmony’. The most dramatic use of all is saved for the long final shot of the film in which a rowing boat slowly disappears into the mist on its way to ‘the far shore’. With this usage, Oppitz remarks, ‘Gardner strikes his best transcendental string’. ³³

Subjects as metaphors

One of the most distinctive features of Gardner's ethnographic film-making praxis is the absence of subjects of rounded, idiosyncratic character. As he said of the Dani early in his career, the subjects themselves are not of interest to him as actual, embodied individuals: what is of interest is what they signify. Some twenty years later, in his conversation with Östör, he goes further, arguing that the idea that one could capture the sense of an individual human being on film is a complete illusion. ³⁴ Certainly, the subjects of Gardner's films are not characters of the kind that one discovers in the films of Jean Rouch or David and Judith MacDougall, that is, individuals full of ideas and opinions, and a range of moods and humours, often mutually contradictory, as in human experience generally.

The closest that Gardner comes to such characters are Weyak and his wife Lakha, and most of all, the young swineherd Pua in Dead Birds , but even they come across more as archetypes rather than rounded individuals. More usually, Gardner's characters are primarily vehicles through which he can explore particular issues or aspects of the human condition that he perceives as being played out within the cultural arena in which those subjects happen to live. By taking away the particularising detail of their lives, Gardner seems to be encouraging us, rather in manner of the hagiographers of Christian saints, to attend more directly to what his subjects’ lives might signify transcendentally.

In Forest of Bliss , the three principal subjects each represent different aspects of life in and around the funeral ghāts, like so many characters in a morality play, and there is no sustained attempt to establish the idiosyncratic person behind or beyond the role that each plays in the film. Even the device of personalising the subjects through naming them, as in Dead Birds , is not employed here. As with the manipulation of material symbols described above, the three subjects are used instead to present a series of fundamental symbolic contrasts. At one extreme is a priestly figure whom Östör identifies as Ragul Pandit, a ‘ritualist’. He represents a series of positive values: wisdom, serenity, purity. His ritual incantations at dawn, reminiscent for Gardner of Gregorian chant, are intended to provide some redemption for the audience at the end of the film, after they have been taken through the valley of the shadow of death. At the other extreme stands the Untouchable, the Dom Raja, a character of ‘utter balefulness’, as Gardner describes him. His ugly and sick body, his broken voice and manner of arrogant contempt accord perfectly with his very worldly business as supervisor of the funeral pyres, an enterprise that he appears to pursues with a ruthless indifference to the bereavement of his clients (figure 9.6).

9.6 Forest of Bliss (1986), subjects as metaphors. Left, the holy Ragul Pandit performs prayers at dawn; right, Dom Raja, the Untouchable cremation businessman, a character of ‘utter balefulness’.

If the Pandit represents the spiritual life and the possibility of renewal, the Dom Raja represents the utter ineluctability and meaninglessness of the material world. Between these two poles, as a sort of trickster, lies a healer and diviner, identified by Östör by name as Mithai Lal, who like the Pandit engages in ritual activity, but who like the Dom can also be vulgar and materialistic. Whereas the other two characters both represent their station in life unambiguously and with utter seriousness, the healer seems to be a joker, capable of both good and evil, banality and religious inspiration, whose mental state verges at some points on insanity. ³⁵

Transitions

As Oppitz astutely pointed out in his review, Gardner employs a range of different types of editorial transition to suggest symbolic significance in Forest of Bliss . ³⁶ Sometimes this significance is achieved through the juxtaposition in a sequence of shots that are reiterative of the same point. One of the most striking examples here is a sequence in which a number of dead animals are shown being dragged down the steps of a ghāt, without ceremony, in order to be disposed of in the river: first a donkey and a cat, and then a dog, are hauled down, their bodies twisting awkwardly and their skulls resounding hollowly, probably with the aid of some acoustic enhancement. But this series of images is preceded and made meaningful by a shot of an elderly blind man also making his way, carefully and elegantly, down the steps of a ghāt. In his rushes log, Gardner records that he had decided long before going to Varanasi that ‘steps would be important indicators of transition: between life and death as much as between river and city’. The implication of this sequence then is that for all their dignity, human beings share the ‘animal fate’ of other creatures, that is, we too, actually or metaphorically, will one day be taken down the steps of a ghāt and disposed of in a river somewhere. ³⁷

