Cyberpositive

Objective

By the end of this volume you understand the text in which the reinterpretation of cybernetics was carried out, and you know the argument that carries it. You know why post-war cybernetics didn't overlook positive feedback but repressed it, and why a homeostatic model can explain nothing that has a history. You recognise the misunderstanding lodged in the word itself, and you know the objection on which Land's picture breaks.

Exposition

The text is called Cyberpositive, appeared in 1994 in the collection Unnatural, and carries two names, Sadie Plant and Nick Land. That joint authorship is more than a footnote. The text comes from the brief period when both poles of the group worked together, before one left and the other took over. What you read here is the moment when the cyberfeminism of Volume 9 and the anti-humanism of Volume 10 speak the same language. In the same year Land delivered the text Meltdown at the Virtual Futures conference at Warwick, driving the same movement to its extreme form.

The conceptual pair is the actual invention. Cybernegative denotes a system that damps its deviation, cyberpositive one that amplifies it. You know both from Volumes 2 and 3. What's new is that here a technical distinction becomes a valuation, and one that flips the sign of an entire discipline.

First clear away the misunderstanding lodged in the name. Positive here means no yes, no optimism, no enthusiasm for technology. It's a sign, as sober as the plus in front of a number. An avalanche is cyberpositive. A bank run is cyberpositive. A tumour is cyberpositive. The term says nothing about the value of a process, it describes only the direction of its loop. And precisely in that sobriety lies the trick of the word. Land affirms the cyberpositive, but he does it under cover of a term containing no affirmation. The stance slips in via the technical. Take the word for a confession and you've misread it. Take it for purely descriptive and you have too.

Now the argument, and it's sharper than the usual account admits. The common story runs that cybernetics restricted itself to negative feedback and overlooked positive feedback. That isn't true. Wiener and his successors knew positive feedback precisely. They had names for it, vicious circle, escalation, runaway. They treated it at length. But they treated it as what destroys a system when control fails. Positive feedback was present in cybernetics, but as a pathology, as a failure case, as what you build against.

Why that came about is told by the discipline's origin. Recall Volume 2. Cybernetics arose from a problem of anti-aircraft gunnery. Its primal scene is a gun hitting an aircraft. Its question is how a system holds its target when the world disturbs it. From that birth follows its entire order of values. It is a science of security, born in war, built out in the Cold War, on control loops, guidance systems, early warning networks. Its ideal is control, its enemy the slipping away. Plant and Land read the discipline against that origin. Post-war cybernetics didn't discover the truth about systems, it built out the part of the truth its clients needed. What it filed as a failure case was not the margin of the phenomenon but its centre.

Here stands the sentence that carries, and that goes beyond mere revaluation. A homeostatic model can explain nothing that has a history. A thermostat has no history. It holds twenty degrees, today as in a thousand years, that's all it can do. A system that corrects every deviation never brings forth anything that wasn't already in its set point. It can remain, it cannot become. With that, everything interesting at all drops out for negative feedback. Evolution, technology, language, capital, culture, all of that is novelty, and novelty is precisely what a control loop by definition prevents. The cybernetics of control has no model for history. It can explain states and no events. Try to describe culture with it and you get a picture without time.

From that follows the shift in the direction of time, and it's the actual yield of this volume. A negative loop needs a set point, and a set point was fixed in the past. The thermostat compares the present with a specification that was already there. It runs from behind. A positive loop has no set point. It compares with nothing, it runs towards something that doesn't yet exist and that only comes into being in the running. It has no memory, it has a direction. With that the question stands in the room that Land takes up in the next volumes. If a system has no goal behind it but one ahead of it, one it generates itself, what then determines it? From the past that can no longer be answered. Exactly at this point his model of time begins, and you can see now that it's no whim but a consequence.

Meltdown draws the consequence and shows at once what the price of this thinking is. The text plays through the earth as a system seized by a technical-capital escalation, and it ends with the announcement that nothing human emerges from the near future. The tone is no warning. The prose itself accelerates, overheats, loses the order of academic speech, you remember Volume 10. The text does what it speaks of. That's consistent, and it's a problem, because a text that enacts escalation rather than asserting it evades testing. You can undergo it and you can hardly contradict it.

The objection is technical in nature and therefore hard to dismiss. Pure positive feedback doesn't exist. Every real runaway ends. The avalanche comes to rest in the valley because the slope runs out. The feedback tone screams until the amplifier hits its power limit. The bubble bursts because the credit dries up. Escalation always needs a reservoir, and reservoirs are finite. What occurs in reality are never pure loops but couplings of amplifying and damping components, and the interesting systems are precisely those in which both interlock. Pure escalation is a mathematical idealisation with no physical instance.

The heavier part of the objection concerns Prigogine. Land appeals to him, and the appeal doesn't deliver what it promises. Prigogine's dissipative structures are not runaway systems, they are stable patterns that form far from equilibrium and hold there. The vortex from Volume 3 doesn't disintegrate, it persists. Such structures arise precisely because amplifying and damping feedback interlace into a form. Prigogine is a thinker of order, not of collapse. Land takes from him the rhetoric of distance from equilibrium and leaves behind the mechanics it comes from. That's no trifle, because the authority of a Nobel laureate here props up a picture his work doesn't cover.

What remains is less than the text claims and more than the objection leaves. The critique of the cybernetics of control lands. A science stemming from targeting and raising stability to a standard is no use for describing what transforms, and its blindness is not accidental but follows from its origin. The shift in the direction of time is a real gain and carries everything that follows. The claim, by contrast, that culture and capital are pure escalation is no inference from those insights but a stipulation that feeds off them. After this volume you hold a sharp critique in your hand and a vision that overshoots the critique. Keeping the two apart is the work that will accompany you through the whole fourth module.

Core Claim

The text Cyberpositive by Plant and Land flips the sign of cybernetics by declaring what it treated as a failure case to be the centre. Its carrying argument is that a science of homeostasis can explain nothing that has a history, because a system without deviation never brings forth anything. From that follows the inversion of the direction of time, since an amplifying loop runs towards a goal it generates itself rather than holding a set point from the past. The price is an idealisation, because pure escalation doesn't occur in reality, and Prigogine, on whom Land calls, describes order rather than collapse.

Bridge to the Next Volume

The direction of time is inverted, the goal lies ahead and gets generated in the running. Volume 16 draws the consequence that made Land famous and notorious. If a process runs from its expected future, then it can be described as something that assembles itself backwards out of the future, and capital acquires the status of an intelligence that uses the human as a passage. There the hyperstition of Volume 14 and the escalation of this volume meet.