In the 2020s we live in an era of automation anxiety and automation fever (Bassett and Roberts, 2020). There are rising concerns around extant computational cultures, near-horizon developments, and longer-term predictions about computational futures. Anti-computing of various kinds is back with a vengeance. A new moment of urgency arises. Once again the contemporary moment is proclaimed as the time of make or break, the time for decisive action. Once again it is argued that technology will in short order crystallize into a good or bad angel, working for good or ill, as Norbert Wiener put it, back in the 1950s – and his work too has undergone a revival. For a flavour of the moment look at what the books say; Democracy Hacked (Moore, 2018), for instance, claims we're in the last chance saloon, ‘if we don't change the system now, we may not get another chance’, whilst People versus Tech (Bartlett, 2018) fears that unless ‘we radically alter our course, democracy will join feudalism, supreme monarchies and communism as just another political experiment that quietly disappeared’. Others’ fears are more existential, reading the contemporary moment as an AI tipping point.

Reading, or, rather, the matter of what is being read, is germane here. This chapter is informed by sixty-four books that are critical, anxious, hostile, concerned, that are actively writing against the computational state we find ourselves in. But this is not a close reading. All the titles were located through Amazon, the majority written within the past five years, but all found in 2020. The contents of some of these inform the arguments here, but the chief interest is in a meta-narrative: how automation anxiety is being framed and sold, how those tropes identified as characteristic of various categories of automation anxiety are expressed, and how they recombine – or divide – in new ways today. Amazon's algorithms thus assisted in a somewhat unofficial form of distant reading (Moretti, 2013). The method – barely one – was simple. I swiped (left and right) across Amazon's quasi-Borgesian ‘shelves’, letting it tell me what other readers read (or bought) when they started where I started, with the books I started with. I also explored Amazon's ‘promoted’ reading suggestions, which often threw up titles further from my concerns than the reader recommendations.

Three or four ‘seed’ books produced the vast majority of the titles. A few emerged on later searches when I used new terms to route around disconnects; cybersurveillance and detox, for instance, didn't quite intersect in expected ways. I make no claims to completeness in undertaking this exercise – nor to having isolated a discrete genre; these are not all anti-computing books, the boundaries of what constitutes a hostile or investigative account are disputable and sometimes blurred, and I certainly have no access to Amazon's algorithms, which occasionally threw up an entirely unexpected suggestion. However, as a probe, this did produce results that were startling, and not least in their quantity. Overall, the search evidenced an explosion of anti-computing taking many forms and evident across a series of genres – I avoided a sharp division made repeatedly, for instance in Paul Mason's Clear Bright Future, between ‘airport’ accounts and ‘serious’ endeavours. This was a cornucopia of interpretation, apostasy, outrage, sorrow, critique, and anger – along with robust demands that readers suffering from computational cultures buy this book, or that one, to help themselves, or perhaps be cured. (Emma Harrison's extant and forthcoming work on connections between mental health and cultures of digital anxiety is germane here.)

The probe thus indicated in outline the rough shape of contemporary anti-computing formations, suggesting key foci, structures of feeling, emotional registers, the orientation and/or bias of currently loudly heard arguments. To consider these formations in relation to the taxonomic categories developed at start of this book is to ask how or if they fit. They may also indicate how a category has morphed, how connection with older or longer-standing formations is made, what of them is really new, and what persists or is revived from earlier moments, or revenant discourses. This is undertaken in this chapter. It leads to a brief reassessment of anti-computing itself, offered by way of a conclusion to the book as a whole. The emphasis is on reconsidering the relationship between computational thinking as a cultural form encapsulating both fatalism about, and a desire for, the further rise of the computational, and the anti-computational impulse, also fatalistic at times, but at others making demands for change. I conclude finally that a certain kind of anti-computational thinking is necessary for truly hopeful computational futures to be envisaged, sought – or fought for.

Distant reading swallows Amazon

I am looking for books attacking Amazon on Amazon. The platform obliges with The Four: The Hidden DNA of Amazon, Apple, Facebook and Google by Scott Galloway (2017). Amazon is of course largely indifferent to the specifics of content. It doesn't care what the books it sells communicate, it's more interested in what I communicate to its algorithms in my fevered search for hostility to, amongst other things, the platform upon which I conduct my searches. The need to capture that data might produce a desire for a better semantic understanding but there is no need for Amazon's bots to appreciate irony, at least if the latter is measured in human terms, only perhaps to understand that it is there.