These serial montages may be contrasted with a technique whereby semiotically significant juxtapositions are incorporated within a sequence that is ostensibly a real-time event but which has actually been constructed in the edit suite. An example in a relatively early sequence shows a boat being rowed upstream, heavily laden with wood. As it progresses, there are two shots, presented as if they were taken from the point of view of someone in the boat, first of vultures whirling in the sky, then of a corpse floating face-downwards in the shallows, anus towards the camera. In commenting on this sequence, Gardner is quite candid in admitting that these two shots were ‘connected by editing, not at all by actuality’. His intention, he explains, was to intimate that this consignment of wood, the purpose of which has not yet been made clear, has ‘some death-related meaning, that it is not just for keeping people warm at night’. ³⁸ Only later do we discover that this is the wood for the funeral pyres.

In this case, the shots of the vultures and the corpse, although only connected to the boat in the edit suite, have been carefully woven into the narrative of the film: not only do they fit very smoothly into the chronological progression of the boat upriver, but they also appear to be representing the point of view of somebody in the boat. But in other cases, Gardner inserts into the middle of a sequence one or more shots that are heavily charged with metaphorical meanings, yet are quite clearly from another place or another time.

A good example here is the lengthy opening sequence in which the healer-diviner Mithai Lal makes his way down the ghāt steps to bathe. Once he has immersed himself in the river, the generally observational quality of the sequence is suddenly interrupted by an abrupt cut to a tightly framed long shot taken from the ‘far shore’ showing a sail going by, with a funeral pyre burning in the background. Then, after returning for another couple of shots of the healer frolicking in the water, there is a cut to two people launching an offering of marigolds laid out upon a large leaf. Shortly afterwards, the bath time routine is again interrupted, this time in an even more startling fashion, by two further shots, one of a dog gnawing at what seems to be a corpse and the other, the prow of a boat shot from above that is accompanied by the grating sound of the unseen rope rowlocks. The exact import of these interpolated shots is not clear at this stage, but Gardner comments that he wants the viewers to become aware right away that the river has many meanings beyond that of simply being a good place to bathe. ³⁹

In all these cases, the juxapositioning for semiotic purposes involves the insertion of one or more metaphorically significant shots within a sequence. But in other cases, it is complete sequences that are associated semiotically by being intercut with one another, in the manner of conventional parallel editing. This occurs in one of the most important scenes midway through the film, in which a sequence of a new boat being launched into the river, garlanded with marigolds, is intercut with a sequence in which a corpse, similarly garlanded, is shown being brought down on a bier and immersed in the river. The meaning of this conjunction was immediately clear to Radhika Chopra, as she describes in her review:

Gardner insists that these events were not brought together in the edit suite but in reality did actually happen simultaneously at the same ghāt. All documentarists live in the hope of being blessed with such epiphanies, but Gardner is only too aware of how such simultaneities can be editorially manufactured. He therefore now worries that despite ‘this sanction of reality’, the conjunction of the two events in Forest of Bliss might be considered just too contrived. He also worries that the two events might be considered not only metaphorically linked, but metonymically, that is, the audience might presume that the deceased once owned the boat, or that the boat is being launched for the purposes of carrying the corpse. But he concludes, ‘that may be part of the price you pay in order to get the effect you want’. ⁴¹

But if all these editorial juxtapositionings refer to contexts of metaphorical meaning that pertain to Varanasi in particular, there are other instances in which they appear to be reaching out to meanings of more general significance. For Gardner's ultimate objective is not just to treat the subject of death in Varanasi but also, as he puts it, euphemistically, ‘journeys to any far shore’. This, as he remarks with good reason, is not easy, given the very literal and specific nature of film as a medium. ⁴²

