The Mechanical Art of Laughter

Abstract

Our aesthetic experiences are today conditioned by machines, which operate at multiple levels: at the moment of conception of a work, at the moment of conservation and distribution of the work, and at the moment of its contemplation. For art today, it is no longer a theoretical question of asking whether the machine can act with freedom in the sense of a game that remains as of yet open-ended—or if humans themselves can still so act in a world entirely conditioned by technology—because the brute fact is that machines are becoming ever more autonomous, and humans ever more dependent upon them. For some artists, therefore, the ideas of autonomy and sacralization are best addressed, not in the posing of serious questions, but rather through the subversive activity of enticing the machine to reveal its comic nature—and wherein we discover, with Bergson, the essentially rigid and mechanical nature of the humorous.

1. The Omnipresent Machine

If one were to be allowed a somewhat impressionistic description of the role of the machine in modern culture, one might begin by noting that it is nearly everywhere, and especially in the production of sound and image, as with photography, cinema, video (considered as a distinct art), and television. The means of cultural production and distribution have long since been rationalized and made autonomous: images are produced, reproduced, and diffused through numerous mechanical, electronic, and algorithmic procedures. Images are in turn becoming less representation than artificially produced simulation. A relay of automated reproduction is set off: from literature to theatre to cinema to television, and back again to the printed page in the form of reviews, take offs, and send ups. At the center of it all is the machine in its role of institutionalized automatism, and operating at several levels: at the moment of preparation of a work, at the moment of conservation and distribution of the work, and at the moment of its contemplation. These devices can be understood as organs, as extensions of our senses, and human perception as a system, using models taken from cybernetics, and conceived of on the basis of computations, actions, and feedback loops.

Since the mid-20th century, in short, the automatic processing of information has brought about a major shift in the nature of the work of art; but even before then, the purely mechanical machine had become inseparable from our own aesthetic experiences, as with the sensation of speed and the rapid change of scenery when traveling by train, automobile, or airplane. Likewise, when considering the intimate emotional spaces in which the machine acts upon us, our current awe at witnessing, for example, the precision of the surgical robot (Ancarani 2012) is simply an addition to the emotions we have long felt in the presence of certain historically impressive machines. The machine has now given us the tools, furthermore, to experience phenomena inaccessible with our basic human senses. For instance, the phenomena of the behavior of certain materials in the experiments of the artists Evelina Domnitch and Dmitry Gelfand (Domnitch and Gelfand 2018) would be completely unknown to us without the technology that allows us to see at such a level; technology is similarly essential in the sonification of electromagnetic activity (Kubisch 2003) or tidal flow (Eacott 2008); and the machines of Felix Luque-Sanchez even allow us to experience a sense of the infinite (Luque-Sanchez 2015). In a world composed of information, machines translate, transpose, code, decode, and transcode phenomenon on our behalf. Indeed, we can enter a universe in which everything is calculated—reference points, forces, illumination, structures, textures, behaviors—and where, lacking mass and without up or down, we can even pass through walls: a so-called ‘virtual reality’.

2. The Subversive Machine

In October 2018, a series of murky but also compelling portraits in a style reminiscent of the 18th century—but generated almost entirely by computer—were offered for sale at Christie’s by the Paris collective Obvious (Fautrel et al. 2018). The portraits were produced using the GAN (Generative Adversarial Networks) technique developed by AI researcher Ian Goodfellow and his team of engineers at the University of Montreal, with human intervention limited to selecting in the first place the large set of existing portraits fed into the system and used by it as examples of the expected output. This marked the first offering of AI-generated art at a major auction house, and with the lead such work selling for $432,500—this a piece entitled Portrait of Edmond Belamy, which was signed in the lower right corner with the equation guiding the entire process, as if by an artist with the charming name of min_g max_d E_x[log(D(x))] + E_z[log(1 − (D(G(z)))] (Elgammal 2018; Schneider 2018).