Amazon of course began with printed books – and perhaps it is surprising that it still sells them. The book refused to die. You knew this, I'm pointing to it because there's a confluence to be observed; here are objects which could be purely informational, that are, by our own desires, still (sometimes) produced on paper, with cardboard covers, using CMYK (cyan, magenta, yellow, black) inks to absorb or reflect colour rather than relying on RGB (red, green, blue) screen display technologies. Reading paper-based books itself begins to represent a slightly tenacious refusal to upgrade, or to be ‘up to date’, and Amazon's book pages, suggesting Kindle but just as willing to sell paper, both escalate the omnivorous logics of the digital and point to a long-term, non-trivial, intrinsic, attachment to other kinds of material; to a form of resistance. They also point to the impurity of ‘the virtual’ as an offering in a material world, since even if books finally become fully digital objects there will still be bodies to read them, and this side of authoring developments signalled by OpenAI's GPT2/3 language models at any rate, (post-)human authors to write them. So, although not surprising, there is something nice about Amazon recommending (or passing on ‘reader’ recommendations, however these are manipulated through encoding) books that are hostile to its materials, to itself, to its logics; books that offer ways to break an addiction to the forms of networked sociality and communication that it itself promotes and lives by. Printed books are far less insistent that readers do not close the covers of a discrete work without thinking about the next one ‘before you go …’. In all, then, what better place to stick around in, if we are looking for where anti-computing has got to by the turning of the first two decades of the 21st century; right now – print publishing turnaround times – and my own tardy writing – permitting.

What is this now? Or, rather, when is it? These are the days of politics dominated by Trump's tweets and Russian interference, Brexit populism, fake news, rising acknowledgement of platform monopoly. We live beyond the knowledge that the Cambridge Analytica scandal gave us, in an era of systematic bucket data capture, concern around screen addiction, trolling, bias, misogyny, addiction, and the rest. Mark Zuckerberg has declared the need for privacy and a desire to step his platform away from the political writ large; a move back to the drawing room, as if that would chase out ‘the political’ (it's always already personal, a chorus of feminists point out). In the UK at least, there is a sense of political crisis, and a moral panic about the internet as partly ‘what did it’ – but also a sense that disengagement is not possible, particularly in COVID/Zoom times).

The above of course is a grossly compacted scenario. I invoke it to stress the force with which the sentiment that we live in unusual times (of political crisis, environmental crisis, moral crisis) is being felt. Which is why it is useful – even salutary – to consider how contemporary computational hostility, arising in response to that formation, part of it as well as mapping onto it, challenges and partially reshapes the more general taxonomy I began with, but also works through its invocation. Today's anti-computing is a recurrent, partly familiar affair, as well as something new. Many of the tropes that inform it are ready to hand, some in more or less complete form, others it is itself reshaping.

Notes from an annotated meta-bibliography

The general taxonomy of anti-computing developed at the start of this book (Figure 3 and Chapter 1) was generated both by paying attention to real-world formations across time and by working with critical theorizations of the media-technological and of medium-technological cultures. A binary division between ontological and political justifications for anti-computational thinking was identified as informing popular thinking, but also as failing to hold, despite its ideological force. It was also recognized as inadequate as a way of grappling with, or categorizing, these formations in order to critically investigate forms of anti-computing. A more elaborated taxonomy of anti-computing was developed which avoided this division. Eight different forms which anti-computing has typically taken were categorized and potentials for cross-cutting them with other classificatory systems (e.g. affective categories, critical theoretical divisions) were explored.

Figure 3 A general taxonomy of anti-computing.

What is offered now is a partly automated reader's report that responds to these categories. As already noted it works with over sixty books which Amazon thinks somebody with an initial interest in something anti-computational should read because other readers did. Following this trail, invoking titles, brief descriptions, and reader comments offered, I explore how these works variously fit into, burst out of, modify, or challenge entirely the original anti-computing categories, which together identified an anti-computational reservoir, a persistent set of tropes, submerged but also ready to re-arise.

(i) Computer technology as control technology

Contemporary examples that fit squarely into this category are Jonathan Taplin's Move Fast and Break Things (2017), and Jamie Bartlett's The People versus Tech (2018). Both make the case for the erosion of democracy by the tech corporations and their products – and both point to the consolidation of power in the new digital economy's heartland. Bartlett's book asks if we have ‘unwittingly handed too much away to shadowy powers behind a wall of code, all manipulated by a handful of Silicon Valley utopians, ad men, and venture capitalists?’ Taplin considers a shift in the balance of power from old industrial capital to Silicon Valley and the platforms, arguing that ‘with this reallocation of money comes a shift in power’. Shoshana Zuboff's The Age of Surveillance Capitalism (2019), an academic publication that crossed over to become a best-seller, also identifies ‘the threat of unprecedented power free from democratic oversight’ of informational capitalism. Franklin Foer's argument in World without Mind (2018) follows similar lines as Taplin and Bartlett, with a focus on who ‘controls knowledge and information’, now in the hands of the ‘titanic powers’.