An example of how Gardner attempts a generalising reference of this kind occurs in a key scene almost halfway through the film. This is set in what we learn, albeit from sources outside the film, is a hospice for the dying. ⁴³ We have been introduced to this hospice before and now return to it for a second time. On this occasion, from a respectful distance, we observe the corpse of a recently deceased person as it is carried down from an upstairs room, wrapped in a white shroud, to the sound of chanting by the attendants. On the ground floor, it is placed on a simple bamboo bier and decorated with a single garland of marigolds, before being carried out to the cremation ghāts. Although we have encountered all the elements of this scene before – even, one supposes, the deceased person, since in the previous scene in the hospice we saw some old women in extremis – this is the first time that they are presented to us in a single, consecutive strand.

However, what distinguishes this from a conventional example of chronologically progressive editing is that towards the beginning of the sequence, just before the corpse is brought down by the attendants, there is a highly subjective point-of-view shot, perhaps even slightly in slow motion, that descends the empty staircase on its own, as it were. As it is strikingly out of character stylistically with the rest of the sequence, this is clearly intended to remind the viewer that death is not just something that happens to people in ‘other cultures’.

Structures

All Gardner's major films are organised around relatively conventional cyclical narrative structures. These typically commence with the posing of some initial issue, problem or mystery and then progressively carry the audience forward to some final culmination, before returning literally or metaphorically to the beginning.

In Dead Birds , the return to the beginning is essentially metaphorical, though the link between beginning and end is also made through an ingenious narrative device, itself also metaphorical. After proposing the common mortality of humans and birds in the opening sequence, as described above, the narrative is thereafter both framed and advanced by the weaving of a band by one of the principal characters, Weyak. This functions as a sort of ‘clock’, as Gardner has put it. ⁴⁴ Weyak starts to weave the band at the beginning of the film, is shown working on it at various points during the film and finally completes it at the end. As such, the progression of the band parallels the temporal progression of the film through a series of events related to the ongoing ritual warfare between Weyak's group and their enemies. These culminate in a revenge killing and a celebratory dance. However, the band also represents the progression of time in a more sinister metaphorical sense since, as we discover about ten minutes into the film, it is a funeral band. As Weyak completes his task following the celebratory dance and disappears over a slight rise, the narration brings the film to an end by returning to the theme of the common mortality of men and birds announced in the opening sequence.

In Rivers of Sand , the link between beginning and end is somewhat more obvious. In this case, the film ends with a return to the opening oral testimony of Omali Inda as well to Gardner's initial pessimistic ruminations about the inevitable ‘painful’ differences that separate women and men. This thematic return to the beginning is reinforced by a very literal visual device. In the opening montage, we are offered a shot, taken from a low angle, of a Hamar man walking towards the camera across one of the eponymous rivers of sand that feature in this film, his sandals scrunching loudly on the gravel. The final shot of the film is of the same man, taken from the same position and angle, and presumably taken at the same time as the first shot, but this time the subject is walking away from the camera, his sandals still scrunching loudly.

The narrative structure of Forest of Bliss is also cyclical, though in the absence of an orientating voice-over to provide signposts along the way, the progress from beginning to end is advanced instead through another classical structural device, the ‘as if’ chronology that we have identified a number of times already this book. In this case, it is based on a 24-hour cycle, running from one sunrise to the next. As the material for this film was shot over a period of some ten weeks, we can be sure that this 24-hour cycle would have been entirely constructed in the edit suite, involving a wholesale rearrangement of chronology. ⁴⁵ This constructed diurnal format fulfils, very effectively, the basic function of all narrative structures, that is, it carries the audience forward, with the aid of subliminal sign-postings afforded by the progression of the day, towards an ending anticipated in advance. Moreover, in this particular instance, the use of this structure is particularly appropriate since it offers yet another echo of the cyclical principle underlying Hindu eschatology as the film moves from daybreak to night and back again, ending with the ‘rebirth’ of the new day.

