Bergen, Norway.
Not least for it’s sheer optimism, I was happy to read Paul Mason’s new book, Postcapitalism. In it, Mason makes two broad and largely commonsensical observations regarding the rapid development of information technologies – first, that as our information technology is getting increasingly more sophisticated, it is possible to automate away more and more jobs; second, that as this happens, more and more people will spend their time producing information of different kind – suggesting that these two development will together force capitalism to evolve, largely without any violence or class struggle, into a “postcapitalist” world order.
Is he right? Probably not.
Mason’s main suggestion is that the more our economy is organized around the production of information and knowledge, the less will it make sense for this economy to take the form of capitalism. Why? Because, as Mason puts it himself, “markets are based on scarcity while information is abundant”. Imagine here the difference between sharing an apple and sharing an idea. If I have an apple and share it with you, I end up having only half an apple. If on the other hand I share an idea with you, I get to keep the whole idea only now you have it too. Thus, while ideas are inherently abundant, apples are inherently scarce – and capitalism, Mason points out, is really only equipped to deal with the production and circulation of the latter category of things.
In economics, this difference between scarce and abundant goods is known as a difference between rival and non-rival goods, and the main reason why our current economic system is so bad at dealing with the latter is because money is itself a rival good. If I share twenty dollars with you, we both have ten and not twenty each. The problem is that, when the exchange of information (as well as other non-rival goods) is mediated by money, we will all have to act as if information is also a rival good, and that sharing it for free consequently diminishes it’s value – which, of course, is true in terms of the money that we use to measure its exchange-value, but horribly misleading in terms of the actual use-value of the information. To quote open source pioneer Aaron Swartz, when we do treat information as a non-rival good, “It’s called stealing or piracy, as if sharing a wealth of knowledge were the moral equivalent of plundering a ship and murdering its crew. But sharing isn’t immoral – it’s a moral imperative. Only those blinded by greed would refuse to let a friend make a copy”. Popular wisdom notwithstanding, then, while it is no real problem to equate apples with oranges, we run into serious troubles when the equation is made between apples and information.
This, I would say, is the gist of Masons diagnosis. Let me now suggest why I don’t think it will lead to “postcapitalism”.
That we are currently living in an economic system modelled on the production and circulation of rival goods, yet witness how this system is increasingly based on the production and circulation of non-rival goods, is no doubt a major contradiction, and Mason is right to suggest that its severity should make us question the very viability of capitalism. Of course, he is not the first to do so, nor to do so precisely on this ground. For example, literary theorist and activist Michael Hardt made a similar case quite a few years ago, suggesting that the shift in capitalism from an older reliance on “material production” (production and circulation of rival goods such as shoes and apples) to a new reliance on “immaterial production” (production and circulation of non-rival goods such as information and ideas) would bring about nothing less than the material conditions of a communist revolution, as the free sharing of information, ideas and images would, as he phrased it, put “the common in communism”. That hasn’t really happened, though.
More than anything, it seems to me, the main problem with this particular line of argument is the assumption that, by demonstrating that capitalism is ripe with contradictions (preferably adding some new contradiction to the list), one has effectively demonstrated that the end is near, too. This, however, is not at all self-evident. After all, precisely this prediction has been made many times before, and we are still living in capitalism. On the other hand, of course, to point this out is just as little an argument for why capitalism could never fall under the weight of its own contradictions. So let’s see if the argument seems plausible, this time around.
At one level, Mason’s critique is the oldest one in the book; in fact, it’s found on the very first page of Marx’s Capital. This is the observation that capitalism rests on a contradiction between use-value and exchange-value. As we’ve seen, in Mason’s argument, this manifests itself as the newly emerging contradiction that an exchange-value modelled on scarcity is increasingly relying on the production of abundant use-values. But of course, in its most everyday expression, the contradiction between use-value and exchange-value has been around since the birth of capitalism. As David Harvey puts it: “Nothing could be simpler. I walk into a supermarket with money in my pocket and exchange it for some food items. I cannot eat the money but I can eat the food. So the food is useful to me in ways that the money is not.” The reason why this amounts to a contradiction in its own right is because those use-values that I buy at the supermarket are ultimately produced in order to maximize exchange-value – and in the long run, the maximization of exchange-value (which in terms of total economy means endless compound growth) will lead to the eventual depletion of use-values, not their maximization. This, at least, was Marx’s prediction when he wrote that “[c]apitalist production … only develops the techniques and the degree of combination of the social process of production by simultaneously undermining the original sources of all wealth – the soil and the workers”, a statement that has turned out to hold water, considering the unprecedented levels of both ecological and human devastation that two centuries of capitalist civilization has brought the world.