The People versus Tech is introduced by the claim that ‘The internet was meant to set us free’. The trope of disillusion or disenchantment recurs powerfully in this category, and with it a sense that freedom was never as present as the hype suggested. Another example is Morozov's 2012 Net Delusion. This ‘shows why internet freedom is an illusion. Not only that – in many cases the net is actually helping oppressive regimes to stifle dissent, track dissidents and keep people pacified.’ Fears about computers controlling humans also arise, although increasingly they fit better in the second category of Liveliness, which considers the replacement of the human by the machine (in Haraway's famous 1980s formulation, humans become less lively as computers take over more control). Computer power as malevolent or indifferent to human futures remains a strong trope, and one that, despite the rise of new forms of AI and the desire of many AI researchers to avoid the unthinking melding of matters of AI consciousness with matters of AI, remains remarkably stable – and does very often operate on this elision. For non-fiction versions, see later. In fictional SF dealing in singularity and/as apocalypse this blending is something of a staple.

(ii) Computers becoming more lively

Four of the books thrown up by Amazon's algorithms are usefully invoked here, suggesting how long-standing tropes revive but also have evolved. Callum Chase's Surviving AI (2nd edition, 2018), lays out the argument that computer autonomy produces human redundancy in existential terms – ‘If we get it right it will make humans almost godlike. If we get it wrong … well, extinction is not the worst possible outcome.’ Gerd Leonhard's Technology vs. Humanity (2016) also fears unthinking AI development: ‘The imminent clash between technology and humanity is already rushing towards us. What moral values are you prepared to stand up for – before being human alters its meaning forever?’ James Barrat's Our Final Invention (2013), meanwhile, explores ‘the perils of the heedless pursuit of advanced AI’. These three books exhibit symptomatic concerns and anxiety around singularity – and could also find a place in the first category.

The Glass Cage, by Nicholas Carr (2016), is more narrowly focused. It explores existing automation and its trajectories and considers the likely impact on human cognitive alertness or agency, in an account that stresses the automation of expertise. Amazon says it ‘shows how the most important decisions of our lives are now being made by machines and the radical effect this is having on our ability to learn and solve problems’. Human deskilling as the result of augmented computer intelligence is a concern relating not only to human cognitive capacities and their attention, but also to other forms or modes of life. It finds a specific form in contemporary automation anxiety. An obvious reference here, and it is referenced, is Martin Ford's Rise of the Robots (2016). This looks at ‘the terrifying societal implications of the robots’ rise …’, arguing that ‘any job that is on some level routine is likely to be automated and if we are to see a future of prosperity rather than catastrophe we must act now’; once again, we're in the last chance saloon.

Rather different forms of anti-computing that also inform this category are found, firstly in Rushkoff's Team Human (2019), a ‘manifesto’ which argues that ‘society is threatened by a vast antihuman infrastructure that undermines our ability to connect’. The living ‘community’ of humans is seen by Rushkoff as threatened by increasingly avaricious and active machines. Turning this around, Meredith Broussard (2018) captures another form of hostility in her look at computational stupidity – or what she terms ‘artificial unintelligence’. This returns to an old form of hostility to AI based on critiquing the justifiability of the claims made for it and for the timescale of advance. Like Weizenbaum and Searle (Chapter 6), Broussard's case is that AI does not (perhaps cannot) deliver on the promises it makes.

(iii) Computerization and the hollowing-out of everyday life and social interaction

This category has exploded. Older tropes come back in and inform the framing of new concerns and are, in this process, substantially revived; three forms of currently virulent anti-computing include addiction tropes and (less seen but still evident) other medicalized responses to pervasive connectedness, digital detox as a suggested response, and concern around noise and silence, screen life, and compulsive sociality, which becomes a figure for the lack of solitude or private life per se. None of these tropes is new (they appear in Leavis’ accounting with technologico-Benthamism, figure in moral panics around television and family life, reach back to Edmond Berkeley and Salon, and more broadly figure in critiques of industrialism itself in all its noise and clamour). However, they are currently being remade, sometimes in radically new as well as in familiar forms, and are expressed as matters of urgency, in registers comprehending anger, sadness – and even despair; reflecting this perhaps are tales of social media and its role in provoking youth suicide.

Of the many books that Amazon throws up for me to pith (since we are hollowing out), those invoked here are chosen because they take a particularly clear line. Psychologist Sherry Turkle's book Reclaiming Conversation (2016) is notable for revisionism; Turkle sees it as marking a turn away from the more positive viewpoint she held in her early engagements with digital culture – via Life on the Screen and The Second Self (2005), although a newly revised edition of the latter is an Amazon suggestion. Turkle now argues that a more critical engagement with the digital is needed. We were young, she says, and so was the internet, but now we need to grow up – there is a ‘backlash’ (interview in Vox, Illing, 2018). The argument of Conversation concerns the failure to maintain meaningful relationships across digital media platforms and within a digitally saturated life. A related work is Scott's Four Dimensional Human (2018), exploring surveillance and data-captured lives. It is claimed that the book asks ‘how do we exist in public with these recoded inner lives, and how do we preserve our old ideas of isolation, disappearance and privacy on a Google-mapped planet?’ Michael Harris's Solitude (2018) also engages with this – ‘In a world of social media and smartphones, true solitude has become increasingly hard to find’ – and goes on to delve ‘into the latest neuroscience to examine the way innovations like Google Maps and Facebook are eroding our ability to be by ourselves’.

