DEFENSE AND ILLUSTRATION OF NEWSPEAK
by Jaime Semprun

Chapter VIII: Newspeak as a necessary outgrowth of communication between machines

The philologist Victor Klemperer, who in Germany in the 1930s had occasion to observe the attempted establishment of a new language, noted “the abundance of constructions belonging to the domain of technology, the mass of mechanizing words.” To illustrate this “tendency toward mechanization and automation,” resulting in a “flagrant mechanization of people themselves,” he offered as a characteristic example the creation of the verb gleichschalten, borrowed from the vocabulary of electromechanics, and which he said made one “hear the click of the button whose pressing lends uniform and automatic attitudes and movements to otherwise distinct human beings.” This verb, whose literal meaning is “to synchronize,” is habitually translated in French as mettre au pas,² but it seems to me the richness of French newspeak would allow us to find a better equivalent for it: one could use, for example, mettre en phase³ or maybe formater,⁴ depending on the context. In any case, the appropriate term most certainly exists, which usage will demand if it hasn’t already.

Indeed, the new idiom currently taking shape does not lack for terminology or turns of phrase drawn from the technological realm, computer science in particular. But could one really go so far as to risk drawing a parallel with the the language of the Third Reich, which was the object of Klemperer’s study? Surely not, for the latter took pains to note that, in this case, it was strictly a matter of the “encroachment of technical expressions on non-technical domains,” domains which were still so vast, diverse, and numerous, that to impose a technologized language upon them demanded an application of the most terroristic possible violence. There was no way the results thus obtained, by regulatory coercion, could be durable, and thus the promoters of this arbitrary and premature newspeak were greatly mistaken in proclaiming that the language of the preceding era, which they referred to as a “mummified past,” was “no longer spoken or understood in the present day.” Despite their proud declarations about the “clarity” and “determination” of their directives, their lack of confidence and technical expertise was betrayed by their contradictory dependence on a mythico-naturalist phraseology peppered with metaphors emphasizing things that sprout and grow spontaneously, without being forced or perverted, without being “artificialized” by intelligence: the “organic truth,” the “mysterious center of the people’s soul, the soul of the race,” and everything that has the heady odor of blood floating around it.

Nothing like that is to be found in French newspeak, where the profusion of technical terms corresponds quite exactly to the expansion of the domains of life that are actually regulated by technological procedure. Likewise, the environmentis no longer evoked in this context under some mystical form of “nature”—an obscure power that eludes intellectual enlightenment and rational calculation—but rather, as we have seen with regard to agroforestry and biodiversity, as a reserve of resources to be protected and managed. You’d have an even harder time equating the present technologization of the language with the one awkwardly cobbled together by the ideologues of the Third Reich, for the simple fact that it has not been imposed on us in an authoritarian manner but responds to an authentic social demand, passed along by all sorts of experts whose very vocation consists in openness to the Other. Once more the daily news is at hand to provide me with an example. A psychiatrist expounds upon how the Internet can serve as a kind of behavioral therapy, helping someone who suffers from social phobia, who is unable to speak in social situations, take the plunge by sending an email, thereby setting the machine in motion, thereby initiating the relationship and test-driving her emotions with a practice run. These three figurative turns of phrase borrowed from the mechanical domain are immediately understood by everyone, they aren’t in any way the province of some specialized professional jargon, which shows once again that the character of newspeak is fundamentally democratic: in one and the same movement everybody speaks like a shrink, and the shrinks speak like everybody else.

Everybody has been test-driven by their daily contact with machines, and therefore everybody—in order to describe what in oldspeak would have been called mental operations or states of consciousness—makes spontaneous use of images inspired by the functioning of electric and above all electronic devices: recollections are no longer engraved upon one’s memory but on one’s hard drive, one is connected to all kinds of networks, one crashes, one channel-surfs, eventually one blows a fuse or cable, etc. My aim here, as I’ve already said, is not to draw up a nomenclature that would necessarily be incomplete. The phenomenon is in any case so obvious I can leave it to the reader to come up with further examples. I will make one last addition, however, because it didn’t come to me by way of the news and because it illustrates so well what I want to talk about. It consists of verbs formed from the Anglo-American terms bug and glitch, words used (in French) as active verbs and no longer as mere nouns, to say that a computer program has bugs or glitches. A child who learns piano and hits the wrong note will say quite naturally that she had a glitch: “J’ai bugué.” And here we see just how profoundly the purists misconstrue the laws governing the evolution of language, and in particular the propagative power of images based on common experience; some people are still out there suggesting that, in order to save some aspect of oldspeak, these English glitches and bugs should be replaced by the native French terms for poxes or infections.