It is also more than an inert formal pattern for ordering the material. Deploying yet another classical narrative strategy, Gardner animates this structure by posing a series of enigmas that he hopes the viewers will be intrigued by and will therefore allow themselves to be carried forward to the next stage of the film. In effect, Forest of Bliss develops by a process of slow disclosure, with each material component of the funeral procedures being introduced independently with, as it were, a question mark attached. Most of these are finally brought together in the scene in the hospice described above, which occurs shortly before the midpoint of the film. Only one major component is still missing, namely the wood, but then there is an immediate cut back to the athletic young boatman whom we have already seen a number of times before, working his way upstream with a large load of logs destined for the funeral pyres.

Thereafter, the pace of the film picks up and corpses on stretchers appear to converge on the Manikarnika cremation ground from all directions to the recurrent chant of ‘rama nama satya he’, sung in a variety of speeds and styles. However, there is still a considerable delay before we finally see a cremation, even though we see everything associated with it: we see wood being stacked, weighed and thrown down to the cremation ground; we see smoking pyres; we see women, a young boy and a dog picking among charred embers, we see stretchers of corpses being washed in the river and then lined up on the ghāt, we see lamenting mourners from a distance and a close-up of the doleful expression of a water buffalo whose symbolic role in this film appears to be to epitomise melancholy.

Even when the chief mourner takes a smouldering straw torch from the Dom Raja's hearth and walks down towards the cremation ground, we are not immediately vouchsafed a clear view of the cremation itself. Instead, there is a very long shot, taken from a boat of the kind used for hauling the firewood, now empty and almost imperceptibly pitching off shore. The empty boat itself takes up most of the image while the chief mourner is barely visible among a knot of other mourners at the top right of frame. Here he circumambulates a pyre with the now-flaming torch and then finally sets light to it. The soundtrack is rather muted compared to the previous tumult of chanting, but there is a low howling of dogs, the ever-present acoustic markers of the frontier between life and death (figure 9.7, left).

9.7 Forest of Bliss (1986), metaphors of transition to the ‘far shore’. Left, while a cremation takes place in the far distance, an empty wooden barge stands offshore, as if waiting to carry the soul of the deceased across the river. Right, in the final shot, a rowing boat disappears into the mist to the sound of grating rowlocks.

Gardner explains that throughout the editing, he was dubious about whether he would use this shot but finally decided to do so because it pulls together a number of threads:

Even then, remarkably, we still do not cut back straight to the pyre. In-between there are two shots, first of a water pipe emptying into the river and then of a boy drinking from it. These shots, Östör points out, are there to remind the viewer that in Varanasi the ordinary is combined with the extraordinary. But the sound of the splashing water runs into the crackling of a fire and finally, we at last see a close-up of a shrouded head licked by flames, at what is almost exactly the three-quarters point of the 89-minute film.

Gardner says that he found this part of the film more difficult to cut than any other. Even while shooting, it had occurred to him repeatedly that he should avoid ‘simply documenting cremation’. He felt that he had an obligation to preserve the dignity of the corpses while at the same time making the subject of death itself interesting. He was convinced that simply showing a long shot of a burning body would be ‘pretty tedious’. ⁴⁷ So now, having finally arrived at the culmination of the event towards which the whole film has been moving up until that moment, he quickly withdraws. After a single shot of the burning corpse, he returns briefly to the mourners lamenting and to the Dom Raja counting his money as the pathetic material goods left by one of his clients are kicked into the dust. Then he wraps up the whole sequence with a montage of the later stages of the cremation, moving very quickly over some of the more challenging moments, such as when the skull is smashed to release the spirit and the brains fall out.