Hooking into addiction tropes, already at hand from gaming, and/or video scares, but less invoked around computer culture/screen culture more broadly until relatively recently, is Mary Aiken (2017), who fears the deleterious effects of everything from ‘screens on the developing child to the explosion of teen sexting, and the acceleration of compulsive and addictive online behaviours (gaming, shopping, pornography) …’. Her The Cyber Effect ‘also examines the escalation in cyberchondria (self-diagnosis online), cyberstalking and organized crime in the Deep Web’. The four horsemen of the internet apocalypse, notoriously predicted by Electronic Frontier spokesman John Perry Barlow in the early 1990s as wreckers of the possibilities for a newly remade society somehow on the edge of old laws (Barlow, 1994), are right here, right now.

Mainlining on addiction is Adam Alter's Irresistible (2017), which argues that previously specific kinds of engagement with the computational have generalized: ‘Welcome to the age of behavioural addiction – an age in which half of the American population is addicted to at least one behaviour. We obsess over our emails, Instagram likes, and Facebook feeds; we binge on TV episodes and YouTube videos … Millennial kids spend so much time in front of screens that they struggle to interact with real, live humans’ – so says the blurb. This book is much invoked for its account of how Silicon Valley parents keep their own children away from the machines they design and market. Will Storr's Selfie: How the West Became Self Obsessed (2018), meanwhile, explores and attacks our addiction to ourselves via selfie culture. As Storr puts it, ‘our expectation of perfection comes at a cost. Millions are suffering under the torture of this impossible fantasy.’

Finally, there are the detox guides. For instance, Marchant's Pause (2018), bringing an ‘important message ready to be heard’, one whose time is right; ‘We check our phones an average of 221 times a day, we have apps that help us sleep and remind us to be mindful whilst we secretly measure our success in “likes”.’ Pause says we should stop doing this stuff. Time to Log Off (2017) brings news from a new anti-computing campaign. It is claimed that ‘[t]his canny little bible will help you log off and wake up to less stress and more time. Enjoy real experiences, real connections and real happiness. Reset your boundaries with carefully crafted exercises, new outlooks and wise words from Tanya Goodin, digital detox specialist and founder of Time To Log Off.’ Goodin, like Matusow (see Chapter 3), might be said to have a ‘gimmick’. Tero Karppi argues that logging off isn't that simple. Her Disconnect (2018) reminds us that ‘Facebook see disconnection as an existential threat – and have undertaken wide-ranging efforts to eliminate it’. The argument is that users’ ability to control their digital lives is, even if they wish for control, gradually dissipating, so that personal anti-computing tactics may not be effective enough these days.

(iv) Computer technologies and the threat to human forms of culture

Some of the older tropes expressing these concerns appear somewhat submerged currently and/or have morphed considerably. An old hostility to gaming versus reading, for instance (e.g. Stallabrass's 1996 attack on interactive multimedia as a shift from the illusion of scene to the illusion of action), has morphed into a concern around excess screen time in general. This produces new links to addiction tropes (often nascent in gaming critiques), particularly in relation to critiques of addiction by design, or the tactics of the software companies to build in the demand for ‘more’ as part of the attention economy. An example is the invitation to keep binge watching; it is not only in gaming that there is to be no game over (see e.g. Lanier, 2018). There is a more general critique of gamification – which links less to creative activity in discrete zones than to the gamification of everyday life or the reduction of experience to the dashboard. This is implicit at least in Rushkoff's work (see earlier) – and though it didn't arise in my search it is also treated with in critiques of bureaucratization such as that by David Graeber (2015). Striking hostility also arises around gaming, but less around the form than around gaming as a cultural practice mired by discrimination and gender hatred; gamergate – when members of the gaming ‘community’ attempted to vote down forms of SF whose rise was perceived to be bending the knee to liberal feminism (see Quinn later) or women – is invoked. A different set of anxieties and hostilities arising (with vertical take-off) around cultural production include deep-fake issues. In part, these revise and rework old Photoshop debates arising in relation to news photography and point to a longer-term anxiety about authenticity and ‘the real’ – in relation to computer art and music, and in relation to fake news and/as public culture (discussed further later).

This category might also encompass hostile responses to (the lauding of) our own remaking as cultural producers. Concerns of this kind are voiced by Andrew Keen, whose 2008 The Cult of the Amateur was invoked despite its relative age. This argues that ‘much of the content filling up YouTube, Myspace, and blogs is just an endless digital forest of mediocrity which, unconstrained by professional standards or editorial filters, can alter public debate and manipulate public opinion’. Written more or less as a response to the moment of Web 2.0 and its promise not of freedom, which came and went with Web 1.0, but of freedom of production and the claimed democratization of creativity, it appears less relevant to contemporary debates.