Such absurdities are the inevitable result of the purist’s will to freeze the language in place and hamper the free play of imagination, which ceaselessly transforms and enriches the lexicon by transposing words from one order of ideas into another. The word bug itself, which in English is a general term for various microorganisms and pests, must have followed a long and winding path (which I won’t retrace here) before including its current reference to a kind of syphilis of the computer. Then it made its way into French, and finally its meaning was enlarged further in order to refer, conveniently and concisely, to any manner of dysfunction, whether of a computer or a person: the computer, to which we had ascribed the aptitude of being user-friendly, has given us in return the ability to have glitches. Admittedly we have here what earlier rhetorics would have called a catachresis, an abuse of language, but you couldn’t write or pronounce two consecutive sentences if your intent was to bring every word involved back to the exact significance they carried at their origin. And as for borrowings from foreign languages, Bréal said quite sensibly, more than a century ago, all there is to say on the matter: “In every age and in every nation there have been purists who protested against these borrowings. But those who really form the language, anxious above all to be understood, and that at least cost, trouble themselves but little about the origin of the materials with which they work.”⁵

It is of course impossible to foresee whether this figurative construction, “I had a glitch,” will enjoy widespread usage. Certain images created this way, from the comparison of two objects or actions, have been adopted because they were accurate or picturesque, because they filled in a gap in the vocabulary or even just expressed what needed to be said more quickly; others, to the contrary, because they introduced an annoying ambiguity, have failed to “take.” Be that as it may, the developmental process applied in the present case, which consists of using a word derived from a particular technological vocabulary in order to broaden its meaning, is perfectly irreproachable. Indeed, whether you consider it a simple expansion or rather as a kind of metaphor, it is consistent with the very nature of language and linguistic development. To persuade yourself of this, consider for a moment the innumerable abstract terms which, in various languages, originating from a word supplied by the specific vocabulary of what used to be called the “mechanical arts.” Thus did the Latin word ordo (which itself gave us “order” and its long series of significations, right up to the most abstract) come from the activity of the textile worker and from the verb signifying ourdir, i.e. to weave, to produce cloth by a particular arrangement of warp and weft. It is therefore entirely natural that computer science, which weaves more than any other technology today the fabric of our existence, should be the principle purveyor of words formed in this manner.

This question of the technologizing effect of newspeak is bound up with the question of its clichés. These ready-made expressions, named (in accordance with the process I’ve just alluded to) with the help of the printer’s term referring to the operation by which a block is cast from the traces of a page composed in movable type, are therefore definitively fixed blocks of words, infinitely usable, what linguistics calls more sophisticatedly—without concession to the picturesque—fixed syntagms. Their frequency within specialized languages forged by totalitarian ideologies, where they serve above all as signs of belonging and fidelity to the party or “movement,” has earned them the denomination langue de bois in French—literally “wooden tongue”—evoking its weight and rigidity, and which in English might be rendered as “bureaucratese.” This expression has itself become something of a cliché, used as a marker, together with the phrase groupthink, to show by contrast how free one’s own thoughts are. There has obviously been no lack of intellectuals eager to advertise their heterodoxy in order to denounce newspeak for harboring a new “bureaucratese,” comparable in its rigidity and poverty to those of the totalitarianisms of the 20th Century: their propagandas made language into an instrument, a lever, a machine, and one is meant to remark this same trait in the technologized language of today. A constant repetition of the same clichés amounts, in this view, to the manifest trampling of individuals under the boot of conformity.

