1 Patterns of representation and forms of resistance

The film’s opening unsurprisingly establishes patterns and tensions that the rest of the work will develop or complicate. It begins with a pictorial establishing shot of a southern French vineyard with a large and small tree in the left foreground and two cypress trees in the distance, on a mound behind two polytunnels. The land in between is occupied by pruned vines. A tractor works away in the middle ground (Fig. 1). When the titles appear, Joanna Bruzdowicz’s austere, discordant cello score comes in. As the camera zooms in, we become aware of an agricultural worker gathering cuttings. A cut takes us to a lateral tracking shot, the first of many in the film, which frames him more closely and follows him until something we cannot yet see causes him to stop and look. Another cut takes us to a point-of-view shot of a young woman’s body, wrapped in a muddied blanket and with a stained face. This is our first vision of Mona, an immobilised figure seen through the eyes of a man (Fig. 2).

A crime-scene sequence ensues: two policemen, and various vineyard workers including some North African migrants look at the body. The police make measurements and take photographs. They interview various male witnesses and establish that the young woman was a tramp and was probably thrown into “la compote”, an expression typically meaning stewed fruit but here meaning a mixture including wine dregs. The film cuts back and forth between shots of the police at work and shots of purplish-red wine dregs being cleaned from village walls and a telephone cabin, almost as if, in the context of the body in the ditch, an act of communal violence were being erased. After a cursory gathering of evidence, the police conclude that the unidentified woman has died a “natural” death from the cold. Two shots show her being zipped into a white body bag. It is as if her case too were being zipped shut.

Another cut takes us to a very different scene, the wind-blown ridges on a sandy beach, as an off-screen narrator (voiced by Varda herself) tells us that the young woman’s body remained unclaimed, that she left no trace but that she was called Mona Bergeron and was remembered by those who had met her recently and whose testimony allows the final weeks of her last winter to be reconstructed. As the narrator speaks, the camera pans left and tilts to show a naked young woman emerging from the water in extreme long shot. The narrator tells us that she knows little about her but thinks she came from the sea, a clear reference to the birth of the goddess Aphrodite, also born from the waves (Fig. 3). We cut to an over-the-shoulder close-up of a young man looking at postcards of naked women in a café and cut again to follow him as he joins an all-male group and speaks to a character called Paulo (Joël Fosse), telling him he can have the young woman on the postcard for two francs. But when you saw a real naked woman, you chickened out retorts Paulo. We cut again, a flashback within a flashback, to another over-the-shoulder shot, this time of Paulo and his male companion watching Mona, once again captured and objectified by male eyes, dressing on the beach. Prefiguring Mona’s later actual rape, Paulo proposes raping her but his companion demurs. Another cut takes us to a zebra crossing. The camera tilts to frame Mona and follows her as she moves left, hitching a lift, in the first of the film’s twelve famous right-to-left tracking shots. As she exits screen left, the shot holds on a stop sign by the road. We cut to inside the lorry cab where a shot-reverse shot construction follows the conversation between Mona and the driver. The area has 90,000 inhabitants in summer but only 3,000 in the winter, the driver says. That makes 87,000 empty beds, says Mona, prompting the driver to ask her suggestively if she is seeking a bed and to evict her from the cab when she responds negatively. We cut to a shot of another road sign. When Mona walks into the shot, the camera again tracks her laterally as she moves from right to left.

We can now pull out some patterns and tensions. Although clearly a fiction, the film knowingly incorporates documentary elements which unsettle the spectator’s understanding of what they are seeing. Varda’s expository voice over is the first example but some subsequent “witness statements” about Mona are given documentary-style to camera, even as others emerge organically within conversations among the characters themselves, as in a fiction. Complicating things further, one character, Paulo’s girlfriend, Yolande (Yolande Moreau), breaks the fourth wall from within scenes with her employer, Aunt Lydie (Marthe Jarnias), to talk about the effect Mona’s apparent romance with another vagabond, David (Patrick Lepcynski), a self-styled “wandering Jew,” has had on her. Because the film presents itself as an investigation in this way, we might expect it to arrive at some truth about Mona, albeit a very different one to the simplistic one that the police reach. But, as scholars have noted, the witnesses reveal more about themselves than they do about Mona (Hayward 2000: 270). The men, especially when in male groups, often objectify her, following the pattern set by Paulo and the police in their different ways in the opening scenes. The women are more likely to project their own desires, frustrations or sense of what a woman should be onto her (Smith 1998: 121-9). A young woman on a farm, for example, envies her apparent freedom but her mother points to her precarious existence. An older woman similarly finds Mona’s freedom better than being badly married, as she pointedly tells her husband. A male garage owner in contrast suggests that Mona, like other women tramps, is a slut and layabout. Jean-Pierre (Stéphane Freiss), the agricultural engineer, is afraid of her, he confesses, because he is disgusted by her. Showing that not all negative reactions come from men, his up-tight wife (Laurence Cortadellas) suggests that she is a runaway who may be a criminal, a drug addict or mentally ill. She is manipulating, she suggests, Jean-Pierre’s colleague, Mme Landier (Macha Méril), the plane tree specialist, who is a sucker for good causes. Mme Landier’s reaction is actually more nuanced. She is taken aback by Mona’s dirt and smell but still accommodates her in her car for an extended period, feeds her and is tormented by guilt when she leaves her at the edge of some woods.

