James Bridle

There was a video on YouTube that I watched over and over again, until it got taken down. Then I found GIFs of it posted to news sites and watched those instead: concentrated bumps of the key moment, freebasing on the uncanny. A man in rubber boots and ield camoulage, a hunting rile slung over one shoulder, walks across the vast expanse of the Siberian tundra in springtime. The ground is green and brown, dense, lush with grasses, and extends perfectly lat in all directions, to the pale blue of a horizon that seems a hundred miles away. He takes long, loping steps, an expedition pace, enough to carry him far across the territory each day. But as he steps, the ground shimmers and ripples; the thick earth turns to liquid, and moves in waves.1 What appears as solid ground is merely a thin carpet of plant matter, an organic crust atop a newly shifting, soupy sea. The permafrost beneath the tundra is melting. In the video, it looks as if, at any moment, the ground might crack, the stalker’s boot might plunge through the surface, and he might be swept down by the undertow, lost beneath the sheets of green.

In fact, the opposite direction is more likely: the ground will thrust upward, spewing wet soil and warm gases into the air. In 2013, a mysterious explosion was heard in the far north of Siberia, and residents one hundred kilometres away reported a bright glow in the sky. Scientists, reaching the spot on the isolated Taimyr Peninsula several months later, discovered a vast, fresh crater, forty metres wide and thirty deep.

Taimyr reaches a peak of ive degrees Celsius in midsummer and plunges to minus thirty in the winter. Its bleak landscape is scattered with pingos: small mounds and hillocks formed as hydrostatic pressure pushes cores of ice toward the surface. As they grow, the pingos shed surface vegetation and shattered ice, coming to resemble ranges of squat volcanoes, cracked and cratered at their crowns. But the pingos, like the permafrost, are melting – and in some cases, exploding. In April 2017, researchers in Siberia installed the irst of a network of seismic sensors on the nearby Yamal Peninsula, whose name means ‘the end of the Earth’. Close to the brand-new port of Sabetta at the mouth of the Ob river, the sensors are capable of measuring movements in the ground over a 200-kilometre radius: they are intended to provide early warning of exploding pingos – and worse – that might damage the industrial infrastructure of the port or the local Bovanenkovskoye and Kharasavay gas deposits.

The establishment of Sabetta as an export point for the vast reserves of Siberian natural gas has been made possible by the same forces that have created the exploding pingos: rising global temperatures. As the Arctic ice melts, previously inaccessible reserves of oil and gas become viable. It’s estimated that 30 per cent of the world’s remaining natural gas reserves are in the Arctic.2 Most of these reserves are offshore, beneath less than 500 metres of water, and are now accessible due precisely to the catastrophic impact of the last century of fossil fuel extraction and dependence. The sensors installed to protect industrial infrastructure are necessitated by the conditions produced by the infrastructure itself. This is positive feedback: not positive for life – human, animal or plant – not positive for sense; but accumulative, expansive, and accelerating.

The underlying, localised form of positive feedback at work here is the release of methane by the melting permafrost: the slushy, trembling tundra. The permafrost that lies beneath the Siberian tundra can extend to depths of over a kilometre, made up of continuously frozen layers of soil, rock and sediment. Locked in this ice is millions of years of life, which is starting to return to the surface. In the summer of 2016, a disease outbreak that killed a young boy and hospitalised more than forty others in the Yamal Peninsula was blamed on the exposure of buried reindeer carcasses by the melting permafrost. The carcasses were infected with anthrax bacteria, which had lain dormant in the ice for decades or centuries, frozen in time beneath the tundra.3 Associated with this deadly bacteria is dead matter, which, as the ice melts, starts to decay, giving off plumes of methane – a greenhouse gas vastly more effective than carbon dioxide at trapping heat in the earth’s atmosphere. In 2006, the Siberian permafrost released an estimated 3.8 million tonnes of methane into the atmosphere; in 2013 that had risen to 17 million tonnes. It is this methane, more than anything else, that is causing the tundra to shudder and explode.

Of course, there is no such thing as a local effect in a networked world. What we perceive as weather in the moment shadows the globe as climate: tiny moments of turbulent activity through which we can barely grasp an unseen, unknowable totality. As the artist Roni Horn has observed, ‘Weather is the key paradox of our time. Weather that is nice is often weather that is wrong. The nice is occurring in the immediate and individual, and the wrong is occurring systemwide.’4 What appears on the tundra as an ever-increasing uncertainty of footing is the destabilisation of the entire planet. The very ground trembles, rots, ruptures, and stinks. It cannot be relied upon.

The exploded pingos and open melt lakes of the Siberian plain, seen from the air, resemble brain scans of spongiform encephalopathy patients, their cortexes pitted and scarred by the death of nerve cells. The prion diseases that cause spongiform encephalopathy – scrapie, kuru, mad cow disease, CJD and their derivatives – are the result of misfolded proteins, scraps of base matter that have become twisted into malformed shape. They spread themselves through the body by converting their properly folded counterparts into their own image. When prion infections reach the brain, they cause rapid-onset dementia, memory loss, personality changes, hallucinations, anxiety, depression, and ultimately death. The brain itself comes to resemble a sponge, hollowed out and denatured, unable to make sense of itself and its end. The permafrost – the permanent frost – is melting. The words don’t make sense any more, and with them go the ways we have to think the world.

On June 19, 2006, representatives from ive Nordic countries gathered on the remote Arctic island of Spitsbergen, part of the Svalbard archipelago, to lay the irst stone of a time machine. Over the next two years, workers dug 120 metres into a sandstone mountain, where they excavated caverns another 150 metres long and 10 wide. The time machine is intended to transport one of humanity’s most precious resources to an uncertain future, bypassing certain horrors of the present. In heat-sealed foil packets, packed into plastic cases on racked industrial shelving, rest millions upon millions of preserved seeds: samples of food crops from regional collections around the world.