I won’t repeat what I said above concerning the difference between the encroachment of the technical terms Klemperer spoke of and the borrowings by which contemporary newspeak enriches its novel turns of phrase. The fact that the ideal of technological rationality might not have been brutally imposed from without but instead has been internalized, integrated into the existence of each individual, also allows us to understand how much the status of newspeak clichés differs from those of previous varieties of “bureaucratese.” To highlight this difference, one can, as it happens, refer to the history of technology and think of everything that separates the first “electronic calculators,” massive and sluggish as they were, to the personal computers of today; or consider the heavy slabs of reinforced concrete, such as were fabricated wherever the initiative for the industrialization of buildings rested with the State, in comparison with the flexibility of using concrete poured directly at the site of construction, as is the case today. And yet, on a superficial level, it could appear that we’re still talking about the same things, that is, about computers and concrete.

The clichés of totalitarian languages—like those heavy slabs, so cumbersome to transport and install—presupposed a centralized organization, a unified power of decision, an efficacy obtained through simplification and excessive standardization. That is no longer where we are. In a world where computers are personal, newspeak is, as it were, homemade. We’ve noted above the degree to which this new language was democratic in its purpose, in its conformity to social demand, and now we will see that it is every bit as democratic in its method of production.

The digital robinsonades habitually proposed under the guise of descriptions of the information society present to us an isolated individual confronted with the luxury of the virtual jungle, the immensity of its possibilities. Equipped with search engines, he carves a path, climbs through tree views, and surfs on the network from chatroom to chatroom. No point dwelling further on this interactive idyll, with which no one by now is unfamiliar.

To describe things this way, adopting the point of view of the private netizen, keen to satisfy this or that need or curiosity, or simply to drift, no more allows us to understand the global functioning of the computer system than would, in the case of capitalism, a description adopting the point of view of the private consumer, beginning with his need to purchase shoes or ham. Just as the primary goal of the market system is assuredly not to satisfy the needs of consumers, but to realize profits, likewise the primary goal of the worldwide computer system is not to inform or entertain the cybercitizens of a programmed society, but rather to make machines communicate with other machines, in a language of binary signals that is specific to them.

Machines have in fact for a long time tended to link up directly to one another, without having to rely anymore on the intervention of fallible human beings. All they lacked in order to do this was a common language that would allow them to record and transmit knowledge, and thus to acquire a collective memory. Computerization was a response to this need. When, in 1872, Samuel Butler contemplated the hypothesis that the machines of his era might be to those of the future what the first saurians were to mankind, he still could not have imagined but in the vaguest sense what would one day be their language. And yet he saw well enough in the earliest wailings produced by these fledgling machines, in the various signals, bells, and sirens, by which they announced their needs to their supervisors and mechanics, the embryo of a sophisticated language—exactly as cries of warning or command had been the first forms of human language. From that moment it was in his view inevitable that the machine people should one day or another manage to assume a superior stage of evolution, going so far as to form an organized society and even, so he believed, declaring their independence.

Butler took care to refute the arguments against his thesis, arguments which amounted more or less to saying that machines, however advanced they might become, would nonetheless remain under our control, would possess no free will of their own, of any kind, and would be unable even to reproduce themselves without our cooperation. To all of this he replied, quite astutely, that we would evaluate the situation more correctly if, rather than basing our reasoning on the existence of each machine taken separately from every other, we would consider them all together, as an already organized collectivity. Then we would see them collaborating in order to reproduce and perfect themselves, and we would observe that, if they do require humans for reproduction, it’s a bit like the way many plants require insects to achieve the same purpose. But whereas insects fulfill this function without any awareness of what they are doing, we are for our part entirely conscious and even proud of serving, as we do, the development of machines.

It’s worth noting, incidentally, that the argument according to which machines don’t reproduce without our cooperation can now be turned precisely on its head: for even when we don’t resort to methods like in vitro fertilization, we no longer reproduce except for with the assistance of various machines and technical apparatuses, beginning with the indispensable ultrasound. And the most confident futurists foresee, in the very near future, secure births taking place in artificial uteruses guided entirely by computer.

