Duchamp, Childhood, Work and Play: The Vernissage for First Papers of Surrealism , New York, 1942

In 1942, in a large nineteenth-century Italianate mansion in central Manhattan, a group of émigré surrealist artists led by André Breton mounted the first major exhibition of surrealist art in the United States. Titled First Papers of Surrealism, the exhibition included paintings and sculptures by more than thirty artists accompanied by examples of so-called ‘primitive art’, and was designed by Marcel Duchamp, who living up to his reputation for iconoclasm and conceptual innovation, produced one of the most audacious exhibition installations of the early twentieth century, threading the entire space of the gallery with his so-called ‘mile of string’ (fig.1).1 While this web-like construction produced the succès de scandale that Breton and the surrealists had hoped for, Duchamp’s intervention must have surpassed their expectations, to the extent that it almost obliterated the view of some of their works, which were supposed to be the centre of attention. The way in which the criss-crossing twine prohibited engagement with the art on show might well relate to Duchamp’s ongoing critique of surrealism, which can be discerned in his work of the 1940s and which was picked up by contemporary critics such as Robert Coates, who, writing in the New Yorker, felt that the process of stringing, which must have become repetitive and tiresome by the time Duchamp had finished, commented obliquely on the tiredness of surrealism.2 Duchamp is generally understood to have been sympathetic to surrealism and broadly in tune with its preoccupations, although he was deeply sceptical of psychoanalysis and of the group protocols established by Breton.3 Four years before First Papers of Surrealism he had installed the group’s last exhibition in Paris, Exposition internationale du surréalisme, and, as will be become clear, arguably made use of the free hand he was given in mounting these two major late exhibitions to subtly (or perhaps not so subtly) question the validity of the group’s commitment to social revolution.4

If the installation of the show was dramatic enough, its opening reception, or ‘vernissage’, which took place on 14 October 1942, was even more unusual. Wealthy art patrons and members of New York’s cultural elite milled around, attempting to make what they could of the strange web or net in which they were caught, peering through it to look at the paintings, while a number of children wove in and out of the guests, eventually carving out a space for themselves in the central area of the exhibition. From all accounts the group of children, led by the eleven-year-old Carroll Janis (son of the art collector Sidney Janis), consisted of six boys dressed in baseball, basketball and football attire, who threw balls among themselves, and six girls who played skipping games, jacks and hopscotch. They were under strict orders from Duchamp to carry on playing throughout the event, and to explain, if questioned, that they were playing on Duchamp’s instructions. Duchamp, incidentally, was nowhere to be seen. As was his custom he had decided not to attend the opening.5 What was Duchamp articulating by allowing a group of unruly kids to overrun the exhibition, not only within the context of surrealism, but more broadly in relation to his understanding of the avant-garde and its relationship to the spaces of the gallery or museum? The cat’s cradle-like installation, quite apart from its iconoclastic role in cancelling out some of the paintings, may have a more direct relationship to the children’s games than has thus far been acknowledged. While Duchamp scholars have tended to see the ’mile of string’ installation as alluding to the displacement and disorientation of the surrealist group at this time – not least because most of the artists had only recently arrived in the US after difficult passages out of wartime Europe – the concept of play was obviously central to the opening of First Papers of Surrealism.6 What this paper examines, then, is the extent to which children’s play and childhood were central preoccupations of Duchamp’s artistic milieu in New York during the latter half of 1942.