(v) The general accident/catastrophe theory

This category gathered together fears about computers spiralling out of control. It is concerned with computer power and complexity and unforeseeable risks. What appears to be emerging in relation to this category is a shift not so much in the magnitude of the perceived threat computation might pose, but in its articulations. Discourse around the Anthropocene, and more generally critiques of untrammelled expansion and its already feared and actually irreversible impacts now more clearly intersect with a particular kind of discussion of computation as part of (disastrous and unchecked and unexamined) growth agendas. This is the background informing writings critical of the degree to which computing is – despite its apparent lightness – energy hungry and environmentally unfriendly, particularly for workers in the global South, as well as for the natural environment in its totality. A recent example of this, slightly early for my Amazon trawl, might be Miller and Maxwell's (2020) work on the mobile phone.

There is also the sharp revival of older tropes framing automatic weaponry; once chiefly expressed via Terminator-style fantasies, these are now critiqued and explored in relation to nascent and actual real-world weapon systems, and/in their covert but also leaked operations. Finally, advances in real-world AI (machine learning) might be behind a revived interest in or new sense of credibility around the actual instantiation of powerful AIs – and with them concerns about the general accident of a ‘hostile’ alien singularity (conscious or not; I refer here to hostility metaphorically) wiping us all out. These concerns cross-list with other titles already named in earlier categories.

(vi) Horrible humans

This label designates that form of anti-computing that doesn't ‘blame’ computers for societal ills but excoriates humans for taking advantage of what computers increasingly enable humans to ‘get away with’. Originally it included concerns chiefly focusing on ‘people’ rather than governments or corporations. This capacious category now expands to accommodate rising dissent and anger about new media organizations (the platforms, chiefly) and their role in the rise of new forms of populism, extremism, intolerance, bigotry, hatred, and aggression in public spheres. Something striking about this form of anti-computing today, perhaps what makes it distinctively of the moment, is the degree to which questions concerning technology, morality, and ethics (e.g. of good and evil) are invoked at many scales simultaneously, and in relation both to personal lives and the conduct of politics and public life in general.

Moral panics (Cohen, 2011), originally defined in 1972 around media content, forms and formats, were of course always political (in that they were ideologically framed, related to dominant ideas and conceptions, and also to material conditions/relations). What is now evident is a form of anti-computing sentiment that amounts to a moral panic about the relationship between the personal and political; what is being critiqued is the (populist) form of the political, of politics, or political life, or public life, or civil society, that societies are (felt to be) letting happen by failing to rein in or control the computational and computational corporations. Amongst the books taking this kind of line is Sarah Jeong's Internet of Garbage (2018), which asks how we filter wanted content from the ‘garbage’ found online: ‘Content platforms and social media networks do not have the power to restrain stalkers, end intimate partner violence, eliminate child abuse, or stop street harassment …’ This both identifies the problem and suggests cures, claiming that it would be possible to ‘cultivate better interactions and better discourse, through thoughtful architecture, active moderation and community management’. Jeong's work is a call for responsibility to be taken up, but explicitly not a call for more radical societal change.

Jaron Lanier, an early Silicon Valley apostate, excoriates the ‘designed in’ toxicity of social media and suggests how to get off it. His Ten Arguments for Deleting Your Social Media Accounts Right Now (2018) comes with a blurb striking for its affective language; the cruel, dangerous, fearful, isolated, tribal, are all invoked:

Lanier is clear that if these are horrible human emotions, they are produced by our dangerous relationship with unethically designed machines and communication models. He goes on to argue that we need to act to help ourselves to get out of this. The ‘ripping apart’ effects of digital media are said to be reorganizing politics and encouraging forms of dangerous populism. An account of this is given in Mike Wendling's Alt Right: From 4chan to the White House (2018). This ‘reveals the role of technological utopians, reactionary philosophers, the notorious 4chan and 8chan bulletin boards, and a range of bloggers, vloggers and tweeters, along with the extreme ideas which underpin the movement's thought’. Other publications look at trolling as a human behaviour generated by the global techno-social culture we live in. Notably, Whitney Phillips's This Is Why We Can't Have Nice Things: Mapping the Relationship between Online Trolling and Mainstream Culture promises to frame trolling in classically horrible human terms, and as mainstream:

Ginger Gorman's Troll Hunting (2019) is also insistent that trolling is widespread, but apparently discerns no coherent logic behind it. The argument, we are told, is that

Trolling produces accusatory responses attacking the forms of computational culture that enable or encourage these kinds of action. What is also evident is the desire to offer cures rather than societal-wide ‘solutions’ (political responses narrowly defined). Self-help is thus very prominent in the suggested readings. An indicative version comes from Sherri Mabry Gordon (2018), who offers teens help in Coping with Online Flaming and Trolling, and the ‘Shock. Disbelief. Pain. Embarrassment’ it causes.