Just 1,120 kilometres from the North Pole, Svalbard is the most northerly year-round settlement on earth, and in spite of its remoteness it has long been an international meeting point. Visited by Norse ishermen and hunters since at least the twelfth century, its ‘discovery’ by Dutch explorers in 1596 opened up the islands to whaling and mineral exploitation. The British landed in 1604 and started hunting walrus; by the end of the century, the Russians arrived seeking polar bear and fox fur. Although driven out in the 1820s by British raids on the Barents Sea, they were to return, like everybody else, for the coal. During the Second World War, the archipelago was evacuated and occupied by a detachment of German troops manning a meteorological station. Cut off in May 1945, it wasn’t until late September that they were picked up by a Norwegian seal hunting vessel, becoming the last German troops to surrender to the Allies.

The discovery of coal deposits at the end of the nineteenth century sharpened questions of sovereignty that had previously been left open. For centuries, the archipelago had functioned as a free territory without laws or regulations, outwith the jurisdiction of any nation. The Svalbard Treaty of 1920, formulated as part of the Versailles negotiations, handed sovereignty to Norway but gave equal rights to all signatories to engage in commercial activities – primarily mining – on the islands. The archipelago was to be demilitarised, and to this day it remains a uniquely visa-free zone: anyone may settle and work on the islands regardless of country of origin or citizenship, provided they have some means of support. Alongside nearly 2,000 Norwegians and almost 500 Russians and Ukranians, Svalbard is home to several hundred non-Nordic people, including Thai and Iranian workers. In recent years, a number of asylum seekers whose applications have been rejected in Norway have made their way to Svalbard to wait out the seven years residency required to gain Norwegian nationality.5

The Svalbard Global Seed Vault – often referred to as ‘the ark’, or the ‘doomsday vault’ – was opened in 2008. As a backup facility to support the work of gene banks around the world, the Svalbard location is doubly suitable. Its zone of geopolitical exception makes it signiicantly easier to persuade national organisations to store their precious – and often conidential – collections there. And buried beneath the permafrost, the vault is also a natural deep freezer: powered by locally mined coal, it’s refrigerated to minus eighteen degrees Celsius, and even if these machines were to fail, the local bedrock remains below freezing all year round. The Seed Vault is an attempt to create a sanctuary that is both geographically and temporally isolated: suspended in neutral territory and the deep time of the Arctic winters.

Seed banks are crucial to maintaining some semblance of genetic biodiversity. They are the fruit of a movement that started in the 1970s, with the realisation that the Green Revolution in agriculture was causing farmers to abandon their usual seeds, locally developed over centuries, for new hybrids. India was believed to have over 100,000 varieties of rice a century ago; today it has only a few thousand. The number of apple strains in the Americas has dropped from

5,000 to a few hundred. Altogether, the UN Food and

Agriculture Organisation estimates, 75 per cent of crop biodiversity has been lost.6 Such diversity is essential to countering the risk of new diseases or pests that might emerge, threatening to wipe out homogenous varieties. The Svalbard collection is intended to provide secure storage for diverse strains in case of catastrophe: technically on long-term loan, its contents are not meant to be accessed unless all other sources have failed. In January 2012, the national seed bank of the Philippines was destroyed by ire, six years after it had been heavily damaged by looding, while those of Afghanistan and Iraq have been completely destroyed by ighting.7 In 2015, the International Center for Agricultural Research in the Dry Areas (ICARDA) requested the irst withdrawal from the vault: 130 of the 325 boxes it had deposited, containing a total of 116,000 samples.

ICARDA was established in 1977, with its headquarters in Aleppo, Syria, and branches across the Middle East, North Africa, and Central Asia. Its work is focused on the particular needs and risks of food security in the region: the development of new crop varieties, water management, conservation, and rural education, particularly that of women. In 2012, rebel ighters in the Syrian civil war seized control of the center’s gene bank twenty miles south of Aleppo, where it maintained a unique collection of 150,000 different populations of wheat, barley, lentil and faba bean seeds from 128 countries. While some of the staff were allowed to remain to maintain the facility, ICARDA was forced to move its headquarters to Beirut, and its access to the collection was cut off.

The ICARDA collection – backed up for now in Svalbard and shortly to be redistributed to Morocco, Turkey, and elsewhere – specialises in crops adapted to the harsh environmental conditions of the Middle East and North Africa. The beneit of the biodiversity inherent in this archive, evolved and engineered by farmers and nature over generations, is not disease and pest resistance, but climate resilience. It is from this resource that scientists hope to mine new genetic traits to moderate the ravages of climate change – for instance, by splicing heat- and drought-resistant crops such as chickpeas and lentils with maize and soybeans to make the latter viable in rapidly changing, and heating, ecosystems.8

This change is so rapid that it has taken even the Global Seed Vault by surprise. The year 2016 was the hottest ever recorded – for the third year in a row, with research indicating that the earth hasn’t been this warm for 115,000 years. In November, scientists reported that Arctic temperatures were up to twenty degrees Celsius higher than average, with sea ice levels 20 per cent below their twenty-ive-year average. In Svalbard, heavy rain fell in place of light snow, and the permafrost started to melt. An inspection of the vault in May of 2017 found that the entrance tunnel had been looded by meltwater, refreezing as it fell below the surface to form an indoor glacier that had to be hacked out to access the seedbank. Intended to function for long periods without human intervention, the vault is now under twenty-four-hour watch, with emergency waterprooing being added to the entrance tunnel, and trenches being dug around the site to channel meltwater away. ‘The Arctic and especially Svalbard warms up faster than the rest of the world. The climate is changing dramatically and we are all amazed at how quickly it is going,’ Ketil Isaksen, a Norwegian meteorologist, told reporters.9