The work of this Paris collective has as one of its precedents the late 1950s Méta-matics of Jean Tinguely. These were painting machines, to be sure, but the paintings produced were an unpredictable result of the co-action between the device itself (with characteristically imprecise drive belts), the audience member’s choice of colored marker, and the amount of time the sheet of paper was left exposed to the machine’s scribblings. The source of whatever creativity that could be said to be involved was thus not obvious, and this was in fact where the real interest in the Méta-matics resided. While each drawing, moreover, was unique, they were not presented as legitimate works of art inasmuch as this would have undercut one of Tinquely’s principle goals with the exercise: in 1959, Tachism was at the forefront of the Parisian art scene, and Tinguely wished to mock the subjective excesses and overly serious discourse of these artists. There was, as well, an implicit denunciation of over-consumption and commercialism:

The drawing machine needed to be cool, funny. The child playing with it had no problem at all, whether it was a work of art or not. And when it began to move, I really liked it, it built up to a certain speed where everything became ridiculous. It became burlesque.1

The Obvious collective, in turn, seems to be mocking the feeding frenzy which characterizes not only the speculative art market but also the community of technophiles; and there is also an inherently subversive quality in the hint that objects produced by artificial intelligence can take on some of the attributes of the ready-made.

In other words, we can trace here the evolution, over a period of sixty years (or more than one hundred years if we go back to Duchamp’s 1913 Bicycle Wheel), of what must now be seen as an alternate artistic strategy—this in contrast to those previously mentioned artists who have embraced the possibilities provided by the machine—for dealing with the brute fact that machines are becoming ever more autonomous, and humans ever more dependent upon them: rather than posing theoretical questions as to whether the machine can act with freedom in the sense of a game that remains as of yet open-ended, or whether humans themselves can still so act in a world entirely conditioned by technology, there is an obvious potency in enticing the machine to reveal its subversive and/or comic nature.

Such an approach can even be placed in the service of a certain cynicism, as with the Cloaca series of ten works, dating from 1992, by Wim Delvoye (Regine 2018): huge, assembly line-like machines which digested a carefully prepared mixture of ingredients in order to produce quite real-seeming feces (up to 80 kg of fecal material produced daily) as an example of technological development always more ‘driven’ (or, if you prefer, ‘pushed’). The output of these systems was actually vacuum packed and sold to the public as a further commentary on our modern industrial economy, and with both scientists and chefs employed as consultants in order to optimize the overall process.

Artists have not lost their sense of humor. Far from falling in line behind a tendency towards monumentalization and stylization that one finds among the worshipers of science and technology, there is to be found a derision with respect to the monumental and its associated aesthetic claims. The machines of these artists continue to probe subjects which, though not serious, are very profound.

3. Towards a Practice of Mechanical Subversion

Some of these works seem to come straight out of Jacques Carelman’s Catalogue of Unfindable Objects (Carelman 1997). Such is certainly the case with the work of Simone Giertz, who presents on YouTube and elsewhere (Giertz 2017) her nutty, do-it-yourself robotic creations. A prime example is the robot that serves breakfast (Giertz 2015). Very approximate both in form (it seems to be held together with tape) and action (the cereal is poured beside the bowl instead of in it, as is the milk), the articulated arm moves in a way that is seemingly deliberate (it identifies the objects to be manipulated, and makes appropriate gripping and tilting motions), but ultimately abrupt and clumsy (the spoon is not dipped quite low enough to actually reach into the bowl and so arrives empty, and only in the general vicinity of the mouth of the inventor, who must therefore stretch her head awkwardly to the side to meet it). The humor of the situation—a technically advanced object which is in fact pathetically inept—is further heightened by the apparent aplomb of its inventor, who continues throughout to read a book without glancing up from it. A similar stoicism is exploited in the video of a makeup machine that scribbles lipstick all over her face; and again, we must be reminded, albeit now it in robotic form, of the machines conceived of in the last century by Jean Tinguely.