There is a line of critique that fits into the category of horrible humans because, whilst attacking the culture of the internet as sexist, racist, and classist, it does not recognize structural inequalities. The tropes reached for here tend to be personalized – they include critiques of those who attack, and assessments of damage done to individuals. Shame Nation (2018), ‘with a forward from Monica Lewinsky’, demands the formation of new forms of internet civility, but again is unabashedly a (self-)help book: ‘An essential toolkit to help everyone – from parents to teenagers to educators – take charge of their digital lives’, we are told.

From a different perspective comes Misogyny Online: A Short (and Brutish) History (2016) from Emma A. Jane. This ‘explores the worldwide phenomenon of gendered cyberhate as a significant discourse which has been overlooked and marginalised. The rapid growth of the internet has led to numerous opportunities and benefits; however the architecture of the cybersphere offers users unprecedented opportunities to engage in hate speech.’ Finally, also clearly hovering between this category and one more overtly political and social, but again placed in here since the tone is one of personal distress and the mode of address is to ‘you the user’, is Zoe Quinn's account of gamergate and after in Crash Override (2017). This promises an ‘up-close look inside the controversy, threats, and social and cultural battles that started in the far corners of the internet and have since permeated our online lives … Quinn provides a human look at the ways the internet impacts our lives and culture, along with practical advice for keeping yourself and others safe online.’

(vii) Standardization/quantification

Developments in this area are critiques of machine learning and what it leads to. Of particular note is automated bias – when machine learning learns from human sources and the results are forms of accelerated or intensified, and apparently more deeply inscribed, or reinstitutionalized bias. One of the best-known examples is Tay, the bot that learned to be a Nazi. Offerings here included Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor, by Virginia Eubanks (2018), and Safiya Noble's Algorithms of Oppression: How Search Engines Reinforce Racism (2018). Attacking in the same vein, but exploring gender bias, is Caroline Criado Perez, with Invisible Women: Exposing Data Bias in a World Designed for Men (2019l; see also Sara Wachter-Boettcher, 2018).

Somewhat differently, there is Cathy O’Neil's Weapons of Math Destruction (2017), in which a ‘Wall Street quant sounds an alarm on the mathematical models that pervade modern life – and threaten to rip apart our social fabric’:

O’Neil's sense of new architectures of unfreedom – of structural developments not amenable to personal solutions, but addressable only by way of some more substantial series of changes, resonates with Morozov's To Save Everything, Click Here: Technology, Solutionism, and the Urge to Fix Problems that Don't Exist. Published in 2014, this continues to arise as a contemporary reference point in reframing and refinding earlier critiques of standardization/solutionism and reinvoking them in relation to data and discourses of ‘smart’. As the blurb tells us: ‘Our gadgets are getting smarter … we're told … it will even make public life – from how we're governed to how we record crime – better. But can the digital age fix everything? Should it? By quantifying our behaviour, Evgeny Morozov argues, we are profoundly reshaping society – and risk losing the opacity and imperfection that make us human.’

(viii) Too much information

The final category in the taxonomy found space for works responding to reviving concerns around over information overload, but also included anxieties about the fetishization of information capture, the fetishism of facts, what is viewed as the overproduction of information – whether as data, text, image – and the prioritization of production and circulation over interpretation. Older terms expressing this concern included fears around data deluge, or information overload. There is also the now largely forgotten term ‘information anxiety’, credited to Richard Saul Wurman who wrote a book of that name (1989). Other early reference points include ‘As We May Think’ (Bush, 1945), and/or Alvin Toffler's 1970s best-seller, Future Shock.

Today, slightly fading now but still thrown up as a reader suggestion, is Nicholas Carr's The Shallows (2011): ‘not since Gutenberg invented printing has humanity been exposed to such a mind-altering technology. The Shallows draws on the latest research to show that the Net is literally re-wiring our brains inducing … superficial understanding …’. These concerns map onto the attention economy and the demands it makes, and underscore ways that connections between this category and others are tighter than they were; the use of automation to handle information that is ‘too much’ for humans, and attendant concerns around autonomy are of note here.

Where this category has re-emerged is around critiques of information politics – the hollowing-out of reflection via always-on news, the prioritization of the virtual – as a mode of social being, the fragmentation of earlier forms of public life (public spheres, commons), and the rise of balkanization. An inaugural trope in this category concerns filtering, and its deleterious effects continue to be observed; the move is from zip codes to filter bubbles and echo chambers. Pariser's Filter Bubble, What the Internet Is Hiding from You (2012), an early swallow in a big summer, is still invoked. The Filter Bubble was part of that rising tide of dismay that later reached huge proportions around fake news, Trump's election, and (in the UK) Brexit. The point to be made here is that information overload surfaces with a new kind of slant; now the worry is at least as much about the various ‘cures’ – attention, selection, automated selection, changes in reading habits towards scanning (sixty-four books, for instance …), nudge, filtered news, and the rest – as about the sheer amount of information. This at least is what surfaces through the Amazon trawl which threw up these latter points, rather than evidencing a concern with information and/as a deluge per se. These concerns do not necessarily link to big data and solutionism, nor to concerns around information or knowledge as complete but, rather, focus on how the flood of information challenges interpretation and raises questions about explainability.