Duchamp, Cornell and childhood

Duchamp, of course, had always been interested in play. Toys or allusions to toys have a role in his early work, for instance in the drawing titled Walking Doll (Bébé marcheur), seemingly based on a product available during the period, which was published in the illustrated journal Le Courrier français in January 1910. A form of mock-advertisement, the image contains a list of the doll’s attributes, which concludes with the assurance that ‘she undresses completely’. A caption accompanying the illustration asserts that the product is from a ‘catalogue for elderly gentlemen’, leaving little doubt that this is a risqué allusion to paedophilia and prostitution.7 As the art historian Arturo Schwarz has observed, this drawing of a doll who undresses appears to prefigure the ‘bride stripped bare’ of Duchamp’s major work, The Large Glass 1915–23 (see Tate T02011). It should be noted that the ‘bride’ who ‘strips’ at the command of the ‘bachelors’ within the complex iconography of The Large Glass (an iconography which has to be deduced from Duchamp’s notes for the work) has other dark, toy-like connotations. The idea of the ‘bride’ seems in fact to have its origins in the wedding tableaux in the country fairs of Duchamp’s youth, where spectators threw balls at the heads of mannequin bridal couples.8 The element of black humour involved here also seems to underpin the bilboquet that Duchamp gave to his friend Max Bergmann as a gift in 1910, and which can be seen as a precursor to the readymades that he produced a few years later. A bilboquet is a traditional French toy consisting of a lathed spike attached with string to a wooden ball. Play involved penetrating the ball with the spike, which seems, in terms of Duchamp’s dialogue with Bergmann, to have been supplied with sexual connotations as a reminder of the amorous adventures shared by the two young artists in Paris.9 It is clear, then, that references to popular amusements and children’s playthings were formative elements of Duchamp’s iconography, and related to this is the fact that Duchamp was deeply preoccupied throughout his life with chess, the pre-eminent game of intellectual combat and strategic forward-thinking, at which he became an international-level player.10 However, in bringing children into the exhibition space in 1942 Duchamp was not just nodding to a preoccupation of his own; he was also responding to the cult of childhood within surrealism, which had manifested itself in various ways in the surrealist art and writing of the 1920s and 1930s.

The surrealist interest in childhood is a massive topic in its own right and can only be touched upon briefly here. Near the beginning of the ‘First Manifesto of Surrealism’, authored by Breton, it is stated that ‘Children set off each day without a worry in the world. Everything is near at hand, the worst material conditions are fine. The woods are white or black, one will never sleep’.11 This late Romantic conception of the child’s uncontaminated imaginative faculties, and Breton’s numerous poetic invocations of the child’s closeness to the wellsprings of the ‘marvellous’, would give way in the 1920s and 1930s to the surrealist fascination with the ‘femme-enfant’, the woman-child who is supposedly closer to irrational forces. It is at this point that the imprint of Freudian psychoanalysis can be detected on the work of Breton and his fellow surrealists, who became compelled to reconcile their idealisation of childish innocence with a darker sense of the child as prey to ungovernable sexual drives.12 The symbolic centrality of the ‘femme-enfant’ was emblematised in a famous photograph of the child prodigy Gisèle Prassinos reading her poetry to a group of surrealists in the 1930s, but the image is ubiquitous in surrealist art. For example, in Max Ernst’s Two Children are Threatened by a Nightingale 1924 (Museum of Modern Art, New York) an adolescent girl from the pages of a Gothic novel dangerously wields a dagger, while in Dorothea Tanning’s Eine Kleine Nachtmusik 1943 (Tate T07346) two adolescent girls appear to spawn disturbances in the natural order, such as the emergence of a massively enlarged sunflower, which seemingly grows from one of them.13 But of all the surrealists the German artist Hans Bellmer is most closely linked with imagery of the sexualised (female) child, the paedophilic associations of his work demonstrating how far the surrealist attitude towards childhood could stretch: from a Romantic veneration of the child’s imaginative openness to a Sadean exploration of its potential ‘perversity’ (the latter of which returns back to Duchamp’s Walking Doll).

Arguably it was the American surrealist Joseph Cornell – with whom Duchamp became especially close in the period leading up to the First Papers of Surrealism exhibition – who most fully exemplified the Romantic pole of the surrealist approach to the child. In Cornell’s object- or tableaux-filled boxes, the most innovative of which were produced in the wake of his discovery of European surrealism in the late 1930s and 1940s, there is a Romantic lyricism, a return to a quasi-Victorian idealisation of the supposed ’innocence’ of childhood, which co-exists with a deeply melancholic sense of loss and a yearning for at-oneness with the child state that can at times verge, in a quite different way from Bellmer, on the proto-paedophilic.14 Cornell’s works, as well as actually utilising children’s playthings – such as in Untitled (Medici Slot Machine) 1942 (private collection), which makes use of wooden blocks – also pay close attention to modes of organisation and miniaturisation that might be seen to be characteristic both of children and the toys that they play with. In 1993 the critic Susan Stewart, discussing the modalities of fantasy at work in the creation of dollhouses, evoked the stunted sexuality and intensely otherworldly private universe that Cornell managed to articulate in his small boxes, few of which exceeded more than two feet square:

Duchamp had himself been concerned with processes of miniaturisation in his own boîtes-en-valise, which he had begun making in 1935, although in this case it was the contents of the boxes – models of his own previous artworks – that were miniaturised (see, for example, From or By Marcel Duchamp or Rose Sélavy (The Box in a Valise) c.1943, Tate L02092). Like Cornell, Duchamp was alert to the toy-like qualities that the miniature versions of his earlier works inevitably possessed, but there was an added twist to Duchamp’s replicas in that they were hand-made versions of what had previously been mass-produced ‘readymade’ items. Once again, Stewart’s reflections illuminate the fantasmatic dimension of the return to the hand-made that was played out by Duchamp and may be exemplified by miniaturised ‘products of mechanized labour’ such as model aeroplanes or automobiles: ‘These toys are nostalgic in a fundamental sense, for they completely transform the mode of production of the original and they miniaturize it; they produce a representation of alienated labour, a representation which itself is constructed by artisanal labour.’16

It would seem, then, that the dialogue between Duchamp and Cornell, as far as the discourse of the miniature and the toy was concerned, turned on their inhabitation of two ‘mental epochs’: Cornell was essentially a Romantic and a symbolist, while Duchamp more properly inhabited the post-industrial moment of the mid-twentieth century. But this is not to deny the obvious impact that one man had on the other. The two had met initially in January 1934 when Duchamp was installing a Constantin Brancusi show at the 57th Street Gallery in Manhattan, and had renewed their acquaintance in June 1942.17 From this point until early 1943 Cornell helped Duchamp construct six of his boîtes-en-valise, and there can be little doubt that the collaboration was mutually beneficial, with something of Cornell’s gentler poetics of childhood rubbing off on the more cerebral Duchamp, and reinforcing his own sense of the artistic potential of toys and games, as he contemplated his First Papers of Surrealism installation.

Cornell, Caillois and ethnographic surrealism

Further aspects of Cornell’s work of the 1940s also help create a context for Duchamp’s deployment of children in the 1942 exhibition opening. Until now Cornell has largely been understood to have been a ‘Bretonian’ surrealist, albeit a cautious and apolitical one, who, given his distaste for Freudianism, felt that surrealism was capable of ‘healthier possibilities’ than it had so far manifested.18 However, it seems highly likely that in the run-up to the First Papers of Surrealism exhibition Cornell would have been aware of an article by the writer and sociologist Roger Caillois titled ‘The Myth of Secret Treasures in Childhood’, which was published in the 1942 issue of the newly inaugurated New York-based surrealist magazine VVV, and which has much in common with Cornell’s work, despite the fact that the two men inhabited profoundly different intellectual universes. Between 1937 and 1939 Caillois had been involved with the College of Sociology in France, a network of French intellectuals spearheaded by Georges Bataille and the former surrealist Michel Leiris, who at the time were gradually moving away from the bourgeois individualism they saw as endemic to Bretonian surrealism, despite its leftist commitments. Caillois shared with fellow collège members a proto-structuralist interest in the analysis of contemporary social life in terms of its underlying mythical determinants and relationship to ritual. However, the central intellectual concerns of the College of Sociology, which were bound up with an exploration of contemporary forms of ‘the sacred’ and an investigation of fascist mentality, were not especially at stake in Caillois’s 1942 essay. Rather it amounted to a poetic account of the way in which children, in accordance with a fetishistic logic, secrete and hoard away ‘magical’ objects of purely personal symbolic value: ‘Treasures are constituted by privileged objects. It is not the sale value of the objects that makes them precious. That is often nil … It is not that they are rare, but that they are coin of another realm … The child therefore keeps the tinfoil that wraps his chocolate bar. He rates “steelies” above all other marbles.’19

Looking at some of Cornell’s boxes of the early 1940s, many of which possess the aura of reliquaries or jewel cases, one finds a remarkably similar set of concerns. Cornell’s L’Egypte de Mlle Cléo de Mérode 1940 (Smithsonian Institution, Washington, D.C.), for instance, was produced as a form of homage to a French ballet dancer of the 1890s, Cléo de Mérode, who was supposedly offered an exotic gift by the Khedive of Egypt if she would consent to visit him. Cornell’s box represents his own version of the prize awaiting Mérode and consists of a set of glass phials and flasks containing sand, jewels, sequins and beads, which poetically evoke the essence of Egypt. In both its form and content the box relates closely to the ways in which Caillois described the function of a child’s ‘secret treasures’:

While Cornell must have read here a confirmation of what he was doing, it is unlikely whether, steeped as he was in Romantic and symbolist poetry, he would have been interested in the finer discursive strategems of Caillois’s analysis in relation to surrealism. Caillois was propounding an account of childhood that was quite opposed to the idealisations of André Breton and even those of Michel Leiris, whose 1938 paper on ‘The Sacred in Everyday Life’ Caillois was partly responding to. Whereas Breton and Leiris upheld the revelatory nature of childish experience in its own right (Leiris saw in his own childish memories ‘the combination of respect, desire, and terror that we take as the psychological sign of the sacred’21 ), Caillois actually saw the amassing of ‘treasures’ and the imaginative flights involved as a means by which the child builds up a strong inner world (ego) as a preparation for adulthood. Cornell would not have been much interested in the adaptive mechanisms of the young mind, but it is fair to assume that the tenor of Caillois’s VVV article helped initiate him into what historian James Clifford has defined as ‘ethnographic surrealism’, an offshoot of mainstream surrealist practice whereby Western social practices are considered to be structurally akin to non-Western ritual practices, and which is primarily identified with Caillois, Leiris and other members of Bataille’s circle.22

In the months leading up to First Papers of Surrealism Cornell was particularly involved with work for the special ‘Americana Fantastica’ issue of the American surrealist-affiliated magazine View, which co-existed alongside the ‘official’ surrealist organ VVV.23 As well as producing an elaborate collage for the cover, Cornell produced a seven-page dossier titled ‘The Crystal Cage (Portrait of Berenice)’ that was inserted in the magazine and which represents a series of visual and textual musings around ‘Berenice’, an imaginary Victorian child-figure who encapsulates a range of Cornell’s associations around the idea of childhood, taking in imagery relating to America, Europe and the Middle East.24 (‘The Crystal Cage’ was also the title of a loose collection of images by Cornell that was stored in a suitcase – similar in style to Duchamp’s boîtes-en-valise – and eventually exhibited in 1946.) Other elements of the ‘Americana Fantastica’ edition also suggest that Cornell had a hand in the editorial process. A double-page spread is designated as a ‘Children’s Page’, taking its lead from puzzle pages in popular children’s books and magazines of the late nineteenth and early twentieth centuries. As well as incorporating a strange collage by Cornell of a child’s doll seemingly covered by litter, the page contains poetic and photographic contributions by the likes of Don Organ and Bern Porter, as well as riddles and vignettes from children’s book engravings reflective of Cornell’s taste.25 All in all, this edition of View gives the impression that Cornell was looking to take stock of a variety of cultural responses to childhood, both by himself and others. This is underlined by the presence in the magazine of a picture by the American photographer Helen Levitt titled Knight in Harlem (fig.2) showing a child dressed up in a makeshift knight’s outfit, which was positioned alongside a reproduction of Cornell’s own Untitled (Medici Slot Machine) of 1942.26 Providing an amusing pendant to the depiction of the Renaissance prince in Cornell’s work, Levitt’s image of a child is in fact highly ambiguous and threatening. Wearing what appears to be a translucent garment, possibly a woman’s stocking, over its head, and unsheathing a wooden sword, the child is hard to identify as either male or female. A few pages further on, there is another equally bizarre photograph by Levitt of a boy in a cardboard hat seemingly shinning up a door lintel.27

Levitt is rarely given much attention within the context of surrealism but her late 1930s and early 1940s photographs of children wearing masks or playing in the streets of New York may be regarded in relation to the work of Eugène Atget and Henri Cartier-Bresson.28 (Levitt had initially seen Cartier-Bresson’s work in the key Documentary and Anti-Graphic Photographs exhibition mounted by Julien Levy at his gallery in New York in 1935 and was heavily influenced by him.29 ) Her photographs of children playing can be indexed not only to surrealist conceptions of the child, as linked to the unconscious and revery, but also to a distinct quasi-anthropological attitude to the phenomenon of childhood that was developing at this time in the US. As the curator Sandra S. Phillips has noted, US psychoanalysts of the period, such as Karen Horney and Harry Stack Sullivan, ‘interwove a child psychology with an anthropology that considered children almost as a different species, with their own development into adulthood’.30 Similarly, in the 1930s two researchers affiliated to the New York Public Library, Ethel and Oliver Hale, collected data on children’s skipping rhymes and games such as hopscotch, and prepared a paper titled ‘From Sidewalk, Gutter and Stoop: Being a Chronicle of Children’s Play Activity’, which trains an anthropological gaze on children. At one point in the unpublished study they write: ‘In many of their games and in much of their play they [children] project imaginary pictures and poses, invent and pretend situations, and, with themselves as the principal actors, create scenes and a world different from their own.’31

