Delivery Lines: Amazon’s Supply Chain
In 2019 Amazon CEO Jeff Bezos made a critical articulation about contemporary capitalism, akin to Vanderbilt’s “The public be damned” or Thatcher’s “There no such thing as society”—quotations of the elite that serve to define their epoch. The quote in question didn’t get quite as much media attention at the time as Bezos’s post-space-travel reminder to Amazon employees and customers that “you paid for this,” though that statement was similarly telling and deservedly ridiculed. Rather, it went relatively overlooked within a speech Bezos gave at a special press conference held for the announcement of Amazon’s Climate Pledge, a set of energy and purchasing goals ostensibly intended to address climate change. In attempting to explain the connectedness of capitalist firms and recruit more signatories to the pledge, Bezos also characterized the neoliberal, data-driven, always-on gig economy that has become relevant to enough of life to define our present era. “We are all a part of each other’s supply chains.”
Production Lines: Labor Surveillance
Amazon has become known for subjecting the people that make up its own vast supply chain—employees as well as “independent” contractors—to demanding performance standards. And judging performance against stated goals requires self-monitoring. Amazon worker performance is tracked via infrastructure that produces data, which then contributes to the dataset that is used to make increasingly robust performance benchmarks. This self-perpetuating cycle of data-driven surveillance has profound implications for safety and general working conditions, both inside and outside of fulfillment centers. This follows the larger “Uberization” of the economy, where digital platforms serve as intermediaries between customers and networked service providers. Workers do not own the surveillance-cum-employment platforms or the data they collect, nor do they have a say in what data is collected and how. With more work happening offsite, remote, or by contractors, it becomes increasingly difficult to distinguish between inside and outside the workplace, on the clock and off, or to tell where the production line is—and how it might be stopped. And as more data on location, health, and consumption is collected and connected to work, performance metrics extend outside of the workplace to the extent that what is being produced is data.
Ledger Lines: Accounting for Scope
Worker surveillance has its analog in a company’s internal processes of accounting for environmental impacts. Measuring a different kind of efficiency, firms carefully track their own operational processes according to categories like energy use, material throughput, and waste. Land-intensive programs to mitigate the environmental impacts of these operations similarly depend on measurements like the carbon sequestration abilities of various non-human organisms or entire ecosystems. Land itself is made to perform labor, and its performance is tracked, measured, and evaluated, quantified, and financialized in the form of carbon offsets, ecosystem services, biodiversity, and other climate mitigation policies. The ability of both forms of labor to become all-encompassing—i.e. you’re always at work, and everything has an environmental impact—leaves much room for the setting of arbitrary boundaries that suit specific purposes.
Like the lines drawn between employee and subcontractor, which allow the company to avoid responsibility for safety, health, and other issues of worker wellbeing, sustainability reporting uses the notion of “scope” to delineate what the company is and isn’t accountable for, or their “lines of responsibility.” In carbon accounting, a widely used three-tier system purports to designate what is and isn’t under a company’s control. But the math gets fuzzy, and the idea of “control” becomes a free pass to absolve responsibility by only claiming territory when it’s convenient or easy. A producer might take responsibility for their own operations: direct emissions, such as the emissions of their own buildings (Scope 1), as well as energy they purchase and consume, such as electricity that powers their factories (Scope 2). But oftentimes indirect emissions, like the emissions of their contractors and subcontractors, or other impacts along the value chain from suppliers or producers, are written out because they are “outside of their control.” Just as no one is checking back on the mechanics of meeting prior, outdated carbon neutrality pledges, no one is really policing the boundaries that determine scope, and where responsibility lies. A company may not directly “control” the work of its subcontractors, but it still exerts influence over them through performance metrics. The scale of Amazon’s aggregate demand influences those in its supply chain who may be technically “out of scope” for the self-monitoring of its climate impacts. In this way, as in much shady accounting, firms can have it both ways. Its measurements of worker safety or climate impact can be both on the books and off the books in the larger system of social, financial, and reputational accounting.