Sandy Flitterman-Lewis, one of the film’s leading feminist interpreters, notes how, with its multiple divergent witness statements, the film constitutes its heroine not as an object of knowledge but as a resistant enigma. Trying to make sense of the elusive Mona, the critically engaged spectator is required to participate in the process of meaning construction so that the socially constructed nature of femininity is brought to the fore. At the same time, by denying any fixity of meaning and appealing to an active spectator, Varda herself refuses any position of directorial mastery (Flitterman-Lewis 1996: 244-5). “Although it offers no positive female role model,” Flitterman-Lewis concludes, “the film represents a highly productive feminist cinematic alternative” (Flitterman-Lewis 1996: 244). For Flitterman-Lewis and other feminist critics then, Mona’s position as vagabond is less central than the way in which her radical non-conformism (her dirtiness, smelliness, refusal to be gracious or grateful, stubborn pursuit of freedom and rejection of domesticity) serves as a catalyst which drives others to express their view of what a woman could or should be. And yet, Mona’s vagabond status also seems to have played an important in the film’s genesis. As Kelley Conway notes, Varda was inspired politically by press reports about the new poor (les nouveaux pauvres) circulating in the 1980s and was more specifically moved by the fact that people were still dying of exposure in the modern world (Conway 2015: 64). The director initially envisaged a plot with “a male protagonist that would explore the interactions of eight people, some of whom were drifters.” An encounter with a young female drifter, Setina Arhab, who would have a cameo as Setina in the film, then led her to centre the film on a female vagabond (Conway 2015: 65). Hayward is familiar with the film’s genesis but argues that the proliferation of points of view of Mona prevents us reading the film as a conventional narrative and stops it simply being “an ideological film about vagrancy” or about les nouveaux pauvres (Hayward 2000: 272). The film’s principal interest is its status as feminist filmmaking and challenge to gendered representational and narrative codes (Hayward 2000). Yet, as Conway’s account of the film’s genesis suggests, Mona’s vagabond status was not simply a representational trope to force gendered attitudes into view but was important in its own right. The film was certainly feminist but also a response to newly visible forms of marginality. Its challenging proliferation of viewpoints, refusal of a stable or knowable truth and indeterminate position between documentary and fiction do not prevent it from being a film about emergent precarity unless we think that all such films should be conventionally realist.

2 Narrative fragmentation, everyday survival and fate

An essential aspect of the film is its refusal of conventional narrative organisation. As Flitterman-Lewis notes, it has a radically fragmented episodic narrative with repeated temporal dislocations (Flitterman-Lewis 1996: 243). Hayward comments on its digressive temporal construction and interweaving of interrupted flashbacks (Hayward 2000: 272-3). Conway similarly notes its fragmented and episodic construction. Short scenes show Mona’s encounters with strangers on the road or where she sleeps. Longer scenes chronicle encounters with individuals more significant in her story (David, her dope-smoking temporary lover; the goatherders who offer her land to cultivate; Mme Landier; Assoun, the Tunisian migrant worker who welcomes Mona into his accommodation and teaches her to prune vines). These scenes are preceded or followed by “testimonies” from the different characters (Conway 2015: 58-62). One extended scene in Nîmes railway station shows several major characters meeting or narrowly missing each other in a reminder of the original idea for a multi-character story (Conway 2015: 61). But the famous interspersed right-to-left tracking shots which typically begin without Mona and continue after she has left the shot cement her elusive centrality and provide a sense of continuity in the otherwise discontinuous narrative.

