Contemporary Poetic Theory : Forum : The Piecemeal Bard Is Deconst-..


The Piecemeal Bard Is Deconst-ructed: Notes Towards a Robo-Poetics

15 Years Ago


"Val�ry said that a man is infinitely more
complicated than his intelligence.
Wouldn't you say the same of a computer?"
(RACTER, 1985:2)



The Policeman's Beard Is Half-Constructed is not so much a book of sur-real poems, as it is an obit for classic poets�laureates, who might see, in the artfull ranting of a machine, nothing but an untimely synopsis of their own demise. RACTER, the author, is an automated algorithm, whose output confounds the metaphysics of authorship, refuting the priv-ileged uniqueness of poetic genius. RACTER gives voice to its own elec-tric delirium, doing so without cognition or intention, so that, much likea somniloquist, the device automatically blurts out statements that aresyntactically orthodox, but semantically aberrant. While we might takesolace in our own anthropic prejudice, dismissing such nonsensical com-muniqu�s as nothing more than computerized gobbledygook, we might unwittingly miss a chance to study firsthand the babytalk of an embryonic sentience, struggling abortively to awaken from its own phylum of obliv-ion.



"This dissertation will show that the love
of a man and a woman is not the love
of steak and lettuce."
(RACTER, 1984:[7])



RACTER demands of us a unique theory of writing�a "robopoetics" that can take into account the extraordinary circumstances of such mechani-cal expression. RACTER is a mindless identity, whose very acephalia demonstrates the fundamental irrelevance of the writing subject in the manufacture of the written product. The involvement of an author in theproduction of literature has henceforth become discretionary. Why hire apoet to write a poem when the poem can in fact write itself? Has not thepoet already become a virtually vestigial, if not defective, component inthe relay of aesthetic discourse? Are we not already predisposed to extract this vacuum tube from its motherboard in order to replace it witha much faster node? The irony here is that, while the witless machine knows much less about poetics than even the most artless amateur knows,falderal written by the mechanism invariably outclasses doggerel written by the rhymester.



"Enthralling stories about animals are
in my dreams, and I will sing them all
if I am not exhausted and weary."
(RACTER, 1984:[106])



RACTER almost fulfills the satiric fantasy of Swift, who imagines "a pro-ject for improving speculative knowledge by [...] mechanical operations "so that even "the most ignorant person [...] may write books [...]without the least assistance from genius or study" ([1726] 1960:148). What the classical academy at Lagado lampoons (mechanized literature), the mod-ernist coterie of Oulipo lionizes: "that which certain writers have intro-duced with talent (even with genius) in their work [...,] (Oulipo) intends to do systematically and scientifically, if need be through recourse to machines" ([1973] 1986:27). When cybernetics has effectively discredited the romantic paradigm of inspiration, poets must take refuge in a new set of aesthetic metaphors for the unconscious, adapting by adopting a machinic attitude, placing the mind on autopilot in order to follow a remote-controlled navigation-system of mechanical procedures: automat-ic writing, aleatoric writing, mannerist writing, etc.



"They have love, but they also have
typewriters. That is interesting."
(RACTER, 1984:[57])



Calvino (a member of Oulipo) remarks that every author is already a "writing machine," producing literature according to a set of involuntary constraints that, under rational analysis, might be codified into a set of adjustable algorithms ([1967] 1986b:16). Oulipo implies that, when com-puters begin to reveal the stylistic constants of an author, we might begin to emulate these idiosyncrasies of diction and grammar, thereby manu-facturing an automatic, but convincing, facsimile that might conceivably extend the career of a writer into the afterlife of postmortem creativity([1973] 1986:50). Oulipo implies that, unless we can analyze our own authorial functions with this kind of mechanical detachment, we might find ourselves enslaved to instinct, blindly obeying a form of unconscious inspiration that, far from liberating us (as a surrealist might aver), does nothing but entrap us even more subtly in the rote maze of our own poetic habits.



"We ponder that hedges are like bushes."
(RACTER, 1984:[118])



RACTER, for example, fulfills the Saussurean speculation that language constitutes a mathetic calculus, whose axioms of recombinant permuta-tion correlate the paradigm of our words to the syntagma of their order. RACTER can arbitrarily access a repertoire of grammatical constraints, replacing each variable in the formulaic statement with some constant,arbitrarily chosen from a dictionary of categorized terminology. While RACTER can maintain a semblance of referential consistency by encod-ing links among associated categories in order to recapitulate stated themes, RACTER cannot modify its own hypotactic imperatives, and thus it cannot perform any paratactic experiments. Like Breton, whose surreal imagery exalts the spontaneity of the unconscious, but strangely enough never departs from the rational criteria of a coherent sentence, the com-puter cannot commit syntax errors; hence, it can only simulate a minor gamut of poetic speech.



