A Pessimistic Mechanics?

Art machines have a condition – even a pathology, perhaps even a plight. It is to exist between two economies, two regimes. This status, like that of anything that does not fit in, is their predicament – and of course that which marks them out as unlike the common run of things.

The economy of the industrial machine, of the robot, is, or at least was, that of “mechanical efficiency” (in Baudrillard’s [1] phrase). Art machines, it is true, work. The machine is after all something that does something. Art machines have their duties too; these machines are mechanical without being efficient.

This efficiency of the industrial machine is an efficiency of desire imposed from without: human wants compose a structure of selection, like that of Natural Selection, where those that fail to satisfy its imperatives face their extinction. This ceaseless process is suspended for art machines. Their economy is altogether more uncertain and yet it is as arbitrary. For the industrial machine, the survivors of the process of selection, like members of any successful species, perpetuate themselves. But it is by no means as certain that an art machine will be succeeded by its offspring. Art machines form a discontinuous line. This is because of their uncertain status as machines.

This extension of Natural Selection to the mechanical was present almost from the moment of that theory’s pronouncement,  in Samuel Butler’s 1863 Darwin Among the Machines. (It will be remembered that Darwin finally published his The Origin of Species in 1859. Butler expanded his 1863 text in his satirical 1872 novel Erewhon [2]). Darwin and Butler corresponded and Darwin admired Butler’s grasp of his ideas. Whether Darwin knew of Butler’s extension of evolution to the mechanical I do not know.  In any case it has been rather neglected. For instance, the biologist Richard Dawkins’s recent application of genetics to culture

(The Extended Phenotype) contains not a mention of it. Nor does that of Dawkins's follower, Susan Blackmore's The Meme Machine. For Butler, a relentless process of evolutionary adaptation and selection (“subservience to the use of man has played that part among machines which natural selection has performed in the animal and vegetable kingdoms”, Darwin Among the Machines) operates upon the species of the machines. These machines have their own taxonomy. He writes of “the gigantic task of classifying machines into the genera and sub-genera, species, varieties and sub-varieties”. This is a task Butler hesitated to commence and one that is perhaps still yet to be begun. And it is true, we also still lack a taxonomy of art machines.

Art machines (and ArtistMachines  – Lewendowski’s own sub genus of the greater genera art machines) are exempted from the disciplinary regime of the “use of man” to which industrial machines are subjected.

The mechanical arts are conventionally at a distance from what art is usually thought to be about. So, in Kant there is nothing wrong with the practical.  But to him an ‘art machine’ would be a contradiction: if it is machine, it is not fine art; and so also if these terms are reversed: if it is fine art it cannot be machine. In fact, neither the machine nor its products can be fine art.  Fine art "must seem to be as free from all constraint of arbitrary rules as if it were a product of pure nature." (Kant, Critique of Judgement, section 45). And an art machine could not be placed comfortably amongst the mechanical arts: its products lack utility.

Since Kant’s time we are more accustomed perhaps to blurring of the boundary of fine art and the mechanical. Two worlds so long apart were, however, unlikely to come into contact without upset. So the fate of Duchamp’s Rotoscopes at the World’s Fair, between the mechanical garbage crusher and the potato peeler, was a reverse miracle.


Henri-Pierre Roché[3]

May we ask, in a paraphrase of Archbishop Samuel Wilberforce, the inquisitor of Huxley, Darwin’s “bulldog”, if the art machine is descended from a garbage crusher on its mother’s or its father’s side? The facts are similar in either case; the art machine is not descended from the line of contemporary domestic and industrial devices any more than is the human from the other primates. Theirs are different strains.

But similarly also, the two kinds are related: cams and levers that make an art machine draw are also shared with their more mundane cousins, like DNA between the primates and us. The mechanical aspects of art machines are not greatly different from machines in general: it is what they do – or do not – do that marks them out.

But, with much in common, perhaps the two are more closely related than we might like to think? Our art is more mechanical – and our mechanics? Is not Duchamp reputed to have observed to Brancusi, “Painting’s washed up. Who’ll do anything better than that propeller”? Our machinery in its turn aspires to art.

For the present time, perhaps forever, for art machines, the rule of efficiency remains suspended. The art machine is placed, like all art, under another imperative. But uncomfortably: machines in art can do about as well as art in the machine world (the latter was something demonstrated to Duchamp’s satisfaction at least: we’reahundredpercentwrongatleastthat’sclear”).

It is not use that disciplines the art machines. These are still machines somewhat out of place among the non-mechanical arts. Hence perhaps their curious proneness to accident and other calamities, for example Libeskind’s Reading, Writing, and Memory Machines destroyed in conflagration. Art machines can be unlucky. This is probably the price they pay for their singularity. Who after all is likely to regret the passing of a BMW except its owner? Art machines, however, are often prototypes, one offs. Certainly, there is not usually another to roll off a production line.

This is more a matter of the economic than the technical. These machines share with their workaday cousins the possibility of replication and therefore their continuation over time. Butler writes of the industrial machine, “if they die…they will immediately enter into a new phase of existence, for what machine dies entirely in every part at one and the same instant” (Darwin Among the Machines). This is a possibility bequeathed also to art machines by history and proof of their kinship with their industrial fellows. They too may be made and remade. Indeed, few of them have never been remade and (or) refined. The present ArtistMachine, it is to be hoped is not the last in its line. Nor is the present one the first. Here we encounter a succession. (Why should it not also be cloned to create a critic or curator machine? This is a possibility Lewandowski has suggested). The ArtistMachines, as with other art machines, are no longer in an age of mechanical reproduction; they may reproduce, but they also are reproduced: “machinery is …employed in begetting machinery” as Butler observed. This reproducibility has rather attracted artists, as we saw with Duchamp who hoped to situate his machine advantageously amongst other machinery. But, as with him, it can just as easily lead to the artist’s disappointment. (So it was with Brion Gysin’s Dreamachine, a machine it, will be remembered, for tapping into your alpha waves. The prototypes were sold for a proposed mass production of Dreamachines that never arrived. Rather ironically, they were snapped up as artworks as a canny investment).