Homing in on the character’s mobility, Conway suggests that “Varda wanted viewers to feel how a woman on the road survives day to day” (Conway 2015: 67). In the director’s words, as reported by Conway, the film asks, “Where does one sleep? What noise do boots make on pavement, in a field, on gravel?” (Varda cit. Conway 2015: 67). The film, Conway notes, wants less to construct a coherent cause-effect narrative with a goal-driven character than “render concretely, in a highly patterned fashion, the material events that make up Mona’s existence” (Conway 2015: 59). In keeping with this observation, the film foregrounds repeated actions which relate to Mona’s material needs: being evicted from vehicles, buildings or camping locations; seeking shelter and trying to keep warm in the winter landscape; finding or eating food and asking for water; hitching; carrying out casual labour for ready cash. There are also scenes of lying around, listening to music, looking at pictures and so on. Mona is not narrowly defined by her material needs even if these recur constantly.

Because these activities contain no inherent forward movement, we might consider Mona’s story to be essentially directionless. But this aspect of the film competes with another, Mona’s movement towards death which is inscribed in the film’s overarching flashback structure but is also present in other elements of it. Hayward argues, for example, that the repeated right-to-left tracking shots following Mona serve as a metaphor for death in implicit contrast to more conventional left-to-right ones which would suggest an opening to the future (Hayward 2000: 272-4). Adding evidence for this reading, the tracking shots are accompanied by Bruzdowicz’s mournful cello score with the music becoming faster and more threatening as Mona approaches her end. Tellingly, two sequences occur by a cemetery, the first when Mona inadvertently pitches her tent by one in the dark, the second when she talks briefly to a prostitute seeking business on the road outside one. Death also stalks the natural world in that Mme Landier and Jean-Pierre are respectively studying and culling parasite-infected plane trees. At the same time, there is a degradation in Mona’s appearance as, after her initial pristine emergence from the sea, she becomes dirtier, and her boots disintegrate. The winter landscape is harsh and inhospitable, a constant mute indication of the difficulty of life for those on the outside who face repeated evictions, as Mona does. But it is not until near the end, when she loses her tent to a fire in an urban squat and is soaked with the earlier mentioned “compote” during the village wine feast at the hands of the male bacchanalian revellers, that her fate seems sealed. The film very skilfully balances the temporality of routine repetition (Mona’s material existence), the contingent time of the accident (the fire, the soaking) and the inexorable time of fate. But in the end, not least because of the overarching flashback structure, it is the latter which prevails. The film seems marked by a profound pessimism. There is a danger that this mood can be downplayed in some feminist readings which justifiably see the film as a triumph of feminist filmmaking. But other critics do emphasise it. At the end of her analysis of the film, Jill Forbes writes, “it is thus impossible not to read Sans toit ni loi as a highly pessimistic vision of women in society and in cinema” (Forbes 1992: 94). Echoing this judgement Carrie Tarr and Brigitte Rollet comment rather darkly: “its [the film’s] box-office success suggests that spectators recognised its representation of a woman’s escape from patriarchal control only through utter solitude, indifference and death” (Tarr and Rollet 2001: 236). This pessimism can only be explained if we replace the film in its context(s) to which I now turn.

3 A film in context: the post-1970s world

In his recently published The Subversive Seventies (2023), Michael Hardt rejects what he calls “standard accounts” which portray the 1960s positively while seeing the 1970s as a time of retreat and defeat scarred by the political violence of leftist groups who had lost their way, and by repressive state and rightist forces (Hardt 2023: 2-7). The 1960s, he explains, were marked by an accumulation of struggles “including colonial and anti-imperial struggles, revolts against state socialist regimes, insurgencies against racial domination, industrial worker rebellions against factory discipline, and more” (Hardt 2023: 10). These struggles made “the ruling global framework of power as well as disciplinary regimes in each country increasingly ungovernable, and precipitated a series of social, economic, and political crises in the early 1970s.” While they still carry lessons for us, they belong effectively to the past, not least because of the changes they drove. In contrast, the 1970s mark the beginning of our time with the emergence of post-industrial society, the shift to post-Fordism and the neoliberal wave of privatisations and attacks on welfare (Hardt 2023: 10). In response to these radical shifts whose effects still structure our present, the progressive and revolutionary movements of the 1970s were forced to develop new and still relevant theoretical models and modes of action (Hardt 2023: 10-11). Despite their great diversity and geographical spread, they converged, Hardt argues, on some core concepts: autonomy, as expressed in a refusal of hierarchies and authority but also of being spoken for; multiplicity and the end of the presumed centrality of industrial workers as vanguard agents of historical change; a democracy of participation rather than one of representation. If these concepts could be knitted together, it was around liberation as a “master concept,” which implied a transformation of self as well as of structures (Hardt 2023: 7-9). “The seventies,” comments Hardt, “may be the last period in which numerous large-scale social and political movements had the audacity and confidence to aim unequivocally at liberation and put in practice the means to achieve it” (Hardt 2023: 9).