Climate change is already occurring, and its effects are as visible and urgent in the landscapes of geopolitics as of geography. The Syrian conlict, which forced the ICARDA scientists to lee to Beirut and call on the Seed Vault for assistance, is itself partly attributable to changes in the environment.10 Between 2006 and 2011, more than half of the Syrian countryside suffered its worst drought on record. More intense and longer lasting than could be explained by natural variations in weather, this drought has been linked to accelerating climate change, and over a few years nearly 85 per cent of rural livestock died, as crops withered. President Bashar al-Assad redistributed traditional water rights to political allies, forcing farmers to dig illegal wells, while those who protested faced imprisonment, torture, and death. More than a million rural villagers led the countryside for the cities. When this rural resentment and demographic pressure met the totalitarian oppression already bearing down on the cities, it provided the inal trigger for an uprising that spread rapidly through the most drought-stressed areas. Media reports and activists have called the Syrian conlict the irst large-scale climate war of the twenty-irst century, connecting climate directly to the vast numbers of refugees arriving in Europe. Scientists are more circumspect about making explicit connections between conlict and climate – but not about the changing climate itself. Even if Syria recovers politically in the next few years, it stands to lose nearly 50 per cent of its agricultural capacity by 2050. There is no going back from here.

Why should we be so concerned with the Seed Vault? The Seed Vault is vitally important because it is a bastion not only of diversity, but of diversity-in-knowing and diversity-as-knowing. The Seed Vault transports things – stuff, knowledge, and ways of knowing – from an uncertain present into an even less certain future. It’s fuelled not merely by the stuff, but by the sheer variety of the stuff, that it carries. The Seed Vault’s fuel is heterogenous; it’s motley and incomplete: because this is the nature of knowledge and the world. It’s a necessary opposition to a monoculture – in this case, not even a metaphor, but a literal monoculture of plant strains engineered for speciic geographic and temporal tasks that, when generalised, fail utterly to accommodate the messy incoherence of the world as it actually is. The climate crisis is also a crisis of knowledge, and of understanding; it is a crisis of communication, and of knowing, in the past, the present, and the future.

In the Arctic regions, everyone is a climate scientist. Archaeologists searching for the remains of ancient cultures are digging into the deep history of the planet to pull out evidence that might assist us in understanding how the earth – and humans – behaved in past periods of rapid climatological change, and thus how we might address them now. On the western coast of Greenland, on the shores of the great Ilulissat Icefjord, the permafrost surrounding the ancient settlement of Qajaa preserves the relics of three civilisations, each of which occupied the same site over the previous three and a half thousand years. These are the Saqqaq, Dorset, and Thule cultures, the irst of which established itself in southern Greenland around 2500 bce, with the subsequent groups slowly supplanting their predecessors until contact with Europeans intensiied in the eighteenth century. The history of each of these cultures comes down to us through middens: layers of kitchen and hunting refuse laid down by generations, sinking into the earth and waiting for archaeologists to delve into them.

These middens have helped us to make sense of population movements and previous environmental happenings. What occurred in the Greenlandic cultures is not culturally unique, but it is archaeologically unique. Unlike Stone Age sites around the world, where only stone remains, the Arctic sites, thanks to the deep freeze of the permafrost, preserve far more information about ancient human material culture. The middens in Qajaa contain wooden and bone arrows, hafted knives, spears, sewing needles and other objects that have not survived elsewhere on the planet. They also contain traces of DNA.11

Like the entangled history and future of the seed banks, understanding how earlier civilisations and cultures adapted, changed, coped or failed to cope under previous periods of environmental stress is one way in which we might be able to respond to our own – if that understanding is not itself destroyed before we can reach it.

In the next century, these unique archaeological deposits – repositories of knowledge and information – will disappear entirely, after thousands of years of stability. Researchers from the University of Copenhagen’s Center for Permafrost drilled into the earth surrounding the Qajaa midden and another site in northeast Greenland and excavated cores of frozen soil, which, packed into plastic bags, were kept frozen on the journey back to the laboratory, where they were examined for signs of heat production. As the earth warms, long-dormant bacteria in the soil start to wake and become active. The bacteria themselves produce heat, causing the soil to warm further, thawing and awakening more bacteria – more positive feedback. As the ice melts and the water starts to drain away, oxygen lows into the layers of soil, breaking them up and degrading them. The newly awakened bacteria start to feed on the organic residues, leaving nothing behind but stone, and venting more warming carbon as they go. ‘When the ice melts and the water is drained’, writes Professor Bo Elberling, leader of the study and head of the Center for Permafrost, ‘there’s no way back.’12

In a report from the Greenland ice sheet in October 2016, Thomas McGovern, a professor of archaeology who has worked on the middens for decades, detailed how the rapid melting of the ice sheet is reducing to mush an archaeological record that stretches back millennia, and which we have barely started to comprehend:

Back in the old days, these sites were frozen most of the year. When I was visiting south Greenland in the nineteen-eighties, I was able to jump down in trenches guys had left open from the ifties and sixties, and sticking out the sides you could see hair, feathers, wool, and incredibly well-preserved animal bones. We’re losing everything. Basically, we have the equivalent of the Library of Alexandria in the ground, and it’s on ire.13

McGovern’s statement is deeply troubling in two particular ways. The irst is the terrible feeling of loss, as the possibility of accessing our own past and knowing more about it slips away from us at the very moment it might be of greatest use. But the second is more existential: it relates to our deep need to discover ever more about the world, to gather and process more data about it, in order that the models that we build of it may be more robust, more accurate, and more useful.

But the opposite is occurring: our sources of data are slipping away, and with them the structures by which we have structured the world. The melting of the permafrost is both danger sign and metaphor: an accelerating collapse of both our environmental and our cognitive infrastructure. The certainties of the present are founded on the assumption of ever-increasing, ever-crystallising geologies of knowledge; it is reassuring to imagine a cooling earth, coming into shape, manifesting in distinct and solid forms. But, as in Siberia, the sponging of the Greenlandic landscape reiterates a return to the luid: the marshy and boggy, the undifferentiated and gaseous. A new dark age will demand more liquid forms of knowing than can be derived from the libraries of the past alone.