Levitt’s presence alongside Cornell in the ‘Americana Fantastica’ edition of View further suggests that American surrealists trained a spotlight on the culture of childhood, and particularly the phenomena of childish games. This can be underlined by a series of experimental films made by Cornell in the 1940s which take children’s party games as their subject. Rarely discussed in the literature on Cornell, this sequence of so-called ‘collage films’ includes The Children’s Party, Cotillion and The Midnight Party, although the films are profoundly interdependent, being composed of roughly the same found footage. This footage, much of which documents a large Halloween party but which also contains images from a circus and of children dancing in a chorus line, was scavenged by Cornell during visits he made to Manhattan curio shops in the 1930s and 1940s and subsequently edited into three films, with extra footage added at undetermined points in the 1940s.32 One short section of The Children’s Party is worth focusing on: after a sequence of images of circus acts, inter-cut with images of partying children, one sees a close-up of a boy wearing a peculiar looking party hat, who, after taking part in an apple dunking ceremony with other children, lifts his face up towards the camera and displays an enormous apple in his mouth. Whether Cornell was conscious of it or not, the vaguely disturbing image of the apple held in the abnormally stretched mouth would have been an inevitable reminder, for an audience schooled in surrealism, of the images of tribal people with masks and facial adornments published in surrealist journals of the 1920s and 1930s such as Documents and Minotaure (not to mention the exoticising images reproduced at the same time in populist magazines).33 At a certain level, then, this film was indeed an exercise in cultural anthropology, with phenomena such as apple dunking carefully documented as examples of Western ritual behaviour. Once again, the tenor of Cornell’s attentiveness to childhood is close to ‘ethnographic surrealism’. It should be stressed that it is unlikely that Cornell sought to interrogate eurocentrism in the way that this particular analogy between Western and non-Western forms of ritual suggests. In line with mainstream surrealist attitudes, Cornell saw the child as an exotic creature whose participation in a distinctively American vernacular culture of party games and circus entertainments provoked nostalgia for the lost land of childhood.

Whatever degree of criticality or sentimentality can be ascribed to these films, they should be seen as part of a shifting, more sociologically sensitive approach to childhood in American surrealist circles, which has rarely been given specific attention. It is this context that has to be borne in mind when analysing Duchamp’s vernissage for First Papers of Surrealism. Indeed, his playful interventions during the opening must partly be understood as a knowing nod to Levitt, Caillois and especially Cornell. Interestingly, the only evidence of the participation of the children in the 1942 exhibition that survives, apart from eyewitness accounts and the testimony of the children themselves, is the catalogue for the show, in which, on the first page, can be read the following succinct account (in French) of what was envisaged for the opening of the exhibition: ‘Vernissage consacré aux enfants jouant, à l’odeur du cèdre’ (‘Opening consecrated to the play of children accompanied by the odour of cedar’).34 It is far from coincidental that this poetic dedication, which can be ascribed to Duchamp, but which has the aura of Cornell’s wistful poetics, is positioned above a short note (in English) on ‘Primitive Art’ that partly serves to acknowledge the collections from which various tribal objects had been borrowed (including those of Max Ernst, André Breton, Claude Lévi-Strauss and Pierre Matisse):

The fact that the natives of the Northwestern Pacific coast head the list is not surprising, given surrealism’s newfound proximity to the region, but it is surely significant that the red cedar tree, or giant cedar, is indigenous to this part of North America, and that Northwestern tribes used this wood almost exclusively in the manufacture of totem poles and masks.36 The significance of cedar and its olfactory qualities will be returned to later, but it is clear that the two statements at the beginning of the exhibition catalogue bring children and so-called ‘primitives’ into structural alignment, indicating that Duchamp was well aware of the prevailing ethnographic zeitgeist.

However, it would not be correct to see Duchamp’s deployment of children at the exhibition opening as straightforwardly in line with the attitude of exoticisation or ‘othering’ that was so endemic to surrealism, and so clearly evident in Cornell’s film The Children’s Party. Duchamp’s encouragement of children’s play chimes well with Levitt’s photographs, with which he would have been familiar, while, at the same time, something of the spirit of Cornell’s ‘Americana Fantastica’ is at stake in the way that Duchamp left the children to play their native games at the vernissage (hopscotch, basketball, baseball), and the event undoubtedly speaks to his friendship with Cornell (whose work was not actually chosen by the selectors for First Papers of Surrealism, but, by letting the children play, it can be surmised that Duchamp included him in spirit).37 However, the element of disruption courted on this occasion by Duchamp would have scandalised the timid Cornell. The French artist was well aware that the playing children would get in the way of the black-tie guests at the opening, and this anarchic dimension is linked to a cooler, more intellectual approach to the deconstruction of art and its institutions.

