13 June 2019

Five ‘Must Read’ Texts to Understand Automation and the Future of Work

by Aaron Bastani

One of the questions I’m most frequently asked is what I’m reading. One of the advantages of writing a book, particularly one that took far longer than originally anticipated, is that I was compelled to reduce my use of Twitter – something I am generally incapable of doing – and engage in several extensive literatures. In the case of Fully Automated Luxury Communism this was relatively wide-ranging with my year-long reading lists including renewable energy, synthetic biology, the digital revolution and more besides.

Scoping out from daily news, increasingly metabolised through social media, and freed from the frenetic mindset that induces – proved incredibly useful. It allowed me to refine long-held ideas into more specific hypotheses about what technological change means in a world of climate systems breakdown, rising inequality and demographic ageing. More importantly, I grappled with what the left should try and do about it.

One of the more rewarding areas for this was in automation. That was primarily the result of a number of outstanding books published in the years leading up to my research. Indeed the field was so advanced that by 2016, when I began to gather as much information as I could, several generalist books on the topic were being widely lauded by the media – with two of the five texts I suggest making it to the New York Times bestseller list.

While my five recommendations – three books and two excerpts – all offer slightly different conclusions, they are all invaluable. If you spend the time reading them you will, in my opinion, have a better grasp on the automation debate than many policy-makers and much of the media. That each text offers a different prognosis is useful. Our interest here isn’t to arrive at a sacred interpretation of what will inevitably happen, but rather a contested overview that can, at a minimum, provide a broader set of parameters about changes ahead and how people people can craft them in a socially just direction. Fully Automated Luxury Communism holds its own diagnosis and prescription, and does so drawing upon all five, but this reading list certainly doesn’t mean you’ll end up agreeing with me. At the very least, however, I hope you’ll understand my workings.

What I know for certain is that topics such as automation, climate change and ageing are simply too big for the staccato and often disjointed analysis of the typical information diet: blogs, twitter threads and comment pieces. David Wallace-Wells’ utterly brilliant The Uninhabitable Earth is one example of this, the thread of its argument more useful than a thousand viral videos in conveying the scale of climate systems breakdown – the complexity of the issue necessitating an appropriate genre to convey its holistic nature. 

But while the book-form remains much needed, life is too short to read stuff which doesn’t teach you something or move you. So here are my top five reads on automation, what it means for workers in the modern economy and what comes next – if we want it.

While I find its policy recommendations weak, the first book is an immense feat of scholarship. The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies, by Andrew McAfee and Erik Brynjolfsson, is a superb overview that responds to the challenges presented by digitisation, automation and artificial intelligence. The idea of a ‘Second Machine Age’ coined by the authors, the first one being the Industrial Revolution, inspired what I term the ‘Third Disruption’. Just as the early 19th century witnessed the emergence of the Industrial Revolution, built as it was around the steam engine, the energy paradigm of fossil fuels and oriented around the social relations of market capitalism and the factory-system, the 21st century is giving rise to something else. For McAfee and Brynjolfsson the present moment is one where changes in cognitive labour, principally as a result of machine learning and big data, resemble the transformations to physical labour after the First Machine Age. It is a neat categorisation that I find analytically strong and fits with changes we can observe, particularly what technological change means for white collar professions such as law, accounting and even medicine. Their analysis of exponential change is also excellent, providing important insights about what Moore’s Law, even if it continues to decelerate, might mean a half century from now.

Perhaps the key finding of the book, and of relevance to those otherwise unpersuaded that technological change spells mass unemployment, is that automation and AI will inevitably serve to further exacerbate inequality. That is the consequence of the market for ‘uniquely human skills’ getting ever smaller, meaning most of us experience a downward pressure on wages while a tiny elite earn massive sums. This small group, who would be wealthy as a result of the high price they command for their labour, would sit alongside a larger social substrate whose affluence results from asset ownership and rents. McAfee and Brynjolfsson refer to this as ‘The Spread’. You don’t have to think that capitalism has a successor system – and I do – to take such arguments seriously. Such a forecast spells major problems for political and social stability under market capitalism, but also increasingly common crises of under-consumption.

The second recommendation is Rise of the Robots by Martin Ford. The Financial Times’ Business Book of the Year for 2015, it is in a league with the Second Machine Age as a work of scholarship – although it more urgently engages with technological change not only meaning inequality but also mass unemployment. So while Ford’s prognosis is somewhat similar to McAfee and Brynjolfsson’s the diagnosis is more radical – and in its own way closer to that of my own. Ford concludes, “The greatest risk is that we could face a “perfect storm” – a situation where technological unemployment and environmental impact unfold roughly in parallel, reinforcing and perhaps even amplifying each other. If, however, we can fully leverage advancing technology as a solution – while recognising and adapting to its implications for employment and distribution of income  – then the outcome is likely to be far more optimistic. Negotiating a path through these entangled forces…may prove to be the greatest challenge for our time”. 