Still, even if we accept that the contradiction between use-value and exchange value is both foundational to capitalism and at the root of much of our ecological and social misery – and I think that this is a proposition that stands up to testing – it doesn’t in any way support the claim that capitalism is about to come to an end. It might very well imply that human society as a decent and ecologically sustainable place is coming to an end, but this is not in itself something that will compel capitalism to “evolve” into something better (as should have been made abundantly clear by Naomi Klein’s latest book, This Changes Everything). No matter how much the internal dynamic of capitalism contradicts our ecological and social needs, capitalism will only crumble under the weight of its own contradictions once its ability to produce economic growth is severely threatened. Otherwise, we will still, unfortunately, have to struggle against it.
In order to be considered plausible, therefore, Mason’s argument will have to demonstrate just this: how the contradiction between scarcity in exchange and abundance in production will actually threaten the logic of the exchange system. If this can’t be demonstrated, there is once again no reason to assume that capitalism will wither away, as Mason predicts, on the backs of “something more dynamic that exists, at first, almost unseen within the old system” – a new
use-value regime that finally will be able to abolish capitalism by “reshaping the economy around new values and behaviours”. Fortunately, there are ways to test most of this against empirical evidence.
The strongest reason provided by Mason as to why his particular version of the exchange-value/use-value contradiction will prove fatal to capitalism is that people have already started to develop alternative exchange systems – ones that are much more in synch with the abundance characteristic of the information economy – within the shells of the old market system. “Almost unnoticed,” he writes, “in the niches and hollows of the market system, whole swaths of economic life are beginning to move to a different rhythm. Parallel currencies, time banks, cooperatives and self-managed spaces have proliferated, barely noticed by the economics profession, and often as a direct result of the shattering of the old structures in the post-2008 crisis.”
Notice, here, how Mason repeatedly tells his readers that these new and supposedly postcapitalist economies are “almost unnoticed”, “barely noticed”, and “almost unseen” – by people in general, and economists in particular. To me, this seems to be an important ideological function of the argument. What apparently is being insinuated is that, if these new economic forms are “barely noticed by the economics profession”, then, surely, this must be because the economics profession could never incorporate them into their theory of capitalism. Their absence from economic theory thus proves their radical potential; their marginality signals their significance.
Yet as with any message that is too insistently repeated – think of how frequently racist parties insist that they are not racists – suspicion lurks that it betrays its opposite. And sure enough, if one actually consults the economics profession, what one finds there is a rather large (not to say “abundant”) literature dealing exactly with this transition from a form of capitalism based on rival goods, to one based increasingly on non-rival goods. What has gone almost unnoticed by the economics profession, one the other hand, is the prediction that this will somehow lead to postcapitalism.
One good example is a book released 2014, The Second Machine Age, by the two MIT-economists Erik Brynjolfsson and Andrew McAfee. Like Mason, these authors argue that we are currently witnessing a second technological revolution, one that is “doing for mental power – the ability to use our brains to understand and shape our environments – what the steam engine and its descendants did for muscle power”, the former being “at least as important for progress and development – for mastering our physical and intellectual environment to get things done – as physical power”. And just like Mason, they identify one crucial difference between the two, which they also locate in the distinction between scarcity and abundance.
For Brynjolfsson and McAfee, however, the story begins with the relation between economic growth and wage developments. Historically, when society has seen increases in productivity, the general result has been an increase in living standard for the majority of the population. The reason behind this, Brynjolfsson and McAfee say, is that whenever capital could produce the same amount of goods from less labour, both production and wages have tended to go up – so that, as the total supply of goods and services increased, effective demand did so as well. Society thus produced more commodities, but it also produced a workforce that could afford to buy more.