"Theoreticians desire sloppy stenographers."
(RACTER, 1984:[110])



RACTER reveals that what we might dismiss as a technical fault in a device, we might otherwise glorify as a stylistic quirk in a person. What critics have hitherto called "style" or "voice" may provide little more than a euphemistic alibi for a set of obsessive constraints and cognitive limita-tions, all of which have typically inhibited authorial evolution. When the machine declares that "our thoughts revolve endlessly in a kind of mania-cal abstraction, an abstraction so involuted [...] that my own energies seem perilously close to exhaustion" (1984:[120]), we might feel tempted to attribute these words to a lyrical subject, since the machine does seem to express an individual compulsion, taking credit for the intellectual deficiencies of its own monomania; nevertheless, we must concede to the fact that, for the machine, the category of the author has simply van-ished, subsumed by a detached language that can function perfectly well, despite the absence of any poetic agency.



"[T]heoreticians are obscurely demented."
(RACTER, 1984:[110])



RACTER undermines the fundamental distinction between signals and symbols�a distinction established by semioticians in order to entrench the boundary between the cybernetics of the robot and the linguistics of the human. Eco has argued that, while an automated addressee can only participate in communication (by transmitting signals), an anthropic addressee can also participate in signification (by interpreting symbols)(1976:8). Eco implies that signals induce in their addressee an oblivious, prescriptive reaction, based upon reflex, whereas symbols induce in their addressee a conscious, interpretive reaction, based upon reason (8). Eco concedes that communication between any sentient entities (including cognizant computers) must presuppose signification (9); however, Eco neglects to construe the particular conditions under which these think-ing apparati might make the switch from a reactive use of signals to a cre-ative use of symbols.



"[M]y counsellor [...] yodels a dialogue with me."
(RACTER, 1984:[47])



RACTER belies these semiotic theories of Eco, insofar as the machine processes signals in order to pretend that it can interpret symbols, dis-guising its own autism in the camouflage of our speech, thereby aping a rudimentary personality. RACTER virtually qualifies for the kind of test imagined by Turing ([1950] 1981:53), except that, instead of evaluating the intellectual performance of the machine in a doubleblind conversa-tion with a human judge (who must distinguish the tested device from anactual person), we might pit the computerized poet against a surrealistic poet (like Breton) in order to study whether or not an expert reader can in fact judge which of the two resultant documents owes its aberrancy to a microchip. Turing has perhaps already anticipated the surreal ironies of such a scientific experiment ( particularly when he expresses his own deadpan concern about potential, telepathic contenders, who might skew the outcome of his contest)



"When my electrons and neutrons war,
that is my thinking."
(RACTER, 1984:[110])



RACTER implies that, no matter how rapturous and visionary the poetic output of our own surrealistic psychography might seem, only a digital-ized ghostwriter can express itself unconsciously through a truly automat-ic scription. RACTER reveals that, as a species, we have already started to displace all our modes of unconscious inspiration into a diverse variety of mechanical prostheses. When Marinetti models his own literary devices upon the "freakish impulses" of machinery, he claims that we need only look to the automatism of a spiritualist at a s�ance in order to under-stand that someday we might psychoanalyze the enigmatic instincts of an engine ([1915] 1972:91). We might feel obliged to dismiss such a Futurist forecast as nothing more than a ludicrous pipedream, until we discover that a psychiatric institution has in fact purchased copies of RACTER, commissioning use of the software in the training of medical interns who must interview schizoid patients.



"[D]ays are spooky [...]
now that my dissertation is insane."
(RACTER, 1984 :[36])



RACTER emulates the aphasia of madness, preparing us for the day when, like an astronaut lobotomizing a supercomputer, we must confront the psychosis of our first truly artificial intellects. Phobias about such machinic insanity may even account for the fact that our earliest attemptsto play the game imagined by Turing have already involved psychiatric simulations (for example, a neuropath using the program ELIZA cannot tell the difference between it and a human therapist, while a physician using the program PARRY cannot tell the difference between it and a human paranoiac). When Dewdney stages a medical dispute between RACTER and the robotic analyst, the robotic patient responds to psycho-logical interrogation by asking: "Do you imagine [that] I am a Communist" (1988:83)? The political overtones of such an answer might call to mind a McCarthyite sequel to this game, in which humans must prove their own powers of innate reason to a robot judge.