Hardt’s discussion of autonomy and multiplicity are of particular interest here. Italy, Hardt notes, played a groundbreaking political and theoretical role which would anticipate developments elsewhere and which can be analysed across three stages. In the first stage, in the late 1960s, there is a peak of industrial worker militancy which coincides with “the autonomy of a large segment of workers from the unions and the traditional parties; and the capacity of autonomous workers’ movement to spill out into the city and, thus, to serve as a base and focal point for general social rebellion” (Hardt 2023: 111). Although diverse social actors are involved, workers and the factory are still considered central. Two key principles being fought for are the refusal of work (liberation not “of” work but “from” work), and workers’ autonomy, not as mediated through parties or unions but as enacted through direct participation (Hardt 2023: 113-4). In the second stage, after the end of the Fiat factory occupations of 1973, there is a realisation that, if a more general social rebellion is sought, the focus must shift from the factory to the social factory (as generated by the increasing capitalist commodification and exploitation of everyday life) and social reproduction cast more broadly. Traditional conceptions of the vanguard role of a largely male led industrial working class are therefore no longer tenable (Hardt 2023: 116-7). In a third stage, once the industrial worker has been decentred, “diverse figures of labour (most of which have long existed) came into full view: service workers, care workers, unpaid domestic labourers, precarious and informal workers, those engaged in legal and illegal forms of work, and so on” (Hardt 2023: 120). The question then becomes not how to organise autonomy in the singular but across a non-hierarchical constellation or multiplicity of networked autonomies (Hardt 2023: 118-22). As Hardt observes, to those attached to the leading role of the industrial working class, “the heterogenous subjectivities mobilised by Autonomia appeared as something like the lumpenproletariat – chaotic, morally suspect and politically dangerous.” Similarly, “to those dedicated to vanguard party structures, these experiments in decentralized and network organisation could only appear as hopelessly incoherent” (Hardt 2023: 122). Nonetheless, the need to think the question of the non-hierarchical articulation of multiplicities would be a core one in the 1970s, notably for feminists and for black, antiracist and Third World activists (Hardt 2023: 149-80).

How does this help us make sense of Varda’s Vagabond? I would argue that the questions of liberation, autonomy, escape from labour and multiplicity, and the need for the articulation of different struggles, all associated by Hardt with the 1970s, float over Varda’s film but without the horizon of hope that still seemed available in that recently ended decade. One of the few things we learn about Mona’s previous existence is that she has worked in an office but taken to the road to escape the control of lower managers. She seeks autonomy and an escape from all forms of labour discipline. She works only casually for cash for immediate needs. But her refusal goes beyond labour to embrace domesticity and the consumer lifestyle that goes with it more broadly. She certainly challenges gendered norms through her behaviour, attitudes and appearance but, without losing its gendered dimension, her challenge is more multidimensional than that. It is a challenge to the consumerist, productivist and reproductive social norms of her time, a social strike. The problem, however, is that her struggle for autonomy is enacted on a hostile physical and social terrain. She is helped by multiple fleeting or longer lasting alliances with a diversity of others – a multiplicity in Hardt’s terms – but none of the alliances are durable enough – sufficiently tightly articulated – to help her survive in the long term.

The alliances are worth unpicking. They are with both women and men and cut across lines of class, ethnicity and profession. Mme Landier, who welcomes Mona in her car and lets her sleep in it overnight, is a university professor. Yolande, who brings Mona back to her employer’s house to sleep, is a housekeeper. Other women offer her brief moments of sympathy or support: the farmer’s wife who helps her pump water; the prostitute who chats with her and buys (stolen) cheese from her (Fig. 4); the nuns who feed her; Aunt Lydie, the old woman, with whom she enjoys alcohol-assisted hilarity. This is a diverse assortment of women from different social classes and with varying degrees of respectability.

