The Stack On Software and Sovereignty

sector ’ s total footprint by 2020, while telecoms infrastructure, PCs and peripherals, will represent much more. While the shift toward Cloud platforms represents efficiencies over “ dead tree media ” and perhaps will ultimately reduce reliance on air travel, it also enables an exponential growth in data flotsam, such as search histories, redundant personal media, legally mandated trails of sales receipts, unfiltered spam, backups of the trails of spam and search, and so on, all of which need to be stored somewhere, and it is now stored online in various databases here and there. Sometime in the future, this Cloud landfill of postcontemporary data junk may provide new insights for digital humanities as to the real nature of global discourse (or existential clues for a future artificial intelligence, itself born of spam perhaps, ⁵⁵ seeking out the meaning of its origin), but until then, it is just more carbon debt. It is estimated that the electricity required to send the trillions of spam e-mails worldwide each year is equivalent to powering 2 million American homes and generates the same amount of greenhouse gases as 3 million cars. But the Cloud layer is not uniform, and how it affects the Earth layer depends on where it is buried. It matters where data centers are located because the available energy mix ranges from dirty fossil fuels to robust renewables sources (Hong Kong hosts one of the dirtiest clouds, while Iceland and Sweden are among the cleanest, and so Iceland has made Cloud hosting a key part of its national industrial strategy). In the United States, many of the largest data centers (such as Google ’ s in Lenoir, North Carolina, Yahoo ’ s in La Vista, Nebraska, Microsoft ’ s in Chicago and Apple ’ s in Apple, North Carolina) are, as of this writing, all at least 50 percent coal powered. ⁵⁶ The incentives to introduce greater energy efficiencies in data centers and to reduce operating costs thereby are considerable, and they represent critical competitive advantages for different Cloud platform players, as will be discussed further in the following chapter. Nevertheless, without significant gains in energy and carbon efficiency (and cost), it may be too expensive for the Cloud to “ grow ” beyond certain thresholds. If so, then less assured paths of innovation will ensue, many of which may further distort infrastructural access between the global North and South.

Even if all goes well, the emergent mega-infrastructure of The Stack is, as a whole, perhaps the hungriest thing in the world, and the consequences of its realization may destroy its own foundation. If growth estimates are correct (or, worse, if they are too modest), then the collective project of constructing The Stack may tip toward an equally comprehensive and self-amplifying exhaustion of resources and systemic collapse. That is the bad scenario, but it is not the only one. The Stack not only consumes energy; it also mediates it and rationalizes its metabolic distribution. As said, energy is not just the driver of The Stack; it is also one of the things that The Stack computes. Energy, regardless of its source, is not likely consumed at the site of its capture, but is instead shuttled around a regional-scale network of production, storage, transmission, distribution, and metering, and in many cases this network is startlingly inefficient.

94

Earth Layer

“ Electricity generation currently accounts for 57% of India ’ s total carbon emissions and will continue to do so until 2020. India ’ s power network is highly inefficient and much of the generated electricity is wasted. The lack of transparency in the grid makes losses difficult to measure, but it is estimated that in 2007 India lost 32% of total generation. ” ⁵⁷ For an economy of 1 billion people, one-third of the energy generated was lost in relay. That kind of signal-to-noise ratio cannot scale. To intervene, hardware consultants and equipment providers evangelize whole-cloth new grids, brought under the larger rubric of centralized information networks, such that grids for electrical “ packets ” working like grids for information packets would realize an “ Internet of energy. ” Even though to date ecstatic marketing visions far outpace reality, under the right pressure, incentive, and circumstance, and with enough patience, that may not remain true. ⁵⁸

Such energy-information networks (however theoretical or actual they may be at the time you read this book) are central to how the Earth layer functions within The Stack. They are also themselves designed as software-hardware platforms that can in principle reduce infrastructural costs and risks, which for some is earned through greater transparency and resilient decentralization and for others by greater centralization and system optimization. In principle, such grids can reduce peak demand crises through better energy storage, making the use of renewables for baseload energy more feasible, while also monitoring use through platform-standard metering making energy markets more liquid, resilient, accountable, and predictable. This may help to undistort market pricing of ecologically expensive energy use as well as the means to mitigate some of its associated risks — in theory, that is. For The Stack, this may allow for the production and consumption of flows of energy that are simple, omnidirectional, and ubiquitous. Any activity that generates more energy than it uses to sustain itself could invest that surplus into a vast metabolic agora where it powers and animates distant and complex projects, its muscular and cellular force captured, stored, and routed to where it can best collaborate with other aggregated exertions to power another unknown project. Yes, you are a battery. Our most visionary plots have the Stack ’ s carbon footprint measured not in debt but in surplus, and likely in our lifetimes or not, the geopolitics of a postscarcity Earth layer is worth articulating and defending as an ideal.