Knowledge derived or uncovered from the past is one approach to coping with the catastrophic impacts of climate change. But our existing technologies and processes should also be capable of shielding us, to some extent, from its excesses – that is, if those technologies and cognitive strategies are not themselves among the earliest victims of climate change.

The Council for Science and Technology, an advisory body to the UK government, published a report in 2009 entitled ‘A National Infrastructure for the 21st century’, examining the future of the country’s communications, energy, transport and water networks. The report emphasised that the UK’s national infrastructure, like the internet, constituted ‘a network of networks’ – and a fragile one at that, fragmented in delivery and governance, unclear in its responsibilities and accountabilities, largely unmapped and chronically under-supported. The root causes of this situation identiied by the study included government siloing, public and private under-investment, and a general lack of understanding of how such complex networks of matter and knowledge even begin to function – let alone how they fail.

The report was clear about one challenge however, which would and must trump all other concerns – the changing climate:

Resilience against climate change is the most signiicant and complex longer-term challenge. The effects of climate change are predicted to cause higher summer and winter temperatures, sea-level rises, a rising intensity of storms, forest ires, droughts, increased looding, heatwaves and alter resource availability, e.g. of water. The challenges for the current infrastructure are both to adapt to such impacts and to support the radical transition to a low carbon economy. The Government’s National Security Strategy, published in March 2008, recognises climate change as potentially the greatest challenge to global stability and security, given expected world-wide impacts. Effective adaptation is key to mitigating this risk, in relation to infrastructure and other areas.14

Again, what is striking about the direct effects of climate change predicted in the report is their luidity and unpredictability:

Pipe systems for both drinking water supply and sewage will be more prone to cracking as climate changes lead to greater soil movement as a consequence of wetting and drying cycles . . . Dams will be more prone to siltation resulting from increased soil erosion, and the slippage risk to soil dams from intense rainfall events will also increase.

Another report for the UK government, published the following year by environmental consultancy AEA, explores the speciic impacts of climate change on information and communications technologies.15 ICT, in this context, is deined as ‘the whole of the systems and artefacts which enable the transmission, receipt, capture, storage and manipulation of voice and data trafic on and across electronic devices’ – that is, everything we might consider to be part or artefact of our contemporary digital universe, from ibre-optic cables, aerials and antennae to computers, data centres, telephone exchanges, and satellites. Outside of the scope of the study, for example, are power lines, despite the essential nature of their services to ICT. (The Council for Science and Technology’s study, on the other hand, notes that ‘one of the limiting factors for the transfer of electricity by overhead transmission lines is their thermal capacity, which is affected by the ambient air temperature. Higher global peak temperatures will reduce those limits and hence the capacity of the network to transfer electricity.’)16

Reports written for governments are often far starker and clearer than governments’ own statements and policies. As in the United States, where the US military has put into action ten-year plans for adapting to climate change even while deniers take charge of the executive branch, so the British reports take climate science at face value, making for startlingly lucid reading on the value of networks:

All of the above artefacts work together as a system – interconnected, interdependent and completely enmeshed in each other and working to absolute rules of inter-operability. ICT is the only sector of infrastructure that directly connects any one user to any other user across time and space using multiple pathways simultaneously and capable of dynamic re-routing in real time. As such, in this case, the national asset is the network rather than any of the individual components – and it is the operation of the network that relies on the whole infrastructure and enables the generation of value . . . whilst the network is the asset at the level of infrastructure, the value of the network is not in the asset itself but in the information that travels on it. Nearly the whole of the economy relies upon the ability to transmit, receive and convert streams of digital data in close to realtime – whether it is the extraction of cash from an ATM, the use of a credit or debit card, sending an email, controlling a remote pump or switch, despatching or receiving aircraft or a mundane phone call.17

Contemporary information networks are both the economic and cognitive frameworks of society: So how will they fare in an era of climate change? And what damage are they doing in the present?

Rising global temperatures will particularly stress data infrastructures that already run hot, as well as the people who work in and around them. Data centres and individual computers generate vast amounts of waste heat, and require corresponding quantities of cooling, from the acres of air conditioning systems on industrial buildings to the fans that cool your laptop when a YouTube kitten video sends the CPU into overdrive. Increased air temperatures bring increased cooling costs – and the possibility of outright failures. ‘iPhone needs to cool down before you use it’ pleads the error message on Apple’s latest phone when the ambient temperature rises above forty-ive degrees Celsius. Such a response can be triggered by leaving the device in a hot car in Europe today, but is projected to become a daily occurrence in the Gulf regions in the second half of the twenty-irst century, following record-breaking heatwaves in 2015, when Iraq, Iran, Lebanon, Saudi Arabia and the Emirates endured daytime temperatures approaching ifty degrees Celsius.

The AEA report on ICT and the climate identiies a number of speciic effects that will be felt by information networks. At the level of physical infrastructure, it notes that much of this network is parasitic upon structures that were not designed for their contemporary uses, nor for the effects of climate change: mobile phone masts grafted onto church steeples, data centres in old industrial units, telephone exchanges constructed in Victorian post ofices. Below the ground, ibre-optic cables run through sewage channels that are becoming incapable of handling increased storm surges and looding; cable landing sites, where the internet comes ashore from undersea data links, are susceptible to rising sea levels, which will be particularly destructive in southeast and eastern England, sites of crucial connections to the continent. Coastal installations will be increasingly susceptible to saline corrosion, while towers and transmission masts will buckle and fall as the ground, attacked by drought and lood, shears and subsides.