There are also several men: there is, first, David, the hippy-like “wandering Jew,” with whom she stays for a time and whose dope she smokes; there is the man with a masters in philosophy who has returned to the land with his partner and child to raise goats and who offers Mona a plot to grow potatoes but expects labour discipline in return; there is, finally, Assoun, the immigrant Tunisian vineyard worker, who teaches her to trim vines and welcomes her into his accommodation until his Moroccan co-workers return and demand she leave. One could perhaps add the mixed group of squatters who hang around in Nîmes station drinking and begging or extorting money. Mona stays with them until, unseen by her, an angry David bursts in, demands money he is owed, and causes the fire that separates Mona from her tent and sleeping bag. The alliances range from the mainstream to the countercultural, from the middle class through the service class to the voluntarily unemployed, and from the salaried through the self-employed (the goatherd) to various lumpen or outsider groups (the wandering hippy, the squatters, the prostitute, the immigrant labourer). But, partly due to the prejudices that Mona forces into view, and partly due to her own refusal to compromise, or that of others, none of the alliances lasts. The film shows multiplicities that seem to have potential for articulation but also the failure of that articulation.

A useful contrast can be established with Varda’s earlier One Sings, the Other Doesn’t (L’Une chante, l’autre pas, 1977) with its story of the long-lasting friendship of two women from very different backgrounds and contrasting trajectories, starting in 1962, picking up in 1972 around the pro-abortion Bobigny protests, and continuing for several years thereafter. Despite both characters going through very difficult times, they are ultimately able to shape their destinies in ways they find satisfactory because of their own determination, their unshakeable friendship and a broader context of feminist solidarity and activism (Smith 1998: 107-14). The contrast with Vagabond, a film from only seven years later, is striking. Certainly, the later film shows a multiplicity of potential alliances where the earlier film is more narrowly about women supporting each other. But the admirably durable solidarities of the earlier work give way to purely ephemeral ones in the later one. Vagabond, as we have suggested, carries many of the aspirations of the 1970s into the following decade but only to show how, without a nurturing broader terrain and the necessary solidarities, they cannot prosper.

If it looks back in this way to the 1970s, Vagabond also anticipates important aspects of the cinema of precarity to come. To begin with, without constituting a magically exact compendium of the social types that will be found across precarity films, it anticipates their multiplicity and the difficulty films still have in imagining the articulation of their different struggles (O’Shaughnessy 2022: 35-7; Sticchi 2021: v-vi, 18-21). Secondly, it prefigures a thematics and iconography of precarity around food, cold and shelter. Thirdly and relatedly, it anticipates the way in which, with the withdrawal or failure of mediations, social protections and collective solidarities, socio-economic or gendered violences will impact directly on the body in its vulnerability and exposure and the immediacy of its needs (O’Shaughnessy 2022: 34). Fourthly, it looks forward to the temporal economy of precarity films where the future is effectively foreclosed and characters are stuck in a decaying or impossible present, a temporal economy which Lauren Berlant tellingly encapsulates in the term “impasse” (Berlant 2011: 199-200; O’Shaughnessy 2022: 34). Finally, and with reference to a smaller but still significant body of films, it anticipates a cinema where women at the social margins becomes key embodiments of precariousness in contrast to an earlier social or political cinema where it was typically the male worker who embodied the subaltern status of the working class. In all these ways, while still being marked by a difference residing in its ability to carry forward earlier aspirations, even when they are condemned to frustration, Vagabond anticipates the cinema of precarity. Yet, even if it sits between the 1970s and the 1990s in this way, and speaks eloquently to its immediate context, it also evokes a much wider temporal sweep, partly because of the historical reach of the figure of the vagabond and partly because of Varda’s own stylistic choices and referential depth of field.