A conclusion

This exercise in distant reading suggests to me that the original taxonomic division of anti-computing, which set out a series of persistent characteristics, does connect to (and does enable us to grapple with and reframe) anti-computing as it arises today. It is justifiable to assert that anti-computing persists across the decades of computational instantiation in recognizable forms, even as it also shifts, morphs, finds new targets, or reshapes older ones. It also changes. The long-standing sub-categories continue to make sense but they strain and bulge, are deformed and reformed as they take in contemporary anti-computing formations. These deformations articulate the form anti-computing takes in the specific moment, in relation to earlier forms, in relation to the time in which it finds itself, in relation to technological developments of that time, and in relation to its political culture.

Chapter 1 also set out some supplementary taxonomies for anti-computing, cross-cutting the main series. One categorized anti-computing strains by dividing various forms, degrees, and kinds of affective engagement to produce an emotional register (see Figure 1). This offers a key to understanding the contemporary formation. The Amazon crawl points to the intensity of responses to the computational currently evident, and the degree to which these continue both to distinguish and to blur distinctions between concerns around the ontology of the computational and those related to issues of social and political power, control, and domination. Anger around gender and race bias and the concern about human hatred is palpable as a framing of many of the works found in the Amazon trail (and clear elsewhere, of course). The strength of the feeling of disgust, not only with the computational as machine culture but with computational culture as it has mirrored the dark side of human culture, is marked. Supplementing this, a more ambivalent orientation – and often one with a less intense affective charge – is evident in the framing of works exploring singularity and/as AI, and the issue of belief and the matter of its relation to caring here raises its head again. Chapter 7 grappled with the paradoxical mix of formal acceptance and emotional and even rational scepticism surrounding singularity's claims. Two emotions – viscerally felt disgust, and an insincere fatalism (rather than a cruel optimism, perhaps) arising out of a form of unbelief – are both in evidence in contemporary anti-computing's affective geography, and their measure as it were is reallocated across the various categories of the main taxonomy. There is remarkable cheer in many of the publications declaring the end of the world, and the help books are (not surprisingly, given their sales imperative) far more willing to countenance ‘solutions’ than their own arguments would appear to suggest is possible.

Algorithms of course do not feel emotion. Moreover, they do not necessarily parse human emotions well. Perhaps it was the emotional inadequacy of Amazon's algorithm that meant that every so often it threw up a blisteringly positive account mixed in with the negative analysis, hostility, dissent, and relentless advice (amongst the celebrations from the Left was Bastani's 2020 Luxury Automated Communism). But this can also be understood as a symptom of the entanglement between these two orientations, pro- and anti-computing, pointing to their intertwined histories, and to the roles each plays in the future of the other, and this seems to me a sustainable view, given the overarching formations and the particular cases explored in this book. Anti- and pro-computing formations, that is, travel on together, each being less visible, or powerful at various times, in relation to the other, even whilst the systematic dominance of the one over the other persists – and has persisted across the decades of the development of computational capitalism, with its relentless prioritization of more growth.

These two sets of observations, on remapped affectivity, and on the entanglement between computational avidity and dissent, inform some concluding remarks. These include a moment of critique and dissent. What I dissent from is disgust and morality as sufficient frames through which to critique the computational, and computational culture in general, in neoliberal times. I also distance myself from self-help (or how to feel better) as a response to computational ills. These forms of anti-computing are, if not encouraged, then given some purchase in current times even by the industry itself, notably by the platforms, for instance. To contest them is to undertake a form of conditional anti-computing which is at times anti-anti-computing. It demands a commitment to radical structural politics rather than personal solutionism, and/or the advocacy of automated cures to computer-assisted human badness (behaviourism at one end, bot-censorship at the other; the age of the super-nudge).

Elaborating this demands back-pedalling slightly to regloss the claim that the anti-computational comes with the computational. Consider that forms of absolute refusal, or the desire to end computing absolutely, are either vanishingly rare or have existential implications. When I began this book I intended to write a chapter on total refusal but it turned out to be almost impossible to find. Even preppers connect, albeit carefully; religious movements against Western modernity, notoriously, use the internet; whilst primitivism as a movement is even, if not conspicuously, not entirely against computer use (see e.g. Zerzan, 2012). Seeking to abolish the computer itself as a mode of production, or the computational as a form of life, or even to abandon all forms of computational thinking, is rarely what is called for, or desired. To some extent this is a matter of practicalities; in an era of intense saturation, where there is no way out, how low can you go? Or, rather, where can you go? Or how low and where can you go in a globalized world encircled with satellites, where older forms of communication shrivelled up and died in the face of the new? One of the arguments of this book has been about forgetfulness and the appearance of naturalness in the technological which becomes an embedded part of life. If this constrains thinking within the horizon of the computational, it is also because life itself is lived within this horizon.