In the electromagnetic spectrum, the strength and eficacy of wireless transmission will be reduced as temperatures rise. The refractive index of the atmosphere is highly dependent on humidity and severely affects the curvature of electromagnetic waves, along with the rate at which they fade. Increased temperatures and rainfall will shift the beams of point-topoint data links – such as microwave transmissions – and attenuate broadcast signals. As the earth warms and becomes wetter, ever-greater densities of wireless masts will be required, and maintenance will become more dificult. Changing types of vegetation may also impact the propagation of information.

Wi-Fi, in short, will get worse, not better. In one scenario, the shifting ground may even reduce the reliability of reference data for telecommunication and satellite transmission calculations. Accuracy falls; broadcasts overlap and interfere; noise crowds out the signal. The systems we have built to collapse time and space are being attacked by space and time.

Computation is both a victim of and a contributor to climate change. As of 2015, the world’s data centres, where exabytes of digital information are stored and processed, consumed about 3 per cent of the world’s electricity – and accounted for 2 per cent of total global emissions. This is about the same carbon footprint as the airline industry. The 416.2 terawatt hours of electricity consumed by global data centres in 2015 exceeded that of the whole United Kingdom, at 300 terawatt hours.18

This consumption is projected to escalate massively, as a result of both the growth of digital infrastructure and the positive feedback from rising global temperatures. In response to vast increases in data storage and computational capacity in the last decade, the amount of energy used by data centres has doubled every four years, and is expected to triple in the next ten years. A study in Japan suggested that by 2030, the power requirements for digital services alone would outstrip the entire nation’s current generation capacity.19 Even technologies that make explicit claims to radically transform society are not exempt. The cryptocurrency Bitcoin, which is intended to disrupt hierarchical and centralised inancial systems, requires the energy of nine US homes to perform a single transaction; and if its growth continues, by 2019 it will require the annual power output of the entire United States to sustain itself.20

Moreover, these igures relect processing power, but do not account for the wider network of digital activities empowered by computation. These activities – dispersed, fragmented, and often virtual – also consume vast resources, and are, by the nature of contemporary networks, dificult to see and string together. Immediate and local power requirements, easily visible to and quantiiable by the individual, are negligible compared to the cost of the network, just as individual waste production and management, apparently mitigated by ethical shopping and recycling, pale in comparison with globalised industrial cycles.

A 2013 report, ‘The Cloud Begins with Coal – Big Data, Big Networks, Big Infrastructure, and Big Power’, calculates that ‘charging up a single tablet or smart phone requires a negligible amount of electricity; using either to watch an hour of video weekly consumes annually more electricity in the remote networks than two new refrigerators use in a year.’21 And this report is not the product of a worthy, well-intentioned environmental group. Rather, it was commissioned by the National

Mining Association and the American Coalition for Clean Coal Electricity: it is a lobbying call for more fossil fuel use, in order to meet inevitable demands.

What the coal giants point out, perhaps unwittingly, is that data usage is qualitative as well as quantitative. What we look at turns out to matter more than how we look at it – and not just to the environment. One industry consultant quoted in the newspapers argued, ‘We need to be more responsible about what we use the internet for . . . Data centres aren’t the culprits – it’s driven by social media and mobile phones. It’s ilms, pornography, gambling, dating, shopping – anything that involves images.’22 As in most proto-environmental claims, the proposed solutions are either appeals to regulation (taxing data), conservative regressions (banning pornography, or switching from colour to black-and-white photographs to save transmission costs) or hapless techno-ixes (like the miracle-material graphene) – all ludicrous, unworkable, and unable to think at the scale of the networks they seek to address.

As digital culture becomes faster, higher bandwidth, and more image-based, it also becomes more costly and destructive – both literally and iguratively. It requires more input and energy, and afirms the supremacy of the image – the visual representation of data – as the representation of the world. But these images are no longer true, and none less so than our image of the future. As the past melts into the permafrost, so is the future rocked by the atmosphere. The changing climate shakes not merely our expectations, but our ability to predict any future at all.

Just after midnight on May 1, 2017, Aerolot’s regular service from Moscow to Bangkok, Flight SU270, hit a pocket of violent turbulence as it approached its destination.23 Without warning, passengers were thrown from their seats, some of them crashing into the ceiling of the aircraft before falling onto their neighbours and into the aisles. Footage recorded onboard showed dazed and bloody passengers lying in the aisles, surrounded by scattered food trays and luggage.24 On landing, twenty-seven passengers were rushed to hospital, several with fractured or broken bones.

‘We were hurled up into the roof of the plane, it was practically impossible to hold on,’ one of the passengers told reporters. ‘It felt like the shaking wouldn’t stop, that we would just crash.’ The Russian Embassy told Reuters that ‘the reason behind the injuries was that some of the passengers had not had their seatbelts fastened.’ Aerolot asserted in a press release that ‘an experienced crew piloted the light. The pilot has more than 23 thousand light hours, and the co-pilot has over 10.5 thousand light hours. However, the turbulence that hit the Boeing 777 was impossible to foresee.’25

In June of 2016, a ‘brief moment of severe turbulence’ over the Bay of Bengal caused injuries to thirty-four passengers and six crew members aboard Malaysian Airlines light MH1 from London to Kuala Lumpur.26 Food trays cannoned out of the galley, and news agencies showed passengers being taken off on stretchers, and wearing neck braces.

Three months later, a United Airlines Boeing 767 en route from Houston to London had to make an emergency landing at Shannon Airport in Ireland following ‘severe and unexpected turbulence’ in the mid Atlantic. ‘It fell four times in a row,’ said one passenger.

It was a tremendous pull on the body. And on the third or fourth time babies started waking up and crying, people were waking up disorientated. I thought: this is not turbulence. This is what feels like a life-threatening drop. This is not like any feeling I have had. This is immediately like an experience of being ired from a cannon. It pulls you down so hard then it stops for a second and then it does that four times in a row. If you didn’t have your seatbelt on you would have smashed your head.27

The light was met by ambulances on the runway, and sixteen people were taken to hospital.