Jeff Bezos’ signature from Amazon’s Annual Letter to the Shareholders, 1997. Source: Wikimedia Commons.
Signature Lines: Promissory Politics
Amazon’s Climate Pledge is just one of many public announcements of environmentally-motivated goals that came to dominate political discussion from the late twentieth century onwards. As innovations in data science made long term climate modeling increasingly robust, and everyday experience made this science increasingly undeniable, it became common practice for businesses, governments, and consortiums of both of these to make public promises, pledges, and agreements that set future target dates for action or major milestones. Anticipated by the military and corporate tradition of scenario building and forecasting, and further enabled by the increasing power and ubiquity of networked computers and accompanying datafication, this promissory politics is encouraged and made sensible by cultures of tracking and measurement in business management and personal self-improvement. Promise-making has become a regular part of the brand management of corporate firms and governmental entities sensitive to investors, employees, customers, and other stakeholders’ concerns about so-called “non-financial” factors, known in industry lingo as “ESG” (environment, social, and governance). But the practice of promising cannot be dismissed as mere greenwashing, as it is so earnestly blind to its own contradictions to appear pathological, and so widespread as to become nearly obligatory.
Corporate promises are voluntary, piecemeal. They may be touted as backed by science, but they are not backed by any kind of governance. They are, instead, presented as better alternatives to government regulation. Amazon’s Climate Pledge itself, and the PR messaging around it, was crafted to achieve a one-upmanship of not only the US government, but of the coordinated efforts of the nations of the world, its target date prominently advertised as “ten years sooner” than the dates set in the Kyoto Protocol. Efforts to gain signatories to the pledge thus represent a different form of coordination: an organizing effort on the part of the capitalist class to stave off regulation and convince the public that an unfettered market can be trusted to solve climate change. Bezos’s purchasing of naming rights to Seattle’s Key Arena, formerly owned by Key Bank’s parent company KeyCorp, and choosing to name it not Amazon Arena but “Climate Pledge Arena,” underscores just how much of a stunt this is; a staged competition between the false binary of market and government in the arena of public policy.
Corporate environmental commitments usually take the form of some future date at which the company expects to reach net-zero carbon emissions for some percentage, or some part, of its operations. Since the practice is still relatively new in the business world, and old press releases have a way of disappearing from public view, progress toward past goals can be difficult to measure. And measurements of current promises are subject to the same creative accounting involved in their drafting, with so much room for interpretation in variables like the cost of carbon, the type of offset and how it is measured, how the scope of “operations” is defined, which emissions count, how “renewable” is defined, what baseline the percentage is judged against, and so on.
Governments who have been in the promise-making business longer tend to leave behind paper trails of announcements, agreements, and agendas that reveal promise-making as an ongoing game of missing targets and setting new ones, a game whose rules require constant moving of goalposts. US President Jimmy Carter’s 1979 target for 20% of the nation’s energy to come from renewable sources by the year 2000 was moved forward by President Barack Obama in 2013, thirteen years after the initial deadline had passed, to a new goal of 2020. The nation’s promises, at least, are renewable. Perhaps in recognition of this embarrassing precedent, President Joe Biden offered a target percentage range of 50–52% for reduction in US greenhouse gas pollution. The precision of this range, appearing to be backed by some kind of math, is belied by the fact that it is measured against pollution levels from the seemingly arbitrary date of 2005.
Image posted on the Instagram page of Jeff Bezos on June 25, 2020, to announce that Amazon has bought the naming rights to Seattle’s arena. Source: Instagram.