For economists, this has in many ways been the big argument for economic growth. But, Brynjolfsson and McAfee argue, with the rise of information technologies, economic growth has been largely disconnected from positive wage developments. For example, they show how between 1999 and 2011 median income in the US fell from almost 56 thousand dollars per year to just above 50 thousand – and this despite the fact that GDP hit a record high in 2011! The reason behind this new disconnect, Brynjolfssen and McAfee argue, has to do with the kind of productivity increases that information technologies are likely to bring about. For example, while it makes a great deal of macroeconomic sense to increase wages when technological development allows society to produce more cars or houses from the same amount of labour input – since all of these new things have to be bought by someone – it makes pretty much no sense to do that when technological development allows people to share information, ideas, or music with each other for free online. Yet, Brynjolfsson and McAfee say, it is precisely this latter kind of technological development that has been driving economic growth the last twenty-or-so years.
Predictably, the result has been a “trickle up” of wealth to a small percentage of the population – economic “superstars” like Bill Gates and Mark Zuckerberg. What’s new about this new group of “economic winners”, Brynjolfsson and McAfee say, is that they make money not so much from producing new things and selling them to people, but by owning and controlling different information technologies that manage to be productive without having to invest in almost any labour (Facebook, for example, has only about nine thousand employees). Instead, the use-values that are produced from these new technologies are produced for free by everyone, all the time – from sharing images on Instagram to sending an email to uploading text on Facebook.
Up to this point, then, the story is more or less the same as Mason’s. Markets are not designed to deal with non-rival goods, because either they force us to behave as if sharing them is like sharing an apple (installing copy-rights and other forms of enclosures in order to create artificial scarcity), or they simply have no way of expressing the economic value of this new abundance, since nothing can both have a market value and be shared for free.
Beyond this basic observation, however, disagreements begin to surface. Mason, as we’ve seen, describes this new sharing economy as a form of “collaborative production” in which “goods, services and organisations are appearing that no longer respond to the dictates of the market and the managerial hierarchy”, mentioning Wikipedia as an example of a product that “is made by volunteers for free, abolishing the encyclopedia business and depriving the advertising industry of an estimated $3bn a year in revenue.” In contrast to this, Brynjolfsson and McAfee argue that, while it is generally true that it’s become harder and harder to get paid for producing information of different kind (from books to music to software), this does not mean that it’s no longer possible to make money from the production of such things. The only real difference is where that money ends up: in the hand of capital (the owners of the means of production) or in the hand of labour (the actual producers). And, as Brynjolfsson and McAfee have shown in the form of falling median wages relative to GDP, a major transference of wealth from labour to capital has indeed taken place.
Now, as credit to Mason’s argument, were this particular trend to completely dominate economic development, we would no doubt be looking at a process that could eventually only spell the end of capitalism. This, indeed, is precisely what Mason has in mind when he references Marx’s short text from Grundrisse, “The Fragment on Machines”. Mason writes: “In an economy where machines do most of the work, the nature of the knowledge locked inside the machines must […] be ‘social’. In a final late-night thought experiment Marx imagined the end point of this trajectory: the creation of an ‘ideal machine’, which lasts forever and costs nothing. A machine that could be built for nothing would, he said, add no value at all to the production process and rapidly, over several accounting periods, reduce the price, profit and labour costs of everything else it touched.” The key phrase here is “costs nothing”, which of course means “requires no labour input” (in production as well as maintenance). And sure enough, a capitalist economy running entirely on machines that require no labour input whatsoever would soon collapse, because no labour means no wages, and no wages means no demand, which in turn means that these “ideal machines” would end up producing commodities without any market value.
The problem is that Mason seems to believe that this is essentially a realistic description of the direction that our economy is taking. It emphatically isn’t. In fact, people are working more than ever; and, following Brynjolfsson and McAfee, it shouldn’t really surprise us that this is the case. After all, if economic growth is increasingly being produced in sectors of the economy that require little or no labour, the immediate result is not that capital becomes weaker but stronger. And capital, though it has been incredibly good at coming up with new technologies that allow us to produce more goods for less labour, is fundamentally not about such technological advances. Instead, it is and will always be about the purchase of labour. As John Stuart Mill once argued, never was a labour-saving device invented that actually saved anyone a minute’s labour. The irony is that this is something that Mason seems to be fully aware of, as he rightly describes how a fully automated capitalism would self-implode due to a lack of effective demand (or at least makes an approving reference to Marx making this point). However, he seems to imagines that capitalism does not get it.