In the meantime, we note that many of the most important positive potential effects of ubiquitous computationally intensive, point-to-point energy flows are on “ non-Stack ” industries. The Climate Group ’ s Smart2020: Enabling the Low Carbon Economy in the Information Age report issues confident, sunny scenarios for carbon savings from ICT in five critical areas: smart grids, transportation, dematerialization, buildings, and information management. The key interventions include the more nimble transmission grids as discussed above, distributed energy storage systems, congestion pricing, vehicle-to-grid charging and energy storage, teleconferencing, desktop virtualization, building and facility management, fine-grain metering, and supply chain and logistical optimization. The conclusion of the report is that if ICT is more deeply integrated into the fabric of industrial economies, especially in China and India, it would realize a total carbon savings that is five times greater than the sector ’ s direct footprint based on projected growth (ICT ’ s direct footprint is estimated to be 1.4 GtCO2e in 2020, but the total ICT-enabled abatement is estimated to be a savings of 7.9 GtCO2e). According to this model, we cannot afford not to accelerate the construction of The Stack. This is the conundrum into which we are thrown: Can The Stack be built fast enough to save us from the costs of building The Stack?

Earth Layer

95

The Earth layer of The Stack is defined by this risk, also perhaps its most critical (and paradoxical) measurement and prediction challenge: the energy costs of planetary-scale computation on one side of the scale versus the energy savings of Internet on the other, the latter either rescuing us from the former or instead guaranteeing a catastrophe already underway. We may conclude that investment must be accelerated so that the costs of building The Stack do not sink the whole enterprise of industrial civilization, but if the energy and carbon costs of The Stack are too great to pay for the construction of the new grids, then the new grids cannot save us from the effects of those same costs. The Stack is in a race against its own physics, like a long-distance spaceship that must carry a prohibitive excess of fuel just to push the weight of that prohibitive excess of fuel. ⁵⁹ If disaster calls, The Stack would also itself be a causality of its own potentially disastrous impacts. Its own machines and materials are also vulnerable to the foreseeable and unforeseeable disruptions brought by the climate change that its own appetite would exacerbate and ensure. “ The rainfall from Typhoon Morakot caused rivers to flood in Taiwan flushing large volumes of sediment into the ocean. This led to several submarine landslides which broke at least nine communications cables 4000m down. It disrupted the Internet and telecommunications between Taiwan, China, Hong Kong and other parts of Southeast Asia ” ⁶⁰ Flooded data centers and compromised undersea cables can knock out whole networks, which is especially concerning when you realize that “ over 95% of global communications traffic is handled by just 1 million km of undersea fiber-optic cable. Rising sea levels increase the risk of flooding of coastal cable facilities and may also affect the stability of the seabed, making cables even more vulnerable. ” The really smart grid is the one that still works once the climatic effects of its construction come back to bite. In the composition of The Stack, we are conducting an experimental live-fire exercise on its Earth layer. Will the armature of The Stack organize a resilient computational geopolitics, or will its own energy thirst, ecological impact, and toxic production footprint finally overwhelm all, leaving behind elaborate labyrinthine ruins? Will planetary-scale computing prove to be, in some guise, the metaplatform of an alternative counterindustrialization, full of effervescent interfaces of metabolic exchange, or instead will it be the final machine, drawing us into its self-immolation?