The most severe episode of clear-air turbulence on record hit United Airlines Flight 826 en route from Tokyo to Honolulu in 1997. Two hours into the light, minutes after the captain turned on the fasten seat belt sign in response to warnings from other aircraft, the Boeing 747 dropped downwards and then rebounded with such force that one of the crew, a purser who had been steadying himself on a countertop, found himself upside down with his feet high in the air.

A passenger whose seat belt was not fastened left her seat, hit the ceiling, and fell into the aisle. She was unconscious and bleeding heavily, and, despite resuscitation attempts by light attendants and a passenger doctor, was pronounced dead shortly after. Her autopsy revealed severe spinal damage. After the plane turned around and landed safely back in Tokyo, ifteen passengers were treated for spine and neck fractures, and another eighty-seven for bruises, sprains, and minor injuries. The airframe was retired and never lew again.

A report by the US National Transportation Safety Board later found that sensors on the aircraft recorded a peak normal acceleration of 1.814 G in the irst sharp ascent, before plunging to an extreme negative G of −0.824. It also sustained an uncontrolled roll of eighteen degrees – without any visual or mechanical cues to the pilot of what was about to occur.28

 Turbulence can be determined to some extent by the study of the weather. The International Civil Aviation Organisation (ICAO) issues daily ‘signiicant weather charts’ that include information about cloud height and cover, wind speed, weather fronts, and possible turbulence. The main indicator used to determine the possibility of turbulence is the Richardson number – that same Lewis Fry Richardson who developed the measure in a series of meteorological papers in the 1920s related to his work on numerical weather prediction. By examining the relative temperatures and wind speeds in different zones of the atmosphere, it is possible to determine the potential turbulence between them, if such measurements are available.

Clear-air turbulence is so named because it comes literally out of the blue. It occurs when bodies of air moving at wildly different speeds meet: as the winds shear against each other, vortices and chaotic movements are produced. While much studied, particularly in the high troposphere where long-haul aircraft operate, it remains almost impossible to detect or to predict. For this reason, it is much more dangerous than the predictable forms of turbulence that occur on the edges of storms and large weather systems, because pilots are unable to prepare, or route around it. And incidences of clear-air turbulence are increasing every year.

While anecdotal accounts of turbulence such as those above may be widely reported, many incidents, while globally significant, are not reported, and igures are hard to come by. An advisory circular on preventing turbulence-related injuries, published by the US Federal Aviation Administration in 2006, states that the frequency of turbulence accidents has increased steadily for more than a decade, from 0.3 accidents per million departures in 1989, to 1.7 in 2003.29 These igures are already severely out of date.

The reason for the increase in turbulence is rising levels of carbon dioxide in the atmosphere. In a paper published in Nature Climate Change in 2013, Paul Williams of the National Centre for Atmospheric Science at the University of Reading and Manoj Joshi from the School of Environmental Sciences at the University of East Anglia lay out the implications of a warming atmosphere on transatlantic aviation:

Here we show using climate model simulations that clear-air turbulence changes signiicantly within the transatlantic light corridor when the concentration of carbon dioxide in the atmosphere is doubled. At cruise altitudes within 50–75°N and 10–60°W in winter, most clear-air turbulence measures show a 10–40 per cent increase in the median strength of turbulence and a 40–170 per cent increase in the frequency of occurrence of moderate-or-greater turbulence. Our results suggest that climate change will lead to bumpier transatlantic lights by the middle of this century. Journey times may lengthen and fuel consumption and emissions may increase.30

The authors of the turbulence study emphasise once again the nature of feedback in this rise in turbulence: ‘Aviation is partly responsible for changing the climate, but our indings show for the irst time how climate change could affect aviation.’ These effects will be felt the most in the busy air corridors of Asia and the North Atlantic, causing disruption, delays, and damage. The future will be bumpy, and we are losing our ability even to predict the shocks.

I grew up in the suburbs of South London, beneath the inbound lightpaths of Heathrow Airport. At 6:30 p.m. every evening Concorde would rumble overhead, inbound from New York, shaking the doors and window frames like a rocket ship. It had been lying for more than a decade at that point; the irst light was made in 1969, and scheduled services began in 1976. Transatlantic lights took three and a half hours – if you could afford a ticket, which at its lowest cost something in the region of £2,000 for a return light.

 In 1997, the photographer Wolfgang Tillmans showed a series of ifty-six photographs of Concorde that correspond almost perfectly with my own memory: a dark arrowhead rumbling across the sky, seen not from the luxury cabin, but from the ground. Writing in the exhibition catalogue, Tillmans remarked,

Concorde is perhaps the last example of a techno-utopian invention from the sixties still to be operating and fully functioning today. Its futuristic shape, speed and ear-numbing thunder grabs people’s imagination today as much as it did when it irst took off in 1969. It’s an environmental nightmare conceived in 1962 when technology and progress was the answer to everything and the sky was no longer a limit . . . For the chosen few, lying Concorde is apparently a glamorous but cramped and slightly boring routine whilst to watch it in the air, landing or taking-off is a strange and free spectacle, a super modern anachronism and an image of the desire to overcome time and distance through technology.31

Concorde made its inal light in 2003, a victim as much of its own elitism as the fatal crash of Air France Flight 4590 into the Parisian suburbs three years earlier. For many, the end of Concorde was the end of a certain idea of the future.