In their conversation, Östör expresses surprise that Gardner should deal so briefly with this stage of the process, particularly since he had been so uncompromising in showing corpses in the river earlier in the film. ⁴⁸ But there is a clear similarity here with the way in which Gardner treated the cremation of the boy killed in Dead Birds some twenty years previously: there too, though Gardner had been totally uncompromising in the build-up to the climax of the funeral, we do not actually see the final consummation of the corpse by flames. Once it has been enclosed within the funeral pyre, the scene is elegantly and swiftly brought to an end by means of a pan off that follows the first wisps of smoke arising from the pyre as they disperse into the sky.

It is also highly characteristic of Gardner that he should not be interested in faithfully documenting the ritual processing of a dead body per se, but rather in interpreting its significance. As he puts it:

If I understand Gardner's intentions correctly, these requirements are not met by the documentation of the cremation itself but rather by its juxtaposition with the shots immediately before and after: that is, by the conjunction beforehand with, first, the shot from afar in the boat of the circumambulating mourner, suggesting tranquility and the imminent dispatch of a soul, then with the shot of the boy drinking water indicating the everyday nature of the event, next by the conjunction with the shots underlining the insignificance of the material possessions left behind, and finally and perhaps most importantly, with the serial montage of shots which follows the cremation ground sequence. This consists of a veritable flurry of metaphors of transition – birds descending to the river and alighting on driftwood, dogs circling one another, various boats now moving downstream from right to left and, on the soundtrack, the ever-ambiguous tolling of bells and the cawing of birds.

The final quarter of the film, covering the period from sunset to sunrise, is largely ‘redemptive’, as Gardner terms it, intimating that life goes on, not despite death but in association with death since the two are connected cyclically. The continuation of life is represented by the children flying their kites, by a brief return to the marigold gardens, by the devotional ritual in a temple and eventually by a return to Ragul Pandit chanting at dawn. But in the midst of this predominantly life-affirming final quarter, there is a disturbing and ambiguous sequence in which Mithai Lal, the healer-trickster, apparently in trance, is shown chanting over a flaming pit. The section is also regularly punctuated by memento mori: by the burial of a child in the river, by the burning funeral pyres and the tolling of bells at night, by dogs on the shore at dawn and further boatloads of wood for the funeral pyres. Finally, the film returns to where it began, both visually and aurally, on the river, with the boat disappearing into the mists to the sound of the grating rowlocks (figure 9.7, right).

Experience, subjectivity and ethnographicness

Through the gradual, unrelenting, almost overwhelming accumulation and superimposition of symbolic juxtapositions, Gardner moves the audience of Forest of Bliss , in a manner analogous to the participants in a religious ritual, that is, without consciously understanding all of the connections, towards the master tropes of Hindu eschatology: if opposites are necessarily connected, as dogs are to marigolds, as birds in the sky are to wood on the river, and as day is to night, if bliss can exist amid squalor, if the river Ganges is both a place of pollution and of purification, if the fire of the funeral pyre both consumes and releases, if in life we are in death, then in death we are necessarily also in life, and the committing of the body or its ashes to the waters of the holy river is not just the end of a cycle, but also a beginning.

It seems that Gardner intends this to be not so much a film about a rite of passage but to be a rite of passage. Through the narrative of the film, he aims to engage and carry the audience down to the cremation ground to confront their own mortality. But then, having forced them to look into the void, he offers the possibility of redemption in the form of Ragul Pandit:

Gardner is aiming, above all, to communicate an experience. However, the meaning of this experience is highly mediated through his own subjectivity rather than of the subjects. It is here that both he and his critics consider that his work may be in conflict with the some of the central tenets of academic anthropology. Reflecting on the report that certain world-renouncing saddhus would go to the cremation grounds to lie down on the pyres in anticipation of their own immolation, even engaging in cannibalism to transgress beyond all social convention, he comments:

In this passage, if somewhat contortedly, Gardner poses an important question: is there, in fact, a necessary conflict between an anthropological approach and the exploration of existential issues of personal importance to a film-maker? Or is it merely, as he suggests here, merely a presumed conflict? My own view is that while there is no necessary conflict between the pursuit of personal existential enquiry through film and anthropological objectives more generally, what anthropology as an intellectual discipline teaches, surely above all else, is that this personal existential enquiry should always be channelled and tempered by sensitivity to local ethnographic contexts.