96

Earth Layer

1. Sensing and Sovereignty; Polities of Supply and Effect

This risk is not equally shared, and political geography is also redrawn in the image of that unevenness, as sensor and sensed and cause and effect each become leverageable sovereign positions. In that governance is bound and determined by how its means allow it to see, measure, and organize its domains, the systems that mediate governance bind them to it just as it is bound by them. Nowhere is this truer than in the computational governance of ecologies, particularly for computational megaprojects such as Planetary Skin, but with the adoption of new media of observation and measurement come new complications. Is sovereignty primarily for the measurement or for the measured? The referent or the referred to? Who owns the data that all these nodes will be generating about themselves and their $U s e r s ? ^ { 6 1 }$ The answer will not only reflect sovereign claims; it can also generate them. For example, between the Himalayas and the Karakoram, near the slippery boundary between India and Pakistan, both governments as well as nongovernmental organizations, have planted a sizable number of sensors to detect ice temperatures, water flow, and other telltale effects of climate change. In these areas, formal and informal borders are unclear and move as the landscape shifts. Fresh water pathways vital to down-mountain settlements also shift, and flows that start on one side of the line may flow into the other. As development researcher, Sally Daultrey, observes the inscription of sensor grids and the flows of data they generate become active participants in these shifts. ⁶² India and Pakistan are largely unwilling to share the data they capture with one another and thus create a secondary border between their data sets. The very placement of sensors in certain locations and the control of the data generated by them become a way to claim some of the always shifting terrain from their rival, not only through the installation outpost sensor hardware, but in the claiming of the virtual profile of water or air that a sensor can sense. In other words, it is not only that when the sensor drifts with the ice pack into the sovereign territory of one of the countries, then the data it produces now belong to that country, but also that wherever one country ’ s sensor drifts into an ambiguous or contested new location, producing data about that location, then that location itself becomes that country ’ s own because its sensor is measuring and enumerating that site. Here the state “ takes ” the territory that it is capable of sensing, as the data generate and guarantee a sovereign claim as much as (in this case, more than) the sovereign claim guarantees the right to install a sensor and capture data.

Sensing begets sovereignty over the site that is sensed, except of course when it is does not. Recently China asked foreign embassies in Beijing and consulates in Shanghai to stop taking and reporting their own weather and air quality readings and asserted an exclusive right to acquire and report that information. ⁶³ In this case, while the site of a diplomatic embassy might be recognized as the sovereign satellite domain of another country, the Earth on which and in which it is situated is not; and neither is the weather that pours down on it. Similarly in the United States, particularly in dry western states, the issue of “ rainwater harvesting ” by individual parties is extremely contentious. State and federal authorities intervene to protect the interests of downstream residents who rely on access to a legally guaranteed flow versus upstream ranchers who wish to cap ture the water that falls on their land for their own use. Those upstream claim that “ the state is stealing the rain, ” while downstreamers claim that those living upstream are “ privatizing the weather. ” Under certain conditions, state management of an ecology might be seen as an injustice, whereas in others, the injustice is the absence of governmental intervention, though sometimes the terms of that ungovernance take unexpected forms. James Hansen, the former director of NASA ’ s Goddard Institute, who has done as much as anyone else to clarify our climatic precariousness, also helped guide a lawsuit based on the public trust doctrine, and inspired in part by the sixth-century Byzantine rule of Justinian I. That doctrine stipulates “ that common resources like water and air are held in trust by the government for future generations ” and that they must make good on this charge. ⁶⁴ The claim of the suit is that the federal government ’ s failure to suppress climate change is a dereliction of its essential sovereign duty and must be corrected as such. Another speculative alternative is Amy Balkin ’ s Public Smog project, which seeks to transform discrete volumes of the air above into new public parks through the issuance of emissions offsets. Maneuvers and innovations such as these may seem odd, but they may also represent a critical path of design intervention into the unstable geopolitical architectures of the Earth layer of The Stack. Wilsonian internationalist mechanisms have had a far too limited ability to enact and enforce effective solutions, as Hu and Obama ’ s flimsy “ compromise ” at the Intergovernmental Panel on Climate Change (IPCC) talks in Copenhagen in 2008 exemplify. Planetary Skin ’ s central bank for carbon could not rush in and save that particular day, as the essential problems of measuring a carbon economy (Who, how, when, where, why?) would require the most powerful national economies to disclose and share sovereign information about their industrial empires in ways that interested neither of the two most powerful prisoners in this particular dilemma. ⁶⁵ We are left knowing both that impending ecological calamity represents perhaps the most significant challenge to the very premise of governance that we face today, and also that the Westphalian-looped state is a dangerously awkward sovereign unit with which to assemble an effective quo rum. The Stack ’ s own mitigation program must also look elsewhere.