The archive of United Nations policy documents reads like an obituary of past commitments: the unmet goals of Agenda 21 (1992–2000) were replaced by the Millennium Development Goals (2000–2015), which were replaced by the current Sustainable Development Goals (2015–2030), praised by UN Secretary-General Ban Ki-moon as a “universal, transformative, and integrated agenda” and a “historic turning point for our world.” In September 2020, the news passed with little fanfare that of the twenty targets negotiated by the world’s environment ministers as part of the 2010 Aichi Biodiversity goals, not a single one had been met. Even the much-lauded Kyoto Protocol and Paris Agreement, the round of promises that still informs most status-quo policy, were at the time of their negotiation already recognized by some involved as insufficient: dependent on increased ambition of each nation’s voluntary commitments, undergirded by blind faith in technological innovation, and unfundable through existing government and philanthropic means. One gets the feeling that each agreement is the kind a person makes simply to end the discussion, and get back to whatever it was they were doing until the deadline passes and we all agree to sit down together again, pretending to make progress.
Signature lines—which somehow make these promises feel official—have conveniently replaced another line of struggle: the picket line. Resulting from existing public and private pressure, like employees and lobbyists working through board consortia, the main function of these signature lines is to allow business to continue as usual. They make verbal commitments that help employees feel better about their jobs and the companies they work for. But these promises usually just involve more feel-good strategies like offsets that don’t address problems with underlying business models, let alone whole economies. They function as a form of creative accounting instead of creating accountability, and allow for deferral of the kind of “rapid, far-reaching, and unprecedented” action that the IPCC now says is needed to avoid a two-degree temperature rise.
The fundamental logic of these promises is familiar to anyone who has ever been in debt and set up a payment plan; a promise to pay in smaller increments towards a future date of completion. Indeed, climate change may be framed as a situation of intergenerational debt, but one with a kind of compound interest in the warming that has already or is or due to occur, such that the overall debt only grows despite any incremental payoffs. All these pledges and targets seek to buy time but nobody gets paid, and the feeling of imminent existential foreclosure remains.
That these kinds of targets always fall conveniently on years that mark decades or half-centuries like 2030 and 2050 reveals their basis more in public relations than a science whose predictive modeling might suggest more specific dates. Projects like Biden’s “30 by 30,” which has the goal of conserving 30% of land for biodiversity by the year 2030, sound more like a business magazine’s annual roundup of young entrepreneurs than a concrete, science-informed plan. To be fair, the farther off a target, the more justifiable its imprecise positioning. And round numbers certainly are more memorable. But given the margins between 1.5 and 4 degrees of warming, such arbitrary selection or nearest-decade rounding could mean the difference between life or death, survival or extinction.
Following Carl von Clausewitz’s dictum that war is politics by other means, such public promises, goals, statements, resolutions, protocols, and agreements can be understood as a continuation or transubstantiation not of politics but the absence of politics. Or the avoidance of politics. Promises are not merely policies that can’t be policed, they are fundamentally non-policies to begin with. Their main effect, if not their actual intent, is neither to initiate, nor even to guide action, but instead to forestall it.
In some sense, politics has always been about making promises. But what we see today is a phenomenon wherein promises replace politics. The willingness to make promises set in the future is directly proportional to the unwillingness to make the political changes needed in the present. Hence the increasing proliferation, and desperation, of promises.
Border Lines: Conservation as Enclosure
In order for all these promises to be kept and conservation goals to be met, huge swathes of land would need to be used, affecting its existing designation, zoning, legal rights, and claims. Much of the world—up to fifty percent of it, according to E.O. Wilson’s influential “Half Earth” proposal—would essentially become a kind of “promise land.” This comes after the wave of land acquisitions in developing countries by foreign wealth funds, private equity funds, agricultural producers, and other major firms in the food and agribusiness industry following the sharp increase in international food prices during 2007–2008. These land grabs were also about water (since land rights are often rights to the water below) and energy, since biofuels are a way for rich countries to meet their renewable energy targets. The widespread practice of trading REDD credits (“Reducing Emissions from Deforestation and forest Degradation”) for carbon offset programs has further fueled this global land rush on a scale some compare to that of the colonial era, giving the name “green grabbing” to this massive appropriation of land and resources for environmental ends.
