“Amazon – it’s the most obvious name ever!” exclaims Danielle Dean, when I ask her if it’s a play on meanings.
The first thing that springs to mind is naturally Amazon Prime whose arrow logo flits past us everywhere.
Danielle takes me by the hand into a sci-fi world that is here already, dipping me into weird neologisms, such as Requesters, Turks, and HITs.
The name ‘Amazon’ has a wealth of history, which her extraordinary multi-channel video installation at Tate Britain exploits (in a good sense!) to the full. An anamorphic projection provides images of a fake version of a rainforest. Her fact/fiction paintings suggest, she says, “how we relate to nature and how that in turn may be influenced by the American dream”.
Over the last two years, Danielle has collaborated with Amazon Mechanical Turk (AMT) workers around the world. AMT is one of many online job platforms which crowdsource labour to fulfil digital tasks, Human Intelligence Tasks (HITs), for Requesters – companies who need data from thousands of individuals.
During the Covid pandemic she sent video camera kits to AMT workers and distance-trained them to create good footage, filming themselves in their homes where they carry out their jobs in the gig economy. “These workers are my main characters,” she says.
AMT is a form of digital Taylorism driven by global capitalism’s need to extract massive quantities of data. Despite perceptions to the contrary, at the very heart of this process, which reaps tech giants such vast profits, are individual human beings (living labour to use Marx’s term). And it involves super exploitation whereby workers may end up earning even less than $1 per hour, depending on how long it takes to carry out a task.
(Amazon’s web-based Mechanical Turk app was launched in 2005 so that humans could carry out tasks that computers had difficulty with. The original Mechanical Turk was a fake chess-playing robot.)
Placing AMT workers, also known as ‘Turks’, in a fictional ‘Fordlandian’ landscape, Danielle draws a comparison with Henry Ford’s failed Taylorist 1930s experiment in Brazil’s Amazon region.
She projects today’s process of “extraction” as a data-driven version of colonial capitalism in upper Amazonia in the early years of the last century, seeing it as a contemporary form of “primitive accumulation”.
“Now,” she says, “it’s the extraction of human data, pure information, which is what AMT does. Each HIT collects data from between 1,000 and 10,000 people. It includes information about human thoughts, desires and emotional states such as depression.
Is this the ultimate penetration of capital into our inner selves, the ultimate form of alienation? Isn’t Big Tech exploiting technologies developed by the digital commons purely to extract vast profit from labour?
“AI always had human labour behind it,” Danielle responds. “While AMT doesn’t have anyone overseeing it, there is no socialist avenue to it in any way. Requesters put their work on the platform which is accessed by individuals who want work and carry out the HIT, the Human Intelligence Task. There is no one to complain to. The Requester has much more power than the worker. There is no union for sure. If you lose a day’s work there is not even a person on a phone.
“Amazon doesn’t care for the wellbeing of the work force, who don’t have access to the work they are doing in the first place. They want people to be like robots. It’s the division of labour that Marx was critical of, even though people don’t work on an assembly line.
“It’s super important to listen to the people who are doing the work. They are spread across the whole world, but they communicate amongst themselves.” There is even Turkopticon, an activist hack that empowers workers, theorised by Danielle’s colleague, Lilly Irani.
“It’s not about the work or technology as such. Some workers, like the amazing Amy, who features in one of the videos, have no criticism of AMT.
“Working at home really suits her. But it’s the way it is used to make huge profits for a few people and the way that workers are treated. AMT is a Ford assembly line on steroids around the world,” Danielle concludes.
“We need to break the logic of alienation by working together and listening to the real people in real situations.”
Art Now: Danielle Dean, curated by Nathan Ladd, is at Tate Britain
5 February – 8 May
Thanks to Eleanor Costello at Tate for her help and enthusiasm
Background facts:
Now, digital Taylorism is represented by wristbands patented by Amazon that are strapped to warehouse workers to track where their hands are at all times and provide ‘haptic feedback’ when they work inefficiently (Novak, 2018).
“This low-paid work arrives via sites such as CrowdFlower, Clickworker, Toluna, and Amazon’s Mechanical Turk, to name a few. Largely unregulated, these sites allow businesses and individuals to post short tasks and pay workers—in cash or, sometimes, gift cards—to complete them. A recent Mechanical Turk listing, for example, offered workers 80 cents to read a restaurant review and then answer a survey about their impressions of it; the time limit was 45 minutes. Another, which asked workers to fill out a 15-minute psychological questionnaire about what motivates people to do certain tasks, offered $1, but allowed that the job could take three hours.
The Atlantic
A 2016 Pew Research Center survey found that 25 percent of workers who earned money from online job platforms like Mechanical Turk, Uber, and TaskRabbit went on these sites because there was no other available work in their area.
(Amazon takes a 20 percent fee on what requesters pay workers, double what it charged in 2015, an increase that some workers on the platform say has caused requesters to offer less money.)
Both Amazon and Tesla doubled their profit during the pandemic. This year, Forbes states that Elon Musk’s net worth is currently $222bn, whereas Amazon executive Jeff Bezos stands at $169.9bn.
Data collection is driven by the perpetual cycle of capital accumulation, which in turn drives capital to construct and rely upon a universe in which everything is made of data. The imperative to capture all data, from all sources, by any means possible influences many key decisions about business models, political governance, and technological development.
Jason Sadowski: Datafication as a political economic regime 2019