Earth Layer

97

We anticipate that with the ongoing convergence of planetary-scale computation with ecological governance, alternative jurisdictional forms will come to augment national domains and may, in crucial ways, have greater importance for those who live within those domains and circumstances than national identity. The alternative sovereignties that they bequeath (or rent ) may simply matter more to important outcomes, but the appearance of any such geographies requires new frames, procedural alliances, and counterintuitive techniques. Like species that fill a new niche after an ecological shift, new ecojurisdictions and their unorthodox spaces and organs are more likely to appear in direct response to an emergency situation than through any parliamentary deliberation. The latter may retroactively codify the former, but is unlikely to first introduce them. According to Schmitt, the state of emergency provides the generative exception (and sometimes vice versa), and as it draws eccentric lines, it forces positions to be taken in relation to them. But these “ emergencies ” should be understood in both senses of the word: first as a crisis for which conventional understandings and instruments are inadequate and therefore require or justify unconventional measures, and second as moments of the emergence of something that was genuinely and qualitatively not there before. For ecojurisdictions, these emerge around both the production of energy and the effects of that production. We see this already in the divided constituencies and alliances represented at the IPCC climate talks. Large oil- and gas-producing countries share certain interests in ecological governance, or in its prevention, that cut across ideology and continental location, whereas countries whose landscapes serve as important carbon sinks also seek common cause, as do those whose circumstance puts them at shared risks for particular kinds of disaster. Shared ecogeographic interests can matter more than party lines when the transversal economic solidarities of energy production, and its consequences, come to outweigh cartographic, historical, or ethnic proximity.

98

Earth Layer

First, we see subdivided energy polities bound by the kinds of energy that they produce and the specific demands of a particular process of extraction. Behind the slow crawl of petrocitizens, will there be something like an OPEC (Organization for the Petroleum Exporting Countries) for solar-, wind-, and geothermal-producing regions around the world? Would that be a consolidation of interests linking, for example, Japan, Iceland, and geothermal powers to leverage how that energy source is enumerated and calculated as a carbon currency or debt relief mechanism within a larger ecological economics? We strongly suspect that they would rely on different math from that offered by say, the oil geopolity convened under the flags of Saudi Arabia, Venezuela, and Texas. As part of a team commissioned by the European Commission, AMO (the research wing of Koolhaas ’ s OMA architectural studio) proposed a new map of Europe (now “ Eneropa ” ) based on a similar conjecture. In their map, different areas of Europe are redrawn such that Spain, France, and Sweden give way to the new regions of Solaria, the Tidal States, and Geothermia, respectively. ⁶⁶ However, perhaps instead of new federalisms, even discontiguous ones based on energy production, might we see instead see a fragmenting of geographies down into superlocal finely grained bioregional localities linked through the much more nimble sensing and calculating tools now at our disposal? Why start with the blunt brush of the latitudinal state scale? Why not instead thousands, or tens of thousands, of smaller compu-ecological microjurisdictions, some stable and others lasting only a season or two, many overlapping and interweaving like the imaginary lines that try to keep Indian ice out of Pakistani data sets? But if so, how do they enforce their varied interests? Whereas bioregions recognize the primacy of grounded situation, changing climates, like the Cloud, do not.

Earth Layer

99

At the same time, coalitions of producers can ’ t represent those bearing the brunt of global energy use. Ecopolities of those affected are equally important as polities of producers, and we already see them emerging by emergency. In the wake of Hurricane Sandy, New York City introduced a system of zones that differentiated proximity to rising coastal water vulnerability. Citizens living in zone A or zone B might be required to evacuate at different times and might be recognized for different rights of return from those not occupying the same block-by-block micropolity. In this case, the tactical invention of emergency ecojurisdictions as a governing geographical superimposition could not be clearer. As the emergencies that give rise to these inventions become less isolated incidents, these initially temporary jurisdictions become increasingly permanent, and the provisionality of the exception becomes normalized and concretized into a new territory that comes to make demands on its neighbors and citizens. For example, the Alliance of Small Island States emerged during the Kyoto Protocol discussions to represent the interests of low-lying nations that would be most egregiously affected by raising sea levels. The federation of (not exclusively) islands is spread across the world, concentrated along the equatorial belt, and has proven a formidable aggregate voice in global climate talks. Perhaps as well we will see an alliance of those threatened by desertification, or a federation of agricultural regions overrun by migrating insects, or, equally likely, a league of those who, in warmer subpolar regions, will become new agricultural powers, such as Russia and Canada.