To support his thesis, moreover, Butler raised two facts that are significantly more remarkable today than they would have been in his own age. The first is that our supposed free will is a decoy, since we wouldn’t know how to survive more than six weeks if we found ourselves abruptly deprived of the machines on which we’ve become dependent, as much morally as materially. The second is that even while they seem to be exclusively at our disposal, the machines are the ones who dictate their conditions to us, and impose on us a way of life consistent with the optimization of their functioning. Which is as much to say that they have domesticated us, and that we serve their purposes far more than they serve ours. I would add that this last affirmation is in no way contradicted by the fact that in our time machines have less and less need of their human servants, as is the case with automation. Indeed, this only proves that they have become still more independent, that they have less need of our help, in short, that they have well and truly left their childhood behind, as Butler predicted they would.

It all seemed fairly audacious at the time, and still today it will be protested with such and such example of services that machines perform on our behalf without any compensation, and the dishwasher will be cited, and the mobile telephone, etc. But this is only to repeat, each time, the error which Butler has already refuted: the error, that is, of considering an object only in isolation, such that its utility gives it the appearance of being benign and of little consequence. By contrast, as soon as one considers it as an integral part of a larger whole, everything changes. And thus the automobile, that most trivial of machines, and practically an archaic one, which everyone agrees is quite useful and even indispensable to our freedom of movement in the world, becomes something else altogether once it is put back into the society of machines, back, that is, into the general organization of which it is a mere element—a cog, if you will. Now a whole complex system reveals itself, a gigantic organism composed of roads and highways, of oil fields and pipelines, of gas stations and motels, of trips conducted via tour bus and vast areas transformed into parking lots, of interchanges and beltways, of assembly lines and offices of “research and development”; but also of police surveillance, traffic signals, driving tests, regulations, standards, specialized surgical care, “the battle against pollution,” mountains of worn out tires, batteries to recycle, sheet metal to be compacted. And in all of this, like parasites living symbiotically alongside the host organism, affectionate, aphid-like ticklers of machines, men go about busily caring for them, maintaining them, feeding them, serving them still when they think they’re moving about of their own volition, since they must be used up and scrapped according to the prescribed rhythm, so that their reproduction—and the operation of the general system of machines—will not for an instant be interrupted.

It seems pointless to me now to apply the same analysis to far more modern machines, whose imbrication is so well-known as to have been given the name interconnectedness. These are actually for the most part simple extensions of the machines’ nervous systems: sense organs, sorts of antennae, or terminals which serve to ensure that their human bearers conform docilely to the injunctions of mechanical life. The very examples these same people give of the use they make of machines illustrates this function quite well: we need each one of these machines in order to respond to the demands of all the others. And the fact that this nervous system, with its geostationary satellites and its computer network, knows the position of each one of these endings at every instant, what they’re recording, the information they’re transmitting, whether it’s a matter of credit cards or mobile telephones, is sufficient to prove that what we’re talking about is indeed one and the same organism, even if its morphology doesn’t correspond to the idea we have of a living organism.

I’ve been expanding Butler’s demonstration with my own contemporary examples, but to conclude I must touch on two particularly significant facts. First, that men today have so thoroughly accepted their servitude to machines, to their reproduction and perfection, that in every circumstance they put the interests of machines ahead of their own. Not only is there no inconvenience they are not ready to endure so long as it is justified, in their eyes, by technological imperatives, but they go so far as to put (calm as can be) their own mere survival in jeopardy, in order not to interfere in any way with the development of the society of machines. In this domain, nothing scares them: not the disruption of the climate, not radioactivity, not the unpredictable behavior of ever more numerous chemical molecules, etc. Whatever the acknowledged effects such contingencies might have on human health, so long as they are somehow or another necessary to the smooth operation of the mechanized world, the humans accept them willingly. And none of this diminishes the prestige of machines. The devotion which they are paid even degenerates among certain people into fanaticism: the world may burn but machines shall reign supreme! Yet this unshakeable faith, according to which all of the problems created by the civilization of machines will be solved⁶ by some ulterior stage of its development, rests upon a perception of our own inferiority, which is not in any case without a certain lucidity. It brings to mind something the short, ugly, malformed marquis said toward the end of the Ancien Régime, showing off the handsome servants that ran his household: “Look what we make of them, and what they make of us.” If we acknowledge, as Carlo Michelstaedter observed in 1912, that “every technological advance impairs the corresponding part of the human body,” then we should also admit that respecting the intelligence of machines is not an act of faith but rather a manifestation of realism. It was already apparent to Butler that “wherever precision is required, man flies to the machine at once,” and where is precision more necessary than in measuring the alterations to which our natural world is being subjected?