When Forest of Bliss was first released, it was the subject of a number of excoriating reviews by anthropologists in the pages of the visual anthropology newsletter of the American Anthropological Assocation. These provoked a series of equally robust responses by Gardner, Östör and various other supporters. The controversy soon began generating more heat than light and quickly became unnecessarily personalised. What was clear, though, was that at least some of the anthropologist reviewers had entirely failed to recognise the film as a complex and finely crafted work that sought to explore funeral practices in Varanasi through cinematic means, using non-verbal metaphor and montage rather than didactic explanatory exposition of the kind that one might expect in a written text. However, at the same time, the critics also raised a number of legitimate points about the ethnographic status of the work that are worthy of closer attention.

First, there are a number of points in the film where ethnographic reality appears to have been subordinated to Gardner's metaphorical purposes, a criticism that has been made about his previous work too, as described earlier in this chapter. In his review of the film, Jonathan Parry, an anthropologist who has carried out extensive fieldwork in Varanasi, points out a number of examples, but here one will suffice: Parry explains that Mithai Lal, the trickster-healer character, has no direct connection with the funeral industry while the temple in which we see him at work is at some distance from the cremation ghāts. In reality, he is a spirit-medium of the Goddess Kali or Durga, and his ministrations are concerned with petty marital squabbles, troubles in business and minor illnesses rather than with major eschatological issues of life and death, as is suggested by his metaphorical function in the film. ⁵²

A second matter of concern relates to the fact there is a complete absence of the indigenous voice in the film. I refer here not just to the literal voice of the subjects, but also more metaphorically to their general views and understandings. But here too Forest of Bliss is entirely consistent with Gardner's praxis in his earlier works. Just as he was only secondarily interested in what the Dani might think about their warfare and its connection to mortality, he is only secondarily concerned with local eschatological understandings in Varanasi. Instead, he approaches these matters through the clues provided by the classical Greek mythology that he encountered in his childhood: thus the Ganges becomes the Styx, the scavenger dogs on its shores become transformations of Cerberus and the eastern bank that ‘far shore’ from which no traveller returns.

A particular example relates to the title of the film. Although it is poetically evocative in some indeterminate way, we do not learn from within the film what this phrase means for Hindus. It could be argued that we are the poorer for this as it could have enriched our viewing if we did. As Chopra helpfully explains in her review, this is the name given to Varanasi in certain classical Sanskrit texts and refers to the supposed abundance of lingas, the phallic symbols connoting Lord Shiva and associated with ananda, the sheer bliss of creation. (It is presumably some of these lingas that we see Mithai Lal assiduously anointing with water from the Ganges in one of the early sequences). However, this blissful aspect of Shiva as Creator exists in necessary assocation with his complementary aspect as Destroyer, as represented in Varanasi by the cremation grounds. ⁵³ Armed with this understanding, one would surely have viewed the film in a very different way.

There is, admittedly, an oblique reference to the interdependence of creation and destruction in Hindu thought in the card immediately adjacent to the main title of the film, which offers a sombre verse from W. B. Yeats's translation of the Upinashads. But much more striking and immediately apprehensible is the sequence that leads up to the title and which culminates in a particularly vicious dog-fight. This suggests anything but bliss and has led many reviewers to assume that the title is ironic. Whatever Gardner's precise intentions here, it is clear from the entry in his notes about the rushes that he was perfectly well aware that Varanasi is viewed as a place of ‘exultation’ by Hindus. So one can only assume that he intentionally decided to give priority to his own very much darker personal view of the city as a place where, as with the world in general, nothing lasts for ever and all living beings end up being either burnt or eaten. ⁵⁴