These kinds of ecojurisdictions arise out of circumstances that may persist for generations to come, but the form they take at different geographical scales and their varying ability to demand and defend claims are heterogeneous and asymmetrical. It would seem unlikely therefore that these quasi-sovereign forms will congeal into one master format as regular as the Westphalian-looped state, and so the rights, claims, and forms of identities that different groups are likely to claim will remain diverse, contradictory, and unevenly effective. At the same time, emergency is as emergency does. Ecological instability remaps self-interest and geopolitical multipolarity in strange ways, and in some cases, the claims of those affected are truly existential, such as for “ drowning nations ” such as the Marshall Islands, Tuvalu, Maldives, and Kiribati. The international law questions provoked by their possible erasure from the surface of the planet are stunning and even bizarre, and no existing legal framework is well suited to answering them. ⁶⁷ If the entire land of a nation is permanently drowned, is it still a nation? Do former inhabitants still possess even fishing rights in the waters now on top of their homeland? What about the right to issue currency or passport? “ Environmental migrant ” is not a globally recognized legal category of refugee today, though it likely will be, but for whatever nation takes them in, what form of citizenship do the displaced retain? What effective sovereignties (e.g., monetary, legal, geographic) does the state that once represented them (or still does) retain in relation to those offered by their new hosts? Do countries that take in persons permanently displaced by climate change get land, water, and commerce rights in return? The exceptions that necessitate similar improvisations regarding sovereign geographies now scrambled by rapid ecological change will only get more painful, and the solutions only more convoluted and violent. Among the apparently uncomputable accidents of The Stack are the ghost sovereignties of the swelling Pacific, even as they exemplify the most critical geopolitical design questions now set in motion by and for its Earth layer.

100

Earth Layer

1. Designing for versus Designing with Emergencies

Many of those design questions can be evaluated only by first evaluating the different positions in which they would situate the designer. There is a world of difference between designing for, against, or with these conflicts; as often as not, attempting one thing may result in an opposite outcome. The Earth layer is held in place by both an absolutist interest in computational transparency and the debilitating contradictions and ultimate impossibility of that transparency. Telescoping between planetary and atomic scales, The Stack introduces synthetic computation deep down, deep up, and deep into abyssal scales of intricate matter. That introduction can begin with the provision of an Address that allows a location to communicate data across scales, and that addressability may or may not be motivated by, for example, an avaricious splitting open of the world, from the atmosphere to the atom, toward a mandate of full-spectrum dominance. But programs for total capture are also vulnerable to their own comprehensiveness. Their interconnectedness can make them brittle. ⁶⁸ The Stack works by vertical integration, across scales and across technological genres. This allows it to function as a core platform for multiple economies at once and to provide universal valuation and exchangeability only to the extent that participants can be represented as similar computational events. Because of this, it can also absorb, dissipate, and deflect forms of risk that might make any one layer more vulnerable, but at the same time, that integration can also distribute and amplify destabilizing forces and factors. This is not a bug; it is axiomatic of platform logics. In the constitution of a durable order through a hyperlinking of earthly energies and forms — things, people, words, cells, molecules — we recognize that this order can and will also be the generator of the disorder that it hopes to regulate. Philosopher Brian Massumi writes, “ The figure of today ’ s threat is the suddenly irrupting, locally self-organizing, systemically, self-amplifying threat of large scale disruption. The form of threat, fed by instability and metastability, is not only indiscriminate, it is also indiscriminable; it is indistinguishable from the general environment. ” The rough bargain of the universal platform, and of fullspectrum governmentality, is that the systemic feedback loops that give the structure life can also be dangerous or even fatal to the whole. For systems that link across scales, small events can quickly and unexpectedly scale up to threaten much larger processes, and supercomputing systems in particular not only provide no guarantee that they can contain those amplifying irruptions, they can easily magnify and intensify them. Besides rolling and interlocking component failure, the circularity of this infrastructural vulnerability is also the path by which the program of absolute transparency and computational omniscience comes to defeat itself by infinite regress.

Earth Layer

101

Some years ago I was at a panel with HP research scientist Stanley Williams, who recounted a remarkable assignment he once had. He was part of a group of computer scientists who had been asked by the US Department of Energy to propose an architecture for a computer that would be capable of a high fidelity predictive simulation of the entire planetary climate, and of monitoring and simulating the entire planetary climate in real time. The group concluded that such a machine would require zettaflop computing, that is, a system capable of $1 0 ^ { 2 1 }$ floating-point operations per second. At the time, the fastest supercomputers were measured in petaflops $( 1 0 ^ { 1 5 } ) ;$ ; on-tap exaflops $( 1 0 ^ { 1 8 } )$ are still years away. Williams discussed how it is possible that zettaflop computing would demand a very different processing architecture from one based on silicon transistors and might instead be based on photonics and nanoswitches hybridized with silicon to increase performance without loss of volume and power. He spoke of layers and layers of nanoscale wiring weaving into dense synaptic computing fields. Could it be done? Somewhat anticlimactically, Williams also shared the conclusion of the group ’ s report: based on current technology technology, the necessary specifications would mean that the computer would not only be roughly the size of Paris, but it would consume so much energy that it would be the single most significant anthropogenic climatic event that it itself would be modeling! Short of fundamental breakthroughs, his anecdote underscores the paradoxical recursivity that undergirds the demand for global ecological omniscience, especially for an accidental megastructure such as The Stack. ⁶⁹