The question then is: does capitalism get it? It seems so. While information technologies have greatly reduced capital’s need to employ labour in certain sectors of the economy – facilitating in the process a huge transference of wealth from labour to capital – this has not in any way kept capital from employing labour in ever-new ways. In particular, capitalism has been very good at producing low-paying jobs in the service industry. In the US, for example, the three biggest employers – in order, Walmart (with 2.2 million employees), Yum! Brands (owner of KFC, Taco Bell and Pizza Hut, with 523 000 employees), and McDonald’s (with 440 000 employees) – are all low-pay, low-security jobs within the service industry (and out of the ten biggest employers, six are in the service industry). Again, this should not surprise us. As Brynjolfsson and McAfee points out, when it comes to replacing workers with robots, “high-level reasoning requires very little computation, but low-level sensorimotor skills require enormous computational resources”, so while robots have long since surpassed humans in playing chess or figuring out which is the best way to drive to the airport during rush hour, we’re yet to see a robot that is cheap and capable enough to replace even the worst McDonald’s employee (this tendency is known as “Moravec’s Paradox”).
Still, it would probably be wrong to attribute the persistence of human labour in the service industry solely, or even mainly, to the inability of robots to make a Big Mac. As a large number of social theorists have noticed (e.g. Arlie Hochchild, Kathy Weeks and Hardt & Negri) most of the what’s being produced in the service industry is largely indistinguishable from the social relations in which it is being produced. As Weeks puts it, service work “involves putting subjectivity to work in jobs that are less about manipulating things and more about handling people and symbols”. In other words, what’s being asked of people in the service industry is not simply that they perform mechanical tasks that robots aren’t (yet) able to perform, but precisely that they don’t act like robots – selling not only Big Mac’s or T-shirts but also their care and their smiles. And recent movies such as Ex Machina and Her notwithstanding – both telling the story of people falling in love with AI:s – we are most likely far from seeing robots whose smiles, touches, compliments or words of comfort produce the same effects as when coming from other humans.
Yet we can’t all be service workers – that, after all, would rather defeat the purpose. According to Brynjolfsson and McAfee, therefore, the best way for individuals to “race with machines”, as they put it, is to do the second thing that robots aren’t very good at, besides being human beings, which is to come up with innovations. “We’ve never seen a truly creative machine,” they write, “or an entrepreneurial one, or an innovative one”. These generic references to “creativity” and “innovation”, of course, really signify something quite specific. For Brynjolffson and McAfee, this is on the one hand the ability to come up with information and ideas that are truly new and possible to sell (like a poem or a journalistic story), on the other the ability to come up with information technology that is truly new and possible to sell (like Facebook or Windows). “These activities”, they write, “have one thing in common: ideation, or coming up with new ideas or concepts. To be more precise, we should probably say good new ideas or concepts, since computers can easily be programmed to generate new combinations of preexisting elements like words.”
Now, while this observation about ideation is obviously true (though also something of a truism), the more interesting thing about these two forms of creativity, it seems to me, is the way in which they differ from each other. This perspective receives no explicit discussion from Brynjolfsson and McAfee, yet it comes out of their description. The first kind of creativity, they say, achieves a market value mainly by having not yet been made redundant by information technology. “We’ve seen software that could create lines of English text that rhymed,” they write, “but none that could write a true poem”, assuring the world’s poets that they can still feel safe. At the same time, this is a rather fragile assurance, because the second kind of creativity is geared precisely towards coming up with technology that will make other people redundant – from open-source encyclopedias like Wikipedia to, indeed, computers that can produce all sorts of creative writing. So while human beings might “race with machines” by being creative in a uniquely human way – thus complementing the machines – there is always the risk that some other human will create a machine that can simulate this ability, rendering it basically useless as a way to earn a living.