The element of calculation in Duchamp’s approach should not be overlooked. Just as he can be seen to be in tune with American surrealism, so it is telling that it was Carroll Janis, the son of Sidney Janis, one of the sponsors of the exhibition and a collector of surrealist art (including that of Duchamp), who led the children’s play.38 Duchamp was not above using the event to consolidate his position in an evolving network of American supporters, with whom the European surrealists had little familiarity. One could equally question whether Duchamp had very much actual interest in the children’s play he set in motion. As already noted, he was absent from the opening itself, as was his customary habit, and the children simply played at his orders. A striking element of control is involved, to the extent that, if one thinks of Duchamp’s interest in game playing as being bound up more closely with the ‘adult’ game of chess, the children at the vernissage may have simply been pawns in a pre-conceived performative gambit. With this in mind, it is important to look beyond the context of the surrealist cult of the child and ask what larger point Duchamp may have been wishing to make at the opening of the 1942 exhibition.

Work, play and the avant-garde

To consider this question it is necessary to turn briefly to the Exposition internationale du surréalisme of 1938, held at the Galerie Beaux-Arts in Paris, for which Duchamp created an even more elaborate and disorientating décor than he did for First Papers of Surrealism four years later. While the full complexity of the installation cannot be discussed here, a few highlights from the central exhibiting space or ‘Central Grotto’ of the exhibition should be noted.

In the four corners of this central space, four large double beds were positioned, almost as a direct incitement to the visitors to give up spectating and begin dreaming themselves. Further elements were clearly designed to get in the way of a seamless viewing experience. Spectators found themselves walking in a carpet of sand and dead leaves up to a level of six inches, and, as well as being placed on the walls, the surrealist works of art were displayed on revolving doors from department stores, which were poignantly static. To compound the disorientation, Duchamp hung 1,200 sacks of coal over the heads of visitors, which, although actually stuffed with newspaper, apparently leaked coal dust and must have been distinctly unnerving for those beneath (fig.3). To add to this, the entire space was plunged in darkness. There were just two sources of light: a workman’s brazier, placed centrally in the exhibiting area, around which visitors huddled, and the flashlights that the artist Man Ray, Duchamp’s longstanding collaborator and the so-called ‘lighting advisor’ for the show, had provided to enable people to illuminate the works on display.39

What is striking about the installation is the way in which a thematic of labour – manual labour – seems to have been in play, almost in counterpoint to the referencing of dreams. The beds, the darkness of the room, the use of flashlights to view the surrealist paintings; all this suggests that the exhibition-goers were ushered symbolically into the space of dreams, which, of course, is entirely in line with surrealist principles. The metaphorical linkages between the brazier, the coal sacks and the flashlights also speak of the idea that the exhibition goers were akin to coalminers, tunnelling into the unconscious. However, the coal sacks and the worker’s brazier simultaneously speak of a different world. The 1930s in France had been a period of constant industrial unrest – which included coalminers’ strikes – and the period immediately prior to the 1938 exhibition saw the demise of the Popular Front, unable to cope with the general industrial and economic downturn. It could be argued that it is precisely this external world that was imported by Duchamp into the Galerie Beaux-Arts in 1938, producing a weird collision of the symbolisation of inner (unconscious) space and brute social reality. The art historian T.J. Demos has interpreted the installation along similar lines, arguing, with reference to the German philosophers Theodor Adorno and Max Horkheimer, that Duchamp’s coal sacks, in their obvious oppressiveness, signalled ‘the wholesale destruction of particularity’ threatened by fascism and capitalism: ‘in their sheer numerical proliferation the sacks of coal bore down on the audience, as if neutralizing the surrealist fetishization of individuality below and expressing the coming absorption of objects by capitalist production.’40 This may be over-stating the case and reducing what was, in spite of everything, a surrealist gesture on Duchamp’s part to the level of literal social critique. However, Demos is right to suggest that, just as Duchamp seems to have been subtly criticising or even undermining surrealism in his 1942 installation in New York, so this process of subversion was at work in Paris in 1938.