Surprisingly it is Ford’s book, with a keener eye to a potentially-crisis ridden future, that became Forbes top business book of 2015. McAfee and Brynjolfsson, by contrast, offer suggestions for individuals to ‘get ahead’ (no bad thing in itself but a tacit admission of the impossibility of structural transformation), and conclusions that a world of 10 billion – wracked by a panoply of ecological and social crises – will mean more of us become landscape gardeners and interior decorators. Somehow I don’t think so.

Next is The Economic Singularity by Callum Chase. Offering an even pithier style than the first two, Chace’s book has a greater sense of certainty about what’s next. If Ford goes further than McAfee and Brynjolfsson then Chace goes beyond Ford, with only around half the book given to observing technological change as it unfolds in the present. Indeed much of the book is intentionally oriented to the future, with Chace offering digestible forecasts for 2021, 2031 and 2041 (a task which he admits can easily come undone). While policy-makers may have heard of the first two books The Economic Singularity is of arguably greater use, offering the bullet-point analysis and suggestions one normally associates with a think-tank report. Fundamentally Chace believes humanity will have to adapt to a radically different world over the next century – regardless of what happens with climate change or ageing. Indeed it will likely be so different that it is difficult to speculate about the kind of technologies, social relations and forms of production that the continuation of Moore’s Law, albeit proceeding at a diminished rate, might mean a half century from now.

4) The fourth book is the Grundrisse by Karl Marx, specifically the Fragment on Machines. A small sub-section of the book, no more than twenty-two pages in length, it is here where Marx not only isolates what technological change under capitalism means for working people, but how such transformation carries an immanent alternative for the ‘emancipation’ of labour. You can read the Fragment here. While the older Marx would go on to repudiate his early work, including the Grundrisse (German for outline or sketch), the last century has proven it a prescient account on the possibilities of technology. 

Perhaps most interesting about the Grundrisse, and of considerable consequence for Marxist politics over the course of the 20th century, is how it remained unpublished in German until 1939. Worse still for the anglophone world, it was unavailable in English until as recently as 1971. Antonio Negri notes how it was the Grundrisse, not the Communist Manifesto or Capital, which was the ‘livre de chevet’ (bedside reading) of every Italian radical towards the end of the 1960s (it first appeared in Italian after 1968). Whenever someone says ‘these ideas have been tried before’ they are not only wrong about the political applications of Marx generally, but also ignore that it wasn’t until the end of the Cold War that a rich vein of his thinking, particularly the 1844 Manuscripts and Grundrisse, received any sustained attention. That is not to create a tension between the early and late Marx, a periodisation offered by Louis Althusser, but simply to recognise that Marxist thinking – for nearly the entire 20th century – failed to integrate some of his most powerful and original insights. One line in the Grundrisse stands out about all others, “The most developed machinery thus forces the worker to work longer than the savage does, or than he himself did with the simplest, crudest tools”. That, in brief, tells the story of how technological change has failed to save labour time or liberate humans for the best part of fifty years. For Marx the explanation as to why is capitalism.

The final recommendation, in a different vein to Marx, is a short essay by John Maynard Keynes titled Letter on the Economic Possibilities for our Grandchildren. Written just a year after the Wall Street Crash in 1929, its optimism couldn’t have been more at odds with the zeitgeist of the decade that followed. Three years before Hitler’s ascent to Fuhrer, and as the economies of Europe and North America entered a historic slump, Keynes predicted that within just a hundred years material scarcity, or as he called it ‘the economic problem’, would be, “within sight of solution”. As a result Keynes predicted that the, “economic problem is not-if we look into the future-the permanent problem of the human race” but rather how to “live wisely and agreeably and well”.

We now know that Keynes was wrong and, at least when it comes to technology redounding to the benefit of greater leisure over time, Marx was right. Indeed these two short essays, Marx’s ‘fragment’ and Keynes’ ‘letter’, are best read side-by-side. Such a comparative reading will leave you better placed to engage with the first three books. If you want to know why liberals generally don’t grasp the politics of technology, then Keynes is the best place to start. He was the protean techno-optimist and was proven completely wrong. And yet his central claim was that the development of capitalism could mean a social system at odds with it, one which he suspected would make the human self accustomed to millennia of toil short-circuit, “…yet there is no country and no people, I think, who can look forward to the age of leisure and of abundance without a dread. For we have been trained too long to strive and not to enjoy.” Marx and Keynes agreed about the ends, the former called it the cultivation of ‘species-being’, while Keynes waxed lyrical about how, “we shall once more value ends above means and prefer the good to the useful” under conditions of near spontaneous abundance.

Once you’ve read the two greats side by side it makes sense to proceed through the more recent books I recommend. At the very least it’ll help you understand why I arrived at the conclusions I have. Perhaps, more than that, you might even be inclined to agree. 

This site uses functional cookies and external scripts to improve your experience.