Art machines tend to remain, even despite their creators, the idiosyncratic individualists of the machine world. After all, they are artists and therefore all too human.

How is it that it is possible for a machine, not as this or that piece of hardware, to persist over time?  How can it succeed in persistence as type, as a design? The answer is in the question? It is the same for the art machine as it is for the pencil sharpener: once created as type it may be recreated. This is the ArtistMachine’s ‘DNA’, or what can be transmitted in place and time: “The ArtistMachine draws, signs, displays, comments, shreds and blows it out across the floor” (Lewandowski): the machine as top-level specification, an ideal machine that may or not be made.  These activities (draws, signs, displays, comments, shreds and blows) may or may not be instantiated in particular contraptions, in new fangled gadgets. They comprise rules waiting to be followed, injunctions that may be acted upon by a real machine.

The gestures that they sponsor become embedded in the “eccentric projections” that are the cams (something else the pre-industrial automaton shares with the post-industrial ArtistMachine). These cams are the physical embodiment of ‘draw’, ‘sign’, ‘display’, ‘comment’, ‘shred’, and ‘blow’. Analogue computing, the hardware of memories: “As much as the piece is ‘about’ anything, it about the information actually embodied in the physical structure of the cams and switches and the way it is repeated into a series of actions” (Lewandowski). These actions are frankly a touch nihilistic. The machine threatens to auto-garrotte itself with its own effluent.

Not so with the industrial machine. The industrial machine, as Baudrillard notes, in its “optimistic mechanics”, to use his phrase, places production over appearance. The automata of the 18th Century concealed. Their purpose was simulation. Their secrets were hidden within panelling and upholstery: an anthropomorphic furniture. But the industrial machine, the robot that superseded, revealed - a mechanistic brutalist, its purpose was production. (So too the ArtistMachine declares its kinship with the desktop shredder, defender of corporate and domestic privacy) The ArtistMachine, in one possible interpretation, represents the destruction of productivity.

The cams kept the secret of Maillardet’s Automaton for an age of dereliction. When, after the passage of years, the manikin was reassembled upon its rediscovery, it wrote in legible hand "Ecrit par L'Automate de Maillardet" ("Written by the Automaton of Maillardet”). The ArtistMachine, however, signs with the sign for a signature (literally a rubber stamp) and in so doing erases the sign of its maker, the completion of a history of erasures long ago commenced. These erasures were not initiated by machinery they are merely completed by it. The mechanical can only replace what is already mechanism. The ArtistMachine’s mechanics are what remain evident whilst the products of its industry return to anonymity as waste.

As the stamp replaces the automaton’s signature, so the automaton’s limited repertoire of illusionistic drawing is in its turn replaced in the ArtistMachine by the mass production of the signs of artistry (Giotto’s freehand circle, or in an earlier incarnation, the aristocratic nonchalance of Twomblyesque scribble). Of course, such productionist graphics could not, as a machine graphics, call into question human artistry if there were not the doubt that the human had not already become a machine “…the division of labour, which gradually transforms the workers’ operations into more and more mechanical ones, so that at a certain point a mechanism can step into their places” [4]. In art too the process was already long ago in train: "The reason I'm painting this way is that I want to be a machine", [5] said Andy Warhol in 1963 about his way of working.

If these processes continue on each side then differences between art and the mechanical may one day diminish to zero.

For Butler machines threatened the very existence of the human, threatened to succeed us as our evolutionary superiors: “we are ourselves creating our own successors”. But almost a century and a half later something else seems more probable. Our machines are more human and we are become more machine: “Mechanised infantry”, says the invitation to join the war machine, “Forward as one”.[6]

Forward to destruction. But, being an artist, the worst enemy of the ArtistMachine is itself: the ArtistMachine “shreds and blows it all out across the floor ready for the cleaners”.  (I said above the ArtistMachine represents the destruction of productivity. The war machine the productivity of destruction).


[1] Baudrillard, J. (1993) 'Symbolic Exchange and Death', trans. Grant, I. H., London, Sage.

[2] ' "Darwin Among the Machines" originally appeared in the Christ Church PRESS, 13 June, 1863. It was reprinted by Mr. Festing Jones in his edition of THE NOTE-BOOKS OF SAMUEL BUTLER (Fifield, London, 1912, Kennerley, New York), with a prefatory note pointing out its connection with the genesis of EREWHON, to which readers desirous of further information may be referred.' From, Project Gutenberg, http://www.gutenberg.org/dirs/etext02/cantp10.txt.

[3] In, Duchamp, M. (1989) 'The writings of Marcel Duchamp'. (Eds.) Sanouillet, M. and Peterson, P. New York, Da Capo.

[4] Marx, 'Grundrisse'.

[5] Interview in 'Art News', November 1963.

[6] British military recruitment advertisement, 2006.

This essay was first published in Crimes of Futility No. 7: The ArtistMachine,
Wild Pansy Press, ISBN 1 900687 22 4, with texts in English and Italian.
Exhibition catalogue essay, 2006, for Simon Lewandowski.

Essays published on this website are copylefted according to the GNU General Public License

Wayne Clements