There is one site where complete decomputerization is countenanced. Certain visions of the end of computing link in to accounts of the end of the human, the end of human civilization, perhaps the end of the Earth. A form of anti-computing shades into apocalypse-thinking, and in this way also perhaps into some of the most anti-human wings of environmental politics; the presumption that the Earth is better off, or only can survive and regreen without us, and without any of our technologies. Derrida (1984) argued that this horizon is unthinkable – even speculatively – and this side of the apocalyptic prediction/desire for apocalypse, anti-computing is less than absolute. It variously seeks a kind of amelioration, or a process of modification, or a new deal. It wants to check impulses, or companies, or expansion into certain areas of life. That doesn't make it altogether – or even necessarily – reformist. The anti-computational may constitute an element of radical, critical, even revolutionary thinking on the technological. Indeed, because anti-computational is essential to understanding the computational, is part of computational culture, it can be used to define and understand another set of divisions within contemporary techno-cultural thinking; that is the divide between anti-computing positions that entail a radical (structural) critique of computational capitalism and those that do not. Here I myself come off any fence I might be thought to still be sitting on and declare for systematic critique.

In case I am misunderstood here, that doesn't mean arguing that real interest, or significance, in the end inheres only in writings or activities exploring radical or systematic critiques of the technological, or computational, moment. On the contrary, exploring the range of forms anti-computing has taken and takes, as intellectual points of view, shared positions, public arguments, as discourses that emerge, submerge, revive, and appear in new registers (personal, political, moral, political, as campaign, as life story, as political event, as software, as fiction) as they travel, and looking at how they travel and how they land has been the point of this book. Understanding these positions and their operations can contribute to understanding and can assist in breaking through that presentism and those forms of ‘realism’ that constrain thinking about the technological. I put this in scare quotes with reference to Le Guin's discussion of more radical realism invoked in Chapter 7, and also with reference to Mark Fisher's (2009) elaboration of capitalist realism as that horizon which appears to prescribe the possible.

Amongst these constraints are those that systematically drive thinking towards the personal and/or towards a particular conception of responsibilization and atomization; social being as individuated being, so that the anti-computational is itself experienced as a personal response and one that responds to the discrete issue rather than a historically instantiated and developing world order; a techno-political economy.

This orientation explains the heavy leaning towards help, the personalization of these modes of address, which are a characteristic of the anti-computing books we are recommending to each other through our Amazon-circulated ‘reading’ practices. They are, of course, at one with computational shifts in social life, social commons, and social being. We are increasingly invited, indeed organized, in so far as possible directed/nudged, into positions in which our sense of where we may respond, our sense of ‘responsibility’ even, and our sense of where the solution to our problem may be found, shifts towards the personal/personalized sphere.

Questioning the implications of this shift in focus, as itself a mode of critical anti-computing, rather than simply locating an anti-computing trope within it, is therefore significant. Anti-computational thinking on the whole breaks with the assumption that if it is bad today it will get better tomorrow. It can help to constitute an effective intervention partly because it does not partake in a particular kind of expectation for the automated revivification of faith in progress through technology, the kind of technological optimism that says technology will win through, will provide the cure, will make everything new again. Instead, it reaches for those other stories, those other tropes and traditions, through which it thinks, across which it invokes or makes an understanding of the new in the context of the pain of the old, and its trouble.

This can produce a nostalgic orientation, and one that is, in the current climate, in danger of a kind of revived parochialism, certainly as it relates to global culture on the internet, beyond the walls many in control want to build. But that other kind of anti-computing can find in this refusal to be forgetful a form of hope, the kind of hope Walter Benjamin found in refusing to let the past – and its costs – be covered over; that sees that it rises, and sometimes helps it to do so (Benjamin, 2006). Perhaps we should follow Terry Eagleton (2015) in distinguishing between hope (which contains possibility and uncertainty) and optimism (which gambles on less, in exchange for more security). If there is a sense of cruel pessimism (contra Berlant's [2011] well-known formulation of cruel optimism) in the fatalistic anti-computational assessments of the place we find ourselves in, there is also the hope that comes with the demands anger makes. Some of the responses to terrorist events involving networks and social media, to algorithmic bias, to the alt right and its channels, those responses that demand solidarity and refuse hatred and that demand an exploration of what structured events rather than simply asking how they circulated, exhibit that hope. Let that be a place to end.