4 The film’s historical depth of field

The historical range of Varda’s film is perhaps best brought out by Florianne Wild. Wild begins by noting that, “the landscape she [Varda’s Mona] traverses is recognizably [the] France of the 1980s.” But she immediately adds, “if the passage of Mona, the nomad, situates events in the here-and-now, much of the film pulls us back to the archaic” (Wild 1990, 92). Mona’s frozen body, she notes, is found next to a prehistoric tumulus and suggests an ancient statue being found (Wild 1990, 92). Showing the multiple encounters of an itinerant character, the film’s narrative structure, “bears comparison with that of the picaresque novel,” a form with its heyday in the early modern period (Wild 1990, 93). The film repeatedly references classical mythology and its pictorial and sculptural representation: this is most obvious when Mona’s “birth” from the sea evokes Aphrodite being born from the waves and Botticelli’s famous painting of the scene (Wild 1990, 93). But it is also implicitly evident in visual allusions to the myth of the Gorgon Medusa, the mere vision of whose head, with its writhing snakes for hair, could turn men to stone. The first such allusion comes when we see Mona looking at the various pictures which she carries with her and in particular one which Wild identifies as the Gorgon (Wild 1990: 99-100). The second occurs when Mme Landier sends Jean-Pierre to inspect Mona. When the latter gets out of Mme Landier’s car, her matted hair is “whipped by the Mistral [the region’s notorious wind] into snake-like strands” (Wild 1990: 100) (Fig. 4). Jean-Pierre later tells his wife that he has been terrified by this “fury”. The visual allusions to Medusa evoke both a history of male fear of and violence towards the powerful woman and an affirmative feminist reclaiming of her as a figure of joy and beauty, notably in the work of famous French feminist, Hélène Cixous (Cixous 1976: 884-5; Wild 1990: 99).

Male violence towards women, albeit in carnivalesque form, returns towards the film’s conclusion when Mona is attacked by men celebrating a village wine harvest, a sequence at once evoking the Roman Bacchanalia and, as Varda notes in a documentary on the film’s shooting included in the Artificial Eye DVD extras, a more specifically local festival, still in existence when the film was made, but with apparent roots in the Middle Ages (Varda et al. 2003). Varda evokes the same period in an interview about the film in Cineaste. She says: “We are such a sophisticated civilization in so many ways, yet, for some of us, it’s still like the Middle Ages. It’s amazing to me that, every winter, some people who do not have homes have to live outdoors… And, every winter, some of them die” (Varda and Edelman 1986: 20).

This sense that a past which should have passed has returned to haunt the present is felt in the film’s narrative and visual economies. It is perhaps most obvious in the narrative when Mona goes to a convent, as a medieval vagabond might have done, to ask for and receive food. It is there more diffusely in the way she repeatedly comes up against doors, walls and fences in a repetition of an almost timeless encounter between the homeless and enclosed spaces. It is also implicitly and increasingly present in the mise-en-scène as the film moves towards its close. As Mona’s boots disintegrate, and the bright red lining falls around her ankles, they look less like modern footwear, Varda notes, than medieval poulaines or pikes (Varda et al. 2003). Similarly, when she loses her tent and rucksack and wraps herself in a blanket to keep warm, it is as if she had been stripped of modern accoutrements and plunged into earlier times (Fig. 6). When, pursued by the men involved in the Bacchanalia, she seeks refuge in a telephone cabin, she is simultaneously in a place of myth, a pre-modern place, and a space of technological modernity. It is partly this complex temporal inscription that allow the film to activate a rich web of allusions. But it is also partly the multi-perspectival way in which each person that Mona meets has their own version of her as romantic, free, vulnerable, lawless, disreputable, sexually available or lazy that enables the film to activate a broad spectrum of historical responses to the vagabond figure. Mona is certainly a contemporary figure and an anticipatory one, but also evokes a whole history of vagabondage and responses to it and demands to be situated in that longer context.

5 The vagabond as lumpen and as anticipatory figure of refusal

Vagabondage has been the object of severe, sometimes murderous repression in Western societies since the late Middle Ages and was only decriminalised in France in the 1990s (Brodiez-Dolino 2022: 3-4). Typically stigmatised, the vagabond has also been represented by writers “as a romantic figure of freedom and critique” (Cresswell 2010: 247) and celebrated by anarchist artists, for example, for its radical non-conformism and commitment to individual liberty (Hutton 1990). For Marx, in contrast, the vagabond belonged to the lumpenproletariat, a group whose analysis evolved over time without ever becoming favourable. For the early Marx and Engels of The German Ideology (1845-1846), the lumpen were “an ahistorical class of vagabonds, criminals and homeless who are not historically determined by any particular epoch” (Villanova 2021: 476). Existing outside the dialectic, they were at best a residual class and at worst a group capable of acting against the interests of the proletariat (Villanova 2021: 476-7). By the time of Capital (first published 1867-1894) and the later Marx, they were resituated as a reserve army of unemployed labour used to cow the working class and existing outside of the capitalist production of value (Villanova 2021: 481). The most famous formulation of their role, however, was articulated in the Eighteenth Brumaire (1852) where Marx sought to account for Louis-Napoleon’s ability to seize dictatorial power in 1851. He cast them as a disreputable yet dangerous amalgam of disparate groups including vagabonds. They lay outside the dialectical collision of classes defined by their position in the relations of production but were susceptible to political mobilisation in the service of reactionary forces (Villanova 2021: 479-80). At the same time, and in a way which durably marked Marxism, the very proliferation of sub-groups attributed to the lumpenproletariat and the “hysteria of naming” it called for allowed for the retention of a cohesive and historically progressive understanding of the working class (Stallybrass 1990: 72-81).