The early 1930s had seen the surrealist movement severing its links with the Communist Party and saying goodbye to one of its former leading lights, Louis Aragon, who defiantly embraced the official communist aesthetic credo of socialist realism. In 1942 Duchamp made a quite explicit allusion to the political contradictions of surrealism in his Compensation Portrait, published in the First Papers of Surrealism catalogue, for which he hijacked the image of a sharecropper’s wife from the Great Depression by the left-wing American photographer Ben Shahn, substituting an image of his face for that of the woman’s.41 Despite the surrealists’ long-term commitment to the left, it had always been difficult to make themselves plausible as ‘men of the people’. For instance, Breton was unable to join a worker’s cell in Paris as part of his induction into the French Communist Party, admitting in 1929: ‘I was asked to make a report on the Italian situation to this special committee of the “gas cell”, which made it clear to me that I was to stick to the statistical facts (steel production etc.) and above all not to get involved with ideology. I couldn’t do it.’42

The gulf between the surrealists as dandies and actual workers is dramatised in the one known photograph that actually shows them in the company of a bona fide manual worker: it is a very early photograph, from 1921, recording an occasion when the Paris dada group, shortly to become surrealists, had managed to drag a token workman off the streets to pose with them for a group portrait.43 It is important to emphasise that, however much they supported the international communist cause, the surrealists were always virulently opposed to the principle of labour and what they saw as the bourgeois romanticisation of work. In 1929, for instance, in the special surrealist issue of Variétés, the surrealist writer André Thirion addressed an imputed bourgeois reader as follows:

The conflicting demands represented by actual labour and ideology were therefore central problems for surrealism. What Duchamp seems to have done in his 1938 exhibition installation was point out this problem in terms of the paradoxical conflation of ‘inner’ and ‘outer’ worlds.

Turning back to Duchamp’s installation for the First Papers of Surrealism exhibition four years later, and more particularly to the role that children played in animating the opening, it becomes apparent that, just as labour was the underlying but submerged thematic ingredient of the 1938 installation, so play should properly be seen as the determining conceptual schema behind the 1942 installation. With this in mind, however much Duchamp’s ‘mile of string’ can be seen to be a complex mythologically tinged allusion to ideas of displacement, its primary allusion is to a gigantic, beserk cat’s cradle. Leaving aside for a moment the relationship of its vernissage to the nexus of anthropologically informed attitudes to childhood that have already been discussed, it needs to be stressed that play as a concept in its own right was very much in the air at the time. The Dutch historian Johan Huizinga’s key work on the subject, Homo Ludens, had been published in Holland in 1938, but it is perhaps more significant that Roger Caillois, the author of the 1942 essay on ‘The Myth of Secret Treasures in Childhood’, would eventually go on to produce the other key mid-century treatise on play, Man, Play and Games (Les jeux et les hommes), first published in French in 1958.45 It becomes clear then that Duchamp seized on the principle of play as a topical issue in relation to surrealist practice, as though programmatically producing a pendant to the earlier show’s concentration on labour. What is particularly striking is that in both instances principles that might seem to be at odds with the lofty aestheticism and interiority of the surrealists – the workings of the industrial world in the case of the 1938 show and the play of children in 1942 – were mobilised precisely as incursions into the space of the gallery. Whereas in 1938 the presence of strikers on the streets of Paris in the preceding years and the immanence of a world war must have seemed pressing realities, to the point at which they virtually intruded into the installation, the incursion of play into the formalities of the 1942 exhibition opening suggests even more strongly the idea of ‘things getting out of control’. It is interesting, in this respect, that the central principle of Huizinga’s thesis, which in turn was adopted by Caillois, was that play necessarily takes place within constraints:

Duchamp himself was habitually pledged to the notion that play is a necessarily rule-bound activity, as can be deduced from his commitment to chess. However, the games that were played during the 1942 vernissage, however much they were wrapped up in themselves, trespassed into the domain of conventional adult social behaviour and became disruptive. Strictly speaking, therefore, it was not so much children’s play that was at issue, as the structural testing of boundaries. Indeed the performative nature of the vernissage strongly related to dada performances, where an attempt at bewilderment and alienation was often at stake and not mere playfulness. T.J. Demos has convincingly interpreted Duchamp’s 1938 installation in relation to the First International Dada Fair held in Berlin in 1920, and this comparison works especially well for the 1942 event when it is considered that no distinction or distance was established between the performance and the audience. Instead a kind of environment was actually created. The 1942 vernissage, with its focus on a unitary structural shift (gallery opening turned into play arena) actually seems to presage some of the Fluxus performances and happenings of the early 1960s, in which the relation between performance and audience was eradicated.47 Significantly, Duchamp had had no involvement in the formative live performances of the dada period.48 He came to the form late, and when he did so in 1942 it was precisely in order to reinsert the possibility of dada into a surrealist context, a point that serves to emphasise Duchamp’s concern with subverting the tenets of surrealism.