With the promise of irruptive emergencies in mind, as well as Williams ’ s figure of a megamachine eaten by its own image of the world as it tries in vain to source enough power to measure and simulate its own power consumption, how then should we think about design in relation to Stack emergencies and emergences? Remember that for our discussion of Schmitt, the exception draws an exemplary space that exists both inside and outside the law at once. It is where authority is absolute but derived less from normal law than from the sovereign decision both granting and granted by the state of emergency, one taken ostensibly to protect the framework of law that it itself supersedes. In relation to this, design must be genuinely suspicious of remedies that valorize the amelioration of symptoms. States of emergency often begin as a “ temporary measure ” — precaution, pilot program, protective custody, probation, prototype — but over time, through both political shift and in the normalization that appears through a design ’ s ingenious accommodations of the emergency, the exception becomes the rule, and the rule becomes the rule of law because it is now is seen as the normal shape of things. In the face of emergencies, design is asked to provide adaptation for those negatively affected and to give form to accommodation such that a sovereign decision would not be fundamentally threatened by the emergency that it has called into being. Design ’ s job too often is to reform the emergency, so that over time, the exception sheds its temporary and provisional character as its effects settle out.

102

Earth Layer

Consider the stakes for what this enables and prevents. Among Isamu Noguchi ’ s most heartfelt projects, but not best known, is a series of schools, playgrounds and meeting halls he designed for the Japanese American internment camp at Poston, California, during World War II. As Bruna Mori writes, Noguchi feared that those camps would be permanent. He presumed that the emergency was perhaps not a temporary episode but the new state of affairs for his friends and extended family. As such, the camps would demand more careful design to make them as livable as possible for the unforeseeable future. Of course, at the time, no one could know whether the intern ments were temporary or not, and it is precisely this well-meaning making permanent of the provisional exception that is worrisome. Had Noguchi ’ s proposals been built to their full intended extent, the camps may very well have become more permanent than they were, and perhaps the freedom of the interned was to some degree aided by the lack of design accommodation undertaken to normalize their camp ’ s exceptionality. Today many well-meaning projects enroll design to work, like Noguchi ’ s did, to provide better comfort to those caught within real emergencies, such as refugees, displaced persons, and disaster victims. ⁷⁰ Without discounting the obvious help that good shelter provides to alleviate the acute misery of such experiences, we must be honest in seeing that accommodating emergency is also how a perhaps illegitimate state of exception is stabilized and over time normalized. If by sponsoring a commission, the perpetrator remains immunized from disruption while the interned and displaced are reformed and reconfigured, then design on behalf of the emergency aids both victims and victimizers. This danger is especially severe if we believe that a declaration of the emergency, and the political reality of its exceptions, is already dubious in its own right (such as an infinite war or terror, an ethnic profile that allows for the suspension of legal rights, or the wholesale capture of an entire country ’ s Internet traffic). Similar soft normalizations are also at work in some of the most easy-to-follow programs for climate change mitigation, including, and especially, those that promote the superi ority of preindustrial economies. Some accept buffering an anthropogenic economy against the tides of its own effects by a minimal decorative alterations, while others spiritualize their basic confusion regarding means and ends and seek solace in folk lore; many combine both in varying guises. Instead we should plot systemic interventions based on deeper scales of operation that might arrest the eventual immiseration of places, species, and landscapes by securing lines of flight for them. “ Restoration, ” yes, but as part of a wider geopolitics of geoengineering. We hope that the short-term self-congratulation of accommodation by innovation can be inverted along its own seams and made successful despite itself. We hope that design can take on a different role from the agent of immunization-through-mitigation on behalf of a bad-faith sovereign and can instead work less in response to the exceptional emergency than on behalf of the emergence itself. This design needs to graduate from forms of practice that defer so much agency to authorities who claim but cannot defend incredible decisions over the arbitrage of exceptions, and needs to stop providing paramedic gestures that give them support.

Earth Layer

103