By the 1970s, as Hardt told us, it was becoming increasingly difficult to subscribe to a teleological account of history or one where the organised Western working class was the privileged agent of historical change, partly because of the diminishing power of that class, and the sense that it had been seduced by consumer capitalism, partly because feminism, Black liberation and anti-colonial movements made the exclusions that maintained its centrality untenable. The multiplicity of non-industrial workers could no longer simply be bundled up and dismissed as lumpen, as Hardt also noted. With this questioning of established hierarchies, institutions and narratives, significant historical and theoretical attention could shift towards previously neglected groups such as prisoners or vagabonds (Brodiez-Dolino 2022: 6). It was in this context that Michel Foucault produced his Surveiller et punir: naissance de la prison (Discipline and Punish: Birth of the Prison, 1975) and Silvia Federici wrote Caliban and the Witch, a book whose gestation began in the mid-1970s and whose original Italian text appeared in 1984, a year before the release of Varda’s film. Vagabonds play important roles in Foucault’s and Federici’s arguments.

In Surveiller et punir, Foucault developed a genealogical account of a new paradigm of social governance, the disciplinary society, emerging in the latter half of the eighteenth century. Before then, he notes, illegal behaviours occurred at all levels of society and tended to revolve around rights and the bending of feudal laws. Vagabondage was an escape route for workers fleeing their masters, mistreated apprentices, and those escaping military service. Even though it merged into sometimes violent criminality, it was often tolerated by the popular classes in the context of their own survival-driven recourse to outlawed practices (Foucault 1975: 98-100). When prosperity increased in the second half of the eighteenth century and especially after the Revolution, when land ownership became private and exclusive, illegal popular practices carried out by vagabonds and others no longer transgressed the limits of rights but threatened property and its protection (Foucault 1975: 103). As the mechanism for appropriating wealth shifted from feudal dues to the exploitation of labour, the production of a disciplined workforce with docile bodies became primordial. These bodies had to respond to the minutiae of rules and feel under constant surveillance (Foucault 1975: 159-199, 228-264). A series of sites of enclosure (prison, factory, school, convent) played an essential role in their production (Foucault 1975: 161-175). In its abhorrence of unproductive or recalcitrant bodies, such a regime necessarily required the large-scale confinement of vagabonds and paupers (Foucault 1975: 166).

Federici’s analysis converges with Foucault’s in important areas. She also sees historical vagabondage as an essential form of resistance to hated waged labour amongst dispossessed peasants (Federici 2021: 104). She similarly foregrounds the disciplinary regimes applied to the body which needed to be made intelligible and accessible to labour exploitation (Federici 2021: 148). She is, however, profoundly critical of Foucault’s blind-spots: his attention to the deployment of power rather than its sources and focus on an abstract asexual subject rather than a gendered one (Federici 2021: 8-10). Her timeframe is different. Foucault’s disciplinary regime emerges in the mid-eighteenth century. Her time of transition is the sixteenth and seventeenth centuries when Europe and its colonies faced a demographic crisis (Federici 2021: 93-102). The response in the colonies was slavery and, in Europe, the rise of a gendered division of labour and a rigid split between production and reproduction, previously inextricably intertwined (Federici 2021: 77-8). While expulsion from the land hit both men and women, their situations were profoundly different. Women found waged labour much harder to access, even as the money economy became predominant, and with vagabondage being less available due to misogyny and male violence, had fewer escape routes (Federici 2021: 76, 110). Prostitution was one possibility but was subject to growing repression. Domestic service occupied on average a third of the female population in England, France, Spain and Italy (Federici 2021: 112). But increasingly, the separation of spheres and the confinement of women to the sphere of reproduction, and their subordination to men, became widespread norms. An assault on women’s bodies, social power and knowledge, the witch hunt was a veritable war against women which coincided with the time of the enclosures. Its terror played an essential role in the installation of the new patriarchal order (Federici 2021: 175).