All of which brings me quite naturally to the second fact I wished to mention. Thanks to their sensory organs, their sensors, machines accumulate data, or pertinent information, which they then transmit as it were over our heads, and to which we in any case do not have access except through them. For these data, which concern the entirety of the conditions under which terrestrial life presently unfolds, are not directly accessible to our all too unrefined senses. When it comes to artificial radioactivity and its steady increase, for example, only machines can record the exact results of the activity of other machines—in this case, nuclear power stations. What do you call a person without a Geiger counter or film badge dosimeter, and who doesn’t have any iodine pills handy, when a radiological emergency is underway? But the same goes for whatever domain you like, from air traffic control to the circulation of viruses, from the disappearance of ozone to the presence of dioxins, the only objective knowledge of which belongs to machines. It is logical, therefore, and necessary, that a language more well-adapted to the transmission of such precise knowledge might definitively supplant the old human language, forged from our so manifestly deficient sensory experience.

This long digression was necessary, as you can see, in order to reveal in its truest light the model which machine language offers our own newspeak, and, in the same stroke, to show just how unfounded are the regrets of those who weep for the lost richness of oldspeak, as though that richness had the slightest use for us anymore. When you have to measure computer performance in gigaflops,⁷ a unit corresponding to one billion operations per second, speaking or writing in oldspeak is akin to lugging yourself down the information highway in a horse-drawn carriage. But this obsolescence of old natural languages does not merely signify, as it might have seemed to back when rationalization was just beginning to subject us to the rigors of metrics and the certainty of statistics, that the positive science established by machines would substitute its own language for ours—that is, that it would penetrate our thinking about objectively defined notions. In reality, things do not happen so simply, and now we must distinguish between two tendencies at work in the formation of newspeak, or rather between two series of apparently contrary consequences that are due nonetheless to one and the same action of rationalization. Most commentators have gone astray in failing to grasp the profound unity of these contradictory effects.

While exact knowledge of phenomena and objective truth are no longer accessible by way of the limited means human language affords to human thought, the former nevertheless retains the function of translating, for use by the general population, what the machines are saying—that is to say, the decisions being taken by artificial intelligence. In order to fulfill this task efficiently, human language must become ever more rigorous, unequivocal, practical. The automation of thought, however, simultaneously produces an entirely opposite effect, because in all its other social uses language finds itself, for the first time in history, emancipated from the always difficult relationships it has maintained with the objective world. Up until now the strain of accounting for that world, or at least saying something about it, even if this something amounted to mere lies or fables, weighed heavily on language. Presently it has been relieved of this burden, and thus of any responsibility with regard to its veracity. As the post-modern philosophers in their time noted quite correctly, nothing is left but “language games.” Human speech can finally detach itself from the material reality in which it remained mired, and it is also, in the same stroke, delivered from the straitjacket of logic, and thus of syntax. What’s more, an equally complete expressive freedom largely makes up for the constraints that rationalization imposes upon modern subjectivity.

This last point is obviously decisive when it comes to unifying our understanding of the many, apparently quite diverse phenomena which, however, all contribute to the rise of newspeak. Interactions between the two idiomatic forms I’ve just defined are no less constant. A wholeheartedly playful, effervescent, and creative newspeak is constantly borrowing terms and phrases from the one that yields to technological constraints; and for its part the latter, in its production of communication geared toward a general population, also draws as a matter of necessity from the spontaneous inventions of “language games.” The profound unity of these two complementary forms of newspeak is readily apparent in the fact that they both conspire to abolish the constraints of our former syntax. It could have been said in an earlier time that oldspeak was the common opus of poets and manual laborers. We can say today that newspeak is the œuvre of the software engineers and creatives who produce our advertising and our culture.

Translated from the French by Louis Cancelmi