There is little left of Concorde in contemporary aircraft: instead, the latest passenger aircraft are the result of incremental advances – better materials, more eficient engines, adjustments to wing design – rather than the radical advance that Concorde proposed. The last of these is my favourite addition: the ‘winglets’ that now adorn the wingtips of most aircraft. These are a recent invention, developed by NASA in response to the 1973 oil crisis and gradually retroitted for commercial aircraft to increase fuel eficiency. They always bring to mind the epitaph of Buckminster Fuller, as written on his gravestone in Cambridge, Massachusetts: ‘Call me trimtab.’ Tiny in-light adjustments, performed at scale. This is what we remain capable of.

History – progress – does not always go up and to the right: it’s not all sunlit uplands. And this isn’t – cannot be –  about nostalgia. Rather, it is about acknowledging a present that has come unhinged from linear temporality, that diverges in crucial yet confusing ways from the very idea of history itself. Nothing is clear anymore, nor can it be. What has changed is not the dimensionality of the future, but its predictability.

In a 2016 editorial for the New York Times, computational meteorologist and past president of the American Meteorological Society William B. Gail cited a number of patterns that humanity has studied for centuries, but that are disrupted by climate change: long-term weather trends, ish spawning and migration, plant pollination, monsoon and tide cycles, the occurrence of ‘extreme’ weather events. For most of recorded history, these cycles have been broadly predictable, and we have built up vast reserves of knowledge that we can tap into in order to better sustain our ever more entangled civilisation. Based on these studies, we have gradually extended our forecasting abilities, from knowing which crops to plant at which time of year, to predicting droughts and forest ires, predator/prey dynamics, and expected agricultural and isheries outputs.

Civilisation itself depends on such accurate forecasting, and yet our ability to maintain it is falling away as ecosystems begin to break down and hundred-year storms batter us repeatedly. Without accurate long-term forecasts, farmers cannot plant the right crops; ishermen cannot ind a catch; lood and ire defences cannot be planned; energy and food resources cannot be determined, nor demand met. Gail foresees a time in which our grandchildren might conceivably know less about the world in which they live than we do today, with correspondingly catastrophic events for complex societies.32 Perhaps, he wonders, we have already passed through ‘peak knowledge’, just as we may have already passed peak oil. A new dark age looms.

The philosopher Timothy Morton calls global warming a ‘hyperobject’: a thing that surrounds us, envelops and entangles us, but that is literally too big to see in its entirety. Mostly, we perceive hyperobjects through their inluence on other things – a melting ice sheet, a dying sea, the buffeting of a transatlantic light. Hyperobjects happen everywhere at once, but we can only experience them in the local environment. We may perceive hyperobjects as personal because they affect us directly, or imagine them as the products of scientiic theory; in fact, they stand outside both our perception and our measurement. They exist without us. Because they are so close and yet so hard to see, they defy our ability to describe them rationally, and to master or overcome them in any traditional sense. Climate change is a hyperobject, but so is nuclear radiation, evolution, and the internet.

One of the main characteristics of hyperobjects is that we only ever perceive their imprints on other things, and thus to model the hyperobject requires vast amounts of computation. It can only be appreciated at the network level, made sensible through vast distributed systems of sensors, exabytes of data and computation, performed in time as well as space. Scientiic record keeping thus becomes a form of extrasensory perception: a networked, communal, time-travelling knowledge making. This characteristic is precisely what makes it anathema to a certain kind of thinking – one that insists on being able to touch and feel things that are intangible and unsensible, and subsequently dismisses the things it cannot think. Arguments about the existence of climate change are really arguments about what we can think.

And we are not going to be able to think much longer. In preindustrial times, from 1000–1750 CE, atmospheric carbon dioxide varied between 275 and 285 parts per million – levels we know from studying ice cores, the same batteries of

knowledge that are melting away in the Arctic today. From the dawn of the industrial age they begin to rise, reaching 295 ppm at the start of the twentieth century, and 310 ppm by 1950. The trend – named the Keeling Curve, after the scientist who started modern measurements at the Mauna Loa observatory in Hawaii in 1958 – is ever upward, and accelerating. 325 ppm in 1970, 350 in 1988, 375 in 2004.

In 2015, and for the irst time in at least 800,000 years, atmospheric carbon dioxide passed 400 ppm. At its current rate, which shows no sign of abating, and we show no sign of stopping, atmospheric CO2 will pass 1,000 ppm by the end of the century.

At 1,000 ppm, human cognitive ability drops by 21 per cent.33 At higher atmospheric concentrations, CO2 stops us from thinking clearly. Outdoor CO2 already reaches 500 ppm regularly in industrial cities: indoors, in poorly ventilated homes, schools, and workplaces, it can regularly exceed 1,000 ppm – substantial numbers of schools in California and Texas measured in 2012 breached 2,000 ppm.34

Carbon dioxide clouds the mind: it directly degrades our ability to think clearly, and we are walling it into our places of education and pumping it into the atmosphere. The crisis of global warming is a crisis of the mind, a crisis of thought, a crisis in our ability to think another way to be. Soon, we shall not be able to think at all.

The degradation of our cognitive abilities is mirrored at scale in the collapse of the transatlantic jet routes, the undermining of communication networks, the erasure of diversity, the melting away of historical knowledge reserves: these are signs and portents of a wider inability to think at the network level, to sustain civilisation-scale thought and action. The structures we have built to extend our own life systems, our cognitive and haptic interfaces with the world, are the only tools we have for sensing a world dominated by the emergence of hyperobjects. Just as we are beginning to perceive them, our ability to do so is slipping away.

Thinking about climate change is degraded by climate change itself, just as communications networks are undermined by the softening ground, just as our ability to debate and act on entangled environmental and technological change is diminished by our inability to conceptualise complex systems. And yet at the heart of our current crisis is the hyperobject of the network: the internet and the modes of life and ways of thinking it weaves together. Perhaps unique among hyperobjects, the network is an emergent cultural form, generated from our conscious and unconscious desires in dialogue with mathematics and electrons and silicon and glass ibre. That this network is currently being (mis)used to accelerate the crisis, as we will see in subsequent chapters, does not mean it does not retain the potential to illuminate.