This brief discussion of Marx, Foucault and Federici as they respond to the vagabond and the lumpen more generally helps us see how Varda’s Vagabond resonates in a longer history. With her refusal to sacrifice her freedom and accept labour discipline, Mona evokes a memory of refusal of the kind described by Foucault and Federici. Her bodily indiscipline, mobility and messiness similarly embody a refusal of the orderly, productive body disposed to labour discipline whose genealogy Foucault and Federici describe. Yet at the same time, her experience of social stigma, sexualised objectification, and threatened and actual sexual violence echoes the historical experience of female vagabondage and the gendered obstacles it faced as described by Federici. Interestingly, Mona enjoys brief solidarities with a prostitute and a domestic servant, two female figures already significant in Federici’s account of the pathways available to proletarian women expelled from the land and denied other employment opportunities. It seems that certain things don’t change. Yet the fleeting nature of these solidarities and, more broadly, those Mona enjoys with other marginalised or simply marginal figures (the hippy, the libertarian goatherd, the squatters, the migrant) has potentially more negative historical echoes of Marx and his accounts of the lumpen as an amalgam lacking the unity and collective consciousness to be seen as an agent of historical advance.

Also taking a sharply opposing line to the classic Marxist position, bringing us up to the present, Kathi Weeks argues that the lumpen should be seen not as having less potential for leading progressive change than the industrial proletariat but in fact as having more. Like Foucault and Federici, she sees the splitting of a stigmatised lumpenproletariat from the proletariat as the outcome of concrete socio-historical processes and not as something simply assumed (Weeks 2023: 327-329). She also agrees with Federici’s critique of historical Marxism’s prioritisation of production over reproduction, seeing it as both analytically misguided and politically counter-productive in the way it divides capitalism’s antagonists and “serves to uphold the twin ideological pillars of capitalism’s dominance: the work ethic and the family ethic” (Weeks 2023: 332). She argues, more broadly, that “historical processes of proletarianization were inseparable from the processes – including, among others, housewifization, racialization, criminalization and disabilization – by which lumpenproletarians were produced as the disavowed cast offs of the working class” (Weeks 2023: 330). The historically dubious proletarian-lumpen distinction is now simply untenable, given the “neo-liberal hodge-podge of increasingly precarious labour contracts, rise of service labour, and more extensive confounding of what is productive and what is reproductive.” She adds that the “itinerant, informal, and occasional workers” traditionally labelled as lumpen are now “increasingly standard” (Weeks 2023: 332). This is most obviously so in the global south but is also true if we are to understand the broader functioning of Western capitalism (Weeks 2023: 331). Confounding assumptions of their political passivity, different lumpen groups have shown their ability to mobilise (Weeks 2023: 334). The very disreputability that once appeared to disqualify them in Marx and Engel’s eyes now makes them better candidates for political radicalism (Weeks 2023: 335-9). Challenging dominant norms of productivity and reproduction, and escaping the privatised family’s containment of erotic, social and political desires, vagabonds are prototypes in this respect (Weeks 2023: 338-339).

6 Conclusion

We can appreciate now how Varda’s Vagabond, a film that looks back to the 1970s and forward to more contemporary times, also operates in a longer timeframe, carrying a memory of vagabonds as figures of historical refusal, ostracization and gendered oppression, yet also reminding us of their capacity to anticipate challenges to the work ethic (with their recalcitrant bodies, defiant mobility, and refusal of productivity) and compulsory domesticity (again through their bodies and refusal of enclosure) and more broadly to foreshadow the long denied capacity of the lumpen for radical challenge, as described by Weeks. Yet, lest we get too far ahead of ourselves, we should remember that, in its portrayal of only passing alliances, and its overarching pessimism and sense of a future foreclosed, the film also bears the mark of its 1980s context. It could indeed be argued that its temporal reach owes much to this same context. The film comes from a period when post-1945 protections were fading and a future of widespread precarity was emerging with inevitable echoes of earlier insecurities. Its transitional moment helps it look back and forth in time. Yet a film can never be explained mechanically by its context. The historical depth of field and referential reach of Varda’s film also owe much to specific creative choices: the way that it straddles the pre-modern, the mythical and the contemporary in its narrative and visual economies; the manner in which it is able to evoke a historically resonant range of responses to the female vagabond through its multi-perspectival construction. Its power, as a feminist masterpiece, has much to do with its use of the female vagabond figure to render oppressive gendered structures visible and to problematise social constructions of femininity. But its use of the same figure also speaks powerfully to an emergent precarity and how we might make sense of it.