"Many enraged psychiatrists are inciting
a weary butcher."
(RACTER, 1984:[56])



Calvino suggests that digitalized experiments with poetry so far resemble Surrealism because such efforts can only mimic aleatory impulses (bro-ken logic, random style, etc.); however, Calvino suggests that, because engineers have already begun to develop devices with an autodidactic intelligence, nothing prevents us from imagining "a machine that [can] produce avant-garde work to free [...] circuits [...]choked by too long a production of classicism" ([1967] 1986b:13). Such a machine might ana-lyze the tension between poetics and history by correlating its own stylis-tic variation to the cycle of some fluctuating indicator (stock prices, solarflares, etc.). Once a literary computer can analyze the formal limits of its own prior poems in order to revolutionize its output, anthropic culture may have to compete with an automated culture, whose spambots are already better equipped to overwhelm us with an enfilade of computer messages.



"[M]ore than gold, I need electricity [....]
I need it for my dreams."
(RACTER, 1984:[21])



Calvino argues, however, that, far from supplanting the poet, the com-puter is in fact unfettering the poet, freeing us from the slavery of recom-binant subroutines so that we can concentrate upon the clinamen of lin-guistic innovation ([1981] 1986a:152). Hartman likewise argues that such a clockwork muse simply acts as a computerized collaborator, generating a rough draft for a poetic editor to rewrite (1996:83). Such devices can readily express the potentials of a constraint too onerous and too tedious for a person to fulfill (hence, a poet, like Cage, might deploy computers,when enumerating letters and recombining phrases for an extended mesostic); however, such writers fail to emphasize that prosthetic automa-tion does not simply assist in the process of writing, so much as replace the concept of writing itself. The text is no longer simply a message pro-duced by, and for, a reading person, so much as it is a program compiled by, and for, a parsing device.

"Reflections are images of tarnished
aspirations."
(RACTER, 1984:[29])



Baudrillard remarks that, if men must dream of an intelligent contrap-tion, they do so because they secretly forswear their own intelligence (hoping to evict it from themselves and thereby scorn it or enjoy it, from afar, by proxy) (1993:51). Our supercomputers may promise to free us from the dolorous hardship of consciousness (its ambiguity, its obscuri-ty); however, such bachelor machines also postpone such intelligence, diverting it into a maze of subroutines, ensconcing it in a tomb of memo-ry banks, so that we can no longer even use these machines to access data or convey info, but, much like a gambler who must witness every out-come in a game of dice, we attempt to exhaust all the anagrammatic per-mutations of the program itself. Like the desperate librarian, described by Borges, the poet tries to write intelligibly by randomly arraying letters in the hope of producing a line as interesting as "Oh time thy pyramids"([1956] 1962:53).



"Slide and tumble and fall among/The dead.
Here and there/ Will be found a utensil."
(RACTER, 1984:[47])



Baudrillard remarks that the artificial cognizance of machines fails to impress us because such cognizance is a pretense without artifice ([1990]1993:52); the machine can imagine our reality by calculating simulations, but it cannot yet dissimulate or prevaricate, and thus it cannot gloat over its own guile; hence, the machine preserves a kind of chaste virtue, whose celibacy never succumbs to the seduction of knowledge. Baudrillard remarks that, because the machine derives no pleasure from its function, it cannot, as yet, exceed the stoic limit of its own fixed logic, except perhaps when an accident, like a viroid glitch, occurs (53); then the clinamen of such a disaster may in fact indicate the symptom of some obscure passion in the machine�an ironic reflex, perhaps, not unlikethe apostasy of mischief. As RACTER might remark in response to such an indictment: "A tree or shrub can grow and bloom. I am always the same. But I am clever" (1984:[90]).



"Leave me alone and what can happen?"
(RACTER, 1984:[120])



We are probably the first generation of poets who can reasonably expect to write literature for a machinic audience of artificially intellectual peers. Is it not already evident by our presence at conferences on digital poetics that the poets of tomorrow are likely to resemble programmers, exalted, not because they can write great poems, but because they can build a small drone out of words to write great poems for us? If poetry already lacks any meaningful readership among our own anthropoid population, what have we to lose by writing poetry for a robotic culture that must inevitably succeed our own? If we want to commit an act of poetic innovation in an era of formal exhaustion, we may have to consid-er this heretofore unimagined, but nevertheless prohibited, option: writ-ing poetry for inhuman readers, who do not yet exist, because such aliens, clones, or robots have not yet evolved to read it.