The network is the best representation of reality we have built, precisely because it too is so dificult to think. We carry it around in our pockets and build pylons to transport it and palaces of data to process it, but it is not reducible to discrete units; it is nonlocal, and it is inherently contradictory – and this is the condition of the world itself. The network is continuously, deliberately and unknowingly created. Living in a new dark age requires acknowledging such contradictions and uncertainties, such states of practical unknowing. Thus the network, properly understood, can be a guide to thinking other uncertainties; making such uncertainties visible must be done precisely so that they can be thought. Dealing with hyperobjects requires a faith in the network, as mode of seeing, thinking, and acting. It denies the bonds of time, place, and individual experience that characterise our inability to think the challenges of a new dark age. It insists on an afinity with the noumenal and the uncertain. In the face of atomisation and alienation, the network continually asserts the impossibility of separation.


1.       ‘Trembling tundra – the latest weird phenomenon in Siberia’s land of craters’, Siberian Times, July 20, 2016,

2.       US Geological Survey, ‘Assessment of Undiscovered Oil and Gas in the Arctic’, USGS, 2009,

3.       ‘40 now hospitalised after anthrax outbreak in Yamal, more than half are children’, Siberian Times, July 30, 2016,

4.       Roni Horn, ‘Weather Reports You’, Artangel oficial website, February 15, 2017,

5.       ‘Immigrants Warmly Welcomed’, Al Jazeera, July 4, 2006, a

6.       Food and Agriculture Organization of the United Nations, ‘Crop biodiversity: use it or lose it’, FAO, 2010,

7.       ‘Banking against Doomsday’, Economist, March 10, 2012, e

8.       Somini Sengupta, ‘How a Seed Bank, Almost Lost in Syria’s War, Could Help Feed a Warming Planet’, New York Times, October 12, 2017,

9.       Damian Carrington, ‘Arctic stronghold of world’s seeds looded after permafrost melts’, Guardian, May 19, 2017,

10.    Alex Randall, ‘Syria and climate change: did the media get it right?’, Climate and Migration Coalition,

11.    Jonas Salomonsen, ‘Climate change is destroying Greenland’s earliest history’, ScienceNordic, April 10, 2015,

12.    J. Hollesen, H. Matthiesen, A. B. Møller, and B. Elberling, ‘Permafrost thawing in organic Arctic soils accelerated by ground heat production’, Nature Climate Change 5:6 (2015), 574–8.

13.    Elizabeth Kolbert, ‘A Song of Ice’, New Yorker, October 24, 2016,

14.    Council for Science and Technology,A National Infrastructure for the 21st century’, 2009,

15.    AEA, ‘Adapting the ICT Sector to the Impacts of Climate Change’, 2010,

16.    Council for Science and Technology,A National Infrastructure for the 21st century’.

17.    AEA, ‘Adapting the ICT Sector to the Impacts of Climate Change’.

18.    Tom Bawden, ‘Global warming: Data centres to consume three times as much energy in next decade, experts warn’, Independent, January 23, 2016,

19.    Institute of Energy Economics, ‘Japan Long- T erm Energy Outlook – A Projection up to 2030 under Environmental Constraints and Changing Energy Markets’, Japan, 2006,

20.    Eric Holthaus, ‘Bitcoin could cost us our clean – energy future’, Grist, December 5, 2017,

21.    Digital Power Group, ‘The Cloud Begins With Coal – Big Data, Big Networks, Big Infrastructure, and Big Power’, 2013, tech- p

22.    Bawden, ‘Global warming’.

23.    Alice Ross, ‘Severe turbulence on Aerolot light to Bangkok leaves 27 people injured’, Guardian, May 1, 2017,

24.    Anna Ledovskikh, ‘Accident on board of plane Moscow to Bangkok’, YouTube video, May 1, 2017.

25.    Aerolot, ‘Doctors Conirm No Passengers Are In Serious Condition After Flight Hits Unexpected Turbulence’, May 1, 2017,

26.    M. Kumar, ‘Passengers, crew injured due to turbulence on MAS light’, Star of Malaysia, June 5, 2016,

27.    Henry McDonald, ‘Passenger jet makes emergency landing in Ireland with 16 injured’, Guardian, August 31, 2016,

28.    National Transportation Safety Board, ‘NTSB Identiication:


29.    Federal Aviation Administration, FAA Advisory Circular 120 -8 8A, 2006.

30.    Paul D. Williams & Manoj M. Joshi, ‘Intensiication of winter transatlantic aviation turbulence in response to climate change’, Nature Climate Change 3 (2013), 644–8.

31.    Wolfgang Tillmans, Concorde, Cologne: Walther Konig Books, 1997.

32.    William B. Gail, ‘A New Dark Age Looms’, New York Times, April 19, 2016,

33.    Joseph G. Allen, et al., ‘Associations of Cognitive Function Scores with Carbon Dioxide, Ventilation, and Volatile Organic Compound Exposures in Ofice Workers: A Controlled Exposure Study of Green and Conventional Ofice Environments’, Environmental Health Perspectives 124 (June 2016), 805–12.

34.    Usha Satish, et al., ‘Is CO2 an Indoor Pollutant? Direct Effects of  Low – to – Moderate CO2 Concentrations on Human Decision – Making Performance’, Environmental Health Perspectives 120:12 (December 2012), 1671–7..

James Bridle 
James Bridle (born 1980) is an artist, writer and publisher based in London. Bridle coined the New Aesthetic; their work “deals with the ways in which the digital, networked world reaches into the physical, offline one.” Their work has explored aspects of the western security apparatus including drones and asylum seeker deportation. Bridle has written for WIRED, Icon, Domus, Cabinet Magazine, The Atlantic and many other publications, and writes a regular column for The Guardian on publishing and technology.