The world’s proposed climate plans—which would require both massive conservation offsetting to meet net-zero demands and rare earth metal mining to increase battery and panel manufacturing—will rapidly expand this global land grab. The entire enterprise amounts to a new enclosure movement, akin to what historians describe taking place in the transition from feudalism to capitalism in late medieval England, except now, as some critics of platform capitalism have already argued, may be from capitalism to some kind of neo-feudalism. It’s as if the inequitable ownership structure of the digital platform economy is being inscribed upon the physical landscape; the boundary lines of the terrestrial platform called Earth being redrawn in its image.
Excursus: Draft Land Acknowledgement for Online Meetings
The powerful tools that allow us to meet together today, across vast distances of space, depend on equally powerful tools of exploitation that affect resources and people in more remote locations. To extract the resources that go into cables, towers, satellites, and devices; to manufacture and distribute products; to host the platforms and run the servers; to power and to cool the equipment; and to store the resulting waste. All of that happens on land. Sometimes underwater, or land that is submerged. But always in place. And all of these places once belonged, or still belong, to many different peoples whose names are too numerous to list here. The ease with which we now click a few buttons and experience each other’s presence obscures the violence required to create and maintain the infrastructure that makes it possible.
While we gather today from different places with their own histories, our gathering depends on even more places with which we all have a direct material connection. We acknowledge those with deeper connections to the lands that are affected by the infrastructure used to gather today, and the ways in which such lands have been and continue to be appropriated via coercive means, ranging from outright theft to treaties or contracts signed under duress within legacies of unjust relations. Our acknowledgement of these relations is a first step toward repairing them. This acknowledgement is in no way intended to replace or diminish the practice of traditional land acknowledgements or the specific acknowledgements of host institutions and participants, but rather it is an attempt to seriously engage them and extend them into an online networked environment where their relevance may not be so obvious.
Battle Lines
If part of the means of production is the land upon which it takes place—the grounds of production—then its ownership matters. If production will increasingly mean not production of commodities in the traditional sense, but production of (commoditized) climate mitigation, then workers must own this means of mitigation. If production increasingly takes place within platforms of data collection and management, then workers must own the means of knowledge production. Furthermore, if workers can also “own,” in the sense of taking responsibility for something, or claiming it as their work, and climate change produces new means of performing, monitoring, and defining labor, then Marx himself—a habitual practitioner of the classical chiasmus and other literary inversions—might have endorsed a contemporary call for ownership of the production of means.
Bezos’s motto—“We are all a part of each other’s supply chains”—is more than just a technical description of the networked organization of firms like Amazon. It neatly encapsulates the dis-organized state of labor among its own suppliers and the broader global precariat, as well as the culture of personal branding and internalized, mediatized person-as-corporation, the ideology of individual consumer responsibility for climate change, and a sense of obligatory participation in the given economic system in order to merely survive.
This global networked “supply chain” may have largely replaced the traditional single-plant assembly or picket line as a locus of labor struggle, but it too contains choke points and places along the chain that can be blocked. A chain, after all, is still a kind of line; one that is only as strong as its weakest link. Only its networked aspect makes it more resilient, as exemplified in Amazon’s own third-party platform model. The supply chain metaphor contains within it the seeds of its own undoing, if only we recognize it not as a statement of corporate social responsibility, but of worker solidarity. The workers of this world have nothing to lose but their supply chains.
Seeing a network means connecting dots. And what we do when we connect dots is draw lines. Capitalism is global, and so are its effects. This much can be plainly seen. These dots have been connected, these lines have been drawn. But the lines themselves are not yet fully seen for what they are: battle lines.
Workplace is a collaboration between e-flux Architecture and the Canadian Centre for Architecture within the context of its year-long research project Catching Up With Life.