Perhaps however it’s time now to call this for what it is: a profoundly ideological vision. Consider, again, the short comment above on computers and poetry. Here, the implicit message seems to be that the only real difference between a computer and a poet is that the former cannot (yet) produce credible poetry; “a true poem”, as the author’s put it – and not that poetry is essentially a form of communication between human beings. Apparently, from the point of view of economics, the only question that matters is that of the Turing test: can you tell the difference between a human and a computer? If you can’t – if the computer has made you believe that a human wrote the poem – then the computer has effectively created poetry. Simple as that. Still, it would seem that for anyone who is seriously interested in poetry, its value lies not in its ability to simulate human emotions but to actually express them – and in so doing invite the reader to share them, however briefly. If we teach our computers to simulate that experience, we haven’t taught them how to produce poetry – we’ve taught them how to lie.
The important point here though is that this ideological blindness regarding the difference between a computer and a poet signals a broader blindness that is inscribed into the very DNA of capitalism. Marx called this blindness “reification”, i.e. the tendency of capitalism to reduce human beings to things. He argued that, even though a particular human practice originated as deeply embedded in and indistinguishable from social relations (say, the reading of poetry or the cooking and serving of food), if it is possible to strip it of all that and sell it in a “pure” way as a product or a service (i.e. something more akin to a thing), this is what capital will do. Thus, in economic terms, all that matters really is the Turing test.
It is in this context, I think, that we should read a recent essay by the French activist group The Invisible Committee, memorably entitled “Fuck Off, Google”. In this essay, the authors make a compelling case for a distinction between “technology” and “techniques”. All human societies and forms of life, they argue, are technical in the sense that they represent “a certain configuration of techniques, of culinary, architectural, musical, spiritual, informational, agricultural, erotic, martial, etc., techniques. And it’s for this reason,” they write, “that there’s no generic human essence: because there are only particular techniques, and because every technique configures a world, materializing in this way a certain relationship with the latter, a certain form of life.”
Different techniques, in other words, are radically incommensurable: there is no universal human being who can pick and choose between them as she pleases, and no world in which techniques can be neatly displayed as so many means towards the same end. Instead, “Every tool configures and embodies a particular relation with the world, and the worlds formed in this way are not equivalent, any more than the humans who inhabit them are. And by the same token these worlds are not hierarchizable either. There is nothing that would establish some as more ‘advanced’ than others. They are merely distinct, each one having its own potential and its own history.” Still, the authors argue, such an equation and hierarchization of different techniques is precisely what the field of technology offers. This is the important point. According to the authors, technology introduces “an implicit criterion making it possible to classify the different techniques”, a way of imagining techniques as if devoid of any ethical, and thus unique, character. In the case of capitalism, this criterion “is simply the quantifiable productivity of the techniques, considered apart from what each technique might involve ethically, without regard to the sensible world it engenders.”
If you want a definition of the difference between a computer and a poet, I would say that this is pretty much it. The only situation in which it would ever make sense to replace a poet with a computer is one in which the ethical character of poetry as a technique of communication has been replaced by the attempt to produce more from less. Understood from the perspective of technology, poetry produces the same thing as a can of tomato soup or a pack of cigarettes, only in a different way: shareholder value for whomever owns the means of its production, the “bottom line” of business. If a computer does this more effectively, it does it better. Understood as a technique, on the other hand, poetry produces an ethical relation between the reader and the poet, so that, when reading Goethe, I am not trying to achieve in a more efficient manner the same thing that I achieve when reading Shakespeare – and even less than when I use my dishwasher. Rather, when reading Goethe, I am seeking to enter a distinct ethical world. For most people, this is a reward in and of itself.
From this perspective, a world in which everything is either mediated or replaced by technology starts to look rather bleak – far from the glimmering visions offered by optimistic economists, in which everything “smart-“ is envisioned as the brave new future of all humanity. In fact, following the Invisible Committee’s suggestion, it seems that such a world must become increasingly “world-less”. As French philosopher Alain Badiou puts it: “Any world, for Plato and for me, only becomes visible, is only thrown into relief, by the differences constructed within it […] But within a horizon in which everything is equivalent to everything else, no such thing as a world is discernible, only surfaces, supports, apparitions without number.”
Perhaps in such a world, it is rather that we will be “post-“ everything but capitalism.