In the same vein, My little piece of privacy by Niklas Roy (Roy 2010) centers on a curtain installed in a storefront window, and meant to prevent the occasional sidewalk passerby from looking in. It is much too small for the job, however—reminding one of the tiny handkerchiefs behind which exotic dancers pretend to hide their dainties—and so must be robotically shuttled back and forth along its curtain rod (this in fact accomplished with a quite sophisticated system consisting of a surveillance camera, computer, and servo drive mechanism) in order to attempt to continuously block the view of said occasional pedestrians as they pass in front of the window. The behavior of the curtain, in turn, evokes a reaction from them, who notice that its movement follows theirs. The interaction sometimes becomes playful, with the goal of the game being to move faster than the curtain, or to find strategies which will trip it up. Once again, there is a disparity: on the one hand, between a task calling for subtlety and discretion, and, on the other, the mechanical system to which it has been assigned. The curtain thus sometimes ends up being jerked back and forth in a frantic and hilarious manner—and we are thus reminded of the crucial connection that Bergson has made between mechanical rigidity and the comic [italics mine]:

Consequently, it is not his sudden change of attitude that raises a laugh, but rather the involuntary element in this change—his clumsiness, in fact. Perhaps there was a stone on the road. He should have altered his pace or avoided the obstacle. Instead of that, through lack of elasticity, through absentmindedness and a kind of physical obstinacy, as a result, in fact, of rigidity or of momentum, the muscles continued to perform the same movement when the circumstances of the case called for something else. That is the reason of the man’s fall, and also of the people’s laughter.2

Or in summary, “the mechanical plastered onto the living”, to use Bergson’s phrase, is the true source of laughter.

The literary realm, likewise, has its examples of algorithmic subversion, and these are clearly Oulipian in spirit. Such is the case with the email novel Rien n’est sans dire (Nothing is without saying) by Jean-Pierre Balpe (Balpe 2001), and the ‘Pipotron’ by the collective Cyber!Campus (Cyber!Campus 1997), an automatic generator of random sentences, hollow phrases, and other gibberish which, at the end of a long day, one might use to plump up the introduction or conclusion of a serious report. Comic and ironic at the same time, the Pipotron produces results not unlike those of a certain all-knowing politician; and here again, the humor stems from the rigid and wooden quality of phrases that do fit into the discourse. More recently, director Oscar Sharp has given us his Sunspring (Sharp 2016), a short science-fiction film whose dialogue was automatically generated by an AI program of the type originally designed to predict, for example, what word one is attempting to type when sending a text message—but trained instead on the scripts of dozens of science fiction films. A movie so generated will of necessity remain utterly directionless and incoherent, and without depth and meaning; but the viewer is nonetheless surprised by some improbable effects. In the first place, one is shocked to discover the extent to which properly formatted but ultimately nonsensical language can arouse in us an anticipation of meaning; but one is also shocked to realize that we have become almost accustomed to such language via the formulaic speech of advertising and politics. Indeed, we are all but startled to discover that our own understanding has perhaps been in an automatic mode as well. In keeping with our theme, furthermore, this surprise ultimately turns into laughter—i.e., the dipositif3 stands exposed—at the beautifully constructed gibberish we have been taking so seriously.

4. Postscript: Unproductive Expenditure and the Free Laugh

In his The Notion of Expenditure (Bataille 1933), Georges Bataille reminds us of the importance to society of what he calls “unproductive expenditure”, and hence it is with the comic machines we have been examining here.

From their derision are born unstable images apt to trigger reflections on the meaning of existence. These so-called “useless” machines do not produce anything except laughter, dreams, and even dread. They are machines of theatrics, machines for communication. Yet they do not willingly admit their madness: as with any robot working in a world where order is disorder, they obey.

Decidedly unlike humans in their form and appearance, yet extremely close in their attitudes, paradoxes, complicity, and the humor they provoke, these mechanical works are sufficiently like us so that in a moment of recognition, the regard of each spectator can be turned back on itself. In this, they allow us to go beyond the appearance of the indivisible and the permanent in favor of a mobile, heterogeneous multiplicity of the identities we give to them.

These games thus allow for the appropriation of a symbolic space where the work of art is, in part, unencumbered by the weight of social constructs. In this sense, the artist builds a space of experimentation where the experiment feeds fictions, and these fictions in turn feed reality.