Both these points relate to a third, namely, intelligibility. There can be no doubt that Forest of Bliss is a challenging film, dense with meaning, which is very difficult to comprehend in its full complexity, particularly on the basis of a single viewing. The detailed interpretations that I have offered here have been made possible not only by repeated viewings, but also with the aid of the conversation between Gardner and Östör which offers a series of vital clues as to the intentions underlying the many diverse elements of the film. As a viewer without any specialist knowledge of Hinduism, my understanding of the film was also greatly aided by the key provided by Chopra in her review of the film, in which she immediately identifies ‘a central principle of Hindu thought – the juxtaposition and interpenetration of opposites’. It is surely no coincidence that she was able to do so, being Indian herself and therefore, one assumes, exposed to these ideas since childhood. But without these extra-filmic aids to interpretation, most audiences will inevitably miss a great deal in viewing the film. Certainly, my experience of showing this film year upon year to students of visual anthropology, also without any specialist knowledge of Hinduism, proves that while they are usually entranced by the film, they generally glean only a small fraction of its significance.

Gardner is far from unique in grappling with this problem since it confronts all ethnographic film authors who make a film about a subject that is culturally unfamiliar to their audience. Nor is there an easy solution. While some audiences may be relied on to make connections, others will not; the same applies at the level of the individual too. If one provides insufficient means for understanding the cultural significance of what the film portrays, then it is likely that audiences will fill that vacuum of incomprehension with their own, usually ethnocentric, interpretations. If, on the other hand, one burdens a film with too much explanation, which will usually take a verbal form, one runs the risk of turning it into an illustrated lecture, thereby undermining the distinctive capacity of film to communicate an experience, even if vicarious, of the reality portrayed. Indeed, too much verbal explanation can obscure rather than clarify, as another ethnographic film about Hindu cremation practices, Releasing the Spirits , made by Timothy Asch and his colleagues in Bali, around much the same time as Forest of Bliss , demonstrates in a particularly cautionary manner. ⁵⁵

Given these difficulties, one cannot but respect the sheer boldness of Gardner's attempt in Forest of Bliss to provide explanatory frames of reference for abstract religious concepts through a combination of non-verbal symbolism, metaphorical association and montage. Equally, however, one cannot escape the conclusion that Gardner's refusal to provide more explicit devices that would allow the audience – on the basis of viewing the film alone and without the benefit of his after-the-fact exegesis – to connect the practices shown in the film with the ideas and relations that underpin them necessarily means that its value as a specifically ethnographic film is diminished. Even so, it is still possible to admire the praxis that underlies it. Indeed, I see no reason why Gardner's distinctive mode of film authorship, albeit more fully anchored in local ethnographic understanding, should not be a source of inspiration to all those ethnographic film-makers who continue to harbour ‘anthropological scruples’.

Robert Gardner as a ‘masterful cutter’

Peter Loizos once characterised Robert Gardner's work as ‘experimental’ on the grounds that it goes beyond currently dominant realist orthodoxies. ⁵⁶ Yet while it may be true that Gardner's work holds promise for the future in this sense, it also represents the continuation of an older tradition of documentary film-making. As both critics and admirers have pointed out, there is an echo in Forest of Bliss of the ‘city symphonies’ of the 1920s and 1930s. However, Gardner himself established the connection between his own approach and that of an earlier generation of film-makers more directly, in the form of a tribute to Basil Wright, a leading figure of the British Documentary Movement of the 1930s, who died in 1987, not long after Forest of Bliss had been released.

The account that Gardner offers of Wright's editorial technique in The Song of Ceylon , released in 1934, when Wright was only 27, could almost be a description of his own technique in Forest of Bliss . This technique, he suggested, consisted of establishing

So is it with Forest of Bliss . For no one who has watched this film with an open mind can fail to be enchanted and transformed by it. By carefully examining the works of the ‘masterful cutter’ who made it, ethnographic film-makers too may learn how to enchant and transform their audiences.