If the 1942 vernissage was effectively a precursor of neo-avant-garde performance art, it is also worth asking how Duchamp’s practice here, and in the installation of 1938, can be indexed to the critic Peter Bürger’s notion of the avant-garde proper. Bürger’s famous contention that the so-called ‘historical’ avant-garde was primarily motivated by a desire to sublate the principles of art and life fits rather cosily onto the account of the incursions of work or play into art that have been discussed. Beyond that, however, it has little explanatory power. More interesting theoretical possibilities arise from considering how the principles of work and play act on each other from one event to the other. The art historian Gavin Grindon has recently argued that the negative cast of Bürger’s theory (the idea that the avant-garde heroically ‘failed’ to dissolve the autonomous institution of art while the neo-avant-garde dismally repeated this failure a second time) obscures what was positive about the ‘refusal of work’ posited by the historical avant-garde.49 According to Grindon, the avant-garde preserved a notion of the autonomous nature of art activity linked to the philosophical conception of play articulated by the eighteenth-century German philosopher Friedrich Schiller in The Aesthetic Education of Man (1795). According to Schiller, as summarised by Grindon, ‘artistic production and aesthetic contemplation resolve the rational and sensuous aspects of man in the form of a play drive’.50 This positive autonomy of art-as-play is preserved by the avant-garde in opposition to the negative valuation ascribed by them to the autonomy of art-as-commodity under capitalism. It is thus necessary for the avant-garde to oppose the idea of labour as something bound up with the production of art-as-commodity. This refusal goes hand-in-hand with the refusal of the role of the artist. ‘Play’ can thus be opposed to the idea of ‘work’ in relation to a more positive and radicalised notion of artistic identity and value.

It might then be argued that Duchamp was keenly aware of the dialectic between work and play as central to the unfolding of surrealism. Duchamp’s thematisation of labour as the occluded ‘other’ of surrealism presented in the 1938 exhibition, reveals, as has been argued, the paradoxical nature of a movement that supported a communist cause but which at the same time refused to engage in ‘work’. Yet by the late 1930s surrealism had already capitulated to the market, and its rhetoric of the refusal of artistic labour had long since subsided. In the 1942 exhibition Duchamp seems implicitly to have posited that the return to the principle of play was the only means of reconnecting with a genuine avant-gardism.

It is time now to return to the peculiarly olfactory dimension of the 1942 vernissage, in other words to the ‘odour of cedar’ that, according to the exhibition catalogue, supposedly accompanied the children’s play. As it happens, smell had played a role in the Exposition internationale du surréalisme in 1938, namely the scent of roasting coffee (a possible allusion to Duchamp’s small painting Coffee Mill of 1910), which had been provided courtesy of a coffee-roasting machine installed by the poet Benjamin Péret.51 In 1942 the reference to the ‘odour of cedar’ in the exhibition catalogue, while connoting tribal objects from the Northwest coast, might be understood as a knowing nod on Duchamp’s part to the values of spirituality and healing that are traditionally linked to the cedar tree.52 Childhood, Duchamp appears to have suggested, is regularly seen in similar terms, as something linked intrinsically to natural purity and spirituality, and, in drawing out these associations, Duchamp alluded to friends such as Cornell, as well to the surrealist veneration for the state of childhood. In the 1942 exhibition Duchamp seems to have asserted, in highly coded terms, that such values are precisely what the art displayed around the walls is unable to deliver. The spontaneity and freshness of childhood was thus mobilised, in structural terms, against the surrealism for whom children were such exotic creatures and whom recent American surrealist discourse had positioned as anthropological specimens to be considered on a par with ‘primitives’. Duchamp thus pointed to something beyond the capacity of surrealism. In a last ditch attempt to recover the movement’s avant-garde impetus, he offered them a tangled cat’s cradle and the faintest whiff of the sensuous qualities of play.